Abstract
Background
Mortality associated with the correction of congenital heart disease has decreased to approximately 2% in developed countries and major adverse events are uncommon. Outcomes in developing countries are less well defined. The World Database for Pediatric and Congenital Heart Surgery was utilized to compare mortality and adverse events in developed and developing countries.
Methods
A total of 16,040 primary procedures were identified over a two-year period. Centers that submitted procedures were dichotomized to low/middle income (LMI) and high income (HI) by the Gross National Income per capita categorization. Mortality was defined as any death following the primary procedure to discharge or 90 days inpatient. Multiple logistic regression models were utilized to identify independent predictors of mortality.
Results
Of the total number of procedures analyzed, 83% (n = 13,294) were from LMI centers. Among all centers, the mean age at operation was 2.2 years, with 36% (n = 5,743) less than six months; 85% (n = 11,307) of procedures were STAT I/II for LMI centers compared with 77% (n = 2127) for HI centers (P < .0001). Overall mortality across the cohort was 2.27%. There was a statistical difference in mortality between HI centers (0.55%) versus LMI centers (2.64%) (P < .0001). After adjustment for other risk factors, the risk of death remained significantly higher in LMI centers (odds ratio: 2.36, 95% confidence interval: 1.707-3.27).
Conclusion
Although surgical expertise has increased across the globe, there remains a disparity with some outcomes associated with the correction of congenital heart disease between developing and developed countries. Further studies are needed to identify specific opportunities for improvement.
Introduction
According to the World Health Organization, noncommunicable diseases account for 17 million premature deaths globally, with 82% occurring in low- and middle-income countries (LMI) and 37% due to cardiovascular disease. 1 The impact of congenital heart disease (CHD) on these grim statistics is less well defined. Certain developing countries have seen a dramatic increase in not only the overall birth rate but also increases in the number of children born with CHD.2-4 Historically, the burden of support for CHD has fallen to nongovernmental agencies, with governmental support being directed to nonhealthcare sectors.5,6 Accurate and real-time data are critical to bringing these inadequacies to the forefront of governmental agencies, which are in the best position to provide the necessary support to improve outcomes related to CHD.
The World Database for Pediatric and Congenital Heart Surgery (WDPCHS) began accepting and tracking heart operations on January 1, 2017. In doing so, it became one of the first global platforms for the exchange of experience and knowledge for children with CHD. The WDPCHS was utilized to analyze and compare the overall hospital mortality for congenital heart operations between LMI and high-income (HI) countries. We aimed to assess differential hospital mortality between the HI countries and the LMI countries for infants and children undergoing surgery for a congenital heart defect for an overall group of index congenital heart procedures, both as a composite cohort and for specific cardiac operations. In addition, we attempted to investigate the incidence of an aggregate class of postoperative complications, evaluating for possible differences between LMI and HI countries. We hypothesized that the LMI group would have a higher incidence of preoperative risk factors (PRF), postoperative complications, and worse overall survival when compared with the HI group.
Material and Methods
The WDPCHS was utilized to identify 16,040 primary procedures submitted over a two-year period (2017-2019). Primary operations included in the analysis are listed in Table 1. Centers that submitted procedures were dichotomized into LMI and HI by the Gross National Income (GNI) per Capita categorization. The World Bank classifies low-income economies as those with a GNI per capita, calculated using the World Bank Atlas method, of $1085 or less in 2021; LMI economies are those with a GNI per capita between $1086 and $4255; upper middle-income economies are those with a GNI per capita between $4256 and $13,205; HI economies are those with a GNI per capita of $13,205 or more (datahelpdesk.worldbank.org). Information collected on specific procedures included demographic, PRF, complications, and hospital mortality via prepopulated data entry forms. Terms and definitions for a specific congenital heart defect were defined by the International Pediatric and Congenital Cardiac Code (ipccc.net). Procedures, PRF, and syndrome are defined in appendices found on the WDPCHS website (www.uab.edu/medicine/wdps/). PRF was categorized into severe, moderate, or no risk groups (Table 2). Defined morbidities were grouped into none or at least one. Mortality was defined as death following the primary operation during the same hospitalization or within 90 days of operation if still an inpatient. Any variable with greater than 15% missingness was excluded. Individuals with ages <216 months (<18 years) were included in the analysis. Individual patient consent was waived if approved by individual centers. All data were stored on a secure server within the University of Alabama (UAB) maintained by the Kirklin Institute for Research in Surgical Outcomes.
