European Journal of Pediatrics

, Volume 167, Issue 5, pp 509–515

Hypercalciuria is the main renal abnormality finding in Human Immunodeficiency Virus-infected children in Venezuela

Authors

  • Corina Gonzalez
    • Departments of Pediatric Infectology and Pediatric Nephrology, Hospital Doctor Enrique TejeraUniversity of Carabobo
    • Division of Kidney Diseases, Children’s Memorial Hospital, Feinberg School of MedicineNorthwestern University
    • Children’s Memorial Hospital, Division of Kidney Diseases
  • C. B. Langman
    • Division of Kidney Diseases, Children’s Memorial Hospital, Feinberg School of MedicineNorthwestern University
  • P. Zibaoui
    • Departments of Pediatric Infectology and Pediatric Nephrology, Hospital Doctor Enrique TejeraUniversity of Carabobo
  • L. Escalona
    • Departments of Pediatric Infectology and Pediatric Nephrology, Hospital Doctor Enrique TejeraUniversity of Carabobo
  • L. F. Dominguez
    • Departments of Pediatric Infectology and Pediatric Nephrology, Hospital Doctor Enrique TejeraUniversity of Carabobo
  • M. A. Rosas
    • Departments of Pediatric Infectology and Pediatric Nephrology, Hospital Doctor Enrique TejeraUniversity of Carabobo
Original Paper

DOI: 10.1007/s00431-007-0538-4

Cite this article as:
Gonzalez, C., Ariceta, G., Langman, C.B. et al. Eur J Pediatr (2008) 167: 509. doi:10.1007/s00431-007-0538-4

Abstract

Kidney involvement in children with Human Immunodeficiency Virus (HIV) infection is increasing in prevalence in parallel with the longer survival of HIV-infected patients and the side-effects of new antiretroviral drugs. However, there are only a few reports describing renal tubular disorders in HIV+ children. This is a cross-sectional, case series study evaluating kidney disease in 26 Venezuelan HIV-infected children. The study cohort consisted of 15 girls and 11 boys, with a median age of 5.9 years (25–75th percentile: 3.6–7.8), who had been treated with antiretrovirals for 2.8 ± 0.4 years, Overall, the patients were short for their age and gender (Z-height: −3.1; 25–75th percentile: −4.94 to −1.98), and 15 showed signs of mild to moderate malnutrition. All of the children had a normal estimated glomerular filtration rate (136 ± 22.6 ml/min/1.73 m2), and glomerular involvement was only observed in one patient with isolated proteinuria. None had nephromegaly. In contrast, tubular disorders were commonly found. Hypercalciuria was detected in 16 of the patients (UCa/Cr = 0.28; 25–75th percentile: 0.17–0.54 mg/mg), with five of these showing crystalluria. Eight children showed hyperchloremia, and three had frank metabolic acidosis. Kidney stones were absent in all, but one boy had bilateral medullary nephrocalcinosis. Conclusion, in Venezuelan children, HIV infection per se, or its specific treatment, was commonly associated with renal tubular dysfunction, especially hypercalciuria and acidosis, potentially leading to nephrocalcinosis and growth impairment. We recommend renal tubular evaluation during the follow-up of children with HIV infection.

Keywords

AcidosisHIV NephropathyHypercalciuriaKidney stonesTubular disorders

Introduction

Human Immunodeficiency Virus-associated nephropathy (HIVAN) was first described in HIV-infected adults with heavy proteinuria, large edematous kidneys and rapid progression to end stage kidney disease (ESKD) [48]. HIVAN was found to occur tenfold more frequently in young African-American males [23], who were mostly affected by focal and segmental glomerulosclerosis (FSGS) [42]. In HIV-infected children, HIVAN is a late event [2], causing nephrotic syndrome [51, 60] with FSGS, mesangial hyperplasia, or tubulo-reticular inclusions appearing in renal biopsy [15, 60]. There is a distinct trend of increased survival in children with HIV, and, consequently, a greater frequency of kidney disease (>40% in USA), with a larger number of HIV children progressing to ESKD [2]. However, kidney involvement is heterogeneous, and, in addition to HIVAN, acute tubular dysfunction, glomerular disease or HIV-associated thrombotic microangiopathy have been described [50].

