Postextubation Arterial Blood Gas to Predict Reintubation in Children With Congenital Heart Disease

Abstract

Background:

Children with cardiac disease frequently receive mechanical ventilation. Drawing an arterial blood gas (ABG) 1–2 h postextubation is common practice. Limited data exist evaluating the utility of ABGs in this context. We hypothesized that postextubation acidemia would be associated with reintubation.

Methods:

We retrospectively evaluated all patients with an ABG postextubation in the pediatric cardiac intensive care unit (PCICU) between June 2022 and December 2024. We defined acidemia as an arterial pH <7.35. Our primary outcome was reintubation within 48 h. Secondary outcomes included duration of noninvasive respiratory support (NRS) and hospital stay.

Results:

We studied 718 subjects, and 91 (13%) had acidemia. Subjects with postextubation acidemia were more likely to be reintubated (21% vs 7.3%, P < .001) and spent longer on NRS (6.7 [2.8–15.9] vs 4.6 [2.2–10.9] days, P = .036). Hospital and PCICU stay were similar. Prior to extubation, demographics, medical history, and surgical history showed no differences, except subjects with acidemia more frequently had delayed sternal closure (34% vs 20%, P = .007) and single ventricle physiology (40% vs 27%, P = .009). Pre-extubation variables were similar between groups. Subjects with acidemia had higher median pre-extubation 24-h fluid balance, higher median 48-h fluid balance, higher final extubation readiness test (ERT) breathing frequency (39 [32–47] vs 35 [26–44] breaths/min, P = .01), lower final ERT tidal volume (6.7 [5.4–7.6] vs 7.1 [6.0–8.3] mL/kg, P = .004), more frequent upper airway obstruction (26% vs 9.4%, P < .001), and required higher NRS at 24, 48, and 72 h after extubation. Logistic regression identified acidemia (odds ratio [OR] 2.8, 95% CI [1.4–5.6], P = .004), ventricular assist device placement (OR 30.2, 95% CI [2.3–396.9], P = .009), and final ERT breathing frequency (OR 1.03, 95% CI [1.01–1.06], P = .006) as factors associated with re-intubation.

Conclusions:

Eighty percent of subjects with postextubation acidemia did not require reintubation, but acidemia was associated with a 3-fold increased risk of reintubation.