Menoufia Medical Journal

ORIGINAL ARTICLE
Year
: 2016  |  Volume : 29  |  Issue : 2  |  Page : 198--208

Study of urinary neutrophil gelatinase-associated lipocalin as a predictor of acute kidney injury in postliver transplant recipients


Hassan Abdel-Hady1, Said S Khamis1, Mahmoud A Kora1, Emad F Abdel-Halim2, Wael M Abdel-Razek3, Mohamed H Badr1,  
1 Department of Internal Medicine, Menoufiya Univerity, Menoufiya, Egypt
2 Department of Clinical Pathology, Faculty of Medicine, Menoufiya Univerity, Menoufiya, Egypt
3 Department of Hepatology, National Liver Institute, Menoufiya Univerity, Menoufiya, Egypt

Correspondence Address:
Mohamed H Badr
Department of Internal Medicine, Faculty of Medicine, Menoufiya University, Shebin El Kom, Menoufiya, 32511
Egypt

Abstract

Objective: The aim of this work is to study the value of urinary neutrophil gelatinase-associated lipocalin (u-NGAL) as an early predictor for the occurrence of acute kidney injury (AKI) in patients undergoing liver transplantation (LT) in comparison with the SCr and RIFLE i (Risk, Injury, and Failure; and Loss; and End-stage kidney disease) classification. Background: Renal dysfunction is common after LT. The incidence of AKI complicating the post-transplant period varies between 48 and 94% and affects both short-term and long-term outcomes. Mortality in those requiring renal replacement therapy may be as high as 40% at 90 days, increasing to 54% at 1 year. Materials and methods: This study was carried on 30 hepatic patients who underwent LT in the National Liver Institute, Menoufiya University, Egypt, in the period from June 2012 to May 2014. Thorough assessment of history, clinical examination and preoperative routine investigations, and baseline renal functions including assessment of GFR by renal scan were performed. Operative data were collected during LT; patients were also assessed 5 days postoperatively for the occurrence of AKI by SCr, urine output and RIFLE classification, and u-NGAL. Results: Fourteen of 30 patients (46.7%) fulfilled the criteria of the RIFLE classification and had AKI; in these patients, there was a significant relation between the RIFLE classification and the cause of LT, preoperative platelet count, the use of basiliximab in the induction of immunosuppression, and day 1, day 2 u-NGAL. u-NGAL levels above the cut-off value of 1300 pg/ml at day 1 and 4440 pg/ml at day 2 are considered good predictors of AKI post-LT, with AUROC 0.77 and 0.77, respectively, with an acceptable accuracy of 66.7 and 73.3%, and high sensitivity and specificity for prediction of AKI post-LT, and correlate significantly with different preoperative and operative parameters. Conclusion: u-NGAL is a valuable marker for early detection of AKI in patients undergoing LT before an increase in SCr occurs. Levels above the cut-off value (1300 pg/ml) have a high sensitivity, specificity, and positive predictive value for AKI after LT.



How to cite this article:
Abdel-Hady H, Khamis SS, Kora MA, Abdel-Halim EF, Abdel-Razek WM, Badr MH. Study of urinary neutrophil gelatinase-associated lipocalin as a predictor of acute kidney injury in postliver transplant recipients.Menoufia Med J 2016;29:198-208


How to cite this URL:
Abdel-Hady H, Khamis SS, Kora MA, Abdel-Halim EF, Abdel-Razek WM, Badr MH. Study of urinary neutrophil gelatinase-associated lipocalin as a predictor of acute kidney injury in postliver transplant recipients. Menoufia Med J [serial online] 2016 [cited 2024 Mar 28 ];29:198-208
Available from: http://www.mmj.eg.net/text.asp?2016/29/2/198/192432


Full Text

 Introduction



Renal dysfunction is common after liver transplantation (LT). The incidence of acute renal failure complicating the post-transplant period varies between 48 and 94% and affects both short-term and long-term outcomes. Mortality in those requiring renal replacement therapy (RRT) may be as high as 40% at 90 days, increasing to 54% at 1 year [1].

The difficulties in early intervention contribute significantly toward the poor prognosis of acute kidney injury (AKI) [2]. Serum creatinine, the commonly used marker for renal injury, is slow and insensitive. It is a marker of function and not injury and cannot detect subtle injuries or delineate a single cause of renal injury. Serum creatinine is diluted by large-volume transfusions and may take days after injury to increase. Therefore, an increase in serum creatinine reflects the sum of injuries in the preceding days and cannot identify the effect of a single renal insult or subtle differences between groups of patients [3].

