In the present study, we found extreme NT pro-BNP levels were positively correlated with creatinine levels and the evaluation of impaired renal function in elderly patients. Elevated NT pro-BNP levels are common in patients with CKD, as it showed in the presence of coronary artery disease (CAD) and left ventricular hypertrophy (LVH). Nitta and colleagues  reported that the mean serum NT pro-BNP concentration was higher in haemodialysis patients than in normal healthy subjects. Yang et al.  demonstrated that BNP levels of patients with CKD showed a positive correlation with creatinine levels, and a high concentration of BNP can be seen in patients with stage 5 CKD without heart failure. After analyzing 277 patients with CKD, Prnjavorac et al.  found that NT pro-BNP increased at the beginning of water overload in patients with CKD. However, Law et al.  demonstrated that BNP levels were not correlated with the presence of renal dysfunction in 113 patients with BNP value >3000 pg/mL. The major reason for this inconsistency maybe that patients with an age ranging from 21 to 102 were enrolled in Law’s study , which may lead to different result regarding BNP levels determined by age. Moreover, our results are similar to the study by Guglin et al. , who also found high BNP levels (4000–20,000 pg/mL) were associated with kidney dysfunction. Although the data are inconsistent, our results suggest in elderly patients, the extreme NT pro-BNP may be more useful for the assessment of the severity of impaired renal function. Certainly, further studies are still needed to ascertain the relationship between extreme BNP levels and impaired renal function in large sample size.
Previous evidence demonstrated that BNP testing is a useful tool for the diagnosis or stratification of decompensated heart failure [13, 14]. Most studies have confirmed that BNP is associated with the severity of HF in a certain range. However, a study regardless of age published in 2010 demonstrated that extreme values of BNP do not correlate with the presence of HF . Our results also showed that there was no significant difference in the distribution by NYHA class and LVEF in different NT pro-BNP levels. There are several possible reasons for this phenomenon. Firstly, immunoassays play an important role in falsely elevating BNP levels, which are used in the clinical setting and do not determine precise molecular forms of these natriuretic peptides. A review showed antibodies against BNP may also recognize and bind to pro-BNP and NT-fragments . Meanwhile, analysis from chromatography-based studies revealed that inactive pro-BNP forms often predominate in HF,. This indicates that the bioactive forms of natriuretic peptides may not be processed proportionally in patients with advanced HF. Secondly, inflammation may be associated with falsely elevating BNP levels. Because one study had demonstrated that inflammatory markers were positively correlated with BNP levels, suggesting a link between elevations of BNP levels and inflammation .
Through a systematic review of 19 studies, Doust and colleagues found that NT pro-BNP is a strong prognostic indicator for both asymptomatic patients and for patients with heart failure at all stages of disease, and for every 100 ng/L rise in NT pro-BNP concentration, there was a corresponding 35% increase in the relative risk of death . Additionally, a previous cohort study showed that NT pro-BNP was found to be a good predictor of mortality in elderly with and without specific cardiac diagnoses . Bettencourt and colleagues  also demonstrated that NT pro-BNP levels were the strongest independent predictor of death or hospital readmission during hospitalization and six months of follow-up. In the extremely elevated geriatric population above 80 years, we found higher levels of NT pro-BNP with higher mortality. These results confirmed that even with the level above 3000 pg/ml and suggested that NT pro-BNP may be also an independent predictor of death in elderly patients with or without cardiac disease.
In the early phase of AMI, NT pro-BNP gene expression increases considerably, the phenomenon also can be seen in human cardiac allograft acute rejection . Therefore, NT pro-BNP is a useful marker of reflection injury of myocardium function and prognostic of myocardial infarction . Moreover, NT pro-BNP level is similar to BNP in circulation of healthy volunteers and it proportionally increse as cardiac function deterioration [23, 24]. After compared the predictive value of NT pro-BNP and other frequently used biomarkers, Melki et al. recently found that NT pro-BNP may be the best single predictor of left ventricular function in patients with non-ST-segment elevation acute coronary syndromes . In the present study, although there was no difference in the ratio of acute myocardial infarction patients in admission in different level of NT pro-BNP, we found the weak correlation between NT pro-BNP and CK-MB, and higher levels of CK-MB accompanied with higher NT pro-BNP, suggests elevated CK-MB levels derived from AMI contribute to the elevation of NT pro-BNP.
In recent years, CRP has emerged as a possible potent risk marker for cardiovascular diseases [26, 27]. The correlation between CRP and BNP levels has also been demonstrated , which means a possible interaction between natriuretic peptides and the systemic inflammatory response. Conversely, Caroline and collegues  found that CRP levels are not independently associated with first cardiovascular events in older individuals, and increasing age seems to attenuate the association between plasma CRP levels and the risk of cardiovascular disease. We also observed the levels of CRP in the study, but didn’t find the correlation between NT pro-BNP and CRP.
In addition, multiple factors are known to influence the circulating levels of NT pro-BNP. The prevalence of possible influencing factors including gender, body mass index (BMI), anemia and medications for heart failure such as ACE inhibitors, angiotensin-receptor blockers, beta blockers [29–31], but in the present study, we didn’t find the correlation between BNP and all of them. Maybe these parameters are just confounding variables, and the underlying mechanism still needs to be further revealed.
There are several limitations in the present study. Firstly, the cutoff of NT pro-BNP for the enrolment and groups was established arbitrarily. Secondly, we only enrolled elderly patients above 80 years, and the population with the characteristics of more comorbidities and more severe conditions, that bring about a more confounding factor.