|
|
||||||||
Ann Thorac Surg 2005;80:44-49
© 2005 The Society of Thoracic Surgeons
a Department of Pediatrics, Division of Pediatric Cardiology, Columbus Childrens Hospital, Columbus, Ohio
b Department of Cardiology, Childrens Hospital Boston, Boston, Massachusetts
Accepted for publication January 20, 2005.
* Address reprint requests to Dr Cua, Department of Pediatrics, Division of Pediatric Cardiology, Columbus Childrens Hospital, 700 Childrens Dr, Columbus, OH 43205-2696 (Email: cuac{at}pediatrics.ohio-state.edu).
| Abstract |
|---|
|
|
|---|
METHODS: PubMed was searched using six different terms individually for articles from January 2003 to October 2004. Manuscripts that compared the classic to modified Norwood were reviewed. Mantel-Haenszel analysis was used to evaluate the relationship between treatment method and mortality stratified across hospitals. The Breslow-Day procedure tested homogeneity of odds ratio across hospitals. Separate analyses were performed for inpatient and interstage periods.
RESULTS: A total of 4,545 citations was screened. Five manuscripts met the criteria. Seventy-two patients undergoing classic Norwood and 84 patients undergoing modified Norwood survived to initial hospital discharge. The Breslow-Day statistic supported homogeneity of odds ratios for survival across hospitals (
2 = 2.09, df = 4, p = 0.72). Odds of interstage death was 11.6 times greater (2.2 to 62.1, 95% CI) for the classic Norwood compared with the modified Norwood procedure. This difference was statistically significant by the Mantel-Haenszel
2 (11.0, p = 0.001). The Breslow-Day statistic supported homogeneity of the odds ratios across hospitals (
2 = 3.1, df = 4, p = 0.53).
CONCLUSIONS: The modified Norwood procedure has a significantly lower interstage mortality compared with the classic Norwood procedure. A large randomized study is needed to determine whether these results remain consistent.
| Introduction |
|---|
|
|
|---|
A recent modification of the Norwood procedure consisting of a right ventricle to pulmonary artery conduit (NW-RVPA) to supply pulmonary blood flow instead of the modified Blalock-Taussig shunt has been reported by various institutions to improve early mortality [13, 5, 11, 2426]. The theory behind this modification is that coronary artery blood supply will be preserved owing to higher diastolic blood pressure that occurs in the NW-RVPA compared with the NW-BT procedure and thus will result in improved early mortality. While these early reports are favorable for the NW-RVPA procedure, the midterm and long-term results are still unknown. Theoretically, if a major cause of interstage deaths was due to poor myocardial perfusion from abnormal coronary anatomy, then the higher diastolic blood pressures in patients undergoing the NW-RVPA procedure should improve coronary perfusion and therefore reduce interstage mortality. Unfortunately, the numbers of patients in each of the institutional experiences comparing the NW-BT to NW-RVPA procedures have been too small to show if there were significant differences for interstage mortality between the two groups.
The goal of this study was to determine if there was a significant difference in interstage mortality, defined as time from hospital discharge after stage I procedure to completion of stage II, between patients undergoing the NW-BT versus the NW-RVPA procedure. Meta-analysis was used to achieve the patient size to assess a statistically significant difference.
| Patients and Methods |
|---|
|
|
|---|
Manuscripts were selected if they met all of the following requirements: compared a cohort of patients undergoing the NW-BT procedure to the NW-RVPA procedure for palliation of single ventricle physiology; documented baseline demographics such as number of patients in each group; age and weight at time of surgery; documented hospital mortality; and documented interstage mortality. Studies without the above data were excluded. In addition to the data above, anatomical diagnosis, ascending aorta diameter, sex, shunt/conduit placed, cardiopulmonary bypass time, presence of a restrictive atrial communication, length of intensive care stay, and number of patients completing a stage II procedure were also recorded if available.
A Mantel-Haenszel analysis was used to evaluate the relationship between treatment method and mortality stratified across hospitals. The Mantel-Haenszel procedure provides a common odds ratio estimate and confidence interval (CI) showing the risk of death using the NW-BT procedure relative to the NW-RVPA procedure. Separate analyses were performed for inpatient and interstage periods. The Breslow-Day procedure tested the homogeneity (consistency) of the odds ratio across hospitals.
The primary analysis used intention to treat to classify patients into treatment groups. A secondary analysis was conducted to adjust for 4 patients who were reassigned to the NW-BT group after an NW-RVPA procedure and 7 patients who had been discharged, but had not yet had their bidirectional Glenn procedure.
| Results |
|---|
|
|
|---|
The operative time period for patients described in the papers ranged from October 1999 to November 2003. Each procedure, NW-BT and NW-RVPA, had a total of 93 patients in their respective cohort. Baseline demographics qualitatively showed minimal differences in age at surgery, weight at surgery, or ascending aorta diameter (Table 1). Diagnosis, shunt or conduit size used, presence of a restrictive atrial communication, cardiopulmonary bypass duration, and intensive care lengths of stay were also recorded if available (Table 2). Statistical analysis was not performed on these data to determine homogeneity between the different manuscript demographics because insufficient information was available for comparison.
|
|
|
2 (4.42, p = 0.04). The Breslow-Day statistic supported homogeneity of the odds ratios for improved survival after NW-RVPA compared with NW-BT across the institutions (
2 = 2.09, df = 4, p = 0.72).
