ORIGINAL ARTICLE Cardiovascular Surgery

Circ J 2009; 73: 10451048

Outcome of Non-Transplant Surgical Strategy for End-Stage Dilated Cardiomyopathy in Young Children
Hisashi Sugiyama, MD; Minako Hoshiai, MD; Atsushi Naitoh, MD; Toshie Kadono, MD; Syoji Suzuki, MD*; Kanji Sugita, MD
Background: The shortage of heart transplantation donors is a problem, but partial left ventriculectomy (PLV) and mitral valve replacement (MVR) are feasible at the optimal timing, even in young children. Methods and Results: From May 1998 to May 2008, 11 children under the age of 3 years were diagnosed with severe dilated cardiomyopathy (DCM). Indications and outcomes of non-transplant surgical strategies were evaluated and 8 procedures were performed in 6 children: 5 PLV and 3 MVR. Two of them underwent MVR after PLV because of deterioration of mitral regurgitation (MR). Age at surgery ranged from 8 months to 2 years 11 months. Four are alive, of whom 1 eventually underwent a heart transplant overseas. Two children died during the study period: 1 who underwent only MVR died of intracranial bleeding during thrombolytic therapy for a thrombus stack valve and the other child died of congestive heart failure because of progressive MR 2 months after PLV. Follow-up after PLV ranged from 2 months to 8 years, and after MVR ranged from 1 month to 4 years. Conclusions: PLV and MVR are feasible and effective and should be considered when heart failure resists conventional therapy. (Circ J 2009; 73: 10451048) Key Words: Alfieli mitral valve repair; Dilated cardiomyopathy; Mitral valve replacement; Partial left ventriculectomy

eta-blocker therapy with angiotensin-converting enzyme inhibitor or/and angiotensin receptor blocker is widely used as conventional therapy for 1 dilated cardiomyopathy (DCM), even in young children. However, it is not always effective for children with severe DCM. Although heart transplantation is a highly effective therapy for patients with end-stage heart failure (HF), the limited number of donors makes it available to only a small percentage of such patients who could benefit from the 2 In addition, mortality while waiting for transoperation. plantation cannot be ignored3 and donors under 3 years of age are in short supply. In Japan, children under 15 years of age are prohibited by law from becoming organ donors. As a result, the youngest recipient in Japan so far was 8 years old. Therefore, in Japan heart transplantation in young children is practically unavailable. Meanwhile, partial left ventriculectomy (PLV) or mitral valve replacement (MVR) are feasible, in even pediatric patients, despite some diffi47 Such procedures can be scheduled for an optimal culties. time, according to the patients condition, and do not carry the risk of rejection or the complications of long-term immunosuppressant therapy. For young children with a poor response to conventional therapy, a non-transplant surgical strategy can be considered.

Editorial p1010 Methods

From March 1998 to October 2007, 11 consecutive children under the age of 3 years were diagnosed with severe DCM: 5 boys and 6 girls, ranging in age at onset from birth to 2 years 11 months (Table1). Patients with systemic organic diseases, such as muscular dystrophy, mitochondria disease or inherited metabolic disorders were excluded, as were children with coronary artery anomaly. A minor congenital anomaly that did not affect hemodynamics was detected in 2 children: a small ventricular septal defect with minimal shunt, and a double-orifice mitral valve with mild mitral regurgitation (MR) and without valvular stenosis, respective. None of the others had any primary mitral valve abnormality. All had symptoms related to congestive HF, such as loss of appetite, tachypnea, pale skin and edema. 8 Cardiac The clinical status of HF was class IV for each child. catheterization and right ventricular myocardial biopsy were performed in all children and the diagnosis of DCM was based on clinical features and biopsy. Significant inflammatory cell proliferation or findings of metabolic disorder was not observed in any of the myocardial specimens. Medical Treatment Our therapeutic strategy for DCM in young children is as follows. In the acute or recurrent phase, large doses of diuretics, vasodilators and phosphodiesterase inhibitors are intravenously administered in the intensive care unit, in conjunction with restricted water and salt intake. If necessary, an appropriate dose of catecholamine is added. In the chronic phase, diuretics, angiotensin-converting enzyme inhibitor or/and angiotensin receptor blocker are adminis-

