Back
Acute kidney injury (AKI) is a common and serious complication among liver cirrhosis patients and is associated with increased morbidity and mortality. MAKE 30 is defined as the occurrence of either rise of creatinine> 40%, requirement of dialysis, death at day 30 and is used as an outcome measure in critically ill patients. As creatinine is an inaccurate and unreliable marker, newer biomarkers are being explored. There is scarce data comparing serum versus urine biomarkers as predictors of AKI in critically ill cirrhotic patients. Some studies show that plasma biomarkers are better at predicting AKI than urinary biomarkers. We planned this study to compare accuracy of serum and urinary FABP and Cystatin C in predicting occurrence of AKI and MAKE 30 in critically ill cirrhotics.
All cirrhotic patients with critical illness admitted to intensive care unit were screened and consecutive patients were enrolled. Patients with AKI or CKD at presentation and/ or those on dialysis were excluded. Clinical data of enrolled patients were recorded and blood and urine samples were stored at baseline and follow-up. Outcomes were recorded at 30 days. Development of AKI or occurrence of either MAKE 30 was considered a positive outcome. Serum and urinary biomarkers of FABP1 and Cystatin C were tested by ELISA technique. Statistical analysis was done by logistic regression as shown in Table 2.
Total of 49 patients were analyzed. Cirrhotic etiologies and their respective frequency obtained were as follows: Hepatitis C: 6; Hepatitis B: 3; Alcoholic Hepatitis: 10; NAFLD/NASH: 7; malignancy: 6; sepsis: 11; others: 6. 32 patients had overlapping conditions. AKI developed in 25 patients. 23 patients required ventilator support. Total of 36 patients developed any of MAKE 30. At 30 day follow up, 26 patients had expired, 4 patients were on dialysis. 6 events with rise in creatinine > 40% were observed. The most frequent MAKE 30 event was death. All 4 serum and urinary biomarkers were associated with AKI, and none with MAKE 30 as shown in Table 2. Serum FABP1 had highest sensitivity whereas urine Cystatin C had highest specificity in predicting AKI in critically ill cirrhotic patients as shown in Table 3.
Table 1. Baseline parameters with average and SD
Baseline
parameters
Average
SD
Hb (g/ dL)
8.219
1.820
TLC ( / ul)
14059
9412.1
PC
94659.456
71082
Urea (mg/ dL)
72.6
60.455
Creatinine (mg/ dL)
1.597
1.483
SGOT (U/ L)
126.37
144.52
ALT (U/ L)
158.56
434.07
Bilirubin (mg/ dL)
14.407
9.571
Albumin (g/ dL)
2.804
0.49
INR
2.3705
1.278
SBP (mmHg)
115.55
21.236
DBP (mmHg)
67.955
11.454
MELD score
28.521
9.17
Abbr.: SD; Standard Deviation, Hb; Hemoglobin, TLC; Total Leukocyte Count, PC; Platelet Count, SGOT; Aspartate Aminotransferase, ALT; Alanine Transamine, INR; International Normalised Ratio, SBP; Systolic Blood Pressure, DBP; Diastolic Blood Pressure
Table 2. Logistic Regression analysis: Association of independent biomarkers with AKI
Descriptive data (ng/ml)
Association with AKI
Biomarker
P value
aOR
CI 95 %
Serum_Cystatin C
2436.994
645.253
0.003
1.001
(0.999-1.002)
Serum_FABP1
29.251
6.695
0.002
1.172
(0.999-1.376)
Urine_Cystatin C
1747.96
2235.55
0.048
1.000
(1.000-1.000)
Urine_FABP1
51.036
405.717
0.025
0.999
(0.997-1.002)
Abbr. aOR; adjusted Odds Ratio, SD; Standard Deviation, CI; Confidence Interval
Table 3: Youden’s index analysis: Test for sensitivity and specificity of serum and urine Cystatin C and FABP1 in predicting AKI
AUC
Asymptotic P value
Asymptotic CI 95 %
Cut off >
Sensitivity
Specificity
Serum_Cystatin C (ng/ml)
0.757
0.623- 0.890
0.890
0.720
0.667
Serum_FABP1 (ng/ml)
0.773
0.001
0.641- 0.905
0.905
0.840
Urine_Cystatin C (ng/ml)
0.629
0.121
0.466- 0.792
0.792
0.520
0.875
Urine_FABP1 (ng/ml)
0.708
0.013
0.563- 0.852
0.852
0.680
0.625
AUC was calculated using Receiver Operator curve
Abbr. AUC; Area under the Curve, CI; Confidence Interval
In critically ill cirrhotic patients admitted to intensive care unit, serum and urinary Cystatin C and L-FABP 1 at baseline predicted development of AKI but were not associated significantly with MAKE 30. Serum FABP1 had highest sensitivity whereas urine Cystatin C had highest specificity in predicting AKI in this population. It is possible that outcomes other than AKI are primarily responsible for MAKE 30. Serum Cystatin C and urine FABP1 can be used to stratify patients at risk of AKI and can help in identifying patients for early institution of reno-protective measures. The impact of early AKI identification on survival needs to be studied further.