Case Report/Series

Introduction 

Clozapine is an atypical antipsychotic, it is the treatment of choice for treatment resistant schizophrenia and more effective than conventional neuroleptic medications. Clozapine is associated with potentially life-threatening side effects, some of which appear early in treatment. 

Myocarditis is an uncommon but serious early adverse event of Clozapine, the majority of reported cases occurring in the first 4-8 weeks.¹ Clozapine induced myocarditis (CIM) can present with mild symptoms, but can progress rapidly to fulminant symptoms and thereafter heart failure and death.¹  These symptoms and signs typically include dyspnoea, palpitations, chest pain, fatigue, flu-like symptoms, pyrexia and tachycardia.

Case Report

A 21-year-old Caucasian male with a two year diagnosis of schizophrenia and previously inadequate responses to Risperidone and Olanzapine was commenced on Clozapine. The patient had previously tolerated Risperidone and Olanzapine and did not experience adverse events, but there was inadequate therapeutic response to both; hence it was decided to commence Clozapine.

On admission, his physical examination, baseline blood investigations (these did not include cardiac markers such as troponin or C-reactive protein (CRP)) and electrocardiogram (ECG) were normal. His medical history was unremarkable and he did not have a family history of cardiac disease. He smoked 15 cigarettes per day.

A rapid Clozapine titration compared with the standard UK titration^[2] was commenced with a target dose of 200 mg/day on day 14. He was not on any other psychotropic medication.

The patient remained asymptomatic in the first 3 days. On day 4, he developed tachycardia (114 BPM). A repeat physical examination and ECG was normal, eventually his heart rate settled to 94 BPM. The tachycardia was deemed to be a benign side effect of Clozapine, and the rate of titration was slowed down as a precaution.

On day 12, the patient reported dizziness when standing and a ‘cold air’ sensation in his chest. Nurses reported that blood pressure was normal with a heart rate of 145 BPM but when reviewed clinically his heart rate was 89 BPM. His titration was continued. 

On day 14, the patient complained that his ‘internal organs were hurting’. His Clozapine dose was 125 mg/day at the time. He reported chest tightness with central pain, pain in his legs and abdomen, intermittent breathlessness and palpitations. The duration of his symptoms was 24-36 hours.  Examination was normal except for a heart rate of 110 BPM. His ECG showed sinus rhythm with no ST segment or T wave changes. Blood tests showed markedly elevated troponin I—1211.5 ng/L (normal range: <34.3 ng/L), CRP—176 mg/L (normal range: 0-10 mg/L) and eosinophil count —1.28 10⁹/L (normal range: 0.02-0.5 10⁹/L).  

The patient was afebrile throughout the titration period.

He was admitted to an acute hospital and a provisional diagnosis of Clozapine induced myocarditis was made. The echocardiogram did not reveal structural abnormalities or damage. An EBV (Epstein Barr Virus) serology was negative. Clozapine was withheld and the patient improved along with the blood markers, after 4-5 days he was discharged back to the psychiatric hospital.

Discussion

CIM is an often overlooked adverse event associated with Clozapine titration. Currently there is no mandatory requirement of laboratory monitoring for detecting myocarditis during Clozapine titration unlike the mandatory requirement for detecting neutropenia, despite roughly similar estimated incidence of the two adverse events at 3%.^[3,4]

This case was unusual because of the very early appearance of symptoms, the patient’s age and atypical symptom presentation. Although CIM is an early adverse event, the onset within 2 weeks of initiation was unusual. Literature suggests that myocarditis typically presents within 4-8 weeks.^[1] The patient was also younger than the reported median age of patients (30).^[1] The symptoms appeared at a low dose of 125mg/day which literature suggests is unusual, although CIM at doses of 50mg/day has been reported.^[5]

Tachycardia and fever are common early side-effects of Clozapine. Tachycardia usually settles after 4-6 weeks of treatment^[6] and fever typically for 2-3 days.^[7] Both symptoms can be suggestive of myocarditis, especially when they co-occur. CIM often presents in a non-fulminant form.^[8] As this case demonstrates many patients may not report symptoms when CIM is mild.  

Increasing age, concomitant administration of sodium valproate and increased rate of dose titration are significant risk factors for CIM.^[9] In this case, the patient was young and sodium valproate was not co-administered. The titration was originally intended to be rapid but slowed down soon after commencement.

Given the clinical difficulties in detecting mild CIM, we suggest that all patients have baseline troponin, CRP, heart rate, blood pressure, temperature, resipatory rate and ECG. If medical history reveals history of heart disease, a baseline echocardiogram can be obtained. If there is history of congestive cardiac failure, then baseline brain natriuretic peptide (BNP) or N-Terminal pro-B-type natriuretic peptide (NTproBNP) should be measured.^[10]

In clinically asymptomatic patients, if there is elevated baseline CRP (>100 mg/L), troponin, BNP or NTproBNP then Clozapine titration should not commence and further advice from cardiology should be sought. 

Weekly CRP and troponin should be done in the first month of titration and levels repeated once after stable dose of Clozapine is reached. The dose increase should not be rapid.

Tachycardia developed should be checked with reference to the baseline heart rate measured before commencing Clozapine. A heart rate of greater than 120 BPM or increase of more than 20 BPM over the baseline pulse rate should lead to the review of physical health, blood monitoring, ECG, and Clozapine titration rate.

An increase in troponin above upper limits or an increase in CRP should trigger consideration of CIM. Literature suggests that troponin levels greater than 2x the upper normal limit are indicative of acute myocarditis.^[9] CRP is raised on average 3 days before any increase in troponin levels is detected.^[9] If the troponin levels are within the normal range and the CRP levels are raised but less than 100 mg/L, clozapine titration can continue, but the pace must be slowed. Troponin levels and CRP levels should be monitored daily and the patient should be closely monitored for clinical signs of developing cardiotoxicity.

We do not recommend routine eaosinophil monitoring as the marker in 90% of cases does not exceed normal limits at the onset of CIM and typically peaks 7 days after cessation of Clozapine.^[11]

Conclusion

Clozapine induced myocarditis often presents with low level cardiotoxicity. Mild symptoms may be missed; however, progression to fulminant myocarditis can be rapid, with high mortality rates.^[1] Myocarditis, including clinically asymptomatic myocarditis remains a risk with Clozapine every time the patient is titrated onto this medication^[12]. Close clinical monitoring, high index of suspicion and monitoring of cardiac parameters will help early detection of adverse cardiac events.

INTRODUCTION

The pituitary gland is a tiny gland located at the base of the brain and is connected to the hypothalamus. Dubbed as the body’s “master gland”, it produces important hormones that control many bodily functions such as those involved in the control of haemodynamics, glucose, fight or flight response, body growth and many more. Any of the pituitary hormones may be affected in pituitary disease, with acute adrenocorticotropic hormone (ACTH) deficiency being the most catastrophic and life-threatening.

Pituitary apoplexy occurs following acute haemorrhage or infarction of the pituitary gland, causing patients to be acutely unwell due to hormonal as well as local compressive effects. These effects cause the usual presentation of pituitary apoplexy such as severe headache, diplopia, visual loss and hypopituitarism.

We report a case of pituitary apoplexy that presented with a 2-week history of loss of peripheral vision and lethargy with stable vital signs.

CASE PRESENTATION

A 49 years of age gentleman complained of loss of peripheral vision in the left eye and lethargy for 2 weeks. The loss of vision was sudden, painless and non-progressive and had caused him considerable difficulties with driving where he would shift into the wrong lane and was honked at. He had no known medical or surgical history of note. Prior to presentation, he had no history of eye pain, eye redness or a history of trauma to the left eye. There were no headaches, neurological deficits or constitutional symptoms.

Clinically, he had bitemporal hemianopia with no other cranial nerve deficits. His Glasgow Coma Scale was 15/15, vital signs were stable and there was no postural change in blood pressure. Examination of other systems was unremarkable. Blood investigations revealed a decreased morning cortisol of 46 nmol/L and a normal thyroid stimulating hormone (TSH) with borderline low free thyroxine. Serum electrolytes, plasma glucose and all other anterior pituitary hormones were within reference range.

A computed tomography (CT) of the brain showed an enlarged pituitary sella with a large well-circumscribed and heterogeneously enhancing mass within. This mass measured 3.5cm x2.7cm x3.5cm (AP xW xCC) and had no calcifications within. It was also compressing onto the optic chiasm.

Two days later, a brain pituitary Magnetic Resonance Imaging (MRI) was done which reported a heterogeneous mass occupying the sella with suprasellar extension measuring 2.7 x2.8 x2.9cm (AP xW xCC) (Figures 1.1 & 1.2). This mass returned mixed solid-cystic intensity with significant enhancement post-contrast administration. There was evidence of layering within the cystic component of this mass. Inferiorly, the right border ended lower than the left (Figures 2.1 & 2.2).

Following consultations with endocrinologists, neurosurgeons and radiologists, a clinical diagnosis of pituitary apoplexy with hypocortisolism and central hypothyroidism was reached. The patient was started on oral hydrocortisone 20mg in the morning and 10mg in the evening; and oral L-thyroxine 100mcg in the morning before he was referred to the neurosurgeon for trans-sphenoidal surgery. While awaiting surgery, no clinical deterioration was reported. An endoscopic trans-sphenoidal surgery successfully took place a week later which revealed an enlarged haemorrhagic pituitary gland (Figure 3.0). The patient was discharged well a week post-surgery.

His histopathology report later confirmed pituitary adenoma where monomorphic tumour cells arranged in nests and trabeculae and some pseudorosettes were seen. The tumour cells exhibited mild pleomorphism with moderate amount of cytoplasm. The stroma was highly vascularised. No necrosis, calcification or mitosis was seen. Immunohistochemistry studies were positive for follicle-stimulating hormone (FSH) and luteinising hormone (LH) and negative for ACTH, growth hormone, prolactin and TSH.

Figure 1.1 and 1.2: MRI brain on coronal view illustrating well-defined and heterogenous suprasellar mass

Figure 2.1 and 2.2: MRI brain on sagittal view illustrating mixed solid-cystic intensity pituitary mass

Figure 3.0: Intraoperative finding showing haemorrhage of the pituitary gland

DISCUSSION

Pituitary apoplexy is a potentially fatal condition caused by haemorrhage or infarction or both. Most cases occur during the fifth decade of life, predominantly in males.¹ In the majority of cases, it is associated with a pre-existing non-functioning macro-adenoma which accounts for 14-54% of pituitary adenomas and has a prevalence of 7-41.3/100,000 population. The standardised incidence rate is 0.65-2.34/100,00.²

The many clinical presentations of pituitary apoplexy result from local compression of adjacent structures or deficiency of pituitary hormones – the former being more common where affected individuals present with headaches, visual disturbances and other symptoms of raised intracranial pressure.³

Subclinical haemorrhages refer to asymptomatic individuals with evidence of pituitary haemorrhage on MRI. In a 2018 retrospective transversal analysis involving 64 patients, 34.38% had subclinical haemorrhage within a non-functional adenoma.⁴ In another retrospective overview by Turgut et al, 186 cases of apoplectic pituitary adenoma presenting with monocular or binocular blindness were published in the last century.⁵ In a case report by Sasaki et al, a 65-year-old gentleman was only diagnosed with pituitary apoplexy following weeks of blood investigations for hyponatraemia and repeat imaging. His only presenting complaints were anorexia, low energy and fever for two weeks.⁶ These studies show that while an early correct diagnosis of pituitary apoplexy is important, it is not necessarily urgent.

On the other end of the spectrum, pituitary apoplexy may also present as a life-threatening situation where patients are unconscious and hemodynamically unstable due to hypopituitarism. In its acutely deficient state, ACTH causes acute adrenal insufficiency hence resulting in hypotension, hypoglycaemia, hyponatraemia and hyperkalaemia. Sometimes, non-specific symptoms precede the symptoms of hypocortisolism. A drop in consciousness level may be due to the tumour’s mass effect transmitting pressure to the brainstem or causing hypothalamic compression.⁷ Espinosa et al reported a 48-year-old gentleman with pituitary apoplexy who presented with the worst headache of his life, requiring urgent neurosurgical intervention which proved to be life-saving.⁸

Complex as it already is, diagnosing pituitary apoplexy may be further complicated when non-specific symptoms can be explained by other causes such as post-general anaesthesia drowsiness, hyponatraemia in a patient on diuretics and headaches in post-partum women receiving spinal anaesthesia.^{9, 10}

While most patients consequently suffer from pituitary insufficiency, the extent, type and duration of therapy differs between patients. Cases of spontaneous recoveries whether a surgical or conservative approach was adopted have also been reported.^{11, 12} However, robust control studies comparing the outcome of surgical with conservative management in patients with pituitary apoplexy have yet to emerge. Nonetheless, studies have proven that visual outcomes significantly improve with surgery.^{13, 14}

Having discussed the varied presentations of pituitary apoplexy, it can be agreed upon that the life-threatening endocrinal condition should be considered in any patient with abrupt neuro-ophthalmic deficits despite the state of clinical stability. This is imperative as prompt medical and surgical management may not only be life-saving, but also significantly improve visual and cranial nerve outcomes.¹⁵

CONCLUSION

Pituitary apoplexy is an endocrinal emergency which requires immediate investigation and treatment. Despite its disastrous pathology, there have been cases where affected patients present with isolated visual disturbances or with no symptoms at all. It is therefore important to have early suspicion of pituitary apoplexy in stable patients with eye complaints as early detection and management are life-saving and significantly improve neuro-ophthalmic outcome.

Introduction

Nocardia sp. are aerobic, gram-positive microorganisms found mainly in soil and stagnant water. Nocardiosis commonly occur in immunocompromised patients, is often multi-systemic and easily mistaken for tuberculosis when involving the lungs ^{1 2}. We report a complex case of disseminated nocardiosis in a patient already suffering from Myasthenia Gravis (MG) and azathioprine-induced myelosuppression, in which selection of antimicrobials and management planning became complex.

Case Report

A 68-year old lady, known to have generalized MG for 12 years, presented to our centre following a week history of diarrhea and lethargy. She describes having loose stools, up to 8 times per day, chills and rigors, but denied per rectal bleeding or melaena. The patient was, on regular oral prednisolone (50mg a day). She had recently ceased her azathioprine 4 months prior due to severe anaemia, linked to myelosuppression following a bone marrow aspiration test revealing markedly reduced erythropoiesis, normal oesophageo-gastro-dudodenoscopy and colonosocopy, and having normal thiopurine methyltransferase levels prior to azathioprine initiation. The patient had several admissions in that 4 months due to her anaemia, requiring packed cell transfusions, and had recently sustained an interhemispheric subdural bleed two weeks prior following a mechanical fall. At the time, hemoglobin and platelet levels had returned within normal range. Clinical examination was unremarkable and her vitals on arrival included an oxygen saturation of 98% on room air with a normal chest radiography on arrival (Figure 1a). The patient was subsequently treated for infective gastroenteritis, and was started on metronidazole.

Figure 1: Chest radiography on (a) initial presentation and (b) 48-hours into admission, showing newly developed diffuse bilateral consolidation.

Unfortunately, 48 hours into her inpatient stay, the patient developed acute dyspnoea in which her oxygen saturation dropped to 77% under room air and examination revealed diffuse, coarse crepitations bilaterally. A repeat chest radiograph confirmed diffuse consolidation bilaterally (Figure 1b). She was subsequently treated for a hospital-acquired pneumonia, and had her antibiotics swapped to intravenous ceftriaxone. Unfortunately, the patient showed little improvement on the ward, which prompted further investigation. Her blood cultures revealed presence of nocardia farcinica which was resistant to ceftriaxone. In view of being immunosuppressed, a computed tomography (CT) imaging of the brain, thorax, abdomen and pelvis (Figure 2) was performed. Imaging showed extensive bilateral lung consolidation and reticulonodular opacities, predominantly in upper lobes. There was also evidence of sigmoid colon diverticulitis with a rim-enhancing collection adjacent to the posterior aspect of the proximal-to-mid sigmoid colon (2.9 x 2.6 x 2.6 cm).

Figure 2: Computed tomography (CT) imaging of (a) the brain, revealing inter-hemispheric hyperdensity consistent with a subdural haematoma, (b) the thorax, revealing extensive bilateral lung consolidation and reticulonodular opacities, predominantly in upper lobes, and (c) abdomen, showing sigmoid colon diverticulitis with a rim-enhancing collection adjacent to the posterior aspect of the proximal-to-mid sigmoid colon (2.9 x 2.6 x 2.6 cm).

Following consultation with our Infectious Disease team, the patient was treated for likely disseminated nocardiosis using intravenous trimethoprim/sulfamethoxazole as a recommended first line therapy. Unfortunately, following 72 hours of trimethoprim/sulfamethoxazole administration, the patient developed severe pancytopenia (haemoglobin 63 g/L, white cell count 2.0 x 10⁹/L and platelets 33 x 10⁹/L) requiring packed red cell and platelet transfusions, as well as immediate cessation of trimethoprim/sulfamethoxazole. Further decisions on antimicrobial proved difficult as it was noted that other effective alternatives against nocardiosis, including amikacin and fluoroquinolones may potentially exacerbate MG symptoms. Furthermore linezolid, another possible alternative, could potentially exacerbate her thrombocytopenia and worsen the pre-existing subdural haematoma. Thus, a decision was made to commence co-amoxiclav and meropenem concurrently despite neither being first-line therapy. The patient subsequently had an uneventful CT-guided drainage of the sigmoid abscess, with the assistance of our General Surgical colleagues, and was kept on prolonged intravenous antibiotic therapy for 3 months. Following improvement in clinical state, she was allowed discharge from hospital with ongoing oral co-amoxiclav and linezolid with regular bloods tests to monitor for myelosuppression as an outpatient, which has not occurred till this date. With support from clinical and radiological improvement, we aim for 12 months of therapy.

Discussion

Various cases of nocardia sp bacteraemia have been reported, in which immune system dysfunction was primarily due to chronic glucocorticoid therapy ^{1 3 4}. A retrospective review of 40 patients from a Chinese tertiary hospital revealed that half of patient with nocardiosis were on corticosteroids prior to infection onset ⁵. An even larger case series reported up to 94% of pulmonary nocardial infection being linked to use of immunosuppressants ⁶.

Management of nocardiosis can be challenging. Often, imipenem, trimethoprim/sulfamethoxazole, amikacin or a combination of these antibiotics are recommended, with treatment duration being guided by clinical and radiological improvement often extending up to a year ⁷. Bactericidal agents including carbapenems and amikacin are often recommended alongside trimethoprim/sulfamethoxazole in cases of disseminated nocardiosis to ensure greater success in treatment ^{7, 8}. The use of amikacin however should be cautioned in cases of MG, and reports have described management using linezolid as an alternative ^{1, 9}.

In our patient, the risk of worsening pancytopenia as illustrated following trimethoprim/sulfamethoxazole administration led to hesitance in using linezolid. Furthermore, the consequences of thrombocytopenia would have been devastating for the patient due to having residual subdural hematoma. Thus, the choice of antibiotics was limited to that of second-line agents including co-amoxiclav and meropenem on top of surgical drainage of existing abscesses, which has also been shown to be beneficial ^{1 5}. However, it should be noted that the risk of pancytopenia remains to be multifactorial in our patient, ranging from pre-exisitng myelosuppression to ongoing sepsis and initiation of culprit medications

Mortality of nocardiosis infections ranges depending on organ involved, rates being under 40% in cases of pulmonary spread, and up to 100% when disseminated to the central nervous system (CNS) ⁶. Unfortunately, nocardia farcinica, the subtype reported in our case, has been shown to be more resistant to antimicrobials and carries a greater risk of dissemination to the CNS, a point being greatly considered as our centre continues to manage the patient ¹⁰.

