Mid-term result of atrioventricular valve replacement in patients with a single ventricle

Transcription

1 Interactive CardioVascular and Thoracic Surgery (2018) 1 6 doi: /icvts/ivy155 ORIGINAL ARTICLE Cite this article as: Sughimoto K, Hirata Y, Hirahara N, Miyata H, Suzuki T, Murakami A et al. Mid-term result of atrioventricular valve replacement in patients with a single ventricle. Interact CardioVasc Thorac Surg 2018; doi: /icvts/ivy155. Mid-term result of atrioventricular valve replacement in patients with a single ventricle CONGENITAL a b c d e f Koichi Sughimoto a,b, *, Yasutaka Hirata c, Norimichi Hirahara d, Hiroaki Miyata d, Takaaki Suzuki e, Arata Murakami c, Kagami Miyaji a and Shinichi Takamoto f Department of Cardiac Surgery, Kitasato University School of Medicine, Sagamihara, Japan Division of Cardiothoracic Surgery, British Columbia Children s Hospital, Vancouver, British Columbia, Canada Japan Cardiovascular Surgery Database, JCVSD-Congenital section, Tokyo, Japan Department of Health Policy and Management, School of Medicine, Keio University, Tokyo, Japan Department of Pediatric Cardiac Surgery, Saitama Medical University International Medical Center, Hidaka, Japan Department of Cardiovascular Surgery, Mitsui Memorial Hospital, Tokyo, Japan * Corresponding author. Department of Cardiovascular Surgery, Kitasato University School of Medicine, , Kitasato, Minami-ward, Sagamihara-City, Japan. Tel: ; fax: ; ksughimoto@gmail.com (K. Sughimoto). Received 10 September 2017; received in revised form 4 April 2018; accepted 7 April 2018 Abstract OBJECTIVES: Atrioventricular valve replacement is the last option to treat the atrioventricular valve regurgitation in single ventricle. This study investigates the mid-term outcomes of the atrioventricular valve replacement based on the Japan Cardiovascular Surgery Database registry. METHODS: From 2008 to 2014, 56 patients [34 males (61%) and 22 females (39%)] with a single ventricular circulation, underwent atrioventricular valve replacement. Questionnaires were collected to review operative data, mid-term mortality, morbidity and redo replacement. Risk factor analysis was performed by the Cox regression model for death and redo replacement. RESULTS: Heterotaxy, a right systemic ventricle and a common atrioventricular valve was present in 46% (26/56), 64% and 57% of patients, respectively. The most common timings for atrioventricular valve replacement were the interstage between the second and third palliations (34%) and after the Fontan operation (34%). Twenty died during the 3.7 ± 2.6-year follow-up. Eleven received redo atrioventricular replacement. The cumulative incidences of redo atrioventricular valve replacement and survival at 3 years were 20% [95% confidence interval (CI) 9 30] and 66% (95% CI 55 80), respectively. Univariable Cox regression analysis revealed that a tricuspid valve was a risk factor Presented at the 31st Annual Meeting of the European Association for Cardio-Thoracic Surgery, Vienna, Austria, 7 11 October VC The Author(s) Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

