Article Text

Download PDFPDF
Original article
Prognostic value of mid-regional pro-adrenomedullin in patients with heart failure after an acute myocardial infarction
  1. IJsbrand T Klip1,
  2. Adriaan A Voors1,
  3. Stefan D Anker2,3,
  4. Hans L Hillege1,
  5. Joachim Struck4,
  6. Iain Squire5,6,
  7. Dirk J van Veldhuisen1,
  8. Kenneth Dickstein7,8,
  9. for the OPTIMAAL investigators
  1. 1Department of Cardiology, University Medical Center Groningen, Groningen, The Netherlands
  2. 2Applied Cachexia Research, Department of Cardiology, Charité Campus Virchow-Klinikum, Berlin, Germany, UK
  3. 3Centre for Clinical and Basic Research, IRCCS San Raffaele, Rome, Italy
  4. 4BRAHMS Aktiengesellschaft, Henningsdorf, Germany, UK
  5. 5Department of Cardiovascular Sciences, University of Leicester, Leicester, UK
  6. 6NIHR Biomedical Research Unit, Glenfield Hospital, Leicester, UK
  7. 7Stavanger University Hospital, Stavanger, Norway
  8. 8Institute of Internal Medicine, University of Bergen, Bergen, Norway
  1. Correspondence to Professor Adriaan A Voors, Department of Cardiology, University Medical Center Groningen, PO Box 30001, 9700 RB Groningen, The Netherlands; a.a.voors{at}thorax.umcg.nl

Abstract

Objective To assess the cardiovascular prognostic value of mid-regional pro-adrenomedullin (MR-proADM) and compare this with B-type natriuretic peptide (BNP) and N-terminal pro-brain natriuretic peptide (NT-proBNP), on death or a composite end point in patients who developed heart failure after an acute myocardial infarction (AMI).

Methods From a subset of 214 patients from the OPTIMAAL study, blood samples were obtained at a median of 3 days after AMI when patients had developed signs and/or symptoms of heart failure (HF) or a left ventricular ejection fraction <0.35%. End points were all-cause mortality and a composite end point, including death, myocardial reinfarction, stroke and/or resuscitated cardiac arrest.

Results Mean age of the patients was 68±10 years and mean follow-up was 918±311 days. During follow-up 31 patients died and 61 reached the composite end point. In multivariable Cox proportional hazard models adjusted for BNP, NT-proBNP and other covariates, a doubling of MR-proADM showed a 3.02 (95% CI 1.66 to 5.49) times increased risk of mortality (p<0.001) and a 1.77 (95% CI 1.13 to 2.78) times increased risk of reaching the composite end point (p=0.013). Receiver operating characteristic curves indicated that MR-proADM (area under the curve (AUC)=0.81) was a stronger predictor of mortality than BNP (AUC=0.66; p=0.0034 vs MR-proADM) and NT-proBNP (AUC=0.67; p<0.001 vs MR-proADM). Furthermore, MR-proADM enhanced significantly risk classification and integrated discrimination improvement in comparison with BNP and NT-proBNP. Finally, changes in MR-proADM over time significantly added prognostic information to the baseline value.

Conclusion MR-proADM is a promising biomarker and has strong prognostic value for mortality and morbidity in patients with HF after an AMI. In this study, MR-proADM had stronger predictive value than BNP and NT-proBNP.

  • Heart failure
  • adrenomedullin
  • mid-regional pro-adrenomedullin
  • natriuretic peptides
  • myocardial infarction
  • cardiac remodelling
  • haemodynamics

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

The development of heart failure (HF) is one of the most important indicators of adverse prognosis in patients who have had an acute myocardial infarction (AMI). Therefore, early identification of patients at risk of adverse outcome should constitute an integral part of the management of an AMI.

Over the years, circulating levels of B-type natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) have proved to be strong prognostic markers in the prediction of death or HF after an AMI.1 However, plasma BNP and NT-proBNP levels appear to be strongly related to age2 and renal function, which themselves carry prognostic value.3 4 Another potential disadvantage is that plasma BNP levels vary over time in patients with chronic HF.5 Since neurohormonal activation has a crucial role in the progression to HF after an AMI,6 additional biomarkers should be considered in the assessment of prognosis in patients with signs of HF after an AMI.