Conclusion

Nocardiosis, although not uncommon in immunocompromised individuals, poses a unique challenge in the MG population due to limitations in suitable antibiotics. Our case highlights the importance of a multi-disciplinary approach, involving various specialties including the infectious disease, neurology, general surgical and microbiology to ensure successful management of such complex cases.

Background:

Evidence suggests that Serotonin has an important role in bladder control through central and peripheral neurological pathways. The three main serotonin receptor sites involved in the micturition pathway are 5-HT1A, 5-HT4, and 5-HT7. 5-HT7 and 5-HT4 are excitatory to acetylcholine release and 5-HT1A is inhibitory. Increased serotonergic activity leads to parasympathetic inhibition, which results in urine retention. It is through this mechanism of action and their effect on pre-synaptic serotonin 1A and peripheral 5-HT3 receptors that SSRIs were observed to have anti-enuretic effect. However, the exclusive role of serotonin in this regard is not fully understood because along with serotonin, other neurotransmitters, particularly acetylcholine are also implicated in micturition physiology. Acetylcholine is released from nerves innervating the detrusor muscle and causes bladder contraction resulting in voiding. Contrarily, adrenergic pathways lead to constriction of the bladder sphincter and promote continence. There have been suggestions that at lower intrasynaptic 5-HT concentrations, there is prevalence of inhibitory control of micturition, whereas excitatory effect is more pronounced at higher concentrations of 5-HT. This may suggest a dose-dependent relationship between Sertraline and urinary side effects. ¹

Case Reports:

Case 1

A 14 year old girl with a diagnosis of moderate depressive episode was prescribed Sertraline 150 mg once daily. On follow up with her community psychiatrist, mum reported that she had been having episodes of bedwetting on a regular basis for almost two weeks. There was no past history of enuresis, no medication changes, or changes to her diet or routine. She had been drinking fluids during the day and had limited fluid intake after 6 pm. On a visit to the Sheffield Children’s Hospital, she had been diagnosed with a urinary tract infection and was prescribed a five-day course of antibiotics. She denied symptoms of abdominal pain, dysuria or fever.

On discussion with the trust pharmacist, it was reported that urinary incontinence is a rare listed side effect of Sertraline with nocturia occurring in 1 in 100 to 1 in 1000.² At further medication review appointments, the patient continued to report being incontinent on approximately every alternate night and had to use incontinence pads. It was agreed with the patient to reduce the dose of Sertraline to 100 mg once daily to test if her urinary incontinence was linked to Sertraline and review after 2 weeks in clinic. The reduction in Sertraline dose to 100mgs once daily did not alter the frequency of bedwetting that continued on most week nights and varied from being partial to full emptying of the bladder. As a result, she was then referred to the Paediatric Community Incontinence clinic for further investigation regarding the sudden onset of these night bedwetting episodes. Concurrently, Sertraline was gradually reduced and stopped. She was switched to Fluoxetine liquid for treatment of her depressive symptoms, which was titrated to a dose of 16 mg once daily. At the community continence clinic, urine dipstick was negative. Systemic examination including a neurological examination was unremarkable. Mum reported that since the change in medication from Sertraline to Fluoxetine, there was a remarkable improvement in her urinary symptoms.

Case 2

A 16 year old boy with a diagnosis of mixed anxiety and depressive disorder was initiated on Sertraline which was gradually titrated to a maximum dose of 200 mg once daily. He reported improvement in his symptoms of anxiety and depression. However, a few days into taking the higher dose, he experienced symptoms of hesitancy with micturition and failure to ejaculate. On reduction of Sertraline to 100 mg once daily, he reported complete resolution of urinary and sexual side effects, while still reporting a reactive and stable mood. Due to his significant progress, he was eventually discharged from CAMHS back to the care of his GP.

Case 3

A 12 year old girl with a diagnosis of Generalized Anxiety Disorder and Attachment Disorder reported three incidents of urinary incontinence whilst being on Sertraline 200 mg once daily. Sertraline was discontinued by the patient against medical advice. No follow up information was available to observe for resolution of symptoms after discontinuation of Sertraline.

Discussion:

Selective Serotonin Reuptake Inhibitors (SSRIs) are a very commonly used class of psychotropic medication in the CAMHS population to treat depression, anxiety, PTSD and OCD. ³ It is evident by the cases discussed above that SSRIs may have a key link in causing symptoms of urinary dysfunction, which may range from nocturnal enuresis to acute urinary retention. This could be explained by Serotonin’s pivotal role in micturition through central and peripheral pathways. There is not enough evidence on the links in a child and adolescent population as most of the studies are on an adult cohort.⁴

Conclusion:

In conclusion, it is important for clinicians to bear in mind the genitourinary side effects of SSRIs, which may be debilitating for patients in the CAMHS population. It is equally important for us as clinicians to educate young people and their parents about these potential side effects and how they can be managed. It has also been observed that higher doses of Sertraline have shown a possible link between onset of urinary side effects. In order to establish a significant causal and dose-related relationship on the onset and severity of genitourinary symptoms, studies with a larger sample size followed up over a longer period would be required.

Case Report:

A 33 year-old ASA1, primigravida, presented to our delivery suite with spontaneous onset of labour at 38 weeks of gestation. Epidural analgesia was commenced to alleviate her labour pains. Subsequently, she underwent an assisted vaginal delivery of a live male baby (weighing 4660 gms) using Keiland’s outlet forceps after 90 min second stage of labour. 10 hours postpartum, she complained of dyspnoea & severe central substernal chest pain. She was noted to have an unusual swelling of face and neck with oxygen saturations of 90 % on room air. Ascultation of chest revealed normal bronchovesicular breath sounds, normal heart sounds with absence of added sounds. Arterial blood gases showed an O₂ tension of 11 kpa, CO₂ tension of 5 kpa and pH of 7.34. The diagnosis of subcutaneous emphysema, pneumomediastinum and small left apical pneumothorax (Hamman’s syndrome) was confirmed on chest X-ray (CXR 1). We ruled out differential diagnosis of pulmonary embolism, Tension pneumothorax, angina pectoris, pericarditis, dissection of aortic aneurysm, mediastinitis, cardiac tamponade, chest infection & oesophageal tear. She was managed conservatively by close monitoring for complications, administration of supplemental oxygen and use of simple analgesics. She demonstrated a complete uneventful recovery over the next 24 hours with normalising of chest signs (CXR 2).

CXR 1: shows pneumomediastinum, extensive surgical emphysema & a left apical pneumothorax.

CXR 2: shows small pneumomediastinum, the surgical emphysema & pneumothorax resolved.

Discussion:

Hamman’s syndrome is named after Louis Hamman (1847-1946), the physician who first described it in 1945. The first reference to this condition was in 1618, when Louise Bourgeois, midwife to the Queen of France, wrote, “I saw that she tried to stop crying out and I implored her not to stop for the fear that her neck might swell”³_.

Hamman’s syndrome usually occurs in the 2^ndstage of labour & is associated with prolonged and protracted labour and larger than usual babies ⁴_. However, the clinical presentation is often delayed to the postpartum phase as was clearly seen in our case. The condition seems to be provoked by any valsalva manoeuver such as vigorous coughing/vomiting/sneezing, forced physical activity & enormous efforts during spontaneous vaginal delivery. Its occurrence is usually related to the expulsive phase of labour when ‘pushing down’ actively raised the intraalveolar pressure. This may subsequently increase the intrathoracic pressure up to 50 mm of Hg or higher¹. Rupture of marginal alveoli with air entering along the perivascular sheath into the mediastinum is the most likely mechanism, in our case. It is probable that, the air tracts through the fascial planes into subcutaneous and retroperitoneal tissues. Other reported mechanisms of Hamman’s syndrome include oesophageal rupture during childbirth, or pneumomediastinum related to asthmatic bronchospasm⁵ or chest infection, or dissection of pneumoperitoneum, secondary to epidural catheter placement or caesarean section^1.

Palpable crepitus on face & neck is suggestive of subcutaneous emphysema & appearance of this emphysema in labour is the hallmark of pneumomediastinum. Other features of pneumomediastinum include substernal chest pain, dyspnoea, voice change, cough, sore throat and tachycardia¹. Hamman’s sign, a fine auscultatory crepitation synchronous with the heartbeat, heard along the left sternal border; is sometimes observed in this condition².

Chest X-ray and CT thorax are the diagnostic tests. Majority of the patients with Hamman’s syndrome have pneumomediastinum & subcutaneous emphysema without any pneumothorax and this requires supportive management with strict monitoring. Our patient demonstrated a small pneumothorax, which was managed conservatively. A surgical intervention in the form of subcutaneous air drainage may occasionally be indicated in severe cases.

Overall most cases have a benign, self-limiting course when the aggravating factors are no longer present. Published data indicates that subsequent pregnancies pose no additional risk of recurrence⁵.

Conclusion:

Since Hamman’s syndrome is a potentially dangerous complication of normal childbirth. We propose that every obstetric anaesthetist and obstetrician should be aware of this syndrome.

Introduction:

The spectrum of psychiatric illness in Systemic lupus erthyromatosus (SLE) include psychotic, depressive, subtle cognitive and personality disorders of histrionic type. The occurrence of psychiatric manifestations in SLE varies widely from 5 to 83%. It is postulated that there is a direct action of the disease on the central nervous system by autoantibodies namely anti phospholipid and anti-ribosome P auto antibodies or cytokines like interleukin 2, interleukin 6, alpha interferon ¹. During the course of the disease side-effects of glucocorticosteroids and hydroxychloroquine or anxious reaction to chronic and potentially lethal illness is postulated to be another mechanism of psychiatric manifestation of SLE . SLE patients are prone to develop myriad of psychological distress in addition to neuropsychitric manifestations which require a social and psychological support. While some of these manifestations are treated by corticosteroids and psychotropic drugs¹ medications with anticholinergic side-effects, like phenothiazines, tricyclic antidepressants and hydroxyzine which enhance the oral dryness should be avoided in SLE.

Clinical scenario:

A 27-year-old male suddenly developed aggressive behaviour for the first time in his life ,while on his work place. The patient had no insight into his illness and was brought to the local psychiatric hospital by his colleagues where he was admitted as a case of acute mania. He was managed with electroconvulsive therapy (EST) in addition to antipsychotic medication as neuro imaging including CT scan and MRI brain were normal. A few days later , the patient was discharged on anti-psychiatric medicines. However, after six months while on antipsychotic medication, he developed a low grade fever .He was admitted to a local hospital where in addition to base line investigations a lymph node biopsy was done which revealed follicular hyperplasia, without any abnormal cell. Patient’s HBV, HCV, HIV were negative. The patient developed anorexia , significant weight loss and progressive difficulty in getting up from a sitting position .He also developed shortness of breath and presented to King Abdul Aziz specialist hospital in Taif, Saudi Arabia virtually in a bed bound state . He was admitted in the intensive care unit of the hospital .The examination revealed pallor, generalised lymph-adenopathy, palmer rash, alopecia and mouth ulcers. The patient had mild pericardial effusion and Mitral regurgitation (MR)++ on echocardiography. Further evaluation showed significant proteinuria. Serum ANA, dsDNA were positive .Lupus anticoagulant was negative. Keeping in view above symptoms and signs the patient was diagnosed as a case of SLE² (Mouth ulcers, Pericardial effusion ,ANA positive , dsDNA positive ) The patient was managed with pulse dose of methylprednisolone 1g intravenously (IV) daily for 5days, followed by oral prednisone 60 mg once daily, which was tapered on follow up . Patient tolerated the treatment well and improved progressively . He became ambulatory and rejoined his job. The psychiatric medications were stopped.

However, on follow up the patient continued to have proteinuria 1.8 gm/24 hr. He was readmitted and the kidney biopsy revealed class IV lupus nephritis. He was given pulse cyclophosphamide 1gm/m² intravenously and later started on tablet Mycophenolate 1.5gm once daily. The proteinuria improved and he is following our clinic for the last two years now .Patient’s follow up investigations are shown in table 1.

Table: 1 Patients’ hospital investigations and results

Discussion:

The correct diagnosis of central or even peripheral nervous system manifestations in patients with SLE can be challenging because of many SLE-related and non-SLE-related processes present in a patient. The index case proved to have acute mania as the first manifestation of SLE which remained under oblivion till he developed serositis another complication of SLE. While this patient came to clinical attention after one year a case of SLE masquerading schizophrenia for 14 years was reported by Funaunchi et al³. In another report, a 14-year-old boy with a two-year history of cognitive dysfunction and behavioural problems SLE was diagnosed after two years⁴ . It appears that the psychiatric symptoms may occur as the first manifestation of juvenile SLE. It will not be out of place to mention that the psychiatric manifestation can be at times dire which could even result in harm to others in a given society. The case of Folie a trios syndrome, characterized by the transfer of delusional ideas from one person to two other persons culminating in murder has been reported in a patient with SLE⁵ . In a significant retrospective data from China (a cohort of 518) neuropsychiatric manifestations in SLE were observed in 96(19%) of the above study cohort . The seizure disorder accounted for the most prevalent disorder of neuropsychiatric manifestations (NP) of SLE followed by cerebrovascular disease and acute confusional states. In the above study, 96 patients with psychiatric symptoms, acute psychosis was observed in 10(11%)patients. Authors in this study were of the opinion that this percentage could have been higher if subtle cognitive dysfunction were included as well. Authors of the same study further concluded that the antiphospholipid antibodies were significantly associated with NP manifestations, especially cerebrovascular disorders⁶.

The autoantibodies have been found to be biomarkers for future neuropsychiatric events in SLE. A prospective study throughout ten years conducted among 1047 SLE patients demonstrated that individuals who had evidence of lupus anticoagulant (LA) were at an increased future risk of intracranial thrombosis. Further, those with anti-ribosomal P antibodies were at an increased future risk of lupus psychosis⁷. The Lupus anticoagulant in the index case was negative, and anti-ribosomal P antibodies were not available . A study by Sanna et al⁸ have shown that an association exists between anti-NR2 antibodies and depressed mood in addition to decreased short-time memory and learning. Authors in this study concluded that antibodies to NMDA receptors thus might represent as one of the several mechanisms for cerebral dysfunction in patients with SLE.

The CT scan brain of the index case was normal. However, massive bilateral calcification of sub-cortical structures in a patient with SLE with the psychotic disorder has been reported^9. The psychiatric diseases are related to vasculitis and non-inflammatory vasculopathy of the small cerebral blood vessels. Further, a study has shown that ninety per cent of the patients with psychosis, organic brain syndrome or generalised seizures had increased IgG antineuronal activity as compared with only 25 per cent of the patients who presented with hemiparesis or with chorea/hemiballismus. The authors in the above study concluded that the diffuse central nervous system manifestations of SLE are a direct result of the interaction of the antibody with neuronal cell membranes¹⁰.

The management neuropsychiatric manifestation in SLE should include treatment of the disease itself and specific psychotropic treatment. The index case had rapid improvement following Glucocorticosteroid therapy. Intravenous infusions of immunosuppressive agents, such as cyclophosphamide, have been found to be effective in such conditions ¹ . Psychotropic drugs may be used, but it is prudent to mention that SLE-inducing drugs, like chlorpromazine, carbamazepine and lithium carbonate must be avoided. Following treatment with steroids, the index case improved and all his antipsychiatric medications were finally stopped and he resumed his job.

To conclude the index case highlights that even though SLE is more frequent among females of childbearing age but males are no way immune to SLE . While evaluating patients with multiple unexplained somatic complaints and psychiatric symptoms SLE ought to be ruled out. The existence of neuropsychiatric manifestations in SLE constitutes an indisputable clinical reality that every practitioner must be able to recognise and treat.

Background

Necrotising soft tissue infections (NSTI) are severe and rapidly progressive, requiring rapid recognitions and early, often surgically-based, management. Mono-microbial types of NSTI (i.e. Type 2 NSTI), which amounts for 20 to 30% of overall cases, are often linked to invasive Group A Streptococcus or Staphylococcus Aureus infections ¹. Rarely, Group B Streptoccocus (GBS), also known as Streptococcus Agalactiae, are implicated ². We report a unique case of NSTI of the lower limbs due to GBS, with acute pericardial dissemination leading to cardiac tamponade, leading to a diagnostic dilemma due to co-existing cardiogenic and septic shock.

Case Report

A 51-year old gentleman of Chinese ethnicity presented with right foot pain and swelling over 2 weeks, associated with chest pain and shortness of breath during that period. He had a 10-year history of poorly controlled diabetes mellitus with a Hba1c level of 8.8 %, hypertension and dyslipidaemia.

He was hypotensive on arrival, with a blood pressure of 91/60 mmHg and hypoxic, requiring high flow oxygen of 15L/min to maintain saturations at 100%. Otherwise other vitals were stable, pulse rate being 72 beats per minutes, respiratory rate 24 breaths per minute and a temperature of 37.4 degrees Celsius.

Clinical examination revealed a gangrenous lateral two toes extending into the lateral malleolus on the right foot, with evidence of pus discharge and associated warmth and crepitus up to hindfoot level on palpation. There was also evidence of dry gangrene in the fourth toe of the left foot, with presence of a small puncture at dorsum of foot with pus discharge. Similarly, crepitus was felt up to midfoot level on palpation of the left side. Bilateral dorsalis pedis and posterior tibialis pulses were palpable but feeble.

Table 1: Blood Investigations on Admission

Figure 1a & 1b: Radiography of the (1a) left foot and (1b) right foot respectively, demonstrating gas within soft tissue bilaterally.

Figure 2: Chest radiography demonstrating cardiomegaly and globular-shaped heart, with loss of left-sided cardiophrenic and costophrenic angles.

Figure 3: Electrocardiogram demonstrating widespread saddle-shaped ST-elevation, consistent with pericarditis.

Figure 4: Parasternal Long Axis view of bedside echocardiography showing evidence of pericardial effusion and right atrial and ventricular collapse

Figure 5: Purulent pericardial aspirate via pericardiocentesis

Initial blood investigations are highlighted in Table 1. HIV Antibody, Hepatitis B Surface Antigen and Hepatitis C Antibody serology were all negative. Lower limb radiography revealed evidence of gaseous shadows bilaterally (Figure 1). The clinical and radiological findings were consistent with necrotising soft tissue infection of bilateral feet, and the patient was advised for extensive wound debridement and possible amputation of the affected sites during an orthopaedic consult.

However, on closer review of the chest radiography, there was evidence of cardiomegaly with a globular-shaped heart (Figure 2). His electrocardiogram on arrival, revealed diffuse ST-segment elevations on majority of leads, ST-segment depression on lead aVR consistent with pericarditis (Figure 3).

A bedside echocardiogram was performed, revealing a massive pericardial effusion, measuring largest at 2 cm in depth, with evidence of both right atrial and ventricular collapse (Figure 4).

An urgent pericardiocentesis was performed, under echocardiographic guidance, which revealed turbid-looking aspirate (Figure 5). Urgent microscopic analysis revealed 45 Cells per mm³, majority of which were lymphocytes, and gram stain showed moderate amounts of pus cells with occasional gram positive cocci. Pericardial fluid was negative for acid-fast bacilli.

A repeat transthoracic echocardiogram was performed post-pericardial drain insertion, revealing minimal remnant pericardial fluid, with the pericardial drain in-situ, and no evidence of any mass or vegetation. Unfortunately, a transoesophageal echocardiography and Computed Tomography (CT) imaging of the mediastinum (to rule out mediastinitis and pneumonitis) was not performed, as management of the NSTI took precedence.

The patient was started on intravenous antibiotics, both tazobactom-pipericillin and clindamycin. There was a delay in performing limb saving wound debridement as the patient was reluctant for invasive management, but had later consented to the procedure which was performed only on day 3 of admission. Tissue cultures were taken peri-operatively. Unfortunately, the patient deteriorated post-operatively due to extensive blood loss and overwhelming septicaemia and succumbed to his illness 72 hours after. Subsequently, it was revealed that cultures from blood, pericardial aspirate and tissue aspirate were positive for GBS infection.

