Introduction: Cardiac surgery-associated acute kidney injury (CSA-AKI) is a common complication associated with increased morbidity and mortality. Tissue inhibitor metalloproteinase-2·insulin-like growth factor-binding protein 7 (TIMP-2·IGFBP7) determines tubular stress markers, which may occur prior to tubular damage. Previous studies on the use of TIMP-2·IGFBP7 for the prediction of CSA-AKI showed divergent results. Therefore, this study aimed to explore the predictive value of TIMP-2·IGFBP7 measurements for the early detection of acute kidney injury (AKI) and short-term adverse outcomes after cardiac surgery. Methods: In the prospective cohort study, blood and urine samples were collected 6–12 h after cardiac surgery. Blood samples to monitor serum creatinine levels were additionally extracted from days 1 to 7. AKI was defined based on the KDIGO consensus guidelines. AKI within 7 days following surgery was the primary outcome. The initiation of renal replacement therapy, in intensive care unit mortality, and the combination of both were secondary outcomes. Results: A total of 557 patients were enrolled; 134 (24.06%) of them developed AKI and 33 (5.9%) had moderate or severe AKI. AKI developed more frequently in elderly patients with diabetes or with higher baseline serum creatinine levels. Patients with AKI had higher EuroSCORE II, Cleveland Clinic Score, and simplified renal index (SRI) than those without AKI. Urinary TIMP-2·IGFBP7 was significantly higher in patients with AKI. The area under the curve was 0.66 in predicting all AKI and 0.70 in predicting stages 2 and 3 AKI. The resulting sensitivity and specificity were 44.0% and 83.9%, respectively, for a calculated threshold TIMP-2·IGFBP7 value of 0.265 (ng/mL)2/1,000. The TIMP-2·IGFBP7 values, SRI score, and age were significantly associated with AKI within 7 days postoperatively. A total of 33 patients reached the composite endpoint; the percentage of patients who reached the composite endpoint in the TIMP-2·IGFBP7 of >0.265 (ng/ml)2/1,000 group was significantly higher than that of ≤0.265 (ng/mL)2/1,000 group. Conclusions: Postoperative implementation of TIMP-2·IGFBP7 improved the prediction of CSA-AKI and may aid in identifying patients at risk of short-term adverse outcomes. We identified an ideal calculated cutoff value of 0.265 (ng/mL)2/1,000 for the prediction of CSA-AKI among all AKI patients.

Cardiac surgery-associated acute kidney injury (CSA-AKI) is a common complication associated with increased morbidity and mortality [1]. Several factors may contribute to the CSA-AKI development, including ischemia-reperfusion injury, inflammation, hypoperfusion, and oxidative stress [2]. The early diagnosis of acute kidney injury (AKI) is essential for the early implementation of strategies to prevent the development of severe kidney damage.

This has been exacerbated by the fact that AKI diagnosis is usually delayed when traditional diagnostic markers such as serum creatinine (sCr) level and urine output are used [3]. The US Food and Drug Administration approved the combination of urine tissue inhibitor metalloproteinase-2 (TIMP-2) and insulin-like growth factor-binding protein 7 (IFGBP7) (NephroCheck®) to be performed as a tool to help an adequate risk assessment and an early detection of AKI [4]. A cutoff value of 0.3 (ng/mL)2/1,000 is used to identify patients at high risk for AKI, and a value >2.0 (ng/mL)2/1,000 identifies patients at the highest risk for AKI [5]. Thus, the predicting capability of NephroCheck® in previous studies was good in predicting more severe AKIs in critically ill patients with respiratory or cardiovascular dysfunction in the intensive care unit (ICU) [6].

