Clinical outcomes of children operated for congenital heart diseases in Rwanda: a 14-year retrospective analysis

Abstract

Background

Congenital heart disease surgery and catheterization for Rwandan children has historically been accomplished with the assistance of visiting external teams and referrals abroad. However, follow-up information on patient-specific clinical outcomes is lacking.

Methods

This is a retrospective cross-sectional study of clinical outcomes among Rwandan children under 15 years of age at the time of heart surgery operated between 2006 and 2020. Data from two Rwandan teaching hospitals were collected from paper and digital patient charts.

Results

Among the 406 children included in our study, 59.1 % were female; 30.5 % and 28.1 % of the children were underweight and stunted, respectively. Tetralogy of Fallot was the most common cyanotic CHD. Of procedures, VSD repair was the most performed operation and PDA closure represented the majority of therapeutic catheterizations. Postoperatively complications such as pneumonia occurred in 6 %, residual shunt in 17 % and residual mild or greater stenosis in 41 %. The overall mortality was 3.7 %, with sepsis as the leading cause of death. Intensive care unit stay longer than seven days, symptomatic heart failure before surgery, and being underweight were significantly associated with complications. Cardiopulmonary bypass over 60 min was associated with mortality among operated children.

Conclusion

Procedural treatment of CHD in Rwandan children was predominantly accomplished by visiting missions during the study period. Of patients with sufficient clinical follow up data, survival was 96 % and complications occurred in 35 %. Nutritional compromise is a correlate of poor outcome. A centralized cardiac registry might mitigate the observed loss of clinical follow up data in 14 % of children.

1. Introduction

Congenital heart disease (CHD) encompasses a diverse group of structural heart anomalies that are among the main causes of disability and mortality in children in developing countries. [1,2]. In developed countries with advanced cardiac facilities, CHD is almost universally diagnosed at an early age with high quality perinatal ultrasonography contributing to accurate diagnosis and prompt management, including appropriately-timed catheter and surgical procedures. The World Health Organization (WHO) reports 1 % of live births globally have CHD resulting in around 1.5 million cases of CHD per year. It is estimated that 500,000 children with CHD are born in Africa each year, the majority in sub-Saharan Africa and have relatively high morbidity and mortality [3,4].

Interventions for CHD are widely performed with excellent outcomes, including balloon atrial septostomy (BAS) for transposition of great arteries (TGA), balloon valvuloplasty for valvular stenosis and device occlusion of patent ductus arteriosus and atrial septal defect (ASD), and may even be performed without a dedicated catheterization laboratory [5]. Staged and complete repairs are well established for repair of cyanotic and acyanotic defects. Early interventions in children with CHD allow the majority to live a normal and productive life if they have access to timely treatment [6]. Conversely, 90 % of those born in resource-limited settings receive late and inadequate medical care [7]. Another challenge for children undergoing cardiac surgery in developing countries such as those in sub-Saharan Africa is lack of adequate post-operative follow-up to assure their clinical outcome [12].

Since 2006, Rwanda has hosted international visiting teams to offer cardiac surgery to its children and young adults. During these short term missions many children with CHD have undergone catheterization or surgical procedures. Others have been referred abroad for interventions either by the governmental or philanthropic support. There is an incomplete knowledge of their outcomes in general and of most current patient health status in particular. This study therefore aims to determine the clinical outcomes of all Rwandan children managed under this model.

2. Methods

2.1. Study design, site, population, and variables

This was a retrospective cross-sectional study of clinical outcomes of Rwandan children operated for CHD in Rwanda or abroad using the available data from local registries and medical charts. Of the total, 90 % had their procedures in Rwanda and remaining 10 % were referred abroad for cardiac surgery.

Two pediatric tertiary centers which provide pediatric cardiology services in Rwanda were considered: University Teaching Hospital of Kigali (CHUK) and King Faisal Hospital, Kigali (KFH).

The original patient group included all Rwandan children operated for CHD between 2006 and 2020 irrespective of their location of surgery who followed up with a cardiologist appointment at a tertiary center in Rwanda.

Operated children in whom the diagnosis was not clearly documented, surgical procedures could not be determined or for whom no follow up information was available were not included in the study.