There are but a few reports describing tubular dysfunction in HIV-infected patients, despite commonly observed microcystic tubular dilatation, acute tubular necrosis or interstitial lesions in kidney biopsies [11]. Kidney damage may be the result of tissue infiltration by lymphocytes, plasma cells or monocytes [50], heparin binding growth factors or chemokine recruitment [11], or due to the HIV viral expression per se [31, 49]. Opportunistic infections and associated malnutrition affect the kidney as well [17, 18]. Further, lactic acidosis, kidney stones, and renal tubular injury are known to be associated with specific antiretroviral therapies [55].

In Venezuela, as in many other developing countries, there has been an unfortunate increase of HIV infection in children. This study was carried out to discern the presence of HIV-related renal disease in a group of HIV-infected children treated at “Dr. Enrique Tejera Hospital” in Valencia, Venezuela.

Population and methods

Patients

During a 10-month period, 26 patients were recruited from of 32 HIV-infected children seen at the “Dr. Enrique Tejera Hospital.” Six patients refused to consent and were therefore not enrolled in the study. Overall, these subjects represented half of the total number of children diagnosed as HIV+ in the State of Carabobo (Venezuela). The participants in the study met the following inclusion criteria: age between 1–14 years, treatment with antiretroviral drugs >3 months, and parental-guardian's consent to participate. At the time of the study, the patients had received a specific treatment for a mean of 2.84 ± 0.4 years in different combinations: (1) reverse transcriptase inhibitors: zidovudine (ZDV; 22 patients), didanosine (ddl, 15 patients), lamivudine (3TC,11 patients), stavudine (d4T, three 3 patients); (2) non-nucleoside reverse transcriptase inhibitors: efavirenz (EFV, three patients); (3) protease inhibitors: nelfinavir (NFV, 20 patients), amprenavir (APV, three patients; lopinavir/ritonavir (one patient). Two other children were also treated with gancyclovir and another one with co-trimoxazole. The viral load was undetectable (<400 copies/mm3) in four children, intermediate (400–100,000 copies/mm3) in 16, and high (>100,000 copies/mm3) in six. In all but one patient, HIV was transmitted vertically.

Design

This was a cross-sectional study within a cohort of HIV-infected children.

Data collection

Information regarding medical history, co-infections, physical exam and treatment was collected from patients, guardians, and subjects’ charts. Nutritional evaluation was based on Waterlow’s formula [65] and the Venezuelan growth references of FUNDACREDESA [22].

Laboratory methods

Blood samples were processed with Bayer Express plus 560 equipment. Biochemistry data were obtained using the following techniques: urea by Talke’s and Schubert’s method [61], creatinine by modified Jaffe’s method [57], chloride by the thiocyanate colorimetric method [28], total calcium, based on the α-cresolphthalein complexone and calcium reaction, visualized with colorimeter [35], phosphate, by the Daly and Ertingshausen methodology [14], and uric acid by the Trinder methodology [47]. The plasma anion gap was calculated by [Na - (Cl+CO2)] formula (normal values: 7–16) [52]. Venous gases were performed simultaneously with the same blood specimen (ABL 5 gas meter, Radiometer, Copenhagen).

First-morning voided urine samples were analyzed by optical microscopy and reactive dipsticks. In addition, creatinine was quantified by the modified Jaffe’s method [57], and urinary total protein using a Spectronic 20 photocolorimeter (the Lowry method) [53]. The protein to creatinine ratio (UProt/UCr in mg/mg) was calculated [52]. Second-morning voided urine samples were tested for creatinine, calcium, uric acid, phosphate, sodium, potassium, chloride, and calcium. A urinary ratio of calcium/creatinine (UCa/Cr) ≥0.21 mg/mg was indicative of hypercalciuria [29, 39, 45]. The urinary anion gap was calculated in cases of metabolic acidosis [3], and the glomerular filtration rate estimated (eGFR) by the Schwartz formula [56].

Kidney ultrasound was performed in all subjects by a single radiologist and Z-scores of kidney length determined, based on published standards [54].

Data analysis

Parametric data were described as means and standard deviations (SD) and non-parametric data as median (md) and percentiles (25th and 75th).