2 2 K. Sughimoto et al. / Interactive CardioVascular and Thoracic Surgery for redo valve replacement [hazard ratio (HR) 6.76, 95% CI ; P = 0.005] and that young age was a risk factor for death (HR 0.77, 95% CI ; P = 0.019). Fourteen patients required a pacemaker implantation. CONCLUSIONS: Valve replacement for uncontrollable atrioventricular valve regurgitation in single ventricular circulation was associated with a moderately high risk of death, redo replacement and pacemaker implantation, whereas valve replacement at a later period and with a larger prosthetic valve size was associated with low mortality. Keywords: Congenital heart disease Valve replacement Single ventricle Fontan INTRODUCTION Atrioventricular valve regurgitation is one of the risk factors for worse outcomes in a single ventricular circulation [1, 2]. Atrioventricular replacement is the last option to treat the atrioventricular regurgitation despite complications related to prosthesis implantation [3, 4]. However, a single ventricle was reported to be one of the predictors of death after atrioventricular valve replacement in children [5]. A paucity of the data is available regarding atrioventricular valve replacement in the single ventricle due to limited experience in a single institution [6]. This study evaluates the mid-term outcomes of atrioventricular valve replacement based on the Japan Cardiovascular Surgery Database registry (JCVSD). MATERIALS AND METHODS Patient demographics Patients with a single ventricular circulation who received prosthetic valve replacement in the systemic atrioventricular position were included. Exclusion criteria were biventricular or one and a half circulation. From 2008 to 2014, 56 patients (34 males and 22 females), with a median age at replacement operation of 2.1 years [interquartile range (IQR) 10 months 10.5 years] and a median weight of 8.7 kg (IQR ), were included in this study. In contrast, 589 atrioventricular valve repairs in patients with a single ventricular physiology during the same period were registered but were not included in the study. The preoperative patient characteristics are summarized in Table 1. Diagnosis Patients were classified into 1 of the 12 primary diagnoses of single ventricular circulation. Heterotaxy in single ventricle was the most common, followed by unbalanced atrioventricular septal defect and hypoplastic left heart syndrome. Data collection Data on the patients preoperative clinical features, operations, postoperative course, late events and survival were obtained through a detailed review of medical records from each institution. These early postoperative data were transmitted to the JCVSD. The JCVSD currently collects clinical information from 82 Japanese institutions specializing in congenital heart disease, covering almost all major congenital heart surgery programmes in Japan. Each participating hospital has received the appropriate approval from the respective institutional review board. Patients who matched the above criteria were identified from the JCVSD database, and questionnaires were sent to the participating institutions and the answers were collected. All the analyses was carried out in a blinded fashion, through the JCVSD. Table 1: Preoperative patient characteristics Number (total = 56) Primary diagnosis HLHS 7 13 Hypoplastic LV 1 2 Hypoplastic RV 1 2 SV TAPVD 1 2 DILV 3 5 Heterotaxy MA 4 7 DIRV 2 4 TA 1 2 Unbalanced AVSD 8 14 Other 5 9 Systemic ventricle RV Indeterminate (including AVSD) LV 6 11 Valve type Common AV valve Tricuspid valve Mitral valve 5 9 Heterotaxy Asplenia Polysplenia 3 5 Unknown Preoperative echo measurement AV valve regurgitation grade None 0 0 Mild 0 0 Moderate Severe Unknown 1 2 AV valve stenosis velocity (m/s) None < Unknown 3 5 EF (%) < > Unknown 3 5 Percentage AV: atrioventricular; AVSD: atrioventricular septal defect; DILV: double inlet left ventricle; DIRV: double inlet right ventricle; EF: ejection fraction; HLHS: hypoplastic left heart syndrome; LV: left ventricle; MA: mitral atresia; RV: right ventricle; SV: single ventricle; TA: tricuspid atresia; TAPVD: total anomalous pulmonary venous drainage.