Data on demographic characteristics, nutrition status, diagnosis, date of surgery, hospital stay, complications, and most recent clinical status were retrospectively recorded from medical charts. Weight and height were plotted against WHO normal values. Underweight and stunting were attributed according to standard WHO definitions. Last clinical status was sometimes aggregated from data in multiple fragmentary clinical reports The intended schedule of postoperative follow up visits was at one week, one month, three months, six months then yearly, However due to limitations of transportation and financial issues not all patients were able to comply. Any identified residual shunt, small or large was included in the total and did not require that it be physiologically significant. We could not standardize this assessment retrospectively. A pre-coded data collection tool was devised to record demographic characteristics and clinical parameters of participants. A password-protected database of study variables was accessible only to the study team.

2.2. Data analysis

A descriptive analysis was applied to summarize quantitative variables as mean, median, and standard deviation. Cross-tabulations were used to examine the relationship between different variables in this study. The chi-square (X²) test of independence was applied to determine whether there is an association between categorical variables. The significance of association was defined as a p-value of <0.05 and a confidence interval of 95 %. Multivariate logistic regression was applied to analyze the effect of several risk factors on dependent variables with a significance level of p = 0.05 and a 95 % confidence interval. variables for the multivariate analysis were selected based on univariate p < 0.05. Data were entered using Microsoft Excel and analysis was performed using SPSS 25.0.

2.3. Ethics

Ethical approval was obtained from the Institutional Review Board (IRB) at Kigali University Teaching Hospital (CHUK), King Faisal Hospital, and the College of Medicine and Health Sciences (Approval No. 135/CMHS IRB/2021) Patient consent was waived due to the retrospective nature of the study.

3. Results

A total of 490 children with CHD who underwent interventional treatment were identified. We excluded 17 participants aged above 15 years at the time of operation, the conventional pediatric age cut-point in Rwanda, The patient inclusion flow chart is shown in Fig. 1.

Fig. 1.

Flow diagram of participant enrollment. CHD: congenital heart disease.

3.1. Demographics and clinical characteristics

Four-hundred-and-six children with CHD who underwent surgical and catheter interventions were included, 239 of them female (59 %). While 30 % of the children were underweight at the time of intervention, children with stunted growth represented 28 %. A small proportion (3.4 %) of operated children had genetic abnormalities mostly represented by trisomy 21. See Table 1 for complete details.

Table 1.

Sociodemographic and clinical characteristics of the included patients. Note the prevalence of suspected malnutrition in the patient group. Abbreviations: CBHI: Community Based Health Insurance, (the widely used national insurance scheme also known as Mutuelles de Santé), RSSB: Rwanda Social Security Board, (an insurance held by government workers), MMI: Military Medical Insurance.

Variables	N	%
Age at time of intervention
0–5 years	241	64
5–10 years	87	23
10–15 years	50	13
Gender
Male	166	41
Female	240	59
Province of origin
Kigali	163	42
East	81	21
North	67	17
West	54	13
South	27	7
Health Insurance
CBHI	258	70
RSSB	38	10
MMI	11	3
Private insurance	47	13
No insurance documented	13	4
Anthropometrics
Underweight	124	30
Stunting	114	28
Symptomatic	168	41
Pulmonary hypertension	67	16
Genetic anomalies (n=14)
Trisomy 21	11	3
Noonan Syndrome	1	<1
Di George syndrome	1	<1
Non identified	1	<1

3.2. Frequency of diagnosis among selected children

Among all operated children, 83 % were diagnosed with acyanotic CHD (predominantly PDA and VSD as the most common diagnoses with 32 % and VSD 17 % respectively). Tetralogy of Fallot was the predominant cyanotic CHD (13 %). See Table 2 for lesion details.

Table 2.

Congenital heart diseases among treated children (combined catheterization and surgery) and their prevalence in the patient group.

Diagnosis	N	%
Acyanotic CHD (n=426)
Patent ductus arteriosus	165	39 %
Ventricular septal defect	88	21 %
Atrial septal defect	53	12 %
Pulmonary valve stenosis	43	10 %
Coarctation of aorta	16	4 %
Atrioventricular septal defect	16	4 %
Sub-valvular aortic stenosis	15	4 %
Mitral valve anomalies	14	3 %
Aortic valve stenosis	6	1 %
Other:	10	2 %
Cyanotic CHD (n=82)
Tetralogy of Fallot	69	84 %
Transposition of great arteries	3	4 %
Pulmonary atresia	3	4 %
Other	7	9 %

3.3. Surgical interventions among operated children

Of the 256 open heart surgeries that were performed, the majority were VSD closure followed by PDA closure, complete TOF repair, and atrial septal defect (ASD) closure as detailed in Table 3.

