Objective: Stroke survivors with dysphagia face difficult decisions about surgically placed feeding tubes, typically a percutaneous endoscopic gastrostomy (PEG). We examined PEG use and outcomes of patients with and without PEG in a community setting. Methods: Ischemic stroke and intracerebral hemorrhage (ICH) stroke cases from the Brain Attack Surveillance in Corpus Christi (BASIC) project were identified (4/2016 to 12/2020). Participants (or a proxy) completed a 90-day interview to assess functional outcome (activities of daily living/instrumental activities of daily living ADL/IADL score: 22 items, range, 1–4; higher=worse) and quality of life (short form stroke specific quality of life scale, 12 items, range 0-5, higher= better). Logistic regression explored factors associated with PEG. Individuals with/without PEG were matched on demographics and clinical characteristics using a propensity score (complete case and two-to-one nearest neighbor matching). Outcomes were compared in PEG-matched groups for the baseline sample (mortality, length of stay) and 90-day survivors (functional outcome, and quality of life). Results: The sample included 2,565 patients with stroke, with 1120 completing a 90-day interview. Overall, 142/2,565 (4.8%) underwent PEG, with descriptive characteristics shown in Table 1. After adjustment, PEG use varied by sex, stroke type, severity, and acute therapies (Table 2). Results for the propensity score matched sample are in Table 3. Patients with PEG were hospitalized 2.35 times longer than those without PEG (p25%) and ninety-day (>35%) mortality were high, with no difference based on PEG use. At 1 year, mortality was higher among PEG recipients (55.4% vs. 44.6%; RR 1.29, 95% CI 1.07–1.55, P=0.007). PEG recipients in the matched sample had worse functional outcome (Median 3.73 vs. 2.91). In other words, on average, participants with PEG were only able to complete ADL/IADLs with a lot of difficulty (score=3) or with help from another person (score=4). Those receiving PEG also reported worse quality of life (Table 3) than those without PEG. Conclusions: Stroke patients receiving PEG have high short- and long-term mortality and a high risk of dependence at 90 days. Limitations include sample size and lack of data on patient/family perceptions of the PEG or functional outcomes beyond 90 days. Randomized studies of alternative strategies for decisions on artificial feeding after stroke may be warranted.
Zahuranec et al. (Thu,) studied this question.