3 K. Sughimoto et al. / Interactive CardioVascular and Thoracic Surgery 3 Table 2: Timing and urgency of atrioventricular valve replacement Variables Statistical analyses Statistical analyses were performed with XLSTAT software (version , Addinsoft SARL, Paris, France) and in R (version 3.3.2; Continuous variables are presented as a median (IQR) for skewed distributions or mean (± standard deviation) for normally distributed variables. Survival was assessed by the Kaplan Meier method and the cumulative incidence was estimated for redo replacement with death as a competing risk. Each time-to-event was measured from the date of the atrioventricular valve (AVV) replacement procedure. The median follow-up time was estimated with the Kaplan Meier method using the censoring distribution. Cox regression models for survival and redo replacement (cause-specific Cox regression) were used for univariable risk factor analysis. Due to lack of significance or a small number of events, multivariable analysis was not possible. The proportional hazards assumption was assessed based on the method of Harrell Lee and via diagnostic plots. For patients without reoperation or death, freedom from reoperation or death was censored at their last known alive date, with the last date of follow-up recorded in April Patient follow-up was 100% for the corresponding patients. RESULTS Number (total = 56) Timing 1st palliation 2 4 Interstage 1st 2nd nd palliation (BCPS) 5 9 Interstage 2nd 3rd rd palliation (Fontan) 3 5 Post 3rd palliation (Fontan) Urgency of the surgery Scheduled Urgent 5 9 Emergent 3 5 BCPS: bidirectional cavopulmonary shunt. Percentage Ninety-six percent (25/26) of institutions responded to the questionnaires. The median follow-up period was 5.0 years (IQR years, measured from those alive at last follow-up). Timing and urgency of the atrioventricular valve replacement The timing of the atrioventricular valve replacement varied from the first palliation to after the third palliation (Fontan) (Table 2). The most common timing was the interstage between the second and third palliations, and after the Fontan operation, both in 19 patients (34%), followed by the interstage between the first and second palliations in 8 patients (14%). Five (9%) and 3 (5%) patients underwent atrioventricular valve replacement along with a second (bidirectional cavopulmonary shunt) or third (Fontan) operation. Concerning urgency of surgery, the majority of Figure 1: Valve size and mortality. patients (48 patients, 86%), underwent the scheduled surgery, whereas 3 (5%) and 5 patients (9%) were operated on an emergency or urgent (within 48 h) basis. Type and size of the prosthesis Among 56 patients, 4 patients (7%) received a bioprosthetic valve whereas the majority of the patients (52 patients, 93%) received a mechanical valve. The choice of the prosthesis was based on institutional or surgeon policy. A variety of prosthesis sizes was selected for the atrioventricular valve replacement, ranging from 16 to 33 mm (Fig. 1). A 23-mm prosthesis was most commonly used in 12 patients. All patients who received a prosthesis of more than 27 mm in size, survived. Surgery and postoperative anticoagulation management Cardiopulmonary bypass and cross-clamp times were 187 ± 74 and 83 ± 41 min, respectively. Circulatory arrest and ventricular fibrillation were used during the bypass in 2 and 2 patients, respectively. Redo atrioventricular valve replacement During the follow-up period, 11 patients required a second atrioventricular valve replacement at a median time of 90 days (IQR 13 days 9.4 months) after the first. Cumulative incidence of redo atrioventricular valve replacement at 3 years was 20% [95% confidence interval (CI) 9 30%] (Fig. 2). Univariable analysis revealed that the tricuspid valve [hazard ratio (HR) 6.76, 95% CI ; P = 0.005] and the interstage between the second and third palliations (HR 4.28, 95% CI ; P = 0.021) were risk factors for redo atrioventricular valve replacement (Table 3). The reasons for the redo replacement were clot formation in the prosthesis in 6 (55%), malfunction of the valve in 3 (27%) and haemolysis in 1 (9%). Mortality During the follow-up period, 20 patients died, whereas of the 11 patients with redo surgery, 5 subsequently died. The smallest child, a 2-month-old girl weighing 2.9 kg, received a 17 mm mechanical prosthesis but died on the day of the CONGENITAL

