Selective head cooling and acute kidney injury in neonates with hypoxic ischemic encephalopathy

Abstract

BACKGROUND:

The reno-protective effect of therapeutic hypothermia in infants with hypoxic ischemic encephalopathy (HIE) is still debatable. We aimed to study the effect of therapeutic hypothermia on the development and progress of acute kidney injury (AKI) in neonates with HIE.

METHODS:

Thirty full term infants with HIE were equally distributed between cooling group (selective head cooling) or non-cooling group (late presentation after 6 hours of birth). Serum creatinine, urine output (UOP), serum neutrophil gelatinase-associated lipocalin (NGAL), and serum cystatin C were measured at baseline, day 4 and day 10 of life.

RESULTS:

The incidence of AKI as per Acute Kidney Injury Network (AKIN) criteria was comparable in cooling and non-cooling groups (40% versus 53%, respectively). Serum creatinine and UOP were significantly improved on day-4 and day-10 samples compared to base-line samples in both groups regardless of cooling. Therapeutic hypothermia was associated with a significant reduction in serum NGAL, but not cystatin C, level in day-4 and day-10 samples compared to the non-cooling group. Serum NGAL and cystatin C did not show a significant decline in day-4 and day-10 samples compared to baseline samples in both the cooled and non-cooled groups indicating an ongoing AKI.

CONCLUSIONS:

Therapeutic hypothermia was associated with less renal impairment when compared to infants with HIE who were not cooled. Continuing kidney injury may persist in asphyxiated newborns despite improvement in serum creatinine and UOP.

Trial Registration Number:

NCT02683915.

Keywords

Asphyxia acute kidney injury cystatin C neutrophil gelatinase-associated lipocalin therapeutic hypothermia neonate

Abbreviations

AKI

Acute kidney injury

AKIN

Acute kidney injury network

DIC

Disseminated intravascular coagulation

HIE

Hypoxic ischemic encephalopathy

NGAL

Neutrophil gelatinase-associated lipocalin

PPHN

Persistent pulmonary hypertension of the newborn

UOP

Urine output

1. Introduction

Perinatal asphyxia is a multisystem disorder, of which AKI is a common sequel. It has been noticed that the incidence of AKI was lowered from a range of 41.7–70% in asphyxiated neonates not subjected to therapeutic hypothermia [1, 2] to a range of 31–38% among asphyxiated neonates subjected to therapeutic hypothermia [3, 4]. We assumed that this lower incidence may be related to a reno-protective effect for the cooling practice.

Observational studies on asphyxiated neonates treated with therapeutic hypothermia identified AKI as an independent predictor for abnormal brain magnetic resonance imaging [3], alongside with prolonged durations of mechanical ventilation and NICU stay [4]. Also, abnormal early renal parameters in asphyxiated neonates are significantly associated with more severe neurodevelopmental deficit at the end of first year of life [5]. Hence, early prediction of AKI in asphyxiated neonates is warranted to dictate the overall prognosis.

The probable mechanism by which therapeutic hypothermia protects against AKI in neonatal HIE is by prevention of ischemic reperfusion injury through decreased metabolic demand, reduction of free radicals production, and limitation of apoptosis [6]. Early diagnosis of AKI in neonates with HIE is challenging. Traditional tools as UOP and arbitrary serum creatinine are not helpful and proved to have many limitations. Although the current definition of AKI depends on rising levels of serum creatinine [7], it has been shown that serum creatinine may not increase until about 25–50% of renal function is lost [8]. In the context of seeking for earlier and more accurate definition of AKI, several biomarkers had been evaluated including serum Neutrophil gelatinase-associated lipocalin (NGAL) and cystatin C. NGAL is a protein covalently bound to gelatinase from human neutrophils that is expressed in renal and extra-renal tissues in response to ischemia. Two previously published studies [9, 10] had reported serum NGAL as a sensitive and early biomarker for AKI in neonates with HIE not treated with therapeutic hypothermia. Cystatin C is a protein of low molecular weight that is freely filtered through the glomeruli and totally reabsorbed by proximal tubules and hence, considered as an accurate marker of glomerular filtration rate (GFR) [11]. Treiber and colleagues found that neonates with HIE had significantly higher cystatin C levels in the umbilical cord and on the third day of life compared to healthy control [12].

We conducted this prospective observational study to evaluate if selective head cooling tends to exert a reno-protective effect and to assess the prognostic accuracy of serum NGAL and serum cystatin C to predict AKI in asphyxiated neonates.