Measurement of serum creatinine remains the most common diagnostic tool for assessment of renal function, but this long-used marker shows insensitivity to small changes in renal function, lacks specificity for renal function, given the influence exerted by a variety of non-renal factors (such as muscle mass and diet), and has a long lag time in response to kidney injury [4]. The limitations of creatinine-based estimates of renal function have probably limited the success of AKI therapeutic trials and the opportunity to alter the therapeutic strategy accordingly (e.g. delaying the initiation of nephrotoxic immunosuppression after transplant in patients with AKI) [4].

A new class of molecular biomarkers has shown significant potential in other clinical settings. These biomarkers include serum cystatin C, urine interleukin 6 (IL-6), serum IL-6, urine IL-18, and urinary neutrophil gelatinase-associated lipocalin (u-NGAL). These molecules have been shown to be early detectors of AKI in a variety of clinical settings such as critical illness [5].

In particular, NGAL is emerging as an excellent biomarker in the urine and plasma, for the early prediction of AKI, for monitoring clinical trials in AKI, and for the prognosis of AKI in several common clinical scenarios. However, biomarker combinations may be required to improve our ability to predict AKI and its outcomes in a context-specific manner [6]. Elevated plasma NGAL (pNGAL) and u-NGAL even without an increase in serum creatinine have been suggested to identify patients with subclinical AKI and patients with an increased risk for RRT, prolonged stay in the ICU, and adverse outcomes [7].

 Aim of the Work



The aim of this work is to study the value of measurement of u-NGAL as an early predictor for the occurrence of AKI in patients undergoing LT versus serum creatinine and urine output (components of the RIFLE classification).

 Patients and Method



This study was carried on 30 hepatic patients undergoing orthotopic LT in the National Liver Institute, Menoufiya University, Egypt, in the period from June 2012 to May 2014. They ranged in age from 2 to 58 years (10 pediatrics, 20 adults); there were 24 males and six females in the study. All of them underwent living donor LT after proper cross matching between donors and recipients. A written consent was obtained from every patient and the study was approved by the local authority of the Institute.

Inclusion criteria

Hepatic patients who underwent LT during the study period were included in this study.

Exclusion criteria

Patients with hepatorenal syndrome, advanced extrahepatic malignancies, autoimmune diseases, end-stage renal diseases, patients with a history of nephrectomy, CKD patients stage 3 or more [glomerular filtration rate (GFR)£60 ml/min], and patients with glomerulopathies were excluded from this study.

Preoperative data

History

The etiology of the liver disease, the cause of LT, either cirrhosis (defined by liver biopsy Metavir score F4) or hepatocellular carcinoma (HCC) (diagnosed by imaging techniques and a-fetoprotein), a history of chronic pre-existing renal disease, diabetes mellitus, and hypertension were the main factors we examined during the study.

Clinical examination

All the patients were examined for manifestations of liver decompensation (ascites, jaundice, splenomegaly, etc.), blood pressure, in addition to a thorough medical examination.

Laboratory investigations

CBC, serum albumin, prothrombin concentration, INR, total bilirubin, direct bilirubin, alkaline phosphatase, a fetoprotein, and baseline renal functions were assessed the day before the operation and they included serum creatinine and blood urea nitrogen (BUN).

GFR measured by 99m Tc-DTPA. Renal scintography (the most accurate method to assess renal function) was also performed before the operation and patients with CKD stage 3 or more (GFR≤60 ml/min) were excluded from the study.

Twenty-four urinary protein excretion was measured for all patients using a Beckman Coulter AU analyzer Beckman Coulter, Inc. California, USA), and patients with 24 h-urinary protein excretion more than 300 mg were excluded from the study to avoid enrollment of patients with glomerulopathies.

From the results of preoperative laboratory investigations, the preoperative MELD score was calculated for adult patients and the pediatric end-stage liver disease score was calculated for pediatric patients using specific equations [8].

Operative data

All the patients had undergone living donor LT, which included three surgical phases: the preanhepatic, anhepatic, and neohepatic phases. During the anhepatic phase, none of our patients needed portocaval shunt either by venovenous bypass or by the Piggy-back technique as the duration of the anhepatic phase did not exceed 3 h.

