Citation

Krishna V, Kim DH. J. Neurosurg. 2007; 107(3): 522-529.

Affiliation

Harvard School of Public Health, Boston, Massachusetts, 02115, USA.

Copyright

(Copyright © 2007, American Association of Neurological Surgeons)

DOI

10.3171/JNS-07/09/0522

PMID

17886550

Abstract

OBJECT: Studies on risk factors for subarachnoid hemorrhage (SAH) show heterogeneity. For example, hypertension has been found to be a significant risk factor in some studies but not in others. The authors hypothesized that differences in the ethnicity of the populations studied could account for these findings. METHODS: A metaanalysis was performed using 17 case-control and 10 cohort studies that met specified inclusion criteria. The authors used a random-effect model to calculate the pooled effect estimates for current smoking, hypertension, and alcohol consumption. A meta-regression analysis was performed using the ethnic composition of the study populations as a covariate. Studies were classified as multiethnic or monoethnic, and the pooled effect estimates were compared. RESULTS: Analysis of the cohort studies yielded a pooled effect estimate or risk ratio of 3.18 (95% confidence interval [CI] 2.37-4.26) for current smoking, 3.05 (95% CI 2.09-4.44) for hypertension, and 2.46 (95% CI 1.42-4.24) for alcohol consumption at a rate of 150 g/week or more. The results were similar for the case-control studies. For current smoking, the ethnic composition of the study population was a statistically significant predictor of heterogeneity among case-control studies (p < 0.001, even after application of the Bonferroni correction). The risk for SAH among current smokers was higher in multiethnic populations (odds ratio 3.832) than in monoethnic populations (odds ratio 2.487). CONCLUSIONS: The results of this metaanalysis suggest that differences in susceptibility to the harmful health effects of smoking may be one cause of the observed differences in SAH incidence for different ethnic groups. The role of ethnicity in risk factors for SAH should be considered in future studies.

Language: en