Predictive value of the videofluoroscopic swallowing study for long-term mortality in patients with subacute stroke

To investigate the usefulness of the videofluoroscopic swallowing study (VFSS) for subacute stroke in predicting long-term all-cause mortality, including not only simple parameters obtained from VFSS results, but also recommended dietary type as an integrated parameter. This was a retrospective study of patients with subacute (<1 month) stroke at a university hospital between February 2014 and September 2019. The independent risk factors were investigated using stepwise Cox regression analysis, which increased the all-cause mortality of patients with stroke among VFSS parameters. A total of 242 patients with subacute stroke were enrolled. The significant mortality-associated factors were age, history of cancer, recommended dietary type (modified dysphagia diet; adjusted hazard ratio [HR], 6.971; P = .014; tube diet, adjusted HR: 10.169; P = .019), and Modified Barthel Index. In the subgroup survival analysis of the modified dysphagia diet group (n = 173), the parameters for fluid penetration (adjusted HR: 1.911; 95% confidence interval, 1.086-3.363; P = .025) and fluid aspiration (adjusted HR: 2.236; 95% confidence interval, 1.274-3.927; P = .005) were significantly associated with mortality. The recommended dietary type determined after VFSS in subacute stroke was a significant risk factor for all-cause mortality as an integrated parameter for dysphagia. Among the VFSS parameters, fluid penetration and aspiration were important risk factors for all-cause mortality in patients with moderate dysphagia after stroke. Therefore, it is important to classify the degree of dysphagia by performing the VFSS test in the subacute period of stroke and to determine the appropriate diet and rehabilitation intervention for mortality-related prognosis.