Adrenomedullin (ADM), a 52-amino-acid peptide, plays a part in endothelial function, causing longlasting vasodilatation via nitric oxide production,7 8 similar to the biological activity of BNP. Plasma ADM levels are raised in HF9 in proportion to the severity of the disease10 11 and to the severity of left ventricular systolic dysfunction.12 13 Measurement of plasma ADM levels is difficult, however, owing to the short half-life of ADM and the existence of a binding protein.14

Like the natriuretic peptides, ADM is derived from a larger precursor peptide (proADM). Recently another more stable part of the ADM precursor peptide was identified, mid-regional pro-adrenomedullin (MR-proADM).15 In a few studies MR-proADM appeared to be promising in predicting prognosis in patients after an AMI16 and in patients with acute HF.17 18 The aim of our study was to compare the prognostic value of MR-proADM with those of BNP and NT-proBNP for mortality and morbidity in patients who developed HF after an AMI. Second, the additive prognostic value of temporal changes of MR-proADM was assessed.

Patients and methods

Patients

This study was conducted as part of a prospective neurohormonal substudy of the OPTIMAAL (Optimal Trial in Myocardial Infarction with Angiotensin II Antagonist Losartan) trial. The OPTIMAAL trial was a randomised controlled trial, comparing the effects of the angiotensin receptor antagonist losartan (50 mg daily) and the ACE inhibitor captopril (50 mg three times daily) on all-cause mortality in 5477 clinically stable patients with definite AMI.19 To be included, patients had to be over 50 years of age and have signs or symptoms of HF during the acute phase as defined by one of more of the following: treatment with a diuretic or vasodilator therapy for heart failure, pulmonary rales, third heart sound, persistent sinus tachycardia (≥100 bpm) or radiographic evidence of pulmonary congestion. Patients with an ejection fraction <35%, end-diastolic dimension >65 mm (optional) and/or new Q-wave anterior wall AMI, a new left bundle branch block, or any reinfarction with previous pathological Q waves in the anterior wall were also eligible for inclusion.

This substudy included 214 patients from six participating centres in the UK, Sweden and Norway that were willing to participate. There were no significant differences between the population of this substudy and that of the OPTIMAAL trial.

Blood sampling

Blood samples were collected using pyrogen-free tubes containing EDTA (Becton Dickinson, San Jose, California, USA) and immediately centrifuged at 3000 rpm for 30 min at 4°C. The platelet-poor plasma was separated and stored at −80°C until analysis. The estimated glomerular filtration rate (eGFR) was calculated using the abbreviated modification of diet in renal disease equation: 186×(creat/88.4)−1.154×(age)−0.203 (×0.742 if female). Blood samples were obtained at baseline (median of 3 days after AMI), after 30 days and 1 year of follow-up. The trial protocol was approved by the local ethics committees and carried out in accordance with the Declaration of Helsinki.

MR-proADM assay

The MR-proADM was detected using a novel commercial assay in the chemiluminescence coated tube format (BRAHMS AG, Hennigsdorf, Germany) as previously described.20 Briefly, tubes were coated with a purified sheep polyclonal antibody raised against a peptide representing amino acids 83 to 94 of preproADM. A purified sheep polyclonal antibody was raised against a peptide representing amino acids 68 to 86 of preproADM, was labelled with MACN-acridinium-NHS-ester (InVent GmbH, Henningsdorf, Germany) and used as a tracer. Dilutions of a peptide representing amino acids 45 to 92 of preproADM in normal horse serum served as standards. The immunoassay was performed by incubating 10 μl samples/standards and 200 μl tracer in coated tubes for 2 h at room temperature. Tubes were washed four times with 1 ml immunoassay wash solution (BRAHMS AG), and bound chemiluminescence was measured using an LB952T luminometer (Berthold, Bad Wildbad, Germany). The MR-proADM assay has been characterised in detail previously. The lower detection limit of the assay was 0.08 nmol/l; the functional assay sensitivity (defined as the lowest concentration detectable with an interassay coefficient of variation (CV) of 20%) is 0.12 nmol/l. The intra-assay CV at 0.5 and 5 nmol/l is 3% and 3.5%, respectively; the interassay CV at 0.5 and 5 nmol/l is 8.5% and 6.5%, respectively.

