Incidence of low-titre factor VIII inhibitors in patients with haemophilia A: meta-analysis of observational studies.

INTRODUCTION A few studies have been focused on low-titre inhibitors in patients with haemophilia A. Although several putative factors have been implicated in the development of these inhibitors, solid data are still lacking. AIM The aim of this study was to perform a proportion meta-analysis on the incidence of low-titre inhibitors in haemophilia A. METHODS We surveyed the PubMed database to identify studies on de novo development of low-titre inhibitors in haemophilia A patients. On the basis of these data, we carried out a proportion meta-analysis to summarize information on incidence and between-study variability. Furthermore, the following three covariates were assessed by meta-regression: (i) mild disease vs. severe haemophilia; (ii) status of previously untreated patient (PUP) as opposed to multi-transfused and (iii) type of factor VIII. RESULTS Our literature search on PubMed extracted 340 eligible articles. From these, we selected 33 patient cohorts that were included in our meta-analysis (19 cohorts for PUPs and 14 cohorts for multi-transfused or unselected patients). The pooled incidence of low-titre inhibitors was 10.3% (95%CI: 8.3-12.5%) for studies including PUPs and 5.8% (95%CI: 2.5-10.4%) for the other studies; the difference was statistically significant (P = 0.003). Meta-regression of 31 patient cohorts found that mild disease and type of factor VIII were not associated with an increased incidence of low-titre inhibitors. CONCLUSIONS Our results confirmed that PUPs show a higher incidence of low-titre inhibitors than the other patients. Furthermore, our data showed that mild haemophilia was not associated with an increased incidence of low-titre inhibitors.