2. Subjects and methods

2.1. Design and setting

This prospective observational study was conducted on neonates with moderate to severe HIE in of Mansoura University Children’s Hospital, between September 2015 and November 2016. Our NICU is a level III NICU which is experienced in therapeutic hypothermia and acts a referral center for management of cases with perinatal asphyxia in the whole delta region.

2.2. Included subjects

Neonates delivered at or after 36 completed weeks’ gestation with moderate to severe HIE were included. Eligibility criteria were: (1) evidence of perinatal depression with at least one of the following: 10 minute Apgar score of 5 or less; persistent need for resuscitative efforts at 10 min after birth; or severe acidosis (defined as pH less than 7.0 or a base deficit of 16 mmol/L or more in an arterial blood sample within 60 min of birth); (2) moderate/ severe encephalopathy (fulfilling at least 3 abnormal signs of 6 categories of modified Sarnat and Sarnat scoring system within the first 6 hours of age) [13].

Infants were included in the non-cooled group if they met the same eligibility criteria; however they were delivered in regional hospitals and transferred to our NICU after the first 6 hours of life.

2.3. Excluded subjects

Neonates with major congenital abnormalities, head trauma causing major intracranial haemorrhage, or birth weight less than 1800 g were excluded.

2.4. Group allocation

Enrolled infants were divided into two groups according to timing of admission to our NICU: Cooled group (neonates who were admitted and underwent therapeutic hypothermia within first 6 hours of life) or non-Cooled Group (identified and referred to us after 6 hours of life) as therapeutic hypothermia is not the standard of care after 6 hours of life.

2.5. Study procedures

Neonates in the cooled group underwent therapeutic hypothermia using Olympic Cool-Cap system (Olympic Medical, Seattle, WA, USA) to a core temperature of 33.5–34.5°C for a period of 72 hours followed by slow re-warming over a period of 8 hours at a rate not exceeding 0.5°C per hour, whereas neonates in non-cooled group were managed in incubators maintaining skin temperature at 36.5–37.2°C.

Neonates in both groups received a uniform standard intensive supportive care as regard fluid management, cardiovascular support, respiratory support, and anticonvulsant therapy. Infants were monitored for arterial blood pressure, heart rate, and respiratory rate. Hypotension was defined as need to volume expander and single inotrope to maintain mean arterial blood pressure above 40 mmHg, while severe hypotension was considered if more than 2 inotropes were needed. All infants underwent routine echocardiography to check cardiac contractility and persistent pulmonary hypertension of the newborn (PPHN). Clinical neurologic daily assessments were done applying Thompson scoring at recruitment and daily for the first week of life [14].

AKI in enrolled neonates was identified and graded using modified serum creatinine based staging of AKIN criteria [15]. Baseline serum creatinine and arterial blood gases obtained in the first 6 hours of life were collected from our data and from the data of referring hospitals for all infants. Base line blood samples for serum NGAL and cystatin C as well as urine output were obtained for all enrolled neonates at 24 hours of life. Serum creatinine was obtained at 48 hours of life for all infants to define and grade AKI. Follow up samples were collected on DOL 4 and 10 of life to measure the effect of intervention. Serum NGAL was determined using ELISA technique (Elabscience Biotechnology Company. Ltd www.elabscience.com. WuHan, PRC. Catalog No: E-EL-H0096) with a minimum detectable dose of human NGAL of 0.1 ng/ml. Serum cystatin C was determined using ELISA technique (Elabscience Biotechnology Company. Ltd www.elabscience.com. WuHan, PRC. Catalog No: E-EL-H0055) with a minimum detectable dose of human cystatin C of 3.75 ng/ml.

We collected demographic data of enrolled neonates including: gestational age, birth weight, and gender. Perinatal data included vaginal delivery, cord prolapse, birth location, Apgar score at 10 minutes, chorioamnionitis, meconium aspiration, and arterial blood gases within the first hour of life. Clinical characteristics at time of enrolment included Thompson score, Sarnat and Sarnat staging, and presence of seizures within the first 24 hours of life. Recorded events and outcomes during admission included hypotension, severe hypotension, DIC, elevated liver enzymes, culture proven sepsis, hypocalcemia, hyponatremia, PPHN, hypercoagulability, the need for transfusion of blood products, days of mechanical ventilation, length of hospital stay, and mortality before hospital discharge.