BNP and NT-proBNP assays

Both assays were carried out in the laboratory of the Pharmacology and Therapeutics group, University of Leicester, Department of Cardiovascular Sciences, using non-commercial assays. For the NT-proBNP assay, described previously,21 a non-competitive assay using unextracted plasma with two antibodies was carried out. Capture antibodies were directed to the C-terminal of human NT-proBNP and detector antibodies were directed to the biotinylated N-terminal. Samples or NT-proBNP standards were incubated in the C-terminal antibody-coated ELISA plate wells together with biotinylated N-terminal antibody for 24 h at 4°C. For the detection streptavidin labelled with methyl acridinium ester was used and the plates were read on a Dynatech MLX Luminometer (Dynex Technologies Ltd, Worthington, UK), with sequential injections of 100 μl of 0.1 M nitric acid (with H2O2) and then 100 μl of NaOH (with cetyl ammonium bromide). Intra-assay and interassay coefficients of variation for NT-proBNP were 2.3% and 4.8%, respectively. There was no significant cross-reactivity between the natriuretic peptides (NT-ANP, BNP or CNP). NT-proBNP values of this assay were compared with a commercially available NT-proBNP assay (Roche Elecsys, Roche Diagnostics, Rotkreutz, Switzerland), and the correlation was 0.9 (data on file, Leicester, UK).

Assays for BNP were based on commercially available antibodies (Phoenix Pharmaceuticals Inc, Belmont, California, USA). BNP was measured on C18 extracts of plasma, dried on a centrifugal evaporator and then reconstituted in assay buffer.22 Antibodies were from Bachem UK Ltd (St Helens, Merseyside, UK), and samples or standards were preincubated with 25–50 ng of the specific antibodies (within ELISA plate wells coated with 500 ng of anti-rabbit immunoglobulin G). All peptide tracers were produced by biotinylation with biotin-X-N-hydroxysuccinimide ester (Calbiochem, Nottingham, UK), followed by purification using reverse-phase high-performance liquid chromatography on a C18 column, developed with an acetonitrile gradient (2%/min). After 24 h of incubation at 4°C, biotinylated tracers of the peptides were added to each well. After a further 24 h of incubation at 4°C, ELISA plates were washed, treated with the streptavidin, labelled with methyl acridinium ester, and this was followed by detection of enhanced chemiluminescence as discussed previously. Intra-assay and interassay coefficients of variation for BNP were 4.0% and 6.8%, respectively. It was confirmed that each assay did not cross-react with the other natriuretic peptides, while maintaining 100% specificity for BNP.

End points

Values of MR-proADM, BNP and NT-proBNP were assessed for the prediction of the primary and secondary end points of this study. The primary end point was all-cause mortality. The secondary end point was the combined end point of death, myocardial reinfarction, stroke and/or resuscitated cardiac arrest.

Statistical analysis

Data are presented as mean±SD (normal distribution) or median±IQR (non-normal distribution). Baseline characteristics were divided into quartiles of MR-proADM. For comparison of differences in baseline characteristics, the χ2 test was used for non-normally distributed variables, the one-way analysis of variance test for normally distributed variables. Univariable Cox regression analysis was performed to calculate the predictive value of all variables presented in table 1 on the two major end points. Levels of MR-proADM, BNP and NT-proBNP were normalised using a double log-transformation, and presented hazard ratios (HRs) refer to a doubling in the levels of these markers. Three multivariable models were constructed, one for MR-proADM, one for BNP and one for NT-proBNP. In these multivariable Cox proportional hazard models well-recognised epidemiological predictors of outcome (eg, age, gender, renal function, previous myocardial infarction, diabetes and treatment group) were entered. Kaplan–Meier survival curves were generated to illustrate survival probability and clinical outcome over time. Log rank tests were used to assess the significance between the levels of MR-proADM above and below the median. To assess the ability of MR-proADM, BNP and NT-proBNP to predict both end points, areas under the curve (AUCs) of receiver operating characteristics (ROC) curves were calculated. The statistical significance of differences in AUCs was estimated using the approach of DeLong et al.23 ROC analysis was limited to an overall period of 31 months since follow-up of the survivors was very similar (minimum 31 months, maximum 40 months), allowing all survivors to be included in the analysis. Also, a separate analysis was performed to evaluate the prognostic effects of changes in MR-proADM between baseline and 1 month. Both baseline levels and the percentage change of MR-proADM between baseline and 1 month were altered as continuous variables in a multivariable Cox regression model for both mortality and the composite end point.

