The influence of renal function on the prognostic value of Nt-pro brain natriuretic peptide in St-elevation myocardial infarction

Abstract

In patients with ST elevation myocardial infarction (STEMI) NT-pro brain natriuretic peptide (NT-pro BNP) [1,2] and estimated glomerular filtration rate (eGFR) [3] were associated with increased risk of mortality. Previous studies examined the association between renal function and NT-pro BNP, with apparently discrepant results [4–7]. In patients undergoing non vascular surgery [8] the discriminative value of NT-pro BNPwasmostpronounced inpatientswithGFRN90 ml/min/1.73 m2and had no prognostic value in patients with GFR b30 ml/min/1.73 m2. No data are so far available on this topic in STEMI patients. Therefore, the present investigation, performed in 646 consecutive STEMI patients all submitted to mechanical revascularization, was aimed at assessing: a) the influence of eGFR on the predictive value of NT-pro BNP and b) the utility of combining eGFR and NT-pro BNP to stratify risk of mortality at short and long term. From the 1st of January 2007 to 30th of June 2010, 646 consecutive patients with STEMI (within 12 hours from symptoms' onset), were admitted to our intensive cardiac care unit (ICCU), after primary PCI [8–16]. Wemeasured glucose (mg/dl), insulin values (mIU/l) [17], glycated hemoglobin (HbA1c, %), troponin I (ng/ml), C-reactive protein (mg/ dl), NT-pro brain natriuretic peptide (NT-pro BNP) (pg/ml) [13] and uric acid (mg/dl) [18]. Acute insulin resistance was measured by means of homeostatic model assessment (HOMA) [19–21]. Worsening in renal function (WRF) was defined as an increase in creatinine ≥0.3 mg/dl, as previously described [22–24]. The study protocol was in accordance with the Declaration of Helsinki and approved by the local Ethics Committee. Informed consent was obtained in all patients before enrollment. [25]. Categorical data are reported as frequencies (percentages) and analyzed by means of χ (or Fisher's exact text, when predicted counts in almost one cell were less than 5). Continuous data are reported as means±SD or medians (95% confidence interval—CI) according to their normal or not normal distribution (assessed by means of Kolmogorov– Smirnovone-sample normality test) andanalyzed bymeans of Student's t-test or Mann–Whitney U-test, respectively. Correlations between Log NT-pro BNP levels and variables (age, log BMI, admission systolic blood pressure, admission LVEF, admission eGFR, admission glycemia, Log TnI, uric acid, Log CRP, gender) were tested by Pearson correlation. A multivariate linear regression analysis was performed to identify factors that were associated with levels of NT-pro BNP for the whole study population, and in stratified analysis according to eGFR. A multivariable logistic regression analysis was carried out considering as outcome in-ICCU death. Model calibrationwas assessed by means of Hosmer–Lemeshow goodness-of-fit test; Nagelkerke pseudo-R square is also reported. To predict long-term mortality we performed Kaplan– Meier event curves, stratified according to eGFR and NT-pro BNP, about clinical cut-off values (60 ml/min/1.73 m,1000 pg/ml, respectively) and in paired combinations. Long time survival was explored, after proportionality of risk assessment, with multivariable Cox regression analysis. In all multivariable analyses, candidate variables were chosen as those considered clinically relevant or that showed a univariate relationship with outcome; nonsignificant ones were dropped by means of backward selection. A two-tailed p value b0.05 was considered statistically significant (PASW 17.0 statistical package, SPSS Inc, Chicago, IL). In our series, 129 patients (20%) exhibited an eGFR b60 ml/min/ 1.73 m (Group B). Group A was younger (pb0.001) and with a lower incidence of diabetes (p=0.002), previous PCI (pb0.001) and previous MI (pb0.001). Higher Killip classes were more frequent in Group B (pb0.001) who showed a lower admission and discharge LVEF (pb0.001 and pb0.001, respectively). A higher mortality rate both in-ICCU and at follow-up was observed in Group B (pb0.001 and pb0.001, respectively). When combining eGFR and NT-pro BNP (Fig. 1), the highest mortality was observed in patients with eGFR b60 ml/min/1.73 m and NT-pro BNP ≥1000 pg/ml (pb0.001). A higher incidence of more advanced coronary artery disease and PCI failure was observed in Group B (p=0.044 and p=0.039, respectively).