 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 105 No. 5 May 2000, pp. 1073-1081
, and
From the * Division of Cardiology, Department of Pediatrics,
Harbor-UCLA Medical Center, Torrance, California; and Division of
Cardiology, Department of Pediatrics, University of California, Los
Angeles School of Medicine, Los Angeles, California.
 |
ABSTRACT |
|---|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Objective. Previous studies have shown that children with congenital heart disease (CHD) who live in nonurban areas or who do not have private insurance are at risk for delayed referral to a pediatric cardiologist. However, the effect of these factors on the age at which cardiac surgery is performed has not been evaluated. This study is designed to evaluate the factors that influence the age at which definitive surgical repair is performed.
Methods. Data on hospital discharges for 1995 and 1996 in California were obtained from the Office of Statewide Health Planning and Development database. Children <18 years who underwent surgical repair for atrial septal defect (ASD), ventricular septal defect (VSD), tetralogy of Fallot (TOF), or atrioventricular canal (AVC) were included in the study. Age at surgery was evaluated using type of CHD, gender, race, type of insurance, surgical centers, urban or rural home location, and distance between home and surgical center as independent variables.
Results. In 1995-1996, 666 children underwent ASD closure
(mean age: 5.1 years; median: 4.0 years), 582 VSD closure (mean age:
2.8; median: 1.1 years), 394 TOF repair (mean age: 1.7; median: .9 years), and 177 AVC repair (mean age: 1.1; median: .6 years). Comparing
median and mean age at surgery, we found: AVC<TOF<VSD<ASD (< indicates younger than). A consistent trend for all 4 types of CHD was
seen indicating that for median age at operation: private insurance<managed care<Medicaid. Gender or race had no effect on age
at operation, although Asians tended to be older at surgery for all 4 types of CHD. There is a significant negative correlation between the
case volume of surgical centers and median age at operation for ASD
(r = 
.37), VSD (r = .49),
TOF (r = .63), and AVC (r = .17). In addition, significant positive correlation was found
between degree of urbanization of home locations (measured by
population density) and median age at operation for ASD
(r = .50), VSD (r = .77), and
TOF (r = .18). No significant correlation was found
between distance to surgical center and age at operation.
Conclusions. Many medical and nonmedical variables play important roles in determining age for definitive repair of CHD in children. Type of insurance, a recognized surrogate for access to care, may play an important role. In addition, centers with higher surgical case volume were more likely to operate at a younger age. Finally, children in urban areas tend to be older at the time of surgery for ASD, VSD, and TOF. Key words: age, surgery, congenital heart disease, children.
Congenital heart disease (CHD) remains one of the most
common forms of severe congenital defect. Advances in pediatric
cardiology, such as the wide availability of echocardiography and
recent development of fetal diagnosis, have significantly decreased the
age of diagnosis and referral of neonates and infants with CHD to
surgical centers. Newly developed techniques for surgical intervention,
myocardial preservation, and postoperative care have allowed complete
repair of many congenital heart defects in the neonatal period or early infancy. For some lesions, such as complete transposition of the great
arteries, lower early postoperative morbidity and mortality are
achieved if an arterial switch operation is performed in the first few
weeks of life.1 However, the optimal age for complete
repair of most other congenital heart defects remains undefined or
controversial. Despite continuing controversies regarding optimal
timing for surgical repair of many forms of CHD, data collection and
research efforts addressing age at the time of cardiac surgery have
been limited.
In previous studies, age at diagnosis and referral has been identified
as an indicator of access to medical care in pediatric patients. In a
retrospective study of 379 561 live births in Dallas County, Texas
between 1971 and 1984, Fixler et al2 found differences in
the prevalence rates of specific cardiac defects among ethnic groups.
However, timing of referral for pediatric cardiac care was not related
to ethnicity, median family income, or household educational level. A
more recent study by Perlstein et al3 focused on the
factors influencing age at referral to a pediatric cardiologist for
children with CHD. Perlstein et al3 found that the major
risk factors for delayed referral were insurance other than commercial,
nonurban home location, and type of CHD. These observations suggest an
important question: does delayed referral to a pediatric cardiologist
result in delayed diagnosis and surgery that might affect clinical
outcomes for children with CHD?
By using a large database containing information on all hospital
discharges in California in 1995 and 1996, we sought to identify the
factors associated with, or influencing the age at which, children
undergo definitive surgical repair of the most common forms of CHD.
