International Journal of Cardiology 167 (2013) 2963–2968

Contents lists available at ScienceDirect

International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard

Circulating biomarkers of collagen type I metabolism mark the right ventricular fibrosis and adverse markers of clinical outcome in adults with repaired tetralogy of Fallot☆ Chun-An Chen a, b, Wen-Yih Isaac Tseng c, d, Jou-Kou Wang a, Ssu-Yuan Chen e, Yen-Hsuan Ni a, Kuo-Chin Huang f, Yi-Lwun Ho b, g, Chung-I Chang h, Ing-Sh Chiu h, Mao-Yuan Marine Su c, Hsi-Yu Yu h, Ming-Tai Lin a, Chun-Wei Lu a, Mei-Hwan Wu a,⁎ a

Department of Pediatrics, National Taiwan University Hospital, Taipei, Taiwan Graduate Institute of Clinical Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan c Department of Medical Imaging, National Taiwan University Hospital, Taipei, Taiwan d Center for Optoelectronic Biomedicine, College of Medicine, National Taiwan University, Taipei, Taiwan e Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, Taipei, Taiwan f Department of Family Medicine, National Taiwan University Hospital, Taipei, Taiwan g Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan h Department of Surgery, National Taiwan University Hospital, Taipei, Taiwan b

a r t i c l e

i n f o

Article history: Received 17 April 2012 Received in revised form 21 August 2012 Accepted 31 August 2012 Available online 19 September 2012 Keywords: Cardiac fibrosis Collagen Magnetic resonance imaging Tetralogy of Fallot

a b s t r a c t Background: Right ventricular (RV) fibrosis is common in patients with repaired tetralogy of Fallot (rTOF). Although accumulating evidence indicates the role of circulating biomarkers of collagen metabolism in left ventricular fibrosis, rTOF data are lacking. This study examined the expression profile and clinical relevance of circulating biomarkers of collagen type I metabolism in rTOF patients. Methods: Serum biomarkers of collagen type I synthesis (carboxy-terminal propeptide of procollagen type I, PICP), degradation (carboxy-terminal telopeptide of collagen type I, CITP), and enzymes regulating collagen degradation (matrix metalloproteinases, and type I tissue inhibitor, TIMP-1) were measured in 70 rTOF and 91 control adults. All patients had complete clinical data and received cardiovascular magnetic resonance scans with late gadolinium enhancement (LGE). Results: Compared to the controls, rTOF patients had higher PICP levels (p b 0.001), PICP:CITP ratios (p b 0.001), and TIMP-1 concentrations (p b 0.001). Increasing PICP levels correlated with higher RV LGE scores (r = 0.427, p b 0.001), lower VO2max (r = − 0.428, p = 0.002), and larger RV volumes. Furthermore, stepwise multivariate linear regression analysis identified RV end-diastolic volume index > 150 mL/m2 (β = 40.52, p = 0.016), RV LGE score (β = 3.94, p = 0.008), and age (β = − 1.77, p = 0.011) as independent correlates of circulating PICP levels. Conclusions: Patients with rTOF exhibited a profibrotic state with excessive collagen type I synthesis and dysregulated degradation. Elevated circulating PICP levels might reflect RV fibrosis, and link to adverse markers of clinical outcome. © 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Regional replacement fibrosis or diffuse interstitial fibrosis is an integral feature of adverse ventricular remodeling in the failing heart [1]. In patients with repaired tetralogy of Fallot (rTOF), regional ☆ Funding This work was supported by a grant (NSC 97-2314-B-002-073-MY2) from the National Science Council, Taiwan, and by a grant (NCTRC200712) from the National Clinical Trial and Research Center, Taipei, Taiwan. ⁎ Corresponding author at: Department of Pediatrics, National Taiwan University Hospital, No. 7, Chung-Shan South Road, Taipei 100, Taiwan. Tel.: + 886 2 23123456x71524; fax: + 886 2 23412601. E-mail address: [email protected] (M.-H. Wu). 0167-5273/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijcard.2012.08.059

right ventricular (RV) fibrosis, as assessed by cardiovascular magnetic resonance (CMR) with late gadolinium enhancement (LGE), is a common phenomenon associated with adverse clinical outcomes [2,3]. Although the mechanisms leading to RV fibrosis may be multifactorial, progressive RV dysfunction caused by pulmonary regurgitation (PR) may play a pivotal role [2]. Other than using LGE as a noninvasive tool for cardiac fibrosis assessment, accumulating evidence indicates that circulating biomarkers from the metabolism of collagen type I are correlated with the extent of cardiac fibrosis [4,5]. In addition, certain biomarkers are linked to the clinical outcomes of ischemic heart disease, hypertension, and cardiomyopathies [4–7]. However, information is scant