Previous studies on the use of tissue inhibitor metalloproteinase-2·insulin-like growth factor-binding protein 7 (TIMP-2·IGFBP7) for the prediction of CSA-AKI showed divergent results. In 2014, Meersch et al. [7] found a cutoff value of 0.3 (ng/mL)2/1,000 showed good sensitivity and specificity at 4 h after cardiac surgery with a cardiopulmonary bypass (CPB). Furthermore, the decline of urinary TIMP-2 and IGFBP7 at discharge was a strong predictor for renal recovery. A meta-analysis including 8 studies and 552 patients confirmed that urinary TIMP-2 and IGFBP7 are helpful biomarkers for the early diagnosis of CSA-AKI [8]. In the following study, Meersch et al. [9] disclosed that implementing a bundle of kidney protective measures in AKI-high-risk patients defined as urinary TIMP-2·IGFBP7 of >0.3 (ng/mL)2/1,000 after cardiac surgery was beneficial in reducing AKI occurrence and progression. Likewise, Engelman et al. [10] found that moderate or severe AKI was reduced by 89% by implementing a predefined staged protocol after identifying patients at high risk for AKI using NephroCheck®. However, the PREDICTAKI trial recently showed NephroCheck® results did not improve the prediction of CSA-AKI based on creatinine levels only. Moreover, no conclusions could be appropriate to predict severe AKI due to the limited number of events [11].

AKI is now recognized as a heterogeneous syndrome associated with exposure (low cardiac output, sepsis, major surgery, toxicity, etc.) and pathophysiology (hypoperfusion, inflammation, etc.) [12]. As AKI etiologies and pathophysiology substantially differ between patient populations, the AKI biomarkers may be very different. We, therefore, aimed to explore the predictive value of TIMP-2·IGFBP7 measurements for the early detection of AKI and short-term adverse outcomes after cardiac surgery.

Study Design and Participants

All adult patients who underwent cardiac surgery (≥18 years) from March 2017 to 2018 at San Bortolo Hospital (Vicenza, Italy) were registered in the study. Patients with CKD stages 4–5, anuria, and AKI preoperatively were excluded. AKI within 7 days after surgery was the primary outcome. The initiation of renal replacement therapy (RRT), in-ICU mortality, and the combination of both were the secondary outcomes.

Blood and urine samples to measure sCr and urinary (TIMP-2·IGFBP7) concentrations were collected 6–12 h after cardiac surgery. Blood samples to measure the sCr levels were additionally monitored from days 1 to 7. All other clinical data for the study were collected from hospital records, including the medical history, baseline demographics, body mass index, urine output, comorbidities, left ventricular ejection fraction, CPB time, and length of hospital stay. In addition, data about RRT and death were also recorded. The http://euroscore.org was used to calculate the European System for Cardiac Operative Risk Evaluation II (EuroSCORE II). The Cleveland Clinic Score [13] and the simplified renal index (SRI) score [14] were also calculated.

Blood and urine samples were analyzed within 1 h after the collection in the local central laboratory. The sCr level was measured using the enzymatic method with an automatic analyzer (Dimension Vista, Siemens Healthcare, Tarrytown, NY, USA). Urine samples were analyzed for TIMP-2 and IGFBP7 using NephroCheck® (Astute Medical, San Diego, CA, USA). The concentration of the two biomarkers was analyzed using the ASTUTE140 meter, which divides these products by 1,000 to report a single numerical test value with units of (ng/mL)2/1,000 (the units for all TIMP-2·IGFBP7 tests and cutoff values in this report).

AKI was defined based on the KDIGO consensus guidelines [15]. AKI stage 1: increase in sCr by ≥ 0.3 mg/dL (≥26.4 μmol/L) or increase by 1.5–1.9 times baseline or urine output of <0.5 mL/kg/h for 6–12 h; AKI stage 2: increase in sCr by 2.0–2.9 times baseline and/or urine output of <0.5 mL/kg/h for ≥12 h; AKI stage 3: increase in sCr by 3.0 times baseline or increase in sCr to 4.0 mg/dL (353.6 μmol/L) or the treatment with RRT and/or urine output <0.3 mL/kg/h for ≥24 h or anuria for 12 h. Moderate to severe AKI was defined as KDIGO stage 2 or 3 AKI. The estimated glomerular filtration rate was determined using the Chronic Kidney Disease Epidemiology Collaboration equation. The baseline sCr level was defined as the sCr value obtained 3–6 months before hospital admission or the sCr value at hospital admission in patients with no previous values (86 patients, 15.44%).

Flowchart

Study design and number of patients in each group (Fig. 1).

Fig. 1.

Study design and number of patients in each group.

Fig. 1.