We assessed the following risk factors for mortality: initial Thompson score, Sarnat and Sarnat staging, and seizures in first day of life, AKI, PPHN, severe hypotension, culture proven sepsis, therapeutic hypothermia, serum NGAL, and serum cystatin C on the first day of life.

3. Statistical analyses

Statistical analysis was performed using SPSS statistical software (version 16; SPSS, Chicago, Illinois). Analytic comparison of continuous and categorical variables was performed using Student-t test or Mann–Whitney test and Chi-square test or Fisher’s exact test, respectively. The diagnostic accuracy of serum NGAL and serum cystatin C for identifying AKI was evaluated using receiver operating characteristics (ROC) curves. A p-value of < 0.05 is considered to be statistically significant. The dependent and independent risk factors potentially associated with in-hospital mortality were studied using binary logistic regression. The regression model included all statistically significant variables in univariate analysis plus all clinically important variables, whether or not they were statistically significant. Hosmer–Lemeshow test was used to assess the performance of the regression model. A kaplan-Meier survival analysis was conducted to compare percentage of survival between cooled and non-cooled group.

4. Results

We recruited 15 infants with HIE in the cooling group (mean gestational age of 38.6±1.4 weeks and birth weight of 3129±431 grams) and 15 infants with HIE in the non-cooling group (mean gestational age of 37.6±1.4 weeks and birth weight of 3062±239 grams). There was no difference in the demographic and baseline characteristics between the two studied groups (Table 1). Acute kidney injury was identified in 6 (40%) infants in the cooling group (5 infants had stage 1 AKI and 1 infant had stage 2 AKI) and 8 (53%) infants in the non-cooling group (4 infants had stage 1 AKI, 3 infant had stage 2 AKI, and 1infant had stage 3 AKI) with no significance between the two groups (Table 2).

Table 1
Demographic and perinatal data between the two groups

Characteristics Cooled group
n = 15 Non-cooled group
n = 15 p-value

Gestational age (weeks) 38.6±1.4 37.6±1.4 0.07

Birth weight (grams) 3129±431 3062±239 0.6

Male sex 8 (53%) 8 (53%) 1.00

Outborn 11 (73%) 15 (100%) 0.03

Vaginal delivery 6 (40%) 5 (33%) 0.7

Apgar score at 10 minutes 4 (4–6) 5 (4–6) 0.94

Chorioamnionitis 2 (13%) 3 (20%) 0.6

Meconium stained liquor 5 (33%) 2 (13%) 0.19

Cord prolapse 1 (6%) 1 (6%) 1.0

Age at recruitment (hours) 5.5 (5–6) 24 (14–24) 0.000

Thompson score at recruitment 10 (6–16) 12 (11–16) 0.17

Seizures in the first 24 hours of age 11 (73.3%) 12 (80%) 0.66

Sarnat staging

Stage II 8 (53.3%) 10 (66.7%) 0.46

Stage III 7 (46.7%) 5 (33.3%)

ABG at birth

PH 6.97±0.18 7.06±0.05 0.08

pCO2 (mmHg) 36 (27.3–52) 36.7 (21–46) 0.57

Base Deficit (mEq/L) 9.5 (7–17) 11 (7–17) 0.98

Characteristics	Cooled group n = 15	Non-cooled group n = 15	p-value
Gestational age (weeks)	38.6±1.4	37.6±1.4	0.07
Birth weight (grams)	3129±431	3062±239	0.6
Male sex	8 (53%)	8 (53%)	1.00
Outborn	11 (73%)	15 (100%)	0.03
Vaginal delivery	6 (40%)	5 (33%)	0.7
Apgar score at 10 minutes	4 (4–6)	5 (4–6)	0.94
Chorioamnionitis	2 (13%)	3 (20%)	0.6
Meconium stained liquor	5 (33%)	2 (13%)	0.19
Cord prolapse	1 (6%)	1 (6%)	1.0
Age at recruitment (hours)	5.5 (5–6)	24 (14–24)	0.000
Thompson score at recruitment	10 (6–16)	12 (11–16)	0.17
Seizures in the first 24 hours of age	11 (73.3%)	12 (80%)	0.66
Sarnat staging
Stage II	8 (53.3%)	10 (66.7%)	0.46
Stage III	7 (46.7%)	5 (33.3%)
ABG at birth
PH	6.97±0.18	7.06±0.05	0.08
pCO2 (mmHg)	36 (27.3–52)	36.7 (21–46)	0.57
Base Deficit (mEq/L)	9.5 (7–17)	11 (7–17)	0.98

Data are expressed as mean±SD, median (interquartile range) or number (%).