Overall, interstage mortality was 19.4% (14 of 72) for the NW-BT procedure and 2.4% (2 of 84) for the NW-RVPA procedure. The risk for interstage death was 11.6 (2.2 to 62.1, 95% CI) times greater for the NW-BT procedure relative to the NW-RVPA. This difference was significant by the Mantel-Haenszel
2 (11.0, p = 0.001) indicating that patients undergoing the NW-BT procedure had increased risk of death after discharge from the hospital while awaiting the bidirectional Glenn procedure. The Breslow-Day statistic supported homogeneity of the odds ratios across hospitals (
2 = 3.1, df = 4, p = 0.53). Qualitatively, the age at which stage II was performed on both cohorts were similar within institutions, but varied among institutions (Table 3).
Sensitivity of the findings was tested by eliminating the 7 NW-RVPA patients who had not yet undergone the second procedure, 4 patients from Azakie and colleagues [2] and 3 patients from Bradley and associates [1]. Furthermore, 4 subjects who had the NW-BT procedure after the NW-RVPA procedure were also crossed over to that group, 2 patients from Bradley and associates [1] and 2 patients from Mahle and colleagues [11]. The results were essentially the same. When adjusting for the 4 patients who crossed over from the NW-RVPA to NW-BT group and excluding the 7 patients in the NW-RVPA group that had not completed the stage II procedure at time of publication, interstage death was 18.4% (14 of 76) for the NW-BT and 2.7% (2 of 73) for the NW-RVPA group. The risk of death was 8.8 (1.64 to 47.3, 95% CI) times greater for the NW-BT procedure relative to the NW-RVPA. This difference was significant by the Mantel-Haenszel
2 (6.5, p = 0.011). The Breslow-Day statistic supported homogeneity of the odds ratios for survival postdischarge after NW-RVPA compared with NW-BT across the institutions (
2 = 3.36, df = 4, p = 0.50). Mahle and coworkers [11] described 27 patients discharged from the hospital after initial surgery with 4 interstage deaths. Twenty-two of the possible 23 patients had subsequently undergone the second stage procedure. This 1 patient was not included in adjusted analysis because of uncertainty which procedure was performed. Inclusion or exclusion of this patient did not change significance.
| Comment |
|---|
|
|
|---|
Numerous reasons have been documented or theorized to be the etiology of interstage mortality; these include decreased coronary perfusion, shunt obstruction, arrhythmias, right ventricular failure, residual arch lesions, acute intercurrent illness, or atrioventricular valve regurgitation. Unexpected death also continues to be a significant documented cause of interstage mortality [12, 14, 22]. Assuming that residual arch lesions, intercurrent illness, and atrioventricular regurgitation would be similar between the two groups despite the different time periods examined, possible reasons for the improved mortality in the NW-RVPA group could be improved coronary perfusion, decreased risk for shunt obstruction, decreased risk for arrhythmias, or nonspecific factors that may be characterized as "experience" with treatment of patients with HLHS.
Hypothetically, the larger conduits used for the NW-RVPA procedure would be less likely to occlude compared with the smaller shunts used for the NW-BT operation. In the papers analyzed, shunt size ranged from 2.5 to 4.0 mm, whereas the conduit sizes were from 4.0 to 6.0 mm. Decreased risk for arrhythmias may also be a possibility for improved mortality in the NW-RVPA group because of improved coronary perfusion; however, this must be weighed against the risk of ventricular arrhythmias due to the right ventriculotomy in the NW-RVPA procedure. Overall improvement in quality of care for patients with HLHS cannot be excluded as a possible reason for the improved mortality, especially since the two cohorts were from different time periods, albeit relatively close to one another. Markedly improved interstage mortality has been reported for the NW-BT group with strict follow-up [12], but there was no evidence that such a protocol was initiated in the five institutions analyzed to account for the differences in mortality.
Another theoretical reason for improved interstage survival for the NW-RVPA procedure could be the higher diastolic blood pressure because of lack of "run off" in the shunt compared with the NW-BT procedure, and therefore improved coronary perfusion. Postmortem studies of patients with HLHS undergoing the NW-BT procedure have attributed the majority of deaths to coronary abnormalities [22]. Furthermore, it has been shown that infants with HLHS have less perfusion and oxygen delivery to the systemic ventricle after the NW-BT procedure compared with infants with fully corrected congenital heart disease [36]. The higher diastolic blood pressures may help overcome the inherent difficulty of supplying coronary perfusion through a hypoplastic ascending aorta despite arch augmentation. One study showed improved interstage mortality with strict follow-up and early intervention if there was systemic desaturation or poor weight gain [12]. One could hypothesize that if the patients who underwent the NW-BT procedure were exposed constantly to decreased coronary perfusion, their cardiac output would eventually decrease. Decreased cardiac output would decrease mixed venous saturation and therefore systemic saturation and decreased cardiac output would also explain the poor weight gain.
