Vikas Singh, Nileshkumar J. Patel, Alex P. Rodriguez, Ghanshyam Shantha, Shilpkumar Arora, Abhishek Deshmukh, Mauricio G. Cohen, Cindy Grines, Eduardo De Marchena, Apurva Badheka, Abhijit Ghatak

The objective of our study was to assess patients with end-stage liver disease undergoing percutaneous coronary intervention (PCI) and determine the rates and trend of complications and in-hospital outcomes. Data were obtained from the Nationwide Inpatient Sample 2005 to 2012. We identified all PCIs performed in patients with diagnosis of cirrhosis during the study period by the International Classification of Diseases, Ninth Revision, Clinical Modification codes. Preventable procedural complications were identified by Patient Safety Indicators. Propensity scoring method was used to establish matched cohorts to control for imbalances and account for differences that may have influenced treatment outcomes. A total of 1,051,242 PCIs were performed during the study period, of these, 122,342 were done on subjects with a formal diagnosis of cirrhosis. Bare-metal stents (BMS) were more likely to be used in patients who presented with ST-elevation myocardial infarction (19.73 vs 13.58, p <0.001), in cardiogenic shock (5.58, vs 2.81, p <0.001), or required intraaortic balloon pump (4.73 vs 2.38, p <0.001). The overall rate of complications was 7.1%, whereas the overall mortality rate over these years was 3.63%. On a propensity-matched analysis the mortality rate was 2 times higher for BMS (5.18 vs 2.35, p <0.001) compared with drug-eluting stents. PCI remains a safe and plausible option for patients with cirrhosis albeit riskier than for the general population. The use of BMS is associated with increased mortality and bleeding complications compared with drug-eluting stents which likely is representative of preferential use of BMS in patients with more advanced end-stage liver disease who are also likely to experience higher postprocedural complications.