Author + information
- Received June 29, 2009
- Revision received August 26, 2009
- Accepted August 31, 2009
- Published online October 27, 2009.
- George J. Mak, MD⁎,†,
- Mark T. Ledwidge, PhD⁎,
- Chris J. Watson, PhD†,
- Dermot M. Phelan, MD⁎,†,
- Ian R. Dawkins, DPhil⁎,
- Niamh F. Murphy, MD, PhD⁎,
- Anil K. Patle, BTech⁎,
- John A. Baugh, PhD† and
- Kenneth M. McDonald, MD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Prof. Kenneth M. McDonald, Director, Heart Failure Unit, St. Vincent's University Hospital, Elm Park, Dublin 4, Ireland
Objectives This study was designed to evaluate the impact of eplerenone on collagen turnover in preserved systolic function heart failure (HFPSF).
Background Despite growing interest in abnormal collagen metabolism as a feature of HFPSF with diastolic dysfunction, the natural history of markers of collagen turnover and the impact of selective aldosterone antagonism on this natural history remains unknown.
Methods We evaluated 44 patients with HFPSF, randomly assigned to control (n = 20) or eplerenone 25 mg daily (n = 24) for 6 months, increased to 50 mg daily from 6 to 12 months. Serum markers of collagen turnover and inflammation were analyzed at baseline and at 6 and 12 months and included pro-collagen type-I and -III aminoterminal peptides, matrix metalloproteinase type-2, interleukin-6 and -8, and tumor necrosis factor-alpha. Doppler-echocardiographic assessment of diastolic filling indexes and tissue Doppler analyses were also obtained.
Results The mean age of the patients was 80 ± 7.8 years; 46% were male; 64% were receiving an angiotensin-converting enzyme inhibitor, 34% an angiotensin-II receptor blocker, and 68% were receiving beta-blocker therapy. Pro-collagen type-III and -I aminoterminal peptides, matrix metalloproteinase type-2, interleukin-6 and -8, and tumor necrosis factor-alpha increased with time in the control group. Eplerenone treatment had no significant impact on any biomarker at 6 months but attenuated the increase in pro-collagen type-III aminoterminal peptide at 12 months (p = 0.006). Eplerenone therapy was associated with modest effects on diastolic function without any impact on clinical variables or brain natriuretic peptide.
Conclusions This study demonstrates progressive increases in markers of collagen turnover and inflammation in HFPSF with diastolic dysfunction. Despite high background utilization of renin-angiotensin-aldosterone modulators, eplerenone therapy prevents a progressive increase in pro-collagen type-III aminoterminal peptide and may have a role in management of this disease. (The Effect of Eplerenone and Atorvastatin on Markers of Collagen Turnover in Diastolic Heart Failure; NCT00505336)
Preserved systolic function is increasingly recognized in heart failure (HF) and is now generally accepted to be present in as many as 50% of new presentations of this syndrome (1–3). Abnormalities in diastolic function have been shown to be a major determinant of symptoms in many cases of preserved systolic function heart failure (HFPSF) (4). Recent data from our laboratory have provided some insight into possible pathophysiological mechanisms explaining the alterations in diastolic function (5). Abnormalities in collagen turnover were shown to be present, and these abnormalities correlated with markers of diastolic function as assessed by Doppler echocardiography (5). This observation may provide a possible therapeutic avenue for a syndrome still deficient in effective disease-modifying therapies.
The natural history of collagen turnover in this condition is unclear. This is now more easily addressed, given that serum levels of markers of collagen turnover have been shown to reflect myocardial fibrosis (6) and clinical outcome (7). Furthermore, it remains unknown whether the observed abnormalities in collagen metabolism can be beneficially modulated by a therapeutic intervention and whether such changes would result in improvements in diastolic function and clinical status. Efforts to date to improve outcome in HFPSF have been somewhat disappointing, with 3 studies showing at best a modest impact on secondary end points (8–10). All studies have sought to modulate the renin-angiotensin-aldosterone system (RAAS) through use of an angiotensin-converting enzyme inhibitor (ACE-I) or angiotensin-II receptor blocker (ARB) therapy. Such approaches have been based on experimental and clinical data indicating that angiotensin-II in particular may be an important stimulus for myocyte hypertrophy and abnormal growth of the cardiac interstitium (11,12).
