Single-ventricle (SV) congenital heart disease (CHD) refers to severe cardiac malformations where the left or right ventricle is underdeveloped and cannot sustain normal circulation. Morbidity and mortality are substantial, but a series of palliative surgeries culminating in Fontan circulation, permitting survival and a reasonable quality of life, is typically realistic for patients in high-income countries (HIC) (1). Important numbers of children with SV CHD are born in low- and middle-income countries (LMIC), but information regarding outcomes in this setting is sparse.
Toward the elucidation of this problem, the International Quality Improvement Collaborative for Congenital Heart Disease (IQIC) collects information regarding CHD surgery in LMIC and uses quality improvement methods to improve outcomes (2). The current study identifies interventions, patient and institutional characteristics, and risk factors for inhospital mortality among SV surgeries at 32 IQIC sites in 20 countries. This study focuses on Glenn and Fontan surgeries. Initial palliations are also described.
Methods
IQIC sites in LMIC participated voluntarily and submitted deidentified data to a REDCap registry (REDCap Cloud, Encinitas, California). Sites under went audits as previously published (3). Approval to conduct research using IQIC data was obtained from the Boston Children’s Hospital Institutional Review Board. The database was searched for: 1) all instances of surgery in SV CHD; 2) Glenn; and 3) Fontan. Surgeries during the period January 2010 through December 2014 were identified and included or excluded as noted in the following text.
Inclusion Criteria
All surgeries performed in patients with SV anatomy were included. All Glenn and Fontan surgeries were included, permitting identification of patients with 2 ventricles that underwent single-ventricle palliation (SVP). Operations without mortality data were excluded.
Statistical Analysis
Glenn and Fontan surgeries were analyzed for associations with in-hospital mortality. Patient age, sex, weight, and body mass index (BMI) were used to classify subjects <18 years as <5th, ≥5th but <15th, or ≥15th percentile for weight- or BMI-for-age percentiles. Weight-for-age percentiles were calculated for subjects up to 9 years, and BMI-for-age percentiles for those 10 to 17 years of age; percentiles were combined for analysis. Generalized estimating equation models were used to evaluate associations between patient characteristics and in-hospital mortality. Risk factors significant at the 0.10 level in univariate analysis were considered for inclusion in a multivariable model; p < 0.05 was required for retention in the final model. To evaluate the impact of institutional volume, the total number of SVP performed at each institution from 2010 to 2014 was divided by the number of years for which the site contributed data to create the institutional average annual volume of SVP, which was then added to the multivariable model.
RESULTS
SVP ranged from <1% to 16% of a site’s CHD surgical volume, and 2,543 surgeries met inclusion criteria. Mortality data were missing in 26 patients, who were not included in analysis. Glenn (53%) and Fontan (27%) were most frequently performed (Table 1), but initial palliations were common (18%), including stage 1 Norwood palliation (S1P) (5%), pulmonary artery banding (6%), and systemic-to-pulmonary shunt (7%). Some surgeries (2%) were not SVP as such (e.g., isolated valve repair). Not all centers performed all surgeries; 4 (13%) reported Glenn and Fontan only. Only 11 (34%) performed S1P.
TABLE 1.
Glenn and Fontan Surgeries in Low- and Middle-Income Countries
| Glenn | Fontan | ||||
|---|---|---|---|---|---|
| Surgeries | Mortality | Surgeries | Mortality | ||
| Total | 1,358 (53) | 77 (6) | 673 (27) | 55 (8) | |
| Female | 536 (39) | 30 (6) | 275 (41) | 30 (11) | |
| Single ventricle | 779 (57) | 40 (5) | 425 (63) | 36 (8) | |
| 2 ventricles | 579 (43) | 37 (6) | 248 (37) | 19 (8) | |
| Age at surgery | |||||
| Glenn | Fontan | ||||
| <6 months | <2 yrs | 183 (13) | 16 (9) | 15 (2) | 1 (7) |
| 6–11 months | 2–3 yrs | 408 (30) | 25 (6) | 144 (21) | 9 (6) |
| 1–2 yrs | 4–5 yrs | 414 (30) | 20 (5) | 182 (27) | 16 (9) |
| $3 yrs | $6 yrs | 353 (26) | 16 (4) | 331 (49) | 29 (9) |
| WHO weight- or BMI-for-age percentile <5th | 745 (55) | 51 (7) | 226 (34) | 21 (9) | |
| Major noncardiac anomaly | 51 (4) | 6 (12) | 23 (3) | 3 (13) | |
| Major chromosomal anomaly | 21 (2) | 3 (14) | 7 (1) | 0 (0) | |
| Major medical illness | 48 (3) | 6 (12) | 28 (4) | 2 (7) | |
| Surgical site infection | 36 (3) | 3 (8) | 14 (2) | 1 (7) | |
| Bacterial sepsis | 62 (5) | 16 (26) | 30 (4) | 11 (37) | |
| Any major infection | 84 (6) | 16 (19) | 40 (6) | 11 (27) | |
| <20 center SVP average annual volume | 167 (12) | 19 (11) | 80 (12) | 8 (10) | |
| 20–59 center SVP average annual volume | 580 (43) | 43 (7) | 200 (30) | 30 (15) | |
| $60 center SVP average annual volume | 611 (45) | 15 (2) | 393 (58) | 17 (4) | |
Values are n (%).
