Acute renal failure in neonatal sepsis

Abstract

Objective

To evaluate the occurrence of acute renalfailure (ARF) and the factors associated with it in cases of neonatal sepsis.

Methods

The case control study was conducted in the referral neonatal intensive care unit of a tertiary teaching hospital. 200 out born neonates with sepsis admitted to the nursery from January to July 2003 were evaluated for presence of ARF (cases) or not (controls). Sepsis was diagnosed on the basis of either a positive sepsis screen (immature: total (I:T) neutrophil ratio >0.2, μ-ESR> age in days + 2mm or>15 mm, CRP>6mg/dl, TLC<5000 cells/mm³; 2 or more positive) or a positive blood culture in symptomatic neonates. ARF was defined as blood urea nitrogen (BUN) >20mg/dl on two separate occasions at least 24 hours apart. Oliguria was defined as urine output <1ml/Kg/hr.

Results

52 out of 200 (26%) neonates with sepsis had ARF; only 15% of ARF was oliguric. The mean gestation of neonates with ARF was similar to those without ARF (36.1±4.1 wks vs. 36.6±3.5 wk; p=0.41). A significantly higher number of babies with ARF weighed less than 2500 gm as compared to those without ARF (86.5%vs 67.6%; p=0.008). The association of meningitis, disseminated intravascular coagulation (DIC) and shock was also significantly higher in neonates with ARF (46.8%vs 26.2%, p=0.01; 65.4%vs 20.3%, p<0.001; 71.2%vs 27.0%, p<0.001 respectively). Mortality in neonates who developed ARF was significantly higher (70.2%vs 25%, p<0.001). Factors including gestational age, weight, onset of sepsis, culture positivity, associated meningitis, asphyxia, shock, prior administration of nephrotoxic drugs were subjected to univariate analysis for prediction of fatality in neonates with sepsis and ARF; only shock was found to be a significant predictor of fatality (p<0.001). ARF had recovered in 22 out of 49 neonates in whom data was available; three patients had left against medical advice. The mean duration of recovery in these 22 neonates was 5.5 days (range 1–14 days). Presence of co-existing morbidities (perinatal asphyxia/congestive heart failure (CHF)/necrotising enterocolitis (NEC) or nephrotoxic drugs did not alter the frequency of recovery of ARF in septic neonates (45.5%vs 44.4%, p=0.944; 41%vs 52%, p=0.308 respectively).

Conclusion

Renal failure occurred in 26% neonates with sepsis. Although ARF in neonates has been reported to be predominantly oliguric, it was observed that ARF secondary to neonatal sepsis was predominantly non oliguric. Low birth weight was an important risk factor for the development of ARF. The mortality being three times higher in neonates with ARF demands a greater awareness of this entity among practitioners and better management of this condition.