The use of estimated glomerular filtration rate in the evaluation of renal function in HIV-positive children in Enugu

Background: Glomerular filtration rate (GFR) is a measure of renal function that estimates the volume of fluid filtered from the renal glomerular capillaries. An assessment of renal function in HIV-positive children is timely due to the high prevalence of HIV-associated nephropathy (HIVAN) in HIV-positive children that progresses to end-stage renal disease (ESRD). Objectives: To estimate the glomerular filtration rate using creatinine-based Schwartz formula. To determine the renal function in HIV-positive children in UNTH, using the estimated glomerular filtration rate (eGFR). Materials and Method: A total of 159 HIV-positive children had serum creatinine estimation and GFR calculated using the Schwartz formula. Values of eGFR of 2SD below the normal for age and gender as determined by National Kidney Foundation (NKF) were taken as low renal function. Results: There were 80 males and 79 females enrolled. The mean eGFR for the study population was 109.44 ± 16.86 mls/min/1.73 m ² . Ninety children (56.6%) had normal renal function while 69 (45.4%) had low renal function for their respective age groups. Those with low renal function were older (P < 0.001) and had lower CD4 cell counts (P = 0.01). The eGFR was inversely related to age (r = -0.30, P = < 0.001) and directly related to CD4 cell count (r = 0.30, P = < 0.001). Conclusion: There is derangement in renal function among HIV-infected children as seen from their eGFR.

Keywords: Estimated glomerular filtration rate, HIV-infected children, immunosuppression, renal function

HIV is a major cause of infant and childhood morbidity and mortality in Africa. ^[1],[2] It affects the kidney manifesting commonly as HIV-associated nephropathy (HIVAN): A chronic kidney disease (CKD) that progresses rapidly to end-stage renal disease (ESRD) ^[2] in patient with increased HIV RNA level ^[3] and decreased CD4 cell count. ^[4] Treatment of patients in early stages of HIVAN with highly active anti-retroviral therapy (HAART) can slow progression of the CKD ^[3],[4],[5] due to HAART- mediated suppression of viral load. ^[4]

The NKF recommends the evaluation for CKD in children using their GFR estimated from the Schwartz formula ^[5] although it overestimates their GFR. ^[6],[7] This estimation of the renal function in HIV-infected children is necessary because the institution of HAART in those with deranged renal function will help to forestall the progression of HIVAN to ESRD. ^[3],[4] Chaparro and colleagues noted a decrease in viral load and improvement in the renal function of patients with HIVAN after institution of HAART. ^[4]

This study was carried out in the University of Nigeria Teaching Hospital, (UNTH), Enugu, in Enugu state. The Pediatric Infectious Disease Clinic of the teaching hospital was established in the year 2005. At the time of the study, there were 236 pediatric HIV-positive children registered in the clinic. This was a cross-sectional descriptive study in which consecutive HIV-infected children attending the Pediatric Infectious Disease Clinic of UNTH, and who were 18 months to 14 years, were enrolled. Ethical approval was obtained from the Health Research and Ethics Committee of UNTH, Enugu before commencing the study.

On presentation to the clinic, a written consent was obtained, data on gender, age, HAART use and its duration, duration of HIV diagnosis (defined as the time duration from the documentation of confirmed HIV infection to the time of the study), the results of CD4 cell count and percent were obtained. The physical examinations of the participants were performed to obtain the vital signs, anthropometric measurements, and clinical status of the participants. These data were used to categorize the participants according to WHO clinical staging system for infants and children. ^[8] The weight and the height of the participants were measured using the combined measuring instrument: HEALTH WEIGHING SCALE and STADIOMETER (RTZ-120A, HECOS, China) with respective sensitivities of 0.1 kilogram and 0.1 centimeters.

Subsequently, blood samples were collected in a non-heparinized bottle and sent to the laboratory for estimation of their serum creatinine, using the Jaffe reaction method. With the Schwartz formula, ^[9] the serum creatinine and height obtained for each participants were used to calculate the respective glomerular filtration rate, taking into consideration the different values of constant “K” for each age group and gender: 0.55 in children 1-13 years and adolescent females above 13 years, 0.70 in adolescent males above 13 years.