Table 3.

Surgical interventions performed. Abbreviations: VSD = ventricular septal defect; PDA = patent ductus arteriosus; TOF = tetralogy of Fallot.

Operation	Open heart surgery (n = 256)	Closed heart surgery (n = 64)	n	%
VSD repair	88	–	88	28
PDA closure	24	37	61	19
Complete TOF repair	53	–	53	17
ASD closure	37	–	37	12
AVSD repair	16	–	16	5
Palliative BTT shunt	–	16	16	5
Resection of subaortic membrane	9	–	9	3
Coarctation of aorta repair	–	9	9	3
Others:	29	2	31	10

3.4. Catheterization

Among 154 interventional catheterisations performed, 104 (68 %) were PDA occlusion, 33(21 %) balloon valvuloplasty,10 (6 %) device ASD closure, and 7 (5 %) were primary coarctation angioplasty.

3.5. Operating teams

COH Belgium and OHI both operate on Rwandan children through domestic missions (62 % and 19 %, respectively). The remaining interventions were performed abroad with the support of various organizations, such as Mécénat Chirurgie Cardiaque (France) and Save a Child's Heart (Israel).

3.6. Complications

The post-procedural complications are detailed in Table 4. Pneumonia was most common after surgery, occurring in 16 children. Interventional catheterization procedures resulted in fewer complications compared to surgical interventions.

Table 4.

Postoperative complications among operated children.

Immediate complications	Frequency	%
Related to surgery
Pneumonia	16	6
Death	14	3.7
Sepsis	9	4
Lung atelectasis	5	2
Massive bleeding	5	2
Pneumothorax	4	2
Hypovolemic shock	3	1
Septic shock	3	1
Systemic hypertension	3	1
Others	16	6
Related to Catheterization
Pneumonia	1	<1
Sepsis	1	<1
Pulmonary edema	1	<1
Right femoral vein thrombosis	1	<1
Congestive heart failure	1	<1
Retention of catheter fragment	1	<1
Death	1	<1

3.7. Mortality

The overall mortality rate was 3.7 % (15/406), with 10 (2.5 %) deaths occurring within 30 days post-operation and 5 (1.2 %) after 30 days. See details in Table 5.

Table 5.

Procedural mortality for operated children.

Variables	n	%
Death (n = 15)
Early (within 30 days)	10	2.5
Late (after 30 days)	5	1.2
Causes of death
Septic shock	5	33
Congestive heart failure	2	13
Low cardiac output	2	13
Hemorrhage	2	13
ARDS	2	13

3.8. Other procedural results

Intraoperatively, median (range) CPB time and aortic clamp time were 65 (19–278) and 40 (8–156) minutes respectively and ICU stay was 2(1–75) days. Hospital length of stay was 5 (1–75) days for surgical patients and 1(1–16) days for catheterization patients.

Sixty-one patients (17 %) had residual shunts, 48 (41 %) had mild or greater residual stenosis, and 4 (2.6 %)had embolization of catheter-delivered devices. Eighteen patients (4 %) had ventricular dysfunction after intervention and one patient had patch dehiscence. The median length of stay on readmission was seven days ranging from one to 49 days. Among 16 patients deemed to require reoperation, six actually received them (see Table 6)

Table 6.

Postoperative outcomes.

Variables	n	%
Readmission	33	8
Hospital LOS on readmission [median (range)]	7 (1–49) days
Need of reoperation	16	4
Postoperative echocardiography findings
Residual shunt	61	17
Residual stenosis	48	41
Ventricular dysfunction	18	4
Reoperations	6	38
Interventional PDA closure	2	33
Complete ToF repair	1	17
Revision of VSD patch dehiscence	1	17
Resection of fibromuscular subaortic tissue	1	17
Surgical removal of retained catheter fragment	1	17

3.9. Correlates of postoperative complications

Children with symptomatic heart failure before the intervention were 2.2 times more likely to have complications compared to those who were not in heart failure (OR = 2.2 95 %CI: 1.3–3.7; p < 0.01). Underweight children were 2.8 times more likely to have complications than to those of normal weight (OR = 2.8 95 % CI: 1.7–4.7; p < 0.01) and stunted children were significantly more likely to have complications than those without stunting (OR = 1.8 95 %CI: 1.1–3.1; p = 0.02). Among children with tetralogy of Fallot, those with hypercyanotic spells before the intervention were 2.5 times more likely to have complications than to those who did not (OR = 2.5 95 %CI: 1.3–4.8; p < 0.01). Patients who were on cardiopulmonary bypass over 60 min were 4.5 times more likely to have complications than those with shorter times (OR = 4.56; 95 %CI: 2.53–8.21; p < 0.002). The eighteen patients who were hospitalized in ICU for seven or more days had significantly more complications than those who spent fewer than seven days in the ICU, among whom 19.5 % had complications (p < 0.01). There was no significant association between age at time of surgery, pulmonary hypertension before surgery, trisomy 21, and complications ( see Table 7).