Table 1

Baseline characteristics of a subgroup of 214 patients from the OPTIMAAL study divided into quartiles of mid-regional pro-adrenomedullin (MR-proADM)

The net reclassification improvement (NRI) and the integrated discrimination improvement (IDI) were also calculated.24 The NRI assesses risk reclassification and is the difference in proportions moving up and down risk strata among case patients versus control participants—that is, those who did or did not develop the disease during follow-up. The aim of the NRI was to examine the prognostic discrimination of MR-proADM on top of the natriuretic peptides without adjustment for clinical risk factors. The IDI is the difference in Yates, or discrimination, slopes between two models, in which the Yates slope is the mean difference in predicted probabilities between case patients and control participants. We used risk categories of <20%, 20–40% and >40%. Status at 950 days' follow-up was used and observations censored (n=9) before 950 days of follow-up were excluded.

A two-sided p value of <0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 16.0 (SPSS Inc), STATA version 9.0 (StataCorp LP) or SAS software version 9.2 (SAS Institute).

Results

Patient characteristics

The baseline characteristics of the 214 patients, according to quartiles of MR-proADM, are described in table 1. Mean age was 68±10 years and mean follow-up was 918±311 days. During follow-up 31 (14.5%) patients died and 61 (28.5%) patients reached the composite end point of either death, myocardial reinfarction, stroke or resuscitated cardiac arrest.

Plasma levels of MR-proADM ranged from 0.38 to 4.02 nmol/l with a median of 0.83 nmol/l. Quartiles of higher MR-proADM levels were significantly associated with increasing age (p<0.001), increased percentage of smokers (p=0.021), Killip class >1 on admission (p=0.005), hypertension (p=0.04) and decreased renal function (p<0.001) (table 1). In addition, levels of MR-proADM also increased with increasing levels of BNP and NT-proBNP (table 1). Levels of MR-proADM were lower in patients who received β blockers (p=0.025), statins (p=0.01) or who were treated with thrombolysis (p=0.015) (table 1).

The Kaplan–Meier survival curves (figures 1 and 2) showed increased mortality (log rank χ2 test 27.6; p<0.001) and worse clinical outcome (log rank χ2 test 7.84; p=0.005) in patients with MR-proADM levels above, compared with MR-proADM levels below the median.

Figure 1

Kaplan–Meier curves demonstrating freedom from death over time according to levels of mid-regional pro-adrenomedullin (MR-proADM) above and below the median (0.83 nmol/l) at baseline (log rank χ2 test 41.4; p<0.001).

Figure 2

Kaplan–Meier curves demonstrating freedom from combined end point of death/myocardial reinfarction/stroke/resuscitated cardiac arrest over time according to mid-regional pro-adrenomedullin (MR-proADM) levels above and below the median (0.83 nmol/l) at baseline (log rank χ2 test 7.84; p=0.005).

Univariable analysis

In univariable analysis, higher levels of MR-proADM (HR=3.84 per doubling; 95% CI 2.5 to 5.9; p<0.001), BNP (HR=1.85; 95% CI 1.28 to 2.66; p=0.001) and NT-proBNP (HR=2.16; 95% CI 1.29 to 3.61; p=0.003) were significantly related to a higher mortality.

The composite end point of death, myocardial reinfarction, stroke and/or resuscitated cardiac arrest was also related to higher levels of MR-proADM (HR=2.2; 95% CI 1.53 to 3.14; p<0.001), BNP (HR=1.4; 95% CI 1.09 to 1.79 p=0.008) and NT-proBNP (HR=1.55; 95% CI 1.12 to 2.15; p=0.008).

