Article Text
Abstract
Background Frailty predicts poor outcomes after stroke. Although obesity increases stroke risk, studies have shown better outcomes in patients with higher BMI, a phenomenon known as the obesity paradox. We investigated the impact of low BMI on various stroke subtypes.
Methods We created cohorts based on ICD-10s for varying stroke subtypes (I60-I63, I66) utilizing the US-based TriNetX database. We assessed outcomes and stroke recurrence in patients with low BMI compared to propensity-score-matched cohorts with high BMI (≥25 kg/m2 or ICD-10:E66) within 90 days.
Results After matching, our study consisted of 1,454,298 patients. There were 727,149 patients with low BMI separated into 5 cohorts (subarachnoid hemorrhage[SAH]:52,920; intracerebral hemorrhage[ICH]:77,212; subdural/extradural hemorrhage[S/EDH]:91,458; Cerebral infarction[CI]:475,156; Transient ischemic attack[TIA]:30,403) and 727,149 patients in the respective propensity-score matched cohorts with high BMI. All patients with low BMI had increased mortality and respirator dependence but fewer seizures, PE/DVT, and MIs. Patients with low BMI and SAH, S/EDH, CI, and TIA were more frequently hospitalized. Critical Care/ICU was utilized significantly more often by patients with low BMI and ICH, S/EDH, CI, and TIA. Emergency endotracheal intubation was performed significantly more often in patients with low BMI and CI or TIA. SAH was significantly more common in patients with low BMI after S/EDH. ICH was more common in patients with low BMI after ICH, S/EDH, and CI. Recurrent CI was significantly more common in patients with low BMI. CI was significantly less common in patients with low BMI after SAH, ICH, S/EDH, and TIA.
Conclusions Patients with low BMI and various stroke subtypes have higher mortality and respirator dependence, but fewer seizures, PE/DVT, and MI. The different recurrence associations of certain stroke subtypes underscore the complex interplay between BMI and stroke outcomes, emphasizing the need for tailored interventions and monitoring strategies for individuals with low BMI and stroke.
Disclosures G. Sioutas: None. O. Shekhtman: None. M. Salem: None. A. Gajjar: None. V. Srinivasan: None. J. Burkhardt: None.