Significant difference between control group (healthy infants) and pre and full-term newborns treated with gentamicin groups ^* P<0.05; ^** P<0.01; ^*** P<0.001. Significant difference between zero day and other days of gentamicin administration of pre and full-term newborn groups ^a P<0.05; ^b P<0.01; ^c P<0.001 Abbreviations: FE Na% (fractional excretion of sodium), LDH (lactate dehydrogenase) and AP (alkaline phosphatase)

 Materials and Methods

Patients and healthy subjects

We identified 29 participants of a total of two hundred participants admitted between February 2009 and September 2011 to the Pediatric Ward of Alborz Hospital, Karaj, Iran. Using the below criteria (5), fourteen pre-term newborns (8 males and 6 females) with gestational age of 32–36 weeks and fifteen full-term newborns (8 males and 7 females) with gestational age of 38–40 weeks who were treated for suspected infections with gentamicin (including neonatal sepsis, pneumonia, prolonged rupture of membrane, or meconium aspiration syndrome) in the first week of life, were enrolled. As the control group, fifteen healthy infants (in the first week of life with gestational age 38–42 weeks) were included in this study. Briefly, inclusion criteria were infants with normal respiratory and metabolic stability through blood gas analysis (pH: 7.25–7.45, pO₂:50–70 mm Hg, pCO₂:≤55 mmHg) and exclusion criteria were history of prenatal asphyxia or shock, renal impairment [Normal Glomerular Filtration Rate (GFR) is 17–60 ml/min/1.73 m²], hypotension, use of other potentially nephrotoxic drugs, and polycythemia (hematocrit level of greater than 65%). The administered gentamicin dose was 1.5 mg/kg/8 hr intravenously over a 30 min period via an intravenous drip infusion in a total volume of 5 ml of 5% dextrose solution over a period of 5 days. The parental informed consent was obtained before study and the study protocol was approved by the Research Committee of Social Security Organization Hospital, Karaj, Iran.

 Collection of urine and blood samples

Urine and blood specimens were obtained from the subjects before and after gentamicin infusion on the 1^st, 3^rd, and 5^th days of treatment. Urine specimens were obtained over a 2 to 3 hr period using adhesive bags. To obtain serum, venous blood samples were collected into test tubes without any anticoagulant. Urine and blood samples were centrifuged (2000×g for 10 min) immediately after collection.

 Biochemical laboratory parameters

All chemicals used were of analytical grade. PNP (ρ- nitrophenol) & PNP-NAG (ρ- nitrophenyl -N-acetyl-β-D-glucosaminide) were obtained from Sigma Chemical co., St. Louis, Mo, USA. All other required chemicals were purchased from Merck, GmbH, Germany. NAG activity in urine was measured at 37^oC, according to the described method (7). Briefly, the reaction mixture consisted of ρ- nitrophenyl-N-acetyl-β-D-glucosaminde (PNP) as substrate in 25 mmol of acetate, or 100 mmol of citrate (pH 4.8) incubated with 0.025 ml of urine and a photometer (Model ECOM-E 6125, Eppendorf Inc. Germany) at 405 nm.

Serum and urinary creatinine and urinary urea, AP, LDH, and microalbuminuria were measured by using commercially available assay kits (Pars Azmun Co. Ltd, Iran) by Jaffe, Urease-GLDH, DGKG, DGKG, and the immunoturbidometry method respectively. All these tests were measured with the Hitachi 902 autoanalyser (Hitachi Ltd., Japan). Sodium was determined using flame photometry (Model EFOX 5054, Eppendorf Inc. Germany). FE Na (%) expresses the fractional excretion of the sodium, which is calculated as follows:

FE (Na) = Urine Na × Serum Cr / Serum Na × Urine Cr.

To correct for variations in urine flow, the urinary enzyme activities (U/l) were normalized to urinary creatinine concentration (g/l) and given as U/g creatinine.

 Statistical analysis

Data are expressed as the mean ± standard deviation (SD). The difference between different days was analyzed using one way analysis of variance (ANOVA) followed by Student-Newman-Keuls test. Statistical significance was accepted at            P <0.05 for all comparisons and in all correlations.

 Results

Table 1 depicts the routine biochemical parameters of newborns treated with gentamicin and the control group (Healthy Infants). For serum creatinine, blood urea nitrogen, and urinary albumin no differences between the control group and treated groups were detected. In pre-term newborns, a statistically significant increase in FE Na (%) was noted on the 3^rd and 5^th days of therapy compared with control group (P<0.01 and P<0.01; respectively). Moreover, in full-term newborns a statistically significant increase was observed in             FE Na (%) the 5^th day of therapy compared with before gentamicin administration (P<0.01). During gentamicin treatment, a statistically significant increase in urinary LDH was noted in both pre-term  and full-term newborns on the 5^th day compared with the control group (P<0.001 and P<0.001; respectively) as well as before gentamicin administration in pre-term and full-term newborns compared with 5^th day of therapy (P<0.001 and P<0.001; respectively). Additionally, a statistically significant increase in urinary AP was noted in both pre-term and full-term newborns on the 5^th day compared with the control group (PP<0.01; respectively). Finally, in pre-term newborns a statistically significant increase was observed in urinary AP on the 5^th days compared with before gentamicin administration (P<0.01). Furthermore, in full-term newborns a statistically significant increase was observed in urinary NAG on 3^rd and 5^th days compared with before gentamicin administration (P<0.05, P<0.001 and P<0.001; respectively).

