Abstract
Background
Biomarkers of myocardial necrosis may be increased in patients with chronic heart failure. We investigated whether ischaemia-modified albumin (IMA), a marker of ischaemia, is also elevated in patients with compensated heart failure, due to dilated cardiomyopathy (DCM).
Methods
We studied 42 patients with DCM and an equal number of age-matched normal volunteers. We assessed IMA serum levels with the albumin cobalt binding test.
Results
IMA was 89.9 ± 13.1 (71–117) KU/L in the patient group and 93.9 ± 9.9 (76–122) KU/L in the control group, with no significant difference between the two (P = 0.11). However, IMA differed significantly according to the New York Heart Association classification (P = 0.003) and was negatively correlated with the left ventricular ejection fraction (r = −0.40, P = 0.014).
Conclusions
We conclude that IMA, a marker of ischaemia, does not differ in patients with clinically stable DCM compared with normal subjects, but varies significantly in relation to the severity of the disease.
Introduction
Cardiac troponins (cTns) are sensitive and specific markers of myocyte injury that is mainly due to ischaemia but also inflammation, neurohormonal activation and oxidative stress of the myocardium. cTns are increased in patients with advanced heart failure in the absence of ischaemia 1 and determine prognosis in ambulatory patients with chronic heart failure (CHF), 2 in advanced CHF 1 as well as in decompensated acute heart failure. 3 Other myocardial proteins such as myosin light chain 1, heart fatty-acid binding protein and creatine kinase (CPK) MB fraction are often present in the serum of stable patients with severe heart failure and are predictors of death or hospitalization. 4 Although at present only natriuretic peptides can be used to assess the prognosis and response to treatment in heart failure, a multimarker approach may be of interest. 5 We investigated whether ischaemia-modified albumin (IMA), a biomarker of ischaemia, which increases following percutaneous coronary intervention 6–9 and in relation to acute coronary syndromes, 10–13 is also elevated in clinically stable, compensated patients with CHF due to dilated cardiomyopathy (DCM).
Methods
All study participants, patients and controls, gave written informed consent and the Ethics Committee of our Institution approved the study protocol. We studied 42 patients with established CHF due to non-ischaemic DCM; all patients were clinically stable over at least the last three months. Patients with significant anaemia (haemoglobin < 11 g/dL and haematocrit < 35%) or renal impairment (creatinine >1.8 mg/dL) and those with low albumin levels (albumin <3.5 g/dL) were excluded; IMA may be higher in anaemic patients 14 and extremely low or high albumin levels may affect IMA. Blood sampling was performed in the outpatients’ clinic during regular follow-up visits. Blood samples were assayed for N-terminal-pro b type natriuretic peptide (NT-proBNP), CPK, the MB isoenzyme of creatine kinase (CKMB), cTn-I, total protein and albumin. Our control group included 42 age-matched normal volunteers; there is evidence that gender has no significant effect on circulating IMA. 15 In our control group, we only assessed IMA serum levels. Serum IMA was measured with the albumin cobalt binding test on an Integra 800 analyzer (Rotkreuz, Roche, Switzerland), which is an indirect method of IMA measurement. Cobalt not bound to the N-terminus of albumin is detected using dithiothreitol as a colometric indicator. Blood samples were collected in serum separator tubes, centrifuged at 3000 rpm for 10 min and stored at −70°C for one month. All samples were tested in one session in triplicate and were thawed only once. According to the manufacturer, expected values determined in a population of 283 healthy individuals range from 52 to 116 KU/L with a 95th percentile at 85 KU/L. The total interassay imprecision (coefficient of variation) was 2.7–5.7% at 56.3–125.9 KU/L for quality control material. In our lab, the variability in IMA measurements was calculated in 15 serum samples as follows: three times consecutively for each sample at day 1, once at day 2 and once at day 3. The within-day coefficient of variation was 6.1% while the between-day variation was 9.22%. Reference normal values for cardiac enzymes and NT-proBNP were as follows: CPK <190 mU/mL, CKMB <3.6 ng/mL, cTn-I <0.1 ng/mL and NT-proBNP < 125 pg/mL.
We used the Mann–Whitney U test for comparison of IMA levels between the patients and the controls as well as for comparison of IMA between DCM patients with positive or negative cTn-I values; we applied the Kruskall–Wallis test to compare IMA levels according to New York Heart Association (NYHA) classification and subsequently the Wilcoxon test for paired analysis. We used Spearman's correlation to examine the relation between IMA levels and the left ventricular ejection fraction; likewise, Spearman's correlation was applied to examine the relation between IMA and albumin, NT-proBNP and cTn-I as well as between NT-proBNP and cTn-I. All statistical calculations were performed in SPSS version 14 package (SPSS Inc, Chicago, IL, USA).
