Pediatr Cardiol DOI 10.1007/s00246-014-0876-8

ORIGINAL ARTICLE

Reverse Ventricular Remodeling and Improved Ventricular Compliance After Heart Transplantation in Infants and Young Children Kanwal M. Farooqi • Leo Lopez • Robert H. Pass Daphne T. Hsu • Jacqueline M. Lamour

•

Received: 28 November 2013 / Accepted: 24 January 2014 Ó Springer Science+Business Media New York 2014

Abstract After heart transplantation (HT) in infants and young children, environmental and intrinsic factors may lead to changes in the geometry and compliance of the donor heart. Serial demographic, clinical, hemodynamic, and echocardiographic data were obtained from HT recipients younger than 4 years of age. Echocardiographic chamber measurement z-scores were compared using recipient body surface area from the time of HT to 1 week, 3 months, and last follow-up visit. Left ventricular enddiastolic volume (LVEDV) z-scores were correlated with pulmonary capillary wedge pressure (PCWP) at each time point. Heart transplantation was performed for 13 children between March 2009 and December 2012, 9 of whom (69 %) were boys. The median age at HT was 8 months (range, 4–43 months), and the mean follow-up period was 13 ± 7 months. Left ventricular end-diastolic dimension zscores decreased significantly (p = 0.03) between HT and 1 week, then increased from 1 week to 3 and 12 months. (-1.32 ± 1.7, -0.71 ± 1.8, 0.41 ± 2.1, 0.79 ± 2.3, respectively). A positive relationship (R2 = 0.48) between the LVEDV z-score and PCPW was present at the last follow-up visit. For infants and young children, the allograft demonstrates appropriate growth by 1 year after HT. Left ventricular compliance improves over time.

K. M. Farooqi (&) Division of Pediatric Cardiology, Department of Pediatrics, The Mount Sinai Medical Center, New York, NY, USA e-mail: [email protected] L. Lopez
R. H. Pass
D. T. Hsu
J. M. Lamour Division of Pediatric Cardiology, Department of Pediatrics, The Children’s Hospital at Montefiore/Albert Einstein College of Medicine, 3415 Bainbridge Avenue, Rosenthal 1, Bronx, NY 10467, USA e-mail: [email protected]

Keywords Pediatric heart transplantation
Cardiac transplant growth
Cardiac transplant ventricular compliance

Introduction Pediatric heart transplantation is the only option for the long-term survival of children with end-stage heart failure. In children, donor–recipient size mismatch, normal somatic growth, and the effect of immunosuppressive agents such as steroids may be factors that affect the growth of the allograft. Early studies of pediatric heart transplant recipients from infants to adolescents demonstrated that left ventricular (LV) volumes increase in proportion to the recipient’s body size. In some cases, the mass-to-volume ratio has been elevated, with and without the presence of LV hypertrophy [2, 19, 20]. The majority of patients in these early studies were treated with triple immunosuppression consisting of cyclosporine, azathioprine, and prednisone. More recent studies have shown that donor–recipient mismatch may contribute to the finding of LV hypertrophy in the immediate posttransplantation period but does not affect longer-term growth of the transplanted heart [6, 9]. In a study of pediatric transplant recipients ranging in age from 0 to 17 years who had been treated with triple-drug immunosuppression consisting of cyclosporine, myocophenolate mofetil (MMF), steroids, the findings showed that right and left ventricular end-diastolic diameters, volumes, and myocardial mass were increased during early follow-up assessment and subsequently decreased during the first year after transplantation. These dimensions returned to within the normal range and then increased appropriately in later follow-up assessments [5].

123

Pediatr Cardiol

We sought to demonstrate the changes in LV dimensions that occurred in a contemporary cohort of infants and young children treated with a steroid-sparing regimen. The younger pediatric population was specifically chosen because the steep growth trajectory in infants and young children may demonstrate more rapid changes in dimensions and adaptive growth than the growth trajectory of older children and adolescents. We also studied the change in compliance of the transplanted heart as it adapts to the recipient physiology.