Specifically, our objectives were to: 1) describe the distribution of
age at surgery for repair of CHD; 2) determine the factors associated
with age at surgery; and 3) identify risk factors for delayed surgical
repair of CHD in children.
In this study, we used a large administrative database to select
children who had surgical repair of common types of congenital heart
defects and analyzed factors associated with age at which these
surgeries were performed.
Database
We used 1995 and 1996 hospital discharge data from the
California Office of Statewide Health Planning and Development
(OSHPD) database. The database contains International
Classification of Diseases, Ninth Revision, Clinical Modification
(ICD-9-CM) discharge diagnosis and procedure codes assigned by
California hospitals to each individual discharge during the year.
Fields are provided for up to 25 diagnoses and 25 procedures. The
database includes routine demographic and administrative information
such as age, gender, race, admission type and source, discharge status,
length of hospitalization, and total hospital charges. This
administrative database contains important and valuable information on
health care use and outcomes and has been used in many health services studies.4-6
Case Selection
Data from the OSHPD database for 1995 and 1996 were obtained in
a format compatible with Microsoft Access 2.0 (Microsoft Corporation, WA). For purposes of this study, patients were selected for age between
birth and 17 years and an All Patient Refined Major Diagnostic Category
code of 05. These Major Diagnostic Categories are mutually exclusive
categories containing all possible principal diagnostic areas. The
diagnoses in each Major Diagnostic Category correspond to a single
major organ system or cause and code 05 refers to the cardiovascular
system. Thus, the selection criteria produced a list of children under
18 years of age with any cardiac diagnosis. Subsequently, the Health
Care Financing Administration Diagnosis Related Group was used to
exclude noncongenital cardiovascular diagnosis (codes 109, 113, 114, 119, 128, 132-134, and 140-143). This procedure eliminated all but
6933 cases in 1995 and 6314 cases in 1996. From this list, patients
were selected whose ICD-9-CM principal diagnosis code was secundum type
of atrial septal defect (ASD; code 745.5), ventricular septal defect
(VSD; code 745.4), tetralogy of Fallot (TOF; code 745.2), or
atrioventricular canal defect (AVC, code 745.6). After patients with
the principal diagnosis of ASD, VSD, TOF, or AVC were selected,
procedure codes were used to identify patients who had cardiac
surgeries. Patients without the ICD-9-CM procedure code for
cardiopulmonary bypass (39.61) were excluded from the study to
eliminate the possibility of ambiguous or erroneous coding of
palliative surgical procedures, transcatheter closure of ASD, VSD, and
other catheterization procedures. Principal procedure codes of 35.51, 35.61, and 35.71 were used to identify surgical closure of ASD; codes
35.53, 35.62, and 35.72 for closure of VSD; code 35.81 for repair of
TOF; and codes 35.54, 35.63, and 35.73 for repair of AVC. Patients with
ASD closure who had other procedures listed (such as VSD closure, AVC
repair, repair of anomalous pulmonary veins, or other miscellaneous
procedures not consistent with simple closure of secundum ASD) were
excluded. Patients with VSD closure who had other procedures listed
(such as AVC repair, TOF repair, or right ventricular outflow
tract/pulmonary valve procedure, or other procedures suggestive of
complex anatomy in addition to VSD) were excluded from the VSD group.
Patients with TOF repair who had AVC listed, or patients with AVC
repair who had TOF repair listed as an additional procedure, were also excluded.
Other Data Sources
Information on land areas and population of the 58 counties in
California were obtained from the 1996 City and County
Extra.7
Data Processing
In this study, the dependent variable age was expressed in days
or years. In the OSHPD database, the age of patient under 4 years old
is recorded as days and years (eg, a 2 years and 2 months old patient
is recorded as 2 years and 60 days). Over 4 years of age, the database
contains only age in years (expressed as an integer). The variable
payer was derived from the expected source of payment, which was coded
into 14 categories in the database. In this study, we simplified the
payer status to 4 categories: Medi-Cal (California's Medicaid program)
or California Children Services (CCS; California's Title V-designated
children with special health care needs agency), managed care health
plans (including health maintenance organization [HMO] and preferred
provider organization [PPO]), private insurance (including non-HMO
Blue Cross/Blue Shield), and others (self-pay, charity care, etc).
