Impact of Kidney Disease in Patients Undergoing Catheter Directed Interventions for Intermediate to High-Risk Pulmonary Embolism.

Background: Catheter-directed interventions (CDIs) for pulmonary embolism (PE) continue to evolve. However, due to the paucity of data, their use has been limited in patients with underlying kidney disease.

Methods: The National Readmission Database (2016-2020) was utilized to identify intermediate to high-risk PE (IHR-PE) patients requiring CDI (thrombectomy, thrombolysis, and ultrasound-assisted thrombolysis). Cohorts were stratified based on the presence of CKD stage ≥3, including ESRD. A Propensity Score Matching (PSM) model was applied to compare outcomes.

Results: From 2016-2020, 20795 patients with IHR-PE underwent CDIs. Most were done in the non-CKD/ESRD population (N:18438, 88.7%), while only 2357 (11.3%) were done in the CKD/ESRD population. After propensity matching, the CKD/ESRD population had higher adverse events, including mortality (7.3% vs. 5.1%, p: 0.036), need for transfusions (52.6% vs. 44.7%, p<0.001), and acute bleeding (15.4% vs. 10.6%, p<0.001). CKD/ESRD population had a higher median LOS (5 vs. 4 days, p<0.001) and total cost ($32935 vs. $29805, p<0.001) in the index admission. Over the study period, total cost decreased in the CKD/ESRD population ($37829 to $31436, p-trend: 0.024) but remained the same in the non-CKD/ESRD population (p-trend>0.05). 180-day readmission rates were higher in the CKD/ESRD population (24.7% vs. 17.5%, p: 0.006). Our subgroup analysis, excluding ESRD patients, showed no significant difference in in-hospital mortality (6.5% vs. 7.3%, p>0.05), but the rates of thoracic or respiratory bleeding (4.5% vs. 2.6%, p:0.012), need for transfusions (52.4% vs.. 43.5%, p<0.001), and AKI (57.1% vs. 23.2%, p<0.001) were higher in patients with CKD undergoing CDIs for IHF-PE.

Conclusion: CKD/ESRD patients requiring catheter-directed interventions for IHR-PE had higher periprocedural mortality and acute bleeding. The presence of ESRD mainly drove periprocedural mortality in our study, while the presence of non-dialyzed CKD was associated with higher rates of non-fatal localized hemorrhage.