A systematic review and meta-analysis of HHV-6 and mortality after hematopoietic cell transplant.

Abstract

Human herpesvirus-6B (HHV-6B) reactivation has been associated with non-relapse mortality (NRM) and overall mortality (OM) following allogeneic hematopoietic stem cell transplant (HCT). We performed a systematic review and meta-analysis to better quantify the association. Studies were included if they systematically tested a cohort of HCT recipients for HHV-6 infection or reactivation and described mortality for patients with and without HHV-6B. Random effects models were used to assess the pooled effect of HHV-6B positivity on each outcome of interest. Bayesian aggregation was additionally performed if models included 10 or fewer studies. Eight studies were included in the NRM analysis, which demonstrated a significant association between HHV-6 detection and NRM (pooled effect: 1.84; 95% CI: 1.29-2.62) without significant heterogeneity (I2 = 0.0%, p = 0.55). A Bayesian aggregation of the raw data used to construct the NRM random effects model supported these findings (95% credible interval: 0.15-1.13). Twenty-five studies were included in OM analysis, which showed a significant positive association (pooled effect: 1.37; 95% CI: 1.07-1.76), though considerable heterogeneity was observed (I2 = 36.7%, p < 0.05). HHV-6 detection is associated with NRM and OM following HCT. Randomized trials are warranted to evaluate if preventing or treating HHV-6B reactivation improves outcomes.