Comparisons were made among 3-payer groups: Medi-Cal/CCS, managed care,
and private insurance. The variable race included white, black, Native
American/Eskimo/Aleut, Asian/Pacific Islander, other, and unknown. The
variable surgical center represents the hospital at which a given
patient's surgery was performed. The degree of urbanization of home
location was measured as the population density of the county of
residence (number of residents per square kilometer). Using the
computer program US Street Atlas, Version 5 (Delorme Inc,
ME),8 we calculated the distance measured as the number of
street or highway miles between patient's home and the surgical center
where surgery was performed for each patient using the zip code of home location and zip code of the surgical center as the 2 geographic endpoints.
Statistical Analysis
For the purposes of this analysis, age at surgery was the main
outcome (dependent) variable. The distribution of age at operation for
each type of CHD was first assessed for skewness of distribution. To
test whether age at surgery was normally distributed, a 1-sample Kolmogorov-Smirnov goodness-of-fit test was applied. The mean (± standard error of the mean [SEM]) and median age at operation were
calculated for each type of CHD. Mann-Whitney U test was used to compare the medians between pairs of groups and Kruskal-Wallis 1-way analysis of variance test was used to compare the medians of
multiple groups. When multiple comparisons were made, Bonferroni's correction was used. Using Cox regression and hazard function analysis,
we calculated the relative hazard of undergoing CHD surgery at a given
age for patients with private insurance, Medi-Cal, and HMO/PPO. We also
used hazard function analysis to compare the relative hazard for
undergoing repair of a cardiac lesion between white and non-white
populations. For all 4 types of CHD selected in this study, we
conducted 2 weighted linear regression analyses. The first weighted
linear regression used median age of patients with repair of specific
type of CHD at a surgical center as the dependent variable and case
volume (number of cases for repair of the specific type of CHD) of the
surgical center as the independent variables. Each center was weighted
by its case volume for repair of the specific type of CHD. The second weighted linear regression analysis used the median age of patients residing in the same county who underwent repair of each specific type
of CHD as the dependent variable, and population density (number of
residents per square kilometer) as the independent variable. Each
county was weighted by the number of children with the specific type of
CHD who resided in the county and underwent surgical repair of the type
of CHD. All statistical analyses were performed using SPSS 7.0 for
Windows (SPSS Inc, Chicago, IL).
Type of CHD
Using the case selection criteria described above, we found 666 cases of ASD closure, 582 VSD closure, 394 TOF repair, and 177 AVC
repair in California for the years 1995 and 1996. Children with ASD
closure had a mean age at operation of 5.1 years (median: 4 years), VSD
closure mean age 2.8 years (median: 1.1 years), TOF repair mean age 1.7 years (median: .9 year), and AVC repair mean age 1.1 year (median: .6 year). Table 1 lists the mean and median
ages at repair of ASD, VSD, TOF, and AVC in days. Ninety-five (54%)
children with AVC also had the diagnosis of Down syndrome. Children
with Down syndrome and AVC tended to be younger at AVC repair compared
with children with AVC without Down syndrome (median age: 187 and 233 days, respectively; P = .19; mean age: 306 and 529 days, respectively; P < .01).
TABLE 1
METHODS
Top
Abstract
Methods
Results
Discussion
Conclusion
References
RESULTS
Top
Abstract
Methods
Results
Discussion
Conclusion
References
Age at Operation (in Days) and Type of Insurance
Comparing the age at operation for different types of CHD using both mean (multiple t test) and median (Mann-Whitney U test), we found a consistent order: AVC<TOF<VSD<ASD (< indicates younger than; all P < .05 with Bonferroni's correction).
Distribution of Age at Surgery
The age distributions of VSD, TOF, and AVC were not consistent with a normal distribution using Kolmogorov-Smirnov goodness-of-fit test. The distributions of age at surgery were skewed toward younger ages for VSD, TOF, and AVC (Fig 1). The standard error of the skewness for VSD was 2, and for TOF and AVC were greater than 2. The peak of the distribution curve was 6 months for repair of AVC, 9 months for TOF, 12 months for VSD, and 4 years for ASD. Because age at surgery was not normally distributed, the following statistical comparisons were based on median age instead of mean age.
|
The percentages of children with unrepaired ASD, VSD, TOF, or AVC are plotted in relation to age and shown in Fig 2. Ninety percent of children with AVC were repaired by 3 years of age, and 90% of children with TOF were operated on by 5 years of age. For children with VSD, 90% were operated on by 8 years of age.