Aldosterone, a potent pro-fibrotic hormone also involved in the RAAS, has been shown to stimulate myocardial fibrosis (13). While ACE-I and ARB therapies may have some indirect impact on aldosterone, direct inhibition may be the most effective means of diminishing the antifibrotic effect of this hormone and improving diastolic function (14–16). Such an approach has been shown to be effective in patients with hypertensive heart disease (HHD) with improvement in diastolic function with spironolactone over 6 months (17), but it has yet to be tested in patients with proven HFPSF and diastolic dysfunction.
Therefore, the primary aim of this study was to assess the natural history of collagen turnover in HFPSF and to determine whether treatment with the aldosterone antagonist eplerenone can have an impact on abnormalities in collagen metabolism in patients with established HFPSF as a result of diastolic dysfunction.
Patients and study design
This was a prospective, randomized, open-label study. All parameters were assessed by persons blinded to treatment. Subjects gave written informed consent to participate in the study. The study protocol was approved by the ethics committee of St. Vincent's University Hospital, which conformed to the principles of the Helsinki Declaration.
The study population consisted of 44 Caucasian patients with proven HFPSF. The diagnosis of HFPSF was based on the following criteria: prior New York Heart Association (NYHA) functional class IV HF admission or symptoms consistent with HF, B-type natriuretic peptide (BNP) >100 pg/ml, left ventricular ejection fraction >45%, and evidence of diastolic dysfunction on Doppler-echocardiographic study. It is expected that cardiac remodeling in HFPSF is a protracted process and accordingly, patients were followed up for 12 months. Furthermore, since it has been shown in a hypertensive population that eplerenone exerts its effect in a dose-dependent fashion (18), we evaluated patients with a dose of 25 mg daily for 6 months followed by a dose increment to 50 mg until the 12-month time point. Figure 1demonstrates the recruitment and follow-up of patients in the study.
Patients were excluded if they were clinically unstable as defined by any change in diuretic dose a month before enrollment or were already receiving eplerenone or spironolactone therapy. Other exclusion criteria were evidence of significant inflammatory disease, hepatic disease, or metabolic bone disease that may alter parameters of collagen metabolism, serum creatinine >200 μmol/l, prior documented left ventricular ejection fraction <45%, hemodynamically significant valvular disease, cor-pulmonale, hypertrophic, restrictive, or constrictive cardiomyopathy, atrial fibrillation or flutter with resting ventricular rate >120 beats/min, severe anemia, clinically significant pulmonary disease as evidenced by hospitalizations, or use of oral corticosteroids for pulmonary decompensation within 12 months or patients who require home oxygen therapy. After informed consent, patients were randomly allocated to 1 of 2 treatment arms: treatment with the selective aldosterone antagonist eplerenone 25 mg daily (with a dose increase to 50 mg daily after 6 months to investigate any dose-response effect) or usual HF treatment. All patients were followed up for 12 months.
All patients were assessed at their baseline visit and at 6 and 12 months. At each visit, a full physical examination, NYHA functional class, and quality-of-life indexes measured by the Minnesota Living with HF Questionnaire were performed by a blinded observer.
Peripheral venous blood samples were drawn at each clinical visit and immediately underwent serum isolation. Venous samples were tested for full blood count, urea and electrolytes, liver function tests, calcium, lipid profile, and BNP. Additional samples were taken for markers of collagen turnover and inflammatory status and immediately separated by centrifuge, with aliquots stored at −80°C. These markers were subsequently simultaneously analyzed by investigators blinded to treatment status of the patients.
Biochemical analysis of markers of collagen turnover
Amino-terminal peptide of pro-collagen type-I (PINP), amino-terminal peptide of pro-collagen type-III (PIIINP), and carboxy-terminal telopeptide of collagen type-I (CITP) were measured by radioimmunoassay using commercial antiserum kits (Orion Diagnostica, Espoo, Finland). Intra-assay variations for determining PINP, PIIINP, and CITP were 7%, <5%, and <8%, respectively.
Carboxy-terminal peptide of pro-collagen type-I (PICP) was measured with a specific enzyme-linked immunosorbent assay based on the manufacturer's method (Takara Biochemicals, Tokyo, Japan). Serum matrix-metalloproteinase (MMP) type-2 and -9 and tissue inhibitors of MMP were analyzed using commercially available 2-site sandwich enzyme-linked immunosorbent assays (Amersham Pharmaceuticals, Buckinghamshire, United Kingdom). Intra-assay variations were <10% for all the above assays. Duplicate measurements were performed in all the above tests. The BNP was measured using the Triage Meter Point of Care assay (Biosite, San Diego, California) in all patients.