Table 2

Clinical course and outcomes of infants in both groups during intervention

Characteristics	Cooled groupn = 15	Non-cooled groupn = 15	p-value
Grade of AKI
0	9 (60%)	7 (46.7%)	0.5
1	5 (33.3%)	4 (26.6%)
2	1 (6.7%)	3 (20%)
3	0 (0%)	1(6.7%)
Hypotension	5 (33%)	7 (46.6%)	0.46
Severe Hypotension	3 (20%)	3 (20%)	1.0
Packed RBC transfusion	1 (6.6%)	4 (26.6%)	0.14
Platelet transfusion	2 (13%)	1 (6.6%)	0.54
DIC	1 (6.6%)	0 (0%)	0.31
Hypercoagulability	1 (6.6%)	0 (0%)	0.31
Elevated liver enzymes	11 (73%)	13 (86.6%)	0.36
Culture-proven sepsis	9 (60%)	11 (73%)	0.44
Hypocalcemia	1 (6.6%)	0 (0%)	0.31
Hyponatremia	3 (20%)	1 (6.6%)	0.28
PPHN	7 (46.6%)	5 (33%)	0.46
Duration of mechanical ventilation (days)	5 (0–9)	6 (4–11)	0.26
Length of hospital stay (days)	16 (13–22)	15 (12–18)	0.63
Mortality	6 (40%)	7 (47%)	0.71

Data are expressed as median (interquartile range) or number (%).

A cut-off baseline serum NGAL level of 190.8 ng/ml was found to have the highest sensitivity of 86%, specificity of 69%, positive predictive value of 71%, and accuracy of 77% in predicting AKI in asphyxiated infants. A cut-off baseline serum cystatin level of 1.6 mg/L was found to have the highest sensitivity of 86%, specificity of 75%, positive predictive value of 75%, and accuracy of 80% in predicting AKI in our asphyxiated infants. Serum samples for NGAL and cystatin C were available for analysis throughout the study period in the cooling group while two infants (13 %) died before day 10 samples in the non-cooling group.

Urine output (Fig. 1A) and serum creatinine (Fig. 1B) were significantly improved on the day-4 and day-10 samples compared to base-line samples in both groups without significant effect for cooling. Serum NGAL level was significantly lower in the cooling group at day-4 (127.2±53.5 versus 257.5±58.8 ng/ml, p < 0.01) and day-10 samples (126.3±29.1 versus 164.4±33.1 ng/ml, p < 0.05) when compared to the non-cooling group (Fig. 1C). There was no significant difference in the level of serum cystatin C at day 4 or day 10 samples between cooling and non-cooling groups (Fig. 1D). Serum NGAL in the Non-cooling group and cystatin C in both groups remained above the cut-off values for AKI on the day-4 samples. Serum NGAL and cystatin C did not show a significant decline in the day-4 and day-10 samples compared to base-line samples in both groups (Fig. 1).

Fig. 1.

Effect of therapeutic hypothermia on (A) urinary output, (B) serum creatinine, (C) serum NGAL, (D) serum Cystatin C at different time points. P < 0.05 compared to (*) non-cooling group, (#) baseline sample, (+) day 4 sample.

Serum NGAL level was significantly lower in infants with AKI at day-4 (144.2±51.5 versus 244.9±47.1 ng/ml, p = 0.015) and day-10 samples (127.7±36.9 versus 244.9±47.1 ng/ml, p < 0.01) when compared to baseline sample in the cooling group (Fig. 2A) but not in the non-cooling group (Fig. 2B). Serum cystatin C level was significantly lower in infants with AKI at day-4 (1.58±0.84 versus 2.88±0.88 mg/L, p = 0.042) when compared to baseline sample in the cooling group (Fig. 2C) but not in the non-cooling group (Fig. 2D). Serum NGAL and serum cystatin C levels were significantly higher in infants without AKI at day-4 sample when compared to baseline sample in the non-cooling group (257.1±82.4 versus 112.2±80.4 ng/ml, p = 0.005; and 2.88±1.23 versus 1.31±0.33 mg/L, p = 0.024 respectively) (Fig. 2C & D).

Fig. 2.

Effect of AKI on (A) Serum NGAL in cooled infants, (B) Serum NGAL in non-cooled infants, (C) Serum Cystatin C in cooled infants, (D) Serum Cystatin C in non-cooled infants at different time points. P < 0.05 compared to (*) non-AKI group, (#) baseline sample (+) day 4 sample.