|
Type of Insurance
Table 1 lists the mean (± SEM) and median age at operation for ASD, VSD, TOF, and AVC of children covered by different types of insurance. When comparing the median age at operation, a consistent trend was found that, for all 4 types of CHD, patients with private insurance were younger at surgery than patients with managed care health plans, and patients with managed care health plans were younger than patients with Medi-Cal/CCS (private insurance<HMO/PPO<Medi-Cal; Fig 3). Using Cox regression analysis, the overall probability for TOF repair for children covered by private insurance was 1.5 times that of children with Medi-Cal or HMO/PPO (P = .05). For AVC, the overall probability of complete repair for children with private insurance was 1.8 times of that for children with Medi-Cal or HMO/PPO (P = .07). No difference was found between children with Medi-Cal and children with HMO/PPO for TOF or AVC. For ASD and VSD, no significant difference was found among children with private insurance, Medi-Cal, or HMO/PPO. Hazard function plots for all 4 types of CHD are shown in Fig 4.
|
|
Gender and Race
There was no statistical difference between males and females regarding median age at operation for any of the 4 types of CHD. Age at operation was also compared among white, black, Asian, and others. Native Americans were excluded from the analysis because of small sample size. In general, Asians tended to be older at surgery for all 4 types of CHD (Fig 5). Compared with the white or black group, the Asian group was older for VSD closure (P < .05). Asians were also older for AVC and TOF repair, although no statistical significance was found when comparing with white or black. There was no difference in age at operation for ASD among the 4 racial groups. No difference was found between white and black patients for repair of ASD, VSD, TOF, or AVC. Hazard function analysis showed no difference in the risk for unrepaired CHD between white and non-white patients in any of the 4 types of CHD.
|
Surgical Centers
In a weighted linear regression analysis, we found that surgical
centers with higher case volumes for repair of each type of CHD tended
to perform operations at a younger age. Negative correlation between
median age and case volume of surgical center was found for repair of
ASD (r = .37), VSD (r = .49), TOF
(r = .63), and AVC (r = .17; all
P values <.01). Fig 6 shows
the scatter diagrams and regression lines of median age at operation
and the case volume of surgical center for the 4 types of CHD.
|
Home Location
To determine the relationship between rural or urban home location and age at operation, we used population density (number of residents per square kilometer of the county) to represent the degree of urbanization of each home location. Using weighted linear regression, significant direct (positive) correlation between the median age at surgery and population density of the county of residence was found for closure of ASD (r = .50), VSD (r = .77), and repair of TOF (r = .18; all P values <.01). However, there was no significant correlation found for repair of AVC.
Distance to Travel
No statistical correlation between median age at operation and distance between patient's home and surgical center was found for ASD, VSD, TOF, or AVC.
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
In the present study, we have demonstrated that a variety of factors influence the age at which children with CHD undergo surgical repair. These factors can be categorized into 2 groups: medical and nonmedical variables. We have shown that medical variables, such as type of CHD, presence of Down syndrome, and differences in practice among pediatric cardiac centers, may influence the age of surgical repair. In addition, we have identified several nonmedical variables that seem to be important, including: race, type of insurance, and home location.
Medical Variables
CHD includes a wide spectrum of congenital cardiac malformations differing in anatomic complexity and physiologic variability. For some forms of CHD, there is evidence suggesting the optimal timing for surgical repair. However, for most forms of CHD, there is not adequate evidence or consensus regarding the age at which definitive surgical repair should be performed.9 Traditionally, many factors are considered when deciding the age for repair, including operative mortality, complications, and long-term outcomes. For example, the relationship between intellectual function and age at repair of CHD has been examined by many investigators.10,11
Of the 4 common forms of CHD selected in the present study, a consensus seems to be developing that better outcomes are achieved with early AVC repair. Yamaki et al12 reported early pulmonary vascular changes in children with AVC and recommended repair AVC within the first 6 months of life to prevent long-term morbidity. Hanley et al13 reviewed the results of surgical repair of AVC and found a drop in mortality from 25% to 3% over the last 20 years and provided support for aggressive approach for repair in the first 3 months of life. In addition, children with Down syndrome are more vulnerable to early pulmonary vascular changes.12 Therefore, early repair of AVC to prevent permanent pulmonary vascular changes in children with Down syndrome has been the practice in many surgical centers. Our data indicate that children with Down syndrome tended to have AVC repair at a younger age, compared with children with AVC who did not have Down syndrome.