Methods Patients A retrospective review of all the patients who underwent transplantation at the Children’s Hospital at Montefiore between March 2009 and December 2012 was undertaken to identify recipients younger than 4 years at the time of transplantation. The age, sex, height, weight, and body surface area (BSA) of both the donor and the recipient were collected. The indication for transplantation, the donor– recipient BSA ratio, and the ischemic time for the transplantation also were retrieved. The posttransplantation course including immunosuppression treatment, rejection history, and catheterization hemodynamics was recorded. Immunosuppression Regimen All but one patient received induction with methylprednisolone and antithymocyte globulin (ATG) and were treated using a steroid-sparing maintenance immunosuppression protocol. Methylprednisolone was administered intraoperatively with the release of cross-clamp at a dose of 20 mg/kg and then continued on postoperative day (POD) 1 at a dose of 2 mg/kg, which was weaned to 0.25 mg/kg by POD 5. During PODs 1 to 5, ATG was administered at a dose of 1.5 mg/kg. Tacrolimus and MMF were instituted on PODs 2 to 3. Tacrolimus was administered twice daily at a total dose of 0.05–0.1 mg/kg/day to achieve levels of 8–12 ng/ml. Twice daily administration of MMF was initiated at a total dose of 1,200 mg/m2/day. Corticosteroids were discontinued 6 ± 2 days after transplantation for 12 of the 13 patients in this study. One patient did not undergo induction with ATG and received prednisone for the first 3 months after transplantation.

(LVEDD), left ventricular end-diastolic volume (LVEDV), and LV mass (LVM). These measurements were adjusted for BSA using a database of echo measurements from normal children, and z-scores were calculated for each measurement [18]. The LVEDD was measured in the parasternal short-axis view at end diastole between the papillary muscles. The LVEDV calculation used the LV cross-sectional area (LVEDA) measured in the parasternal short-axis view at end diastole and the LV length (LVEDL) measured from the apical four-chamber view at end diastole from the midpoint of the mitral valve annulus to the apical endocardium. The LVEDV was calculated using the area length method as follows: LVEDV = 5/6 9 (LVEDA 9 LVEDL). The LVM was calculated by subtracting the endocardial LVEDV from the epicardial LVEDV and multiplying the difference by the myocardial specific density (1.05 g/ml). The LVM/V ratio was calculated by dividing the LVM by the LVEDV. The LV dimensions were collected at three time points: from transplantation to 1 week, 3 months, and last follow-up visit. Cardiac Catheterization Cardiac catheterization with endomyocardial biopsy was performed routinely for all patients after transplantation to obtain right-sided hemodynamics and to assess for cellular rejection. The catheterizations were performed initially 2 weeks after transplantation and then monthly afterward. Venous access was obtained either from a femoral vein or via an internal jugular approach. The pulmonary capillary wedge pressure (PCWP) was used as a surrogate for LV enddiastolic pressures. The cardiac catheterization performed in closest proximity to the designated time point at which LV dimensions were measured was used for correlation analysis. The time between the echocardiograms and the cardiac catheterizations were 6 ± 4 days at 1 week, 11 ± 9 days at 3 months, and 28 ± 21 days at the last follow-up visit. Statistical Analysis The z-scores were compared between donor and recipient at 1 week, 3 months, and last follow-up visit by one-way analysis of variance (ANOVA), with a p value lower than 0.05 considered significant. The PCWP and LV measurements were correlated using regression analysis.

Echocardiographic Measurements

Results

The echocardiographic dimensions of the donor were collected from the data provided at the time of the donor heart acceptance. The LV dimensions of the recipient that were recorded included left ventricular end-diastolic dimension

Patient Characteristics

123

Heart transplantation was performed for 13 children younger than 4 years between March 2009 and December

Pediatr Cardiol

2012. Nine of these patients (69 %) were boys. The median age at transplantation was 8 months, and the mean followup period was 13 ± 7 months. The mean recipient BSA was 0.38 ± 0.09 kg/m2, and the mean donor–recipient BSA ratio was 1.15 ± 0.25. The patient characteristics including primary diagnoses are detailed in Table 1. At the time of the first biopsy, 10 patients were inpatients, and all patients were receiving furosemide. Two of the inpatients were receiving inotropic support. None of the patients in this study group were hypertensive during the entire follow-up period. There was no significant cellular rejection (greater than 1 R) noted on the endomyocardial biopsies performed during the entire follow-up period. Of the 13 biopsies, 4 at the initial time point showed evidence of myocardial edema or inflammation. Linear Growth The patients showed appropriate linear growth over time. The change in BSA over time is plotted in Fig. 1. Table 1 Characteristics of the 13 study patients Median age at transplantation (months)