2964

C.-A. Chen et al. / International Journal of Cardiology 167 (2013) 2963–2968

on these biomarkers' expression profiles and the clinical correlates in patients with rTOF. Therefore, we investigated the serum levels of biomarkers related to collagen type I synthesis (carboxy-terminal propeptide of procollagen type I, PICP), degradation (carboxy-terminal telopeptide of collagen type I, CITP), and enzymes regulating collagen turnover (matrix metalloproteinases, MMPs, and their type I tissue inhibitor, TIMP-1) in an adult cohort of rTOF. The relationships with LGE and markers of adverse outcome were assessed to delineate the potential clinical implications of these biomarkers of collagen type I metabolism. 2. Methods

intraclass correlation coefficients of intraobserver and interobserver agreement of 0.97 and 0.93, respectively. 2.5. Other investigations B-type natriuretic peptides were measured on the same day of CMR examination (Triage BNP test, Biosite, San Diego, California, USA). Aortic root dimensions at the sinotubular junction were determined from 2D echocardiographs [11], and an aortic root z score >2 indicated aortic root dilatation. Using Doppler echocardiography, LV diastolic function was assessed by peak early and late transmitral filling velocities, the ratio of peak early to late transmitral filling velocities, and the deceleration time of the early transmitral wave. We performed 24-h Holter ECG and cardiopulmonary exercise testing (upright bicycle ergometer) within 6 months of CMR examination. Peak oxygen consumption (VO2max) was measured and expressed as the percentage of predicted VO2max for age and sex.

2.1. Participants 2.6. Adverse markers of clinical outcome Between July 2007 and June 2009, consecutive patients with rTOF who received CMR examinations in our institute were eligible for this study. To avoid potential confounding effects on the natural metabolism of collagen, candidates were excluded if they met the following criteria: (1) age at CMR examination b18 years; (2) b10 years between complete repair and enrollment; (3) significant hemodynamic residuals other than PR, including a peak pressure gradient across the right ventricular outflow tract (RVOT) > 40 mmHg by echocardiography; (4) taking cardiac medications within 6 months of enrollment; and/or (5) having alcohol abuse, systemic hypertension, diabetes mellitus, liver disease, renal insufficiency, metabolic bone disease, autoimmune disease, and received operation or having a history of trauma within 6 months of study enrollment. After excluding 19 patients (11 patients were b18 years of age, 4 had hepatitis, 3 had a RVOT pressure gradient >40 mmHg, and 1 patient had a follow-up duration b10 years), a total of 70 patients were finally enrolled to the study. Ninety-one healthy volunteers of similar age and sex served as the control participants. The following clinical data were collected from patients' medical records and at the time of clinical evaluations: age at complete repair, with or without shunt creation before complete repair, types of RVOT reconstruction, body weight and height, blood pressure, and New York Heart Association functional class. All study participants provided written informed consent, and the institutional review committee approved the study protocol. This study conformed to the principles of the Helsinki Declaration. 2.2. Measurements of circulating biomarkers of collagen metabolism Venous blood samples were obtained from both patients and control participants. Serum was isolated within 30 min of phlebotomy and stored at −80°C prior to simultaneous analysis of biomarkers of collagen metabolism. Serum PICP concentration was determined by sandwich enzyme immunosorbent assay using the METRA EIA kit (Quidel, San Diego, California, USA), and CITP was determined using a commercial radioimmunoassay (Orion Diagnostica, Espoo, Finland) [7]. Intra- and inter-assay variations for determining PICP and CITP were all b6%. The PICP:CITP ratio was considered an index of the degree of coupling between the synthesis and degradation of collagen type I [7]. Serum levels of MMP-1, -2 -3, and -9, and TIMP-1 were analyzed using commercially available 2-site sandwich enzyme-linked immunosorbent assay (Quantikine R&D Systems, Minneapolis, Minnesota, USA) [8]. Intra-and inter-assay variations for determining MMPs and TIMP-1 were all b8%. 2.3. CMR imaging All scans were performed with a 1.5-T system (Sonata, Siemens, Erlangen, Germany) on the same day of blood sampling. A short-axis contiguous stack of balanced steady-state free-precession cine images from the atrioventricular ring to the apex was acquired. The following parameters were measured: RV and left ventricular (LV) end-diastolic volume index (EDVi); RV and LV end-systolic volume index (ESVi); RV and LV ejection fraction (EF); and RV and LV mass index. The PR fraction was quantified by non-breath-hold velocity-sensitive phase images in a plane transecting the long axis of RVOT [9]. 2.4. LGE imaging Using a segmented inversion-recovery sequence, LGE imaging was acquired 10 min after intravenous injection of gadolinium-diethylenetriaminepentaacetic acid (0.2 mmol/kg). Short-axis views from the cardiac base to the apex, as well as images of the RV and LV long-axis planes and 4-chamber views were acquired. We assessed the extent of LV LGE using a standard 17-segment model, with segments scored on a 5-point scale [10]. For the RV, we adopted the Babu-Narayan SV et al. segmentation system, which divided RV into 7 segments, and graded RV LGE according to the linear extent of enhanced myocardium in any given segment (0 = no LGE; 1 = up to 2 cm; 2 = 2 to 3 cm; 3 = greater than 3 cm) [2]. Enhancement in trabeculation was scored as 0 = no LGE; 1 = enhancement of one trabecula; 2 = enhancement of 2 to 4 trabeculae; and 3 = enhancement of more than 4 trabeculae. The superior and inferior RV insertion points were each scored as either 0 or 1 for the absence or presence of LGE, respectively. The maximum RV LGE score was 20. The reproducibility of RV LGE scores was high, with

