Causes of hypercapnic respiratory failure and associated in-hospital mortality

Abstract

Background and Objective: Hypercapnic respiratory failure (HRF) can occur due to severe respiratory disease but also because of multiple coexistent causes. There are few data on the prevalence of antecedent causes for HRF and the effect of these causes on prognosis, especially where study inclusion has not been biased with respect to primary diagnosis, interventions received or clinical outcome. We sought to determine the prevalence of pre-specified conditions among patients with HRF and to determine the effect of these causes on in-hospital mortality. Methods: Cross-sectional study of patients with HRF from 2013 to 2017. Inclusion criteria were PaCO2 >45 mm Hg and pH ≤7.45. Causes of interest were identified using diagnosis codes from hospital records. We used directed acyclic graphs to inform logistic regression models for the outcome of in-hospital death. Results: We identified 873 persons with HRF in the study period. Mean (SD) age was 69 years and 50.4% were males. Acidosis (pH <7.35) was present in 488 (55%) cases. Most (83%) had one or more of the following: obstructive lung disease, lower respiratory tract infection, congestive cardiac failure, sleep disordered breathing, neuromuscular disease, opioid or benzodiazepine use. In-hospital mortality was 12.8%. Obstructive lung disease and cardiac failure were associated with a lower risk of death, whereas respiratory tract infection and neuromuscular disease were associated with increased risk of death. Conclusion: HRF is associated with a range of potentially causative conditions, which significantly impact hospital survival. Systematic evaluation of patients with HRF may increase detection of treatable comorbidities.