Background The Prognostic Nutritional Index (PNI) has been reported to be correlated with long-term outcomes after gastrointestinal tumor surgery. However, to our knowledge, only a few studies have shown that the PNI is related to cardiovascular diseases. Therefore, we aimed to assess the association between the PNI and long-term outcomes in patients with coronary artery disease (CAD) undergoing percutaneous coronary intervention (PCI). Methods This was retrospective observational study. A total of 3561 patients with CAD after PCI were retrospectively enrolled in the CORFCHD-ZZ study from January 2013 to December 2017. The patients (3519) were divided into three groups according to PNI tertiles: the first tertile (PNI < 47.12, n = 1173), the second tertile (47.12 ≤ PNI < 51.50, n = 1185), and the third tertile (PNI ≥ 51.50, n = 1161). The mean follow-up time was 37.59 ± 22.24 months. The primary endpoint long-term mortality, including all-cause mortality (ACM) and cardiac mortality (CM).Secondary endpoints were major adverse cardiovascular events (MACEs) and major adverse cardiovascular and cerebrovascular events (MACCEs). Result In our study, the incidences of ACM in the first, second, and third tertiles were 3.8%, 1.8% and 1.4%, respectively ( P < 0.001). The incidences of CM occurring in the first, second, and third tertiles were 1.7%, 3.1% and 2.1%, respectively ( P < 0.001).There was statistically significant different in primary endpoints incidence. MACEs occurred in 139 patients (11.8%) in the first tertile, 121 patients(11.1%) in the second tertile and 123 patients(10.8%) in the third tertile( P = 0.691). MACCEs occurred in 183 patients (15.6%) in the first tertile, 174 patients(14.7%) in the second tertile and 160 patients(13.85%) in the third tertile( P = 0.463).There was no statistically significant different in secondary endpoints incidence. Kaplan–Meier analyses showed that elevated PNI was significantly related to long-term CM (log rank, P < 0.001) and long-term ACM (log-rank, P < 0.001). Cox regression analyses suggested that compared with the patients in the first tertile, the risk of ACM was decreased to 60.9% (HR = 0.609, 95% CI: 0.398–0.932, P = 0.029) in the second tertile and 40.3%(HR = 0.403, 95% CI: 0.279–0.766, P = 0.003) in the third tertile, while the risk of CM was decreased to 58.8%(HR = 0.588, 95% CI: 0.321–0.969, P = 0.038) in the second tertile and 46.6%(HR = 0.466, 95% CI: 0.250–0.870, P = 0.017) in the third tertile. Multivariate Cox regression analyses showed that the PNI was an independent predictor of long-term ACM and CM. Conclusion Our finding shown that PNI is an independent predictor in CAD patients after PCI,the higher the PNI, the less occurring adverse event. Therefore,PNI may be an new biomarker to predict long-term outcome of CAD patients after PCI.