The following clinical indicators in relation to circulating biomarkers were assessed, including (1) RVEDVi>150 mL/m2; (2) RVESVi>80 mL/m2; (3) RVEFb 47%; (4) LVEFb 55%; (5) exercise intolerance (≤70% of predicted VO2max); and (6) sustained tachyarrhythmia related to right heart volume overload. Based on previous studies, these indicators may serve as criteria for pulmonary valve replacement [12–14], and therefore are considered as markers of adverse clinical outcome for our present study. 2.7. Statistical analysis Data are expressed as percentage, mean ± standard deviation, or median (25th– 75th percentile), as appropriate. Patients were divided into normal and high PICP

Table 1 Patient characteristics. Median (25th–75th percentile) or mean ± SD Age (years) Male sex n (%) Body mass index (kg/m2) Age at repair (years) Shunt before repair n (%) Types of RVOT reconstruction Direct closure of RV n (%) RVOT patching n (%) Transannular patching n (%) Follow-up since repair (years) Systolic/diastolic blood pressure (mmHg) Echocardiographic data Pulmonary stenosis (mmHg) Aortic root z score Aortic root dilatation n (%) Transmitral E wave (cm/s) Transmitral A wave (cm/s) Transmitral E/A ratio DT of E wave (ms) QRS duration (ms) B-type natriuretic peptide (pg/mL) Exercise testing† Peak heart rate (bpm) VO2max (mL/min/kg) VO2max (% of predicted) CMR data RVEDVi (mL/m2) RVESVi (mL/m2) RVEF (%) RV mass index (g/m2) PR fraction (%) LVEDVi (mL/m2) LVESVi (mL/m2) LVEF (%) LV mass index (g/m2) LGE data RV LGE n (%) RV LGE scores LV LGE n (%) LV LGE scores

24.8 (21.3–30.8) 39 (56) 22.3 ± 3.6 4.1 (3.0–5.5) 11 (16) 8 (11) 30 (43) 32 (46) 20.7 (17.6–26.2) 114 ± 13/64 ± 11 14.9 (10.7–23.1) 2.1 ± 1.7 38 (54) 97.4 ± 22.0 57.6 ± 13.1 1.75 ± 0.48 201.8 ± 46.4 149 (140–162) 20.9 (10.5–35.5) 166 (154–178) 25.0 ± 3.6 70.1 ± 13.0 99 (81–124) 55 (47–77) 40.4 ± 7.4 19.2 ± 5.8 26.4 ± 18.8 61 (53–68) 20 (17–25) 65.8 ± 6.2 52.4 ± 12.4 70 (100) 6 (5–9) 13 (19) 0 (0–0)

†n = 50. A, peak late transmitral flow velocity; DT, deceleration time; E, peak early transmitral flow velocity.

C.-A. Chen et al. / International Journal of Cardiology 167 (2013) 2963–2968 groups based on the 90th percentile value of the PICP concentrations in control participants. Continuous variables were analyzed using the two-sample t test or the Mann– Whitney U test, after testing for normality. Categorical variables were analyzed by the chi-squared test or Fisher's exact test, as appropriate. The Spearman rank correlation coefficient was used for univariate correlation analysis. Receiver operating characteristic (ROC) curves enabled determination of the overall performance of serum PICP levels in relation to each marker of adverse clinical outcome. Significant correlates of circulating PICP levels were identified using stepwise multivariate linear regression. All statistical analyses were performed using the R statistical software, version 2.11.1 (R Foundation for Statistical Computing, Vienna, Austria) [15]. A value of p ≤ 0.05 was considered statistically significant.

