Effects of Undernutrition and Predictors on the Survival Status of HIV-Positive Children after Started Antiretroviral Therapy (ART) in Northwest Ethiopia

Malnutrition and human immunodeficiency virus/acquired immunodeficiency syndrome have complex and multidirectional relationships. Ethiopia is one of the countries hardest hit by the HIV epidemic as well as malnutrition. This study was aimed at assessing the effects of undernutrition on the survival status of HIV-positive children who received HIV/AIDS care in Northwest Ethiopia. Materials and Methods. A facility-based retrospective follow-up was conducted from January 1, 2009, to December 31, 2020. The data was entered into EpiData version 4.2.0. Then, the entered data was exported to STATA 14 software for further analysis, and the Kaplan-Meier survival curve was used to estimate survival time after the initiation of ART. The Bivariable and multivariable Cox regression analyses were conducted to identify predictors of mortality associated with undernutrition. Results. The mean (±SD) age of participant children was found 118.4 (±38.24) months. The overall mortality rate in this study was determined as 5.4 per 100 child-years (95% CI: 3.6, 5.8). Children with CD4 cell counts below the threshold [AHR = 1.6; 95% CI (1.19, 7.85)], advanced WHO clinical stages (III and IV) HIV [AHR = 4.5; 95% CI (2.80, 8.40)], and being severe stunting at the beginning [AHR = 2.9; 95% CI (1.80, 6.40)] were significantly associated with mortality of HIV-positive children. Conclusion. The findings of the current study indicated that HIV-positive children on ART had a high rate of mortality. Baseline undernutrition has the mortality of children who had CD4 counts below a threshold, advanced WHO HIV clinical staging (III and IV), and being severe stunting (HAZ ≤ −3 Z score) which were found to be independent predictors for mortality of undernourished HIV.


Introduction
Malnutrition and human immunodeficiency virus (HIV)/ acquired immunodeficiency syndrome (AIDS) have complex and multidirectional relationships that cause progressive immune system damage [1]. Both are frequently intertwined and have a synergistic effect [2]. Malnutrition increases viral replication and accelerates the progression of HIV disease by decreasing CD4 T cells, suppressing delayed hypersensitivity, and altering β-cell responses [3,4]. For this reason, malnutrition has a poor prognosis for clinical outcomes [5]. However, HIV/AIDS accelerates the progression of immune impairment and the occurrence of opportunistic infections [6] by increasing the risk of appetite loss, hastening the decline of CD4 T cell concentration, and causing markers of microbial translocation (16sDNA), intestinal damage (iFABP), monocyte activation, and increased proteolysis in HIV-positive children [6][7][8][9][10]. Children with HIV/AIDS have a reduced appetite and ability to consume food, as well as a higher incidence of diarrhea, resulting in malabsorption and nutrient a loss, which makes malnutrition a common phenomenon [11,12]. In 2018, an estimated 1.7 million children (aged <15 years) were living with HIV worldwide [13,14]. In the same year, in Ethiopia, 44229 children were diagnosed with HIV and 2055 died related to AIDS [9,15]. Undernutrition in all its forms remains a global public health burden that accounts for 45% of all deaths for people living with HIV [2,16]. Children are more vulnerable to undernutrition than adults with HIV; hence, rapid viral replication and a higher rate of CD4 cell destruction are inevitable due to immature immunity [17]. Evidence suggests that even relatively small losses in weight (5%) are associated with a decrease in the survival rate of HIV-positive children [18]. Epidemiologically, the incidence of stunting is declining too slowly, while wasting still has a great impact on too many young children worldwide [19]. In 2017, 13.8 million children were wasted, of whom 4 million were severely wasted [13,20]. Most of all were in sub-Saharan countries [21]. Each year, over a million children die and develop severe acute malnutrition (SAM), and it presents mainly in marasmus, kwashiorkor, or marasmus kwashiorkor [22,23]. Similarly, in comparison to kwashiorkor, wasting was more commonly reported [8,24], as was early bone growth failure and an increased risk of death in HIV-positive children [1,25].
Several studies investigated whether lower CD4 cell count, age, wealth index, adherence, food insecurity, and social support were risk factors for malnourished HIVpositive children's mortality [1,2,15,[26][27][28]. Ethiopia is one of the countries hardest hit by the HIV epidemic as well as malnutrition [1]; the prevalence of undernutrition among HIV/AIDS children ranges from 12.3 to 46.8% [19,29] and is blamed for 57% of deaths [6,13]. However, evidence of the effect of undernutrition on the survival status of HIVpositive children who are <15 years of age remains sparse and inconclusive [11,29]. This study aims to assess the effects of undernutrition and its predictors on the survival of HIV-positive children who received antiretroviral therapy (ART) in selected public health facilities in Northwest Ethiopia.

