Summary
Background
Access to cardiac surgery is limited in low-income settings, and data on patient outcomes are scarce.
Aims
To assess characteristics, surgical procedures and outcomes in patients undergoing open-heart surgery in low-income settings.
Methods
This was a cohort study (2001–2011) in two low-income countries, Cambodia and Mozambique, where cardiac surgery had been promoted by visiting non-governmental organizations.
Results
In Cambodia and Mozambique, respectively, 1332 and 767 consecutive patients were included; 547 (41.16%) and 385 (50.20%) were men; median age at first surgery was 11 years (interquartile range [IQR] 4–14) and 11 years (IQR 3–18); rheumatic heart disease affected 490 (36.79%) and 268 (34.94%) patients; congenital heart disease (CHD) affected 834 (62.61%) and 390 (50.85%) patients, with increasingly more CHD patients over time ( P < 0.001); and the number of patients lost to follow-up reached 741 (55.63%) and 112 (14.6%) at 30 days. A total of 249 (32.46%) patients were lost to follow-up in Mozambique, remoteness being the only influencing factor ( P < 0.001). Among patients with known vital status, the early (< 30 days) postoperative mortality rate was 6.10% ( n = 40) in Mozambique and 3.05% ( n = 18) in Cambodia. Overall, 109 (8.18%) patients in Cambodia and 94 (12.26%) patients in Mozambique underwent re-do surgery. In Mozambique, a further 50/518 (9.65%) patients died at a median of 23 months (IQR 7–43); in Cambodia, a further 34/591 (5.75%) patients died at a median of 11.5 months (IQR 6–54.5).
Conclusions
Cardiac surgery is feasible in low-income countries with acceptable in-hospital mortality and proof of capacity building. Patient outcomes after cardiac surgery in low-income countries remain unknown, given the strikingly high numbers of lost to follow-up.
Résumé
Contexte
L’accès à la chirurgie cardiaque est limité dans nombre de pays en voie de développement et le devenir des patients opérés a été peu décrit à ce jour.
Objectifs
Étudier les caractéristiques, les procédures entreprises et le devenir des patients ayant une chirurgie cardiaque dans des pays en voie de développement.
Méthodes
Étude de cohorte (2001–2011) dans deux centres au Cambodge et au Mozambique, où des programmes de chirurgie cardiaque ont été mis en place par des organisations non gouvernementales.
Résultats
Un total de 1332 patients ont été inclus au Cambodge, et 767 au Mozambique. Parmi les patients, 547 (41,16 %) et 385 (50,20 %) étaient de sexe masculin. L’âge médian lors de la première intervention était de 11 (IQR 4–14) et 11 (IQR 3–18) ans, respectivement au Cambodge et au Mozambique. Les étiologies comportaient : cardiopathie rhumatismale dans 490 (36,79 %) et 268 (34,94 %) cas ; cardiopathie congénitale dans 834 (62,61 %) et 390 (50,85 %) cas au Cambodge et au Mozambique, avec un nombre croissant de patients avec cardiopathie congénitale au cours de l’étude ( P < 0,001). Cent douze (14,6 %) et 741 (55,63 %) patients ont été perdus de vue à 30 jours au Mozambique et au Cambodge. Au Mozambique, la distance du domicile au centre chirurgical était le seul facteur associé avec le statut perdu de vue ( P < 0,001). La mortalité postopératoire précoce (< 30 jours) était de 40 (6,10 %) au Mozambique, et de 18 (3,05 %) au Cambodge (en tenant compte des patients au statut vital renseigné). Le nombre de ceux nécessitant une chirurgie redux était de 109 (8,18 %) au Cambodge et de 94 (12,26 %) patients au Mozambique. Au Mozambique, 50 patients parmi 518 (9,65 %) sont décédés secondairement à une médiane de 23 mois (IQR 7–43). Au Cambodge, la mortalité tardive était de 34/591 (5,75 %) à une médiane de 11,5 mois (IQR 6–54,5).
Conclusions
Le devenir des patients opérés du cœur demeure inconnu dans certains pays en voie de développement étant donné le nombre de patients perdus de vue.
Background
Access to cardiac surgery remains limited in many developing countries, especially in low-income settings . Therefore, the natural history of severe rheumatic heart disease (RHD) and congenital heart disease (CHD) in need of cardiac interventions still remains of importance in countries with no access to cardiac surgery . Timely interventions are needed to improve outcomes in these young patients, and the majority of the world’s population aged < 15 years live in low- and middle-income countries with poor access to cardiac interventions .
Humanitarian efforts have led non-governmental organizations (NGOs) to launch surgical programmes in low- and middle-income countries, in an attempt to fill the gap in these fragile healthcare systems . The methods of these NGOs vary, with some providing overseas treatment, while others carry out fly-in fly-out missions with varying levels of capacity building . The data on the outcomes of these initiatives are scarce; publications focus either on the methods of the NGO missions or on patient short-term outcomes from a single centre .
We aimed to assess two cardiac surgery programmes conducted over 10 consecutive years in two low-income countries, by describing patient characteristics, surgical procedures, temporal trends and patient outcomes.
Methods
Objectives
The main objective was to describe patient characteristics, interventions and early postoperative mortality in two low-income countries; we also explored factors associated with loss to follow-up in these settings.
Settings
Cambodia and Mozambique are two low-income countries with a gross national income per capita in 2014 of 1010 USD and 630 USD, respectively. Cambodia is an East Asian country of 15.41 million inhabitants, with a life expectancy at birth of 72 years. Mozambique is a sub-Saharan African country of 26.47 million inhabitants, with a life expectancy at birth of 50 years .
