Outcome after Discharge from Hospital of Children with Complicated Severe Acute Malnutrition and Predictors Factors of Non-Response during Outpatient Treatment, in Senegal ()
1. Introduction
Malnutrition is a public health problem, particularly in developing countries. Acute malnutrition affects 52 million children under the age of 5 with 17 million in severe form worldwide. Asia and Africa are most affected by the burden of malnutrition (96%), with two-thirds and one-quarter of children under five suffering from acute malnutrition respectively [1] [2] . It has a negative impact on health, increases the risk of mortality and causes intellectual or cognitive deficiencies in children, preventing them from reaching their full potential [3] . Severe acute malnutrition can be defined by a weight-for-height index of less than −3 z-score, a brachial circumference of less than 115 mm, and bilateral pitting edema in children aged 6 to 59 months [4] . The risk of death is 9 to 11 times higher in severe acute malnutrition. However, most deaths occur in children with mild to moderate malnutrition [5] [6] . Malnutrition is implicated in approximately 45% of deaths in children under 5 years of age, accounting for more than 3 million deaths per year [7] . Studies in Africa and Asia have reported recovery rates of 25% to 95% and 50% to 93% respectively in children followed for SAM in hospital and community settings. [8] . In Senegal, 8% of children suffer from acute malnutrition, and 1% of them have a severe form [9] . Fewer or no studies have examined the outcome of children who were hospitalized for severe acute malnutrition with complications discharged from hospital. The aim of this study was to assess the outcome of children hospitalized for severe acute malnutrition with complications after discharge.
2. Methods and Materials
2.1. Study Design
This is a descriptive and analytical cross-sectional study conducted from 1st March to 31st November 2021 at the Albert Royer’s Children Hospital (HEAR). This is the largest pediatric hospital in the country and is of the highest standard for our country with medical and surgical ward. Children of all ages are referred to it from secondary or peripheral health facilities as well as from other hospitals of the same standing (level III). It has a nutritional recovery and education center. Most children from the Dakar region with severe pathologies, or even patients from other regions, are treated at CHNEAR.
2.2. Populations
All children aged 6 - 59 months who were hospitalized for complicated severe acute malnutrition at Albert Royer’s Children Hospital were taken as the study population.
2.3. Eligibility Criteria
All children aged 6 - 59 months hospitalized for complicated severe acute malnutrition discharged from hospital, whose nutritional management was continued on an outpatient clinic, and whose parents were reachable and willing to participate in the study. While those children aged 6 - 59 months with congenital malformations were excluded.
2.4. Data Collection
Patients were identified from registries and hospitalization records for the years 2018 and 2019. Parents were called by telephone and an appointment was scheduled at their convenience. For all included patients, we collected epidemiological data (age, gender, address, feeding; mother’s age, education, marital status and occupation, clinical manifestations (palor, oedemas, complications such as fever, cough and/or breathing difficulties, diarrhea, vomiting,) and nutritional status (weight, height, weight for height or body mass index for age) on the medical record and during appointment. Children were weighed using the Seca digital scale with tare function. Height was measured with the UNICEF height gauge, in the supine position for patients who were less than equal to 24 months and were unable to stand by themselves and in the standing position for children who were more than 24 months and can stand by themselves developmentally. Height, weight, and age data were used to calculate Z-scores for the weight-for-height (W/H) indices according to reference curves provided by the World Health Organization (WHO). Z-scores were calculated with the Anthro program (WHO Anthro, Geneva, Switzerland) [9] .
2.5. Statistical Analysis
Qualitative variables are described according to frequencies, proportions and their confidence intervals. Quantitative variables are described according to the mean and standard deviation when the distributions were normal, in median and quartiles in the opposite case. A univariable logistic regression was first performed to select the variables (with a p-value < 15%) potentially associated with malnutrition. In the second step, these selected variables were introduced into a multi-variable model. A top-down stepwise regression strategy was used to identify factors associated with persistent malnutrition with a p-value < 5%. All analyses were performed with R software.
3. Results
A total of 155 children were hospitalized during the two years for severe acute malnutrition with complications. Among them, 16 children died during hospitalization, 15 parents of children could not be reached, and 24 parents declined the invitation to participate in the study (Figure 1). One hundred and three children were included in the study. The sex ratio was 1.23.
The mean age of children at hospitalization was 16.41 ± 10.11 months and 46.03 ± 13.43 months at the time of the study. The age groups 6 - 12 months (28.16%) and 12 - 18 months (25.24%) were the most frequent. The mean age of the mothers was 29.75 ± 5.73 years. The sociodemographic characteristics of the children and their mothers are shown in Table 1.
3.1. Medical Complications and Clinical Features of SAM during Hospitalization
At hospitalization, the children had marasmus (76.70%), kwashiorkor (15.53%) and a mixed form (7.7%). The mean z-score of weight for height was −4.76 ± 0.32. The main infectious complications were dominated by pneumonia (47.67%) including 6 cases (4.65%) of tuberculosis, anemia (47.57%) and hydro electrolytic disorders (40.78%) with 15.53% of severe acute dehydration with shock. Two children (2.34%) had HIV infection.