Discussion

GBS is a common microorganism, often colonising the gastrointestinal and reproductive tract ³. Rarely, GBS colonises the skin and can cause necrotising fasciitis, i.e. necrotising soft tissue infection (NSTI), with only 22 cases having been reported in the past ii. Majority of these patients are either immunocompromised or have other predisposing factors including recent thoracic intervention or trauma ^{4, 5}.

GBS-related infections of cardiac structures are rare, as a whole, with 2 to 3% of cases presenting as native valve endocarditis and far less as pericarditis, mycotic aneurysms and intraventricular abscesses ³. Parikh et al reviewed the types of microorganisms isolated from purulent pericarditis samples and revealed that only 5% were due to streptococcal organism, sans Streptococcus Pneumoniae, possibly less so due to GBS ⁶. Our literature search revealed only one case of GBS-related purulent pericarditis reported although the case was not linked in any way to a NSTI to our knowledge ⁷.

Our case was unique as, at the time of writing, there were no other reports of GBS-related lower limb NSTI in combination with mediastinal involvement. There have been only a handful of cases of necrotising fasciitis reported with mediastinal involvement, the majority of which were supra-diaphragmatic with only one reporting NSTI of the lower limb due to Aspergillus infection ^{8 9}.

The similarity in culture results obtained from blood, tissue aspirate and the pericardial fluid in our patient suggest dissemination of GBS from the NSTI, possibly via a haematogenous route, although bacterial-related pericardial dissemination can also occur via direct spread from infected foci from neighbouring intra-thoracic structures or sub-diaphragmatic ⁴. The possibility of multi-routed spread should remind clinicians that, albeit rare, mediastinal involvement in NSTI is a possible complication of such disease.

Conclusion

This case highlights the rare possibility of cardiac involvement in cases of NSTI, and the possibility of cardiac tamponade causing cardiogenic shock masquerading alongside septic shock. It also highlights the importance of combining clinical findings with ancillary testing, including bedside echocardiography, when faced with challenging cases of sepsis to help look for possible foci of infection.

Introduction

Epidural anaesthesia is one of the favored and effective treatment options for labour pain. It is usually safe and only a handful situations lead to absolute contraindications to this technique such as patient’s refusal, lack of expertise and equipment, severe coagulopathy and infection at the site of puncture (1). However, as with any other technique and procedure, epidural anaesthesia is not flawless. The side effects and complications include hypotension, pruritus, inadequate analgesia, post puncture headache, nerve damage, infection, and epidural haematoma (1,2). Headache is common in one third of the patients after lumbar puncture however, the frequency is less in epidural anaesthesia as the fluid is injected in and not removed in the latter (3). Accidental dural damage and subsequent headache following epidural anaesthesia is uncommon and is an important cause of morbidity which can limit patient severely. Further, in rarest of rare cases Pneumocephalus can develop after epidural anaesthesia which has rarely been reported. We report a patient who developed Pneumocephalus after receiving epidural anaesthesia for labour pain.

Case Report:

A 39 year old female presented to our Emergency Department with severe headache not responsive to analgesics. The headache started developing 10 to 12 hours after she was given an epidural which was attempted three times for labour pain which was four days prior at a nearby medical center . The severity of the headache did not change with lying or the upright position. She had no symptoms of vomiting, no fever and no confusion. Neurological examination and vital signs were unremarkable. The site of the spinal anaesthesia did not reveal any swelling or any signs of infection. An urgent head CT scan was performed which revealed Pneumocephalus denoted by numerous left fronto-parietal extra axial air locules (Figure 1 and Figure 2). MRI spine revealed mild subcutaneous oedema at the site of the needle insertion without any haemorrhage or collection. The patient was admitted and treated conservatively for six days and follow up serial head CT scans showed complete resorption of the Pneumocephalus and the patient’s symptoms resolved completely. The patient was discharged and the follow up was uneventful.

Figure 1: Pneumocephalus seen as locules of air (black color) in the left fronto-parietal region denoted by arrows (Axial section)

Figure 2: Multiple pockets of air seen in the Saggital section marked by arrows demonstrate the Pneumocephalus.

Discussion:

Pneumocephalus is the presence of air in the intracranial cavity. It can be acute ( less than72 hours ) or delayed (more than 72 hours). The most common site is the frontal region (4). Plain skull x-rays can detect Pneumocephalus of about 2 ml, whereas it requires only 0.5 ml of air to be detected by a CT scan (5). Pneumocephalus is most commonly a result of traumatic brain injury, surgical intervention of the brain or infection (5). Trauma accounts for up to 75% percent of the total cases. Chronic infections of ENT especially otitis media also amounts to a number of significant cases. Surgical procedures of brain, spine and ENT like sinus surgery, nasal polypectomy and nasal septum resection accounts for the causes. The incidence after supratentorial craniotomy has been reported to be 100% (6, 7). However, it is very unusual for pneumocephalus to develop post epidural anaeasthesia possibly due to ball valve mechanism in which the air enters the space through the CSF leakage which allows input but not output. Headache post lumbar puncture and epidural anaesthesia is relatively not uncommon but certain situations may demand a more thoughtful approach (3).

In our patient we suspect there was a puncture of the dura during epidural anaesthesia which led to air being trapped and siphoned upwards in an inverted soda bottle fashion. This is supported by the meta-analysis done by Choi et al. which states the incidence of accidental dural puncture in epidural insertion to be 1.5% and among those 52 % will have post puncture headaches (8). In another extensive study performed over ten years, the overall incidence of accidental dural puncture and postdural puncture headache were 0.32% and 0.38%, respectively (9). The authors further stressed that if more than one attempt was required to identify the epidural space, the accidental dural puncture rate increased to 0.91%. In our patient we witnessed the same wherein three attempts were made to identify the epidural space which increased the risk of dural injury and subsequent leaking. Pneumocephalus usually gets absorbed without any clinical manifestations. The conservative treatment involves placing the patient at rest, avoiding Valsalva manoeuver, administering analgesics. With these measures, reabsorption was observed in 85% of cases after 2–3 weeks (5). Use of oxygen mask, nasal catheter, hyperbaric oxygen sessions and good hydration have also been reported. If conservative measures fail to provide the desired results then specific treatment like a epidural blood patch or even surgical closure of the dural gap is indicated (3, 10).

Background

This case highlights a rare and interesting medical condition bought about by liquorice ingestion.  While there have been many previous reports of liquorice toxicity secondary to eating confectionary, I have only found one case report of liquorice toxicity secondary to liquorice tea ingestion and the patient there had only a mild case of hypokalaemia and did not require hospital admission unlike patient X.¹

Case presentation

Patient X is a usually fit and well 49-year-old woman who was admitted following a collapse.  Prior to this collapse she had experienced a 3-week history of gradual onset, slowly worsening dull headache and a feeling of tingling in her hands which increased over this timeframe.  On the day of admission, she experienced nausea.  Her past medical history included migraines, for which she self-medicated with paracetamol, ibuprofen and codeine as necessary.  She was also using Cerazette.  She was in full time work, was a lifelong non-smoker and used alcohol very rarely.  She had no family history of note.

On examination, Patient X was a slim individual who was hypertensive with a blood pressure of 170/100 mmHg.  Her other observations were normal.  Her general and neurological examinations were unremarkable.

While initially the medical team felt that Conn’s syndrome was a likely cause of the abnormalities, this was reconsidered on a ward round where, following research on the internet into her abnormalities, patient X brought it to the attention of the team that she had been consuming between six and eight liquorice tea infusions per day for the past two months and had been consuming around three a day for a significant period time prior to this (roughly 18 months).  The diagnosis was made based on her history. 

Investigations

A venous gas demonstrated a metabolic alkalosis with a pH of 7.57.

Blood tests revealed hypokalaemia (K+ = 2.2 mmol/L) and hypophosphataemia (PO4 = 0.35mmol/L).  No other abnormalities were detected. 

More specialist assays were undertaken, and demonstrated that her plasma renin was 2.3 ng/mL/hour (normal range 0.2 – 3.3 ng/mL/hour).  Her morning supine plasma aldosterone level was 29 pg/mL (normal range 30 – 160 pg/mL).  Her morning plasma cortisol level was 612 nmol/L (normal range 138-635 nmol/L).

In addition, the patient also underwent a CT head and a CT renal angiogram, both of which were normal.

DIFFERENTIAL DIAGNOSIS

• Apparent mineralocorticoid excess
• Exogenous mineralocorticoid excess
• Liddle’s syndrome
• Congenital adrenal hyperplasia
• Cushing syndrome
• Liquorice

Treatment

Intravenous potassium and phosphate replacement, cessation of liquorice intake and amlodipine 5mg OD.

Outcome and follow-up 

The patient was discharged with a blood pressure of 132/66 mmHg on amlodipine, a potassium level of 2.9, and a phosphate level of 1.16.  She was given oral potassium supplementation to be taken 3 times per day.  After two weeks, the amlodipine was stopped as she became mildly hypotensive.  Her blood pressure was 120/60 following cessation of amlodipine.  Three months later her plasma potassium level was 4.6 without supplementation.  Mrs X required no follow up although she reports an ongoing feeling of tingling in her hands. 

Discussion

Liquorice is an extract of the roots of the Glycyrrhiza glabra plant and has been used as both a confectionary flavouring agent and a herbal remedy.  It is also commonly used as a laxative, and as a flavouring agent in chewing gums, sweets, and food products.

The active ingredient in liquorice is glycyrrhetinic acid which inhibits the enzyme 11-β-hydroxysteroid dehydrogenase.  This enzyme converts cortisol into inactive cortisone within the distal tubule of the kidney and so in liquorice toxicity there is a build-up of cortisol in distal tubular cells.²  This results in increased mineralocorticoid like activity as there are structural similarities between cortisol and aldosterone, with increased Na+ and water retention in conjunction with increased H+ and K+ excretion.³  Here hyperaldosteronism occurs, but with a low or low-normal plasma aldosterone and renin level.  Serum glycyrrhetinic acid levels can be measured with enzyme-linked immunosorbent assay (ELISA) and high-performance liquid chromatography (HPLC).   Urinary glycrrhetinic acid levels can be measured with gas chromatography-mass spectrometry (GC-MS).⁴

Pseudoaldosteronism secondary to liquorice consumption is a relatively rare occurrence.  Case reports demonstrate a range of clinical manifestations from an asymptomatic patient fortuitously diagnosed to those with more severe presentations such as rhabdomyolysis, hypertensive encephalopathy, asthenia, paralysis, heart failure, and cardiac arrhythmias such as polymorphic ventricular tachycardia and ventricular fibrillation secondary to hypokalaemia.  For these reasons, it has been suggested that the public should be made aware of the potential dangers associated with liquorice consumption. ^5-13

The combination of alkalosis hypokalaemia and hypertension suggests increased mineralocorticoid activity leading to increased renal tubular Na+ reabsorption along with increased k+ and H+ excretion.  Both primary and secondary hyperaldosteronism cause these abnormalities, the former via an appropriate response (renin release) to decreased renal perfusion pressure or decreased sodium concentration in the ultra filtrate, the latter an inappropriate release of aldosterone from the adrenal cortex, often a result of an adrenal adenoma.

Other genetic syndromes, such as Bartter’s or Gitelman’s, cause hypokalaemia with alkalosis but without hypertension.¹⁴

Features of low potassium include generalised weakness and lethargy, ascending paralysis, and rhabdomyolysis.^15-17   Decreased intake is rarely a cause of low potassium as the western diet usually contains significantly more potassium than is needed and because the renal tubular reabsorption mechanism can be extremely effective in limiting potassium excretion.¹⁷

The maximum recommended dose of liquorice is 100mg/day although cases of liquorice toxicity have been reported in association with doses as low as 80mg/day.  Each liquorice tea bag contains approximately 500mg of glycyrrhetic acid, of which approximately 20mg is ingested per infusion.

This is a relatively rare occurrence and it has been suggested that certain groups are more susceptible to toxicity than others – for example those with 11-β-hydroxysteroid dehydrogenase deficiency.¹⁸  It is also thought that those with essential hypertension are also more at risk.¹⁹

LEARNING POINTS/TAKE HOME MESSAGES

• Consumption of liquorice can cause pseudoaldosteronism.
• The clinical picture is similar to that of primary aldosteronism, but is characterised by low levels of both aldosterone and renin.
• While liquorice toxicity can be asymptomatic, clinical manifestations are wide ranging and include cardiac arrhythmias, rhabdomyolysis, weakness and paralysis.
• Pseudohyperaldosteronism caused by liquorice consumption is reversible and generally resolves upon cessation of liquorice consumption.  Prior to resolution, potassium supplements are usually necessary.

A 23 y/o male was brought to our Emergency Department after having a seizure. He was alert and his vital signs were stable. He is known to have epilepsy and is on regular anti-epileptic medication for three years. He is being followed up at a neighborhood medical center at his native village . On physical examination numerous brown papules were seen over his nose and both cheeks in a butterfly pattern which correspond to facial angiofibromas (Figure 1). Ash Leaf Hypomelanotic macules were seen over his extremities (Figure 2). Few hyperpigmented café au lait macules were observed over his trunk (Figure 3). A big fibroma was also seen over his scalp (Figure 4). Areas of thick leathery texture of orange peel known as Shagreen patches were observed on back (Figure 5).

Figure 1: Facial angiofibromas

Figure 2: Ash Leaf spot

Figure 3: Cafe au lait macule

Figure 4: Scalp fibroma

Figure 5: Shagreen patch

A Brain CT scan revealed multiple subependymal giant cell astrocytomas. Laboratory investigations were normal.

This patient was clinically diagnosed as Tuberous Sclerosis Complex having a myriad of skin lesions.¹

Tuberous sclerosis complex is an autosomal-dominant, neurocutaneous, multisystem disorder characterized by cellular hyperplasia and tissue dysplasia.² Seizures are commonly encountered in Emergency Room however, conspicuous lesions as described above must alert the physician to guide the patient for a multidisciplinary approach.³

Introduction

Warfarin is a widely used anticoagulant, primarily for the prevention of thrombosis and thromboembolism. Warfarin is used as a prophylactic agent in conditions such as atrial fibrillation and coronary artery thrombosis.¹

Although effective and safe, treatment with Warfarin is associated with risks. Because of its narrow therapeutic index, patients require regular blood monitoring for the international normalised ratio (INR) to determine the safe yet effective dose of Warfarin.

Warfarin has interaction with several medications which affect the availability of Warfarin. One class of drugs which is very commonly prescribed is the selective serotonin reuptake inhibitors (SSRIs) antidepressants.

Due to their supposedly favourable side-effect profile, e.g. less cardiotoxicity, and safety in overdose, SSRIs have become the first-line antidepressant  ², preferred over tricyclic antidepressants (TCA). However, SSRIs, have other serious side-effects including increased tendency to bleed, particularly gastrointestinal bleeding. SSRIs may increase the risk of bleeding due to the secondary effect of serotonin release which is essential for platelet aggregation. ³ This effect is especially significant when combined with anticoagulants. ¹

Several of the SSRIs are inhibitors of the cytochrome P 450 enzyme system, which is responsible for the metabolism of some medications, including Warfarin. Both SSRIs and anticoagulants are frequently prescribed in the elderly population.

Fluvoxamine is a SSRI which is licensed for use in the treatment of depressive disorder, obsessive-compulsive disorder (OCD) ¹, and is also used in the treatment of social phobia. While an interaction between Fluvoxamine and Warfarin is to be expected because of Fluvoxamine’s inhibitory action on a number of cytochrome P 450 enzymes, there have not been many case reports of the interaction between Fluvoxamine and Warfarin.

A Medline search revealed only two case reports of an interaction between Warfarin and Fluvoxamine. ^4,5

We report a case of an elderly man who developed elevated INR when he was started on Fluvoxamine for the treatment of depression, while on Warfarin.

Case Report

A 75-year old male was admitted to the acute psychiatric unit with complaints of anxiety, depressed mood and suicidal ideation. In the previous months, he had developed a pre-occupation that his bowels were not functioning properly and that he would not be able to open his bowels. He was using excessive laxatives secondary to this preoccupation. He also described other depressive symptoms: anhedonia, insomnia with early morning wakening, poor concentration, and low motivation.

The patient was diagnosed with depression two years previously, requiring Electroconvulsive therapy (ECT). He was discharged on Mirtazapine 45mg and Venlafaxine-XL 150mg. Following a deterioration in mental state, the Venlafaxine-XL dose was increased to 225mg three months before this admission, without much improvement. Risperidone and Olanzapine were trialled as an adjunct without beneficial effect and were discontinued. Compliance with medication was reportedly good.

The patient had multiple physical health complaints: previous myocardial infarctions, hypertension and paroxysmal atrial fibrillation. He was prescribed tamsulosin, bisoprolol, perindopril, atorvastatin, and warfarin.

The patient’s preoccupation about bowel movements was, for the most part, deemed to have obsessive quality: he accepted that these worries were repetitive and came to his mind against his wishes. He said that he would rather not have these worries but was unable to distract himself.

The marked subjective anxiety, according to the patient, was entirely linked to the preoccupation. However, when the patient became agitated, it was difficult to persuade him to appreciate the anomalous nature of his thoughts. At such times he insisted that there was something definitely wrong with his bowels and nothing could help him.

Prior to admission, the patient was treated at different times with different antipsychotic medications, which made little impact on the symptoms. Following his admission he was referred to the psychology services, which concluded that he was too unwell to meaningfully participate in psychological therapies. The patient, too, was not keen on this option. He also declined ECT and was deemed to have the capacity to make the decision. 

Venlafaxine-XL was switched, in light of the patient’s cardiac risk history, to an SSRI known to have an effect on obsessional symptoms. Accordingly, a dose of Mirtazapine was decreased to 30mg, and Venlafaxine-XL was tapered off over ten days.

Fluvoxamine was started at 50mg/day, and the dose was titrated to 150 mg/day over the next week. The dose was further increased to 200 mg/day.

INR was previously stable on warfarin dose of 5mg per day: blood results were between 2.32-2.68. Following fluvoxamine-initiation, INR started increasing:  2.98 after six days. INR increased further, to 3.82, with increase in Fluvoxamine’s dose

Warfarin-dose, as a result, was decreased, initially from 5mg to 4mg; however, INR remained above range (3.75). With further reduction in Warfarin’s dose, to 3mg/day, INR reduced but was still above range (3.51). INR eventually stabilised when warfarin-dose was reduced to 2mg/day. The dose adjustment took place over ten days.

Discussion

Management of depression in elderly population requires careful consideration of the choice of psychotropic medication, as elderly patients are more likely than younger patients to be on multiple medications for associated physical health problems, which increase the potential for drug interactions. Half-lives of drugs are also extended in the elderly.

The patient was prescribed Warfarin for the management of atrial fibrillation. Warfarin is a racemic mixture of S-warfarin and R-warfarin, of which S-warfarin is more potent than R-warfarin. ⁶

Warfarin has the potential to cause pharmacokinetic drug interactions (drugs affecting hepatic cytochrome P 450 enzyme system, which metabolises warfarin), which are thought to be more clinically relevant than pharmacodynamic interactions (highly protein bound drugs displacing Warfarin from its binding site) for warfarin. ⁶

Warfarin is metabolised by a number of P450 isoenzymes such as 2C9/2C19, 1A2, and 3A4. ⁶ Of these 2C9 is thought to be crucial, as it metabolises the more potent S-isomer. Isoenzyme 1A2 is the major route for the metabolism of the R-isomer, while 3A4 and 2C19 are considered to be minor routes.