For TOF, many centers now perform complete repair in the neonatal period or early infancy instead of palliating these children with a systemic-to-pulmonary artery shunt. Pigula et al14 reported low morbidity and mortality of early complete TOF repair in a series of 99 infants younger than 90 days of age. However, some centers continue with a conservative approach toward the surgical management of TOF with initial palliation in the neonatal period and complete repair after 18 months of age.15
Clinical decisions regarding repair of VSD often depend on the number, location, and size of the defects. Because of anatomic and physiologic variability, the timing of VSD repair for individual patients may vary. It is not generally controversial that patients with large nonrestrictive VSDs and concomitant heart failure or failure to thrive or VSDs in the subpulmonic position (risk of aortic cusp prolapse and aortic regurgitation) should receive earlier surgical repair. In contrast, controversy exists over repairing restrictive VSDs. The optimal timing of surgery for these patients has not been established.16
Children with ASD are usually asymptomatic. Therefore, the age at which patients with ASDs are diagnosed may vary significantly. The general approach in our institution is to repair ASD electively between 3 and 4 years of age to avoid the anxiety and stress that accompanies surgery performed in close proximity with starting school.
An interesting finding of this study is that centers performing larger number surgeries tended to operate earlier for all 4 types of CHD. We speculate that this finding can be explained either by selective referral or a systematic difference in the practice philosophy of surgical centers. It has been shown that pediatric cardiac surgical centers with larger case volumes have lower mortality rates for many forms of CHD.6,17 It is possible, therefore, that a greater number of selective referrals for younger patients are made to centers with larger case volumes because of better surgical outcomes associated with such centers. It is also possible that centers with large case volumes and more experience in operating on young patients may adopt a more aggressive approach toward early definitive repair of CHD. Whether the apparent negative correlation between case volume and age at surgery is attributable to selective referral of young patients to large centers or differences in practice approaches requires further investigation.
Nonmedical Variables
Nonmedical variables that influence the age at operation for children with CHD are predominantly related to access to medical care (financial or nonfinancial variables), and socioeconomic/cultural issues. In a study examining the age at referral of children with common forms of CHD, Perlstein et al3 found the major risks for delayed referral were nonurban location, insurance other than commercial, and type of CHD. In our study, we found that for repair of TOF or AVC, patients with Medi-Cal were older than patients with managed care, and patients with managed care were older than patients with private insurance at the time of definitive repair. In the hazard function analysis, the cumulative hazard of undergoing surgical repair for patients with private insurance for TOF was 1.5 times greater than patients with HMO/PPO or Medi-Cal (P = .05). For AVC, similar differences were found (1.8 times greater for patients with private insurance; P = .07). Thus, we found a trend that patients with private insurance were operated on earlier than patients with HMO/PPO, and that patients with HMO/PPO were operated on earlier than patients with Medi-Cal (Fig 3). This finding is consistent with the report by Perlstein et al3 on age at referral for CHD. However, in our study no statistically significant effect of insurance type was found for patients with VSD or ASD. One explanation for the differences between our study and the study by Perlstein et al may be that we used median age at repair instead of mean age (as used by Perlstein et al) because of the extreme skew of the age distributions. Other distinctions between our study and the study by Perlstein et al include differences in practice patterns between physicians in Arizona versus physicians in California and criterion of age at referral versus age at surgery.
In a weighted linear regression analysis, we found a direct correlation between the population density of the county of residence and median age at operation for ASD, VSD, and TOF (but not for AVC). In contrast, Perlstein et al3 found that urban children were referred at a younger age than nonurban children. Although there are methodological differences between the 2 studies (definition of urban vs rural and statistical methods used for analysis), the population characteristics of California and Southern Arizona are also different. We speculate that urban children tended to be older at surgery in California attributable to the socioeconomic conditions that are prevalent in the inner cities of many large metropolitan areas in California. The population characteristics and socioeconomic indicators, particularly in large urban areas, may be significantly different between California and Arizona. For example, the population characteristics of California cities, compared with other large cities in the United States, show a faster population growth (growth rate: 22.2% vs 6.5%), more whites and Hispanics, and greater percentage of foreign-born residents (29.8% vs 15.3%).18
We have also noted that Asian children tended to be older at surgery for VSD, TOF, and AVC repair (Fig 5). Although the meaning of this tendency is unclear, this finding raises the question whether this difference between the Asian group and other races is attributable to poor access to care or to the cultural background of the Asian population. More studies are needed to better understand differences among racial groups regarding access to care for children with CHD.
