Study objectives: we performed a meta-analysis to assess the usefulness of HLA testing for Narcolepsy diagnosis in four major ethnical groups: Asians, Afro-Americans, Amerindians and Caucasians. Methods: PubMed, EMBASE, Web of Science, Scopus and Cochrane databases were searched for articles in English and French published before October 2017 on HLA class II alleles in Narcolepsy. We included case-control studies, cross-sectional and retrospective cohort studies with patients diagnosed following the International classifications of sleep disorders (1990–2012) and ethnically matched controls. Following PRISMA guidelines, two investigators independently extracted data according to the inclusion criteria listed in PROSPERO CRD42017058677. A third researcher was consulted for discrepancies. We extracted and pooled adjusted OR using random-effect models. We verified the strength of the association between HLA-DQB1*06:02 and the worldwide distribution of Narcolepsy type 1 (NT1) and type 2 (NT2); furthermore, we pooled the OR measuring the association between HLA-DQB1*06:02 and NT1, NT2 and hypersomniacs. Results: We identified 511 titles. Of these, 12 case-control studies were included, for a total of 2077 NT1 patients, 235 NT2 patients, 161 hypersomniacs and 7802 controls. In the population-stratified analysis, HLA-DQB1*06:02 conferred an increased risk for NT1 (OR: 24.1, IC: 14.6–39.5, p < 0.001) and NT2 (OR: 3.9; IC: 2.2–6.8, p < 0.001). For NT1 the pooled estimated positive Likelihood Ratio (LR+) was 5.94 (IC: 3.71–9.51) and the negative Likelihood Ratio (LR-) was 0.23 (IC: 0.16–0.33); for NT2 LR+ was 3.35 (IC: 2.08–5.38) and LR- 0.72 (IC: 0.63–0.81). Moreover, for hypersomniacs LR+ was 1.436 (IC 0.668–3.089) and LR- 0.903 (IC 0.714–1.142). Conclusions: Our data support the preponderant role of HLA-DQB1*06:02 in susceptibility to NT1/NT2 across all ethnicities. HLA-DQB1*06:02 negativity should make clinicians cautious in excluding other diagnoses.
- Human Leukocyte Antigens
ASJC Scopus subject areas