Results

26 HIV-infected children, 15 girls and 11 boys, with a median age of 5.9 years (25th–75th percentiles: 3.6–7.8 years) were studied. As a group, these children were short for their age and gender (Z-height = −3.1; 25–75th percentile: −4.94 to −1.98), but the weight was more appropriate (Z-weight : −1.02; 25–75th percentile: −1.33 to −0.37], based on contemporary Venezuelan FUNDACREDESA standards [22]. Mild or moderate malnutrition was observed in 15 of the subjects by the Waterlow’s formula [65] (Table 1). Other characteristics of these patients, such as CDC Pediatric HIV classification [8] and existing co-infections, are shown in Table 1.
Table 1

Patient characteristics at time of study

 

n

Nutritiona

 Normal

11

 Mild malnutrition

11

 Moderate malnutrition

4

HIV stageb

 A1

7

 A2

1

 B1

4

 B2

2

 B3

5

 C3

7

Co-infectionsc

 HVB

2

 HVA

2

 EBV

1

 CMV

1

 Toxoplasmosis

1

 Tuberculosis

1

aWaterlow’s formula [65] based on the Venezuelan growth references of FUNDACREDESA [22]

bSee Schafer et al. [54]

cHVB, Hepatitis B virus; HVA, hepatitis A virus; EBV, Epstein-Barr virus; CMV, cytomegalovirus

Renal functional parameters are summarized in Table 2. All subjects had a normal eGFR (136 ± 22.6 ml/min/1.73 m2). Mild proteinuria (UProt/UCr = 0.7 mg/mg) in one child (patient 10) was the only marker of glomerular damage in the entire study group.
Table 2

Selected renal functional parameters in 26 HIV-infected children

Parametera

Mean ± SD or Median (25–75th percentiles)

P Cr (mg/dl)

0.44 ± 0.07

eGFR (ml/min/1.73 m2)

136 ± 22.6

P Na (mEq/L)

139 ± 6.4

P K (mEq/L)

4 ± 0.38

P Cl (mEq/L)

109.5 (106-113)

P pH

7.36 ± 0.04

P Bicarbonate (mEq/L)

25.3 ± 2.99

FENa%

0.62 (0.25–1.13)

UCa /Cr (mg/mg)

0.28 (0.17–0.54)

UProt/Cr (mg/mg

0.04 (0.02–0.11)

aP, Plasma; Cr, creatinine; eGFR, estimated glomerular filtration rate; FENa%, fractional excretion of sodium (%); U, urine; Prot, protein)

Two children, patients 7 and 9 (stage C3 and stage B3, respectively), had a mild metabolic acidosis, with a bicarbonate of 21 mEq/L, normal plasma anion gap, and negative urine anion gap; whereas in one patient (no. 5), who was stage C3 and received AZT + 3TC + amprenavir, non-renal metabolic acidosis (pH: 7.33, bicarbonate: 19 mEq/L) with increased plasma anion gap (+17) was observed. Other electrolyte disorders were not uncommon: hypernatremia (two patients), hyponatremia (three), and hyperchloremia (eight). None of the children had hypokalemia or hypophosphatemia.

Further evaluation of the renal tubular function (Table 3) demonstrated isolated hypercalciuria in 16/26 HIV-infected children, irregardless of the disease stage (five, A1; one, A2; two, B1; one, B2; four, B3; three, C3). Five hypercalciuric patients (cases 2, 3, 8, 9, 19) had associated calcium-oxalate crystals in the urine, whereas patient 21, also hypercalciuric, presented with microscopic hematuria. All three children with metabolic acidosis also presented hypercalciuria, with high urinary sodium in two. Hyperphosphaturia (TRP <75%) was observed in patient 3, in addition to hypercalciuria. In the entire group of patients, urinary sodium excretion did not correlate significantly with UCa/Cr ratio (p = 0.28), but four cases (cases 7, 9, 18, 21) had increased natriuresis associated with hypercalciuria (Table 3).
Table 3

Tubular disorders in 26 HIV-infected children. Abnormal values are given in bold (Ca-Ox calcium oxalate)

Case

Stage

Treatmenta

P Cr (mg/dl)

Urine

Acidosis

FENa%

U Ca/Cr (mg/mg)

U Prot/Cr (mg/mg)

