Treatment and mortality of hemophagocytic lymphohistiocytosis in critically ill children: A systematic review and meta-analysis.

OBJECTIVE

Risk factors of mortality in critically ill children with hemophagocytic lymphohistiocytosis (HLH) are not well described. This systematic review aims to determine overall mortality of critically ill children with HLH, and describes etiologies, treatment, and pediatric intensive care unit (PICU) support employed.

DATA SOURCES: PubMed, Embase, Web of Science, CINAHL, and Cochrane Library from inception until February 28, 2022.

STUDY SELECTION: Observational studies and randomized controlled trials reporting children aged 18 years or below, diagnosed with HLH and admitted to the PICU.

DATA EXTRACTION: Etiologies, treatment modalities, PICU therapies, and mortality outcomes were summarized. Random-effects meta-analysis was performed.

DATA SYNTHESIS: Total 36 studies (total patients = 493, mean age: 49.5 months [95% confidence interval (CI): 30.9-79.5]) were included. Pooled mortality rate was 32.6% (95% CI: 23.4-42.4). The most frequent etiologies for HLH were infections (53.3%) and primary HLH (12.8%), while the remaining cases were due to other causes of secondary HLH, including autoimmune diseases, malignancy, and drug-induced and idiopathic HLH. Pooled mortality rate was higher in primary than secondary HLH (72.2%, 95% CI: 57.8-84.5 vs. 23.9%, 95% CI: 14.4-35.02; p < .01). Univariate analysis found that treatment with etoposide was associated with higher mortality, while intravenous immunoglobulins (IVIGs) were associated with lower mortality. Conversely, multivariable analysis adjusted for etiology demonstrated no association between etoposide and IVIG use, and mortality. Twenty-one studies (total patients = 278) had detailed information on PICU therapies. Mechanical ventilation (MV), continuous renal replacement therapy, and inotropes were used in 107 (38.5%), 66 (23.7%), and 51 patients (18.3%), respectively. Need for MV was associated with increased risk of mortality (mean difference = 28%, 95% CI: 9-47).

CONCLUSION

Critically ill children with HLH have high mortality rates and require substantial PICU support. Collaborative work between multiple centers with standardized data collection can potentially provide more robust data.