Smoking Causes Fatal Subarachnoid Hemorrhage A Case-Control Study of Finnish Twins

Background and Purpose: One of the largest twin studies to date suggested that subarachnoid hemorrhage (SAH) is mainly of nongenetic origin, but the causal effect of environmental factors on SAH is yet unknown. We hypothesized that if only one of the twins experience fatal SAH, they do not share the most important environmental risk factor for SAH, namely smoking. If true, such finding would suggest that smoking causes SAH. Methods: Through the nationwide cause-of-death register, we followed 16 282 same-sex twin pairs of Finnish origin from the older Finnish Twin Cohort between 1976 and 2018 and identified all participants who died from SAH. For the baseline, we collected risk factor information about smoking, hypertension, physical activity, body mass index, alcohol consumption, and education. We classified the pairs as monozygotic, dizygotic, or of unknown zygosity. We examined the within-pair risk factor differences in the pairs discordant for SAH, that is, where one twin died from SAH and the other did not. We computed both individual (whole cohort) and pairwise (discordant pair) hazard ratios and 95% CIs. Results: During the 869 469 person-years of follow-up, we identified 116 discordant and 2 concordant (both died from SAH) twin pairs for fatal SAH. Overall, 25 of the discordant twin pairs were monozygotic. For the whole cohort, smoking (occasional/current) was associated with increased risk of SAH death (hazard ratio, 3.33 [CI, 2.24-4.95]) as compared with nonsmokers (never/former). In the pairwise analyses for discordant twin pairs, we found that the twin who smoked had an increased risk of fatal SAH (hazard ratio, 6.33 [CI, 1.87-21.4]) as compared with the nonsmoking twin. The association remained consistent regardless of the twin pairs' zygosity or sex. Conclusions: Our results provide strong evidence for a causal, rather than associative, role of smoking in SAH.