Prevalence of dysglycemia following paediatric cardiac surgery: a systematic review and meta-analysis

Abstract

BACKGROUND: Postoperative dysglycemia is a common but under-recognised complication in paediatric cardiac surgery, driven by physiologic stress, hormonal changes and cardiopulmonary bypass. Reported prevalence varies widely, and no meta-analysis has synthesised the evidence. Therefore, this study aimed to estimate the pooled prevalence of postoperative dysglycaemia and describe associated postoperative outcomes, such as intensive care unit (ICU) stay and mechanical ventilation duration. METHODS: We conducted a systematic review and meta-analysis following PRISMA 2020 guidelines and the Joanna Briggs Institute (JBI) methodology for prevalence studies. Eligible studies included paediatric patients (≤18 years) undergoing cardiac surgery reporting postoperative dysglycaemia (hyperglycaemia or hypoglycaemia). A random-effects model was used to calculate pooled prevalence and 95%CIs, with logit transformation applied to prevalence proportions and heterogeneity assessed using I² statistics. Subgroup analyses were performed by region/continent, study design, age group and glycaemic threshold. All included studies were critically appraised using the Newcastle-Ottawa Scale for observational studies and the Cochrane Risk of Bias 2 tool for randomised trials. The protocol was registered with PROSPERO (CRD420251161893). RESULTS: 22 studies involving 3864 patients were included. The pooled prevalence of postoperative dysglycemia was 68.6% (95% CI 52.8 to 81.0), with hyperglycaemia accounting for most cases (71.2%) and hypoglycaemia observed in 8.1%. Prevalence was higher in Asia and North America (~75%) than in Europe (45.2%), though not statistically significant. Studies using lower hyperglycaemia thresholds (126-139 mg/dL) reported significantly the highest prevalence (82.3%, p<0.032). The pooled mean ICU stay among dysglycemic patients was 180.3 hours (95% CI 82.3 to 279.3), and mean mechanical ventilation duration was 94.6 hours (95% CI 5.2 to 184.1). Sensitivity analyses showed that no single study disproportionately influenced the overall pooled estimates, and no publication bias was detected. The certainty of evidence assessed using the GRADE approach was moderate for overall dysglycemia prevalence. CONCLUSION: Dysglycaemia is highly prevalent following paediatric cardiac surgery, particularly hyperglycaemia, with prolonged ICU and mechanical ventilation times observed among affected patients. These findings underscore the need for vigilant perioperative glucose monitoring, standardised definitions of dysglycaemia, and evidence-based glycaemic management protocols. Future multicentre studies are warranted to establish optimal glycaemic thresholds and evaluate targeted interventions to improve postoperative outcomes in this vulnerable population.