LAUSR

RATIONALE Acute hypercapnic respiratory failure has been shown to be associated with worse outcomes for various disease states, but less is known about patients with compensated hypercapnic respiratory failure. Although these patients have a normal pH, it remains unknown whether chronically elevated partial pressures of carbon dioxide (PaCO2), irrespective of etiology, put patients at risk of adverse events. OBJECTIVES To understand the burden of and clinical factors associated with morbidity and mortality in patients with compensated hypercapnic respiratory failure. METHODS We performed a query of the electronic medical record (EMR) to identify patients hospitalized at the University of Michigan from January 1 - December 31, 2018 who had compensated hypercapnia, using a PaCO2 ≥ 50 mmHg and pH 7.35 - 7.45 on arterial blood gas (ABG). We obtained demographic and clinical data from the EMR. Survival probabilities for PaCO2 subgroups (50.0-54.9; 55.0-64.9; ≥65.0 mmHg) were determined using the Kaplan-Meier product limit estimator. Cox proportional hazard models were constructed to test the association between PaCO2 and all-cause mortality. RESULTS We identified 491 patients with compensated hypercapnia. The mean age was 60.5 ± 16.2. Patients were 57.4% male and 86.2% white. The mean pH and PaCO2 were 7.38 ± 0.03 and 58.8 ± 9.7 mmHg respectively. There were a total of 1,030 hospitalizations, with 44.4% of patients having 2 or more admissions. The median numbers of cumulative hospital and ICU days were 21.0 (IQR 11.0-38.0) and 7.0 (IQR 3.0-14.0) respectively. 217 patients (44.2%) died over a median of 592 days. In univariate analysis, every 5-mmHg increase in PaCO2 was associated with a higher risk of all-cause death (HR 1.09; 95% CI 1.03-1.16; p=0.004). This association was maintained after adjusting for age, sex, BMI, and the Charlson comorbidity index (HR 1.09 for every 5-mmHg increase in PaCO2; 95% CI 1.02-1.16; p=0.009). There was a statistically significant interaction between PaCO2 and BMI on mortality (p= 0.01 for the interaction term). CONCLUSIONS Patients with compensated hypercapnic respiratory failure have high mortality and healthcare utilization with higher PaCO2 associated with worse survival. Obese hypercapnic patients have higher risk of death with increases in PaCO2.