4 4 K. Sughimoto et al. / Interactive CardioVascular and Thoracic Surgery Figure 2: Cumulative incidence of redo replacement. AV: atrioventricular; CI: confidence interval. Figure 3: Kaplan Meier freedom from death. AV: atrioventricular; CI: confidence interval. Table 3: Univariable Cox regression models: cause-specific Cox regression model for freedom from redo replacement Table 4: Univariable Cox regression models: Cox regression model for freedom from death Variables a HR (95% CI) P-value Age (per year increase) 0.90 ( ) 0.10 T valve 6.76 ( ) Size (per mm increase) 0.86 ( ) Interstage ( ) Post 3rd stage 0.16 ( ) a Proportional hazards assumption was met for each variable. CI: confidence interval; HR: hazard ratio; T valve: tricuspid valve. Variables a HR (95% CI) P-value Age (per year increase) 0.77 ( ) T valve 1.91 ( ) 0.15 Size (per mm increase) 0.85 ( ) Interstage ( ) 0.82 Post 3rd stage 0.57 ( ) 0.28 a Proportional hazards assumption was met for each variable. CI: confidence interval; HR: hazard ratio; T valve: tricuspid valve. operation. Freedom from death at 3 years was 66% (95% CI 55 80%) (Fig. 3). Univariable analysis revealed that a younger age (HR 0.77, 95% CI ; P = 0.019) and a smaller prosthesis size (HR 0.85, 95% CI ; P = 0.01) were risk factors for death (Table 4). Postoperative complications Permanent pacemaker implantation was required in 14 patients (25%). Bleeding was noted in 8 (14%). Extracorporeal membrane oxygenation was necessary in 7 patients (13%) after the operation. Other complications included sepsis in 4 (7%), chylothorax in 2 (5%), renal failure in 2 (5%) and haemolysis in 1 patient (2%). No patient underwent an orthotopic heart transplant. DISCUSSION Valvular function continuously worsens in patients with a common atrioventricular valve during the single ventricle palliation period unless repair is performed at an appropriate time [7]. Although various techniques have been attempted to prevent atrioventricular valve regurgitation [1, 2, 8], valve replacement is the last available option to treat uncontrollable atrioventricular valve regurgitation [9]. Valve replacement is, however, accompanied with comorbidities, such as clot formation leading to valve malfunction, thromboembolism, and risk of bleeding [3, 4]. In the present study, 9 out of 56 patients (16%) experienced a clot in the prosthesis or valve malfunction, which is probably due to difficulties in controlling the parameters, prothrombin time and international normalized ratio control, for small children. Generally, the long-term survival after the mitral valve replacement is worse than after repair in children [10]. Furthermore, valve replacement at the atrioventricular valve position may result in deteriorating ventricular function due to the fact that excision of the subvalvular apparatus and papillary muscles is sometimes necessary [11]. Patients in Fontan circulation whose ventricular function is deteriorating and whose end-diastolic volume is elevated, are at a higher risk of death [12]. In our study, half of the patients exhibited impaired ventricular function with an ejection fraction of less than 50%. Given that the atrioventricular valve regurgitation often causes ejection fraction to be overestimated, these patients were considered to be at a higher risk. An

5 K. Sughimoto et al. / Interactive CardioVascular and Thoracic Surgery 5 earlier timing for intervention when the ventricular function is still preserved may bring about better outcomes. In children, replacement with a smaller prosthesis would require further replacements as the child grows. Brown et al. [5] described the result of mitral valve replacement in children, suggesting that a univentricular heart was one of the predictors of death. For these reasons, atrioventricular valve replacement in patients with a single ventricular circulation is likely to be avoided. The long-term results of the atrioventricular valve replacement have rarely been reported due to the limited number of cases in a single institution [9]. The JCVSD registry consists of 82 cardiac centres that provide the perioperative clinical data [13, 14], and the database can be used to analyse outcomes in a particular patient group of interest. The result obtained from the present study contains important information about patients with a single ventricular circulation, undergoing atrioventricular valve replacement. First, we found that nearly 50% of the patients had heterotaxy. Among these patients, asplenia was most common (11 patients, 20%) followed by polysplenia (3 patients, 5%). Jacobs [15] described complications associated with heterotaxy in Fontan patients, such as problems with variable anatomy of the sinus node and conduction system, potential obstruction of pulmonary venous return, and tendency