The psychotropic medications that are thought to have the potential for significant pharmacokinetic interaction with warfarin include Fluoxetine, Fluvoxamine, Quetiapine, and Valproic Acid. ⁷ Venlafaxine is also considered a high-risk drug in patients taking warfarin. One study found that fluvoxamine and venlafaxine were associated with a more than double risk of having an INR value of 6 or more. ³

Fluvoxamine by dint of its inhibitory actions on 2C9/2C19, 3A4, and 1A2 ⁸ inhibits all the isoenzymes that metabolise Warfarin and can be said to have the maximum potential for pharmacokinetic interaction with Warfarin.

National Collaborating Centre for Mental Health guidelines on depression in adults with chronic physical health problems advise avoiding SSRI in patients taking warfarin or heparin and instead offering an alternative antidepressant such as mirtazapine. ⁹ Therefore, caution is needed when prescribing these medications to patients who fail to respond to other safer options.

At the time of his admission, the patient was on a combination on Mirtazapine and Venlafaxine XL, both at high doses, for several months. Neither of the antidepressants has significant action on the P450 isoenzyme system, although both are substrates of some enzymes, such as 2D6, 1A2 and 3A4. ⁸ The patient’s INR was within normal limits when he was on these two medications (despite the aforementioned potential effect of venlafaxine on INR values).

During the period of admission, a decision was taken to change the antidepressant regime, for clinical reasons. The two antidepressants considered were Fluvoxamine and Sertraline.

It was felt that despite its higher potential to cause pharmacokinetic interaction with Warfarin, Fluvoxamine is chosen ahead of Sertraline (which inhibits 3A4 and 2D6). The more potent action of Fluvoxamine on the sigma₁ receptors, which account for its significant anxiolytic properties and therapeutic action in the delusional depression ¹⁰ was felt to have the potential of benefit given the patient’s clinical symptoms.

 It was also felt that Fluvoxamine would also be of more benefit than Sertraline with insomnia which was patient’s frequent complaint given its side-effect of somnolence. ¹¹

Fluvoxamine was started at a low dose (50 mg/day) after Venlafaxine XL was completely withdrawn. The dose was titrated rapidly over the next one week, to 150 mg/day. The INR showed an upward trend within five days of commencing Fluvoxamine, and it exceeded three by the time the dose of Fluvoxamine was increased to 150 mg/day. This necessitated a reduction in Warfarin’s dose from 5 mg to 2 mg/day. The INR stabilised to between 2 and 3 when warfarin dose was more than halved.

Fluvoxamine has a half-life of 9-28 hours, and steady-state levels reach roughly after ten days. ⁸

The half-life is increased by almost 50% in the elderly. ³ The trajectory of increase in the INR was consistent with the pharmacokinetic of Fluvoxamine: it is a potent inhibitor of CYP4501A2, with a relatively little affinity for the other isoenzymes. The increase in INR was not dramatic, which is different from previous reports. ⁴ Throughout the period of Fluvoxamine –titration and Warfarin dose adjustment, there was no clinically untoward incident.

In conclusion, this case shows the need for close monitoring when warfarin is combined with another drug which significantly enhances warfarin’s anticoagulant effect. At the same time, it can be done safely, even in patients such as the elderly, who can be at a higher risk of adverse effects of interaction, when appropriate steps to monitor the patient are in place. The case also demonstrates the necessity of using clinical judgment while applying the guidance in individual patients.

Introduction

Metastatic bone disease is a relatively common event in the advanced stages of many malignancies.¹ Bone-modifying agents decrease the incidence of skeletal-related events (SREs) such as spinal cord compression and bone fracture, as well as the need for skeletal radiotherapy or surgery.²

Bone modifying agents such as intravenous bisphosphonates (IV BPs) (e.g. pamidronate and zoledronic acid) and denosumab are approved for prevention of SREs. IV BPs are primarily used and effective in the treatment and management of cancer related conditions such as multiple myeloma (MM), and breast cancer with skeletal metastases, because they reduce bone pain, hypercalcemia, and the risk of pathologic fractures.³

Denosumab, a receptor activator of nuclear factor kappa-B ligand (RANKL) inhibitor, represents a breakthrough in the treatment of osteoporosis, MM, and bone metastases. The Food and Drug Administration (FDA) approved it in 2010 for the prevention of SREs in patients with bone metastases and in 2011 for the prevention of endocrine-therapy induced bone loss in patients taking aromatase inhibitors for breast cancer and in patients with non-metastatic prostate cancer.

Three international, randomised, double-blind, double-dummy phase III studies have evaluated denosumab versus zoledronic acid for the treatment of SREs in breast and prostate cancers, and in combined solid tumours and MM. Denosumab’s superior efficacy over zoledronic acid was demonstrated in the studies of patients with advanced breast or prostate cancer, as well as in a pre-specified integrated analysis of all patients enrolled across the three studies.⁴

In the 2014 position paper of the American Association of Oral and Maxillofacial Surgeons (AAOMS), the nomenclature “bisphosphonate-related osteonecrosis of the jaw” changed to “medication related osteonecrosis of the jaw” (MRONJ). MRONJ is defined as cases in which all of the following 3 characteristics are present⁵:

• current or previous treatment with antiresorptive or antiangiogenic agents
• exposed bone or bone that can be probed through an intraoral or extra-oral fistula in the maxillofacial region that has persisted for longer than 8 weeks
• no history of radiation therapy to the jaws or obvious metastatic disease to the jaws

Other terminologies used previously include “denosumab related osteonecrosis of the jaw” (DRONJ), and “antiresorptive agent-induced ONJ” (ARONJ).

The aetiopathogenesis of MRONJ related to denosumab therapy remains enigmatic, and hypotheses have focused on reduced bony turnover, infection, toxicity of the soft tissue, and antiangiogenesis. The epidemiology also remains unclear, and reported incidence varies widely.⁶ Overall, it is estimated that bone necrosis can develop in about 0.7-1.9% of patients with malignancy who are given high-potency IV BPs (such as zoledronic acid), and in 0.01–0.1% of those with osteoporosis who take low-potency oral BPs (such as alendronate). Data relevant to denosumab given subcutaneously in patients with metastatic cancer and osteoporosis seem to replicate those when IV high-potency BPs are administered.⁷ The risk of osteonecrosis of the jaw (ONJ) is higher in patients exposed to concomitant antiagiogenic medication. The individuals’ risk of ONJ is further determined by factors such as the potency of agent, cumulative dosage or duration of antiresorptive treatment, route of administration, comorbidities and local factors such as periodontal disease.^8,9 Oral hygiene plays a significant role with evidence supporting a strong correlation between bacteria associated with periodontal disease and MRONJ.¹⁰

MRONJ typically manifests as painful and often infected areas of necrotic bone, which subsequently may lead to severe chronic pain and facial disfigurement. This adversely affects the ability to eat, speak and lowers the quality of life. Adverse events related to RANKL inhibitors are usually considered to be infrequent and low in occurrence. Unfortunately from our recent clinical experience at Sheffield Teaching Hospitals' Trust, there have been several new cases presented in a very short period of time. In this paper we present a case series of MRONJ related to denosumab therapy since adverse events of denosumab in the mandible or maxilla have received relatively little attention.

The aim of this article is to highlight the elevated risk of MRONJ in patients receiving denosumab treatment and educate all health care providers involved in the management of such patients. Furthermore, the mechanisms of denosumab, comparison with bisphosphonates and the reported management strategies are reviewed.

Mechanism of Denosumab

Denosumab is an antiresorptive agent that exists as a human IgG2 monoclonal antibody and inhibits the binding of the receptor activator of nuclear factor kappa-B ligand (RANKL) to RANK (Receptor Activator of Nuclear Factor kappa-B). The binding normally signals the proliferation of osteoclasts, as RANK is expressed on the surface of osteoclasts and their precursors, whereas its ligand, RANKL, is a membrane bound protein expressed by bone marrow stromal cells, osteoblasts and T-lymphocytes. The activation of RANK is integral to the function of osteoclasts. Osteoprotegerin binds to membrane bound RANKL on osteoblast which in turns decreases the osteoclastic activity and in theory negatively effects bone turnover. Denosumab acts similarly to osteoprotegerin but has a higher affinity for RANKL.^11-13

Denosumab follows nonlinear, dose-dependent pharmacokinetics. The bioavailability of one subcutaneous denosumab injection is 61% and serum concentrations are detected within 1 hour. Maximum serum concentrations occur in 5-21 days and cessation of osteoclast activity occurs within six hours of the subcutaneous injection. The normal function is restored approximately six to nine months later, whilst bone turnover returns to normal shortly after this.¹⁴ Based upon monoclonal antibody pharmacokinetics, denosumab is most likely cleared by the reticuloendothelial system with minimal renal filtration and excretion thus avoiding nephrotoxicity. Its elimination half-life is 32 days, and it does not incorporate into bone.¹⁵

It is currently marketed as Prolia® and Xgeva®, approved by FDA. Prolia® is administered subcutaneously every six months and has shown to reduce the incidence of new vertebral, non-vertebral, and hip fractures in osteoporotic patients.^16,17 Xgeva® is also effective in reducing SRE related to metastatic bone disease from solid tumours when administered intravenously on a monthly basis.^17,18

RANKL Inhibitors and BPs Pharmacokinetics

There are fundamental differences between denosumab and BPs with regard to their mode of action. Denosumab is an antibody and acts extracellularly whereas BPs act intracellularly. As such, BPs must be present in the circulation and available for reuptake into bone for prolonged periods to function.¹⁹ There is not any evidence of drug recycling with RANKL inhibitors, and therefore it is suggested that their adverse effects can be reversible with discontinuation, in fact leading to a transient rebound phenomenon, which can be restored, with subsequent treatment.^14,20 On the other hand, recycling of BPs in the circulation system has been proposed as a reason for the long duration of action even after cessation which can be up to 12 years.

The US FDA-approved manufacturer’s package insert for both zolendronate and pamidronate states that “there are no data available to suggest whether discontinuation of bisphosphonate treatment reduces the risk of ONJ in patients who require dental procedures during therapy and that clinical judgment of the treating physician should guide the management plan of each patient based on individual benefit/ risk assessment”. The package insert for denosumab does not address the issue of treatment continuation in patients who develop MRONJ to date.

Denosumab is a circulating protein capable of distributing throughout extravascular space. It is expected to reach all sites within bone including intracortical sites unlike with BPs. BPs have strong affinity for hydroxyapatite and bone mineral which limits their even distribution throughout the skeleton, particularly to sites deep within the bone.^19,21 This can explain the more profound inhibition of bone remodelling with denosumab than that seen with BPs.

Case Series

Case 1

A 55 year-old lady referred to a dedicated Oral Surgery nerve injury clinic for an opinion and management of her left sided inferior alveolar nerve (IAN) paraesthesia. The patient presented with a history of numbness in the left sided inferior alveolar nerve distribution following removal of the left mandibular second premolar (LL5) in July 2014. She was asymptomatic until she had the LL5 removed and since had suffered with constant pain and numbness. A year later, she had removal of the left mandibular first molar (LL6) and gave a history of recurrent infections and excruciating pain in her mandible over the past two months. On presentation she had an obvious submental swelling and left sided IAN anaesthesia.

Medically she was diagnosed with breast cancer in 2011, for which she underwent wide local excision followed by chemotherapy. She then was placed on unknown clinical trial that we identified at the time to be denosumab trial, following liaison with the Oncology team. She is currently receiving intravenous denosumab every three months.

Clinical examination revealed a grossly mobile anterior mandible with widespread bony necrosis and associated osteomyelitis. Sensory testing revealed complete anaesthesia in the left sided IAN distribution secondary to MRONJ.

An OPG (Orthopantogram) and CBCT (Cone-Beam Computerised Tomography) revealed an extensive patchy area of ill-defined bone loss in the anterior mandible extending posteriorly to the premolar/molar areas bilaterally (Fig 1).

Figure 1 A) OPG showing non-healing sockets in the left mandible with extensive bony destruction together with periosteal reaction extending to the right mandible as shown by the arrows.

Rather interestingly, the bony destruction was evident bilaterally with the patient only having had extraction of teeth in the left mandible (Fig 1). This could be the case of spontaneous ONJ in the right mandible or an extensive ONJ arising from simple extractions on the left side.

Figure 2 3D reconstruction of the CBCT image demonstrating extensive bony destruction involving the lower border of anterior mandible in keeping with a spreading chronic bony infection and clinical presentation of submental swelling as showing by arrows.

Case 2

A 66-year-old female referred by her general medical practitioner (GMP) with a 3-month history of delayed healing following a tooth extraction in the left posterior mandible. She had moderate to severe discomfort and reported multiple previous infections and purulent discharge from the area, which treated with multiple courses of antibiotics. In addition, she reported discomfort from the root treated right mandibular first and second premolar teeth (LR4 and LR5).

Medically she was diagnosed with breast cancer over 10 years ago for which she underwent resection followed by chemotherapy. Three years ago, she diagnosed with metastatic deposits and therefore has been receiving intravenous denosumab every six weeks since then. Other medications include steroids, chemotherapy agents, antihypertensives and analgesics. She did not receive any radiotherapy or BPs treatment in the past.

Clinical presentation revealed a heavily restored dentition with chronic generalised periodontal disease. There was evidence of widespread bone loss clinically and radiographically. The slow healing socket in the left mandible was visible but did not have any exposed bone (Fig 3). The lower right first and second premolar teeth (LR4 and LR5) were clinically and radiographically sound.

Figure 3. Non-healing socket in the left posterior mandible with no evidence of exposed bone or suppuration as showing by white arrow. Gingiva recession (black arrows) is evident in the LL6 and LL5 teeth in keeping with chronic periodontal disease.

Figure 4 Coronal sections of CBCT A and B showing multiple lytic areas within the inferior cortex of the mandible and incomplete healing of the extraction sockets.

On follow-up appointments, the patient suffered multiple repeated infections in the right and left posterior mandible and due to deteriorating periodontal disease, the LR4, LR5, LR6 were extracted by her own general dental practitioner (GDP) due to severe mobility. All three extraction sockets failed to heal (Fig 5) leading to an extensive area of exposed bone in the right mandible, extending from the lower right first premolar (LR4) to lower left first molar (LL6) region. Conservative management was embarked which included antibiotics, chlorhexidine mouthwash and routine oral hygiene appointments. Selective sharp bone trimming and three sequestrectomies were undertaken. At the same time, liaison with the patient’s oncologist resulted in cessation of the denosumab therapy and complete resolution of her oral symptoms.

Figure 5 Clinical picture of exposed necrotic bone (white arrows) following simple extractions of periodontally involved teeth.

Case 3

A 76-year-old lady referred to the Oral Surgery department by her GDP with a 3-month history of a non-healing lower left first premolar (LL4) socket. The patient was treated with two courses of antibiotics prior to referral which provided only temporary relief to her symptoms.

Medically she was diagnosed with breast cancer 10 years ago and recently commenced intravenous denosumab for metastatic disease. She also receives hormone therapy and palliative radiotherapy to the spine.

On clinical examination, there was a partially healed LL4 socket with a rather granulomatous appearance. There was no clinical evidence of suppuration or bony exposure. Radiographs confirmed the absence of bony infill in the socket. Local debridement and biopsy of the granulomatous tissue was performed to exclude any metastatic disease. Biopsy report confirmed the presence of inflammation tissue.

Figure 6 CBCT scan; A and B sagittal views, C axial view and D 3D reconstruction. Extensive periosteal reaction extending from the midline of the mandible to the left molar region is evident in keeping with chronic osteomyelitis secondary to MRONJ.

Liaison with the microbiologist suggested a long-term antibiotic course to arrest osteomyelitis. Further liaison with the oncology team, resulted in denosumab being stopped for 4 months. On subsequent review appointments, patient’s symptoms improved however, there is now an area of exposed bone in the LL4 region as shown in Fig 7.

Figure 7 Clinical photo illustrating exposed bone (white arrow) in the LL4 region without evidence of local infection.

Case 4

A 65-year-old lady referred to the Oral Surgery department by her GDP with a history of a sore upper mouth and jaw underneath the dentures which is unable to wear.

Medically she was diagnosed with disseminated breast malignancy including bone metastases 3 years ago and for that, she is on exemestane and IV Denosumab monthly.

Clinical examination revealed multiple draining sinuses in the anterior maxilla. There was a partially healed LL4 socket with a rather granulomatous appearance and tenderness on palpation. There was neither discharge from the area nor any exposed bone. Radiographs confirmed the absence of bony infill in the LL4 socket. Local debridement and biopsy of the granulomatous tissue was performed to exclude any potential malignancy and it was confirmed as inflammation tissue.

Figure 8 CBCT scan; A axial view, B and C 3D reconstruction. A 25mm fragment of right anterior maxilla is beginning to sequestrate. This extends from the anterior margin of the right maxillary sinus approximately to the position of the upper left lateral incisor, crossing the midline. The sequestrated fragment involves the lateral margin of the nasal cavity. There is bilateral moderate mucosal thickening in the maxillary sinuses. Extensive periosteal reaction extending from the midline of the mandible to the left molar region is evident in keeping with chronic osteomyelitis secondary to MRONJ.

Table 1 Summary of cases

Discussion

ONJ associated with antiresorptive therapy deserves distinction from other causes and diseases/medications associated with the development of osteonecrosis of the jaw. AAOMS recently published stage specific treatment recommendation for MORNJ.²² The various stages and suggested stage-specific treatment strategies are not evidence-based, and in particular, stage 0 disease is not universally accepted. AAOMS recommendations echoed those stated in previous years for BRONJ, namely supporting conservative therapy, with aggressive surgery offered only to symptomatic patients. In contrast, the MRONJ guideline report from the German Dental and the German Oral and Maxillofacial Associations refrains from recommending therapy at least for certain stages of the disease. This might be attributed to the pitfalls of current MRONJ criteria. Furthermore, due to poor guidelines specifically related to RANKL inhibitors, no agreement exists on a universally acceptable therapy strategy of such cases.

Management strategies are largely based on expert opinion rather than experimental data. It includes prevention, conservative and surgical modalities. Prevention of the condition is the gold standard. It is highly recommended all patients have a comprehensive dental examination and preventive dentistry (pre-emptive extraction of unsalvageable teeth and optimised periodontal health) before commencing antiresorptive therapy.^23,24 Oral hygiene should be kept meticulous during the course of therapy as periodontal disease and associated bacteria claim to be implicated in this condition and also observed in these cases.

The success rate of conservative treatment regimens range from less than 20% ^25,26 to above 50%^27,28 although some cases become chronic and develop complications.²⁹

Microbial cultures from areas of exposed bone are not always helpful since normal oral microbes are isolated. However, when there is extensive soft tissue involvement, microbial cultures may help to define comorbid oral infections, which may guide the selection of an appropriate antibiotic regimen.³⁰

Regardless of the stage of disease, areas of necrotic bone that are a source of chronic soft tissue irritation and loose bony sequestra should be removed or recontoured so that soft tissue healing can be optimised. This is in line with our clinical experience. The extraction of symptomatic teeth within exposed, necrotic bone should be considered as it appears unlikely that extraction will worsen the established necrotic process. Otherwise, surgical resection of necrotic bone should generally be reserved for refractory or advanced cases.³¹ Resection may occasionally result in even larger areas of exposed and painful infected bone.³²

A recently published MISSION study⁷ reported that the AAOMS system misclassified/ underestimated the severity of the disease at a rate of about 1 in 3, in particular in patients suffering from MRONJ stage 1 and 2. The authors conclude that these findings may explain why the treatment of stage 3 ONJ, namely surgery with success rate over 85%^33,34, has been deemed to be more predictable and therefore yields more favourable outcomes than the treatment of stages 1 and 2.³⁵

Denosumab is characterised by reversibility of its effect after treatment discontinuation, in contrast with bisphosphonates. This is in line with our findings since cessation of denosumab in two cases helped to improve their symptoms significantly.

