Perioperative outcomes of paediatric congenital cardiac surgery in Johannesburg, South Africa

Abstract

Background: Congenital cardiac diseases are the commonest congenital birth defects with children undergoing surgery at increased risk of mortality. Outcomes following congenital cardiac surgery in children has been well documented in high-income countries for many years, but little is known from the African continent. Aim: The aim of this study was to determine factors associated with perioperative mortality in patients that underwent congenital cardiac surgery at an academic institution. Methods: This retrospective cross-sectional study was conducted at Charlotte Maxeke Johannesburg Academic Hospital spanning a period of 12 years (2006-2017). The surgical, ICU, and anaesthetic records of patients that underwent surgery were reviewed. The records of all congenital cardiac disease paediatric patients undergoing surgical repair were reviewed. Associations between demographic, surgical and other factors were sought with mortality. A multivariable regression analysis was performed for the factors which had p-values of 0.1 and less in the univariable regression analysis. A Cox regression analysis was performed for mortality over time. All hypothesis testing accepted a p-value of <0.05 as statistically significant. Results: There was an 11% (188/1701) in-hospital mortality rate overall. Patients that died were primarily younger, 0·33 years (IQR 0·13 - 1·25), had lower median weight of 5kg (IQR 3·45 – 8·20), longer cardiopulmonary bypass (179 min) (IQR 16·0 - 272·5) and aortic cross clamp (82·5min) (IQR 35.0-139.0) times. Palliative surgeries had the highest mortality rate with 41% of all mortalities coming from this category. iv Cardiopulmonary bypass time (p < 0·001), ICU length of stay (p <0·001), aortic cross clamp time (p = 0·001), days on the ventilator (p < 0·001), and the STAT 2020 mortality score (p = 0·04) were independently predictive of perioperative all-cause in- hospital mortality in a multivariable regression model. The model yielded an area under the curve of 90%. Conclusion: Perioperative mortality following congenital cardiac surgery was higher in our study group compared to high income countries but it was similar to that of other low-middle income countries.