Free-of-charge cardiac interventions are not available in the healthcare system in these two countries. Several paediatric cardiac surgery NGOs – including Chaîne de l’Espoir, France, which co-promoted this work – have built a partnership with two private hospitals: the Centre de Cardiologie de Phnom Penh (CCCPP) in Phnom Penh, Cambodia, and the Maputo Heart Institute (Instituto do Coração [ICOR]) in Maputo, Mozambique. These NGOs include senior cardiac surgeons with expertise in valve repair in RHD, who had been working in Northern and Southern countries for over 20 years before the study period . These were the two only centres to provide cardiac surgery in the corresponding countries during the study period.
Inclusion criteria and data collection
All consecutive patients undergoing open-heart surgery at ICOR and CCCPP between 1 January 2001 and 31 December 2011 were included retrospectively. Patients who had been operated on overseas, those who underwent percutaneous interventions during the study period, and those with non-cardiac (i.e. thoracic) surgery were not included.
Data collected included demographic characteristics, underlying aetiology, surgical procedure, distance between the place of residence and the surgical centre, and early (i.e. < 30 days) postoperative mortality. In addition, data on the use of secondary prophylaxis were collected in patients with RHD at follow-up. In Mozambique, the Aristotle basic complexity score was computed for patients with CHD . We calculated expected early postoperative mortality by using the Aristotle basic complexity score with regard to the underlying CHD and the procedure performed.
In Cambodia, passive surveillance of surgical patients was part of the standard of care, with new visits recorded on the database. The definition of ‘lost to follow-up’ in Cambodia was the absence of a second visit at the institution (CCCPP).
In Mozambique, cross-sectional active follow-up was attempted in 2012, targeting all patients who had not attended the clinic for over 2 years. Patients and their parents were contacted by telephone on three occasions.
The methods comply with the STROBE statement . The local ethics committees of ICOR and CCCPP approved the study and waived written consent from the patient and parent/guardian. This study complies with the Code of Ethics of the World Medical Association (Declaration of Helsinki).
Statistical analysis
All participants’ characteristics are described as medians and interquartile ranges (IQRs) or proportions, as appropriate. Categorical variables were compared using the χ 2 test or Fisher’s exact test, and continuous variables were compared using Student’s t test or the Wilcoxon rank sum test, as appropriate. A two-sided P value < 0.05 was considered to indicate statistical significance. All data were analysed at the Paris Cardiovascular Research Centre, INSERM 970, Paris, France, with the use of Statistical Analysis System software, version 9.4 (SAS Institute, Cary, NC, USA).
Results
Patient characteristics
The study population comprised 1332 patients included in Cambodia plus 767 included in Mozambique ( Fig. 1 ). In Cambodia and Mozambique, respectively, median age at first surgery was 11 years (IQR 4–14) and 11 years (IQR 3–18), and 547 (41.16%) and 385 (50.20%) patients were male. The two main underlying aetiologies were: CHD in 834 (62.61%) and RHD in 490 (36.79%) patients in Cambodia; and CHD in 390 (50.85%) and RHD in 268 (34.94%) patients in Mozambique. Endomyocardial fibrosis was present in Mozambique only, where it affected 49 (6.39%) patients. Patient characteristics were different between the two countries, with a male predominance in Mozambique and differences in the underlying heart diseases ( Table 1 ). There were increasingly more CHD and fewer RHD patients over the study period in both centres ( P < 0.001) ( Fig. 2 ). Patients with CHD were younger than those with RHD in both countries: median age 3 years (IQR 1–8) in CHD patients and 16 years (IQR 12–23) in RHD patients ( P < 0.001) in Mozambique; median age 6 years (IQR 3–11) in CHD patients and 14 years (IQR 12–17) in RHD patients ( P < 0.001) in Cambodia.
| Characteristics | Mozambique ( n = 767) | Cambodia ( n = 1332) | P |
|---|---|---|---|
| Age (years) | 11 (3–18) | 11 (4–14) | 0.20 |
| Male a | 385 (50.20) | 547 (41.16) | < 0.0001 |
| Rheumatic heart disease b | 268 b (34.94) | 490 (36.79) | 0.4 |
| Mitral valve disease | 259 (96.64) | 475 (96.94) | |
| Aortic valve disease | 90 (33.58) | 145 (29.59) | |
| Tricuspid regurgitation | 67 (25.00) | 53 (10.82) | |
| Congenital heart disease c | 390 c (50.85) | 834 (62.61) | < 0.0001 |
| Fallot’s tetralogy | 111 (28.46) | 221 (26.50) | |
| Ventricular septal defect | 114 (29.23) | 197 (23.62) | |
| Persistent ductus arteriosus | 48 (12.31) | 285 (34.17) | |
| Atrial septal defect | 55 (14.10) | 145 (17.39) | |
| Valve disease | 64 (16.41) | 128 (15.35) | |
| Other | 105 (26.92) | 17 (2.04) | |
| Other aetiologies | 109 (14.21) | 8 (0.60) | < 0.0001 |
| Endomyocardial fibrosis | 49 (6.39) | 0 | |
| Infective endocarditis d | 9 (1.17) | 0 | |
| Ischaemic heart disease | 26 (3.39) | 0 | |
| Other | 25 (3.26) | 8 (0.60) |
a Missing data in three patients in the Cambodian cohort.
b One hundred and thirty patients in Cambodia and 81 in Mozambique had a combination of mitral and aortic valve disease.
c Ninety two patients in Mozambique and 150 in Cambodia had combined congenital heart disease.
d Without congenital heart disease or rheumatic heart disease.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