(Received October 2, 2008; revised manuscript received December 25, 2008; accepted January 5, 2009; released online April 9, 2009) Department of Pediatrics, *Second Division of Surgery, Faculty of Medicine, University of Yamanashi, Yamanashi, Japan Mailing address: Hisashi Sugiyama, MD, Pediatric Cardiology, National Cardiovascular Center, 5-7-1 Fujishirodai, Suita 565-8565, Japan. E-mail: shisashi@hsp.ncvc.go.jp All rights are reserved to the Japanese Circulation Society. For permissions, please e-mail: cj@j-circ.or.jp

Circulation Journal Vol.73, June 2009

1046
Table1. Characteristics of the Children Undergoing Surgical Interventions Patient Sex no. 1 2 3 4 5 6 Age at onset Age at PLV Age at MVR Duration (onset-last FU) Status of PLV Duration procedure (surgery to last FU)

SUGIYAMA H et al.

Outcome Alive Alive HTX Alive Death Death

M 2yr11mo 3yr6mo M 11mo 2yr1mo 3yr2mo F 5mo 8mo F Birth 2yr4mo 5yr2mo F 3mo 10mo F Birth 1yr

9yr4mo Elective 8yr10mo 4yr7mo Semi-urgent 3yr6mo 8mo (surgery to HTX) Urgent 6mo (surgery to HTX) 9yr10mo Elective 7yr5mo 9mo Elective 2mo 1yr (surgery to death) 1mo (surgery to death)

PLV, partial left ventriculectomy; MVR, mitral valve replacement; FU, follow-up; yr, year(s); mo, month(s); HTX, heart transplantation.
7,1013 valve repair. Indications for MVR were: (1) more than moderate MR with a severely dilated left atrium on echocardiography; (2) >10% FS or >25% EF on echocardiography; (3) >1,000pg/ml BNP, and (4) previous mitral repair (Alfieri procedure) or coexisting mitral stenosis. All the criteria had to be fulfilled. MVR was performed by the usual methods, with minimum resection of the mitral valve apparatus. Written informed consent was obtained from the patients parents or guardians before surgery.

Results
Of 11 children, 6 were judged as end-stage DCM, meaning HF that was resistant to aggressive medical treatment and rapid progression, and eventually they underwent a total of 8 surgical interventions: 5 PLV and 3 MVR. One child, who underwent urgent PLV, eventually required heart transplantation 5 months later. The follow-up after PLV ranged from 2 months to 8 years 10 months, and after MVR ranged from 1 month to 4 years 8 months. The age at PLV ranged from 8 months to 2 years 11 months and at MVR ranged from 1 year to 5 years 2 months. No patient simultaneously underwent PLV and MVR. No acute complications related to PLV occurred in any of the 5 children. Clinical data at the time of the PLV are shown in Table2. The CTR ranged from 70% to 78%. The percentage predictive value of the left ventricular dimension ranged from 165% to 198%. FS on echocardiography was less than 10% in all children. Serum BNP levels ranged from 1,110 to 4,360pg/ml. On echocardiography, 2 had severe MR and 3 had moderate MR. MR was considered to be caused by chordal tethering in all the children undergoing PLV. In patient 1, cardiac function gradually recovered after PLV and no further surgical intervention has been necessary. Clinical status of HF at the last follow-up was class I. In patient 2, cardiac function gradually improved with time; however, MR deteriorated 6 months after PLV with clinical complaints, and MVR (ATS-AP 20mm, ATS Medical Inc, Minneapolis, MN, USA) was performed 13 months after PLV. Clinical status of HF at the last follow-up was class III. In patient 3, PLV was performed at 7 months of age, but her clinical condition deteriorated shortly after recovery of cardiac function and she eventually underwent heart transplantation overseas 6 months after PLV. No technical difficulty related to the previous PLV procedure was reported at the time of heart transplantation. Patient 4 had a stable clinical condition was stable until approximately 5 years after PLV. Mitral and tricuspid regurgitation became severe, with EF >40% of on echocardiography. MVR (ATS 20mm) was carried out without any difficulties at 5 years 3 months of age, which was 2 years and 9 months after PLV. Clinical
Circulation Journal Vol.73, June 2009