Serum concentrations of interleukin (IL)-6, IL-8, tumor necrosis factor (TNF)-α, monocyte chemoattractant protein-1, and high-sensitivity C-reactive protein were analyzed using an ultrasensitive electrochemiluminescence immunoassay following instructions from the manufacturer (Meso Scale Discovery, Gaithersburg, Maryland). Plates were analyzed using a Meso Scale Discovery Sector Imager 2400 instrument.
Two-dimensional imaging, targeted M-mode, and Doppler ultrasound measurements were obtained in all patients using standard techniques (19). Left ventricular filling pressures were noninvasively assessed through analysis of the ratio of peak E-wave (E) to maximum velocity of lateral mitral valve annulus (E′) ratio (E′/E) measured at the lateral mitral valve annulus. All echocardiography data were based on the mean of 3 measurements on consecutive cardiac cycles. An experienced blinded sonographer carried out all Doppler-electrocardiographic imaging.
Demographics are presented as mean ± SD for continuous normal variables, median and interquartile range for non-normal continuous variables, and as frequencies and percents for nominal/categorical variables. Comparisons between eplerenone and control groups were made on changes over the study period using independent 2-sample ttests for continuous normally distributed data, Mann-Whitney tests for skewed continuous, and chi-square tests for categorical. Within-group tests, comparing baseline to 6 and 12 months, were conducted using paired sample ttests and paired sample Wilcoxon tests where appropriate. The relationships between markers of collagen turnover and markers of inflammation were assessed using Spearman's Rho nonparametric correlation coefficient.
We powered this pilot study to show a difference in PICP levels between the study groups over 12 months. We based our power calculations on data from our laboratory on the relative differences and sensitivities of PICP values between different phases of diastolic dysfunction (5) and assumed that eplerenone would prevent an increase of 25% in PICP levels over 1 year. Thus, we expected an absolute difference of 102 ng/ml with an average SD of 101 ng/ml, and at least 16 patients per arm were required for 80% power with 2-sided significance set at 0.05. However, because this is a pilot study with a paucity of data on which to base power calculations, we decided to recruit at least 20 patients per arm. Analyses were carried out using SPSS version 13 statistical software (Statistical Package for the Social Sciences, SPSS Inc., Chicago, Illinois).
Demographic details for the population are presented in Table 1.As outlined, this is an older group of HF patients (mean age 80 ± 7.8 years) typical of a community-based population with HFPSF. A substantial proportion (91%) of the population had background hypertension, and a large number were receiving therapies that modulate the RAAS. Fifty-nine percent of patients had atrial fibrillation.
The criteria for diagnosis of HFPSF in this population are presented in Table 2.Approximately one-half had a prior hospitalization for acute decompensated HF, confirmed by a cardiologist. All patients were symptomatic, but the majority (87%) were relatively well, being in NYHA functional class II. The BNP was >100 pg/ml in all when clinically stable, with 55% having levels >200 pg/ml. All patients had increased left atrial volume indexes (mean 48 ± 18 ml/m2) in the absence of significant mitral valve pathology, consistent with the presence of diastolic dysfunction.
Markers of collagen turnover
Sequential changes in collagen parameters are presented in Table 3.There was little change in markers of collagen turnover or enzymes involved in their metabolism in the control group between 0 and 6 months. However, at 12 months, there were significant increases in PIIINP (p = 0.003), PINP (p = 0.025), and MMP-2 (p = 0.003) levels compared with baseline.
Eplerenone attenuated the increase in PIIINP observed in the control group at 12 months (change from baseline −0.50 ± 2.01 μg/l for eplerenone vs. 2.13 ± 2.78 μg/l for control group, p = 0.006) (Fig. 2).Additionally, there was a nonsignificant attenuation in the rise of CITP levels in the eplerenone-treated group (change from baseline −0.49 ± 2.38 μg/l for eplerenone vs. 1.5 ± 3.46 μg/l for control group, p = 0.09).
In the control group, there was no change in inflammatory markers observed after 6 months (Table 4).However, there was a significant increase in IL-6 and TNF-α in the control group at 12 months. Eplerenone had no significant impact on inflammatory markers in comparison with the control group, although there were significant increases in IL-6 and -8 over time compared with baseline.