8

Range (months)

4–43

Male/female (n)

9/4

Changes in LV Dimensions The changes in LV dimensions over time are shown in Table 2 and Fig. 2. The mean LVM, LVEDD, and LVEDV were within the normal range at all time points. The LVEDD z-scores increased significantly (p = 0.03) from the time of transplantation to the last follow-up visit. No significant change in the LVEDV (p = 0.11) or LVM (p = 0.23) z-score occurred during the follow-up period (Fig. 2). Figure 3 compares the mean LVM/V ratio z-scores over time. The mean LVM/V ratio z- score was elevated compared with normal 1 week after transplantation, with a zscore of 2.77. The LVM/V ratio decreased by the 3-month time point (z-score, 1.37) and continued to normalize over time, reaching a normal value at the last evaluation (zscore, -0.06). Relationship of LVEDV and PCWP The relationship between LVEDV and the PCPW was determined at each time point, as shown in Fig. 4. No correlation between LVEDV and PCWP was found 1 week after transplantation. At the last follow-up visit, LVEDV and PCWP showed a significant correlation (R2 = 0.48).

Indication for transplantation (n) Cardiomyopathy

12

Congenital heart disease

1

Donor BSA (kg/m2)

0.43 ± 0.14 2

Recipient BSA (kg/m )

0.38 ± 0.09

Table 2 Left ventricular (LV) dimension z-scores LV dimension

Donor

1 Week

3 Months

Last followup

Donor–recipient BSA ratio

1.15 ± 0.26

LVEDD

–1.32 ± 1.7

–0.71 ± 1.8

0.41 ± 2.1

0.79 ± 2.3

Ischemic time (min)

209.7 ± 62.1

LVEDV

–1.1 ± 1.2

–0.48 ± 2.1

0.29 ± 2.1

0.78 ± 2.6

Follow-up (days)

395 ± 228

LVM

0.11 ± 0.9

0.75 ± 1.5

1.03 ± 1.0

1.37 ± 1.2

2.77 ± 3.6

1.37 ± 2.5

–0.06 ± 2.5

Values are expressed as mean ± standard deviation unless otherwise noted BSA body surface area

LVM/V

LVEDD LV end-diastolic dimension, LVEDV, LV end-diastolic volume, LVM, LV mass, LVM/V

Fig. 1 Change in body surface area (BSA) over time

123

Pediatr Cardiol

Discussion

Fig. 2 Change in left ventricular (LV) dimensions after transplantation. All values are expressed as z-scores. LVEDD, LV end-diastolic diameter; LVEDV, LV end-diastolic volume; LVM, LV mass

Fig. 3 Changes in left ventricular (LV) mass–volume ratio z-scores over time

Fig. 4 Regression analysis of the correlation between left ventricular end-diastolic volume (LVEDV) and pulmonary capillary wedge pressure (PCWP) (mmHg). The R2 value was 0.0099 at the initial followup time, 0.3676 at the 3-month follow-up time, and 0.4768 at the last follow-up time

123

In characterizing the changes that occur in a transplanted heart as it adjusts to the donor circulation, we sought to determine how LV dimensions evolve over time. The relationship between the LVEDV and the PCWP also was analyzed using data from echocardiograms and cardiac catheterization to understand better the change in compliance of the donor heart. The LV dimensions were normal within 1 week after transplantation in this group of infants and young children undergoing transplantation. The LVEDD z-score increased significantly during the first year after transplantation but remained within the normal range. This growth was independent of factors such as donor–recipient BSA ratio, age at transplantation, sex of the recipient, and ischemic time. These findings confirm results reported in other studies of pediatric patients that included adolescents and also used various methods for normalization of the echocardiographic variables [2, 6]. Our results differ from those reported by Delmo Walter et al. [5], who found that right ventricular (RV) and LV dimensions were higher at the initial time point (30 days after transplantation) than measurements obtained 1 year after transplantation. Prior studies have described LV hypertrophy in the early posttransplantation period. However in our contemporary cohort of very young patients receiving a steroid-sparing immunosuppressive protocol, the LVM was normal 1 week after transplantation. The LV mass in our cohort increased over time but remained within a normal range during the follow-up period. Zales et al. [20] reported similar findings, with LVM and the LVM/V ratio remaining within normal limits in older patients.