Figure. 99mTc-tetrofosmin myocardial scintigram shows a more than moderate perfusion defect at the posterolateral wall in an infant with dilated cardiomyopathy.

tered orally and then the -blocker, carvedilol, is started and increased stepwise from 0.05 to 0.6mgkg1day1. When cardiac function deteriorated despite aggressive medical therapy, surgical interventions were considered. Surgical Interventions PLV was chosen for children with a severely dilated left ventricular cavity, whereas MVR was chosen for children with severe MR who had undergone previous mitral repair or had coexisting mitral stenosis. Indications for PLV9 were: (1) >3 months elapsed since onset; (2) requiring more than a medium dose of intravenous inotropic agent or mechanical ventilation support; (3) cardiothoracic ratio (CTR) on chest X-ray >0.70; (4) left ventricular end-diastolic volume on ventriculography >350% of normal predictive value or >150% of normal predictive value of left ventricular dimension on echocardiography; (5) <10% of fractional shortening (FS) or <25% of ejection fraction (EF) on M-mode echocardiography; (6) >1,000pg/ml of brain natriuretic peptide (BNP); and (7) perfusion defect in the posterolateral segment of the left ventricle on myocardial scintigraphy (Figure). Exact evaluation by myocardial scintigraphy could not be done in all the children because of their small hearts. When a patients data fulfilled 6 of the 7 criteria, PLV was chosen. Our standard PLV procedure is as follows. The left ventricle is incised, starting near the apex. The as much myocardium as possible is resected from the posterolateral wall between the 2 papillary muscles, followed by Alfielis mitral

Non-Transplant Surgery for Pediatric DCM

Table2. Clinical Data Before and at Last FU After PLV Patient no. 1 2 3 4 5 CTR (%) Before Last FU 78 70 70 72 70 48 63 63 65 69 LVDd (mm) Before Last FU 59.3 47.2 51.2 58.3 52 36.1 47.3 40.6 39 35.6 FS (%) Before Last FU 6 43 8 9 5 8 9 26 8 10 BNP (pg/ml) Before Last FU 1,290 4 1,110 393 1,620 1,580 4,360 129 3,050 403 HF status Before Last FU IV IV IV IV IV I III I (HTX) III Death

1047

CTR, cardiothoracic ratio on chest X-ray; LVDd, left ventricular diastolic dimension on M-mode echocardiography; FS, fractional shortening; BNP, brain natriuretic peptide; HF, heart failure. Other abbreviations see in Table1. Table3. Onset Data for Children With and Without Surgical Interventions Patient no. Age at onset CTR (%) 71 64 67 60 70 71 72 68 70 72 65 % of LVDd FS (%) MR BNP (pg/ml)

Surgical cases 1 2yr11mo 2 11mo 3 5mo 4 Birth 5 3mo 6 Birth Medical cases 7 2yr3mo 8 2yr8mo 9 5mo 10 1yr4mo 11 6mo

178 4 144 8 181 8 186 11 200 9 121 27 146 17 115 7 186 16 186 9 177 10

Mild 2,790 Severe 1,950 Moderate 538 Mild 27 Mild 1,907 Severe 436 Moderate 2,400 Moderate 704 Severe 1,857 Mild 527 Mild 380

MR, mitral regurgitation. Other abbreviations see in Tables1,2.