2965

group. Serum levels of MMP-1, -2, and -3 were similar between patients and control participants. 3.3. Associations between biomarkers of collagen metabolism and LGE

3. Results

Right ventricular LGE score was positively correlated with serum PICP level (Fig. 2A) and the PICP:CITP ratio (Fig. 2B). Negative correlation between RV LGE score and CITP level was also noted (Fig. 2C). Conversely, the LV LGE score did not show a relation to any biomarkers of collagen metabolism in all study patients and those with LV LGE.

3.1. Patient characteristics

3.4. Clinical correlates of serum PICP level

The demographic, clinical, and image data of the 70 patients are summarized in Table 1. All patients were classified New York Heart Association functional class I, and none had concurrent use of cardiac medications. Only one patient had a history of sustained supraventricular tachycardia, and none had ventricular tachycardia. In our study cohort, all patients exhibited RV LGE. Typical RV LGE locations are shown in Fig. 1. In contrast, LV LGE was uncommon, and total LV LGE scores were low.

Since only one patient had a history of sustained arrhythmia, the relationship between clinical arrhythmia and biomarkers of collagen metabolism would not be investigated in the following analysis. Using the 90th percentile of serum PICP concentrations of the control participants as a cutoff value for PICP level, 21 patients with rTOF were classified as having a high PICP level (PICP > 102.7 ng/mL). Patients with a high PICP level were younger, more likely to be male, and had shorter follow-up duration (Table 3). These patients had higher RV mass index and RV LGE score, and RVEDVi also tended to be increased compared to those with a normal PICP level. Furthermore, patients with a high PICP level exhibited significantly lower exercise capacity (Table 3). Using univariate regression analysis, a negative correlation was found between PICP level and age (r = −0.384, P = 0.001), as well as between PICP level and the percentage of predicted VO2max (r =−0.428, p = 0.002). In addition, PICP positively correlated with RV mass index (r = 0.287, p = 0.016) and LVESVi (r = 0.248, p = 0.039) (Table 3).

3.2. Biomarkers of collagen metabolism Serum levels of biomarkers of collagen metabolism in patients with rTOF and control subjects are shown in Table 2. The most significant difference in the levels of these biomarkers was serum PICP, which was highly elevated in the patient group, compared to the controls (87.7 [68.9–115.6] vs 58.9 [48.4–71.9] ng/mL, p b 0.001). Serum CITP levels and the ratios of PICP:CITP were also higher in the patients than in the controls (CITP 3.6 [3.2–4.2] vs 3.1 [2.9–3.8] ng/mL; PICP: CITP ratio 23.1 [17.5–30.7] vs 16.5 [13.6–21.7]; both p b 0.001). The profiles of serum MMPs and TIMP-1 levels showed a pattern of decreased collagen degradation in patients with rTOF (Table 2). Compared with control participants, serum TIMP-1 concentrations were higher, but MMP-9 concentrations were lower in the patient

3.5. Relationships between serum PICP level and adverse markers of clinical outcome ROC curve analysis showed that serum PICP level is linked to adverse markers of clinical outcome, including significant RV dilatation (RVEDVi > 150 mL/m 2 or RVESVi > 80 mL/m 2; Fig. 3A and 3B) and

Fig. 1. Typical LGE (arrows) include the following regions: RVOT (A, B), region of surgical patching of the ventricular septal defect (C, D, G), superior and inferior RV insertion points (E), trabeculated myocardium (F, G), and the LV apex (H). Panels A, C, and H are presented in the LV outflow tract view, and the remaining panels are in the LV short-axis views.

2966

C.-A. Chen et al. / International Journal of Cardiology 167 (2013) 2963–2968

Table 2 Serum biomarkers of collagen metabolism in patients with rTOF and control participants.