We observed significant correlations between IL-6 and MMP-2 (r = 0.477, p = 0.001) and between IL-6 and PIIINP (r = 0.298, p = 0.049) at baseline. We also observed correlations between TNF-α and the following biochemical markers at baseline: MMP-2 (r = 0.438, p = 0.003), PICP (r = 0.37, p = 0.015), and PIIINP (r = 0.468, p = 0.002). Significant correlations were observed between IL-6 and the following biochemical markers at 12 months: tissue inhibitors of MMP-1 (r = 0.402, p = 0.011), PICP (r = 0.51, p = <0.001), PINP (r = 0.315, p = 0.048), and PIIINP (r = 0.339, p = 0.035).
At baseline, both groups demonstrated minor abnormalities in diastolic function, with a marginal prolongation of isovolumetric relaxation time, deceleration time, and an E/A ratio of <1 in both groups (Table 5).Additionally, the left atrial volume index was increased in both groups in the absence of significant mitral valve disease.
Over time in the control group, there was no significant change in any of the parameters of diastolic function other than a reduction in E′. This reduction in E′ was also observed in the eplerenone group, and there was no between-group difference. However, eplerenone produced a significant reduction in deceleration time compared with control over the course of the study (change from baseline −82 ± 51 ms for eplerenone vs. −7 ± 73 ms for the control group, p = 0.032) (Fig. 3).
Clinical measurements and BNP
At baseline, there were no clinically significant differences between groups in terms of clinical indexes, biochemical indexes, and quality-of-life score (Table 6).
In the control group, BNP levels progressively decreased over time. In the eplerenone group, BNP also decreased along with the expected increase in urea and potassium over the course of the study, although the between-group differences were not significant.
This is the first randomized, prospective study assessing the impact of eplerenone, an aldosterone receptor antagonist, on patients with established HFPSF associated with diastolic dysfunction. The report provides original observations on the natural history of collagen turnover in this population over 1 year and the impact of eplerenone on these markers, inflammatory status, and measures of ventricular function. In the control group, there is evidence of progressive remodeling of the interstitium indicated by an increase in markers of collagen turnover, namely, PIIINP, PINP, and MMP-2, that was associated with a subtle decline in diastolic function. Eplerenone had a modest impact on this natural history by attenuating the increase in PIIINP. While there was an associated reduction in deceleration time, there were no other significant changes in markers of diastolic function compared with the control group. In summary, these data indicate that the natural history of HFPSF with diastolic dysfunction is characterized by a progressive increase in markers of collagen turnover. Selective aldosterone antagonism had a modest but measurable impact on this natural history, without significant impact on diastolic function.
Recent data from our group and others have demonstrated that abnormalities in collagen metabolism resulting in excessive myocardial fibrosis may be a central pathophysiological abnormality in patients with HHD and HFPSF (5,20–23). Such observations have predominantly relied on serum markers of collagen turnover, which may be a practical surrogate of collagen content within the myocardium (6,24,25). In particular, it was found that breakdown products of the major fibrillar collagens of the heart, PICP and PIIINP, were significantly elevated, thereby suggesting an increase in collagen synthesis in patients with established HFPSF (5). Such observations have predominantly relied on serum markers of collagen turnover, which may be a practical surrogate of collagen content within the myocardium (24,25). The data presented here extend these observations by providing original information on the natural history of these markers over time in HFPSF with diastolic dysfunction. No significant changes were noted at 6 months, which is in contrast to the natural history of collagen turnover after acute myocardial infarction (26). That likely reflects the differences between an acute healing response in infarction compared with progressive remodeling of the interstitium as might be expected to occur in our patient group. At 12 months, significant changes in markers of collagen turnover were observed in the control group. Changes in collagen-I metabolism, the major collagen form in the myocardium, were observed, with an increase in PINP and CITP. However, no sequential change in PICP was observed. Alterations in collagen-III metabolism were also noted, with a significant increase in PIIINP. In addition, MMP-2, a gelatinase that has profibrotic properties, increased over time (27). Taken together, these data highlight a disease process that is progressive despite treatment in a high number of patients with RAAS-modifying agents.