MRONJ has been reported to occur after a mean administration period of 39.3 months and 35 infusions in oncology patients.²³ It is interesting that all published cases of denosumab-related ONJ occurred early after commencement of therapy, independent of the number of previous administrations.^36,37 In our experience, all patients developed MRONJ within the first 3 months of teeth extractions; well ahead of the reported period and number of administrations of denosumab.

Furthermore, all four cases have had extensive lytic lesions developed following removal of a single tooth. The common radiographic findings in all cases include:

• non-healing extraction socket
• areas of focal and diffuse sclerosis
• thickened lamina dura
• early sequestrum formation
• reactive periosteal bone
• osteolysis of cortical and spongious bone

These findings, although common in MRONJ cases, have had extensive bony involvement and rapid progression of ONJ, demonstrating a far more aggressive nature of the disease compared to that seen with BPs.

In our experience, not all patients are adequately informed of the risks and adverse events of denosumab therapy. This highlights the importance of educating patients and inter-professional communication regarding the prevention and best management of MRONJ cases. In one of the cases, the lack of patient education concerning denosumab side effects and the failure of inter-professional communication had a detrimental effect on the patient’s overall management and subsequently patient’s oral health.

Table 2 Important Points

Conclusion

We present our experience with denosumab-related ONJ from Sheffield Teaching Hospital’s NHS Trust. This case series should contribute to the existing sparse clinical literature on this topic. The pathogenesis, treatment and outcome of ONJ are complex and multifactorial. Patients treated with denosumab may be more prone to developing ONJ even without a precipitating dental event. ONJ may have a more aggressive profile and develop significantly earlier in patients receiving denosumab. Prevention of ONJ still remains the most important goal, and this is most directly accomplished by avoiding invasive dental procedures and establishing inter-professional communication.

Introduction

Tuberculosis is an important and major infectious disease worldwide, especially in developing countries with an estimated global case fatality rate of 13% in 2007. The World Health Organization estimated that there were 13.7 million prevalent cases of TB infection worldwide, with each year bringing about 9.27 million new cases, 44% of which are new smear-positive cases¹. Musculoskeletal tuberculosis is rare, chest wall tuberculosis is rarer and involvement of costochondral junction is among the rarest forms of tuberculosis. Tubercular costochondritis usually presents with insidious onset non-specific pain and swelling, resulting in delay in the diagnosis. Diagnosis is usually made by typical radiological findings and microbiological and histological evidence of tuberculosis. Treatment consists of antitubercular therapy with or without surgery.

Case report

30 year male, smoker, from low socio economic status presented with history of low grade fever, malaise and anorexia which began gradually two months back. For about one month he had pain in right side of chest just adjacent to lower part of sternum. The pain had started insidiously, gradually worsened with time, and was dull and aching in character. Pain was localized to the right lower parasternal area, occasionally radiating to the back. The pain was aggravated by physical activity and deep inspiration and was relieved to some extent by ordinary anti- inflammatory medications. There was significant history of pulmonary tuberculosis in the patient’s sister 1 year ago. There was no history of cough, haemoptysis, fever with chills or history of tuberculosis in the past.

On general physical examination, the patient was weak and febrile. On local examination, there was a bulge in the right lower parasternal area, corresponding to the right 9th costochondral junction. On palpation, there was severe tenderness in the same area. There was no peripheral lymphadenopathy and abdomen was soft and non-tender with no organomegaly. Chest examination was unremarkable. The rest of the examination was normal.

In terms of investigations, the chest radiograph, ECG, haemogram, kidney function tests and liver function tests were normal. ESR was high (34), Mantoux test was positive(15mm). Sputum for AFB was negative. Axial CT chest demonstrates expansion of the left 9th costal cartilage with soft tissue thickening on both the inner and outer aspect of the cartilage (Fig.1). FNAC (Fine Needle Aspiration Cytology) of the costochondral junction revealed Mycobacterium tuberculosis and on culture and histopathology of aspirated material revealed tubercular granuloma (Fig.2). A final diagnosis of tubercular costochondritis was made and the patient was treated with anti tubercular drugs for 9 months. Patient's symptoms improved after 2 weeks of treatment and swelling and tenderness subsided. Post treatment axial CT demonstrated complete resolution of soft tissue abnormality previously seen around the costal cartilage (Fig.3).

Figure 1: CT chest showing evidence of costochondritis.

Figure 2: Typical tubercular granuloma with central caseous necrosis on histopathology.

Figure 3: Post ATT CT showing complete resolution of costochondritis.

Discussion

Tuberculosis is very common infectious disease in developing countries like India, resulting in significant morbidity and mortality. Musculoskeletal tuberculosis is relatively uncommon and accounts for 1 to 2% of all the tuberculosis patients^2,3 and accounts for about 10% of all extrapulmonary TB infections⁴. Tuberculosis of the chest wall constitutes 1 to 5% of all cases of musculoskeletal TB^5,6. TB abscesses of the chest wall are usually seen at the margins of the sternum and along the rib shafts, and can also involve the costochondral junctions, costovertebral joints and the vertebrae⁷. In one study⁸, on the basis of the part of the rib involved, the roentgenographic findings for patients with rib tuberculosis was divided into four categories: Costovertebral (35%), Costochondral (13%),Shaft (61%), and Multiple cystic bone. TB of the thoracic wall is usually caused by reactivation of some latent foci of primary tuberculosis formed during hematogenous or lymphatic dissemination but direct extension from contiguous mediastinal lymph nodes can also occur⁹. In developing countries such as ours, tuberculosis is endemic and all the rare forms of the disease have been described, but in developed countries, resurgence of tuberculosis due to HIV may be responsible for atypical presentations .

Thoracic wall tuberculosis mostly presents insidiously with pain and swelling, but the diagnosis of chest wall TB is often delayed due to lack of specific symptoms and signs and gradual course¹⁰. Approximately less than 50% of chest wall TB patients may have active pulmonary TB^11,12. Imaging techniques like X-Rays and CT scan play an important role in diagnosis and follow up of these patients. According to a study done by Vijay YB et al¹³, radiological signs may not be present initially at the time of presentation with symptoms, abscesses or sinuses may be present much before the imaging modalities detect them. Computed tomography (CT) scan is more valuable for localization and detection of bone destruction and soft tissue abnormalities. Atasoy et al demonstrated the role of Magnetic Resonance Imaging (MRI) for early detection of marrow and soft tissue involvement in sternal tuberculosis due to high contrast resolution of MRI¹⁴.Diagnosis is usually confirmed by finding acid fast bacilli (AFB) and positive AFB cultures of bone (positive in up to 75% of cases), and caseous necrosis and granuloma on histopathology¹¹.

Complications of thoracic wall tuberculosis include secondary infection, fistula formation, spontaneous fractures of the sternum, compression or erosion of the large blood vessels, compression of the trachea and migration of tuberculosis abscess into the mediastinum, pleural cavity or subcutaneous tissues as discharging sinus. Chest wall TB needs to be differentiated from benign and malignant tumors [chondroma, osteochondroma, fibrous dysplasia, lipoid granuloma, chondrosarcoma, myeloma multiplex]¹¹, metastatic carcinoma, lymphoma or other kinds of infection^15,16.

The treatment of choice of chest wall TB is still not clear. Whether antitubercular therapy alone or surgical debridement (or excision based on lesion extension) combined with antitubercular therapy should be done needs further studies. But the general rule is if there are any complications, surgery is the treatment of choice followed by antitubercular therapy. We conclude that the diagnosis of thoracic wall tuberculosis is a challenge for physicians and is suspected by gradual onset clinical features and confirmed by microbiology, histopathology and radiography findings. Early diagnosis and treatment are important to prevent complications caused by bone and joint destruction.

Introduction

Meningiomas are common intracranial neoplasms with a wide range of histopathological appearances. The WHO classification of tumours of the central nervous system recognises 15 subtypes of meningiomas, of which meningothelial, fibrous and transitional subtypes are most common. Lymphoplasmacyte-rich meningiomas (LPM) are rare WHO subtype that belong to Grade I meningiomas.¹ The estimated incidence is less than 1% of all meningiomas.² LPM usually occurs in young and middle age patients, with most common locations being cerebral con­vexities, skull base, parasagittal area within the superior sagittal sinus, cervical canal, optic nerve and tentorium.³ Histopathological examination shows extensive infiltrates of lymphocytes and plasma cells often obscuring the meningothelial component.

Case report

A 21-year-old man presented with a history of headache since 4 months. It was a dull pain not associated with vomiting, seizures or visual symptoms. The patient did not have any features suggestive of cranial nerve involvement. Physical examination was unremarkable except for the presence of papilloedema. Non-contrast CT scan showed a large isodense lesion with peri- lesional oedema and eccentric enhancing nodular component in the right fronto-parietal region (Figure 1). A radiological diagnosis of glioma with mass effect and shift to left was rendered. A right frontoparietal free bone flap craniotomy was performed. Operatively, a well encapsulated tumour probably arising from the dura mater was found. Gross total removal of the tumour was done and the excised tumour was sent for histopathological examination with a provisional clinical diagnosis of meningioma.

Histopathological examination revealed a tumour arranged as sheets and whorls of meningothelial cells without any mitoses or atypia. A dense infiltrate of lymphocytes and plasma cells was seen in large areas of the tumour (Figure 2).

On immunohistochemistry, tumour cells were positive for epithelial membrane antigen (EMA) (Figure 3) and vimentin. The lymphoplasmacytic infiltrate contained mixture of CD3 and CD20 positive lymphocytes. A diagnosis of lymphoplasmacyte- rich meningioma was given.

Figure 1. Non-contrast CT scan showing a large isodense cystic lesion with perilesional oedema and eccentric enhancing nodular component in the right frontoparietal region

Figure 2: Tumour arranged as sheets and whorls of meningothelial cells without any mitoses or atypia. A dense infiltrate of lymphocytes and plasma cells seen in large areas of the tumour (H & E x 100)

Figure 3: Tumour cells positive for epithelial membrane antigen (x 200)

Discussion

Meningiomas are common neoplasms accounting for 24-30% of all primary intracranial tumours. They arise from the arachnoidal cells, and are typically attached to the inner surface of the duramater.¹ Most of the meningiomas are benign, corresponding to WHO grade I and associated with a favourable clinical outcome. LPM is a rare low grade histopathological subtype of meningioma, usually seen in younger patients, with the mean age of onset being 34 years.^4,5 The patients with LPM have variable clinical manifesta­tions according to the location of the tumour. The common presentations include headache, hemiparesis, seizure, vomit­ing, dizziness, visual disturbance, dyscalculia, dysgraphia and slurred speech.³ Although the natural history of LPM is often over one year, few cases might occur in short duration due to inflammatory cell infiltration and oedema.⁶ Systemic haematological abnormalities such as hyperglobulinemia and iron refractory anaemia have been documented in some patients with LPM, believed by some to be due to the plasma cell infiltrate.^3,6,7

Radiologically, LPMs are usually globular, highly vascular, contrast- enhancing, and dural based tumours. The typical characters of LPM on MRI are isointense lesions on T1-weighted images and hyperintense lesions on T2-weighted images, with a strong homogenous enhancement after the administration of gadolinium; obvious peritumoural brain oedema and dural tail signs.³ Sometimes, cystic component and heterogeneous enhance­ment may also be encountered, making pre­-operative diagnosis difficult, as in our case.⁸

On microscopic examination, this tumour is characterised by a conspicuous infiltrate of lymphocytes and plasma cells, sometimes completely obscuring the tumour cells. The massive infiltration of lymphocytes and plasma cells has been postulated to play a central role in the development of brain oedema associated with LPM. The origin of this tumour (neoplastic or inflammatory) is unclear, so it is considered closer to intracranial inflammatory masses rather than typical meningiomas.⁷

The differential diagnoses include collision tumour of meningioma and plasmacytoma, inflammatory pseu­dotumour, idiopathic hypertrophic pachymeningitis (IHP), and intracranial plasma cell granuloma.^3,7 The use of staining for EMA and vimentin is useful in indicating the meningothe­lial origin of the tumour, and differentiates LPM from other intracranial lesions.⁹

The pathological findings of IHP usually include thickened fibrotic dura mater with marked infiltration of lymphocytes and plasma cells, occasionally accompanied with small islands of meningothe­lial proliferation mimicking those of LPM. Localised nodular lesion can some­times rule out this diagnosis in that IHP usually shows diffused lamellar thickenings or plaque-like features.⁴

Chordoid meningiomas often contain regions that are histologically similar to chordoma, with cords or trabeculae of eosinophilic, vacuolated cells in a background of abundant mucoid matrix background.³ Detailed histological studies can aid the dif­ferential diagnosis. The plasma cell component is not neoplastic and thus plasmacytoma with reactive meningothelial hyperplasia or a collision tumour involving meningioma and plasmacytoma can both be excluded.¹⁰

The knowledge of this rare entity is important to avoid its underdiagnosis as an inflammatory pseudotumour or plasma cell granuloma and overdiagnosis as a plasmacytoma.

CASE REPORT

A 61-year-old lady underwent a modified radical mastectomy and axillary clearance in 2008 for a carcinoma of the left breast. Histopathology examination revealed two tumours within the left breast; a 16mm Grade 2 lobular carcinoma with probable vascular invasion and a 9mm Grade 1 infiltrating ductal carcinoma with no vascular invasion. She had clear resection margins. 21 out of 34 removed lymph nodes were positive for metastatic deposits. The tumour was oestrogen receptor positive and HER2 negative. She was staged as T1 N3a Mx and the tumour had a Nottingham Prognostic Index of 5.32. Metastatic workup revealed no distant metastasis.

Postoperatively, she required aspiration of a seroma but her recovery was otherwise satisfactory. She received adjuvant chemotherapy in the form of three cycles of Fluorouracil, Epirubicin and Cyclophosphamide and 3 cycles of Docetaxel. In addition, she had postoperative radiotherapy to the chest wall and supraclavicular fossa (40 Gy in 15 Fractions over 3 weeks) and hormonal therapy with Letrozole 2.5mg once daily.

The patient opted to undergo a prophylactic right mastectomy in 2010. She was regular in follow up and appeared to be free of disease recurrence for 6 years.

Her past surgical history included abdominal hysterectomy and bilateral salpingo-ophorectomy for fibroid disease as well as varicose vein stripping. She is a non-smoker and doesn’t consume alcohol. She had a family history of colon and cervical cancer in her uncle and sister respectively.

The patient visited the surgical outpatient clinic complaining of abdominal cramps, altered bowel habits and fatigue of a few months duration. There was no associated rectal bleeding, haematemesis, melaena, weight loss or urinary symptoms. Physical examination was unremarkable but she was noted to have gradually worsening renal function. Her symptoms were at first attributed to side effects of intravenous antibiotic treatment, which she received during an admission for cellulitis. She had already undergone an upper GI endoscopy which showed oesophagitis and ulceration; biopsies were within normal limits. She received treatment with proton pump inhibitors but her symptoms persisted.

A non-contrast abdominal CT scan was done, on account of her poor renal function, which showed bilateral hydronephrosis and thickening of the postero-superior aspect of the bladder wall. Considering the limitations of the non-contrast study, there were no other abnormalities. A colonoscopy was also done to investigate her altered bowel habit and it revealed a benign-looking stricture in the sigmoid about 25cm from the anal verge which was easily bypassed by the scope.

Figure 1. Benign stricture on flexible sigmoidoscopy

Biopsies of the sigmoid stricture showed an infiltrate of small to medium sized tumour cells in the submucosa, which had an Indian file pattern. They were positive for AE1/AE3 (pancytokeratins) and negative for CD68. They were positive for CK7 and negative for CK20, strongly positive for oestrogen receptors and HER2 negative. Taken in conjunction with the patient’s past history of an invasive lobular carcinoma of the breast, the appearance was consistent with a metastatic lobular carcinoma.

Figure 2. Clusters and cords of cells with positive cytoplasm for the cytokeratin immunostain CK7. Although the classical ‘Indian filing’ of lobular carcinoma is not well seen, the image clearly demonstrates that the large bowel glands are negative (normally CK20+, CK7-) and that the infiltrate is beneath the glandular mucosa (i.e. not originating from dysplastic glands within the mucosa and raising the possibility of infiltration from outside the bowel wall). The magnification is x200. Lobular carcinoma is usually CK7 +, CK20 -, ER +.

Figure 3. The same cells with their nuclei staining positively with an immunostain to oestrogen receptors. There are a few short chains of ‘Indian filing’ with the cells appearing rather rectangular in shape with straight margins. You can make out slight ‘moulding’ of the nuclei as they press against one another. The magnification is x 400.

Figure 4. Haematoxylin and Eosin section at 400 magnification. This shows a diffuse infiltrate of single cells with eccentric nuclei.

The patient required a right nephrostomy and a cystoscopy with left double J ureteric stent insertion to address her hydronephrosis and deteriorating renal function before undergoing restaging of her disease.

DISCUSSION

In patients with history of breast cancer, isolated GI metastases are less common than benign disease processes or second primaries of the GI tract.^1.2 In a retrospective review, 73 out of 12001 cases of breast cancer had gastrointestinal metastases, out of which 24 were to the colorectum³ and invasive lobular carcinoma was the commonest histological subtype. ^3.4 However, sixteen percent of patients with breast cancer have GI metastases at postmortem examination¹.

There might be a long interval of time between the diagnosis of breast cancer and development of gastrointestinal metastasis which together with their rare occurrence and nonspecific clinical and radiological manifestations adds to the diagnostic challenge. The median interval between the diagnosis and the development of GI metastasis was reported to be 6 years (range 0.25 to 12.5 years) by Schwarz et al ⁵with 25 years being the longest reported in the literature.⁶ Because of this long interval the history of a primary breast cancer can be missed. This also highlights the importance of long term follow up and maintaining an index of suspicion when these patients develop GI symptoms.

In our case, the interval between the diagnosis of breast cancer and colonic metastasis was 81 months. Her GI symptoms were initially attributed to side effects of antibiotic treatment for cellulitis and dyspepsia before investigating her with a colonoscopy. Even at colonoscopy the appearance was that of a smooth benign-looking stricture which did not seem to harbour any sinister pathology

Histological examination is probably the most reliable tool to make a diagnosis and it is prudent in such cases to compare the specimen with the original breast tumour. In this case, there were two primary tumours; an invasive ductal carcinoma as well as a lobular carcinoma but the metastatic disease favoured the lobular component, which is consistent with other published reports in the literature. The reasons why metastases favour lobular carcinoma are poorly understood. One explanation is the loss of E-cadherin expression, a molecule involved in cellular adhesion, in invasive lobular carcinoma⁷. A similar case in which the primary was a mixed ductal and lobular type with lobular subtype colonic metastasis was reported by Uygun et al.⁸ Immunohistochemistry can also help in establishing a diagnosis. Metastatic breast cancers tend to be positive for Oestrogen or Progesterone receptors as well as Gross Cystic Disease Fluid Protein-15.^{9, 10} It is, however, worth noting that primary colonic cancers can be oestrogen receptor positive in 30 to 70% of cases.¹¹

Accurate histopathological diagnosis probably saved our patient an unnecessary surgical treatment for a primary colonic neoplasm as the main focus of her treatment should be systemic therapy for metastatic breast cancer.

CONCLUSION

GI tract metastases from breast cancer are a rare occurrence. The patients may present after a long interval from the original diagnosis and the clinical and radiological features are nonspecific with the diagnosis often established on histological examination. Moreover, the history of breast cancer may not be elicited in all cases and these patients may present to a gastroenterologist or colorectal surgeon rather than a breast surgeon or oncologist. Therefore, remaining vigilant to this possibility is advised in any patient with a history of breast cancer who presents with unexplained GI symptoms.