We conducted a subgroup analysis to compare the course of both serum NGAL and serum cystatin C in neonates with AKI in both groups. Serum NGAL was significantly lower in infants with AKI underwent cooling at day-4 (144.2±51.5 versus 257.9±45.3, p = 0.003) and day-10 samples (127.7±36.9 versus 201.9±5.3, p = 0.026) when compared to the non-cooling group but not at baseline sample. Levels of serum cystatin C were comparable among neonates with AKI in cooling group and non-cooling group at all times. Levels of serum NGAL and cystatin C were matched among the whole cohort in neonates with culture-proven sepsis when compared to non-septic patients at all times.

There was no significant difference in the clinical course, duration of mechanical ventilation, length of hospital stay or mortality between cooling and non-cooling group (Table 2). A kaplan-Meier survival analysis did not show a statistical significant difference in the proportion of survival between infants in the cooling compared to the non-cooling group (log rank 0.42, p = 0.51).

In a univariate analysis, high level of baseline serum cystatin C and presence of PPHN were significantly associated with mortality before hospital discharge in neonates with HIE (Table 3). After implementing a multivariate analysis (Table 3), presence of PPHN was the single independent predictor associated with mortality before discharge in neonates with HIE. The observed and predicted values for the logistic regression model were matched as identified by Hosmer–Lemeshow test(p = 0.52).

Table 3

Predictor of mortality in studied infants

	Univariate analysis		Multivariate analysis
	Risk ratio (95% CI)	p-value	Risk ratio (95% CI)	p-value
Seizures at first day	0.63 (0.05–7.7)	0.74
Sarnat stage	0.36 (0.08–1.6)	0.18
Thompson score	0.89 (0.75–1.05)	0.16
AKI	0.34 (0.08–1.5)	0.16	0.49 (0.006–41.0)	0.75
Serum NGAL ng/ml (At recruitment)	0.99 (0.9–1.01)	0.72	1.0 (0.99–1.1)	0.54
Cystatin C mg/L (At recruitment)	0.44 (0.20–0.96)	0.039*	0.2 (0.05–1.3)	0.13
Severe hypotension	1.0 (0.01–1.01)	0.051	2.2 (0.17–2.2)	0.55
PPHN	0.04 (0.01–0.28)	0.001*	0.01 (0.001–0.26)	0.006*
Neonatal sepsis	1.5 (0.3–6.9)	0.60	1.03 (0.3–33.8)	0.98
Therapeutic hypothermia	0.76 (0.18–3.2)	0.71	0.22 (0.006–8.5)	0.41

5. Discussion

Hypoxic ischemic encephalopathy is commonly associated with AKI during the neonatal period. The incidence of AKI ranges from 30% to 70% in asphyxiated neonates due to variability in the parameters used to define AKI among different studies [16 –18]. Moreover, AKI has been associated with increased morbidity and mortality in asphyxiated neonates [16, 19]. Therapeutic hypothermia was initially introduced to improve survival and neurodevelopmental outcome in term neonates with hypoxic ischemic encephalopathy. More recently, it has been shown that therapeutic hypothermia has beneficial effects on other body systems beyond its neuro-protective value [20 –22]. We aimed to study the effect of therapeutic hypothermia on the development and progress of AKI in term infants born with HIE.

We found that baseline serum NGAL level of more than 190.8 ng/ml and baseline serum cystatine C level of more than 1.6 mg/L had high sensitivity and specificity for detecting AKI in neonates with HIE. Closer to our findings, previous studies have shown that a serum NGAL level of more than 140.7 ng/ml after the first 24 hours of life and serum cystatin C level of 1.69 mg/l in the first 3 days of life had the highest sensitivity (88.9% and 82% respectively) and specificity (95% and 94% respectively) for predicting AKI in asphyxiated neonates [12, 23]. Furthermore, the previous two studies have indicated that serum NAGL and cystatin C are more sensitive parameters for diagnosing kidney injury in asphyxiated neonates compared to serum creatinine.