List and Percentage of Cases.
Abbreviations: ASO, arterial switch operation; AVSD, atrioventricular septal defect; CAVSD, complete atrioventricular septal defect; PAPVC, partial anomalous pulmonary venous connection; TAPVC, total anomalous pulmonary venous connection; TOF, tetralogy of Fallot; VSD, ventricular septal defect.
Preoperative Risk Factors.
Abbreviations: AV, atrioventricular; ECMO, extracorporeal membrance oxygenation; IABP, intra-aortic balloon pump; PRF, preoperative risk factor; VAD, ventricular assist device.
All statistical analyses were performed using SAS 9.4. The statistical significance was assessed using an alpha level of 0.05. Continuous data were presented as mean (±standard deviation) or median (interquartile range) where appropriate. Categorical data were presented as frequency (percent). Comparisons were evaluated using Student t test for continuous variables or Wilcoxon rank sum. Chi-square or Fisher exact tests were used to compare categorical variables. Additionally, Fisher exact tests were calculated to determine whether the proportion of patients dying differed by income group for the 14 operative groups. Risk factors for hospital mortality were examined using multiple logistic regression, incorporating covariates describing type of surgery, income group, PRF, STAT Mortality risk category, and demographics such as age, gender, and race.
Results
Patient Characteristics
The study cohort consisted of 16,040 individual procedures. The procedures included are listed in Table 1. The most common operations were ventricular septal defect (VSD) repair and tetralogy of Fallot repair at 45.3% and 17.5%, respectively. The smallest percentage of cases included Ebstein repair and unidirectional Glenn/Hemi-Fontan at 0.47% and 0.19%, respectively. Table 3 provides descriptive statistics of the demographics. Briefly, 57% of patients were male, 55% were Asian, and 64% were greater than six months old at the time of surgery (Table 3).
Patient Characteristics.
Based on the World Bank Classification for income categories, the cohort was divided into HI countries (17%; n = 2,746) and LMI countries (83%; n = 13,294). Table 4A and 4B breaks the entire cohort down by both STAT Mortality risk category and RACHS Mortality risk group. There was a statistically significant difference across STAT categories for the HI and LMI groups (P < .0001; Table 4A). A majority of the cases in the LMI group were in the STAT 1 category (64.13% vs 45.93% in the HI group; Table 4A). There was a higher percentage of STAT 4 cases in the LMI group versus the HI group (8.09% vs 2.88%; Table 4A). STAT 5 cases in the HI cohort had a larger percentage of cases versus the LMI cohort (5.07% vs 0.37%, respectively; Table 4A). Similar findings were seen when using the RACHS categories (P < .0001; Table 4B). A vast majority of the LMI cases were in the RACHS 1 + 2 (86.26% vs 66.93%). There was a higher percentage of RACHS 6 cases in the HI group when compared to the LMI group (5.07 vs 0.37%, respectively; Table 4B).
Distribution of STAT Categories Among Income Groups.*
Abbreviation: STAT, The Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery.
*Patients with STAT group unknown or uncategorized were excluded from this analysis.
Distribution of RACHS Categories Among Income Groups.*
Abbreviation: RACHS, risk adjustment for congenital heart surgery.
*Patients with RACHS group unknown or uncategorized were excluded from this analysis.