Study Design and Setting.
A facility-based retrospective follow-up study was employed among 721 HIV-positive children who started HIV/AIDS care in two hospitals and two health centers since January 1, 2009, to December 31, 2020, which are representative of the Benishangul-Gumuz region. All these health facilities provide health care services for all Oromia, Amhara, and Benishangul regions [30]. Following HIV/AIDS care initiated in the regions between January 1, 2009, and December 31, 2020, there were 2968 HIV/ AIDS care started populations. More than 737 of these populations were children living with HIV/AIDS, and the data were drawn from two hospitals and two health centers (Assosa Hospital N = 359 and Pawe Hospital N = 315; Felege Selam Health Center N = 21 and Gilgel Beles Health Center N = 37) which were included for the final data analysis.

Populations.
The records of all HIV-infected children, whoever started ART in four health institutions, were the source of the population. The records of all HIV-infected children receiving ART between January 1, 2009, and December 31, 2020, were considered as the study population. All HIV-infected children who had at least one month of ART follow-up from January 1, 2009, to December 31, 2020, were included.

Data Collection Instruments.
Before the data collection procedure was deployed for children, the standard anthropometric measurements were taken from all subjects at recruitment and at follow-up after being discharged from the clinic. MUAC was measured using a color-coded MUAC tape on the left arm. Data abstraction tools (checklists) were prepared using Ethiopia's Federal Ministry of Health Pediatrics ART follow-up and medical history sheet combination [35]. Six diploma nurses and four BSc public health officers were recruited for data collection and supervision. One day of training was given for data collection and supervision for all facility data collectors.

Data Collection Procedures and Quality Assurance.
To assure the quality of the data, data collectors and supervisors were trained about how and what information they should collect from the medical records for one day. The checklist was pretested on 5% of randomly chosen charts that were not included in the actual study. After the pretest, the necessary modifications to the data collection tool were made. Strict follow-up and supervision were carried out during data collection by the principal investigator, and feedback was given daily. Individual records with incomplete data during data collection were excluded. The collected data was first checked and cleaned for completeness.  with the log-rank test shows differences in the hazards of death or undernutrition in HIV-positive children on different covariates. Being HIV-infected children at baseline stunting (HAZ ≤ −3 Z score), being on WHO clinical stage (III and IV), and having CD4 count below threshold had survival differences between undernourished and counter peers do have (Figures 1-3).

Bivariable and Multivariable Cox Regression Analysis.
As shown in Table 4 in the multivariable Cox regression analysis, after adjustment and controlling of certain confounding in the final model, three variables were found significantly associated with mortality of undernourished children. HIV positive children with CD4 cell counts below the threshold were 1.6 times at higher risk of death as

