PREVENTING READMISSIONS FOR SEVERE COPD EXACERBATIONS: DOES HYPERCAPNIA MATTER?

TOPIC: Obstructive Lung Diseases TYPE: Original Investigations PURPOSE: Acute exacerbation of chronic obstructive pulmonary disease (AECOPD) is responsible for substantial hospitalization costs. This is driven not just by the index hospitalization, but also high rates of readmission frequently seen in such patients. There is little data to determine target pCO2 at discharge to prevent future hospitalizations. We aimed to identify if the arterial pCO2 level prior to discharge was associated with time to readmission for patients admitted to the medical intensive care unit (MICU) with AECOPD. METHODS: We performed a retrospective cohort study of patients 18 years and older admitted to the MICU with an attending physician diagnosis of AECOPD between Jan 2011 and Jan 2018. Patients had to have at least 3 years of follow up within our health system after index admission. Patients were excluded if the primary reason for MICU admission was not AECOPD. Charts were obtained by searching the ICD-9 and ICD-10 billing codes for AECOPD. Data were collected on demographics, time to readmission, ABG results including on MICU admission, highest pCO2, and pCO2 on hospital discharge. RESULTS: 214 patient charts met the criteria for the queried ICD-9 and ICD-10 codes. 72 charts were excluded because they were not admitted primarily for AECOPD. 85 of the remaining charts were excluded because they were not admitted to the MICU. 31 charts were excluded because they did not have 3 years of follow up. 2 charts were excluded due to death on index admission and 1 chart was excluded due to discharge to home hospice. Of the 23 remaining patients in the analysis, 9 patients in total were readmitted during follow up. 16 patients (70%) were male. The average age was 64 years. 83% were African American, 13% were Hispanic, and 4% were white. 78% were active smokers. Survival analysis by Kaplan-Meier graphing showed that the time to readmission curves for patients whose pCO2 on discharge was above vs below 60mmHg were non-overlapping. However, the log-rank and Wilcoxon p-values were 0.38 and 0.33 respectively. Similar analyses using pCO2 values from ABG on MICU admission and highest pCO2 showed no statistical significance. CONCLUSIONS: Although the Kaplan-Meier survival curves for patients whose pCO2 at discharge that were above and below 60 were non-overlapping, log-rank and Wilcoxon testing showed no statistical significance for time to readmission in patients with MICU admission for AECOPD. A limitation of this study was that despite the 6-year enrollment period, a low number of patients were found to have AECOPD requiring MICU admission. This relatively low number may have contributed to underpowering and the negative result of this study. CLINICAL IMPLICATIONS: Our data suggest that it may be prudent to treat hypercapnic respiratory failure in patients admitted with AECOPD until the arterial pCO2 levels are under 60mmHg to limit time to readmission. Further study needs to be looked into this claim. DISCLOSURES: No relevant relationships by Richard Lenhardt, source=Web Response no disclosure on file for Avantika Nathani; No relevant relationships by Agata Parfieniuk, source=Web Response No relevant relationships by Jay Pescatore, source=Web Response No relevant relationships by Sairam Raghavan, source=Web Response No relevant relationships by Jean Luis Reinoso Abreu, source=Web Response