status of HF at the last follow-up was class II. In patient 5, cardiac function slightly recovered from 8% to 10% of FS at 1 month after PLV; however, severe MR developed and she died of progressive HF 2 months after PLV while under consideration for MVR. In patient 6, severe MR and moderate mitral stenosis with moderately impaired left ventricular function, which was approximately 40% of EF on echocardiography, was noted at birth. She did not gain body weight because of the HF, so MVR (ATS 20mm) was chosen as the surgical intervention rather than PLV. Shortly after surgery, a thrombotic stack valve was noticed on echocardiography and tissue plasminogen activator inhibitor was intravenously administered as thrombolytic therapy. Unfortunately, she died of intracranial hemorrhage as a complication of the thrombolytic therapy 1 month after surgery. So far, more than 5 years have passed after PLV for patients 1, 2 and 4. Two of them attend primary school without any complaint. Five children who responded to conventional medical therapy and were exempted from surgical interventions are alive in good clinical condition (class I or II). There were no remarkable differences in the onset data for children with and without surgical interventions (Table3).

Discussion
Childhood DCM can improve spontaneously with or without medical therapy in approximately 30% of cases 2,1416 However, the prognosis for young over several years. children after a diagnosis of DCM is generally poorer and 2,16 so it is difficult to deterearlier than for adult patients, mine the indications and timing of surgical interventions. Recently, surgical procedures other than heart transplantation have been developed and PLV is a surgical option currently used for the treatment of end-stage HF in adult
Circulation Journal Vol.73, June 2009

1719 We have reported successful cases of PLV in patients. young children and shown that the procedure is feasible 7,9,10 However, body growth even in very young children. needs to be considered, as well as the short-term outcome. The increase in body size associated with normal growth in children places greater demands on the heart, so we use Alfielis methods, which allow for valvular growth with age, 1113 However, instead of using valvular ring annuloplasty. the longer-term outcomes after the procedure are still largely 47,10 unknown. MVR is another option for cases of DCM with severe 2022 To a greater or lesser degree, MR coexists with MR, severe DCM and the mechanism of MR is considered to involve tethering of the mitral leaflet by the papillary muscles, with dilatation of the mitral annulus. Mitral insufficiency leads to a vicious cycle of increasing volume overload of the dilated left ventricle, thus leading to progression of annular dilatation, worsening of MR and volume overload. This cycle can be interrupted by correcting the mitral valve insufficiency. Mitral valve reconstruction in children with congestive HF may lead to clinical and functional improvement, but it is believed to be associated with a substantial perioperative risk. In general, MVR is thought to be inferior to mitral valve repair, in terms of cardiac function, because of the interruption of the annular-chordal-papillary muscle continuity. However, mitral valve repair cannot completely prevent MR, which may recur with time and body growth. In the present study, Alfieli mitral valve repair was done at the time of PLV. Although all 5 patients undergoing PLV had significant MR before surgery, they had only mild MR immediately after surgery, so Alfieli mitral valve repair was considered to be effective in the short term. However, in 2 children MR deteriorated with time because of valvular annulus re-dilatation and in them MVR eventually improved cardiac function. Sartipy et al report that edge-to-edge mitral