The data were used to classify the study population into various categories. Advanced HIV disease was defined as WHO clinical stage 3 and 4 while stage 1 and 2 defined non-advanced HIV disease. ^[10] For immunological category, severe immunosuppression was defined as CD4 cell count of less than 200 cells/mm in those subjects equal to or more than five years old, CD4% of less than 15% for those between 36 and 59 months, CD4% of less than 20% for those between 18 and 35 months. ^[11] The anthropometric values of weight and height were used to classify the nutritional status of the children using the World Health Organization weight-for-height z-score nutritional status parameters. ^[12] Values greater than -1SD up to +1SD defined well-nourished children while malnutrition was defined with the index values -1SD and below. ^[13] Using the WHO age stratification for age-related CD4 values, ^[14] the HIV-positive children were stratified into three different age groups; 18 months – 35 months, 36 month – 59 months, and 5 years – 14 years.

The estimated GFR (eGFR) was used to determine the renal function of the children. Normal renal function for children (mean eGFR ± SD in mls/min/1.73 m ² ) was taken as 133.0 ± 27.0 (for males and females aged 2-12 years old), 140.0 ± 30.0 (for males aged 13-21 years old), and 126.0 ± 26.0 (for females 13-21 years of age). ^[5] Values of eGFR less than 2SD below the mean for the age and gender denote a low renal function. ^[5]

Demographic and laboratory data were analyzed using the Software Package for Social Science (SPSS) version 15.0 for Windows ^® . Continuous variables such as the age, weight, duration of HIV diagnosis were analyzed and expressed as mean and standard deviations, median, and range. Comparison of means was done using independent ‘t’ test and ANOVA. For categorical variables, chi-squared test was used. In order to ascertain any relationship between eGFR and subject’s characteristics, Pearson correlation coefficient was used. Significant levels were set with P-value of < 0.05 and 95% confidence interval.

There were 159 HIV-positive children enrolled consisting of 80 males and 79 females. The mean age of the study population was 79.4 ± 39.7 months (95% CI, 74.1 to 84.6, P =< 0.0001). The mean weight was 21.7 ± 7.6 kg (95% CI, 20.5 to 22.9, P =< 0.0001). The older subjects were heavier as shown in [Table 1]. Most of the subjects (88.1%) were well-nourished as 140 children had weight-for-height z score above-1SD while 19 children (11.9%) were malnourished with scores of -1SD and below. However, of all the subjects, only 4.4% (7) had severe malnutrition (weight-for-height z score less than -3SD).

The proportion of subjects on HAART was 77.4%; 4.4% had advanced HIV disease while 23.3% were severely immunosuppressed. The mean duration of HIV diagnosis was 21.1 ± 15.8 months (95% CI, 18.6 to 23.6, P =< 0.0001), and the mean duration of HAART treatment was 16.4 ± 16.0 months (95% CI, 13.9 to 18.9, P =< 0.0001). The mean CD4 cell count was 696.0 ± 446.5 cells/mm ³ (95% CI, 626.0 to 765.9, P =< 0.0001), and the younger subjects and the males had a higher CD4 cell count [Table 1].

The mean eGFR of the study population was 109.4 ± 16.9 mls/min/1.73 m ² (95% CI, 106.8 to 112.1, P =< 0.0001). The younger subjects had significantly higher eGFR than the older subjects, P = 0.01 [Table 2]. As shown in [Table 3], the mean eGFR was higher in males (P = 0.01), non-advanced HIV disease (P = 0.10), those not on HAART (P = 0.30), in malnourished children (P = 0.10), and in subjects with severe immunosuppression (P = 0.09). The correlation between eGFR and age of subjects was negative (r = -0.30, P =< 0.001), while between that and CD4 cell count was positive (r = 0.30, P =< 0.001) as shown in [Table 4].

Using the eGFR to define renal function, 90 (56.6%) had normal renal function while 69 (43.4%) had low renal function. None of the subjects had eGFR of less than 60 mls/min/1.73 m ² . Children with low renal function were significantly older (t = 0.00, P =< 0.0001), had shorter duration of HIV diagnosis, longer duration of treatment with HAART, and lower CD4 cell count [Table 5]. The proportion of the subjects with normal renal function in each age group is shown in [Table 6]. More males than females had normal renal function, (X ² = 3.35, P = 0.07; [Table 6]). There was no significant difference in the renal function of those on HAART and those not on HAART, (X ² = 0.06, P = 0.80; [Table 6]). Majority of the subjects had normal renal function irrespective of the degree of HIV disease severity, (X ² = 0.60, P = 0. 40; [Table 6]) and degree of immunosuppression, (X ² = 1.34, P = 0.20, [Table 6]).