Limitations
In the present study, we used a large administrative database to analyze the medical and nonmedical variables associated with age at operation for common forms of CHD. Therefore, the analysis was limited with respect to detailed clinical information. For example, we found no differences in age at operation for VSD among the 3 payer groups. However, no information regarding the size and location of VSDs was available in the database. In addition, it was not possible to determine with reliability which patient with TOF had previous palliative surgeries. Without the ability to include important medical variables, conclusions regarding both medical and nonmedical variables are less secure. Nevertheless, our study represents an initial attempt to understand the complicated issue of determining the age at surgical repair for children with CHD. Future studies using clinical databases that include detailed medical variables are clearly indicated.
The list of variables evaluated in this study is not exhaustive. We attempted to include as many relevant variables as possible, limited as we were to the information available in the database. In addition, attributable to the complex nature of the variables, which were obtained from a variety of sources in various formats, we were not able to conduct a multivariate analysis to evaluate the relative contribution of each variable in determining age for surgical repair. Such analysis must await the development of a large scale, prospective clinical database.
In addition, some patients may have undergone palliative surgeries before the complete repair of CHD studied here. For example, patients with TOF may or may not have undergone a previous shunt procedure, and patients with VSD or AVC may or may not have undergone a previous pulmonary artery banding. Because of the structure of the OSHPD database, we were not able to distinguish patients with from those without previous surgeries. However, palliative procedures for patients with TOF, VSD, or AVC are usually performed on an urgent or semiurgent basis, compared with repair of these defects on an elective or semielective basis. It was our intention to study factors that determine the age for elective or semielective complete surgical repair, because we hypothesized that for these procedures both many medical and nonmedical variables have important role (compared with urgent procedures, where medical variables may be much more important).
| |
CONCLUSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Many medical and nonmedical variables play important roles in determining age for definitive repair of CHD in children. Medical variables such as type of CHD, presence of Down syndrome, and the practice patterns of surgical centers were significant variables in determining age at repair of CHD. Centers with higher surgical case volume were more likely to operate at a younger age. In addition, nonmedical variables, such as type of insurance, a known surrogate for access to care, may play an important role. In California, children who live in urban areas tended to be older at surgery for ASD, VSD, and TOF. We believe that this study demonstrates that health services research of this type has the potential to enhance our understanding of important issues related to access to health care for children with complex and/or severe medical conditions. Whether delay in complete repair of CHD in some ethnic groups or patients without private insurance results in difference in clinical outcomes remains an important issue to be further investigated.
| |
FOOTNOTES |
|---|
The abstract of this article was presented at the 48th Annual Scientific Session of the American College of Cardiology; March 9, 1999; New Orleans, LA.
Received for publication Nov 6, 1998; accepted Jul 27, 1999.
Reprint requests to (R-K.R.C.) Division of Cardiology, Department of Pediatrics, Harbor-UCLA Medical Center, 1000 W Carson St, Torrance, CA 90509. E-mail: rkchang{at}ucla.edu
| |
ABBREVIATIONS |
|---|
CHD, congenital heart disease; OSHPD, Office of Statewide Health Planning and Development; ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; ASD, atrial septal defect; VSD, ventricular septal defect; TOF, tetralogy of Fallot; AVC, atrioventricular canal; CCS, California Children Services; HMO, health maintenance organization; PPO, preferred provider organization; SEM, standard error of the mean.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
This article has been cited by other articles:
 |
|
 |
|
  R.-K. R. Chang, A. Y. Chen, and T. S. Klitzner Female Sex as a Risk Factor for In-Hospital Mortality Among Children Undergoing Cardiac Surgery Circulation, September 17, 2002; 106(12): 1514 - 1522. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Jenkins, K. Gauvreau, J. W. Newburger, T. L. Spray, J. H. Moller, and L. I. Iezzoni Consensus-based method for risk adjustment for surgery for congenital heart disease J. Thorac. Cardiovasc. Surg., January 1, 2002; 123(1): 110 - 118. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Marino, S. Anaclerio, R. Di Donato;, R.-K. R. Chang, and T. S. Klitzner Age at Operation for Children With Atrioventricular Canal Pediatrics, July 1, 2001; 108(1): 217 - 217. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2000 American Academy of Pediatrics. All rights reserved. |
||
|
||