The main finding of our study is that therapeutic hypothermia has a protective effect on acute kidney injury as evident by a significant drop in serum NGAL (Fig. 1C), but not cystatin C (Fig. 1D), on the 4th day and 10th day of life in all recruited infants with HIE, as well as in subgroup of infants defined to have AKI, treated with 72 hours of brain cooling compared to control infants. We also found that serum NGAL level dropped significantly on day 4 and day 10 of life compared to baseline in neonates defined to have AKI whom have been treated with therapeutic hypothermia (Fig. 2A) but not in control neonates with AKI (Fig. 2B). Moreover, neonates who apparently had no AKI, based on their serum creatinine level, had a significant rise in their serum NGAL level on the 4th day of life compared to baseline when not subjected to therapeutic hypothermia (Fig. 2B). Furthermore, serum cystatin C level dropped significantly on day 4 of life compared to baseline in neonates with AKI treated with therapeutic hypothermia (Fig. 2C) but not in control neonates with AKI (Fig. 2D).

In a rat model of ischemic reperfusion injury, hypothermia (33°C) was associated with the lowest risk of kidney injury compared to normothermia (37°C) or hyperthermia (40°C) [24]. In adults exposed to hypothermia during coronary artery bypass surgery, rapid re-warming to 37°C was associated with higher incidence of kidney injury compared to gradual re-warming to 34°C as measured by serum creatinine level and creatinine clearance [25, 26]. In a meta-analysis of nineteen trials including 2218 patients with cardiac arrest, hypothermia was found to decrease the risk of mortality but has no effect on the incidence of kidney injury or the need for dialysis [27]. However, this meta-analysis had limitations of including adult patient’s only, small sample size trials, and lacked a homogenous definition for AKI among patients included.

Previous report has shown no difference in decreasing the incidence of AKI when selective head cooling was compared to total body cooling in asphyxiated neonates [28]. Furthermore, a meta-analysis of Tagin et al., demonstrated that the risk of mortality or major neurodevelopmental disability was reduced by both modalities significantly when compared to normothermia [29].In a randomized control design, 120 term neonates with HIE were equally divided to receive either whole body therapeutic hypothermia using cool gel packs or standard care aiming to study the effect of therapeutic hypothermia on the incidence of AKI as measured by the level of serum creatinine and fractional excretion of sodium. Authors reported a lower incidence of AKI in the therapeutic hypothermia compared to the standard group (32% versus 60%, p < 0.05) together with lower incidence of mortality (26%, versus 60%, p < 0.05) [21]. Though, authors of the previous study did not report a significant decline in serum creatinine over time in hypothermia treated group compared to standard group.

An important observation in our results was the persistent high level of serum NGAL in non-cooled infants and cystatin C in cooled and non-cooled infants throughout the three measured samples despite the apparent improvement in urinary output and decline in serum creatinine. This is of particular interest as it may indicate an ongoing kidney injury for a longer time than expected using the current definitions of AKI which depends on changes in serum creatinine level. Our observation highlights the necessity of considering the proper time for administration of nephrotoxic drugs and the proper duration for dose adjustment if ongoing kidney injury is still suspected.

Although the value of NGAL and cystatin C has been proved to be sensitive and specific predictor for early detection of AKI, their value in estimation of long term outcomes such as the need for renal replacement therapy, development of chronic kidney disease, and mortality is still questionable. In a longitudinal follow up study, 174 children who had AKI and survived to hospital discharge were followed for 5 years outcome. Authors reported that in spite of the initial improvement of kidney function, more than 50 % of children had signs of CKD by 3 to 5 years of life and mortality was as high as 20% most of them (65%) died within 12 months after the initial hospitalization [30].

We found no statistically significant differences in levels of serum NGAL and cystatin C between neonates with and without culture-proven sepsis at all times. Similar to our finding, two previous studies found no significant difference in levels of serum NGAL [31] or urinary cystatin [32] among critically ill neonates with and without sepsis. In contrast Smertka et al. demonstrated that septic neonates had significantly higher serum and urinary NGAL levels compared to non-septic group [33]. To our knowledge, no previous studies evaluated the impact of sepsis upon NGAL and cystatin C neonates with HIE.

We had only one case that required peritoneal dialysis in the non-cooling group and none of our cases was discharged with impaired renal function. We did not find a significant effect for brain cooling in decreasing infants’ mortality which may be attributed to small sample size. Furthermore, no significant effect of AKI on predicting mortality was elicited. A univariate analysis of our variables revealed that high baseline cystatin C level and persistent pulmonary hypertension of a newborn were significantly associated with increased mortality before hospital discharge. In a logistic regression model, persistent pulmonary hypertension of the newborn was the only predictor for mortality after controlling for other variables. However, these results have to be interpreted with caution as our study was neither designed nor powered to study such outcomes. It will be warranted to design a multi-center study on an adequate sample of infants with HIE who develop AKI to evaluate the role of NGAL and Cystatin C as early biomarkers for prediction of mortality and long term renal morbidity.

