Background: Previous studies have documented an obesity-stroke paradox, suggesting stroke patients who are overweight or obese (measured by body mass index, BMI) have a lower risk of post-stroke mortality as comparing those with normal (or lean) BMI.
Purpose: This systematic review aims to synthesis the evidence regarding the association of BMI with mortality in stroke patients.
Method: A search was conducted using databases MEDLINE, EMBASE, Ovid Nursing Database and CINAHL Plus from conception to July 2015. In addition, a hand search was done of the references of the eligible studies. Two reviewers independently evaluated the methodological quality of the studies reviewed.
Result: Eleven studies met the inclusion criteria: eight cohort studies and four experimental studies. Seven studies examined both ischemic and hemorrhagic stroke while 5 examined ischemic stroke only. Ten studies examined all-cause mortality, and one study examined recurrence of stroke plus stroke and vascular-related mortality. Mean age of the participants ranged from 64 to 78.8. One study recruited postmenopausal women and another study used US-Japanese born men. Sample sizes per study ranged between 304 to 21,884 stroke survivors. Mortality follow-up ranged between 14 days to 16 years. Findings from four studies reported obese patients had a significantly lower risk of mortality compared to normal/lean patients; while four studies found non-significant results. However, the remaining three studies found that a significantly higher risk of mortality was reported in obese patients. In addition, five of the seven studies that had an underweight category reported that the underweight group had a significantly higher risk of mortality than normal weight patients.
Conclusion: There is not enough evidence at present to support the association of lower all-cause mortality in overweight or obese stroke patients. Based on BMI measures in the existing literature, our findings suggest that obesity is not a protective factor for stroke survivors. Our review also highlights the importance of cautious interpretation of the association between BMI and all-cause mortality, while also taking into account potential etiologies of unintentional weight loss among stroke patients.