Preoperative risk factors were grouped into three categories based on general clinical contribution to risk (Table 2); low or no preoperative risk (78.4%; n = 12,579); moderate risk (17.3%; n = 2,776); and high risk (4.27%; n = 685). The high-risk preoperative group included: cardiopulmonary resuscitation (0.7%), mechanical circulatory support (0.12%), renal failure requiring dialysis (0.07%), and respiratory failure requiring mechanical ventilation (3.86%) (Table 2). The moderate risk group included diagnoses such as failure to thrive (12.11%), malnutrition (2.76%), and respiratory failure not requiring ventilation (1.29%) (Table 2). Patients undergoing operative repair in LMI countries had a statistically significantly greater number of moderate PRF than those from HI countries (18.96% vs 9.32%, P < .0001), while this association was opposite for the severe PRF (3.13% vs 9.8%, P < .0001) (Table 5).
Preoperative Risk Factor Category for the High and Low/Middle-Income Cohorts.
Abbreviations: PRFCAT, preoperative risk factor category; HI, high income; LMI, low/middle income.
Mortality
Overall mortality for the entire cohort was 2.3% (Table 3). There was a significantly higher incidence of mortality when comparing the LMI to the HI cohort across some of the major surgeries including: Tetralogy of Fallot repair (2.25% vs 0.28%; P = .0075), coarctation repair (3.4% vs 0%; P = .001), Fontan (6.26% vs 0.35%; P < .0001), ASO (9.58% vs 1.15%; P < .0001), and Norwood operation (49% vs 1.4%; P < .0001) (Table 6). Results of Fisher exact test concluded that a greater proportion of individuals died in the LMI group when compared to the HI group (2.64% vs 0.55%, P < .0001). There was statistically significantly greater mortality in LMI for all PRF categories (Table 7). When comparing STAT categories, we found there was statistically significantly greater mortality in the LMI cohort compared to the HI cohort for STAT categories 1, 2, 3, and 5 (Table 8).
Mortality stratified by procedure and economic category.
Abbreviations: ASO, arterial switch operation; AVSD, atrioventricular septal defect; HI, high income; LMI, low/middle income; PAPVC, partial anomalous pulmonary venous connection; TAPVC, total anomalous pulmonary venous connection; TOF, tetralogy of Fallot; VSD, ventricular septal defect.
Mortality Stratified by PRF Category and Economic Category.
Abbreviation: PRFCAT, preoperative risk factor category.
Mortality Stratified by STAT Category and Economic Category.
Abbreviation: STAT, The Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery.
Table 9 gives the overall results of the multivariable logistic regression analysis. Income level of the country, STAT Group, age, race, and PRFCAT were highly significant predictors of hospital mortality. Since the overall hospital mortality is very small (2.27%), the relative risk of death may be approximated by the odds ratio. Mortality from surgery is 2.36 times more likely in a middle-income country than in a HI country (Table 9). Similarly, as evident from Table 9, a child with severe PRF is about three times more likely to die from surgery compared with a child with no such risk factors.
Risk Factors for Hospital Mortality (Multivariable Logistic Regression).
Abbreviations: PRFCAT, preoperative risk factor category; STAT, The Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery; OR, odds ratio.
Postoperative Complications
There was a higher incidence of postoperative complications in the HI group (Table 10). Patients undergoing surgery in the HI group had a higher incidence of arrhythmia requiring drug therapy, bleeding requiring reoperation, severe cardiac dysfunction, mechanical circulatory support, phrenic nerve injury, reintubation, seizure, unplanned open chest, unplanned reoperation, vocal cord dysfunction, mediastinitis, superficial wound infection, pulmonary hypertension, and cardiac arrest (P < .0001) (Table 10).
Postoperative Complications.*
Abbreviations: OR, operating room; IABP, intra-aortic balloon pump; VAD, ventricular assist device; ECMO, extracorporeal membranous oxygenation; PA, pulmonary artery.
*Patients were excluded when postoperative complications were missing or unknown.
Comment
This large retrospective review using the WDPCHS showed that outcomes continue to improve at the international level. The overall mortality found in the cohort of 2.3% is not significantly different from other database studies in the developed world. 7 This positive trend demonstrates that continued humanitarian investment in developing autonomous surgical programs is well worth the endeavor as it is not always feasible to establish programs where patients can be flown internationally to centers with long-standing, excellent surgical programs. 8 Although the results of this study are overwhelmingly positive, there were several key areas that were identified that require further investigation.