Discussion
In this retrospective cohort study, the effects of undernutrition and its predictors for survival status among HIVpositive children on ART were determined. At the end of chronic successive follow-up, about 90 (12.48%) patients were deceased, and 14 (1.94%) patients were absconded. The review data on the national level revealed a mortality rate of 5 to 8% at 6 months and 24 months after ART started [11,23]. The overall mortality rate of this study was 5.4 per     International Journal of Pediatrics 100 child-years (95% CI: 3.6, 5.8), which is consistent with findings from the Felege Hiwot Comprehensive Specialized Hospital, Northwest Ethiopia (four deaths per 100 childyears) [37] and Northwest Ethiopia 4.4 per 100 child-years (95% CI: 3.2, 6.0) [13]. However, the mortality rate reported in this study is higher than studies reported from Mekelle Hospital, Northern Ethiopia (1.4 deaths per 100 child-years) [38], Wolaita Zone health facilities, Southern Ethiopia (2.1 deaths per 100 child-years) [39], Northwest Ethiopia (3.2 deaths per 100 child-years [29], the Asia-Pacific region (1.9 deaths per 100 child-years) [40], Zimbabwe (2.9 deaths per 100 child-years) [41], and Congo (3.4 deaths per 100 childyears) [42]. Conversely, the mortality rate found in this study is much lower than that of Debre Tabor General Hospital and Dessie Referral Hospital (6.3 deaths per 100 childyears) [38], Addis Ababa (8.8 deaths per 100 child-years) [13], Southwest Ethiopia (11.2 deaths per 1000 personyears) [11], and Kenya (8.4 deaths per 100 child-years) [43]. Explanations for variation in the incidence of mortality rate might be due to the difference in health care awareness of the community, sample size, study settings, study period, and/or characteristics of study participants. The present study explored predictors of mortality among study participants. In this study, children who had CD4 counts below the threshold showed a higher risk of death than their counterparts. This finding is consistent with other previous   [11,18,19,29,37,44], India [45], Congo [42], Tanzania [46], Bangladesh [47], and Malaysia [48], which all indicate that low CD4 count was an independent predictor of mortality. Serious and lifethreatening opportunistic infections, such as central nervous system toxoplasmosis and cryptococcal meningitis, are more common in children with severe immunodeficiency. This hastened a sharp reduction in CD4 count; this has declined the survival probability of children since care started. Advanced WHO HIV clinical staging (III and IV) was another strong predictor of mortality. Children with advanced WHO HIV clinical stage (III and IV) at the time of ART initiation have a higher risk of death as compared to their counterparts with mild status (i.e., WHO HIV clinical disease stage (I and II)). Similar results were reported from previous studies conducted in Ethiopia [8,37,49,50], two rural hospitals in South Africa [5], Tanzania [38], Zimbabwe [41], Kenya [43], Eastern India [45], and Malawi [51], which all indicated that advanced WHO HIV clinical disease stages were a predictor of mortality. For those living with HIV, as WHO HIV clinical staging becomes more advanced, the risk of developing and recurrence of opportunistic  International Journal of Pediatrics infection also increases, which might be associated with the cause of death. In this study, severe stunting was also an independent predictor of mortality among HIV-positive children on ART. Children who were severely stunted before ART initiation have a higher risk of death as compared to those who were not stunted. This is consistent with findings in Northwest Ethiopia [13], Burkina Faso [4], and Tanzania [22]. Reasonably, malnutrition is a common complication of HIV infection. Severe stunting is associated with a weakened immune system and complicates the treatment of diseases by affecting intestinal absorption of drugs and the ability to absorb various nutrients, besides causing dysregulated lipid metabolism and increased proteolysis in the body [1,8,46].

Limitations of the Study
There were inconsistencies in determining causes of death in this study, particularly for individuals who died at home. Some of those who died without being notified may have been counted as "lost to follow-up." In addition, important determinants of death, such as viral load and nutritional deficiencies, were also not taken into account.

Conclusion
The findings of the current study indicated that HIVpositive children on ART had a high rate of mortality. In this study, nutritional status was found to have a significant effect on the survival of HIV-positive children on ART. Baseline CD4 counts of less than the threshold, advanced WHO HIV clinical disease staging (III and IV), and severe stunting were found to be independent predictors of mortality in HIV children on ART. This calls for the government to give due attention to strengthening HIV/AIDS treatment and care modalities and ensure that the relevant nutritional support for children at risk is provided appropriately.