Background

Flecainide is a class IC antiarrhythmic drug indicated for patients with supraventricular arrhythmias. It has proarrhythmic effects and although toxicity is rare, mortality rate can be as high as 10%.

Objectives

Flecainide is primarily metabolized by CYP2D6. We present a case of flecainide toxicity contributed by CYP2D6 polymorphism.

Methods

Patient was born at 29+5 weeks and was initially treated with propranolol for supraventricular tachycardia (SVT). However, he developed a recurrence of SVT and was switched to oral flecainide.

7 days after flecainide initiation, he became bradycardic. Electrocardiogram (ECG) showed sinus node dysfunction and atrioventricular block. The bradycardia was refractory to drugs, transcutaneous and transvenous pacing. In view of poor perfusion and severely reduced biventricular function on 2D-echocardiogram, extracorporeal membrane oxygenation (ECMO) was initiated. He was treated with IV calcium chloride and lipid emulsion and subsequently reverted to sinus rhythm with return of arterial pulsatility. Patient was supported on ECMO for 12 days and started on sotalol for SVT control.

Results

While other causes of flecainide toxicity were ruled out, investigations revealed supratherapeutic serum flecainide concentrations. Since flecainide is metabolised by CYP2D6 which exhibits polymorphism, a gene test was sent out to determine patient’s genotype. Patient has CYP2D6*10X2/*36 genotype that is predicted to have intermediate CYP2D6 enzyme activity which is associated with lower flecainide clearance.

Conclusion

Based on patient’s genotype, 25% dose reduction of flecainide is recommended by some national guidelines. This case report highlights the importance of ECG and flecainide levels monitoring when treating neonates. CYP genotype testing can further identify at-risk patients of toxicity.

Background

Oncology patients admitted to the paediatric intensive care unit (PICU) are at high risk of mortality.

Objectives

This study aims to describe the risk factors for mortality in these patients.

Methods

This retrospective study included all PICU oncology admissions from 2011-2017. Clinical and laboratory data were extracted. Primary outcome was PICU mortality. PICU survivors and non-survivors were compared, using Chi-squared and Wilcoxson rank-sum tests as deemed appropriate. Covariates included in the multivariable logistic regression for mortality included: type of admissions, previous bacteraemia and PRISM (Paediatric Risk of Mortality) III scores.

Results

Of 200 patients identified, the most common diagnoses were brain tumours (73/200, 36.5%) and acute lymphoblastic leukaemia (36/200, 18.0%). Median age and PRISM III scores were 7 (3, 12) years and 3 (0, 5) respectively. Septic shock was the main reason for emergent admissions (25/97, 25.8%). Overall mortality was 18/200 (9.0%). Mortality was higher in emergent cases than elective cases [17/97, 17.5% vs 1/103, 0.9%, p<0.001], in those with active bacteraemia [6/25, 24.0% vs 12/175, 6.9%, p<0.001], previous bacteraemia [10/35, 28.6% vs 8/165, 4.8%, p<0.001] and neutropenia [7/37, 18.9% vs 11/163, 6.7%, p=0.020] compared to those without. Independent risk factors for mortality were emergent admissions (adjusted odds ratio [aOR]: 10.3, (95% confidence interval [CI]: 1.26, 84.03); p=0.030) previous bacteraemia (aOR: 3.3, (95% CI: 1.06, 10.38); p=0.039) and higher PRISM III scores (aOR: 1.2, (95% CI: 1.02, 1.29); p=0.021).

Conclusion

Emergent PICU admissions, previous bacteraemia and higher PRISM III scores were independent risk factors for mortality in our centre.

Background

The Pediatric Acute Lung Injury Consensus Conference (PALICC) definition is currently used to diagnose pediatric acute respiratory distress syndrome (PARDS).

Objectives

We aimed to determine the clinical characteristics of patients identified to have PARDS and evaluate whether these patients could be better defined by alternative diagnoses.

Methods

We screened all pediatric intensive care unit (PICU) admissions daily over a year (2018). Patients who fulfilled the PALICC definition were prospectively recruited. We interrogated the clinical history and course of illness to identify factors which (1) made the diagnosis of PARDS unlikely or (2) altered the course of illness in patients with PARDS.

Results

Over the one-year period, there were 594 PICU admissions. 47/594(8%) patients were identified to have PARDS. Despite meeting the PALICC criteria, 5/47(11%) patients were suspected to have alternate diagnosis to PARDS [congestive cardiac failure/ fluid overload (despite normal 2D-echocardiography), pulmonary hypertension, pulmonary hemorrhage, lung contusion with drained hemopneumothorax]. 28/47(60%) patients had underlying comorbidities, of which 8/47(17%) were likely to confound the evolution of PARDS – these were patients who had underlying pulmonary hemosiderosis, pulmonary lymphangiectasia, restrictive lung disease and immunodeficiency. Together, these patients were more likely to develop severe PARDS [5/13(38%) vs 3/34(9%); p=0.028], and had increased mortality [4/13(30%) vs 3/34(9%); p=0.080] compared to the general PARDS cohort, though it was not statistically significant.

Conclusion

A significant proportion of patients fulfilling criteria for PARDS are suspected to have alternate diagnoses (false positives) or underlying diseases that potentially alter the course of illness.