PICP (ng/mL) CITP (ng/mL) PICP:CITP ratio MMP-1 (ng/mL) MMP-2 (ng/mL) MMP-3 (ng/mL) MMP-9 (ng/mL) TIMP-1 (ng/mL)

Patients

Controls

(n = 70)

(n = 91)

87.7 (68.9–115.6) 3.6 (3.2–4.2) 23.1 (17.5–30.7) 4.2 (2.4–7.0) 226.9 (153.3–261.6) 12.2 (8.2–16.6) 289.1 (214.4–409.0) 168.4 (145.7–180.3)

58.9 (48.4–71.9) 3.1 (2.9–3.8) 16.5 (13.6–21.7) 3.8 (2.4–6.0) 206.9 (181.3–227.6) 13.8 (8.5–20.3) 375.4 (278.0–505.5) 107.3 (91.8–124.2)

p value

b0.001 b0.001 b0.001 NS NS NS 0.001 b0.001

NS, not significant.

exercise intolerance (Fig. 3C). In contrast, PICP level was not predictive of decreased RVEF (b47%) or LVEF (b55%). Stepwise multivariate linear regression analysis was performed to identify significant correlates of serum PICP level after adjusting for potential confounding factors. Variables included in the regression model were those identified as significant above by univariate analyses, including markers of adverse outcome (RVEDVi > 150 mL/m 2, RVESVi > 80 mL/m 2, and the percentage of predicted VO2max ≤ 70%) and other univariate correlates (age, sex, follow-up duration, LVESVi, RV mass index, and RV LGE scores). Regression analysis results showed that RVEDVi > 150 mL/m 2 remained significantly correlated with serum PICP level (β = 40.52, p = 0.016), independent of RV LGE scores (β = 3.94, p = 0.008) and age (β = − 1.77, p = 0.011). 4. Discussion This study yielded two novel findings: (1) the rate of collagen type I synthesis significantly exceeds its rate of degradation in patients with rTOF in whom PR was the predominant hemodynamic residual. Such uncoupling of collagen degradation from its synthesis may be related to unique MMP and TIMP expression, and (2) greater circulating PICP levels not only correlated with more extensive RV LGE, but are also related to significant RV dilatation and exercise intolerance. These findings collectively suggest that altered collagen type I metabolism would be a possible pathophysiological mechanism for in vivo RV fibrosis. Furthermore, elevated PICP levels are linked to adverse markers of clinical outcome in patients with rTOF. Circulating PICP levels have been reported as a reliable index of monitoring collagen type I synthesis within the myocardium in a number of cardiac diseases that manifest as LV dysfunction [4,16]. But, the role of this indicator in relation to RV dysfunction remains unclear. This study demonstrated that serum PICP concentrations are elevated in rTOF patients and might reflect the extent of in vivo RV fibrosis assessed by LGE. LGE has been widely considered a marker

of regional fibrosis [17]. Though, LEG is less effective to visualize diffuse myocardial fibrosis, its presence usually indicated a concomitant increase in diffuse myocardial fibrosis [18]. Previous study had also shown that the extent of diffuse LV fibrosis was not statistically different between patients with rTOF and healthy controls [19]. Since we also observed the rarity of LV LGE in our patients, the elevation of serum PICP concentration is unlikely from LV fibrosis. Therefore, although LGE scores might underscore the contribution from diffuse myocardial fibrosis, the LGE scores of our study patients still could reflect the extent of RV fibrosis, and hence, correlate well with circulating PICP levels. Increased PICP:CITP ratios were found in rTOF patients. Such a profile suggests that, in addition to abnormally upregulated collagen synthesis, collagen degradation is uncoupled from its synthesis. Furthermore, discordant MMP and TIMP expression may lead to reduced MMPs activity, being unable to degrade overproduced collagen. This explains why the amount of circulating CITP is disproportionate to elevated PICP concentrations, despite the increases in CITP absolute concentration. Increased TIMP-1 is associated with LV dysfunction in patients with hypertensive heart disease [20]. However, we did not find a correlation between TIMP-1 levels and ventricular function in our patients. Aortic root dilatation, which may alter the MMP and TIMP expression, is common in patients with rTOF [11]. Previous studies on the dilated ascending aortic tissue of patients with Marfan syndrome, which has numerous histological similarities with the aorta in rTOF patients [21], showed a MMP and TIMP expression profile (either elevated or unchanged MMP-9 levels, and relatively unchanged TIMP-1 levels in the aortic tissue) that is completely different from that observed in our patients [22]. Therefore, the specific portfolio of serum biomarkers of type I collagen metabolism expressed in our patients is unlikely to reflect the process of aortic remodeling. In our patients, whose hemodynamic burden of PR was substantial, the finding that severe RV dilatation was related to abnormally high PICP levels may support the inference that elevated PICP levels reflect adverse RV remodeling in response to chronic volume overload. Although negative ventricular–ventricular interactions in rTOF patients may lead to LV remodeling [23], thus altering circulating PICP levels as well, we found generally preserved LV function in our patients, and except for LVESVi, circulating PICP levels had no direct correlation with either LV systolic or diastolic function. The relationship between circulating PICP level and LVESVi became insignificant after RV function and other relevant factors were taken into consideration in multivariate analysis. Therefore, we suggest that it is the remodeling of RV, and not that of LV, that contributes to elevated PICP levels. Besides, our data showed that age is negatively related to serum PICP levels. This finding has been reported before [24,25]. Although the exact explanation for this phenomenon is still unclear, the negative correlation between age

Fig. 2. Relationship between the circulating biomarkers and RV LGE scores.