The length of surgery ranged from 8 to 19 h. The estimated blood loss (measured by the amount of blood suctioned during the operation and the number of soaked pieces of gauze and towels used during the operation), amount of blood transfusion, amount of blood products transfused, and the number of intraoperative medications administered (diuretics, terlipressin, and basiliximab) were assessed during the operation.

Postoperative data

The patients were transferred to the ICU, with regular monitoring of vital signs, intravascular volume, the level of consciousness, and signs of blood loss.

On the first postoperative day, urine output in ml/kg/h, serum creatinine, BUN, and u-NGAL were all measured 6 h after the operation. In one case, the urine sample was obtained 7 h after the operation because the patient was anuric. The urine samples were collected aseptically in sterile containers and then any impurities were removed by centrifugation. The samples were then stored at −20°C to measure u-NGAL.

On the second postoperative day, 24 h after the previous samples were collected, urine output, BUN, and serum creatinine were measured in the same manner and urine samples were stored after centrifugation for the measurement of day 2 u-NGAL.

After the collection of urine samples for the measurement of u-NGAL, they were rewarmed and the test was carried out using the typical human NGAL ELISA kit (Boster Immunoleader, BioPorto Diagnostics A/S, Tuborg Havnevej 15, st., 2900 Hellerup, Denmark) by fourth-generation ELISA.

Then, on day 5, urine output, serum creatinine, and BUN were measured again in the same manner.

During the ICU stay, the patients were observed for any indication for hemodialysis (i.e. RRT) as follows:

Biochemical indications

Refractory hyperkalemia more than 6.5 mmol/l.Serum urea more than 200 mg/dl.Refractory metabolic acidosis (pH≤7.1).Refractory electrolyte abnormalities: hyponatremia or hypernatremia, and hypercalcemia.

Clinical indications:

Urine output less than 0.3 ml/kg for 24 h or absolute anuria for 12 h.AKI with primary graft nonfunction or multiple organ failure.Refractory volume overload.End organ damage: pericarditis, encephalopathy, neuropathy, myopathy, and uremic bleeding.Creation of intravascular space for plasma and other blood product infusions and nutrition.

Then, patients were classified into two groups according to the RIFLE classification: RIFLE-positive (14 patients) and RIFLE-negative (16 patients) groups [9]. Data were analyzed first for all patients as one group and then reanalyzed after the differentiation of adult patients (20 patients) from pediatric patients (10 patients).

Statistical analysis

Results were collected, tabulated, and statistically analyzed using an IBM personal computer and the statistical package SPSS (version 17; SPSS Inc., Chicago, Illinois, USA).

Two types of statistics were calculated

Descriptive statistics

Number (N), percentage (%), mean (X), and SD were determined.

Analytic statistics

Qualitative data were analyzed using χ2, and whenever one cell of the expected was equal to or less than 5, Fisher's exact test was used. Quantitative data were analyzed using the t-test (Student's test) for comparison of quantitative variables among two independent groups and the Mann–Whitney U-test (nonparametric test) for comparison between two groups that were not normally distributed.

Significance level (P) value

A P value greater than 0.05 was considered NS and a P value of up to 0.05 was considered significant (S).

 Results



Fourteen of 30 patients had developed AKI according to the RIFLE classification of AKI [9]. Eight (57.1%) of these patients were at risk stage, four patients (28.6%) at injury, and two patients (14.3%) at the failure stage.

Their mean age was 37.1 ± 22.9 years. Twelve were males and two were females, and their mean body weight was 62 ± 34.8 kg; four (28.6%) of these patients had diabetes. The complete demographics and laboratory data of the enrolled patients are shown in [Table 1].{Table 1}

There was an insignificant difference between RIFLE positive and negative patients in age, sex, body weight, diabetes, and the etiology of liver disease, whereas there was a significant difference between RIFLE positive and negative patients in the cause of LT (P = 0.001). In six patients (42.9%) of the RIFLE (positive) group, the cause of LT was liver cirrhosis and in eight patients (57.1%), it was HCC. There was a significant difference between both groups in the platelet count (P = 0.017), and the mean platelet count in RIFLE (positive) patients was 77 ± 39 × 103/cm 2.

Also, the administration of basiliximab intraoperatively was found to significantly reduce the incidence of AKI, where overall 14 patients (46.7%) had received the drug, 10/14 (71.4%) did not develop AKI and 4/14 (28.6%) developed AKI (P = 0.01).