Increased serum levels of PIIINP may reflect synthesis or degradation of collagen-III and have been associated with HHD and worsening phases of diastolic function (25,28,29). That may reflect changes in relative collagen I to III content, potentially contributing to stiffness in the myocardium (30,31). We have previously shown that PIIINP levels are increased in more severe diastolic function and also in HFPSF (5,32). Alterations in PIIINP may also have prognostic relevance. This marker has been shown to be a strong predictor of outcome in reduced systolic function HF (7) and also an indicator of those likely to respond to aldosterone antagonist therapy (16). While statistical significance was only observed in this study with PIIINP levels, it is possible that there are corresponding changes in collagen-I not reflected in serum markers in this small patient population.
What remains uncertain is the stimulus for abnormalities in collagen turnover in this syndrome, a critical issue to help define potential beneficial therapeutic interventions. There is a substantial body of evidence linking the RAAS to myocardial fibrosis (11,15,16,20,33). In particular, angiotensin-II has been linked with myocyte hypertrophy and excessive growth of the cardiac interstitium (11,34). As a result, initial attempts to modify this process focused on the impact of ACE-I and ARB therapy on myocardial fibrosis in experimental models of HHD and HFPSF. In general, results were positive and led to the application of such strategies to patients with HHD (20,23) and HFPSF (33). While small single-center studies were encouraging, all 3 large multicenter studies published to date designed to assess the impact of these therapies have not shown a significant impact on their primary end points (8–10). The explanation for these disappointing results is unclear, but several issues need to be considered. While all patients studied were defined as having HFPSF, it appears from the CHARM (Candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity) study dataset that many did not have diastolic dysfunction, diluting the chances of showing an impact of therapy (4). An alternative explanation could be that, in several of these studies, there was a high use of agents such as ACE-I, ARB, and beta-blockers in the placebo group that could have modified the pathological process, thereby making it more difficult for additional RAAS-modulating therapy to demonstrate an impact. Finally, it is likely that angiotensin-II is only one of several mediators of progressive myocardial remodeling in HFPSF. Besides angiotensin-II, there is now a growing interest in the potential role of biochemical mediators of inflammation, ischemia, and oxidative stress in the myocardium (35,36).
Aldosterone is a well-established pro-fibrotic hormone (13). Several studies have underlined its potential role in growth of the cardiac interstitium (13,34,37). Furthermore, myocardial levels of aldosterone have been shown to be elevated in experimental models of HFPSF (12). While an important component of the RAAS, levels of aldosterone are not completely suppressed over time by ACE-I and ARB therapy, which also may in part explain why these therapies have not provided anticipated benefits in large-scale clinical trials. Therefore, more direct inhibition of aldosterone may be required to negate the profibrotic effects of this hormone. In this regard, therapy with specific antagonists of aldosterone has resulted in improvement in parameters of diastolic function (17).
The impact of eplerenone on PIIINP was not associated with any consistent impact on parameters of diastolic function. A modest improvement in deceleration time was observed, but the mean E′ value was also reduced. It should be noted that the major abnormalities in diastolic dysfunction observed in this group are more consistent with impaired relaxation than with major alterations in ventricular compliance, where abnormalities are likely to be more reflective of alteration in collagen metabolism. Such relatively modest diastolic dysfunction with limited impact by eplerenone is consistent with the observation that there were no changes in markers of clinical status compared with the untreated group. However, of note is that the population was clinically stable and the majority were in NYHA functional class II, with only moderately elevated natriuretic peptide levels and low Minnesota Living With Heart Failure Questionnaire scores. Similar to other trials in this population, there was a high background use of ACE-I, ARB, and beta-blocker therapy, all of which through modulation of the RAAS could have hindered the ability of eplerenone to demonstrate additional clinical benefit. It may be that the impact of eplerenone would be more impressive in a later stage of this syndrome, with features of diastolic dysfunction more consistent with altered compliance reflecting the impact of pathology within the interstitium.
An alternative explanation for fibrosis in this population is low-grade systemic inflammation. This has already been defined in HHD (38), a frequent forerunner of HFPSF. Indeed, low-grade inflammation as evidenced by elevated urinary levels of TNF-α has been shown to identify a hypertensive population at higher risk of target organ damage (39). In this regard, it is of interest that we see correlations between several markers of collagen turnover and inflammation and an increase of several inflammatory markers over time in our population. In particular, a rise in IL-6 is seen in both groups over time. Additionally, there is also a rise in TNF-α in the control group over time. As eplerenone therapy did not significantly change the inflammatory response in our study, it is possible that several mechanisms are at play in furthering collagen remodeling in this syndrome.