C.-A. Chen et al. / International Journal of Cardiology 167 (2013) 2963–2968

2967

Table 3 Relationship of serum PICP level to demographic, clinical, and image data in patients with rTOF.

Age (years) Male sex n (%) Age at repair (years) Transannular patching n (%) Follow-up since repair (years) Pulmonary stenosis (mmHg) Aortic root z score Transmitral E wave (cm/s) Transmitral A wave (cm/s) Transmitral E/A ratio DT of E wave (ms) QRS duration (ms) B-type natriuretic peptide (pg/mL) Peak heart rate at exercise (bpm)† VO2max (% of predicted)† RVEDVi (mL/m2) RVESVi (mL/m2) RVEF (%) RV mass index (g/m2) PR fraction (%) RV LGE score LVEDVi (mL/m2) LVESVi (mL/m2) LVEF (%) LV mass index (g/m2) LV LGE scores

Normal PICP level (n = 49)

High PICP level (n = 21)

p value*

Spearman Correlation r (p)*

26.2 (22.7–32.8) 23 (47) 4.2 (3.2–5.5) 21 (43) 23.0 (19.4–28.4) 14.9 (9.0–23.9) 2.0 ± 1.8 95.6 ± 22.7 57.4 ± 9.9 1.70 ± 0.40 204.1 ± 46.5 149 (141–164) 17.9 (8.1–32.1) 166 (154–173) 72.4 ± 13.3 96 (76–114) 53 (45–72) 40.9 ± 7.1 18.2 ± 5.1 24.2 ± 18.1 6 (5–8) 59 (53–65) 19 (16–23) 65.9 ± 5.7 52.5 ± 13.9 0 (0–0)

22.6 (20.0–28.7) 16 (76) 3.8 (2.9–4.8) 11 (52) 17.7 (16.3–22.2) 15.5 (13.9–26.5) 2.6 ± 1.4 101.6 ± 20.0 58.0 ± 19.0 1.89 ± 0.61 196.6 ± 47.0 149 (133–166) 23.6 (12.2–41.8) 180 (143–186) 62.8 ± 8.6 115 (74–153) 69 (38–94) 39.1 ± 8.1 21.5 ± 6.9 31.6 ± 19.9 10 (7–12) 64 (54–76) 20 (17–23) 65.7 ± 7.5 52.1 ± 8.2 0 (0–2)

0.022 0.035 NS NS 0.008 NS NS NS NS NS NS NS NS NS 0.023 0.068 NS NS 0.033 NS 0.001 NS NS NS NS NS

−0.384 (0.001) – −0.227 (0.059) – −0.398 (0.001) 0.211 (0.08) NS NS NS NS NS NS NS NS −0.428 (0.002) 0.207 (0.086) NS NS 0.287 (0.016) NS 0.427 (0.001) 0.219 (0.069) 0.248 (0.039) NS NS NS

*Only p value b 0.1 is shown. †n= 50. A, peak late transmitral flow velocity; DT, deceleration time; E, peak early transmitral flow velocity. NS, not significant.

and PICP levels may confound the relationship between RV remodeling and PICP levels since progressive RV dilatation with increasing age is common in patients with significant PR [26]. In multivariate analysis, we identified that both severe RV dilatation (RVEDVi > 150 mL/m 2) and age are related to greater circulating PICP levels independently. This result further substantiated the relation between adverse RV remodeling and PICP level after adjustment for age. Results from this study may have some clinical implications. Our present study provided evidence for the first time supporting the notion that circulating PICP level links to severe RV dilatation and RV fibrosis, both have been shown to be predictive of adverse clinical outcomes in rTOF [2,3,27]. Longitudinal follow-up study is needed to substantiate the role of circulating PICP level in the prediction of future adverse cardiac events. Besides, collagen type I overproduction and dysregulated degradation in patients with rTOF may imply