Introduction

Multiple Sclerosis (MS) is an inflammatory, demyelinating disease of the central nervous system (CNS) that affects approximately 400,000 people in the United States and 2.5 million worldwide.¹ There are rare variants of this disease that can profoundly delay diagnosis and treatment. Examples of such variants include: Tumefactive MS, Acute Disseminated Encephalomyelitis, Neuromyelitis Optica, Marburg’s MS and Balo Concentric Sclerosis.² These variants have a uniquely aggressive presentation and do not exhibit classic MS features.² Classic MS features include relapsing and remitting sensory and motor impairments, optic neuritis and pain. These aggressive variants are more likely to present with symptoms similar to neoplasm such as motor impairments and seizures. When dealing with these aggressive MS variants diagnostic options include Magnetic Resonance Imaging (MRI), Single Photon Emission Computed Tomography (SPECT) scan, MR spectroscopy and Cerebrospinal Fluid (CSF) analysis.³

Invasive tests such as brain biopsy are not warranted unless absolutely necessary. In MS, a biopsy must not be completed in order to confirm a diagnosis. However, to confirm a diagnosis of cancer a biopsy is required.

We present a rare case of Tumefactive MS that exhibited a clinical picture identical to brain metastasis. This was diagnosed with surveillance MRI and CSF analysis in the absence of a brain biopsy.

Case presentation

A 48-year-old African American female was brought in by emergency medical services after falling with a brief loss of consciousness. Associated symptoms included dull chest pain, diaphoresis and shortness of breath. While in the emergency department she also developed nausea, vomiting and dizziness. The patient reported no similar previous episodes and denied precipitating events. There was nothing else to note on review of systems. The past medical history included hypertension with no previous surgeries and family history included breast cancer of the mother diagnosed at age 47. The patient denied tobacco, alcohol and intravenous drug use. She noted an allergy to iodine.

On physical examination the patient was afebrile, normotensive and tachycardic with an oxygen saturation of 89% on room air. She was alert and oriented but pale and diaphoretic with mild left sided chest pain. Cardiac examination revealed a normal rhythm tachycardia and no murmurs were heard. Her neurological examination showed a normal mental status, normal cognition/comprehension and that Cranial Nerves II-XII were intact.

Laboratory findings included haemoglobin of 9.8 g/dL, 30.8% haematocrit and potassium of 3.3 mmol/L. Electrolytes were otherwise normal. Cardiac workup showed a normal ECG and slightly elevated cardiac enzymes of 0.319 ng/mL.

Given the patients tachycardia and desaturation, a stat Ventilation-perfusion (V/Q) scan was completed (patient had an iodine allergy). The V/Q scan revealed a perfusion defect suggesting pulmonary embolism (PE) as the cause of symptoms. Subsequently the patient was placed on appropriate anticoagulation.

Head CT (computed tomography) showed a left centrum semiovale round hypodense lesion measuring 1.4 cm, a left basal ganglia round hypodense lesion measuring 1.0 cm and a left occipital lobe round hypodense lesion measuring approximately 1.0 cm (Figure 1). No midline shift was seen. MRI showed multiple hypointense T1/hyperintense T2 nonenhancing lesions, mainly within the left cerebrum (Figure 2 A-F). The three largest lesions within the left posterior centrum semiovale (2A), left globus pallidus (2B) and left posterior corona radiata adjacent to the occipital horn (2C) measured 1.5 cm, 1.0 cm and 1.0 cm respectively. Perilesional vasogenic oedema was seen in all except the basal ganglia lesion. There were bilateral cerebral scattered foci of hyperintense FLAIR/T2 signals (D-F). The imaging suggested a differential diagnosis which included metastasis, infection or primary CNS malignancy.

Further work up in search for possible malignancy was completed. Skin map revealed no concerning nevi. Mammogram showed no tumor. CT of the abdomen and pelvis revealed a 2.6 cm indeterminate hypodense lesion in the left lobe of the liver (Figure 3A) along with an enlarged fibroid uterus (17x 7 x 14 cm). Liver biopsy was considered but a repeat MRI and ultrasound showed the lesion to be cystic, so this was deferred following surgical oncology recommendations (Figure 3B). For the hypertrophic uterus found on imaging, gynecology felt no further workup was necessary as they attributed the findings to a fibroid uterus.

Tumor markers CA 27-29, CA 19-9, CA 125 and AFP were all sent and came back negative. Initial lumbar puncture with CSF analysis was not completed secondary to possible complications that could be incurred while on necessary PE anticoagulation.

Due to a non-focal neurological examination, she was discharged on Levetiracetam 500 mg for seizure prophylaxis and Dexamethasone 4 mg for perilesional oedema. Over subsequent months the patient did well without headaches, vision changes or seizure like activity. On subsequent visits to the clinic, she had no evidence of focal neurological deficits except for mild bilateral symmetric hyperreflexia. Given that the metastatic work up remained negative, we considered obtaining a baseline Positron emission tomography (PET) scan to ensure we were not missing any possible metastasis.

She subsequently went back to work full-time and reported no symptoms. Repeat MRI of the head (Figure 4 A-C) showed predominantly T1 hypointense and T2 hyperintense (A-B) lesions with significant decrease in size from MRI done three months ago. These lesions demonstrated no enhancement to incomplete ring enhancement, with diminished vasogenic oedema (A). These findings suggested an inflammatory demyelinating process so a lumbar puncture was obtained after anticoagulation was held. CSF analysis was done using Isoelectric Focusing (IEF) and immunoblotting methodology. This revealed a normal myelin basic protein but with eight oligoclonal bands restricted to the CSF. These findings solidified the suspicion of Tumefactive MS.

Figure 1. Head CT without contrast: left centrum semiovale round hypodense lesions measures 1.4 cm with perilesional vasogenic edema

Figure 2. MRI of brain showing axial T1-weighted (A-C) hypodense lesions of the left centrum semiovale(A), left basal ganglia(B) and left occipital lobe(C). Axial T2-weighted (D-F) views show multiple hyperdense lesions corresponding to the same locations. Perilesional vasogenic edema is seen.

Figure 3A. Thorax CT without contrast. 2.6 cm left lobe liver lesion.

Figure 3B. MRI of abdomen showing coronal T2-weighted half-Fourier acquisition single-shot turbo spin-echo (HASTE) hyperdense lesion. A mildly enlarged liver measuring 18.7 cm in craniocaudal span. Simple 2.8 x2.4 cm cyct in the medial segment of left lobe.

Figure 4. MRI of brain (3 month after initial scans) showing axial T-2 weighted (A-B) hyperdense lesions of the left centrum semiovale(A) and left basal ganglia(B). There is irregular peripheral enhancement. Considerable decrease in size is seen from previous MRI (Figure 2). Left posterior centrum semiovale, left globus pallidus and left occipital lobe lesion measure 1.3 cm, <1 cm and <1 cm respectively. Vasogenic edema is diminished in comparison to previous study.

Discussion

Tumefactive MS lesions are defined as solitary demyelinating plaques greater than 2 cm.⁵ Lesions are difficult to distinguish between primary or metastatic given similarity of imaging features.⁵ Imaging features suggestive of Tumefactive MS include incomplete ring enhancement, absence of mass effect and absence of cortical involvement.^{6 7} Kim describes that CT hypoattenuation of magnetic resonance enhancing lesions was found to be highly specific for distinguishing Tumefactive MS lesions from CNS cancer pathology.⁶ It has been shown that SPECT using I-IMP is useful for diagnosing CNS malignancy.³ This is because there would be increased uptake in comparison to the MS lesions - implying increased metabolic activity.³ However this study has its limitations in diagnosis. In a few isolated cases I-IMP was found in greater quantities in MS tumor-like lesions.³

The imaging studies for this patient established a concern for metastasis, infection or primary malignancy. Extensive cancer workup was completed as previously discussed. Since all tumor markers were negative a baseline PET scan was considered however, was not done secondary to insurance denial. Due to the asymptomatic presentation of her disease, a primary differential diagnosis of brain metastasis and anticoagulation therapy for PE, a CSF analysis was not considered until much later. We were able to use surveillance MRI and CSF analysis to see some resolution of these lesions and confirm the diagnosis. Brain biopsy was never warranted but in unique symptomatic cases it may have been.⁶

The cornerstone of diagnosing MS is the demonstration of lesions in both time and space - termed the McDonald Criteria.⁸ The revised criteria allow a diagnosis of MS, “possible MS” or “not MS”.⁸ This is what made the diagnosis of our patient difficult, as no clinical symptoms or attacks were evident. It was demonstrated that over the course of three months the lesions seen on MRI evolved. From the size of 1.5 cm, 1.0 cm and 1.0 cm they became 1.3 cm, <1.0 cm and <1.0 cm respectively (Figure 2, Figure 4). This was likely the effects of steroids that the patient was on due to her vasogenic oedema. Here an evolution in time and space is demonstrated which excluded brain metastasis and infection. This brings into discussion the diagnostic value of surveillance MRI, which in our case was helpful and appropriate as the patient did not have clinical symptoms.

Conclusion

The diagnosis of Tumefactive MS can be extremely difficult and time consuming. As seen in our case, it can mimic other conditions. Our patient was able to be diagnosed with MRI surveillance and CSF analysis. The definitive diagnostic test for MS is a brain biopsy but this is not preferred due to the invasiveness of the procedure. With the advent of newer diagnostic tests such as SPECT, MR Spectroscopy, surveillance MRI and CSF analysis, diagnosis can be attained and treated presumptively.

CASE

A 79 year old lady presented to the accident and emergency department with severe abdominal pain. On admission she was hypotensive and hypothermic. Blood tests demonstrated raised inflammatory markers and white count, but were otherwise unremarkable. A CT scan revealed no abnormalities. She was treated with intravenous fluids and empirical antibiotics.

She had multiple co-morbidities, including ischaemic heart disease, hypertension and chronic kidney disease.

Figure 1 – Upper Oesophagus

Figure 2 – Distal Oesophagus

Figure 3 - Gastro-eosophageal Junction

Figure 4 – Stomach (in retroflexion)

Figure 5 - Duodenum

Three days into her admission she had a single episode of hematemesis and a gastroscopy was arranged. Endoscopic features were as per figures 1- 5. Histology taken at the time showed necrotic tissue with evidence of candidiasis. Her treatment was optimised with a two-week course of fluconazole with the dose adjusted for her renal function and parenteral nutrition, with good clinical response. She was discharged after a two week hospital admission. A repeat gastroscopy 10 weeks later showed complete resolution of endoscopic features with no evidence of perforation or stricture formation.

DISCUSSION

The images seen at endoscopy demonstrate a region of oesophageal ulceration progressing to a diffuse, circumferential, black discoloration of the distal esophageal mucosa, with an abrupt transition to normal mucosa at the gastro-esophageal junction (Figs. 1-3). These endoscopic features, in the absence of a history of ingestion of caustic substances, are diagnostic of Acute Oesophageal Necrosis (AON), also known as ‘Black Oesophagus’. Whilst histology confirming necrosis is not necessary to make the diagnosis, it is confirmatory.

AON was first described in 1990 by Goldberg et al, since which over one hundred cases have been reported in the literature¹. Population studies have suggested the incidence of this condition to be between 0.08% and 0.2%, although interestingly one post-mortem series of 1000 patients failed to reveal any cases^2-4. There is a male preponderance, with an incidence four times greater than that for women and a peak incidence during the sixth decade of life^{5, 6.}

The aetiology of this condition is not entirely clear; however case reports to date suggest that this is almost exclusively observed in those who are systemically unwell, usually in the context of multi-organ dysfunction^5-7. It has been postulated that necrosis most commonly occurs as a consequence of hypo-perfusion caused by a low flow state in those with underlying vascular disease. This is likely to account for the predilection for the distal third of the esophagus, which is relatively less vascular⁵. Individual cases have occurred in association with bacterial, viral and fungal infections, whilst malnutrition, malignancy and immune-compromise appear to be important factors^{3, 5, 6}.

The most common indication for the gastroscopy that makes the diagnosis of AON is hematemesis and melena, accounting for over 75% of cases⁶. It is therefore likely that AON is significantly under reported as endoscopy is often precluded in those who are clinically unstable. Further it is not clear whether hematemesis is a universal symptom of this condition; it is conceivable that AON may go undiagnosed in those in whom this is not a feature.

Whilst AON has no specific treatment, its presence is indicative of significant systemic compromise and predicts a poor prognosis. This diagnosis should alert physicians that close monitoring and aggressive treatment is required to optimise patient outcomes. There is no clear role for the use of anti-acid therapy, however this is commonly used in management due to patient symptoms, which usually includes hematemesis. Similarly, candidiasis may occur in conjunction with AON, whilst it is not thought to be causative, treatment is considered prudent given the poor prognosis associated with this condition.

For those that recover from their acute systemic insult the prognosis appears to be good. The long-term sequale of this condition includes oesophageal stenosis due to structuring. Evaluation with a repeat gastroscopy if therefore indicated if dysphagia develops.

CONCLUSION

The clinical course of AEN is variable, with an associated mortality of 32%⁵. The severity of the underlying clinical condition appears to be the most important factor in determining prognosis. There is no specific treatment for AON. The current body of experience suggests aggressive management of abnormal physiology optimises outcomes^{5, 6}. Antibiotics, antifungals and nutritional support should be considered on an individual basis.

Background

SSRIs (Selective Serotonin Uptake Inhibitors) are very commonly used in Depression and Anxiety. Though considered as safest antidepressants, they have some common side effects which include gastrointestinal side effects, headache and at times sexual dysfunction. Yawning is one of the rare side effects of SSRIs. SSRIs were found to be the commonest cause of not so common drug induced yawning in a meta-analysis^1. Isolated cases of intractable yawning have been reported with citalopram² fluoxetine, citalopram and sertraline³ in the literature .Excessive yawning can cause injury to Temporo-Mandibular Joint (TMJ) ⁴. Paroxetine has also been shown to cause intractable yawning⁵. Yawning possibly helps in thermoregulation and is an unconscious effort by the body to cool the brain ^{6, 7}. It is known that yawning can be contagious. Reading, talking, seeing someone yawn or even thinking about yawning can induce yawning in the subjects⁸. Susceptibility to contagious yawning is different for different individuals depending upon their ability to process information about self⁹.

Case

A 60 year old postman presented with his first episode of depression. He attended the GP who started him on sertraline (an SSRI). He developed serious headaches and did not notice any therapeutic benefit. He was then referred to the psychiatric services for further management. He was assessed, Sertraline was stopped and Cipramil 20mg was introduced. He was reviewed after 2 months and the dose was increased to 40 mg to which he responded partially but relapsed within 4 months. There were no changes in his psycho- social circumstances. Cipramil was stopped and he was started on fluoxetine 20 mg. Once again the response was partial and was overshadowed by midnight insomnia and increased sleepiness in the daytime. Fluoxetine was increased to 40 mg and he was reviewed after 4 months when he reported clear and significant improvement in his depression but complained of “excessive yawning spells” causing him problems at his work place. The psychiatrist was surprised at the number of times he yawned at the Out Patient Clinic review. On further discussion it became clear that this side effect had become highly troublesome. He complained that his jaw was in severe pain. He was unable to do his delivery rounds and was having clear episodes of attention lapses leading to letters being put to wrong addresses. He was transferred to “sorting” the post at sorting counters and was taken off delivery rounds. Even here the intractable yawning continued and he was committing sorting errors. By now it was affecting his colleagues too and they also started yawning (it is known to be contagious).It was affecting his self-confidence and was extremely embarrassing in all social situations to an extent that he started avoiding social interactions. He was drowsy all the time. He was clearly suffering more due to excessive yawning than due to depression. He was unable to perform his employment duties and was signed off sick. At that point the dose of fluoxetine was reduced to 20 mg .After a couple of weeks his yawning reduced significantly but was still disruptive to his routines. He was advised to slowly taper off fluoxetine over next 4 weeks. Unfortunately his depression relapsed and his GP restarted him on Fluoxetine 20 mg. He was reviewed by the psychiatrist after a couple of weeks. Once again he reported return of intractable yawning.

Fluoxetine was stopped once again and he was started on Mirtazapine 15 mg. There was very little response. The dose was increased to 30 mg after around two weeks. This led to him to experience nausea and vomiting. Unfortunately Mirtazapine too had to be stopped.  He was then tried on amitriptyline 50 mg which improved his sleep and symptoms of Depression. He was reviewed in the outpatient clinic after a couple of months .He did not develop any side effects and responded quite well. He then started his job starting from part time to full time within 6 weeks. After 6 months on the same dose of amitriptyline, did not have any symptoms of depression and was finally discharged from the mental health services.

Discussion

SSRI is the first line antidepressants used in the treatment of depression and Anxiety disorders. They are known to have least side effects and safest when it comes to overdosing. Intractable Yawning is quite an unusual and uncommon side effect.  One has to be conscious of the fact that it may cause yawning that can be pathological and can cause severe disruption of patient’s life. It can contribute to poor compliance. It is quite easy to overlook and ignore this side effect as yawning usually seems to represent sleep problems which is also a significant feature of the associated depression itself.

Excessive yawning can cause Jaw/facial pain. It can even cause dislocation of temporo-mandibular-joint. It can cause severe problems with one’s work and self-esteem. The sufferer might be misunderstood for being inattentive, indolent and sluggish. It might affect relationships with spouse/friend/relatives and especially at place of work. It can be misunderstood by doctors and lead to unnecessary tests and investigations. One has to be aware when prescribing SSRIs in patients who are driving or are involved in handling heavy machinery, athletes, airline pilots, surgeons, life guards, air traffic controllers and many other professionals. Due to its contagious nature, it’s not only the patient who is affected but also others around him. Excessive yawning can adversely affect the level of arousal, the level of concentration and work efficiency leading to poor performances in tasks requiring undiverted attention.

Hence excessive or intractable yawning has to be kept in mind while prescribing the so called most safe anti-depressant class of medication, the SSRIs, in this case fluoxetine.

Case Report

Blue discolouration of the skin can have a multitude of causes, including Mongolian spots, blue naevi, the naevi of Ito and Ota and metallic discolouration¹ or the use of drugs such as minocycline. Here we report the case of a 61 year old gentleman who developed blue macular skin lesions that were not attributable to any obvious cause and may be the result of an unidentified occupational exposure.

A 61 year old Caucasian gentleman developed blue macular skin lesions over a 14 year period. The very first lesion appeared in the middle phalanx of the right middle finger (figure 1). It was light blue and pinpoint, eventually darkening and increasing in size to approximately 1mm x 1mm, at which point becoming permanent and non-evolving. The lesion had no notable associated features and the patient was in otherwise good health.

Figure 1: The first blue macular lesion on the middle phalanx of the right middle finger.

Figure 2: A blue macular lesion on the terminal phalanx of the left middle finger.

Figure 3: A lesion on the anterior abdomen from which a punch biopsy was taken.

Figure 4: Haematoxylin and Eosin stained slide at 10x magnification. Abdominal skin biopsy showing dermal interstitial and perivascular distribution of black coloured pigment deposits

At present, he has approximately thirteen blue macular lesions in total, all of which have developed in the same manner. They are distributed predominantly on his hands with one on his left forearm and one on the right abdominal flank. New spots still continue to arise on his hands (figure 2).

A punch biopsy of the abdominal lesion (figure 3) was carried out. The histological findings were those of skin with normal intact epidermis and the presence of black coloured pigment granular deposits, located largely within the papillary dermis and occasional smaller deposits in the superficial reticular dermis (figure 4). The deep reticular dermis and subcutaneous fat were normal.  The pigment had a perivascular distribution and in dendritic histiocytic cells, with close association to fibroblasts. Histiocytic cells form part of the mononuclear phagocyte system and these cells are abducted mainly for phagocytosis removal or storing material². Apart from the pigment, the remaining skin was normal. The use of light microscopy alone does not identify all substances on examination of a Haematoxylin and Eosin (H&E) stained section of tissue. Applying polarisation light microscopy enables the identification of numerous structures, for example crystals, pigments, bone and amyloid³. However, the black coloured material here was non polarisable (no refractile foreign material could be identified).  These appearances as seen on light microscopy alone are most frequently seen where there is a history of tattoo artistry, but tattoo pigment is typically identified as showing reflective properties using polarisation⁴. Interestingly, the patient had no clinical history of deliberate tattooing and other causes were considered.