[Table 1] also shows that there was a significant difference in day 1 u-NGAL, day 2 urine output, and day 2 u-NGAL (P = 0.028, 0.008, and 0.029, respectively); the mean u-NGAL was 4158.6 ± 3426.9 pg/ml on day 1 in the RIFLE positive group, and 6785.7 ± 4104.9 pg/ml on day 2; the mean urine output on day 2 was 1.1 ± 0.6 ml/kg/h.

[Table 2] shows that in adult patients (n = 20), 10/20 (50%) were RIFLE positive and had AKI and there was a statistically significant difference between RIFLE positive and negative groups in the cause of LT, with the highest incidence among patients with HCC (80%), followed by liver cirrhosis (20%), whereas transplanted patients with HCC and portal vein thrombosis had no AKI (P = 0.001).{Table 2}

Also, there was a statistically significant difference between both groups in the platelet count, prothrombin concentration, alkaline phosphatase, a-fetoprotein, and preoperative BUN (P<0.05).

However, there was a statistically insignificant difference between both groups in the presence of diabetes, the etiology of liver disease, preoperative hemoglobin level, white blood cells count, albumin, total bilirubin, creatinine, GFR, and MELD score (P > 0.05).

[Table 2] also shows that the duration of operation was significantly longer in RIFLE positive patients; patients with AKI had significantly greater blood loss during the operation. Plasma transfusion was significantly lower in RIFLE positive patients and basiliximab use also significantly decreased the incidence of AKI (P<0.05).

The intraoperative amount of blood transfusion and terlipressin use did not differ significantly between the two groups (P > 0.05).

[Table 2] also shows that there was a significant difference in day 1 u-NGAL and day 2 u-NGAL, whereas there was an insignificant statistical difference between both groups in urine output and serum creatinine on day 1, day 2, and day 5 (P > 0.05).

[Table 3] shows the demographic, preoperative, intraoperative, and postoperative data in relation to AKI in pediatric patients (n = 10) and it shows that 4/10 (40%) fulfilled the RIFLE criteria and had AKI (RIFLE positive group) and 6/10 (60%) were RIFLE negative. It also shows that there was a statistically significant difference between the two groups in the platelet count and total bilirubin, and the pediatric end-stage liver disease score was significantly higher in the RIFLE positive group (P<0.05). However, sex, body weight, preoperative creatinine, and GFR were statistically insignificant between both groups.{Table 3}

The intraoperative parameters of the pediatric patients showed a statistically insignificant difference between both groups. [Table 3] also shows that there was a statistically significant difference between pediatric RIFLE positive and negative patients in the day 1 u-NGAL level and the mean u-NGAL level was higher in RIFLE positive patients, day 2 u-NGAL, day 1 urine output, and day 2 urine output.

However, there was a statistically insignificant difference between RIFLE positive and negative patients in day 1 serum creatinine, day 2 serum creatinine, and day 5 serum creatinine (P > 0.05).

[Figure 1] shows that In RIFLE positive patients, the mean u-NGAL on day 1 was 4158.6 pg/ml and it increased to 6785.7 pg/ml on day 2, whereas in RIFLE negative patients, the mean u-NGAL on day 1 was 1795 pg/ml and it increased to 3634 pg/ml on day 2.{Figure 1}

[Figure 2] shows the mean serum creatinine level in the postoperative period and it shows that, in RIFLE positive patients, the mean serum creatinine level was 0.6 mg/dl preoperatively and increased by 50% from the baseline to 0.9 mg/dl on the first postoperative day, remained at 0.9 mg/dl on the second postoperative day, and decreased to 0.7 mg/dl on the fifth postoperative day. However, in RIFLE negative patients, the mean SCr level was 0.6 mg/dl initially, increased to 0.8 mg/dl on the first postoperative day, decreased to 0.7 mg/dl on the second postoperative day, and remained at 0.7 mg on the fifth postoperative day.{Figure 2}

