Abstract
Background
In clinical setting, ischemic heart disease is a challenging problem in hemodialysis (HD) population. Coronary flow reserve (CFR) measured by 13 ammonia positron emitting tomography (13NH3PET) is an established and reliable modality for detecting coronary artery disease. Furthermore, some prior studies show CFR is an important and independent predictor for cardiovascular event and mortality. On the other hand, HD patients with malnutrition status have poor prognosis. We have reported about the relationship between cardiovascular events and geriatric nutrition risk index (GNRI). Now, we wonder the predictability of combination of CFR and GNRI.
Methods and result
We collected 438 consecutive HD patients who received 13NH3PET in our hospital suspected for ischemic heart disease. 29 patients were excluded due to undergoing coronary revascularization within 60 days, 103 patients were excluded due to incomplete database. In total, 306 HD patients were classified into 4 group according the median value of CFR (1.99) and GNRI (97.73); Low CFR Low GNRI group (n=77), High CFR and Low GNRI group (n=76), Low CFR High GNRI group (n=78) and High CFR High GNRI group (n=75). We collected their follow up data up to 1544 days (median 833 days) about all-cause mortality and cardiovascular (CV) mortality. Surprisingly, there is no mortality event in High CFR High GNRI group. We analyzed about all-cause mortality, CV mortality. Kaplan-Meyer analysis shows there are statistically intergroup differences in each (all-cause mortality; log rank p<0.01, CV mortality; log rank p=0.02). Furthermore, we calculated area under the curve (AUC) analysis, net reclassification improvement (NRI) and integrated discrimination improvement (IDI)m adding GNRI and CFR on conventional risk factors. There are intergroup differences for all-cause mortality in AUC [conventional risk factors, +GNRI, +GNRI+CFR; 0.70, 0.72 (p=0.29), 0.79 (p<0.01)], NRI [+GNRI; 0.32 (p=0.04), +GNRI+CFR 0.82 (p<0.01)] and IDI [+GNRI; 0.01 (p=0.05), +GNRI+CFR 0.09 (p<0.01)].
Conclusion
HD patients with low CFR and malnutrition status has statistically significant poorer prognosis comparing HD patients with high CFR and without malnutrition status. Adding combination of GNRI and CFR on conventional risk factors improves the predictability of HD population's prognosis.
Funding Acknowledgement
Type of funding source: None
INTEGRATING HYPEREMIC MYOCARDIAL BLOOD FLOW AND CORONARY FLOW RESERVE FOR PREDICTING CARDIOVASCULAR MORTALITY OUTCOMES IN PATIENTS WITH KNOWN OR SUSPECTED ISCHEMIC HEART DISEASE