There was a significant association between mortality and income group following the repair of a congenital heart defect for a group of procedures utilized to definitively repair or palliate a congenital heart defect. Children that are repaired in HI countries are less likely to die following surgery than those in LMI countries. It is noteworthy that in some operations, in particular repair of VSD, AVSD, and PAPVC, the hospital mortality is low in LMI countries and not significantly different than in HI countries. However, as the underlying malformations and their repair or palliation become more complex, the survival difference between LMI and HI countries becomes more pronounced (TAPVC, Fontan procedure, ASO, Norwood operation, severe risk preoperative risk category, and STAT 3 or higher category). These findings underscore the challenges of these patients in LMI countries; as these patients frequently present late, have other comorbidities, are in a poor nutritional state and demonstrate failure to thrive.9,10 This poses significant challenges to the heart teams in LMI settings at the outset of care for more complex and higher risk operations that benefit from a team of experienced providers with sufficient resources to deliver effective care. However, with appropriate training and resource commitments, excellent outcomes can be achieved. 11
These findings emphasize the importance of the WDPCHS to generate rigorous analyses based on the global experience of congenital heart surgery for the benefit of developing countries, as a means of benchmarking their outcomes against those of experienced centers in more resource-rich countries.12,13 The finding of a marked increase in hospital mortality in particularly high-risk categories compared to HI countries is not surprising given the limitations of resources and training. A natural consequence of this observation is the quandary of whether an initial lower risk palliative operation might mitigate some of the risks. Unfortunately, this is often not a realistic programmatic option given that in many instances people simply cannot return to the hospital for multiple procedures due to socioeconomic and geographical constraints. 12
There was a significant association between the existence of severe and moderate PRF prior to the operative intervention for a congenital heart defect. The finding that centers from LMI countries in general had a somewhat higher proportion of patients with moderate or worse risk factors (22% vs 19%, Table 5) underscores the important challenges of a somewhat sicker group of patients at baseline compared to HI settings. Additionally, it suggests that the truly high-risk patients in the developing world are likely not being offered repair at the same rate as the HI cohort. This supports recognition by the care teams in these developing countries that they do not have the infrastructure yet in place to safely manage some of the highest risk patients.14-17
The higher incidence of most postoperative complications in patients from HI countries (Table 7) can partially be explained by the higher percentage of patients in the HI cohort within the severe risk preoperative category. It also highlights the ongoing improvement in the delivery of care in countries with limited resources.
Limitations
Our study has inherent limitations related to the WDPCHS structure and available data. Detailed identification and analysis of the preoperative anatomic and physiologic characteristics was not possible, impeding understanding of the rationale behind specific surgical management plans. The WDPCHS does not provide longitudinal follow-up across multiple encounters, limiting insights into possible staged approaches or the risk for reintervention after different operative approaches. The data are currently not independently validated and the analyses rely heavily on the individual center for completeness and accuracy. We cannot establish how representative these two groups are of all LMI and HI countries. Finally, results are limited to short-term, in-hospital outcomes and 90-day mortality. Despite these limitations, our study provides an important update on the spectrum of operations and surgical outcomes in a large contemporary international cohort. The database will continue to evolve and provide centers with data for benchmarking and quality improvement. Ultimately, leading to a platform for knowledge and expertise exchange to improve center and national outcomes.
Conclusion
Although surgical expertise has increased across the globe, there remains a disparity for survival outcomes between HI and LMI countries associated with the correction and palliation of some but not all forms of CHD. Developing countries often face challenges related to resources and other aspects of care. Further studies are needed to identify specific opportunities for narrowing this gap.
Footnotes
Abbreviations
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Correction (July 2023):
The middle initial in Dr Jeffery P. Jacobs’s name was missed to be included. It has been included now.