for development of atrioventricular valve regurgitation. Bhaskar et al. [16], from the Melbourne group, described the survival of patients with atrial isomerism undergoing cardiac surgery. They reported that among 182 patients with isomerism, 49 (47%) with left atrial isomerism (polysplenia) and 60 (77%) right atrial isomerism (asplenia) were directed to single ventricle palliation. One of the independent predictors of mortality was asplenia, and survival was better for patients with the left rather than the right atrial isomerism. Among 26 heterotaxy patients with single ventricular circulation in our study, 10 died. In the present study, a right systemic ventricle was most common (36 patients, 64%), followed by atrioventricular septal defect type ventricle (14 patients, 25%), and a left systemic ventricle was the least (6 patients, 11%). This result was reasonable, as a right systemic ventricle is more likely to cause valve insufficiency by the pressure or volume overload [2]. Regarding the replaced anatomical valve, a common atrioventricular valve was most common (32 patients, 57%). This result was also consistent with a previous report by King et al. [7], wherein the unbalanced atrioventricular valve is found to experience a continual decline in valve function. In terms of the timing of the atrioventricular valve replacement, the interstage between the second and third palliations, and after Fontan operation were the most common (both 34%), and showed the highest risk for redo replacement. This result was assumingly because the survivors of the atrioventricular valve replacement during the inter-stage between second and third palliations will need a size-up at later stages while 1 patient out of 19 received the redo valve replacement after the Fontan procedure. As for the urgency of the surgery, although the majority of procedures (86%) were scheduled, 5 (9%) and 3 (5%) patients had to undergo emergency or urgency surgery, respectively. Of these 8 patients, 4 patients died after the valve replacement (201 days in median, IQR 52 days 1.90 years). Although a mechanical prosthesis was chosen in most patients, 4 received bioprosthetic valves ranging from 16 to 29 mm in size. Of those, 1 patient, who was diagnosed with asplenia with unbalanced atrioventricular septal defect, died 565 days after the valve replacement in the interstage between the second and third palliations. Younger age and smaller size were found to be significant risk factors for death. The most commonvalvesizewas23mm,andtherewasnomortalityin patients who received a valve with the size of 27 mm and larger. Of the 7 patients who received extracorporeal membrane oxygenation, 5 had right ventricle (RV) systemic chamber, and 4 died 1 71 days after the operation. Of these 4 who died, 3 had small prosthetic valves ranging from 16 to 18 mm in size. Caldarone et al. [17] previously described the longterm results after mitral valve replacement in children aged less than 5 years, stressing that an increased ratio of prosthetic valve size to patient weight was one of the multivariable predictors of death. Therefore, an appropriately-sized prosthesis should be selected at the time of the atrioventricular valve replacement. With regard to the redo valve replacement, 11 patients required a second valve replacement. A tricuspid valve was associated with the risk of a redo valve replacement. Patients with a smaller valve may eventually need a larger prosthesis. Kanter et al. [18] reported the result of redo mitral valve replacement in children with low morbidity. However, in their series, in addition to the results of the systemic valve of the biventricular heart, a surprisingly high percentage of patients underwent heart transplants. In Japan, however, the chance of a heart transplant for children is extremely limited [19] and patients with a single ventricular circulation, especially younger children with a ventricular assist device, who aim for a heart transplant, face a very high hurdle. Pacemaker implantation was the most common comorbidity (25%); for those who needed pacemaker implantation, the size of the valve ranged from 16 to 25 mm. Of those, 6 patients (42%), with a prosthesis size between 17 and 22 mm, died. Patients who received a smaller prosthesis were more likely to need a pacemaker and be associated with high mortality. Considering these results, older children without heterotaxy syndrome would make good candidates for an atrioventricular valve replacement with a larger-sized valve, while ventricular function is still preserved. Limitations This study was based on the