Multivariable analysis

Two multivariable models were made for all three markers. The first model was adjusted for age and gender. MR-proADM proved to be independently associated with mortality (HR=3.02; 95% CI 1.66 to 5.49; p<0.001) and the composite end point (HR=1.77; 95% CI 1.13 to 2.78; p=0.013) (table 2).

Table 2

Multivariable Cox regression survival analysis of the predictive value of log mid-regional pro-adrenomedullin (MR-proADM), B-type natriuretic peptide (BNP) and N-terminal pro-brain natriuretic peptide (NT-proBNP) on death and the composite end point of death/myocardial reinfarction/stroke/resuscitated cardiac arrest

After adjusting for age, gender, previous AMI, diabetes, eGFR and treatment group in a multivariable analysis, MR-proADM remained independently associated with mortality (HR=3.51; 95% CI 1.38 to 8.94; p=0.008) but not with the composite end point (HR=1.78; 95% CI 0.94 to 3.35; p=0.075). BNP and NT-proBNP were not significantly associated with mortality or the composite end point in both adjusted multivariable models (table 2).

ROC, NRI and IDI: analysis

The ROC curve for MR-proADM yielded an area under the curve (AUC) of 0.81 (95% CI 0.73 to 0.9) in predicting mortality (figure 3). For BNP the AUC was 0.66 (95% CI 0.55 to 0.78; MR-proADM vs BNP, p=0.0034; figure 3) and for NT-proBNP 0.67 (95% CI 0.56 to 0.77; MR-proADM vs NT-proBNP, p=0.0004; figure 3).

Figure 3

Receiver operating characteristics curves for mid-regional pro-adrenomedullin (MR-proADM), B-type natriuretic peptide (BNP) and N-terminal pro-brain natriuretic peptide (NT-proBNP) to predict mortality. Areas under the curves were 0.81 for MR-proADM, 0.67 for NT-proBNP and 0.66 for BNP (MR-proADM vs BNP, p=0.0034; MR-proADM vs NT-proBNP, p=0.0004).

For the composite end point the areas under the ROC curves were 0.66 for MR-proADM (95% CI 0.58 to 0.75), 0.60 for BNP (95% CI 0.52 to 0.69; MR-proADM vs BNP, p=0.22; figure 4) and 0.61 for NT-proBNP (95% CI 0.58 to 0.75; MR-proADM vs NT-proBNP; p=0.14; figure 4).

Figure 4

Receiver operating characteristics curves for mid-regional pro-adrenomedullin (MR-proADM), B-type natriuretic peptide (BNP) and N-terminal pro-brain natriuretic peptide (NT-proBNP) to predict the composite end point of death/myocardial reinfarction/stroke/resuscitated cardiac arrest. Areas under the curve were 0.66 for MR-proADM, 0.61 for NT-proBNP and 0.60 for BNP (MR-proADM vs BNP, p=0.22; MR-proADM vs NT-proBNP, p=0.14).

When combining MR-proADM and BNP or MR-proADM and NT-proBNP in predicting mortality, using a binary logistic model, the areas under the curves were 0.79 (MR-proADM + BNP) and 0.81 (MR-proADM + NT-proBNP). The prediction of the composite end point of death, myocardial reinfarction, stroke and/or resuscitated cardiac arrest yielded AUCs of 0.67 (MR-proADM + BNP) and 0.66 (MR-proADM + NT-proBNP).

Patients, who developed a composite end point and those who did not, were classified separately into risk categories. The NRI for the addition of MR-proADM in a prediction model with BNP was 11.5% in the group with events versus 14.6% in the group without events (p=0.0133), whereas these figures for NT-ProBNP were 8.2% and 13.2% (p=0.035), respectively. The IDI for MR-proADM was 3.3% (p=0.04) and 3.9% (p=0.005) when compared with BNP and NT-proBNP, respectively.