[Table 4] and [Table 5] indicate that D1 u-NGAL shows a significant negative correlation with the platelet count (r = −0.42, P = 0.02), preoperative GFR measured by renogram (r = −0.58, P = 0.03), intraoperative plasma transfusion (r = −0.42, P = 0.02), and postoperative urine output on day 1 (r = −0.58, P = 0.001); also, D1 u-NGAL showed a significant positive correlation with body weight (r = 0.42, P = 0.02) and total bilirubin (r = 0.38, P = 0.03). However, D2 u-NGAL showed a significant positive correlation with the level of total bilirubin (r = 0.34, P = 0.04) and a significant negative correlation with the postoperative urine output on day 1 (r = −0.39, P = 0.033) and day 2 (r = −0.35, P = 0.03).{Table 4}{Table 5}

[Table 5] shows that D1 u-NGAL was statistically highly significant in diabetic patients (P = 0.000); also, D2 u-NGAL was statistically significant in these patients (P = 0.001). There was a statistically significant difference in the mean value of D2 u-NGAL among the different causes of LT (P = 0.01) and the mean value was higher in patients with HCC (7795 ± 4082 pg/ml).

[Figure 3] shows the validity of D1 u-NGAL and D1 serum creatinine in the prediction of AKI in day 1 after LT and shows that D1 u-NGAL at a cut-off value of 1300 pg/ml – the value at which the test yields the highest sensitivity and specificity – has a sensitivity and specificity of 71.4 and 62.5%, respectively, in the prediction of AKI after LT, with AUC (0.77), positive predictive value (PPV) (62.5%), and negative predictive value (NPV) (71.4%) with an acceptable accuracy of 66.7% (P = 0.013).{Figure 3}

However, D1 serum creatinine at a cut-off value of 0.28 mg/dl had a sensitivity and specificity of 57.1 and 62.5%, respectively, in the prediction of AKI after LT, with AUC (0.62), PPV (57.1%), and NPV (71.4%) with a lower accuracy of 22.2% (P = 0.28).

[Figure 4] shows the validity of day 2 u-NGAL and serum creatinine in the prediction of AKI after LT and it shows that D2 u-NGAL at a cut-off value of 4440 pg/ml has a sensitivity and specificity of 71.4 and 75%, respectively, in the prediction of AKI after LT, with AUC (0.77), PPV (71.4%), and NPV (75%) with a higher accuracy of 73.3% (P = 0.013), but D2 serum at a cut-off value of 0.635 mg/dl with a sensitivity and specificity of 71.4 and 50%, respectively, in the prediction of AKI after LT, with AUC (0.554), PPV (55.6%), and NPV (66.7%) with a lower accuracy of 25% (P = 0.618){Figure 4}

 Discussion



Renal dysfunction is common after LT. The incidence of acute renal failure complicating the post-transplant period varies between 48 and 94% and affects both short-term and long-term outcomes [1]. This study aimed to evaluate u-NGAL as an early novel predictor of AKI in the post-LT period.

Our decision to use u-NGAL was made primarily on the basis of how poor a marker of renal dysfunction serum creatinine is in patients with cirrhosis and how increases rises may reflect significant loss of functional nephrons [9].

We studied 30 hepatic patients (20 adults and 10 pediatrics) who had undergone live donor LT.

We defined renal failure in our study by the mean of the RIFLE classification, whereas Umbro et al. [0] used the AKIN criteria and eGFR used the MDRD formula for the diagnosis of AKI. Barri et al. [1] carried out a study on 1050 patients who underwent LT using changes in serum creatinine from baseline as the main marker for AKI. They used three different definitions to diagnose post-LT AKI. Defining AKI as an increase in serum creatinine of more than 0.5 mg/dl resulted in the highest incidence of AKI (78%). The second definition of AKI was an increase in serum creatinine of more than 1 mg/dl and this resulted in an incidence of AKI 46%. When AKI was defined as an increase of serum creatinine of more than 50% from baseline to above 2 mg/dl, the lowest incidence of AKI (14%) was found [1]; however, Niemannet al. [2] had defined renal failure by an increase in serum creatinine more than 50% of baseline and eGFR using the Cockroft–Gault formula.

In the present study, we found that AKI had developed in 14/30 patients (46.7%); in the adult group, the incidence of AKI was 10/20 (50%) whereas in pediatric patients, the incidence of AKI was 4/10 (40%). Wagener et al. [2] found that 37/92 patients (40.2%) developed AKI, Niemann et al. [2] found that 27/59 patients (46%) developed AKI, and Portal et al. [3] found that 30/95 patients developed AKI (31.6%) after excluding patients undergoing RRT before transplantation.

