Personal use of hair dyes and risk of leukemia: a systematic literature review and meta-analysis

Abstract

The objective of this study was to examine the association between personal use of hair dyes and the risk of leukemia. We conducted a systematic literature review of epidemiology studies reporting leukemia-specific cancer risks among hair dye users, and estimated the meta-relative risk (meta-RR) and corresponding 95% confidence interval (95% CI) of leukemia, comparing hair dye users to nonusers. When data from all 20 studies that met the inclusion criteria were combined, ever use of hair dye was associated with a nonstatistically significant increased risk of leukemia, meta-RR = 1.09 (95% CI: 0.97–1.22). When restricted to studies that adjusted for smoking, ever use of hair dye was not associated with leukemia, meta-RR = 0.99 (95% CI: 0.76–1.29). A statistically significant increased risk of leukemia was associated with permanent hair dye use (meta-RR = 1.19 [95% CI: 1.07–1.33]), dark hair dye use (meta-RR = 1.29 [95% CI: 1.11–1.50]), hair dye use among males (meta-RR = 1.42 [95% CI: 1.01–2.00]), hair dye use pre-1980 (meta-RR = 1.49 [95% CI: 1.21–1.83]), and hair dye use for ≥15 years (meta-RR = 1.35 [95% CI: 1.13–1.62]). Overall, findings suggest that ever use of hair dye is not a significant risk factor for leukemia. Certain hair dye use characteristics were associated with a statistically significant increased risk, but further research is required to determine whether these associations truly reflect a risk of leukemia due to methodological limitations in the underlying studies.

The purpose of this study was to synthesize and examine the evidence on the relative risk of leukemia among hair dye users under various exposure scenarios. Findings suggest that personal hair dye use is not a significant risk factor for leukemia when data from all studies were combined. Upon stratification, permanent hair dye use, dark hair dye use, hair dye use pre-1980, hair dye use among males, and hair dye use for ≥15 years were associated with a statistically significant increased risk of leukemia. Further research is required to determine whether these associations truly reflect a risk of leukemia due to methodological limitations in the underlying studies.

http://ift.tt/2xdRYdj