To our knowledge, this is the first study to use serum NGAL and cystatin C as sensitive biomarkers for kidney injury to assess the effect of therapeutic hypothermia on AKI in neonatal HIE. We acknowledge the following limitations of our study: (1) being a case control study on a relatively small sample size, however conducting a large randomized control trial may not be feasible plus being unethical as therapeutic hypothermia has become evidence based standard of care for these infants; (2) the lack of randomization between groups; (3) all infants in the non-cooled compared to 73% of infants in the non-cooled group were delivered in regional hospitals which may lead to variation of initial management between both groups.

In conclusion, therapeutic hypothermia is associated with less risk for AKI in infants with HIE. Continuing kidney injury may persist in asphyxiated newborn despite improvement in serum creatinine and UOP as evident by persistent elevation of serum NGAL and cystatin C.

Conflict of interest

The authors have no conflicts of interest relevant to this article to disclose.

Funding

No funding was secured for this study

Financial Disclosure

The authors have no financial relationships relevant to this article to disclose.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The principle investigator obtained an informed consent from the parents of all infants included in the study upon admission. The study was approved by the Institutional Review Board of Mansoura Faculty of Medicine.

Footnotes

Acknowledgment

We would like to thank the parents and families in the Neonatal Intensive Care Unit for accepting to participate in this study.

Compliance with ethical standards

References

Aggarwal

, Kumar

, Chowdhary

, Majumdar

, Narang

Evaluation of renal functions in asphyxiated newborns. J Trop Pediatr 2005;51:295–9.

Kaur

, Jain

, Saha

, Chawla

, Parmar

, Basu

, et al. Evaluation of glomerular and tubular renal function in neonates with birth asphyxia. Ann Trop Paediat. 2011;31:129–34.

Sarkar

, Askenazi

, Jordan

, Bhagat

, Bapuraj

, Dechert

, et al. Relationship between acute kidney injury and brain MRI findings in asphyxiated newborns after therapeutic hypothermia. Pediatr Re. 2013;75:431–5.

Selewski

, Jordan

, Askenazi

, Dechert

, Sarkar

Acute kidney injury in asphyxiated newborns treated with therapeutic hypothermia. J Pediatr 2013;162:725–729.e1.

Hadzimuratovic

, Skrablin

, Hadzimuratovic

, Dinarevic

SM.

Postasphyxial renal injury in newborns as a prognostic factor of neurological outcome. J Matern Neonatal Me. 2014;27:407–10.

Polderman

KH.

Mechanisms of action, physiological effects, and complications of hypothermia. Crit Care Med 2009;37:S186–202.

Jetton

, Askenazi

DJ.

Update on acute kidney injury in the neonate. Curr Opin Pediat. 2012;24:191–6.

Sweetman

DU.

Neonatal acute kidney injury – Severity and recovery prediction and the role of serum and urinary biomarkers. Early Hum Dev 2017;105:57–61.

Raggal

, Khafagy

, Mahmoud

, Beltagy

SE.

Serum neutrophil gelatinase-associated lipocalin as a marker of acute kidney injury in asphyxiated neonates. Indian Pediat. 2013;50:459–62.

10.

Sarafidis

, Tsepkentzi

, Agakidou

, Diamanti

, Taparkou

, Soubasi

, et al. Serum and urine acute kidney injury biomarkers in asphyxiated neonates. Pediatr Nephro. 2012;27:1575–82.

11.

Bagshaw

, Bellomo

Cystatin C in acute kidney injury. Curr Opin Crit Care 2010;16:533–9.

12.

Treiber

, Gorenjak

, Pecovnik Balon

Serum cystatin-C as a marker of acute kidney injury in the newborn after perinatal hypoxia/asphyxia. Ther Apher Dial 2014;18:57–67.

13.

Shankaran

, Laptook

, Ehrenkranz

, Tyson

, McDonald

, Donovan

, et al. Whole-Body Hypothermia for Neonates with Hypoxic-Ischemic Encephalopathy. N Engl J Me. 2005;353:1574–84.

14.

Thompson

, Putermana

, Linley

, Hann

, van der Elst

, Molteno

, et al. The value of a scoring system for hypoxic ischaemic encephalopathy in predicting neurodevelopmental outcome. Acta Paediat. 1997;86:757–61.

15.

Koralkar

, Ambalavanan

, Levitan

, McGwin

, Goldstein

, Askenazi

Acute kidney injury reduces survival in very low birth weight infants. Pediatr Res 2011;69:354–8.