therapeutic benefits from intervening in collagen turnover. Blockade of the renin–angiotensin–aldosterone system could suppress myocardial collagen synthesis and improve chances of survival in patients with heart failure [28]. A recent randomized clinical trial also demonstrated beneficial effects of ramipril on biventricular long-axis function in patients with rTOF [29]. Further studies to determine the potential benefits from pharmacological modulation of collagen type I metabolism in rTOF are mandatory. This study has some limitations. First, the limitation to serum sampling of biomarkers of collagen metabolism is that the myocardium may not be the only source of these biomarkers. However, the stringent exclusion criteria used in this study would help eliminate significant changes in these biomarkers originating from non-cardiac sources. Second, we relied on circulating biomarkers and LGE to assess cardiac fibrosis, although no histological changes had been

Fig. 3. ROC curves for serum PICP concentrations in relation to adverse markers of clinical outcome. Area under curve (AUC) and its 95% confidence interval (in parenthesis) are shown.

2968

C.-A. Chen et al. / International Journal of Cardiology 167 (2013) 2963–2968

validated. Quantitative analysis of extracellular matrix components may be possible in surgical specimens at the time of pulmonary valve replacement in the future. Third, only adverse markers of clinical outcome were assessed in this study, but not the real clinical outcome. Finally, we enrolled patients with no more than mild residual RVOT stenosis, and the study population was relatively young with a good functional class. Whether the expression profiles of collagen metabolism seen in our study patients can also be observed in a broader range of patients warrants further study. In conclusion, increased collagen type I synthesis and dysregulated collagen degradation might be involved in pathological processes associated with ventricular remodeling in patients with rTOF. Elevated concentrations of circulating PICP are correlated with image evidence of RV fibrosis, and are linked to RV dilatation and decreased exercise capacity.

[11] [12]

[13]

[14]

[15]

[16]

Acknowledgments

[17]

The authors would like to thank Ms. Chiu-Yi Hsu and Ms. Yi-An Shr for assistance with the CMR data analysis. All authors of this manuscript have certified that they comply with the Principles of Ethical Publishing in the International Journal of Cardiology.

[18]

[19]

[20]

References [1] Mewton N, Liu CY, Croisille P, Bluemke D, Lima JA. Assessment of myocardial fibrosis with cardiovascular magnetic resonance. J Am Coll Cardiol 2011;57: 891-903. [2] Babu-Narayan SV, Kilner PJ, Li W, et al. Ventricular fibrosis suggested by cardiovascular magnetic resonance in adults with repaired tetralogy of fallot and its relationship to adverse markers of clinical outcome. Circulation 2006;113:405-13. [3] Wald RM, Haber I, Wald R, Valente AM, Powell AJ, Geva T. Effects of regional dysfunction and late gadolinium enhancement on global right ventricular function and exercise capacity in patients with repaired tetralogy of Fallot. Circulation 2009;119:1370-7. [4] Querejeta R, Lopez B, Gonzalez A, et al. Increased collagen type I synthesis in patients with heart failure of hypertensive origin: relation to myocardial fibrosis. Circulation 2004;110:1263-8. [5] Izawa H, Murohara T, Nagata K, et al. Mineralocorticoid receptor antagonism ameliorates left ventricular diastolic dysfunction and myocardial fibrosis in mildly symptomatic patients with idiopathic dilated cardiomyopathy: a pilot study. Circulation 2005;112:2940-5. [6] Radovan J, Vaclav P, Petr W, et al. Changes of collagen metabolism predict the left ventricular remodeling after myocardial infarction. Mol Cell Biochem 2006;293: 71-8. [7] Ho CY, Lopez B, Coelho-Filho OR, et al. Myocardial fibrosis as an early manifestation of hypertrophic cardiomyopathy. N Engl J Med 2010;363:552-63. [8] Tolosana JM, Mont L, Sitges M, et al. Plasma tissue inhibitor of matrix metalloproteinase-1 (TIMP-1): an independent predictor of poor response to cardiac resynchronization therapy. Eur J Heart Fail 2010;12:492-8. [9] Rebergen SA, Chin JG, Ottenkamp J, van der Wall EE, de Roos A. Pulmonary regurgitation in the late postoperative follow-up of tetralogy of Fallot. Volumetric quantitation by nuclear magnetic resonance velocity mapping. Circulation 1993;88:2257-66. [10] Cerqueira MD, Weissman NJ, Dilsizian V, et al. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: a statement for healthcare professionals from the Cardiac Imaging Committee of the Council on

[21]

[22]

[23]

[24] [25]

[26] [27]

[28]

[29]

