OP38 – 3035: Incidence of and risk factors for neurological complications of cardiac bypass surgery in children with congenital heart disease

Abstract

Objectives Mortality associated with cardiopulmonary bypass (CPB) for correction or palliation of congenital heart defects (CHD) has decreased. Emphasis is shifting towards morbidity as a primary outcome. This case review conducted in a centre currently providing paediatric cardiac services was designed to establish the current incidence of and risk factors for post-operative neurological complications. Methods All children with CHD undergoing surgery involving CPB (including those needing circulatory arrest) between 01/04/2003–31/03/2010 were included. Patients were identified from the electronic procedural database, and discharge summaries were reviewed independently by two consultant paediatric neurologists, with case note review if information required clarification, and for all with neurological complications (NC). Data on haematology, EEG and cerebral CT/MRI images were retrieved and cross- referenced. Each procedure was categorised according to consensus definitions of the Multi-Societal Database Committee for Paediatric and Congenital Heart Disease. Results Of 922 operations, there were 19 (2.06%) deaths. Longer cardiac bypass and cross-clamp times were associated with death (p=0.004, p=0.032 respectively), but not with NC, which occurred in 46 (5%), including seizures (n=9, of which 3 were post-arrest and 5 were hypocalcaemic); vocal cord palsy (n=6), eye movement disorder (n=5), new or evolving intraventricular haemorrhage (n=2, both on ECMO), Horner's syndrome (n=1), peripheral neuropathy (n=1), chorea (n=1), headache (n=2) and phrenic nerve palsy (n=19). There were no pyramidal/extrapyramidal disorders. All NC other than Horner's syndrome and vocal cord palsy were transient. Post-operative haematocrit was lower in those with NC (0.357±0.07) than in controls (0.389±0.075) matched for date of birth, gender and diagnosis (p=0.049). Conclusions The range of observed NC following CPB may be changing, and only detailed analysis will provide an insight into the processes resulting in this change. Further studies should examine pre- and peri-operative factors leading to lower haematocrit, a potentially modifiable risk factor for NC. Mortality associated with cardiopulmonary bypass (CPB) for correction or palliation of congenital heart defects (CHD) has decreased. Emphasis is shifting towards morbidity as a primary outcome. This case review conducted in a centre currently providing paediatric cardiac services was designed to establish the current incidence of and risk factors for post-operative neurological complications. All children with CHD undergoing surgery involving CPB (including those needing circulatory arrest) between 01/04/2003–31/03/2010 were included. Patients were identified from the electronic procedural database, and discharge summaries were reviewed independently by two consultant paediatric neurologists, with case note review if information required clarification, and for all with neurological complications (NC). Data on haematology, EEG and cerebral CT/MRI images were retrieved and cross- referenced. Each procedure was categorised according to consensus definitions of the Multi-Societal Database Committee for Paediatric and Congenital Heart Disease. Of 922 operations, there were 19 (2.06%) deaths. Longer cardiac bypass and cross-clamp times were associated with death (p=0.004, p=0.032 respectively), but not with NC, which occurred in 46 (5%), including seizures (n=9, of which 3 were post-arrest and 5 were hypocalcaemic); vocal cord palsy (n=6), eye movement disorder (n=5), new or evolving intraventricular haemorrhage (n=2, both on ECMO), Horner's syndrome (n=1), peripheral neuropathy (n=1), chorea (n=1), headache (n=2) and phrenic nerve palsy (n=19). There were no pyramidal/extrapyramidal disorders. All NC other than Horner's syndrome and vocal cord palsy were transient. Post-operative haematocrit was lower in those with NC (0.357±0.07) than in controls (0.389±0.075) matched for date of birth, gender and diagnosis (p=0.049). The range of observed NC following CPB may be changing, and only detailed analysis will provide an insight into the processes resulting in this change. Further studies should examine pre- and peri-operative factors leading to lower haematocrit, a potentially modifiable risk factor for NC.