Prognostic impacts of nutritional status on long-term outcome in patients with acute myocardial infarction

Abstract

Patients with acute myocardial infarction (AMI) still remain at high risk of morbidity and mortality worldwide in spite of recent advances in percutaneous coronary intervention (PCI) techniques and devices. Therefore, risk stratification at the time of hospitalisation is important to predict outcomes in these patients. Several nutritional indices, such as the geriatric nutritional risk index (GNRI) and the prognostic nutritional index (PNI), have been used in various fields. Malnutrition has been recognised as an independent predictor of adverse prognosis in patients with heart failure. However, the association between nutritional status and long-term outcome in patients with AMI remains unclear. Thus, the aim of the present study was to investigate the prognostic impact of nutritional status assessed by the GNRI and PNI on long-term outcome among patients with AMI. This was a prospective observational study that enrolled 552 consecutive patients with AMI, who had been admitted to, and discharged from, Fukushima Medical University between January 2010 and September 2018. The diagnosis of AMI was based on the World Health Organization multinational monitoring of trends and determinants of cardiovascular disease (MONICA) criteria. Patients were registered within 72 hours after the onset of symptoms if their level of creatine phosphokinase increased to more than twice the normal range, as previously reported. We evaluated nutritional status at hospital discharge using the GNRI and PNI scores as previously described. Namely, lower GNRI or PNI indicate poor nutritional status. We divided these patients into three groups based on each nutritional index: GNRI (low, GNRI <92; intermediate, 92 GNRI< 98; and high, GNRI 98) and PNI (low, PNI< 35; 35 PNI <38; and PNI 38). The patients were followed up until 2019 for all-cause death. The status and/or dates of death of all patients were obtained from the patients’ medical records, attending physicians at the patients’ referring hospital, or by contacting patients by telephone. We were able to follow up on all patients. Survival time was calculated from the date of hospitalisation until the date of death or last follow-up. Written informed consent was obtained from all study subjects at discharge. The study protocol was approved by the ethics committee of Fukushima Medical University, and was carried out in accordance with the principles outlined in the Declaration of Helsinki. During the follow-up period (mean 1424 days), there were 88 all-cause deaths, including 24 cardiac deaths. In the Kaplan–Meier analysis, as shown in Figure 1, all-cause mortality progressively increased from the high-score group to the intermediate and low-score groups (P< 0.001). After adjusting for other potential confounding factors, the Cox proportional hazard analysis revealed that low GNRI (vs. high GNRI) and low PNI (vs. high PNI) were independent predictors of all-cause mortality in patients with AMI (hazard ratio (HR) 2.732, 95% confidence interval (CI) 1.487–5.019, P1⁄4 0.001; HR 3.361, 95% CI 1.759–6.422, P< 0.001, respectively). In addition, receiver operating characteristic curve analysis (Figure 2) demonstrated that a GNRI cut-off value of 92.7 predicated one-year mortality with a sensitivity of 75% and a specificity of 72% (area under the curve (AUC) 0.80, 95% CI 0.70–0.89, P< 0.001) and a PNI cut-off value of 42.5 predicated one-year mortality with a sensitivity of 63% and a specificity of 69% (AUC 0.68, 95% CI 0.57–0.78, P1⁄4 0.004). The analysis also showed that the