Association between HTLV-1 infection and adverse health outcomes: a systematic review and meta-analysis of epidemiological studies

Background: Human T-cell lymphotropic virus type 1 (HTLV-1) is a human retrovirus that causes a lifelong infection. Several diseases, including an aggressive form of leukaemia, have been designated as associated with HTLV-1, whereby having HTLV-1 is a necessary condition for diagnosis. Beyond these diseases, there is uncertainty about other health effects of HTLV-1. We aimed to synthesise evidence from epidemiological studies on associations between health outcomes and HTLV-1.

Methods: For this systematic review and meta-analysis, we searched Embase, MEDLINE, MEDLINE In-Process, and Global Health for publications from their inception to July, 2018. We included cohort, case-control, and controlled cross-sectional studies that compared mortality or morbidity between people with and without HTLV-1. We excluded studies of psychiatric conditions, of symptoms or clinical findings only, of people who had undergone blood transfusion or organ transplant, and of population groups defined by a behavioural characteristic putting them at increased risk of co-infection with another virus. We extracted the risk estimates (relative risks [RRs] or odds ratios [ORs]) that reflected the greatest degree of control for potential confounders. We did a random-effects meta-analysis for groups of effect estimates where case ascertainment methods, age groups, and confounders were similar, presenting pooled estimates with 95% CIs and prediction intervals.

Findings: Of the 3318 identified studies, 39 met the inclusion criteria, examining 42 clinical conditions between them. The adjusted risk of death due to any cause was higher in people with HTLV-1 when compared with HTLV-1-negative counterparts (RR 1·57, 95% CI 1·37-1·80). From meta-analysis, HTLV-1 was associated with increased odds of seborrheic dermatitis (OR 3·95, 95% CI 1·99-7·81), Sjogren's syndrome (3·25, 1·85-5·70), and, inversely, with lower relative risk of gastric cancer (RR 0·45, 0·28-0·71). There were a further 14 diseases with significant associations or substantially elevated risk with HTLV-1 from single studies (eczema [children]; bronchiectasis, bronchitis and bronchiolitis [analysed together]; asthma [males]; fibromyalgia; rheumatoid arthritis; arthritis; tuberculosis; kidney and bladder infections; dermatophytosis; community acquired pneumonia; strongyloides hyperinfection syndrome; liver cancer; lymphoma other than adult T-cell leukaemia-lymphoma; and cervical cancer).

Interpretation: There is a broad range of diseases studied in association with HTLV-1. However, the elevated risk for death among people with HTLV-1 is not explained by available studies of morbidity. Many of the diseases shown to be associated with HTLV-1 are not fatal, and those that are (eg, leukaemia) occur too rarely to account for the observed mortality effect. There are substantial research gaps in relation to HTLV-1 and cardiovascular, cerebrovascular, and metabolic disease. The burden of disease associated with the virus might be broader than generally recognised.

Funding: Commonwealth Department of Health, Australia.