Table 6: Renal function by age, gender, use of HAART, HIV disease severity, and degree of HIV disease severity Click here to view

This study showed that eGFR values of the participants were within normal ranges and none had chronic kidney disease, which is defined by the National Kidney Foundation (NKF) as the presence of kidney damage or a glomerular filtration rate (GFR) less than 60 mls/min/1.73 m ² for 3 months or longer. ^[5] This could be because of various factors such as the high CD4 cell count noted among the participants, which is known to be a predictor of normal eGFR. ^[14] A low CD4 cell count is a risk factor for direct HIV infection of the kidney in HIV-positive children resulting in renal parenchymal disease. ^[4],[15] Thus, HIV-positive children with high CD4 cell count are less likely to develop renal parenchymal disease and subsequent low eGFR. Other contributory factor to normal eGFR in the index study may be the use of HAART among the study participants. Treatment with HAART in HIV-infected subjects has been shown to suppress viral load and improve the renal function. ^[4] In the study by Wools-Kaloustian and colleagues, ^[14] despite the high CD4 cell count of the subjects, the prevalence of CKD was 11.5%, which could be because the subjects were HAART-naïve, further demonstrating the probable effect of HAART treatment on the eGFR. The subjects with non-advanced HIV disease have a lower tendency to the development of HIV renal parenchymal disease. ^[4],[15] We found that a greater majority of the subjects had non-advanced HIV disease, which may also explain the normal eGFR. Esezobor and co-workers ^[15] documented eGFR less than 60 mls/min/1.73 m ² in their study population, where 75% of their subjects had advanced HIV disease.

The eGFR showed an inverse relationship with age, with younger subjects having higher values. This could be attributed to the significantly higher CD4 cell count in the younger age group, and renal function is preserved in HIV-positive children with high CD4 cell count. Interestingly, this study also showed that the higher the CD4 cell count, the higher the eGFR values. The female subjects had lower eGFR than males, and this may be attributed to their comparably lower CD4 cell count. Although the serum creatinine-based Schwartz formula is influenced by muscle mass, which is less in females than in males, ^[3],[11] this study has, however, not demonstrated an apparent relation of body mass/gender to the eGFR. It may, therefore, be surmised that there may be multifactorial influence on the eGFR values in children.

The non-finding of subjects with CKD in the index study could also be explained by the method of estimation of GFR. Serum creatinine-based Schwartz formula overestimates GFR in children, ^[6],[7] and it is influenced by muscle mass, which varies with age, race, and gender. ^[3],[11] It is easy to use, but best predicts stages 3-5 CKD and, therefore, may not detect an early decline in renal function. ^[6] Esezobor and co-workers ^[15] documented eGFR of less than 60 mls/min/1.73 m ² in 13.3% of their subject, using Cystatin C-based formula. On the other hand, Wools-Kaloustian and colleagues ^[14] found 11.5% using the serum creatinine and the Modification of Diet for Renal Disease (MDRD) study equation. Thus, serum Cystatin C and MDRD study formulae may estimate GFR better than the Schwartz formula, which may have contributed to the finding of eGFR of greater than 60 mls/min/1.73 m ² among all the subjects in the index study.

Although the index study did not demonstrate eGFR less than 60 mls/min/1.73 m ² among the subjects, 43.4% of the subjects had low renal function (eGFR below the normal for age and gender). These subjects with low renal function were older, mostly females, and had lower CD4 cell count. Interestingly, those with low renal function had shorter duration of HIV diagnosis and longer duration of treatment with HAART. This shows that other factors such as age, gender, and CD4 cell count may influence the effect of time duration of HIV diagnosis and HAART treatment on renal function.

There is derangement in renal function among the HIV-positive children. A low eGFR, which denotes slow renal function, was found in older age, female gender, and subjects with low CD4 cell count. The eGFR decreases with age and increases with high CD4 cell count. We recommend routine monitoring and baseline renal function estimation in HIV-positive children. It is further recommended that another validated method of estimation should be employed for a more comparable result.

Source of Support: None, Conflict of Interest: None

[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]