Pulmonary artery catheter receipt among cardiac surgery patients from the national inpatient sample (1999-2019): Prevalence, predictors and hospitalization charges

Despite limited evidence to support its efficacy, use of pulmonary artery catheter (PAC), a relatively expensive medical device, for monitoring clinical status and guiding therapeutic interventions, has become standard of care in many settings, and especially during and after cardiac surgery. We examined the prevalence and predictors of PAC use and its association with hospitalization charges among cardiac surgery patients generally and for each selected subgroup of high-risk or complex surgical procedures. We conducted an analysis on 1,442,528 records from the National Inpatient Sample (1999-2019) that included cardiac surgery patients >= 18 years of age. Subgroups were categorized based on the presence of specific disorders like tricuspid or mitral valve disease, pulmonary hypertension, heart failure, or cardiac surgery combinations. Multivariable regression models were constructed to assess predictors of PAC use as well as PAC use as a predictor of loge hospitalization charges controlling for patient and hospital characteristics. Based on International Classification of Diseases procedure codes, PAC use was prevalent among 7.15 % of cardiac surgery hospitalizations, and hospitalization charges were estimated at $191,345, with no differences according to PAC use. Overall, being female, having Charlson comorbidity index (CCI) > 0, and non-payer (versus Medicare) status were independently associated with PAC use. Among the subgroup with the selected conditions, being female, having CCI>0, and being a Medicaid (versus Medicare) recipient were independently associated with PAC use, whereas elective admission was inversely related to PAC use. Among the subgroup without the selected conditions, having a CCI >0, elective admission, and non-payer (vs. Medicare) status were independently associated with PAC use. PAC use was not independently related to hospitalization charges overall or among subgroups. In conclusion, approximately 7 % of cardiac surgery hospitalizations received a PAC, with no differences in charges according to PAC use and disparities in PAC use driven by sex, elective admission, CCI and health insurance status. Large randomized trials are required to characterize the safety, efficacy, and cost-effectiveness of PAC use among distinct groups of patients undergoing cardiac surgery.