1048

SUGIYAMA H et al.
with utilization of heart transplantation. J Am Coll Cardiol 2004; 44: 391397. Nield LE, McCrindle BW, Bohn DJ, West LJ, Coles JG, Freedom RM. et al. Outcomes for children with cardiomyopathy awaiting transplantation. Cardiol Young 2000; 10: 358366. Berger S, Tweddell JS, Frommelt PC, Weinhause L. Partial left ventriculectomy for dilated cardiomyopathy in a newborn. J Thorac Cardiovasc Surg 1999; 117: 10171019. Gradinac S, Jovanovic I, Dukic M, Hercog D, Jovovic L, Kokai G, et al. Partial left ventriculectomy in a two-year-old girl with dilated cardiomyopathy. J Heart Lung Transplant 1999; 18: 381383. Vricella LA, Gundry SR, Larsen RL, Bailey LL. Successful myocardial volume reduction in a 9-month-old infant. Ann Thorac Surg 2000; 69: 12531255. Yoshii S, Hosaka S, Takahashi W, Amano H, Abraham SJ, Kaga S, et al. Partial left ventriculectomy in an infant with dilated cardiomyopathy. J Thorac Cardiovasc Surg 1999; 117: 616618. Ross RD. Grading the graders of congestive heart failure in children. J Pediatr 2001; 138: 618. Yoshii S, Hosaka S, Suzuki S, Osawa H, Akashi O, Tada Y, et al. Indications for partial left ventriculectomy in pediatric dilated cardiomyopathy. Circ J 2002; 66: 337340. Sugiyama H, Komai T, Hoshiai M, Tan T, Kadono T, Suzuki S, et al. Mid-term outcome after partial left ventriculectomy in pediatric patients. Pediatr Cardiol 2003; 24: 5459. Fucci C, Sandrelli L, Pardini A, Torracca L, Ferrari M, Alfieri O. Improved results with mitral valve repair using new surgical techniques. Eur J Cardiothorac Surg 1995; 9: 621626. Bhudia SK, McCarthy PM, Smedira NG, Lam BK, Rajeswaran J, Blackstone EH. Edge-to-edge (Alfieri) mitral repair: Results in diverse clinical settings. Ann Thorac Surg 2004; 77: 15981606. Kherani AR, Cheema FH, Casher J, Fal JM, Mutrie CJ, Chen JM, et al. Edge-to-edge mitral valve repair: The Columbia Presbyterian experience. Ann Thorac Surg 2004; 78: 7376. Kinugawa Y, Nakazawa M, Monma K, Miura S. The natural history of idiopathic dilated cardiomyopathy in the young. Acta Cardiol Paediatr Jpn 1992; 7: 504509 (in Japanese). Taliercio CP, Seward JB, Driscoll DJ, Fisher LD, Gersh BJ, Tajik AJ. Idiopathic dilated cardiomyopathy in the young: Clinical profile and natural history. J Am Coll Cardiol 1985; 6: 11261131. Venugopalan P, Houston AB, Agarwal AK. The outcome of idiopathic dilated cardiomyopathy and myocarditis in children from the west of Scotland. Int J Cardiol 2001; 78: 135141. Batista RJ, Verde J, Nery P, Bocchino L, Takeshita N, Bhayana JN, et al. Partial left ventriculectomy to treat end-stage heart disease. Ann Thorac Surg 1997; 64: 634638. McCarthy JF, McCarthy PM, Starling RC, Smedira NG, Scalia GM, Wong J, et al. Partial left ventriculectomy and mitral valve repair for end-stage congestive heart failure. Eur J Cardiothorac Surg 1998; 13: 337343. Irie H, Isomura T, Nomura F, Horii T, Hoshino J, Makinae H, et al. Novel left ventriculoplasty for nonischemic dilated cardiomyopathy with functional mitral regurgitation. J Thorac Cardiovasc Surg 2006; 131: 233235. Calafiore AM, Gallina S, Di Mauro M, Gaeta F, Iaco AL, DAlessandro S, et al. Mitral valve procedure in dilated cardiomyopathy: Repair or replacement? Ann Thorac Surg 2001; 71: 11461152. Geha AS, El-Zein C, Massad MG. Mitral valve surgery in patients with ischemic and nonischemic dilated cardiomyopathy. Cardiology 2004; 101: 1520. Gummert JF, Rahmel A, Bucerius J, Onnasch J, Doll N, Walther T, et al. Mitral valve repair in patients with end stage cardiomyopathy: Who benefits? Eur J Cardiothorac Surg 2003; 23: 10171022. Sartipy U, Albage A, Mattsson E, Lindblom D. Edge-to-edge mitral repair without annuloplasty in combination with surgical ventricular restoration. Ann Thorac Surg 2007; 83: 13031309. Miura T, Eishi K, Yamachika S, Hashizume K, Tada S, Yamane K, et al. Isolated undersized mitral annuloplasty for functional mitral regurgitation in non-ischemic dilated cardiomyopathy. Circ J 2008; 72: 17441750. Lee S, Chang BC, Youn YN, Kwak YL, Yoo KJ. Changes in left ventricular function and dimension after surgical ventricular restoration with or without concomitant mitral valve procedure. Circ J 2007; 71: 15161520. Claus M, Beling M, Grohmann A, Borges AC, Baumann G. Longterm results after partial left ventriculectomy. Int J Cardiol 2003; 89: 223230. Stolf NA, Moreira LF, Bocchi EA, Higuchi ML, Bacal F, Bellotti G, et al. Determination of midterm outcome of partial left ventriculectomy in dilated cardiomyopathy. Ann Thorac Surg 1998; 66: 15851591.