Sensitivity, specificity, positive and negative predictive values for death of MR-proADM, BNP and NT-proBNP, at the optimal cut-off level as derived from the ROC curves, are presented in table 3. At the optimal cut-off level, both negative and positive predictive values were highest for MR-proADM.

Table 3

Sensitivity, specificity, positive and negative predictive values for predicting death of mid-regional pro-adrenomedullin (MR-proADM), B-type natriuretic peptide (BNP) and N-terminal pro-brain natriuretic peptide (NT-proBNP) at the optimal cut-off level as derived from the receiver operating characteristic curves

Changes in MR-proADM over time

Median (IQR) plasma levels of MR-proADM, BNP and NT-proBNP at baseline, 30 days and 1 year are presented in table 4. Median concentrations of all neurohormones are highest at baseline and decrease over time. For a total of 197 patients separate analyses were made on the predictive value of an increase in MR-proADM between baseline and 30 days after AMI on both end points. A percentage change in MR-proADM between baseline and 30 days, when adjusted for baseline MR-proADM levels, showed a significant increase in both mortality (HR=1.02 (95% CI 1.01 to 1.03) per 100% increase; p=0.002) and the composite end point (HR=1.02 (95% CI 1.01 to 1.03) per 100% increase; p<0.001). However, in a multivariate model, adjusted for age, gender, previous myocardial infarction, diabetes, eGFR and treatment group, a percentage change in MR-proADM was not significantly related to an increased mortality (HR=1.01 (95% CI 1.0 to 1.03) per 100% increase; p=0.315) and the composite end point (HR=1.01 (95% CI 1.0 to 1.02) per 100% increase; p=0.092).

Table 4

Median (IQR) plasma levels of mid-regional pro-adrenomedullin (MR-proADM), B-type natriuretic peptide (BNP) and N-terminal pro-brain natriuretic peptide (NT-proBNP) at baseline, 30 days and 1 year after an acute myocardial infarction complicated by heart failure

Discussion

The most important finding of this study is the strong prognostic value of MR-proADM for death and cardiovascular events in patients with clinical signs and symptoms of HF or left ventricular dysfunction after an AMI. This novel biomarker also appeared to have a stronger predictive value for mortality than BNP and NT-proBNP. These observations are very similar to our previous study with C-terminal provasopressin (copeptin).25

ADM is a 52-amino-acid peptide that is found in the adrenal medulla, heart, brain, lung, kidney and gastrointestinal organs. Also, ADM is highly expressed in endothelial cells and has come to be known as a secretory product of the vascular endothelium, together with nitric oxide and endothelin.26 Having similar effects to nitric oxide, ADM is a potent vasodilator with inotropic and natriuretic properties, the secretion of which has been shown to be stimulated by both cardiac pressure and volume overload.27 The level of circulating ADM is raised in relation to the severity of HF.28 The prognostic value of ADM after an AMI was suggested in two smaller studies.13 29 Richards et al described a modest prognostic value of ADM for mortality in 121 patients after an AMI.13 Katayama et al found that plasma adrenomedullin concentrations are a strong predictor of patient prognosis in the setting of AMI.29 However, the prognostic value of ADM may have been limited by its instability and rapid elimination from the circulation.14 MR-proADM is the inactive part of the ADM precursor peptide and is more stable in the circulation. Its prognostic value has been demonstrated in patients with an acute myocardial infarction.16 This is the first study to describe the strong prognostic value of MR-proADM in patients with HF after an AMI.

In this study, patients with higher levels of MR-proADM had increased risk for mortality and the composite end point. The Kaplan–Meier curves show that the difference in events becomes apparent immediately after the onset of the study. This is probably related to the high event rate that occurs immediately after an acute AMI.