Limitations of this study include all the limitations of the reviewed manuscripts including the inherent shortcomings of a meta-analysis. Etiologies of interstage deaths were not documented, so we can only speculate on causes. Two studies did not record time to stage II; therefore, we cannot rule out the possibility that NW-BT patients may have had "more time to die" if they had their stage II performed later compared with the NW-RVPA patients. None of the studies were randomized prospective trials comparing the two different procedures. In general, the two cohorts of patients in the different studies had surgery during different time periods as institutions transitioned from NW-BT procedure to the NW-RVPA procedure. Advances in preoperative, operative, and postoperative care cannot be ruled out as a possible reason for improvement in interstage deaths.
In conclusion, this study revealed that there was a significant difference in interstage mortality between the NW-BT versus the NW-RVPA operation after review of the available literature. Midterm and long-term results for this procedure are still unknown, and the concern for ventricular dysfunction and arrhythmia due to the right ventriculotomy will likely be unanswered in the immediate future; however, the decreased surgical and interstage mortality rate makes this procedure an attractive option. Only a large, multicenter, randomized, controlled trial will be able to truly show whether there are any advantages of one procedure over the other.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Caspi, T. W. Pettitt, T. Mulder, and A. Stopa Development of the pulmonary arteries after the norwood procedure: comparison between blalock-taussig shunt and right ventricular-pulmonary artery conduit. Ann. Thorac. Surg., October 1, 2008; 86(4): 1299 - 1304. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Atallah, I. A. Dinu, A. R. Joffe, C. M.T. Robertson, R. S. Sauve, J. D. Dyck, D. B. Ross, I. M. Rebeyka, and the Western Canadian Complex Pediatric Therapies F Two-Year Survival and Mental and Psychomotor Outcomes After the Norwood Procedure: An Analysis of the Modified Blalock-Taussig Shunt and Right Ventricle-to-Pulmonary Artery Shunt Surgical Eras Circulation, September 30, 2008; 118(14): 1410 - 1418. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Scheurer, J. W. Salvin, V. L. Vida, F. Fynn-Thompson, E. A. Bacha, F. A. Pigula, J. E. Mayer Jr, P. J. del Nido, D. L. Wessel, P. C. Laussen, et al. Survival and clinical course at fontan after stage one palliation with either a modified blalock-taussig shunt or a right ventricle to pulmonary artery conduit. J. Am. Coll. Cardiol., July 1, 2008; 52(1): 52 - 59. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Bradley, C. C. Erdem, T.-Y. Hsia, A. M. Atz, V. Bandisode, and J. M. Ringewald Right Ventricle-to-Pulmonary Artery Shunt: Alternative Palliation in Infants With Inadequate Pulmonary Blood Flow Prior to Two-Ventricle Repair. Ann. Thorac. Surg., July 1, 2008; 86(1): 183 - 188. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Pizarro, C. D. Derby, J. M. Baffa, K. A. Murdison, and W. A. Radtke Improving the outcome of high-risk neonates with hypoplastic left heart syndrome: hybrid procedure or conventional surgical palliation? Eur. J. Cardiothorac. Surg., April 1, 2008; 33(4): 613 - 618. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Lai, P. C. Laussen, C. L. Cua, D. L. Wessel, J. M. Costello, P. J. del Nido, J. E. Mayer, and R. R. Thiagarajan Outcomes After Bidirectional Glenn Operation: Blalock-Taussig Shunt Versus Right Ventricle-to-Pulmonary Artery Conduit Ann. Thorac. Surg., May 1, 2007; 83(5): 1768 - 1773. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. N. Fenton, K. Lessman, K. Glogowski, S. Fogg, and K. F. Duncan Cerebral Oxygen Saturation Does Not Normalize Until After Stage 2 Single Ventricle Palliation Ann. Thorac. Surg., April 1, 2007; 83(4): 1431 - 1436. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Booth Surgical Outcome for Hypoplastic Left Heart Syndrome AAP Grand Rounds, September 1, 2006; 16(3): 29 - 30. [Full Text] [PDF] |
||||
![]() |
O. Reinhartz, V. M. Reddy, E. Petrossian, M. MacDonald, J. J. Lamberti, S. J. Roth, G. E. Wright, S. B. Perry, S. Suleman, and F. L. Hanley Homograft Valved Right Ventricle to Pulmonary Artery Conduit as a Modification of the Norwood Procedure Circulation, July 4, 2006; 114(1_suppl): I-594 - I-599. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |