HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Karl F. Welke Christopher Komanapalli Ross M. Ungerleider Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Welke, K. F. Articles by Ungerleider, R. M. Search for Related Content PubMed PubMed Citation Articles by Welke, K. F. Articles by Ungerleider, R. M. Related Collections Congenital - cyanotic

Ann Thorac Surg 2007;84:624-629
© 2007 The Society of Thoracic Surgeons

---

Original Articles: Cardiovascular

Population-Based Perspective of Long-Term Outcomes After Surgical Repair of Partial Atrioventricular Septal Defect

^a Division of Cardiothoracic Surgery, Oregon Health and Science University, Portland, Oregon
^b Department of Medical Informatics and Clinical Epidemiology, Oregon Health and Science University, Portland, Oregon
^c Division of Pediatric Cardiology, Oregon Health and Science University, Portland, Oregon

Accepted for publication March 26, 2007.

^_* Address correspondence to Dr Welke, Division of Cardiothoracic Surgery L353, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd, Portland, OR 97239–3098 (Email: welkek{at}ohsu.edu).

Presented at the Fifty-third Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 8–11, 2006.

	Abstract

Top Abstract Introduction Patients and Methods Results Comment Discussion Acknowledgments References

 
Background: This investigation was designed to determine long-term survival, reoperation rates, and functional status after surgical repair of partial atrioventricular septal defect (PAVSD).

Methods: This population-based cohort study with cumulative, prospective follow-up by questionnaire and medical record review included all patients aged younger than 19 years old in the state of Oregon who underwent surgical repair of a PAVSD from 1958 to 2000. The incidence of early death, late death, and reoperation for left atrioventricular valve pathology were determined. Patient-reported health status as measured by the Medical Outcomes Study Short Form 12 (SF-12) was obtained for patients without Down syndrome when they were aged older than 15 years.

Results: Repair of PAVSD was done in 133 patients. Median follow-up was 8.7 years for a total of 1541 person-years. Mean age at the initial operation was 5.2 ± 5.1 years (median, 3.4 years). Mean weight was 19.2 ± 16.0 kg (median, 13.2 kg). Survival was 95% at 30 days, 87% at 10 years, and 78% at 30 years. Reoperation for left atrioventricular valve pathology was done 15 patients (11.3%). Lower weight, absence of Down syndrome, and lack of mitral valve cleft repair were significantly associated with undergoing reoperation. Patient-reported health status was obtained in 35 patients. For this group, the mean SF-12 summary scores for the physical component (52.8 ± 9.0) and the mean mental component (50.3 ± 11.0) were not significantly different from age-adjusted norms.

Conclusions: The survival rate for this simple cardiac defect is lower than the general population. In addition, the reoperation rate is significant. Despite this, in general, patients without Down syndrome can expect normal functional health status.

	Introduction

Top Abstract Introduction Patients and Methods Results Comment Discussion Acknowledgments References

 
Health care has entered an era where patients, their families, employers, and insurance companies are asking questions about the adequacy of care. Against this background, there is a growing population of adults who underwent repair of congenital heart defects as children and were told their defects were completely repaired. These patients are now resurfacing both with residual pathology and questions about how their defects might impact their future. In addition, patients and the families of patients currently undergoing surgery desire information about the expected long-term outcomes of their lesions.

Unfortunately, the lack of organized, prospective, long-term data collection and the dearth of studies accurately describing the long-term outcomes of these patients leave providers unprepared to answer such queries. Retrospective studies of adult patients seen in specialized clinics cannot define denominators. Single-institution studies may retrospectively track cohorts of patients for short or long periods of time but include specific populations of patients that may not be representative of patients at other institutions or the broader population. The primary outcome tracked in most studies is short-term mortality, which is only one of many outcomes important to patients. As a result, there is a paucity of data that can be referenced when discussing the answers to questions about long-term outcomes after surgical repair.

The present investigation attempts to add to the literature on long-term outcomes for one relatively common congenital heart defect, partial atrioventricular septal defect (PAVSD). We chose this lesion because favorable outcomes are expected despite a scarcity of long-term data. Our purpose is to use the information available about one congenital cardiac lesion to demonstrate the kind of outcomes studies we need to do in the future so that we can provide patients and their families the information they desire. For this analysis, we used a statewide, population-based database of all patients who underwent repair of PAVSD during a 43-year period. We determined long-term survival, reoperation rates, functional health status after surgical repair of PAVSD, and as feasible, risk factors associated with these outcomes.

	Patients and Methods

Top Abstract Introduction Patients and Methods Results Comment Discussion Acknowledgments References

 
The Oregon Health and Science University Institutional Review Board approved this study. The need for patient consent was waived.

In 1982, we instituted a population-based registry to enroll all Oregon residents who had surgical repair from 1958 to 2000 of major congenital heart defects when they were younger than 19 years old [1]. The surgical procedures were performed at five Oregon hospitals; 0.7% of these procedures occurred at hospitals outside Oregon. Over time, this registry has been expanded to include 12 major heart defects, including PAVSD.

To form the registry, medical records departments in all Oregon hospitals that performed cardiac or thoracic surgery were asked to identify cases by using procedure and diagnostic codes of hospital admissions. Computerized records, card files of hospital admissions, and surgical logs were searched to identify cases. Since 1982, data have been added to the registry prospectively with yearly ascertainment of surgical cases. The second author or a research assistant, who had undergone training and monitoring to maintain data quality and integrity, performed chart abstraction.

To obtain long-term follow-up information, subjects were traced through next of kin, physicians, employment records, Department of Motor Vehicles registrations, city and telephone directories, and the Social Security National Death Index. Follow-up status of all individuals in the registry was determined by a mailed questionnaire every 2 years. Because of funding limitations, the last formal follow-up cycle for this cohort began in 1993 and ended in 1998. Some additional information accrued after that time and is included in this analysis. Individuals who did not complete the mailed questionnaire were contacted by telephone for a formatted interview. In addition to assessing functional status, the questionnaire asked the individual or their family about specific events such as endocarditis, recurrent operation, or any hospitalization. Affirmative answers to reoperations and all hospitalizations were confirmed through medical records or patients’ physicians. Death certificates or hospital records of all deaths were obtained.

Patient-reported health status as measured by the Medical Outcomes Study 12-item Short-Form Health Survey (SF-12) was obtained for patients without Down syndrome when they were older than 15 years of age. The SF-12 is a standardized, validated self-administered tool that measures patient functional health in eight domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These eight subscores are summarized by a physical component scale (PCS) and a mental component scale (MCS). Possible raw scores range from 0 to 100, with 100 representing the most unrestricted functionality. We report norm-based scores, which are based on United States population means of 50, with a standard deviation of 10 [2].

The cumulative incidences of death and reoperation for left atrioventricular valve pathology were estimated by the Kaplan-Meier method. These were defined as elapsed time from the date of operation to the date of the event or, in subjects free of reoperation or death, the date of the last date of contact with the subject. Cohort differences in proportions were determined by the ² test. Significance levels were assessed using the Breslow-Gehan method. The same analysis sequence was used to evaluate the association of tested variables with the incidence of reoperation. Age (<2 years of age versus 2 years of age), weight (<10 kg versus 10 kg), and cardiopulmonary bypass time (<60 minutes versus 60 minutes) were evaluated as dichotomous variables. Comparisons of characteristics by decade were made using ² test for linear trend for discrete data and analysis of variance for continuous data.

A cohort-expected survival curve was created using the method of Hakulinen (R 2.2.1 statistical computing package) [3, 4]. Cohort-expected survival curves are useful for graphic comparisons, sample size, and forecasting. This method generates a matched control for each subject in the study and then computes the expected survival for the matched controls. Hazard rates for mortality were stratified by age, gender, and calendar year. Each control was followed up until death or censoring of its matched case. The one-sample log-rank test was used to determine if the observed cohort had equivalent survival to the baseline population [5].

We used the estimated cumulative incidence function to create a curve representing the probability of having a reoperation for left atrioventricular valve pathology. This estimate adjusts for both censored cases (patients who are alive but have not had a reoperation at their latest follow-up) and those who have died without having a reoperation. This differentiation is important. Patients who are alive have the potential to have a reoperation. Those patients who are dead do not. If deceased patients are treated as censored, the Kaplan-Meier procedure overestimates the probability for reoperation [6].

	Results

Top Abstract Introduction Patients and Methods Results Comment Discussion Acknowledgments References

 
From 1958 through 2000, 133 patients from Oregon younger than 19 years old underwent surgical repair of PAVSD (Table 1). The number of patients undergoing operation increased over time, reflecting not only the population growth of Oregon but also changes in diagnosis and management (Table 2). The mean age at the initial procedure was 5.2 ± 5.1 years (median, 3.4 years). Over the time period of the study, mean age decreased from 9.2 to 3.4 years old (p < 0.001; Table 2). Mean weight was 19.2 ± 16.0 kg (median, 13.2 kg). Most patients were white, which reflected the Oregon population, and 24.8% had Down syndrome.

View larger version (9K): [in this window] [in a new window]   Fig 1. Comparison of observed survival (solid line) after repair of partial atrioventricular septal defect with the cohort-expected survival (dashed line) in the United States population is shown by a Kaplan-Meier plot with 95% confidence limits (dot-dash line).

View larger version (14K): [in this window] [in a new window]   Fig 2. Estimated probability of having a reoperation for left atrioventricular valve pathology after repair of partial atrioventricular septal defect (solid line). The cumulative incidence function is presented with 95% confidence limits (dot-dash line).

	Comment

Top Abstract Introduction Patients and Methods Results Comment Discussion Acknowledgments References

 
We used a statewide, population-based database of all patients who underwent repair of PAVSD during a 43-year period to determine long-term survival, reoperation rates, and functional health status. The survival rate for this simple cardiac defect is lower than that of the general population. In addition, the reoperation rate for left atrioventricular valve pathology is significant. Despite this, patients without Down syndrome can generally expect normal functional health status. This investigation highlights the need for long-term outcomes data so that patients and their families have more appropriate information on the true results of our "cures" than can be obtained from traditional in-hospital and 30-day follow-up studies.

The literature relating to PAVSD consists of single-institution retrospective cohort studies with primarily short-term or mid-term follow-up [7–17]. The long-term mortality and reoperation rates we found were consistent with one of the larger single-institution retrospective studies, a cohort of patients who underwent surgery for PAVSD at the Mayo Clinic during a similar time period (1955 to 1995) [14].

Multiinstitutional data are available from The Society of Thoracic Surgeons (STS) Congenital Heart Surgery Database. The STS database reported a 0.5% (2/404) in-hospital mortality rate for PAVSD for calendar year 2005; however, mortality rates and reoperation rates after discharge or 30 days are not collected [18].

It is imperative to recognize the limitations of our study. Our database contained both retrospective (before 1982) and prospective (since 1982) data collected from a variety of sources over many years and therefore suffers in part from the biases of retrospective data. In addition, despite multiple methods, follow-up data were not available on all patients. Follow-up of population cohorts for more than 4 decades is expensive and increasingly difficult given mobility and increasing public distrust; therefore, 71% follow-up for this long time interval should be viewed positively.

Information on comorbidities was limited. During the time these operations were done, no clinical database was available to assist with complete, accurate collection of all relevant variables. Most notably, we did not have information on preoperative left atrioventricular valve regurgitation. The diagnosis and management of PAVSD also evolved during the time period of this study. The use of angiography and echocardiography changed substantially. Decision-making and operative techniques, both specific to PAVSD and applicable to all congenital heart defects, changed as well.

Notwithstanding these limitations, our study was population-based. Population-based studies provide the most accurate quantitative estimates of diseases or conditions in a population. This design eliminates selection bias because all individuals in the affected population, in this case the state of Oregon, are included. The population under study is the population at risk. As a result, unlike single-institution studies, clinical trials, multiinstitution studies, or randomized controlled trials, the findings represent the whole population of our state. Trials report what could happen to a select group of patients under carefully controlled conditions. Population-based studies report what actually happened to an entire population. Although differences between the population of Oregon and other states may limit generalizability across the whole United States, our findings are likely to be more generalizable than those from studies of more limited populations.

Much of the early data used in this study were retrospectively collected, but our study did benefit from prospective collection of all data since 1982 and cumulative, prospective follow-up. In addition, we used objective risk factors and measured hard outcomes: death and reoperation. Although it is important to investigate other outcomes such as left atrioventricular valve regurgitation and stenosis, left ventricular outflow obstruction, and supraventricular arrhythmias, the assessments of these outcomes are more subjective. As a consequence, the definitions of these outcomes, the technologies used for measurement, and the strategies used for treatment have changed over time. We chose to measure outcomes that were objective.

Finally, we used the SF-12 to collect patient-reported health status. This general measure of health status captures information missed by disease-specific measures, thereby providing a better overall assessment of a patient’s mental and physical health. Such data are helpful when counseling patients and their families on how their lives could be affected by any functional limitations resulting from a disease or operation.

When we are asked to provide information about the outcomes of the procedures we perform, it is no longer acceptable to render our best guesses for limited outcomes based on small retrospective studies and personal anecdote. Patients deserve answers backed by evidence-based data describing the outcomes important to them: mortality, morbidity, reoperation, neurologic status, and functional status. Our study provides some of this information; however, the larger benefit of our study may be gained by recognizing what it lacks. As a registry, our procedure-based database contained a select list of data elements on all patients in the population. The information collected was decided on by a limited group of individuals and some was collected retrospectively. Future databases should be designed by multidisciplinary panels and include national rather than state populations. They should be disease-based rather than procedure-based so that patients and their families can be given information on the outcomes of all available treatment options. The data should be prospectively collected using standard nomenclature to eliminate the biases associated with retrospective data and the variability inherent in personal definitions.

Most important, systems must be devised for long-term follow-up of patients. Short-term operative mortality is a vital outcome measure, but we expect that most of our patients will live beyond the traditional 30-day follow-up period. Data on long-term neurologic and functional status are essential to help patients and their families make difficult decisions about treatment options. The health care system must recognize that if our specialty is to continue to advance and patients are to get the information to which they are entitled, collection of such data is an essential component of patient care.

	Discussion

Top Abstract Introduction Patients and Methods Results Comment Discussion Acknowledgments References

 
DR JOHN H. CALHOON (San Antonio, TX): Karl, this is a very nice paper and manuscript. It was obviously put together with much thought and effort. These studies are clearly needed to help us as we evaluate our future surgical efforts. I have three questions, and I think I know the answer, but could you tell us why Down syndrome patients had less reops, or did you have any clues to that? Second, although it is possibly beyond the scope of your study, how do you decide or your colleagues decide how much cleft you should close or whether you should even close the cleft at all in some rare cases? And third, you kind of outlined it at the end of your paper but, briefly, maybe highlight the metrics that you wish to include or capture in your future databases to help with your study—the things that were missing in this one, so to speak. Thank you again for a nicely presented paper.

DR WELKE: Thank you, Dr Calhoon. Regarding Down syndrome patients having fewer reoperations, our finding is consistent with other papers that have looked both at partial and complete atrioventricular canal. It may be that Down patients have less severe pathology than non-Down patients, which are more difficult to repair and manage.

As far as how much cleft to close, there is always a fine line between closing enough cleft and creating stenosis. We are careful to evaluate our patients with a postoperative echocardiogram in the operating room to make sure that we have not created any stenosis. Of course, there are special situations, like parachute valve and other oddities, where you would not want to close a cleft at all.

As far as metrics and future databases, it is important that we participate in The Society of Thoracic Surgeons (STS) database as much as possible and support that effort. Unfortunately, at present the STS database only collects data out to 30 days postoperatively or for the length of the hospital stay, whichever is longer. We need to devise databases and systems of collection that allow us to track these patients over the longer term. The outcomes measured need to be those important to patients: mortality, morbidity, functional status, and neurologic status. This is going to require not only our involvement but also participation from insurance companies and perhaps governmental sources in order to make sure that this is a component—perhaps a required component—of care.

DR JEFFREY JACOBS (St. Petersburg, FL): Karl, I congratulate you again for a very nice presentation. I think the points that you made about modifying our databases to have improved methodologies of follow-up are extremely important, and as you stated, the STS database, while being the largest congenital heart surgery database in the world, stops follow-up at discharge plus 30 days, whichever is longer. The challenge I think to creating a database system that has longer-term follow-up and, ideally, lifelong follow-up for patients with congenital heart disease really is twofold: one, obtaining the funding to support that database; and two, engaging the participation of the appropriate cardiology societies who actually do the patient care for the long-term follow-up. I was wondering if you had any thoughts as to how we could meet those two challenges?

DR WELKE: Thank you, Dr Jacobs. The funding question is an interesting one. One of the ways to think about it is if we do not have long-term outcomes data for these procedures, we are essentially performing operations for which we do not know the results. It is difficult to pay for a procedure when you are not sure what is going to happen to the patient long term. If one uses that argument, one could say that the people who are paying for these operations (ie, insurance companies and the government), should be interested in the long-term results to assure that they are getting what they are paying for. Having a dialogue with those groups is important. At the same time, we have to take personal responsibility for tracking our own patients and so we have some say in this as well.

Input from our cardiology colleagues is crucial, especially if data collection is done on a larger scale than in a small state such as Oregon. We should be having these types of presentations and discussions in their meetings as well and emphasize the collaboration that we have in taking care of these patients. Any database that is designed for the future should not be designed solely by surgeons but should be a multidisciplinary effort so that all parties have a stake and are able to include the elements that they feel are most applicable to the relevant outcomes.

	Acknowledgments

Top Abstract Introduction Patients and Methods Results Comment Discussion Acknowledgments References

 
We are thankful for the assistance of Mara Tableman, PhD, Professor of Statistics, Department of Mathematics & Statistics, Portland State University.

	References

Top Abstract Introduction Patients and Methods Results Comment Discussion Acknowledgments References

 

1. Morris CD, Menashe VD. Twenty-five-year mortality after surgical repair of congenital heart defect in childhood JAMA 1991;266:3447-3452.[Abstract/Free Full Text]
2. Ware JE, Kosinski M, Keller SD. A 12-item Short-Form Health Survey (SF-12): construction of scales and preliminary tests of reliability and validity Med Care 1996;32:220-233.
3. Hakulinen T. Cancer survival corrected for heterogeneity in patient withdrawal Biometrics 1982;38:933-942.[Medline]
4. R Development Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3–900051–07–0, 2005. Available at: http://www.R-project.org. Accessed, March 2, 2007.
5. Harrington DP, Fleming TR. A class of rank test procedures for censored survival data Biometrika 1982;69:553-566.[Abstract/Free Full Text]
6. Gooley TA, Leisenring W, Crowley JC, Storer BE. Why the Kaplan-Meier fails and the cumulative incidence function succeeds when estimating failure probabilities in the presence of competing risksIn: Crowley JC, editor. Handbook of Statistics in Clinical Oncology. New York, NY: Marcel Dekker, Inc; 2000. pp. 513-523.
7. Rastelli GC, Weidman WH, Kirklin JW. Surgical repair of the partial form of persistent common atrioventricular canal, with special reference to the problem of mitral valve incompetence Circulation 1965;31(suppl):I31-I35.
8. Losay J, Rosenthal A, Castaneda AR, Berhard WH, Nadas AS. Repair of atrial septal defect primimResults, course, and prognosis. J Thorac Cardiovasc Surg 1978;75:248-254.[Abstract]
9. Goldfaden DM, Jones M, Morrow AG. Long-term results of repair of incomplete persistent atrioventricular canal J Thorac Cardiovasc Surg 1981;82:669-673.[Abstract]
10. Portman MA, Beder SD, Ankeney JL, van Heeckeren D, Liebman J, Riemenschneider TA. A 20-year review of ostium primum defect repair in children Am Heart J 1985;110:1054-1058.[Medline]
11. Cethami EL, Midgley FM, Perry LW. Long-term results after surgical repair of incomplete endocardial cushion defects Ann Thorac Surg 1989;48:413-416.[Abstract]
12. Giamberti A, Marino B, di Carlo D, et al. Partial atrioventricular canal with congestive heart failure in the first year of live: surgical options Ann Thorac Surg 1996;62:151-154.[Abstract/Free Full Text]
13. Baufreton C, Journois D, Leca F, Khoury W, Tamisier D, Vouhe P. Ten-year experience with surgical treatment of partial atrioventricular septal defect: risk factors in the early post-operative period J Thorac Cardiovasc Surg 1996;112:14-20.[Abstract/Free Full Text]
14. Najm HK, Williams WG, Chuaratanaphong S, Watzka SB, Coles JG, Freedom RM. Primum atrial septal defect in children: early results, risk factors, and freedom from reopertion Ann Thorac Surg 1998;66:829-835.[Abstract/Free Full Text]
15. El-Najdawi EK, Driscoll DJ, Puga FJ, et al. Operation for partial atrioventricular septal defect: a forty-year review J Thorac Cardiovasc Surg 2000;119:880-890.[Abstract/Free Full Text]
16. Cope JT, Fraser GD, Kouretas PC, Kron IL. Complete versus partial atrioventricular canal: equal risks of repair in the modern era Ann Surg 2002;236:514-521.[Medline]
17. Al-Hay AAA, Lincoln CR, Shore DF, Shinebourne EA. The left atrioventricular valve in partial atrioventricular septal defect: management strategy and surgical outcome Eur J Cardiothorac Surg 2004;26:754-761.[Abstract/Free Full Text]
18. The Society of Thoracic Surgeons Congenital Heart Surgery Database–Sixth Harvest–(2002–2005). Durham, NC: The Society of Thoracic Surgeons (STS) and Duke Clinical Research Institute (DCRI), Duke University Medical Center; 2006.

This article has been cited by other articles:

				R. Formigari, G. Michielon, M. C. Digilio, G. Piacentini, A. Carotti, A. Giardini, R. M. Di Donato, and B. Marino Genetic syndromes and congenital heart defects: how is surgical management affected? Eur. J. Cardiothorac. Surg., April 1, 2009; 35(4): 606 - 614. [Abstract] [Full Text] [PDF]

				U. K. Chowdhury, B. Airan, A. Malhotra, A. K. Bisoi, M. Kalaivani, R. M. Govindappa, and P. Venugopal Specific issues after surgical repair of partial atrioventricular septal defect: actuarial survival, freedom from reoperation, fate of the left atrioventricular valve, prevalence of left ventricular outflow tract obstruction, and other events. J. Thorac. Cardiovasc. Surg., March 1, 2009; 137(3): 548 - 555.e2. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Karl F. Welke Christopher Komanapalli Ross M. Ungerleider Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Welke, K. F. Articles by Ungerleider, R. M. Search for Related Content PubMed PubMed Citation Articles by Welke, K. F. Articles by Ungerleider, R. M. Related Collections Congenital - cyanotic