In interpreting these data, several issues need to be considered. This is a small study, in a well-defined HFPSF population of patients who were symptomatically well. Furthermore, the majority were already on therapy known to modulate the RAAS, thereby likely making it more difficult for benefit to be shown with eplerenone. Although we saw an effect of eplerenone on PIIINP, we failed to see a change in PICP, which had been used to power the study. It should also be noted that changes in collagen markers are not exclusively confined to the myocardial extracellular matrix. Finally, although the analyses were performed in a blinded fashion, it is important to note that the control group was untreated rather than placebo treated.
This study presents original data on the progressive increase in markers of collagen turnover and inflammation in patients with HFPSF and diastolic dysfunction. Eplerenone treatment for 1 year attenuated increases in PIIINP with no impact on other markers of collagen turnover and no significant impact on diastolic dysfunction. While no clinical benefit was observed in this small study, the data nonetheless demonstrate an impact on the pathophysiology of this syndrome and thereby a possible benefit of this therapy in the management of this subtype of HF. We await the results of the TOPCAT (Treatment of Preserved Cardiac function heart failure with an Aldosterone antagonist Trial [NCT00094302]) study, which will assess the clinical and biochemical impact of aldosterone antagonism in this patient population.
Dr. Mak received grant support from the Irish Heart Foundation(The Noel Hickey Bursary) sponsored by Pfizer. Drs. Ledwidge and McDonald have received honoraria from Pfizer.
- Abbreviations and Acronyms
- angiotensin-converting enzyme inhibitor
- angiotensin-II receptor blocker
- B-type natriuretic peptide
- carboxy-terminal telopeptide of collagen type-I
- maximum velocity of lateral mitral valve annulus
- heart failure
- preserved systolic function heart failure
- hypertensive heart disease
- matrix metalloproteinase
- New York Heart Association
- carboxy-terminal peptide of pro-collagen type-I
- amino-terminal peptide of pro-collagen type-I
- amino-terminal peptide of pro-collagen type-III
- renin-angiotensin-aldosterone system
- tumor necrosis factor
- Received June 29, 2009.
- Revision received August 26, 2009.
- Accepted August 31, 2009.
- American College of Cardiology Foundation
- Hogg K.,
- Swedberg K.,
- McMurray J.
- Lenzen M.J.,
- Scholte op Reimer W.J.,
- Boersma E.,
- et al.
- Persson H.,
- Lonn E.,
- Edner M.,
- et al.
- Martos R.,
- Baugh J.,
- Ledwidge M.,
- et al.
- Diez J.,
- Panizo A.,
- Gil M.J.,
- Monreal I.,
- Hernandez M.,
- Pardo M.J.
- Cleland J.G.,
- Tendera M.,
- Adamus J.,
- Freemantle N.,
- Polonski L.,
- Taylor J.
- Zile M.R.,
- Brutsaert D.L.
- MacFadyen R.J.,
- Barr C.S.,
- Struthers A.D.
- Zannad F.,
- Alla F.,
- Dousset B.,
- Perez A.,
- Pitt B.,
- RALES Investigators
- Mottram P.M.,
- Haluska B.,
- Leano R.,
- Cowley D.,
- Stowasser M.,
- Marwick T.H.
- Weinberger M.H.,
- Roniker B.,
- Krause S.L.,
- Weiss R.J.
- Lubien E.,
- DeMaria A.,
- Krishnaswamy P.,
- et al.
- Brilla C.G.,
- Funck R.C.,
- Rupp H.
- Querejeta R.,
- Lopez B.,
- Gonzalez A.,
- et al.
- Diez J.,
- Laviades C.,
- Mayor G.,
- Gil M.J.,
- Monreal I.
- Laviades C.,
- Mayor G.,
- Diez J.
- McGavigan A.D.,
- Moncrieff J.,
- Lindsay M.M.,
- Maxwell P.R.,
- Dunn F.G.
- Poulsen S.H.,
- Andersen N.H.,
- Heickendorff L.,
- Mogensen C.E.
- Marijianowski M.M.,
- Teeling P.,
- Mann J.,
- Becker A.E.
- Diez J.,
- Querejeta R.,
- Lopez B.,
- Gonzalez A.,
- Larman M.,
- Martinez Ubago J.L.