Exploration of the associations between dysphagia, liquid intake restrictions, and comorbidities in acute stroke: a retrospective chart review

Objective: To examine relationships between documented dysphagia, liquid intake restrictions, and comorbidities of aspiration pneumonia and urinary tract infection (UTI) during acute stroke hospitalization. Design: Exploratory retrospective chart review with binary logistic regression analysis. Setting: Large urban tertiary care hospital system including two Comprehensive Stroke Centers. Subjects: = 1,216) admitted with acute ischemic or hemorrhagic stroke between 2021 and 2023. Interventions: No experimental intervention. Clinical liquid intake restrictions were recorded as none, thickened liquids, or nil per os (NPO) for more than one day. Main outcome measures: Frequencies and odds ratios for comorbid diagnoses of aspiration pneumonia and UTI given documentation of dysphagia (ICD-10 R13.10), liquid intake restrictions and covariates of age, sex, stroke type, lesion location and lateralization, NIHSS score, length of stay (LOS), smoking, and dentition. Univariate analyses followed by binary logistic regression models for variables with significant univariate associations. Results: Dysphagia was documented in 31.7 percent of patients. NPO orders were documented in 23.5 percent and thickened liquids recommendations in 3.5 percent. Aspiration pneumonia and UTI occurred in 9.4 percent and 10 percent, respectively. After controlling for non-independence across several variables of interest, the final regression models showed that liquid intake restrictions were significantly associated with both comorbidities. Female sex and age ≥ 60 were also significantly associated with UTI. Conclusions: Orders for liquid intake restrictions were significantly associated with increased odds of aspiration pneumonia and UTI. The data suggest a need for improved documentation, timely swallow imaging assessments, and prospective research exploring how the timing of diagnoses, dysphagia assessment and liquid intake recommendations impact the observed relationships.