Another important observation was that the prognostic value of MR-proADM for mortality was better than that of both BNP and NT-proBNP. In patients with HF, circulating natriuretic peptide concentrations are established risk markers for increased risk for adverse prognosis. Khan et al found that the predictive values of MR-proADM and NT-proBNP for death and HF were similar in 983 patients after an acute myocardial infarction.16 In 786 outpatients with chronic heart failure (CHF), Adlbrecht et al found that MR-proADM may be an important prognostic humoral marker in non-ischaemic patients with CHF.30 Melander et al found, in studying 5067 participants without cardiovascular events, that both MR-proADM and NT-proBNP were strong predictors of coronary events.31 Our paper differs from the paper by Khan, since this paper describes patients who had all developed signs or symptoms of HF after an AMI. Also, this study is a multicentre randomised controlled trial. Compared with the papers by Adlbrecht et al and Melander et al, our study is related to patients after a MI and not outpatients with CHF or patients without cardiovascular events. Also, this study presented data on serial measurements, whereas the previous three studies had only single measurements.

The AUCs of the ROC curves of 0.66 and 0.67, respectively, for BNP and NT-proBNP to predict mortality, were similar to those of other CHF studies.17 32 33 In this study the AUC of MR-proADM, to predict mortality, was 0.81. This makes the prognostic value of MR-proADM compared with both BNP and NT-proBNP even stronger.

The explanation for the apparent superiority of MR-proADM over both BNP and NT-proBNP in this study is unclear. A possible explanation is that BNP and NT-proBNP are strongly correlated with age and renal function,2 3 two factors which show powerful association with adverse outcome after an AMI.4 However, the correlations with age and renal function were similar for all three markers, suggesting that differential correlations with these parameters do not explain this observation (data not shown).

A second additional and new finding of this study was the added value of a second measurement of MR-proADM after 1 month. A percentage increase in MR-proADM between baseline and 1 month was related to a significant increase in both mortality and the composite end point. Serial measurements in a ‘MR-proADM-guided-therapy’ study could give added value to this observation. In a recent study, Dhillon et al investigated the prognostic value of admission and discharge MR-proADM levels in 745 patients with non-ST-elevation MI on major adverse cardiac events (including all-cause mortality, HF hospitalisation and/or recurrent AMI).34 The difference between the two sampling times revealed a univariate association with death but no other cardiac events. Also, there was no significant association seen in the multivariate analysis.

The limitations of this study are those inherent in any small, observational study of this nature. Only 214 patients were included and only 31 deaths occurred during follow-up. These observations should be tested further in a larger population of patients with HF after an AMI. Another important limitation is that data on infarct size and left ventricular function and dimensions were not available for this study. Finally, both BNP and NT-proBNP were measured using non-commercial assays. The correlation between our assay and a commercial assay for NT-proBNP was 0.9, which assumes that our results would have been similar if commercial assays had been used. However, we do not have similar comparative data for the BNP assay, so we cannot rule out the possibility that the worse prognostic performance of BNP is due, at least in part, to poorer analytical methods.

The clinical implications of our findings could be important. BNP and NT-proBNP are currently widely used as established risk markers in patients with CHF, since BNP-guided treatment may improve outcome in patients with HF. Further study is required to confirm our findings, and to assess whether the clinical use of MR-proADM might further improve outcome.

In conclusion, MR-proADM is a novel and promising biomarker and has strong prognostic value for mortality and morbidity in patients with HF after an AMI. In this study, MR-proADM had stronger predictive value than BNP and NT-proBNP.

References

View Abstract

Footnotes

  • Funding This study was supported by an unrestricted grant form Merck Research Laboratories, Bluebell, Pennsylvania and by research grants from the Brandenburg Ministry of Economics, Germany and the European Regional Development Fund (EFRE/ERDF). DJvV and AAV are clinical established investigators of the Netherlands Heart Foundation (D97-017 and 2006T37).

  • Competing interests Disclosures and potential conflicts of interest: 1. AAV received a research grant from B.R.A.H.M.S., a biotech company that developed the mid-regional pro-adrenomedullin (MR-proADM) assay. 2. JS is employed by B.R.A.H.M.S. 3. B.R.A.H.M.S holds patent applications with JS as co-inventor on the use of mid-regional pro-adrenomedullin for diagnostics. 4. SDA has received consultant honoraria from B.R.A.H.M.S.

  • Patient consent Obtained.

  • Ethics approval This study was conducted with the approval of the local ethics committees and done in accordance with the Declaration of Helsinki.

  • Provenance and peer review Not commissioned; externally peer reviewed.