Prognostic value of circulating chromogranin A levels in acute coronary syndrome.

In their letter to the editor, Jeske et al. make some comments relevant to our article entitled: ‘Prognostic value of circulating chromogranin A level in acute coronary syndromes’ recently published in the European Heart Journal.[1] First, the observation that chromogranin A (CgA) levels in our patients were only moderately elevated compared with levels in the normal population is consistent with prior findings in other cohorts of patients with acute coronary syndromes (ACS).[2,3] Despite of this, even only mildly to moderately elevated CgA levels discriminate well between patients with a favourable and poor prognosis in various types of heart disease,[2-4] suggesting that CgA elevation reflects detrimental processes specific to the subgroup of patients with acute myocardial ischaemia and a poor prognosis, and not processes found in ACS patients in general. Of note, fourth quartile CgA levels (>33.7 U/L) in our patients were clearly increased compared with reference values (<18 U/L), demonstrating that a proportion of ACS patients has clearly elevated CgA levels. As these patients also had the highest number of events during follow-up, CgA seems to identify the subgroup of ACS patients with an unfavourable prognosis.

Jeske et al. also raise an important question regarding the impact of comorbidities on CgA levels in patients with heart disease. We clearly show that CgA levels in our patients are influenced by a number of factors, including reduced renal function. However, by multivariate Cox proportional hazard regression analysis used in our study, we adjusted for these confounders and found that CgA levels provided independent prognostic information to conventional risk markers. We acknowledge that we lack information on some of the conditions mentioned by Jeske et al., including gastric inflammation by H. pylori, atrophic gastritis, and benign hypertrophy of the prostate, but we believe it is unlikely that these conditions had a major impact on mortality or the secondary endpoints in our study. In fact, we believe that the presence of these factors would have tended to obscure the association between CgA levels and events in our study. Similarly, the influence on CgA levels by medications generally not considered harmful to patients with heart disease would also be expected to attenuate the association between CgA levels and outcome in our patients. Thus, the major novel finding of our study is that CgA levels, although being influenced by comorbidities and medications, still provided independent prognostic information in patients with ACS. More importantly, CgA also provided incremental information to the information obtained by estimating left ventricular ejection fraction, and measuring the contemporary cardiac biomarkers, troponin T, and pro-BNP.

Finally, as is the case for all biochemical markers, analytical issues must be taken into account when evaluating CgA measurements.[5] However, as all our measurements were performed in serum and with the same method, this should not be a problem in our study.

Funding

Funding to pay the Open Access publication charges for this article was provided by Karolinska Institutet, Stockholm, Sweden.