Clinical Cardiology of the American Heart Association. Circulation 2002;105: 539-42. Niwa K, Siu SC, Webb GD, Gatzoulis MA. Progressive aortic root dilatation in adults late after repair of tetralogy of Fallot. Circulation 2002;106:1374-8. Buechel ER, Dave HH, Kellenberger CJ, et al. Remodelling of the right ventricle after early pulmonary valve replacement in children with repaired tetralogy of Fallot: assessment by cardiovascular magnetic resonance. Eur Heart J 2005;26: 2721-7. Geva T. Repaired tetralogy of Fallot: the roles of cardiovascular magnetic resonance in evaluating pathophysiology and for pulmonary valve replacement decision support. J Cardiovasc Magn Reson 2011;13:9. Geva T, Gauvreau K, Powell AJ, et al. Randomized trial of pulmonary valve replacement with and without right ventricular remodeling surgery. Circulation 2010;122:S201-8. R Development Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2006. Available at: http://www.R-project.org. Lopez B, Querejeta R, Gonzalez A, Sanchez E, Larman M, Diez J. Effects of loop diuretics on myocardial fibrosis and collagen type I turnover in chronic heart failure. J Am Coll Cardiol 2004;43:2028-35. Moon JC, Reed E, Sheppard MN, et al. The histologic basis of late gadolinium enhancement cardiovascular magnetic resonance in hypertrophic cardiomyopathy. J Am Coll Cardiol 2004;43:2260-4. Iles L, Pfluger H, Phrommintikul A, et al. Evaluation of diffuse myocardial fibrosis in heart failure with cardiac magnetic resonance contrast-enhanced T1 mapping. J Am Coll Cardiol 2008;52:1574-80. Broberg CS, Chugh SS, Conklin C, Sahn DJ, Jerosch-Herold M. Quantification of diffuse myocardial fibrosis and its association with myocardial dysfunction in congenital heart disease. Circ Cardiovasc Imaging 2010;3:727-34. Ahmed SH, Clark LL, Pennington WR, et al. Matrix metalloproteinases/tissue inhibitors of metalloproteinases: relationship between changes in proteolytic determinants of matrix composition and structural, functional, and clinical manifestations of hypertensive heart disease. Circulation 2006;113:2089-96. Tan JL, Davlouros PA, McCarthy KP, Gatzoulis MA, Ho SY. Intrinsic histological abnormalities of aortic root and ascending aorta in tetralogy of Fallot: evidence of causative mechanism for aortic dilatation and aortopathy. Circulation 2005;112: 961-8. Ikonomidis JS, Jones JA, Barbour JR, et al. Expression of matrix metalloproteinases and endogenous inhibitors within ascending aortic aneurysms of patients with Marfan syndrome. Circulation 2006;114:I365-70. Davlouros PA, Kilner PJ, Hornung TS, et al. Right ventricular function in adults with repaired tetralogy of Fallot assessed with cardiovascular magnetic resonance imaging: detrimental role of right ventricular outflow aneurysms or akinesia and adverse right-to-left ventricular interaction. J Am Coll Cardiol 2002;40:2044-52. Melkko J, Niemi S, Risteli L, Risteli J. Radioimmunoassay of the carboxyterminal propeptide of human type I procollagen. Clin Chem 1990;36:1328-32. Piovesan A, Berruti A, Osella G, et al. Serum levels of carboxyterminal propeptide of type I procollagen in cancer patients with osteoblastic and osteolytic bone metastases. Int J Biol Markers 1994;9:243-6. Bouzas B, Kilner PJ, Gatzoulis MA. Pulmonary regurgitation: not a benign lesion. Eur Heart J 2005;26:433-9. Knauth AL, Gauvreau K, Powell AJ, et al. Ventricular size and function assessed by cardiac MRI predict major adverse clinical outcomes late after tetralogy of Fallot repair. Heart 2008;94:211-6. Zannad F, Alla F, Dousset B, Perez A, Pitt B. Limitation of excessive extracellular matrix turnover may contribute to survival benefit of spironolactone therapy in patients with congestive heart failure: insights from the randomized aldactone evaluation study (RALES). Rales Investigators. Circulation 2000;102:2700-6. Babu-Narayan SV, Uebing A, Davlouros PA, et al. Randomised trial of ramipril in repaired tetralogy of Fallot and pulmonary regurgitation The APPROPRIATE study (Ace inhibitors for Potential PRevention Of the deleterious effects of Pulmonary Regurgitation In Adults with repaired TEtralogy of Fallot). Int J Cardiol 2012;154:299-305.