BMI ¼ body mass index; SVP ¼ single-ventricle palliation; WHO ¼ World Health Organization.
Most surgeries (1,713, 67%) were performed in true SV CHD, most commonly tricuspid atresia (20%) and double-inlet ventricle (17%). There were 150 surgeries (6%) for hypoplastic left heart syndrome. Most Glenn and Fontan surgeries were performed in SV anatomy, but a significant minority (43% and 37%, respectively) were in 2-ventricle anatomy.
Most Glenn surgeries (767, 57%) were performed after 1 year of age, with a median age of 1 year. Nearly one-half (331, 49%) of Fontan surgeries were performed after 6 years of age, with a median age of 5 years. Initial palliations were mostly (61%) performed after 1 month of age. The exception was S1P, which was more commonly performed before 30 days of age (76%).
Malnutrition, a World Health Organization weight- or BMI-for age percentile <5th, occurred in 48% of surgeries. Only 36% of surgeries were performed in patients ≥15th percentile. Malnutrition was common among Glenn (55%) and Fontan (34%) surgeries.
Major noncardiac and chromosomal illnesses were present in only 4% and 2% of patients, respectively. Post-operative infections occurred in 9% of all surgeries: 3% surgical site and 7% sepsis with some overlap between categories. The prevalence did not vary substantially with type of surgery.
Overall survival was 89%. Survival after Glenn surgery was 94%. Multivariable analysis of Glenn surgeries identified bacterial sepsis (odds ratio [OR]: 6.9; p < 0.001) and annual SVP volume <60 surgeries (<20 surgeries OR: 4.7; p = 0.005; 20 to 59 surgeries OR: 3.3;p = 0.03) as associated with higher in-hospital mortality after Glenn. Survival after Fontan surgery was 92%. Multivariable analysis among Fontan surgeries identified male sex as associated with lower odds of mortality (OR: 0.5; p = 0.02). Bacterial sepsis (OR: 6.6; p < 0.001) and annual SVP volume of 20 to 59 surgeries (OR: 3.6; p = 0.005) were associated with a higher likelihood of mortality after Fontan. Age at Glenn or Fontan was not associated with mortality.
DISCUSSION
Publicly available data show that in LMIC, the rank of CHD deaths relative to other diseases is rising (4). Fortunately, there is increasing attention to this challenge (5,6). This study illuminates that the full range of SVP is performed in LMIC, and such interventions represent up to 16% of the volume at some sites. Encouragingly, overall survival is 89%.
Surgeries were performed at a later age than in HIC. Among initial palliations, this was likely due to a number of factors. Prenatal diagnosis and postnatal screening is unusual in LMIC (7). Many infants with CHD deteriorate at home, and recognition is delayed. Distance to care and impediments to transportation may be considerable (8), and regional hospitals may have limited ability to care for CHD. The cumulative result is malnutrition, infection, poor perfusion, and end-organ injury, increasing the risk of death even with technically successful surgery.
Glenn and Fontan surgeries were also performed at a later age than in HIC. Typically, Glenn is performed between 3 and 6 months of age in HIC, whereas only 14% underwent Glenn surgery before 6 months of age in participating centers. Similarly, recent publicly available Society for Thoracic Surgery data show that over 60% of Fontan surgeries were performed between 2 and 4 years of age, whereas 75% of Fontan surgeries were performed after 4 years of age in this study. Importantly, however, a later age at Glenn or Fontan was not associated with mortality.
The late age at surgery may be due to a number of reasons. Late diagnosis may occur, but other patients may have balanced circulation permitting expectant management. Family finances are also important when considering timing of CHD surgery. Families must pay for part or all of surgery in many LMIC, and a delay may simply reflect a family saving to tolerate the costs. It is possible that in some instances, survival through infancy may have been a prerequisite to permit adequate family preparation. Finally, limited operating room availability may cause delays, because a significant backlog of cases is an unfortunate reality in some LMIC.
Many SVP were performed where the anatomy ostensibly included 2 normally sized ventricles. Data regarding decisions to pursue a particular strategy were not obtained, and the reasons are likely complex. Technical concerns, such as challenging cardiac anatomy, may play a role. Limited expertise and imaging infrastructure, and operating rooms oriented toward adults may diminish ability to perform complex repairs.
Other factors may influence the decision to pursue a SVP strategy. Complex biventricular repairs often necessitate future surgeries, representing a significant familial, societal, and financial commitment. The limited availability and expense of conduits pose a further challenge. These limitations can make Glenn surgery the most feasible alternative, offered as a single procedure in a scenario that might have permitted anatomic repair.