Pediatr Cardiol

In studies noting LV hypertrophy during the initial postoperative period, regression and resolution occurred 1 year after transplantation [9, 17]. The initial increased LVM was related to a higher donor–recipient size mismatch, with donor–recipient weight ratios greater than 1.2 associated with increased LVMI [9]. The mean decrease in LVMI after the immediate posttransplantation period was found to be related to a donor–recipient weight ratio greater than 1.5 [17]. Steroid administration and cyclosporineinduced hypertension are hypothesized in some studies to be the mechanisms behind LVH in the adult population, although other studies have not found this association to be significant [10, 13, 14]. Cyclosporine was not part of our immunosuppression regimen, and our patients were not hypertensive during the duration of the follow-up period for this study, perhaps explaining our findings. There are reports of tacrolimus possibly resulting in a hypertrophic obstructive cardiomyopathy in pediatric patients after liver and small bowel transplantations. This finding has not been described after cardiac transplantation [1, 3]. At the earliest time point after transplantation, the LVM/ V ratio was found to be abnormally high. This ratio normalized over time. Postoperative myocardial inflammation partially secondary to cardiopulmonary bypass as well as reperfusion injury results in fluid retention in the myocardium, which may partially explain this observation. Myocardial edema or inflammation was in fact noted in multiple biopsies at the initial time point. Findings have shown LV mass to increase concurrently with a decrease in LV compliance after ischemia and reperfusion on the cardiopulmonary bypass in canine models [11, 12]. As expected, this myocardial edema appears to resolve over time. The compliance of a transplanted heart in infants or young children has not been investigated previously. Over time, the pressure–volume relationship in our cohort transitioned toward improved compliance. In the early postoperative period, small increases in LVEDD and LVEDV correlated with large changes in the PCWP, suggesting poor compliance of the LV. Myocardial edema has been shown to alter the pressure– volume relationship of the LV in a manner that decreases the ability of the LV to distend [4]. In animal models, an increased myocardial water content and an increased LV mass have been associated with decreased LV compliance [4, 15, 16]. In our cohort, there appeared to be improvement in the stiffness of the ventricle over time, with increases in the LVEDD and LVEDV resulting in less significant increases in the PCWP. Taking into consideration the immediate postoperative period and the multiple possible sources contributing to myocardial edema of the transplanted heart, these findings

are not surprising. Ischemic time, fluid administration in the operating room, inflammation of the myocardium after having undergone bypass, and hypotonic cardioplegia perfusing the coronary arteries all may contribute to the poor compliance of the heart [8]. As the myocardium recovers and diuresis occurs, this edema improves. This appeared to be a gradual process that continued to improve during the follow-up period of 14 ± 6 months. Although impaired compliance has been related to transplant rejection, the trend toward normal and the failure to demonstrate significant rejection on endomyocardial biopsy makes this an unlikely mechanism for the change in the pressure– volume relationship [7]. Because this was a retrospective study, a lag occurred between some of the echocardiograms and the cardiac catheterizations performed. As the patients were farther from the time of the transplantation, the catheterizations and echocardiograms became less frequent. This rendered it challenging to obtain data from the two studies that were temporally better related. Ideally, the hemodynamic data and the chamber dimensions would be measured within a short period to avoid changes in loading conditions affecting our results. In addition, our small sample size limited our data. In conclusion, we report normal growth of the transplanted heart in our cohort of infants and young children. There was a trend toward improved compliance of the transplanted heart over time. Further follow-up evaluation is needed for a better understanding of whether this trend continues over time, taking into consideration the possibility of long-term rejection influencing compliance of the transplanted heart.

