The Effect of Stroke Severity on Timing-Related Outcomes in Gastrostomy Tube Placement

Tuesday, March 7, 2023

4:03 PM – 4:12 PM

Location: Phoenix Convention Center, 222AB

Presenting Author(s)

Jaimie Fan, AB

Medical Student
Cooper Medical School of Rowan University

Disclosure(s): No financial relationships to disclose

Author/Co-author(s)

KH

Krystal Hunter, PhD, MBA

Biostatistician
Cooper University Hospital
SA

Sabina Amin, MD

Interventional Radiologist
Cooper University Hospital

Disclosure(s):

Jaimie Fan, AB: No financial relationships to disclose

Purpose: To determine whether stroke severity mediates outcomes in early versus late gastrostomy tube placement, with outcomes including length-of-stay, in-hospital mortality, degree of disability, and discharge disposition.

Materials and Methods: This retrospective cohort study utilized 2002-2019 data from the National Inpatient Sample (NIS), an all-payer administrative healthcare database. 3,723 patients admitted for acute ischemic stroke and who received a gastrostomy tube during the same hospital admission were identified by ICD-9 and ICD-10 codes. Two cohorts (early versus late placement) were defined as the following: patients who received g-tube placement within 7 days of hospital admission (n = 1,763) (early) versus placement after 7 days of admission (n = 1,960) (late). Outcomes, including length-of-stay, in-hospital mortality, and discharge disposition were determined based on pre-existing variables in the NIS database. The degree of disability following stroke was determined by the administrative stroke outcome variable (ASOV), a validated proxy for the modified Rankin Scale. First, a comparison of these four outcome variables between the two primary cohorts (early vs late g-tube placement), was conducted. The cohorts were then stratified by stroke severity (mild/moderate vs severe) using the Administrative Data Stroke Scale (ADSS), a validated proxy of the National Institutes of Health Stroke Scale (with cutoff of ADSS = 12.5 corresponding with NIHSS = 15).

Results:

Initial comparison of early vs late g-tube placement cohorts revealed significant differences in favor of early g-tube placement, including a shorter length of hospital stay (p < 0.001), more frequent discharge to home (p = 0.008), and decreased mortality (p = 0.035). However, following stratification by stroke severity, only patients with severe stroke (corresponding with NIHSS > 15) show significant differences in the following outcomes in favor of early g-tube placement: the percentage of patients discharged to home (p = 0.005) and disability outcome (ASOV) (p = 0.024).

Conclusion: Early g-tube placement (within 7 days of hospital admission) is associated with more favorable disposition and disability outcomes in patients with severe stroke (ASOV >12.5; NIHSS > 15), but not in patients with mild or moderate stroke (ASOV < or = 12.5; NIHSS < or = 15).