Outcome in Children Admitted to the First PICU in Malawi*

**Background:** Over the past decades, childhood mortality has decreased in low- and middle-income countries (LMICs), yet millions of children still die from preventable causes. Pediatric intensive care units (PICUs) are beginning to emerge in sub-Saharan Africa, but data on outcomes and determinants of mortality in these settings are scarce. The opening of the first PICU in Malawi in 2017 provided an opportunity to study outcomes and identify factors associated with mortality to guide resource allocation and improve care.

**Methods:** This was a review of a prospectively constructed PICU database at Queen Elizabeth Hospital in Blantyre, Malawi, a tertiary government hospital serving 5.5 million people. All children admitted to the six-bed PICU between August 1, 2017, and July 31, 2019, were included, except those admitted as part of a clinical trial. Data on demographics, diagnoses, clinical examination findings, laboratory results (blood gas, full blood count, blood cultures), interventions, and outcomes were collected. The primary outcome was death in PICU. Univariate analysis assessed associations between variables and mortality; variables with p<0.1 were entered into two multivariable logistic regression models (one clinical, one clinical plus laboratory). Stratified analyses were performed for neonates (<28 days) and older children.

**Key Results:** Of 531 included admissions, 149 children died (28.1%). Neonates comprised 31.4% of admissions (167/531) and had significantly higher mortality (88/167; 52.7%) than older children (61/364; 16.8%; p≤0.001). Surgical admissions accounted for 84% of cases. The most common diagnoses were gastroschisis (13.3%), foreign body aspiration (7.7%), and sepsis (7.2%). On univariate analysis, gastroschisis (OR 3.6; 95% CI 2.2–6.0), trachea-esophageal fistula (OR 4.3; 95% CI 1.8–10.2), and sepsis (OR 3.5; 95% CI 1.8–6.9) were associated with higher mortality, while Wilms tumor (OR 0.1; 95% CI 0.0–0.7) was associated with lower mortality. In the final multivariable model (clinical and laboratory), independent predictors of mortality were: neonatal age (AOR 4.0; 95% CI 2.0–8.3), decreased mental state (AOR 5.8; 95% CI 2.4–13.8), post-cardiac arrest (AOR 2.9; 95% CI 1.0–8.0), severe hypotension (AOR 6.3; 95% CI 2.0–19.1), lactate >5 mmol/L (AOR 4.2; 95% CI 1.5–11.2), pH <7.2 (AOR 3.1; 95% CI 1.2–8.0), and platelets <150×10^9/L (AOR 2.4; 95% CI 1.1–5.2). In neonates, only severe hypotension (OR 4.9; 95% CI 1.3–18.9), acidosis (OR 5.2; 95% CI 1.4–18.5), and lactatemia (OR 4.8; 95% CI 1.3–18.3) were associated with mortality on univariate analysis. Median length of stay was 3 days (IQR 1–5 days).

**Clinical Implications:** The 28.1% mortality rate is substantially higher than the 2–3.5% reported in high-income country PICUs but consistent with other new PICUs in sub-Saharan Africa (e.g., 25% in Mozambique, 37.5% in Kenya). Older surgical patients (e.g., oncological surgery, foreign body aspiration) had excellent outcomes, supporting prioritization of these patients for limited PICU beds. Neonates, particularly those with congenital anomalies like gastroschisis, had poor outcomes, suggesting that neonatal surgical patients may require specialized approaches beyond standard PICU care, including parenteral nutrition and improved perioperative pathways. Sepsis was a major contributor to mortality in both medical and surgical patients, with most isolates being Gram-negative bacteria, underscoring the need for infection prevention bundles and antibiotic stewardship. The identified clinical and laboratory predictors (hypotension, decreased mental status, lactatemia, acidosis, thrombocytopenia) are consistent with established risk scores from high-income settings and can help identify children at highest risk in low-resource environments. The limited clinical signs predictive of mortality in neonates highlight the critical importance of basic laboratory tests (blood gas, lactate, platelet count) in this vulnerable group. Limitations include missing data, focus on PICU mortality only (not long-term outcomes), and potential selection bias from non-routine laboratory testing. These findings provide an evidence base for admission criteria, resource allocation, and quality improvement initiatives for emerging PICUs in sub-Saharan Africa.