References 1. Atkison P, Joubert G, Barron A, Grant D, Paradis K, Seidman E, Wall W, Rosenberg H, Howard J, Williams S, Stiller C (1995) Hypertrophic cardiomyopathy associated with tacrolimus in paediatric transplant patients. Lancet 345:894–896 2. Bernstein D, Kolla S, Miner M, Miner M, Pitlick P, Griffin M, Starnes V, Rowan R, Billingham M, Baum D (1992) Cardiac growth after pediatric heart transplantation. Circulation 85:1433–1439 3. Chang RK, Alzona M, Alejos J, Jue K, McDiarmid SV (1998) Marked left ventricular hypertrophy in children on tacrolimus (FK506) after orthotopic liver transplantation. Am J Cardiol 81:1277–1280 4. Cross CE, Rieben PA, Salisbury PF (1961) Influence of coronary perfusion and myocardial edema on pressure–volume diagram of left ventricle. Am J Physiol 201:102–108 5. Delmo Walter EM, Huebler M, Stamm C, Alexi-Meskishvili, Weng Y, Berger F, Hetzer R (2011) Adaptive growth and remodeling of transplanted hearts in children. Eur J Cardiothorac Surg 40:1374–1383 6. Delmo Walter EM, Huebler M, Schubert S, Lehmkuhl H, Weng Y, Berger F, Hetzer R (2012) Influence of size disparity of

123

Pediatr Cardiol

7. 8.

9.

10.

11.

12.

13.

14.

transplanted hearts on cardiac growth in infants and children. J Thorac Cardiovasc Surg 143:168–177 Hsu DT, Spotnitz HM (1990) Echocardiographic diagnosis of cardiac allograft rejection. Prog Cardiovasc Dis 33:149–160 Hsu DT, Weng ZC, Nicolosi AC et al (1993) Quantitative effects of myocardial edema on the left ventricular pressure–volume relation: influence of cardioplegia osmolarity over two hours of ischemic arrest. J Thorac Cardiovasc Surg 106:651–657 Kertesz NJ, Gajarski RJ, Towbin JA, Geva T (1995) Effect of donor–recipient size mismatch on left ventricular remodeling after pediatric orthotopic heart transplantation. Am J Cardiol 76:1167–1172 Kimball TR, Witt SA, Daniels SR, Khoury PR, Meyer RA (1996) Frequency and significance of LV thickening in transplanted hearts in children. Am J Cardiol 77:77–80 Lazar HL, Haasler GB, Collins RH et al (1982) Mechanisms of altered ventricular compliance following ischemia, using twodimensional echocardiography. Curr Surg 39:253–255 Lazar HL, Haasler GB, Collins RH et al (1985) Compliance, mass, and shape of the canine left ventricle after global ischemia analyzed with two-dimensional echocardiography. J Surg Res 39:199–208 McKoy RC, Uretsky BF, Kormos R, Hardesty RL, Griffith BP, Salerni R (1988) Left ventricular hypertrophy in cyclosporineinduced systemic hypertension after cardiac transplantation. Am J Cardiol 62:1140–1142 Rowan RA, Billingham ME (1990) Pathologic changes in the long-term transplanted heart: a morphometric study of

123

15.

16.

17.

18.

19.

20.

myocardial hypertrophy, vascularity, and fibrosis. Hum Pathol 21:767–772 Salisbury PF, Cross CE, Rieben PA (1960) Distensibility and water content of heart muscle before and after injury. Circ Res 8:788–793 Salisbury PF, Cross CE, Katsuhara K et al (1961) Factors which initiate or influence edema in the isolated dog’s heart. Circ Res 9:601–606 Shirali GS, Lombano F, Beeson WL, Dyar DA, Mulla NF, Khan A, Johnston JK, Chinnock RE, Gundry SR, Razzouk AJ (1995) Ventricular remodeling following infant-pediatric cardiac transplantation: does age at transplantation or size disparity matter? Transplantation 60:1467–1472 Sluysmans T, Colan SD (2005) Theoretical and empirical derivation of cardiovascular allometric relationships in children. J Appl Physiol 99:445–457 Zales VR, Wright KL, Muster AJ, Backer CL, Benson DW Jr, Mavroudis C (1992) Ventricular volume growth after cardiac transplantation in infants and children. Circulation 86:II272– II275 Zales VR, Wright KL, Pahl E, Backer CL, Mavroudis C, Muster AJ, Benson DW Jr (1994) Normal left ventricular muscle mass and mass/volume ratio after pediatric cardiac transplantation. Circulation 90:II61–II65