We found that the incidence of AKI was significantly related to the cause of LT, with the highest incidence in patients with HCC (57.1%), but it was insignificant in other studies [11],[12],[13],[14].

This was in agreement with the result of Umbro et al. [0], who reported that a high MELD score (an index of the most severe degree of ESLD among patients awaiting OLT) was statistically significant between patients who developed early acute renal dysfunction and others, but in our study, the MELD score was insignificant ([Table 2]); this may have been because the patients enrolled in our study had much lower MELD scores than those in other studies as the scores in our patients ranged from 8 to 20.

Iglesiaset al. [4] found that the platelet count was statistically highly significant between patients with AKI and others; our result was in agreement with theirs as the platelet count also showed a statistically significant difference between patients with and without AKI, and patients with AKI showed lower platelet counts (mean = 77 ± 39 × 103/cm 2) than AKI-negative patients. This may be because of the higher portal venous pressure in these patients, but data about on portal vein pressure assessment were not available in our study.

Portal et al. [3] and Umbro et al. [0] found that serum bilirubin also differed significantly between patients with AKI and others, but we found that this was insignificant in adults ([Table 2]) and significant in the pediatric group ([Table 3]) taking into consideration that most of the children who underwent transplantation because of cholestatic liver disease had high levels of serum bilirubin.

The studies of Pawarode et al. [5] and Campbell et al. [6], and other studies reported that preoperative renal dysfunction is an important predictor of post-LT AKI; also, Yalavarthy et al. [7] observed that preoperative creatinine more than 1.5 mg/dl was predictive of the need for postoperative RRT and also the risk of postoperative infection. In our study, we could not evaluate these opinions as none of our patients had serum creatinine more than 1.5 mg/dl and no patients enrolled in the study needed RRT because of the small number of patients in our study; also, preoperative serum creatinine, BUN, and GFR (by renogram) were statistically insignificant between AKI patients and others in both pediatric and adult groups.

Interestingly, in this study, we found that the use of basiliximab in induction of immunosuppression early on the day of the operation (it is administered in two doses: the first within 2 h of the start of the transplant operation and the second 4 days after the transplant) may significantly decrease the risk of AKI post-LT as in the AKI group, 4/14 (28.6%) patients had been administered the drug (P = 0.01) ([Table 1]) and these results are clear in adult patients as only 2/10 (20%) of the AKI group had been administered the drug ([Table 2]). Thus, basiliximab-containing protocols are recommended to delay the use of calcineurin-containing immunosuppression in patients with borderline renal functions and thus delay calcineurin nephrotoxicity.

All the previous studies studied intraoperative data as a strong factor that can affect the postoperative renal outcome. O'Riordan et al. [8], Poratl et al. [3], and Wenger et al. [3] found that the operative duration (the duration of anesthesia) did not differ significantly in AKI patients and others. Our results are not in agreement with theirs as we found that the duration of operation in the AKI group was significantly longer than that in the other group, especially in adult patients (P = 0.028), whereas in pediatric patients, the duration of operation was insignificant [3,13,18].

O'Riordan et al. [8], Portal et al. [3], and Wagener et al. [2] also found that intraoperative blood loss showed an insignificant statistical difference between patients with and without AKI, whereas Wenger et al. [3] found a statistically significant difference between them. The results of the present study are in agreement with the results of Wenger as we found that the amount of blood loss was higher in patients with AKI (P = 0.05) [3,12,13,18].

O'Riordan et al. [8], Portal et al [3], and Wenger et al. [3] found that the amount of RBCs transfusion was nonsignificant whereas Iglesias et al. [4]found that it was significant. In the present work, we also found it to be insignificant (P = 0.125) [3,13,14,18].

O'Riordan et al. [8] and Wenger et al. [3] also found that intraoperative plasma transfusion showed an insignificant relation to the occurrence of AKI after LT, whereas Iglesias et al. [4] found a highly significant difference between patients with and without AKI in the intraoperative FFP transfusion. In this study, the results are in agreement with the results of Iglesias as we found that patients without AKI had received more units of FFP, whereas in patients with AKI had received less amount of FFP, especially adult patients ([Table 2]). This may be attributed to the fact that correction of prerenal factors can significantly decrease the incidence of AKI [3,14,18].