Table 7.

Factors associated with postoperative complications among operated children.

Predictors	Complication		OR (95 % CI)	p-value
	Yes	No
Age at the time of operation (y)
0–5	53 (24 %)	169 (76 %)
5–10	22 (29 %)	53 (71 %)	1.3 (0.7–2.4)	0.35
10–15	9 (20 %)	37 (80 %)	0.8 (0.4–1.7)	0.53
ICU stay >7 days
Yes	18 (100 %)	0 (0 %)		<0.01
No	66 (20 %)	273 (81 %)
Symptomatic Heart failure
Yes	36 (34 %)	69 (66 %)	2.2 (1.3–3.7)	<0.01
No	47 (19 %)	199 (81 %)
PAH
Yes	17 (31 %)	38 (69 %)	1.6 (0.8–3.0)	0.16
No	67 (22 %)	237 (78 %)
Stunting
Yes	32 (31 %)	70 (69 %)	1.8 (1.1–3.1)	0.02
No	51 (20 %)	205 (80 %)
Underweight
Yes	42 (37 %)	72 (63 %)	2.8 (1.7–4.7)	<0.01
No	42 (17 %)	203 (83 %)
Tet spells
Yes	18 (40 %)	27 (60 %)	2.5 (1.3–4.8)	<0.01
No	66 (21 %)	247 (79 %)
CPB above 60 min
Yes	30 (50 %)	30 (50 %)	4.6 (2.5–8.2)	<0.001
No	52 (18 %)	237 (82 %)

3.10. Factors associated with mortality among operated children

Children who were on cardiopulmonary bypass for over 60 min were four times more likely to die than those who were not (OR = 4.0; 95 %CI: 1.4–11.8; p < 0.01). Patients who were hospitalized for seven days or longer were 9.7 times more likely to die than those who were hospitalized for fewer than seven days (OR = 9.7; 95 %, CI: 2.7–34.3; p < 0.01) (see Table 8).

Table 8.

Factors associated with mortality among operated children.

Predictors	Death		OR (95 % CI)	P-value
	Yes (n = 15)	No (n = 391)
Age at the time of operation (y)
0–5	12 (5 %)	229 (95 %)
5–10	0 (0 %)	87 (100 %)
10–15	1 (2 %)	49 (98 %)	0.4 (0.1–3.1)	0.37
ICU stay >7 days
Yes	4 (22 %)	14 (78 %)	9.7 (2.7–34.3)	<0.001
No	11 (3 %)	374 (97 %)
Heart failure before surgery
Yes	6 (5 %)	109 (95 %)	1.78 (0.6–5.3)	0.30
No	8 (3 %)	259 (97 %)
PHTN
Yes	3 (5 %)	64 (96 %)	1.27 (0.4–4.7)	0.71
No	12 (4 %)	327 (97 %)
Stunting
Yes	3 (3 %)	111 (97 %)	0.63 (0.2–2.3)	0.48
No	12 (4 %)	279 (96 %)
Underweight
Yes	5 (4 %)	119 (96 %)	1.14 (0.4–3.4)	0.81
No	10 (4 %)	272 (96 %
Tet spells
Yes	4 (8 %)	49 (92 %)	3.0 (0.9–10.2)	0.07
No	9 (3 %)	334 (97 %)
CPB above 60 min
Yes	6 (10 %)	54 (90 %)	4.02 (1.4–11.8)	0.011
No	9 (3 %)	326 (97 %)

4. Discussion

The management of CHD remains a challenge in LICs due to a lack of infrastructure and a scarcity of highly skilled medical personnel [15]. In Rwanda, visiting international teams have historically played a major role in providing surgery and catheterizations along with referrals abroad for a limited number of patients. Post-operative follow-up is administered by local physicians in a context of limited capacity in providing optimal care, both in terms of qualified human resources and appropriate equipment. Further, the few cardiologists working in the health facilities of Rwanda are all based in Kigali. This study was conducted to assess clinical outcomes among children undergoing interventional treatment of CHD since the initiation of the mission-based cardiac surgery program in Rwanda in the context of the aforementioned challenges.