1

B3

AZT+ddI+Nelfinavir

0.49

Normal

No

0.19

0.18

0.02

2

A1

AZT+ddI+Nelfinavir

0.44

Ca-Ox crystals, Microhematuria

No

0.81

0.44

0.15

3

C3

AZT+ddI+Nelfinavir

0.30

Ca-Ox crystals

No

0.1

0.78

0.01

4

B2

AZT+ddI+Nelfinavir

0.43

Normal

No

0.38

0.49

0.01

5

C3

AZT+3TC+Amprenavir

0.42

Normal

Yesb

0.49

1.47

0.04

6

C3

AZT+3TC

0.49

Normal

No

0.06

0.18

0.17

7

C3

AZT+ddI+Nelfinavir

0.46

Normal

Yesc

4.6

1.38

0.11

8

B1

AZT+ddI+Amprenavir

0.60

Ca-Ox crystals

No

0.46

0.56

0.04

9

B3

AZT+ddI+Nelfinavir

0.40

Ca-Ox crystals

Yesd

1.86

1.02

0.02

10

A1

AZT+ddI+Nelfinavir

0.35

Leukocyturia

No

0.4

0.44

0.69

11

A1

AZT+Nelfinavir+3TC

0.36

Normal

No

0.15

0.30

0.05

12

A1

AZT+ddI+Nelfinavir

0.45

Normal

No

0.08

0.08

0.02

13

B1

Nelfinavir+d4T

0.57

Normal

No

1.13

0.01

0.12

14

B1

ddI+Nelfinavir+d4T

0.48

Normal

No

0.63

0.25

0.16

15

C3

AZT+3TC+Efaviez

0.40

Normal

No

0.65

0.03

0.11

16

A1

AZT+ddI+Nelfinavir

0.44

Normal

No

0.25

0.18

0.02

17

A1

AZT+ddI+Nelfinavir

0,35

Normal

No

0.33

0.44

0.07

18

B3

AZT+ddI+Nelfinavir

0.46

Normal

No

1.34

0.26

0.02

19

B3

Nelfinavir+3TC+d4T +Efaviez

0.40

Ca-Ox crystals

No

0.68

0.54

0.04

20

C3

AZT+3TC+Amprenavir

0.37

Normal

No

0.89

0.03

0.11

21

A2

AZT+Nelfinavir+3TC

0.48

Microhematuria

No

1.48

0.43

0.04

22

A1

AZT+ddI+Nelfinavir

0.45

Normal

No

0.35

0.87

0.03

23

C3

3TC+Efaviez+Lopinavir/Ritonavir

0.54

Normal

No

1.13

0.17

0.02

24

B2

AZT+Nelfinavir+3TC

0.52

normal

No

1.42

0.12

0.06

25

B3

AZT+Nelfinavir+3TC

0.48

Normal

No

0.62

0.24

0.01

26

B1

AZT+Nelfinavir+3TC

0.36

Normal

No

0.12

0.07

0.05

P, Plasma; Cr, creatinine; eGFR, estimated glomerular filtration rate; FENa%, fractional excretion of sodium (%); U, urine; Prot, protein

aAZT, Zidovudine; ddl, didanosine; 3TC, lamivudine; d4t, stavudine

bBlood: pH, 7.33; bicarbonate, 19; anion-gap, +17. Urine: pH, 6

cBlood: pH, 7.37; bicarbonate, 21; anion-gap, +1. Urine: pH, 5; anion-gap, −10.3

dBlood: pH, 7.28; bicarbonate, 21; anion-gap, +10. Urine: pH, 5; anion-gap, −53

Kidney ultrasound examinations revealed preserved cortico-medullary differentiation and the absence of renal calculi in all subjects. One patient had medullary nephrocalcinosis. Further, nephromegaly was not observed and, indeed, HIV-infected children, as a group, had smaller kidneys that paralleled the reduction in their body mass (Z-score right kidney length = −1.25 ± 1; z-score left kidney length = −1.23 ± 1.1).

Discussion

Despite commonly reported HIV-nephropathy in infected populations [2, 51, 60], our children with HIV disease had preserved glomerular function – with a high frequency of isolated hypercalciuria – even when compared to healthy Venezuelan children [32]. Idiopathic Hypercalciuria (IH) is also the most frequent underlying disorder in adults [10] and children who are kidney stone formers [30]. Although we did not observe any renal calculi in our pediatric HIV patients, HIV-infected stone-former adults have been described with prominent hypercalciuria and increased saturation of calcium oxalate in their urine [37]. In parallel, this difference in the nephrolithiasis pattern between children and adults is also characteristic of populations with IH [59].