Results
In the patient group, 35 were men (83.3%) and seven were women (16.6%), and their age was 46 ± 14 (17–75) years (Table 1). Nine (21.4%) were in class I in NYHA classification, 18 (42.8%) were in class II and 15 (35.7%) were in class III. The left ventricular ejection fraction was decreased, with a mean of 28 ± 2% (15–40%). Peak myocardial oxygen consumption was 24 ± 8 (10–40) mL/kg/min. All patients were on optimal medical treatment – 39 (92.8%) were on b-blockers, 31 (73.8%) were on angiotensin-converting enzyme inhibitors and seven (16.6%) on angiotensin-II receptor antagonists, 37 (88%) were on loop diuretics and 30 (71.4%) on spirinolactone, and only 11 (26.2%) were on digitalis and 10 (23.8%) on amiodarone and 22 (52.4%) on antiplatelet or anticoagulant therapy. Regarding the risk factors, nine (21.4%) patients were hypertensive, five (11.9%) were diabetic, 14 (33.3%) were smokers, eight (19%) were dyslipidaemic and eight (19%) had a family history of coronary artery disease. In the control group, 29 (69%) were men and 13 (31%) were women, with a mean age of 46 ± 1 (19–80) years.
Baseline characteristics of the patients with dilated cardiomyopathy
Data are expressed as mean ± standard deviation (min–max)
Hb, haemoglobin; Ht, haematocrit; TP, total protein
IMA was 89.9 ± 13.1 (71–117) KU/L in the patient group and 93.9 ± 9.9 (76–122) KU/L in the control group, with no significant difference between the two (P = 0.14) (Table 2). IMA serum levels differed significantly in relation to the NYHA classification (79 ± 13 KU/L for class I, 90 ± 11 KU/L for class II and 97 ± 11 KU/L for class III, P = 0.0075); post hoc paired comparison revealed a significant difference between class I versus II (P = 0.02), I versus III (P = 0.028) and II versus III (P = 0.04). Likewise, there was a significant negative correlation of IMA with the left ventricular ejection fraction (r = −0.40, P = 0.014). IMA serum levels did not differ significantly between patients with positive and negative cTn-I values (P = 0.75). There was a significant negative correlation with albumin (r = −0.71, P < 0.0001). No significant correlation was found between IMA levels and either cTn-I (r = 0.25, P = 0.24) or NT-proBNP (r = 0.29, P = 0.19). Likewise, there was no significant correlation between NT-proBNP and cTn-I (r = 0.27, P = 0.30).
Biomarkers in dilated cardiomyopathy
Data are expressed as mean ± standard deviation (min–max)
IMA, ischaemia-modified albumin; NT-proBNP, N-terminal b-type natriuretic peptide; CPK, creatine kinase; CKMB, MB isoenzyme of creatine kinase; cTn-I, cardiac troponin-I
Discussion
We found that in patients with compensated CHF due to DCM, IMA, a marker of myocardial ischaemia, in the serum does not differ compared with age-matched normal volunteers; IMA, however, differs significantly according to the NYHA classification and is negatively correlated with the left ventricular ejection fraction.
cTns, I and T, are often elevated in patients with CHF due to any cause – DCM, coronary artery disease, valvular and congenital heart disease 16 ; in addition, troponin levels do not differ in patients with ischaemic or non-ischaemic causes of heart failure. 1 Increased cTns indicate ongoing myocyte injury or death, and excessive adrenergic stimulation through renin, angiotensin, aldosterone and endothelin signalling pathways, abnormalities in calcium handling, inflammatory cytokines, nitric oxide, and oxidative and mechanical stress may be responsible for this. 17 We found increased cTn-I values in some of our patients with DCM and these were all clinically stable patients; similarly, it has been previously shown that nearly one-fourth of the ambulatory patients with CHF present abnormal cTn-T values. 2
IMA, as assessed by the albumin cobalt binding test, is considered a marker of myocardial ischaemia, in contrast to biomarkers of myocardial injury (CPK, CKMB and cTn-I and T), which are released when cardiac necrosis occurs. Various studies have shown that albumin undergoes a considerable reduction in its capacity to bind cobalt when exposed to an ischaemic insult; ischaemia, through hypoxia, acidosis, sodium and calcium pump disruptions, and free radical injury may induce changes in the binding capacity of the NH2 terminus of the albumin to bind metals such as cobalt. IMA is a recently developed biomarker of transient myocardial ischaemia 18 ; IMA increases rapidly, within minutes, and remarkably during myocardial ischaemia. We found that IMA levels in patients with DCM were similar to those of normal subjects. We also found that IMA, in stable DCM patients, does not differ among patients with positive and negative cTn-I; this possibly implies that myocardial necrosis in compensated patients occurs without preceding transient ischaemia; this may be of help in patients with CHF and underlying coronary artery disease, where an increase in enzymes may be associated with either CHF per se or a subclinical new coronary event; in the latter case, one would expect IMA to increase along with cardiac enzymes. 10–13 What is of interest, however, is that although the IMA in our patients was in the normal range, when compared with controls, it varied significantly in relation to the NYHA classification and was negatively correlated with the left ventricular ejection fraction, both parameters reflecting disease severity.
In conclusion, in patients with compensated heart failure due to DCM, IMA serum levels do not differ compared with age-matched normal subjects but vary significantly in relation to the severity of the disease.
DECLARATIONS