This is the first study of its magnitude to analyze postsurgical outcomes of Rwandan children operated for CHD, addressing this broad knowledge gap and guiding future research in the field and clinical practice in Rwanda and other low resource settings. As Rwanda's cardiovascular healthcare environment is actively transitioning from mission-based to locally managed care, this serves as a benchmark for care in the country.

4.1. Nutrition and demographics

Among the 406 children enrolled, 30 % were underweight and 28 % were stunted at the time of intervention. Similar findings were reported by Ross et al. with 25 % of stunted and 26 % of underweight children in the USA at the time of surgery [16]. This finding highlights the impact of malnutrition among children with CHD in both high-income and low-income countries. Battle et al. conducted a study at Mulago Hospital, the main Ugandan referral hospital in a similar health setting to CHUK, reporting 43 % and 45 % of underweight and stunted children with CHDs, respectively [17]. The majority of operated children in our study were under five years of age (64 %), and predominantly female (59 %). Similarly, Chinawa et al. reported that most cardiac interventions in children with CHD in Nigeria occurred below 5 years of age and primarily in females (59 %) [18].

4.2. Cardiac lesions

In our series, the most common cardiac lesion in children who underwent surgical interventions was VSD followed by PDA (28 % and 19 %, respectively). In a study conducted by Murni et al. among Indonesian children, VSD was also the predominant heart defect repaired with 29 % [19]. Different results were reported by Larrazabal et al., with PDA closure being the common surgical intervention at 34 % among Guatemalan children [20]. Variation in the proportion of CHD might be due to disparity in available infrastructure and qualified personnel, as some countries, such as Rwanda, rely heavily on humanitarian medical organizations, whilst others utilize cardiac centers and patients have easy access to timely surgical and catheterization interventions. Humanitarian medical organizations set the criteria for selecting cases that are easily correctable due to policies of rational resource allocation. This also accounts for the discrepancies in specific cases treated in cardiac centers worldwide. The notable absence in our series also of infant and ductus-dependent cardiac lesions is typical of reports from this region. Even relatively simple repairs such as for aortic coarctation are rare in our setting, where prostaglandin is rarely available.

4.3. Catheter interventions

Of the 154 interventional catheterizations performed, PDA closure and balloon valvuloplasty constituted 68 % and 21 % of the cases, respectively. Similarly, a study conducted in Uganda on a larger number of patients who underwent catheterization procedures showed the predominance of PDA followed by balloon valvuloplasty (51 % and 12 %, respectively) [21]. Our observed survival of 99 % in the catheterization group compares favorably with centers in developed economies. One death of a five-month-old with complex CHD (ToF with severe right ventricular outflow obstruction and PDA) occurred two months after the pulmonary valve dilatation, however. Similar results were reported in a study conducted among Egyptian children with two children (0.7 %) dying due to general anesthesia-provoked hypercyanotic spells [22]. A previous Rwandan study conducted by Senga et al. showed favorable results with excellent clinical status at follow-up among 20 and 9 patients who underwent PDA closure and balloon valvuloplasty, respectively [23]. Regular missions of cardiac catheterization in Rwanda show similarly good results for both lesions. They also show the effectiveness of local cardiac follow-up and indicate the importance of the collaboration between the Rwanda Ministry of Health and humanitarian teams that perform cardiac catheterization.

4.4. Surgery

Cardiac surgery in Rwandan children has until recently been dominated by international visiting teams, such as Chain of Hope (COH), Belgium and Open Heart International (OHI), which completed 62 % and 19 % of pediatric cardiac interventions, respectively. We registered an overall mortality of 6 % among children who underwent cardiac surgery. Of these, 10 of 15 died within 30 days post-operatively. Similar findings were reported in a study conducted among 531 children with CHD operated in France by Mécénat Chirurgie Cardiaque from developing countries [28]. Vijarnsorn et al. reported a mortality rate of 6 % among children operated for CHD in Thailand [29]. A low mortality rate of 1.82 % was reported by Rohde et al. among indigenous children in Australia, a developed country with very advanced cardiac surgery and interventional cardiology services [26]. Comparatively, Onan et al. reported a mortality rate of 10.7 % in Turkey explained by the initiation of the cardiac surgery program, predominantly palliative rather than corrective procedures among operated children [20].