repair without annuloplasty in combination with ventricu23 lotomy carries a high operative risk and poor prognosis. Considering all the results, an alternative mitral valve repair technique, such as the non-circular ring and subvalvular procedure, might be considered instead of the Alfieri method at 19,24,25 the time of PLV. As seen in the present cases of pediatric DCM, deterioration of HF is rapid, so the decision to perform a surgical intervention should be made without too much delay, particularly because emergency surgery results in a poor prognosis. However, on the basis of the onset data for both the non-surgical and surgical cases (Table3), onset data alone cannot predict the necessity for a surgical intervention. Previous studies also report that onset data did not predict 14,15 In the present study, the children who did not prognosis. undergo a surgical intervention were all responders to initial medical therapy. Therefore, we allow at least 3 months from onset to surgical intervention to eliminate responders to conventional HF therapy. High concentrations of plasma BNP (>1,000pg/ml), >150% of the normal predicted value of the left ventricular diameter, >70% CTR and <30% EF, and the necessity for medium doses of catecholamine or mechanical ventilation may all be indicators that PLV is appropriate. In addition, a markedly enlarged left ventricular cavity with a spherical shape is a favorable factor for PLV, because the PLV procedure is theoretically based on Laplaces law. In the present study, 5 of 11 children were managed by aggressive medical therapy without surgical intervention and 8 survived without heart transplantation. The overall outcome of our therapeutic strategy is comparable with that for heart transplantation. A previous study revealed that 26 and in this mortality more than 2 years after PLV is low, study, more than 5 years have passed for 3 of the 5 children who underwent PLV and their cardiac function gives them good quality of life. In contrast, mortality after heart transplantation gradually but steadily increases because of chronic rejection. At the very least, PLV and MVR in childhood have a role as palliative bridge therapy to heart transplanta27 Non-transplant surgical approaches, such as PLV or tion. MVR, are a feasible option, particularly for rescue end-stage cases, and all the more if heart transplants for children are not possible for whatever reason. We consider that a nontransplant surgical strategy might be an appropriate permanent repair in combination with aggressive drug therapy. Further accumulation of experience and longer follow-up are needed to fully assess the effects of this strategy. Study Limitations The present study has several of the limitations of an observational and non-randomized study. As much as possible, we pursued medical therapy before considering a surgical option for severe DCM. Surgical intervention was performed only if cardiac function steadily deteriorated. Consequently, it is impossible to compare the outcome of medical therapy with that of surgical intervention. We would like to emphasize the usefulness of the combination of medical and nontransplant surgical therapy rather than each therapy alone. References
1. Bruns LA, Chrisant MK, Lamour JM, Shaddy RE, Pahl E, Blume ED, et al. Carvedilol as therapy in pediatric heart failure: An initial multicenter experience. J Pediatr 2001; 138: 505511. 2. Tsirka AE, Trinkaus K, Chen SC, Lipshultz SE, Towbin JA, Colan SD, et al. Improved outcomes of pediatric dilated cardiomyopathy

3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

19.

20. 21. 22. 23. 24.

25.

26. 27.

Circulation Journal Vol.73, June 2009