The fact that our subjects were fasting at the time of the urine collection does not support the absorptive origin of hypercalciuria in them [4]. Further, most of the study children did not demonstrate an exaggerated natriuresis, a well-known factor producing or worsening hypercalciuria [41]. Thus, it is more likely that direct HIV infection per se or its specific treatment (HAART) was involved in the elevation of urinary calcium excretion in our patients. Alternatively, recurrent systemic infections, which are known to elevate circulating cytokines that, in turn, lead to excessive bone turnover and release of calcium (resorptive hypercalciuria), may have been responsible [25, 40].

HIV-specific treatments have been linked with kidney stone disease. Crystalluria is commonly seen in HIV-infected adults treated with indinavir sulphate, a HIV protease inhibitor [16, 63, 64]. This drug produces persistent leukocyturia and crystalluria in children and adults [6, 15] and characteristically precipitates to form deposits inside the cortical and medullary collecting ducts [26, 33], causing nephrolithiasis [16]. Most of our patients were not treated with indinavir. However, nelfilavir, used in 77% of our patients, has also been reported to crystallize [20], and both nelfinavir and indinavir are proven components of kidney stones in patients on protease inhibitor therapy [20]. Thus, it is reasonable to assume that nelfinavir caused leukocyturia in patient 10 and may be related with the commonly seen crystals in urine our group.

Chakraborty et al. [9], noted persistent acidosis in 34 of 202 HIV+ children in advanced stages of the disease. Of these, 16 had an elevated serum anion gap acidosis, while the other 18 had normal serum anion gap acidosis with positive urinary anion gap (suggesting distal RTA). Acidotic subjects in their series were more immunocompromised than our contemporary population of HIV-infected children and, in addition, they often received co-trimoxazole [9]. In our study, patient 5, who received AZT + 3TC + amprenavir, had a non-renal metabolic acidosis picture with an increased plasma anion gap (Table 3), possibly secondary to viral reverse transcriptase inhibitors [5]. We were unable to measure lactate levels in this patient to confirm a potential mitochondrial dysfunction caused by those drugs [7], which is the postulated mechanism leading to this subtype of metabolic acidosis in HIV-infected subjects [24, 34].

We also observed two patients with mild metabolic acidosis who were in the B3 and C3 stages of HIV infection and received AZT + ddI + nelfinavir (Table 3). Their plasma anion gap was normal, whereas the urinary anion gap was negative, suggesting the presence of a proximal renal tubular acidosis (RTA). Unlike some investigators, who observed generalized renal tubular dysfunction in HIV+ adults [58] due to antiretroviral-induced tubulo-interstitial damage [19, 46], we did not find hypokalemia, hypophosphatemia, glycosuria or other markers of complex proximal tubular dysfunction in this study. The lack of distal RTA in our series may be explained by the uncommon use of sulfur/sulfone-containing antibiotics in the contemporary management of our patients [36]. Blood sodium and chloride disturbances in some of our patients may be related to multiple factors, including malnutrition [62].

Based on the advanced stages of HIV infection in many of our patients, we expected to find glomerular involvement [50]. Despite reports of this condition in the literature [43, 44], typical HIVAN with enlarged kidneys and glomerular signs was virtually absent in our group. Only one child had mild proteinuria in a random first morning-voided urine, a method validated to quantify proteinuria in children with HIV nephropathy [1]. Ethnic differences between our Venezuelan population (a mix between European, mostly Spaniard, black African and Venezuela’s native Indians) and those reported by others may, in part, explain our findings. In the USA, HIVAN is commonly seen in minority populations, mainly in African Americans [12, 21, 27] with south Saharan ancestors [13], although not exclusively [38].

Although less recognized than glomerular disease, tubular abnormality in the form of hypercalciuria was common in our HIV-infected group of children and has reported to be common in other studies. While we are uncertain of its natural history, such metabolic disturbances may be responsible for an increased risk of nephrolithiasis, and may possibly add to the burden of their somatic growth impairment in the future. Further studies need to be performed to investigate the role of bone loss in the hypercalciuria of an HIV-infected pediatric population. Since the lifespan of HIV-infected children is increasing, we recommend routine screening of renal tubular function in the regular follow-up of this population.

Copyright information

© Springer-Verlag 2007