4.5. Adverse events

Our study reveals an increased incidence of complications among operated children who spent over 60 min on CPB and had ICU stays longer than 7 days. Early postoperative complications are known to lead to longer ICU and hospital stays. A study conducted by Agarwal et al. found a strong association between long duration of CPB and long ICU stay with a high incidence of postoperative complications after CHD surgery [30].

A total of 8 % children were readmitted to the hospital. Similarly, Benavidez et al. found that 11.3 % of children were readmitted following surgery [26]. While residual shunt was reported in 15 % of children enrolled in our study following surgery, Bibevski et al. reported 57 % of children with a residual shunt at the time of discharge following surgical closure of VSD [27]. Our findings show the effectiveness of surgical interventions performed by highly experienced cardiothoracic surgeons from medical teams conducting regular missions in Rwanda.

In our study, children with findings of inadequate nutrition were prone to major complications. This is likely due to the decreased function of the immune system, impaired wound healing and cardiovascular system instabilities seen in the context of malnutrition [31]. Similar findings were reported by Ross et al. with an increased rate of complications, post-operative infections, and prolonged hospital stay among underweight and stunted children at the time of CHD surgery [16]. In this study, children who underwent cardiac surgery and who required CPB with a duration longer than 60 min experienced increased mortality. Likewise in a study conducted by Knowles et al., a long duration of CPB was associated with increased mortality among operated children [32]. Prolonged CPB may cause physiologic changes leading to post-operative fluid overload and edema with more serious effects in pediatric populations [33].

4.6. Potential Benefit of a CHD registry

Establishment of a cardiac registry would vastly simplify a study such as this by assuring accurate, systematic collection of followup data. It would allow tracking, monitoring CHD trends and inform policy makers, and thereby be an enhancement to research and improve patient care. It could even serve as a tool to establish national or regional CHD treatment guidelines.

4.7. Limitations

Although this study offers a general picture of postoperative outcomes, we acknowledge its limitations, such as missing data in a large number (14 %) of potentially eligible participants. Clinical documentation on paper was easily misplaced. Failure to achieve follow-up, however could well be a marker of morbidity and mortality, which would lead to an artificially favorable portrayal of outcomes. Further, the absence of follow-up data from other hospitals not included in this study, such as Rwanda Military Hospital, the University Teaching Hospital of Butare, or private clinics, limits study findings. Finally, no attempt was made to assess or compare the case mix or clinical performance of the individual contributing teams who provided surgical services.

4.8. Future directions

Cardiac care for children in Rwanda has undergone substantial advancements. Notably all hospitals nationwide have adopted electronic medical record (EMRs) system to facilitate efficient and accurate data collection and storage. The government has demonstrated its commitment to progress through sustainable investments in infrastructure, specialty training programs for health care personnel and locally performed cardiac surgery. ‘South-to-south collaboration to strengthen the health workforce: the case of pediatric cardiac surgery in Rwanda’; An ongoing study is being conducted to assess this new model. These promising efforts will ultimately enhance outcomes for children with CHD in Rwanda.

5. Conclusion

Surgery and interventional catheterization for CHD in Rwandan children until recently relied heavily on the work of international visiting teams or referral abroad. Procedures for various CHD were performed with an overall survival of 96 % despite frequent delays to intervention. Pre- and post-surgical predictors of complications and mortality, such as malnutrition, ICU stay over seven days, and CPB time over 60 min, are becoming better understood and represent potentially modifiable factors for more successful future outcomes. However, the lack of a national CHD registry restricts a more comprehensive overview on surgical outcomes of children with CHD. It would be optimal, as procedures are increasingly undertaken by local professionals, for program results to be systematically recorded and measured.

CRediT authorship contribution statement

Providence Akingeneye: Writing – original draft, Project administration, Investigation, Data curation, Writing – review & editing, Resources, Methodology, Formal analysis, Conceptualization. David J. Bradley: Supervision. Joseph Mucumbitsi: Supervision. Naphtal Nyilimanzi: Supervision. Yves Mutabandama: Supervision. Emmanuel K. Rusingiza: Conceptualization, Supervision.

Funding sources

This reseach did not receive any specific grant from funding agencies in the public, commercial, or not for profit sectors.

Declaration of competing interest

The authors declare no competing financial interests or personal relationships that could influence the work reported in this manuscript.