This study aimed to investigate hypoalbuminemia-related factors in patients with stroke in the neurocritical care unit (NCU) and to establish a nomogram model for hypoalbuminemia prediction. Data from 902 patients hospitalized for stroke in the NCU who had normal albumin levels on admission from March 2018 to December 2020 were analyzed. Logistic regression was used to identify the risk factors associated with hypoalbuminemia. A novel nomogram prediction model for hypoalbuminemia was constructed, and it was validated in an independent set (n = 233). The nomogram incorporated seven risk factors, including age, high comorbidity, mechanical ventilation, feeding routes, total protein, high-sensitivity C-reactive protein, and fibrinogen levels, and its area under the curve for predicting hypoalbuminemia in the developmental set and validation set were 0.856 and 0.870, respectively. The novel nomogram facilitates risk prediction of hypoalbuminemia among patients with stroke in the NCU, which might be useful for selecting treatment strategies and management.
Keywords: hypoalbuminemia; nomogram; risk factors; stroke.