Discussion

The discovery of black coloured deposits in the dermis excludes the diagnoses of Mongolian spots or blue naevi and the naevi of Ito and Ota, all of which are disorders of dermal melanocytes. Another important differential is malignant melanoma, but it is not the diagnosis as the histopathological findings did not find any evidence of dysplasia or malignancy.

In a disorder known as anthracosis, similar findings of black coloured deposits can be seen in other organs such as within the lung and draining lymph nodes. It is often found in smokers and urban populations and reflects the deposition of carbon which is the most commonly identified exogenous mineral substance within tissue sections. The skin is not a site where such carbon pigment is typically seen and therefore, this is not a credible diagnosis in this case.

Agyria is a condition that occurs as a result of silver particle impregnation of skin leading to blue-grey skin discolouration. Silver exposure may be due to occupational or surgical exposure (by use of silver sutures) or medication with silver salts. On interview, the patient denied any occupational exposure to silver and the use of silver salts. Although the patient had had previous shoulder surgery, silver sutures are no longer used in modern day surgical practice and therefore this cannot be the cause of his skin discolouration.

Unfortunately, histological examination of paraffin embedded tissue sections can only confirm the presence and distribution of an exogenous substance and it is not possible to precisely differentiate the exact type of material which is present. The use of an electron probe micro analyser may have been useful in identifying the substance, however, such equipment is not currently available and was not used in this case.

Interestingly, in tattoo artistry, carbon black may be used to give blue tattoos their colour⁵ and this is also a component of tyres and industrial rubber products⁶. This provided us with a link to occupational exposure, given that this gentleman is a tyre worker and has been involved in both the manufacture and assembly of tyres for 34 years. Carbon can cause discoloration of the skin, depending on the extent of deposition.

It is notable that in his 34 years of working with tyres, this gentleman did not routinely use gloves or protective uniform until only 10 years ago. This was as workplace safety precautions were not as strongly enforced in previous times. He admitted to have been in direct contact with the materials involved in tyre building and also suffered accidental superficial cuts on his hands whilst working, which may be a route by which carbon may have been introduced into the dermis. This is supported by the observation that the majority of the blue macular lesions were on the hands. Adding credibility to this theory is the identification of a colleague of this gentleman’s (who did not wish to be identified), whose job also involved the manufacture and assembly of tyres, who also has a similar single blue macular lesion on his hand.

In addition to this we have identified a forum on the internet⁷ that reports other similar cases of blue pin-point macular lesions appearing on the skin of tyre factory workers – some of whom worked for the same tyre company that this gentleman did. This may suggest that there is an association between exposure to a chemical, possibly carbon black, involved in the manufacture of tyres, and the appearance of these blue macular lesions.

In this case report, the identity of the material deposited and the route by which it accumulated in the dermis is unclear, but may have been related to an occupational exposure – this was in keeping with the general consensus upon presentation of this case at the West Midlands Dermatology Conference at New Cross Hospital Wolverhampton. We welcome any new case reports or literature that may be able to shed further light on this subject.

Introduction

Pellagra is a nutritional disorder due to an insufficiency in vitamin B₃ also called vitamin PP (‘Pellagra preventing’). The disease was characterized by the following ‘3 D’s’: ‘Dermatitis’, ‘Dementia’ and ‘Diarrhoea’. When it is misdiagnosed, it can lead to the fourth ‘D’ which is ‘Death’.¹ It was very common in past centuries, particularly in populations that had an exclusively maize diet. Nowadays, this diet problem is rare in developed countries, so the disease is less frequent. However, many recent studies suggest that the disease has not been eradicated and can be under-diagnosed. Alcohol, drugs and malabsorption seem to be the new aetiologies of the disease. So, it is important to recognize the ‘3 D’s’ triad in such situations to avoid fatalities.

Observation

A 61-year-old patient presented to the Internal Medicine Department with an 8-month history of deterioration in her general state. She had a medical history of Epilepsy treated by Phenobarbital since she was 16.

Review of systems revealed gastrointestinal symptomatology consisting of intermittent diarrhoea (6-7 watery stools a day without blood), dysphagia and diffuse abdominal pain. The patient also reported a skin photosensitive eruption affecting her hands and feet.

Physical examination showed a listless patient with a low Body Mass Index (17 kg/m²).

On dermatological examination, symmetric, well-defined, brown-coloured and scaly eruption was observed on the dorsa of her feet (Picture 1) and hands (Picture 2). The mucous examination showed a commissural Cheilitis and Glossitis.

Picture 1: Brown well-defined patches on feet.

Picture 2: Brown pigmentation and scales of hands.

Picture 3: Hands aspect after treatment.

The nervous system examination revealed a pyramidal and cerebellar syndrome. The response to neurocognitive tests was altered suggesting Dementia.

The rest of physical examination was normal. In particular, there were neither peripheral lymph nodes, nor spleen or liver enlargements, nor abdominal mass.

The laboratory tests (glucose, calcaemia, creatinine, liver function tests, urine analysis, haemoglobin, hematocrit, sedimentation rate, C-reactive protein and protein electrophoresis) were within normal limits.

Oesogastroduodenal endoscopy was normal. The cranial, thoracic and abdominal CT scans were normal.

The diagnosis of Pellagra was made on dermatological, abdominal and neurological signs. The patient was treated by Niacin (1000 mg/day) and multivitamin complex. Phenobarbital was discontinued and switched to Clobazam. The patient’s symptoms started to improve quickly. Ten days after the treatment began the skin lesions (Picture 3) and gastrointestinal signs completely disappeared.

Discussion

Pellagra was diagnosed clinically in our patient based on the skin aspects. The skin lesions have been described since 1771 by Frapolli whose name was given to the disease: Pellagra which means rough skin in Italian.¹ The typical lesions consist of a brown pigmentation and scales with a photosensitive distribution and well-defined borders as seen in our patient. The face, the neck and the dorsa of the hands are the preferential locations.² The skin lesions are not always found, and cases of Pellagra Sine Pellagra have been described.³ The extra-cutaneous manifestations are less specific, but their association with pellagrous skin lesions are sufficient to reach a diagnosis. The neurological involvement is classically a Dementia syndrome, but ‘Pellagrous Encephalopathy’ can also consist of delirium, insomnia, depression, cerebellar and extrapyramidal syndrome.⁴ The gastrointestinal signs are non-specific; they can be Glossitis, dysphagia, nausea, vomiting and abdominal pain. An intractable diarrhoea may occur in advanced stages of disease and can quickly lead to death.⁵

In 1929, Goldberger attributed such clinical manifestations to a Niacin (vitamin B₃ or PP) deficiency. Niacin is a precursor for two important coenzymes namely ‘Nicotinamide Adenine Dinucleotide (NAD)’ and ‘NAD-Phosphate’ which are essential for many oxidative reactions. This probably explains why Pellagra affects tissues with a high rate of cell turnover such as the skin and digestive tract.⁴

Niacin can be directly absorbed by the gastrointestinal tract or synthesized from Tryptophan.

Primary Pellagra occurs when the diet is deficient in Niacin or Tryptophan as in poor populations with an exclusive maize diet (which contain Niacin but in an indigestible form).

Despite sufficient dietary Niacin, Secondary Pellagra may be caused by a problem in Niacin absorption or metabolism. Many causes of Secondary Pellagra were identified as alcohol, intestinal malabsorption, carcinoid tumours, Hartnup’s Syndrome, Anorexia Nervosa and drugs (Table 1).⁵ Our patient did not consume alcohol and had no biological signs of malabsorption, but she was taking Phenobarbital for about 45 years. In the English literature, we found only two cases of Phenobarbital-induced Pellagra, and with a fatality in one case.^{6, 7} The underlying mechanism of Pellagra caused by Phenobarbital, or other antiepileptic drugs, is an alteration in Niacinamide synthesis due to an enzymatic induction.⁵

Table 1: List of drugs predisposing to Pellagra.

The treatment is first based on correction of predisposing factors. In our patient, it consisted of using another antiepileptic drug instead of Phenobarbital. Second-line treatment is a vitamin therapy based on Niacin. There is no consensus on the doses, form and duration of the treatment, but the minimal dose is 300 mg of Niacin/day. A multivitamin complex containing other B-vitamins is often necessary because of the frequency of other deficits in such patients.² With treatment, the skin lesions and gastrointestinal symptoms generally disappear within a few hours or days, as in the case of our patient, and it is a good argument for a retrospective diagnosis of pellagra.¹ Testing for Niacin levels or urinary metabolites is not frequently available and it’s not necessary for the diagnosis.

Conclusion

We described a typical case of Pellagra in which the ‘3 D’s’ were present. In such a case, we should begin the treatment before the results of the laboratory investigations are known. The improvement of all symptoms within a few days is sufficient to confirm the diagnosis. However, the ‘3 D’s’ triad is not always present, and the clinician should consider the diagnosis in face of unexplained abdominal or neurological signs in certain patient groups.

INTRODUCTION

Anaesthetic management of a patient with a tracheal tumour is challenging, as the airway is shared with the surgeon and patency must be maintained despite airway manipulation.

Several anaesthetic techniques have been used in patients requiring tracheal resection and reconstruction. Cardiopulmonary bypass standby after femoral artery and vein cannulation and then intravenous/inhalational induction while oxygenating the patient has been considered to be a reasonable approach. Intratracheal tumours are challenging to anaesthetists because of the difficulty in establishment of a patent airway before commencement of surgery. The principal anaesthetic consideration is ventilation and oxygenation in the face of an open airway.

CASE REPORT

A 56 year old male with no other co-morbidities presented to the Thoracic Oncology department with a history of progressive dyspnoea and orthopnoea.On examination he was found to have dyspnoea at rest and could not complete full sentences while talking. Change of position made no difference to his symptoms.

Routine blood investigations which included full blood count, renal and liver functions, coagulation profile, ECG and 2DEcho were within normal limits. PFT showed a typical intrathoracic obstructive picture.

His chest X-ray showed bilateral hyperinflated lungs suggesting airtrapping. CT of the chest showed an intratracheal growth about 4 cm above the carina almost completely obstructing the lumen. An awake flexible bronschoscopy confirmed the CT scan findings. A 2.7mm flexible bronschocope was passed with difficulty beyond the tumour to visualise the carina. Excision of the intratracheal tumour was planned with possible resection and anastomosis of the involved tracheal segment. A careful perioperative plan was  discussed and decided in agreement with the thoracic surgeons, anaesthetist, cardiovascular surgeons and the rest of the team members.

Flexible and rigid bronchoscope, a Sanders venturi, an additional anaesthesia machine and various sizes of reinforced and normal endotracheal tubes and tracheostomy tubes were kept ready.

Preoperatively the patient had incentive spirometry and bronchodilator nebulisation and intravenous steroids. An awake epidural at T9-10 level and radial artery cannulation were done under local anaesthesia without any problems. Two 16 gauge peripheral IV lines were sited under local anaesthesia.

After adequate preoxygenation anaesthesia was induced with IV propofol along with oxygen and sevoflurane with BIS monitoring.  As mask ventilation proved to be easy the patient was paralysed with suxamethonium. There was no difficulty in ventilation after muscle paralysis. An 8.5 number COETT portex tube was placed in  the trachea with the cuff just beyond the cords to avoid possible trauma to the tumour. Since there was preoperative evidence of airtrapping, ventilator settings were set to an I:E ration of 1:3 with a tidal volume of 550ml, respiratory rate of 12-14 per minute and PEEP of 4. At these ventilator settings the airway pressures were reaching up to 22 cm of  H20 and ETCO2 reaching a maximum of 40mmHg. Anaesthesia was maintained with oxygen: air with sevoflurane and atarcurium for muscle paralysis.

Flexible bronchoscopy was done to confirm the position of the endotracheal tube, which showed that the ETT was adequately above the tumour.

A laryngeal drop procedure was done in the supine position with neck extension to facilitate mobilisation of the trachea for resection anastomosis. After the laryngeal drop procedure a right thoracotomy was done in the left lateral position. At this point of the procedure, the patient was ventilated with low tidal volumes of 300 and respiratory rate of 16-20 to keep the ETCO2 at around 40.The right lung was surgically retracted and the trachea  was exposed up to the carina. A repeat bronchoscopy was done through the ETT to help identify the upper and lower extent of the tumour. The trachea was then opened below the tumour, after which a 6.5 reinforced tube was introduced through the left bronchus to aid ventilation of the left lung. This ETT was withdrawn intermittently to help visualisation and aid surgical excision of the tumour and  sleeve resection of the trachea. The left lung was ventilated till partial closure of the trachea. The left-sided tube was then removed. Ventilation resumed through the orotracheal tube with intermittent occlusion of the defect with gauze by the surgeon.  The orotracheal tube was adjusted under vision before closure of the trachea to position it above the anastomotic site. The trachea was sutured and the thoracotomy incision closed without any adverse event. The neck was kept in a flexed position to avoid tension on the tracheal anastomotic area.

The patient was then extubated in the immediate postoperative period without any problems and the recovery was uneventful.

DISCUSSION

Anaesthetic management of a patient with a tracheal tumor is challenging, as the airway is shared with the surgeon, and patency must be maintained despite airway manipulation^{1, 2}. Several anaesthetic techniques have been used in patients requiring tracheal resection and reconstruction ^3–5.

Primary tracheal masses are very rare and mostly malignant, occurring in 0.2 in 1,00,000 persons per year ⁶ and among these squamous cell carcinomas form the main bulk. Cardiopulmonary bypass standby after femoral artery and vein cannulation and then intravenous/inhalational induction while oxygenating the patient with the oxygen inhalation has been considered to be a reasonable approach ⁷. Byrne JG et al (2004) advocated planned use of CPB to facilitate complete resection of thoracic malignancies after careful patient selection ⁸.

These patients are often mistaken to have asthma and require treatment with inhaled corticosteroids and beta agonists ⁹. They are generally treated for many years for asthma or COPD, unless a CT scan or endoscopic procedure is done for the symptoms ¹⁰. Intratracheal masses usually start getting symptomatic when 75% or more of the tracheal lumen is obstructed. Tracheal lesions present at lower level can have more complicated management of airway, anaesthesia and surgery for successful and safe removal of the mass. ¹⁰

Intratracheal tumours are challenging to anaesthetists because of the difficulty in establishment of a patent airway before commencement of surgery. The principal anaesthetic consideration is ventilation and oxygenation in the face of an open airway. Ventilation can be managed in many ways, including manual jet ventilation, high frequency jet ventilation, distal tracheal intubation, tracheostomy, spontaneous ventilation and CPB.¹¹

Knowledge of various techniques available for management of such cases is vital. In order to have a successful and safe outcome it is extremely important to have good communication between the anaesthetic, surgical and intensive care team.

The challenge in managing such cases lies in establishing and maintaining a patent airway and also preventing seepage of blood and tumour particles distally into the tracheobronchial tree during the surgery.

There is a possibility of total airway obstruction during ventilation attempts using positive pressure because airway obstruction has a fixed and dynamic component. Dislodgement of the tumour, possibly from trauma following intubation causing total obstruction, should also be considered.

Thus an intratracheal tumour was successfully removed without any complications and by avoiding CPB. This case report also highlights the importance of proper planning and good communication between team members to ensure a successful and safe outcome.

Case report

A 36 year old multiparous woman presented to the Labour Ward at 33 weeks’ gestation with lower abdominal pain. She had mild asthma and her obstetric history included 2 previous normal vaginal deliveries.

At 16 weeks’ gestation she was found during antenatal scanning to have a right ovarian cystic lesion measuring 59x34x50mm. This was monitored and a repeat scan at 25 weeks’ showed it had increased in size to 73x55x47mm.

At 32 weeks’ she was diagnosed with a DVT and was commenced on therapeutic enoxaparin (stopped two days before current presentation). (D-Dimer > 4000micrograms/L). An inferior vena cava filter was inserted. The patient declined any surgical intervention of the cystic lesion during pregnancy and an early elective Caesarean Section with surgical management of the cyst at 34 weeks’ gestation was planned. She had no symptoms or signs suggestive of a PE and was not formally investigated for one prior her current presentation.

On this presentation at 33 weeks’ gestation, her pain suddenly worsened with associated hypotension and evidence of foetal distress and so an emergency exploratory laparotomy was performed.

Admission haematology results: haemoglobin 10g/dL, platelets 158 x 10⁹ /L, INR 1.0, APTT 28.4 seconds, APTT ratio 1, fibrinogen 3g/L.

She underwent a rapid sequence induction in the supine wedged position using thiopentone and suxamethonium. She was a Grade 1 intubation and anaesthesia was maintained with oxygen/nitrous oxide/sevoflurane and muscle relaxation was achieved with atracurium. A ruptured, torted right ovarian mass containing an estimated one litre of altered blood was noted. At Caesarean section a live male infant was delivered. A right oophrectomy was then performed. The infant was subsequently intubated and transferred to the Neonatal Intensive Care Unit.

Oxytocin 5IU followed by a 40IU infusion over 4 hours was administered following delivery of the baby. Effective haemostats was achieved, the uterus appeared well-contracted and the patient’s abdomen was closed. Surgical blood loss was estimated as 600mls (excluding blood within the ovarian mass).

Thirty minutes following completion of surgery the patient, fresh blood was noted vaginally. A HemoCue* reading was taken as 5.9g/dL. Four units of blood and two units of FFP were transfused whilst a second laparotomy was performed. Fresh blood was noted intra-abdominally and the uterus was markedly atonic. Ergometrine 500mcg and carboprost trimethamine 250mcg were administered intramuscularly, as well as, misoprostol 800mg vaginally. A hysterectomy was performed due to the rate of bleeding and the evident haemodynamic instability.

Coagulation studies: platelets 27 x 10⁹/L, INR 1.4, APTT 101.8 seconds, APTT ratio 3.6 and fibrinogen 1g/L.

A further 4 units of blood, 4 units of FFP, 1 pool of platelets and 2 units of cryoprecipitate were transfused. Factor VII was also administered on advice from the Haematologist.

An internal jugular central venous catheter was inserted and a noradrenaline infusion started. Initial attempts to insert an arterial cannula were unsuccessful as peripheral pulses were difficult to palpate. Venous blood gas readings revealed hyperkalaemia (K+ 6.4mmol/L) which was treated with sodium bicarbonate, 10mls 10% calcium chloride and a continuous 50% dextrose and insulin infusion. Her abdomen was packed and percutaneous drains were inserted. Anaesthesia, close monitoring and resuscitation continued.

Ongoing output from drains prompted a third exploration after an hour. There was generalised oozing particularly around the bed of the resected ovary in the pouch of Douglas.

Coagulation profile: platelets 50 x 10⁹/L, INR 1.4, APTT 89.6 seconds, APTT ratio 3.1 seconds, fibrinogen 1.4g/dL.

A further 10 units of blood, 3 pools of platelets, 5 units of FFP and 3 units of cryoprecipitate were transfused. Sequential coagulation profiles and thromboelastography were used to guide transfusion.

During this exploration, ECG revealed a broad complex tachycardia with no palpable pulse confirming cardiac arrest likely secondary to hypovolaemia and/or hyperkalaemia. Return of spontaneous circulation was achieved after 3 cycles of continuous cardiac massage, 4 direct current shocks and adrenaline 1mg.

A femoral artery cut-down was performed and arterial cannula was inserted by a general surgeon. IABP monitoring commenced.

A fourth exploration was carried out around two hours later for ongoing blood loss. Again only generalised oozing was noted particularly from the oophrectomy site. Her abdomen was re-packed.

Coagulation profile: haemoglobin 7.4g/dL, INR 1.2, APTT 48.9 seconds, APTT ratio 1.7, fibrinogen 1.9g/dL, platelets 94 x 10⁹/L.