Wagener et al. [2]measured the discriminatory power of urinary NGAL/urine creatinine ratios to predict AKI as indicated by the area under the curve of the ROC curve (AUROC = 0.800), which was very good 3 h after reperfusion of the liver graft, with a sensitivity equal to 83.5% and specificity equal to 67.5%, and good 18 h later, indicated by AUROC (0.636), with sensitivity equal to 68.1% and specificity equal to 59.7%.

In our study, we measured the validity of u-NGAL alone in predicting AKI after LT in the first 48 h postoperatively and we found that D1 u-NGAL was considered useful at a cut-off value of 1300 pg/ml, with a high sensitivity and specificity (71.4 and 62.5%, respectively), in the prediction of AKI after LT, with AUROC (0.77), PPV (62.5%), and NPV (71.4%) with an acceptable accuracy of 66.7% (P = 0.013) ([Figure 2]). D2 u-NGAL is considered very useful at a cut-off value of 4440 pg/ml, with a high sensitivity and specificity (71.4 and 75%, respectively), in the prediction of AKI after LT, with AUROC (0.77), PPV (71.4%), and NPV (75%) with a higher accuracy of 73.3% (P = 0.013) ([Figure 3]).

Also, in our study, we found that the levels of D1 and D2 u-NGAL showed statistically significant difference in patients with AKI than other patients as well as the D1 urine output, whereas serum creatinine and urine output (the main component of RIFLE classification) in both D1 and D2 showed no statistically significant difference between them ([Table 2]). U-NGAL in our study showed a significant positive correlation with body weight and total serum bilirubin level, which confirms the injurious effect of bilirubin on the kidney [12],[13].

Niemann et al. [2], Portal et al. [3], and Wagener et al. [2] all studied u-NGAL and its role in the prediction of AKI after LT, but they did not study the factors that may influence or correlate with the levels of u-NGAL in AKI patients; for example, we found that diabetic patients enrolled in our study showed a highly significant difference in the levels of D1 and D2 u-NGAL than non diabetics, which may be because of the effect of diabetes on the kidney or other factors, as it is known that in diabetic nephropathy, urinary NGAL is increased not because of increased secretion but because of less reuptake in proximal tubular cells [9].

Also, we found that there was a statistically significant difference in the mean value of D2 u-NGAL among different causes of LT (P = 0.01) and that the mean value was higher in patients with HCC (7795 ± 4082 pg/ml) ([Table 5]). NGAL expression has also been associated with certain neoplasias and is implicated in the metastasis of breast cancer [0], but with HCC, further researches are needed.

We assume that the NGAL that we measured derived from distal nephrons as a response to renal ischemia–reperfusion injury and not from decreased reuptake (or systemic secretion from the liver graft), but further studies are required to elucidate the origin of urinary NGAL after LT.

The use of intraoperative hemodialysis has been described to be useful during OLT cases with impaired renal function or kidney failure by several centers. Sedra and Strum [1] performed over 140 cases of OLT using intraoperative hemodialysis with zero percentage intraoperative mortality and 48 h postoperatively.

Hong et al. [2] attempted an infusion of terlipressin at a dose of 1–4 µg/kg/h in the early postoperative period and compared the renal outcome with the control group, and found that patients who had terlipressin (n = 21) showed an excellent renal outcome and none of the patients developed AKI. In our study, among the patients who had received terlipressin (n = 18), 10 developed AKI and the remaining eight did not develop AKI.

Logically, u-NGAL in our study shows a significant negative correlation with the preoperative GFR and postoperative urine output.

 Conclusion



u-NGAL is a valuable marker for the early detection of AKI in patients undergoing LT before an increase in serum creatinine occurs. At a cut-off value of 1300, it has a high sensitivity and specificity and PPV for AKI after LT.

Conflicts of interest

There are no conflicts of interest.[22]