Despite challenges delivering CHD care in LMIC, survival among SVP was high at 89%. Although a learning curve is a certainty among any surgical program, some observed mortality was certainly in part due to the aforementioned factors outside of the control of the surgical center. Considering that nearly 75% of patients with hypoplastic left heart syndrome in HIC have a prenatal diagnosis and plans to optimize perinatal care (9), the fact that there is not a larger disparity in mortality between HIC and LMIC speaks to the ability to resuscitate infants and deliver quality surgery.
Poor nutrition was common in this study. In many LMIC, malnutrition, independent of CHD, is a problem affecting many children, and overlap between poor nutrition and increased caloric demands from CHD likely resulted in the nutritional status described in this study. Interestingly, poor nutrition was not associated with outcomes.
As described in other studies regarding congenital heart surgery in LMIC, smaller-center surgical volume was identified as a risk factor for mortality. This association has been described among HIC as well (10). Although the association is important, it must be noted that many smaller centers have excellent results.
Study Limitations
The data audit was limited to 10% of key variables. Secondary variables, such as major medical illness, were not audited, with potential for misclassification. Primary diagnostic data were not reviewed. Therefore, verification of anatomy was not possible. Comparing various surgical strategies for specific lesions was not also possible, because data were collected by surgical episode. Similarly, only data for short-term surgical outcomes were available. Importantly, this study does not address possible survivorship bias, nor does it delineate specific assets and limitations of various sites. The IQIC comprises LMIC centers throughout the world, each with a unique patient population, range of experience, resources, and challenges. Identifying nuanced associations with outcomes was not possible.
CONCLUSIONS
SVP is commonly performed among surgical centers in LMIC with good short-term results. The preceding observations regarding pre-operative status, age at surgery, surgical strategy, and outcomes reflect a complex interplay among patient, family, and economic, institutional, and regional factors. This study broadens understanding of care for patients with complex SV CHD in LMIC at the current time, and encouragingly demonstrates the success of surgical programs despite malnutrition, late presentation, and other challenges inherent in the environment. Increasing investment and collaborative approaches such as IQIC should be encouraged to optimize outcomes for complex CHD surgery.
Acknowledgments
The authors wish to gratefully acknowledge the IQIC centers that contributed data to this study, the IQIC team members who donated their time and effort, and the entire collaborative’ s efforts toward improving outcomes for children with CHD.
Funded by Kobren Family Chair for Patient Safety and Quality, Boston Children’s Hospital. Dr. Jenkins has received research funding from NuMed. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
References
- 1.Atz AM, Zak V, Mahony L, et al. Longitudinal outcomes of patients with single ventricle after the Fontan procedure. J Am Coll Cardiol 2017;69: 2735–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jenkins KJ, Castaneda AR, Cherian KM, et al. Reducing mortality and infections after congenital heart surgery in the developing world. Pediatrics 2014;134:e1422–30. [DOI] [PubMed] [Google Scholar]
- 3.Schidlow DN, Jenkins KJ, Gauvreau K, et al. Transposition of the great arteries in the developing world. J Am Coll Cardiol 2017;69: 43–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zimmerman M, Smith A, Sable C, et al. Relative impact of congenital heart disease on morbidity and mortality in infancy around the globe: the Global Burden of Disease Study. Circulation 2017; 136:A14666. [Google Scholar]
- 5.Sciarra AMP, Croti UA, Batigalia F. Information technology implementing globalization on strategies for quality care provided to children submitted to cardiac surgery: International Quality Improvement Collaborative Program–IQIC. Rev Bras Cir Cardiovasc 2014; 29:89–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Jacobs JP, Maruszewski B, Kurosawa H, et al. Congenital heart surgery databases around the world: do we need a global database? Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2010;13:3–19. [DOI] [PubMed] [Google Scholar]
- 7.Kouame B, N0 guetta-Brou I, Kouame GY, et al. Epidemiology of congenital abnormalities in West Africa: results of a descriptive study in teaching hospitals in Abidjan: Cote d0 Ivoire. African J Paediatr Surg 2015;12:51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Karmegaraj B, Kappanayil M, Sudhakar A, Kumar RK. Transport: the crucial link in the care of neonates with CHD at an IQIC Center; 7th World Congress of Pediatric Cardiology Cardiac Surgery Abstracts. Cardiol Young 2017;27:S1–653. [Google Scholar]
- 9.Donofrio MT, Moon-Grady AJ, Hornberger LK, et al. Diagnosis and treatment of fetal cardiac disease: a scientific statement from the American Heart Association. Circulation 2014;129:2183–242. [DOI] [PubMed] [Google Scholar]
- 10.Pasquali SK, Li JS, Burstein DS, et al. Association of center volume with mortality and complications in pediatric heart surgery. Pediatrics 2012; 129:e370–6. [DOI] [PMC free article] [PubMed] [Google Scholar]