She was transferred to the ICU eight hours from the start of the primary operation. Estimated total blood loss was 13,200mls. Transfusions continued and her abdomen was closed two days later. She received haemofiltration therapy for acute kidney injury. She recovered to her premorbid level with no neurological deficit before being discharged with her baby a few weeks later. In total, she was transfused 64 units of blood, 35 units of FFP, 10 pools of platelets and 11 units of cryoprecipitate. Histology of the ovarian mass revealed high grade clear cell carcinoma for which she received chemotherapy but unfortunately she died two years later.

Discussion

The association between cancer and thromboembolism has been well established for many decades.¹Ovarian cancer patients have one of the highest incidences of VTE amongst cancer patients, particularly clear cell carcinoma (CCC). One study found the incidence of thromboembolic complications to be significantly higher in CCC when compared with other epithelial types of ovarian cancer (27.3% vs 6.8%).²

The pathological mechanism behind the hypercoaguable state induced in ovarian cancer patients appears to be largely multi-faceted. A variety of procoagulant subtances may be involved. Of particular interest is tissue factor (TF), a potent procoagulant found in endothelial and blood cells, as well as, in tumour cells. TF may play an important role in this hypercoaguable state through activation of the coagulation cascade.³ TF is frequently over-expressed in ovarian cancer tissue and there is research suggesting it influences tumour progression.^3,4

A hyperviscosity syndrome may also be seen in association with ovarian cancer which favours thrombosis and may accelerate tumour progression and metastasis.DVT is a recognised complicating factor of ovarian cancer which may adversely affect the course of the disease possibly as a component of this hyperviscosity syndrome^5,6 Ovarian cancer patients are also vulnerable to developing cerebrovascular complications and carry a higher risk of developing ischaemic strokes which has a significant impact on morbidity and mortality.⁷

The hypercoagulable state seen in cancer patients can have a spectrum of manifestations that ranges from DIC to massive thromboembolism. DIC in these instances is usually chronic and subclinical.⁸

The degree of coagulopathy which was seen in this case could not be attributed solely to dilutional effects incurred through fluid resuscitation. Instead we propose the acute severe coagulopathy was a consequence of procoagulant factors inherent to neoplastic tissue, including TF, which were suddenly released into the circulation following rupture and surgery to the ovarian tumour. The overall result would be widespread activation of the clotting cascade and a consumptive coagulopathy.

This case aims to increase awareness of a refractory coagulopathy which may be seen following rupture and/or surgical resection of a malignant ovarian tumour. The presence of an ovarian cyst especially in conjunction with evidence of vascular thrombosis in pregnancy should raise suspicion for an ovarian malignancy and hence vigilance for this potential complication. Anticipation of such a severe coagulopathy and associated significant blood loss should pre-empt early establishment of invasive monitoring, ample intravenous access and ensuring quick access to blood and blood products. Bedside coagulation tests such as thromboelastography are useful guides to blood product transfusion.⁹ Prompt mobilisation of resources, multi-disciplinary input and effective team work were crucial factors which facilitated the management of this case.

Acute, severe DIC associated with ovarian intratumoural bleed which resolves following resection of the tumour has been reported previously.¹⁰ This is the first case to the best of our knowledge occurring following ovarian cancer torsion and rupture during pregnancy.

Polyarteritis nodosa (PAN) is a rare vasculitis in childhood. Since first described by Kussmaul and Maier in 1866 ¹, there have been approximately 140 pediatric case reports in the literature. Traditionally, children were classified as having one of three forms: infantile, cutaneous, and systemic. Infantile PAN is now recognized as a severe form of Kawasaki disease. Criteria for a diagnosis of systemic PAN in childhood have been proposed but not validated².

Cutaneous PAN (cPAN) is recognized as a separate entity but there are no diagnostic criteria for cPAN². cPAN is characterized by disease affecting the skin with no major organ system involvement. The cutaneous symptoms are similar to systemic PAN and mild fever, muscle, joint, and peripheral nervous system involvement may also occur. Fever, rash, and musculoskeletal symptoms are common in children and cPAN needs to be differentiated from other diagnostic entities. Definitive diagnosis is by histopathologic evidence of necrotizing inflammation of the medium and small-sized arteries. There is a paucity of knowledge of the spectrum of clinical presentation and management of children with cPAN. Here we describe a case of cPAN and summarize the clinical manifestations, laboratory data and treatment regimens of our patient.

Case report

This 10 year old female adolescent presented with pain in both the elbow joints followed by pain in the left knee joint and both the ankle joints in a course of 8 days and fever for the past 2 days .On admission her vitals were stable, both the elbow joints were tender and the knee and ankle joints were swollen and tender .She had multiple subcutaneous nodules over extensor aspect of both her forearms, both her tibial shins and few on her thighs. Systemic examination showed presence of a soft PSM of grade 2 intensity over left sternal edge. Blood investigations showed leucocytosis, elevated CRP, elevated ASLO titres and 2Decho revealed a mild tricuspid regurgitation. An initial diagnosis of acute rheumatic fever was made and child was started on penicillin and Aspirin. But child continued to have excruciating arthralgia and hence a rheumatologist opinion was taken. Child was advised a skin biopsy from the nodular lesions which showed small and medium vessel vasculitis suggestive of cutaneous polyarteritis nodosa. Hence she was stopped with aspirin therapy , given pulse therapy with methyl prednisolone and continued with penicillin therapy .Her arthralgia subsided within a day of pulse therapy and the subcutaneous nodules gradually disappeared .On discharge child was put on oral steroid therapy and penicillin prophylaxis and advised regular follow up.

Fig1 showing segmental fibrinoid necrosis with inflammatory infiltrates of small artery.

Fig2 showing leukocystoclastic vasculitis with fragmentation of neutrophils in and around blood vessels.

DISCUSSION:

cPAN is not common in the pediatric population with approximately 140 cases reported in the literature. Disease is limited to skin, joints, and muscles in the majority with a minority having nerve involvement. Constitutional symptoms are common. Most children have a chronic and relapsing benign course.

The precise etiology of cPAN remains to be unknown. However, an immune mediated mechanism has been postulated. Several infectious and noninfectious conditions have been associated both to initiation and relapse of the disease^3,4,5. Among them, streptococcal infection has been commonly implicated^6,7. Although some evidence of streptococcal infection as an initiating factor for cPAN is present, caution must be exercised when interpreting elevations in the serologic markers of streptococcal infection in the absence of an appropriate clinical presentation.

Cutaneous and systemic PAN share the same histopathologic features of necrotizing arteritis of small and medium sized vessels. Kussmaul and Meier described the first case of systemic PAN in 1866 ¹. Early reports ^8,9 confirm that cPAN is a separate entity to systemic PAN. We have limited our definition of cPAN to disease affecting the skin, muscle, joints, and peripheral nervous system, with corresponding biopsy confirmation. Any evidence of visceral involvement, either clinically (central nervous system, pulmonary, cardiac, gastrointestinal, or renal), radiographically (abnormal angiography), or by histology (visceral biopsy) were classified as systemic PAN. Nakamura et al ¹⁰ proposed further restriction of the definition of cutaneous PAN in that any extracutaneous involvement such as peripheral neuropathy and myalgias must be limited to the same area as skin lesions. Systemic PAN and cPAN appear to be fairly distinct entities on a clinical continuum. There are only 5 reported cases of cPAN evolving into systemic PAN ^11,12.

On review of treatment regimens reported in the literature, most children respond to corticosteroids. Penicillin should be considered in children with increased ASO titres ^13,14. Recent case series report success with low-dose methotrexate, cyclophosphamide, intravenous immunoglobulin, and biologic therapies ^15,16.

In summary, cPAN can be challenging to diagnose and manage. A diagnosis of cPAN should be considered in a child with fever, tender subcutaneous nodules, livido reticularis, and arthralgias/arthritis. Most children respond to corticosteroids and have a benign course, but some children may be corticosteroid dependent or corticosteroid resistant, necessitating other immunosuppressive agents including DMARDs and biologic therapy. Multicentre pediatric vasculitis disease registries are necessary to inform development and standardization of best clinical practice for childhood cPAN.

INTRODUCTION

Although it is well appreciated that pancreatitis is frequently secondary to biliary tract disease and alcohol abuse, it can also be caused by drugs, trauma and viral infections, or even be associated with metabolic and connective tissue disorders.¹ Knowledge of the true incidence of drug-induced acute pancreatitis is dependent on the clinician’s ability to exclude other possible causes, and by promptly reporting the occurrence. Based on individual case reports and case control studies of drug-induced acute pancreatitis, the estimated overall incidence ranges from between 0.1 and 2% of pancreatitis cases.^2,3 In particular, drug-induced acute pancreatitis is of mild severity and usually resolves without significant complications.⁴

Attempts have been made to categorize the risk of drugs causing acute pancreatitis. A previous classification system described by Mallory and Kern Jr. categorized drugs associated with acute pancreatitis as definite, probable, or possible.⁵ Trivedi et al. proposed a newer classification system for commonly used medications associated with drug-induced pancreatitis. Class I drugs are those medications with at least 20 reported cases of acute pancreatitis and at least one case with a positive rechallenge. Drugs with fewer than 20 but more than 10 reported cases of acute pancreatitis, with or without a positive rechallenge, are designated into Class II. While those medications with 10 or less reported cases, or unpublished reports in pharmaceutical or FDA files, are grouped into Class III.⁶

Acute pancreatitis as a result of either doxorubicin or cyclophosphamide, or a combination of both, or fluorouracil or epirubicin is a rare occurrence and has seldom been reported in the literature. Even the drug package labels registered with the FDA do not indicate the possible occurrence of pancreatitis. In this case report, we present a rare occurrence of drug-induced acute pancreatitis after the completion of the first cycle of the chemotherapy protocol involving cyclophosphamide and doxorubicin in a patient with stage 3 breast cancer, with recurrences of acute pancreatitis after re-challenging with cyclophosphamide and a derivative of doxorubicin, given individually on two separate occasions.

CASE PRESENTATION

A 58 year-old female presented to the emergency room with a one day history of severe, diffuse, deep-seated abdominal pain that radiated to her back, associated with nausea and vomiting, and was unrelieved despite the intake of NSAIDs. There was no reported fever, chills, diarrhea, dysuria, or antecedent trauma. Her medical history is notable for well-controlled hypertension, hyperlipidemia and hypothyroidism for which she takes amlodipine, atorvastatin and levothyroxine. She was recently diagnosed with left-sided breast cancer, Stage III, two months prior to admission and underwent a left modified radical mastectomy. Three days prior to the hospital visit, she was given her first cycle of chemotherapy with Doxorubicin 60 mg/m² and Cyclophosphamide 600 mg/m² along with Pegfilgrastim 6 mg and Fosaprepitant 150 mg. She is a former cigarette smoker, drinks alcohol infrequently, and denies illicit drug use. Her family history is unremarkable.

Physical examination revealed stable vital signs without a fever (36.6^◦C). She had non-icteric sclerae and a dry oral mucosa. Chest exam revealed a well-healed left mastectomy scar and an infusaport located on the right anterior chest wall. Her breath sounds were clear bilaterally. Her heart sounds were normal. Her abdominal exam was significant for mild tenderness to palpation in the epigastric area without palpable masses, organomegaly or ascites. There was no evident ecchymosis observed. The extremities were warm to touch with intact and symmetrical pulses, and without bipedal edema.

Initial work-up revealed an elevated leukocyte count of 42,000 with 80% neutrophils and 17% band forms. Basic metabolic panel was normal except for mild hyponatremia of 129 mEq/L. Serum amylase and lipase were markedly elevated at 2802 units/L and >2000 units/L, respectively. Liver function panel was normal (Alk phos 63 U/L [ref range 30-115 U/L], GGT 21 U/L [ref range 3-40 U/L], total bilirubin 0.90 mg/dL [ref range 0-1.5 mg/dL]). The coagulation profile was within normal range. Imaging of the abdomen with a CAT scan with intravenous and oral contrast showed haziness in the pancreatic fat plane suggestive of pancreatic inflammation, with no gallstones, focal abscesses, hepatic masses, or biliary ductal dilatation (Figure 1). Right upper quadrant ultrasound was essentially normal (Figure 2).

Figure 1. Coronal view of CT scan of abdomen and pelvis with IV and oral contrast showing haziness in the peripancreatic fat plane.

Figure 2. Sonogram of the right upper quadrant of the abdomen showing gallbladder devoid of gallstones and non-dilated common bile duct.

She was admitted to the medicine unit with the assessment of acute pancreatitis likely secondary to doxorubicin and cyclophosphamide. Intravenous fluid hydration with normal saline was initiated. She was kept NPO (nothing per os) and was started on empiric Imipenem and IV Esomeprazole. Her abdominal pain was controlled with intravenous morphine and her nausea with Ondansetron as needed. The serial basic chemistry panel was monitored and electrolyte deficits were replaced accordingly. Further work-up was performed to identify other possible etiologies of pancreatitis. The lipid panel was within normal limits (cholesterol 169 mg/dL [0-200 mg/dL], HDL 74 mg/dL, LDL 71 mg/dL and triglycerides 54 mg/dL [0-150 mg/dL]). The serum calcium levels remained within the normal range throughout her hospital stay. An abdominal sonogram demonstrated absence of gallstones or dilatation of the common bile duct, with a normal appearing liver parenchyma and pancreas. During her stay in the medicine unit, the patient’s abdominal pain improved and she was gradually started on an oral diet, which she tolerated well. Her serum electrolytes remained stable, while her serial CBC revealed progressively decreasing trends in leukocytes, hemoglobin, hematocrit, and platelet count, findings which were attributed to her prior chemotherapy. Repeat serum amylase and lipase both trended downward. The patient was discharged with follow up in the Oncology clinic. A month later, she was started on another chemotherapy regimen that consisted of weekly administration of Paclitaxel 80 mg/m² which, over the next two months, the patient completed without any complications. Then, after explaining the risks of recurrent pancreatitis, the patient consented to have a trial of cyclophosphamide 500 mg/m² along with fluorouracil 500 mg/m². Five days after receiving the chemotherapy, the patient developed acute pancreatitis which was attributed to cyclophosphamide. She again made a full recovery at that time. Three weeks later, her chemotherapy regimen was again changed, to epirubicin 90 mg/m² and fluorouracil. Four days after receiving this regimen, she again, for the third time, had a recurrence of acute pancreatitis. At this time, a repeat abdominal sonogram revealed a 4mm echogenic focus adherent to the anterior gallbladder wall with a comet tail sign, suggestive of cholesterol crystals lodged within Rokitansky-Aschoff sinuses of the gallbladder wall. There were no visible gallstones. A subsequent MRI of the abdomen with contrast revealed a small rounded hypointensity in the dependent portion of the gallbladder wall that was suggestive of a gallstone, however, there was no biliary obstruction, choledocholithiasis or an obstructing pancreatic mass. At this point, chemotherapy was stopped and anastrozole along with radiation therapy was initiated. The patient continues to be followed regularly and has had no recurrence of pancreatitis since her last episode.

DISCUSSION

The case presented described the development of acute pancreatitis in a patient with breast cancer three days after receiving the chemotherapy regimen consisting of cyclophosphamide and doxorubicin. After re-challenging the patient with cyclophosphamide, and again a few weeks later with a derivative of doxorubicin, epirubicin, acute pancreatitis recurred on each occasion. Despite the presence of cholelithiasis detected on the abdominal MRI, the temporal presentation of acute pancreatitis after chemotherapy exposure is highly suggestive of the role these chemotherapeutic agents played in triggering these three acute attacks. Acute pancreatitis was diagnosed based on the clinical suspicion and symptoms suggestive of the acute pancreatitis and was supported by the marked elevation in serum amylase and lipase, as well as, the radiologic evidence of pancreatic inflammation, both of which are markers of acute pancreatitis.

Chemotherapy-induced acute pancreatitis involving cyclophosphamide and doxorubicin either alone or in combination, is quite rare that even the drug labels registered with the FDA do not indicate acute pancreatitis as one of the possible complications. This scenario highlights the importance of drug surveillance and prompt reporting in order to maintain a credible drug safety database.

Though the drug latency, which is the interval between the initial exposure to the drug and the development of acute pancreatitis, differs variably, the present case is considered to have an intermediate latency (1-30 days). Other drugs may have short (< 24 hours) or long (>30 days) latency periods. Examples of drugs with short latency are acetaminophen, codeine, erythromycin and propofol. Intermediate latency drugs include L-asparaginase, pentamidine and stibugluconate. Drugs with long latency are estrogen, tamoxifen, valproate and dideoxyinosine.⁷

Based on the revised classification of Badalov et al, the combination of cyclophosphamide and doxorubicin is classified as Class IV drugs, which have the weakest association with acute pancreatitis due to limited information and the lack of adequate detailed case reports. Fluorouracil, which has been known to cause a gastrointestinal ulcer, is also categorized as a Class IV drug, while epirubicin, which is derived from doxorubicin, has not been classified, as it has not been reported before to cause acute pancreatitis. In implicating drugs in the etiology of acute pancreatitis, two conditions must be considered to weigh the strength of the association between the causality and the disease process, namely: a positive rechallenge test resulting in the recurrence of pancreatitis and a similar latency period between the drug exposure and development of the disease.⁷

The combination of drugs rather than a single agent was implicated for drug-induced pancreatitis in a previous case report that described the development of acute pancreatitis shortly after the second cycle of the chemotherapy regimen composed of cyclophosphamide, doxorubicin, and vincristine in a patient with mediastinal immunoblastic lymphoma. The pancreatitis episode resolved over 48 hours without complications.⁸

Another case was described in a patient with breast cancer developing acute pancreatitis four days after the third cycle of chemotherapy, which involved docetaxel and carboplatin.⁹

Toprak et al. reported the occurrence of acute pancreatitis in a patient with multiple myeloma after the initial treatment with the triple regimen chemotherapy protocol consisting of vincristine, doxorubicin, and dexamethasone. In this case report, symptoms suggestive of acute pancreatitis started to manifest on the first day of the treatment, with resolution following discontinuation of the drugs.¹⁰

Other antineoplastic agents for breast cancer associated with drug-induced pancreatitis are alemtuzumab, trastuzumab and tamoxifen. Extended use of these medications may cause chronic pancreatitis as a result of their causing repeated clinical or subclinical episodes of acute pancreatitis.⁶ Most cases of drug-induced pancreatitis follow a mild clinical course.⁷

In a retrospective study involving 1613 patients diagnosed with acute pancreatitis in a gastroenterology center, the incidence of drug-induced pancreatitis had been reported in 1.4% of patients treated for acute pancreatitis. It has been observed that a higher incidence of drug-induced acute pancreatitis occurs in elderly or pediatric patients, and in those patients with inflammatory bowel disease or AIDS.¹¹

The pathophysiology behind drug-induced pancreatic injury remains unclear. Potential mechanisms underlying such pancreatic injury might be related to drug hepatotoxicity which can be secondary to intrinsic toxicity of the drugs affecting the tissue, or due to an idiosyncratic reaction. In the vast majority of the cases, an idiosyncratic reaction could be the main pathway for tissue injury through a hypersensitivity reaction or production of toxic intermediate metabolites. Idiosyncratic reactions have a longer latency period of months to years before the onset of pancreatitis while the onset of hypersensitivity reactions is earlier (i.e. 1-6 weeks).⁷

CONCLUSION

Due to a variable latent period between the initial drug exposure and the onset of clinical symptoms, drug-induced pancreatitis must remain as a differential diagnosis in patients receiving chemotherapy regimens and presenting with the constellation of symptoms typical of acute pancreatitis. Due to the unclear pathogenesis of chemotherapy-induced pancreatitis, post-marketing surveillance and adverse drug reporting are paramount in elucidating the effect these drugs have on the pancreas.