3.2. Present Nutritional Status
The mean duration of follow-up after hospital discharge was 29.63 ± 8.59 months. Three children (2.91%) died after hospital discharge. Slightly more than two-thirds (75.73%) of the children had made a full recovery. The mean weight-for-height ratio z-score was −1.34 ± 1.08. The median weight-for-height ratio z-score was −1.22 with extremes of −3.98 and 1.73. The nutritional status of the children is shown in Figure 2.
In univariate analysis (Table 2), there were no significant differences according to gender (p = 0.59), feeding (p = 0.19), address (p = 0.38), mothers’ level of education (p = 0.25) and the presence of edema (p = 0.85), anemia (p = 0.96) and HIV infection (p = 0.42). In contrast, children who were 6 - 12 months old at hospitalization (p = 0.04) and those with tuberculosis (p = 0.01) were the most affected by malnutrition. After multivariate analysis by top-down stepwise logistic regression method (Table 3), the most predictive factors for nonrecovery were ablactation before 2 years of age (AOR: 12.21; IC 95% [6.82 - 18.44]; p = 0.04) and tuberculosis (AOR: 21.06; [12.54 - 41.09]; p = 0.03).
![]()
Table 1. Child and mother socio-demographic characteristic.
![]()
Table 2. Risk factors of non-recovery (univariate analysis).
![]()
Table 3. Risk factors of non-recovery (multivariate analysis).
4. Discussion
This work showed that the recovery rate of children at least one year after hospitalization for severe acute malnutrition with complications is quite satisfactory at 75.73% and that ablactation before 24 months and the existence of tuberculosis on admission are the two most predictive factors for non-recovery. This rate is in line with the international standard which predicts a recovery rate for severe malnutrition of more than 75% [10] . In sub-Saharan Africa, a meta-analysis found a combined prevalence of 71.2%, with better results in East, Central and North Africa with 71.4%, 84.7% and 78.9% recovery respectively. West Africa (60.9%) and South Africa (64.7%) had the lowest rates [8] . In Ethiopia, two meta-analyses reported a cumulative prevalence of nutritional recovery of 70.5% and 72.02% with variations according to region. The best results were noted in the Oromia region (80.29%) [11] [12] . The recovery rate was much better in Zambia and Zimbabwe, 86.6% and 87.8% respectively for HIV-infected and uninfected children [13] . The rate of lost to follow-up ranged from 0.0% to 45% [14] . It was 6% in Nigeria [15] , 34.2% in Burkina Faso [16] , and 10% in Ethiopia [17] . Although WHO recommends inpatient follow-up for SAM with complications and continued management in the community after resolution of complications and for uncomplicated SAM [18] , hospitalized children continued to be followed up in the hospital after discharge by medical staff. This may contribute to the loss of sight, especially since the majority of our study population lived in suburban areas, which were generally poorly developed, whereas the hospital was located in an urban area. A study in Mali found a higher proportion of dropouts and noncompliance with appointments among children followed up in health facilities compared with those followed up by community health workers close to their homes [19] . In the work of Kangas ST et al., the recovery rate was slightly better for children followed by community health workers than for those followed by health workers [20] .
In our study, delayed recovery was associated with ablactation before two years of age and the existence of pulmonary tuberculosis at admission. The existence of disease or infection during follow-up has been reported in Burkina Faso [20] , South Africa [21] , and Ethiopia [22] .
The interaction between infection and nutrition has been reported by WHO and operates in a bidirectional manner. Infection alters nutritional status through decreased food intake secondary to the anorectic effect of cytokines, increased losses due to increased protein catabolism, fever and chills, loss of urinary nitrogen, and inflammation-related malabsorption [23] [24] [25] [26] . The association between tuberculosis and malnutrition is well known [27] and delayed nutritional recovery in tuberculosis has been reported in several studies [28] [29] .
In our study, 3.88% of children remained severely malnourished. This rate is much lower than that reported in Nigeria (18.1%) [30] . In our study, 2.91% of children died after discharge. This rate is slightly higher than that reported in Burkina Faso (2.2%) [16] , but much lower than that found in Jigawa state in Nigeria (6.6%) [30] . The low death rate could be explained by the follow-up of children at the hospital level by pediatricians who detect and manage morbidities early.
5. Conclusion
In conclusion, the nutritional recovery rate of children formerly hospitalized for severe acute malnutrition with complications is globally satisfactory with low mortality. The existence of tuberculosis and ablactation before the age of two is factors of non-recovery of nutrition. It is therefore essential to respect the feeding recommendations for infants and young children and to insist on primary prevention of tuberculosis.
6. Limitations
Limitations of this study include the unavailability of some parents of formerly malnourished children to participate in the study, and the absence of a harmonized protocol for outpatient follow-up of children, with a lack of clinical data in some medical records during outpatient follow-up.
Acknowledgements
We would like to acknowledge all the children and parents who agree to participate in this study. Special thanks go to nutritional recovery and education center manager and the supervision on duty.