16.

Agras

, Tarcan

, Baskin

, Cengiz

, Gürakan

, Saatci

Acute renal failure in the neonatal period. Ren Fail 2004;26:305–9.

17.

Alaro

, Bashir

, Musoke

, Wanaiana

, Bashir

Prevalence and outcomes of acute kidney injury in term neonates with perinatal asphyxia. Afr Health Sci 2014;14:682–8.

18.

Gupta

, Sharma

, Bagla

, Parakh

, Soni

, Pramod

Renal Failure in Asphyxiated Neonates. Indian Pediatr 2005;42:928–34.

19.

Chevalier

, Campbell

, Brenbridgea

Prognostic factors in neonatal acute renal failure. Pediatrics 1984;74:265–72.

20.

Joy

, Pournami

, Bethou

, Bhat

, Bobby

Effect of therapeutic hypothermia on oxidative stress and outcome in term neonates with perinatal asphyxia: A randomized controlled trial. J Trop Pediatr 2013;59:17–22.

21.

Tanigasalam

, Bhat

, Adhisivam

, Sridhar

MG.

Does therapeutic hypothermia reduce acute kidney injury among term neonates with perinatal asphyxia? – a randomized controlled trial. J Matern Neonatal Med Taylor & Franci. 2016;29:2545–8.

22.

, Denault

, Dupuis

, Vadeboncoeur

, Harel

, Arsenault

, et al. A pilot study: The Noninvasive Surface Cooling Thermoregulatory System for Mild Hypothermia Induction in Acute Myocardial Infarction (The NICAMI Study). Am Heart. 2005;150:933.

23.

Baumert

, Surmiak

, Więcek

, Walencka

Serum NGAL and copeptin levels as predictors of acute kidney injury in asphyxiated neonates. Clin Exp Nephrol 2017;21:658–64.

24.

Delbridge

, Shrestha

, Raftery

, El Nahas

, Haylor

JL.

The Effect of Body Temperature in a Rat Model of Renal Ischemia-Reperfusion Injury. Transplant Proc 2007;39:2983–5.

25.

Boodhwani

, Rubens

, Wozny

, Nathan

HJ.

Effects of Mild Hypothermia and Rewarming on Renal Function After Coronary Artery Bypass Grafting. Ann Thorac Sur. 2009;87:489–95.

26.

Massoudy

, Wagner

, Thielmann

, Herold

, Kottenberg-Assenmacher

, Marggraf

, et al. Coronary artery bypass surgery and acute kidney injury – Impact of the off-pump technique. Nephrol Dial Transplan. 2008;23:2853–60.

27.

Susantitaphong

, Alfayez

, Cohen-Bucay

, Balk

, Jaber

BL.

Therapeutic hypothermia and prevention of acute kidney injury: A meta-analysis of randomized controlled trials. Resuscitatio. 2012;83:159–67.

28.

Atici

, Celik

, Gulasi

, Turhan

, Okuyaz

, Sungur

MA.

Comparison of selective head cooling therapy and whole body cooling therapy in newborns with hypoxic ischemic encephalopathy: Short term results. Türk Pediatr Arşiv. 2015;50:27–36.

29.

Tagin

, Woolcott

, Vincer

, Whyte

, Stinson

DA.

Hypothermia for Neonatal Hypoxic Ischemic Encephalopathy. Arch Pediatr Adolesc Me. 2012;166:558–66.

30.

Askenazi

, Feig

, Graham

, Hui-Stickle

, Goldstein

SL.

3–5 Year Longitudinal Follow-Up of Pediatric Patients After Acute Renal Failure. Kidney In. 2006;69:184–9.

31.

El-Farghali

, El-Raggal

, Mahmoud

, Zaina

GA.

Serum Neutrophil Gelatinase-associated Lipocalin as a predictor of acute kidney injury in critically-ill neonates. Pakistan J Biol Sc. 2012;15:231–7.

32.

, Li

, Zhou

, Yan

, Zhu

, Pan

, et al. Impact of sepsis on the urinary level of interleukin-18 and cystatin C in critically ill neonates. Pediatr Nephro. 2013;28:135–44.

33.

Smertka

, Wroblewska

, Suchojad

, Majcherczyk

, Jadamus-Niebroj

, Owsianka-Podlesny

, et al. Serum and urinary NGAL in septic newborns. Biomed Res In. 2014;2014:717318.