answers to the questionnaires, from 96% of all participating institutions. Also, the study is limited by its retrospective nature, and its short follow-up period. As this is a multi-institutional study, there is bias in terms of procedure, decision-making and institution or surgeon preferences. We could not evaluate differences in product type of the prosthesis or postoperative cardiac function. CONCLUSION Valve replacement for uncontrollable atrioventricular valve regurgitation in patients with a single ventricular circulation is one of the last available options, entailing a moderately high risk of death, redo replacement, and pacemaker implantation. Valve replacement in a later period with a larger prosthetic valve size had low mortality. Therefore, older children without heterotaxy syndrome would be good candidates for the atrioventricular valve CONGENITAL

6 6 K. Sughimoto et al. / Interactive CardioVascular and Thoracic Surgery replacement with a larger-sized valve, while the ventricular function is still preserved. ACKNOWLEDGEMENTS We thank all participating institutions for supplying clinical data for the study and the JCVSD for providing technical support. Conflict of interest: none declared. REFERENCES [1] Sano S, Fujii Y, Arai S, Kasahara S, Tateishi A. Atrioventricular valve repair for patient with heterotaxy syndrome and a functional single ventricle. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2012;15: [2] Tsang VT, Raja SG. Tricuspid valve repair in single ventricle: timing and techniques. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2012;15:61 8. [3] Kojori F, Chen R, Caldarone CA, Merklinger SL, Azakie A, Williams WG et al. Outcomes of mitral valve replacement in children: a competingrisks analysis. J Thorac Cardiovasc Surg 2004;128: [4] Bradley SM, Sade RM, Crawford FA, Stroud MR. Anticoagulation in children with mechanical valve prostheses. Ann Thorac Surg 1997;64:30 4. [5] Brown JW, Fiore AC, Ruzmetov M, Eltayeb O, Rodefeld MD, Turrentine MW. Evolution of mitral valve replacement in children: a 40-year experience. Ann Thorac Surg 2012;93: [6] Mahle WT, Cohen MS, Spray TL, Rychik J. Atrioventricular valve regurgitation in patients with single ventricle: impact of the bidirectional cavopulmonary anastomosis. Ann Thorac Surg 2001;72: [7] King G, Gentles TL, Winlaw DS, Cordina R, Bullock A, Grigg LE et al. Common atrioventricular valve failure during single ventricle palliation. Eur J Cardiothorac Surg 2017;51: [8] Sughimoto K, Konstantinov IE, Brizard CP, d Udekem Y. Polytetrafluoroethylene bridge for atrioventricular valve repair in single-ventricle palliation. J Thorac Cardiovasc Surg 2015;149: [9] Mahle WT, Gaynor JW, Spray TL. Atrioventricular valve replacement in patients with a single ventricle. Ann Thorac Surg 2001;72: [10] Moss RR, Humphries KH, Gao M, Thompson CR, Abel JG, Fradet G et al. Outcome of mitral valve repair or replacement: a comparison by propensity score analysis. Circulation 2003;108:II90 7. [11] Sarris GE, Miller DC. Valvular-ventricular interaction: the importance of the mitral chordae tendineae in terms of global left ventricular systolic function. J Card Surg 1988;3: [12] Ohuchi H, Miyazaki A, Negishi J, Hayama Y, Nakai M, Nishimura K et al. Hemodynamic determinants of mortality after Fontan operation. Am Heart J 2017;189:9 18. [13] Miyata H, Murakami A, Tomotaki A, Takaoka T, Konuma T, Matsumura G et al. Predictors of 90-day mortality after congenital heart surgery: the first report of risk models from a Japanese database. J Thorac Cardiovasc Surg 2014;148: [14] Hoashi T, Miyata H, Murakami A, Hirata Y, Hirose K, Matsumura G et al. The current trends of mortality following congenital heart surgery: the Japan Congenital Cardiovascular Surgery Database. Interact CardioVasc Thorac Surg 2015;21: [15] Jacobs ML. Complications associated with heterotaxy syndrome in Fontan patients. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2002;5: [16] Bhaskar J, Galati JC, Brooks P, Oppido G, Konstantinov IE, Brizard CP et al. Survival into adulthood of patients with atrial isomerism undergoing cardiac surgery. J Thorac Cardiovasc Surg 2015;149: [17] Caldarone CA, Raghuveer G, Hills CB, Atkins DL, Burns TL, Behrendt DM et al. Long-term survival after mitral valve replacement in children aged <5 years: a multi-institutional study. Circulation 2001;104:I [18] Kanter KR, Forbess JM, Kirshbom PM. Redo mitral valve replacement in children. Ann Thorac Surg 2005;80: [19] Nishimura N, Kasahara M, Ishikura K, Nakagawa S. Current status of pediatric transplantation in Japan. J Intensive Care 2017;5:48.