References

1Lima EQ, Zanetta DM, Castro I, Massarollo PC, Mies S, Machado MM, Yu L. Risk factors for development of acute renal failure after liver transplantation. Ren Fail 2003; 25:553–560.
2Niemann CU, Walia A, Waldman J, Davio M, Roberts JP, Hirose R, Feiner J. Acute kidney injury during liver transplantation as determined by neutrophil gelatinase-associated lipocalin. Liver Transpl 2009; 15: 1852–1860.
3Wenger U, Neff TA, Oberkofler CE, Zimmermann M, Stehberger PA, Scherrer M, et al. The relationship between preoperative creatinine clearance and outcomes for patients undergoing liver transplantation: a retrospective observational study. BMC Nephrol 2013; 14:37.
4Sirota JC, Walcher A, Faubel S, Jani A, McFann K, Devarajan P, et al. Urine IL-18, NGAL, IL-8 and serum IL-8 are biomarkers of acute kidney injury following liver transplantation. BMC Nephrol 2013; 14:17–23.
5Devarajan P. Novel biomarkers for the early prediction of acute kidney injury. Cancer Ther 2005; 3:477–488.
6Devarajan P. Biomarkers for the early detection of acute kidney injury. Curr Opin Pediatr 2010; 23:194–200.
7Haase M, Devarajan P, Haase-Fielitz A, Bellomo R, Cruz DN, Wagener G, et al. The outcome of neutrophil gelatinase-associated lipocalin-positive subclinical acute kidney injury: a multicenter pooled analysis of prospective studies. J Am Coll Cardiol 2011; 57:1752–1761.
8In collaberation with The online MELD and PELD calculator. Available at: http://optn.transplant.hrsa.gov/resources/MeldPeldCalculator. [Accessed 2014 Aug 22].
9Mehta RL, Kellum JA, Shah SV, et al. Acute Kidney Injury Network (AKIN): report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007; 11:R31.
10Umbro I, Rossi M, Tinti F, Fiacco F, Piselli P, Ianni S, et al. Stages of early acute renal dysfunction in liver transplantation: the influence of graft function. Transplant Proc 2012; 44:1953–1955.
11Barri YM, Sanchez EQ, Jennings LW, Melton LB, Hays S, Levy MF, Klintmalm GB. Acute kidney injury following liver transplantation: definition and outcome. Liver Transpl 2009; 15:475–483.
12Wagener G, Minhaz M, Mattis FA, Kim M, Emond JC, Lee HT. Urinary neutrophil gelatinase-associated lipocalin as a marker of acute kidney injury after orthotopic liver transplantation. Nephrol Dial Transplant 2011; 26:1717–1723.
13Portal AJ, McPhail MJ, Bruce M, Coltart I, Slack A, Sherwood R, et al. Neutrophil gelatinase —associated lipocalin predicts acute kidney injury in patients undergoing liver transplantation. Liver Transpl 2010; 16:1257–1266.
14Iglesias JI, DePalma JA, Levine JS. Risk factors for acute kidney injury following orthotopic liver transplantation: the impact of changes in renal function while patients await transplantation. BMC Nephrol 2010; 11:30–45.
15Pawarode A, Fine DM, Thuluvath PJ. Independent risk factors and natural history of renal dysfunction in liver transplant recipients. Liver Transpl 2003; 9:741–747.
16Campbell MS, Kotlyar DS, Brensinger CM, Lewis JD, Shetty K, Bloom RD, et al. Renal function after orthotopic liver transplantation is predicted by duration of pretransplantation creatinine elevation. Liver Transpl 2005; 11:1048−1055.
17Yalavarthy R, Edelstein CL, Teitelbaum I. Acute renal failure and chronic kidney disease following liver transplantation. Hemodial Int 2007; Suppl 3: S7–S12.
18O'Riordan A, Wong V, McQuillan R, McCormick PA, Hegarty JE, Watson AJ. Acute renal disease, as defined by the RIFLE criteria, post-liver transplantation. Am J Transplant 2007; 7:168–176.
19Nielsen SE, Schjoedt KJ, Astrup AS, Tarnow L, Lajer M, Hansen PR, et al. Neutrophil gelatinase-associated lipocalin (NGAL) and kidney injury molecule 1 (KIM1) in patients with diabetic nephropathy: a cross-sectional study and the effects of lisinopril. Diabet Med 2010; 27:1144–1150.
20Chappell WH, Abrams SL, Franklin RA, LaHair MM, Montalto G, Cervello M, et al. Ectopic NGAL expression can alter sensitivity of breast cancer cells to EGFR, Bcl-2, CaM-K inhibitors and the plant natural product berberine. Cell Cycle 2012; 11:4447–4461.
21Sedra AH, Strum E. The role of intraoperative hemodialysis in liver transplant patients. Curr Opin Organ Transplant 2011; 16:323–325.
22Hong SH, Lee JM, Choi JH et al. perioperative assessment of terlipress ininfusion during living donor liver transplantation. J Int Med Res 2012; 40:225–236