As shown in Figure 1, a statistically significant decrease was observed in urinary NAG on first (185±60), 3^rd (75±30), and 5^th (43±5) days compared with before gentamicin administration in pre-term newborns (250 ± 55; P<0.01, P<0.001, and P<0.001; respectively). In contrast, in full-term newborns a statistically significant increase was observed in urinary NAG on first (29±2), 3^rd (35±3), and 5^th (61±4) days compared with before gentamicin administration (25±4; P<0.05, P<0.001, and P<0.001; respectively). During gentamicin treatment, a statistically significant increase in urinary NAG was observed in the full-term newborns on the 5th day of therapy (61±4) compared with the control group (26.5±2.5; P<0.05). Additionally, in the pre-term newborns a statistically significant increase in urinary NAG was observed before (250±55) and on first (185± 60) and third (75±30) days after gentamicin administration compared with control group (26.5±2.5; P<0.001, P<0.001, and P<0.001; respectively).

 Figure 1. Mean values ± SD of urinary NAG (Units per gram Creatinine) in pre and full-term infants as well as control groups (Healthy Infants) on days 1, 3, and 5 after gentamicin administration. Time zero is defined as the time of gentamicin administration

Significant difference between control group (Healthy Infants) and pre and full-term newborns treated with gentamicin groups             ^* P<0.05; ^** P<0.01; ^*** P<0.001

Significant difference between zero day and other days of gentamicin administration of pre and full-term newborn groups ^a P<0.05; ^b P<0.01; ^c P<0.001

 Discussion

In pediatric clinical practice, the most commonly used conventional blood parameters of kidney injury assessment remain serum creatinine and urea (3). Our study, however, found no significant differences in these markers between neonatal patients treated with gentamicin and control group. On the                 other hand, a previous study reported that aminoglycosides nephrotoxicity caused non-oliguric renal failure with a slow rise in serum creatinine (8). This contradicts other studies which speculate that gentamicin-induced nephrotoxicity, which occurred in proximal tubules and serum creatinine, is a marker of glomerular filtration in renal function         (7, 9). Moreover, Tugay et al (10) showed that gentamicin can cause a significant increase in the urine albumin ratio during treatment, which is contrary to our data. We attribute this difference between results to study design and the sample size.

As shown in Figure 1, urinary NAG in the full-term newborns may be useful for evaluation of minor effects on tubular function. These results are consistent with other studies showing that urinary NAG can act as indicator of renal damage in early stages (7, 11). In contrast, a previous study demonstrated a 10-day gentamicin treatment in neonates without any changes in NAG activity (12). The difference between our results and the mentioned study may be due to the study design and differences in laboratory methods used for measurement of biomarkers. Furthermore, clinical studies have demonstrated that gentamicin treatment in pre-term neonates results in increased urinary NAG and claim that it could be used for the assessment of antibiotic-induced renal injury in these patients (13, 14). In the present study, however, we found that urinary NAG was decreased during gentamicin administration. Possibly, activity of urinary NAG in pre-term neonates reflects the maturity of renal tubular epithelium (15), therefore, urinary NAG may be used for evaluation of kidney maturity after birth. Overall, the difference between above studies may partly be due to the differences in study design, sample size, dose administration, length of hospitalization, and other confounding factors.

Urinary electrolyte excretion in sick newborns depends on nutrition, electrolyte balance, and the rate of diuresis as well as the arterial blood pressure; and since sick newborns are usually salt-losers and at risk of developing a negative Na balance (5), adjustment of normal electrolyte balance is difficult. Since evaluation of renal function through urinary electrolyte excretion has the previously mentioned limitations, the increase in FE Na% during treatment with gentamicin in both groups, was not considered a definite biomarker of renal dysfunction in our study.

 As a final remark, the results of our study also showed that in both full and pre-term newborns, urinary LDH and AP increased on the 5th day of gentamicin administration. Since AP is a brush–border membrane enzyme with high sensitivity for tubular damage while LDH is found in most parts of the nephron, it can be assumed that these urinary enzymes may be helpful in renal injury detection in the early stages.

Limitations of this study are due to the small patient sample size, the untreated pre-term newborn group, and ROC curve analysis. Thus, future studies with larger sample size and ROC curve analysis with sensitivity and specificity of all urinary enzymes are mandatory to confirm the gentamicin induced renal injury in this population.

 Conclusion

In conclusion, the most obvious finding to emerge from this study is that urinary NAG as well as urinary LDH and AP may be useful particularly in full-term newborns for evaluation of tubular function when monitoring of serum gentamicin is not possible. In addition, in pre-term infants urinary LDH and AP may be helpful as biomarkers for evaluation of renal insufficiency.

 Acknowledgment

We are grateful to the Hospital Manger as well as Pediatric Unit staff at Alborz Hospital, Karaj, Iran, the participants, and their families