Study design and number of patients in each group.

Close modal

Statistical Analysis

Continuous variables were described as means ± standard deviations or medians and interquartile ranges. Percentages were calculated for categorical variables. Patients were grouped based on the KDIGO criteria values, and the Mann-Whitney test was used for a two-group comparison. Categorical variables were compared between groups using Fisher’s exact test or the χ2 test.

To assess the predictive ability of TIMP-2·IGFBP7 for the endpoints, empirical receiver operating characteristic curves, area under the receiver operating characteristic curve (AUC), sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) were calculated. Logistic regression modeling was performed to analyze the relationship between preoperative and intraoperative variables with AKI within 7 days postoperatively. Finally, all p values were two-sided, and a p value of <0.05 was considered statistically significant. The SPSS Statistics V26.0 (IBM, USA) was used for analyses.

Patient Characteristics and Procedures at Baseline

During the study period, 714 adult patients underwent cardiac surgery: 157 of them were excluded, 15 due to CKD stages 4–5 and 33 due to anuria; 19 patients had AKI preoperatively, 65 due to incomplete data; 5 refused to participate; and 20 withdrew from the study. Finally, the remaining 557 patients were enrolled in the study from March 1, 2017, to March 31, 2018.

The baseline characteristics of the patient population stratified by AKI status are summarized in Table 1. No differences in gender, body mass index, and left ventricular ejection fraction were observed between the group with and without AKI. However, patients in the AKI group were found to be significantly older (70.4 vs. 65.11 years, p < 0.001). Furthermore, AKI is promoted more frequently in patients with diabetes or higher baseline sCr levels.

Table 1.

Baseline characteristics for all patients and by AKI status (n = 557)

Total (N = 557)AKI (N = 134)No AKI (N = 423)p value
Age, years 66.4±11.8 70.4±9.96 65.11±12.06 0.000* 
Male, n (%) 393 (70.6) 99 (73.9) 294 (69.5) 0.333 
BMI, kg/m2 25.85±4.06 26.24±3.88 25.74±4.13 0.468 
Preoperative creatinine, mg/dL 0.93±0.27 1.02±0.35 0.9±0.23 0.000* 
Baseline eGFR, mL/min/1.73m2 83.44±14.72 80.32±15.66 84.39±14.29 0.005* 
Comorbidities 
 Hypertension, n (%) 175 (31.4) 42 (31.3) 133 (31.4) 0.983 
 Coronary artery disease, n (%) 123 (22.1) 28 (20.9) 95 (22.5) 0.292 
 Diabetes, n (%) 40 (7.2) 14 (10.4) 26 (6.1) 0.012* 
 Peripheral artery disease, n (%) 25 (4.5) 18 (5.2) 7 (4.3) 0.308 
 COPD, n (%) 15 (2.7) 4 (3.0) 11 (2.6) 0.753 
 Previous heart surgery, n (%) 33 (5.9) 9 (6.7) 24 (5.7) 0.314 
 LVEF, % 58.85±11.32 55.98±13.73 59.66±10.5 0.112 
CPB time, min 127.94±58.22 134.60±68.57 126.26±55.39 0.392 
EuroSCORE II 7.31±2.48 6.01±2.53 0.002* 
Cleveland Clinic Score 4.29±2.04 3.38±1.20 0.004* 
 Stage 1 (scores 0–2), n (%) 26 (19.4%) 94 (22.2%) <0.001* 
 Stage 2 (scores 3–5), n (%) 79 (59%) 315 (74.5%) 
 Stage 3 (scores 6–8), n (%) 20 (14.9%) 14 (3.3%) 
 Stage 4 (scores 9–13), n (%) 9 (6.7%)  
SRI score 2.10±1.07 1.53±0.78 0.001* 
 Stage 0 (scores 0–1), n (%) 44 (32.8) 245 (57.9) <0.001* 
 Stage 1 (scores 2–3), n (%) 80 (59.7) 171 (40.4) 
 Stage 2 (scores 4–5), n (%) 10 (7.5) 7 (1.7) 
 Stage 3 (scores 6–8)  
LOIS, days 3.89±7.93 6.32±7.68 3.27±7.9 0.022* 
LOS, days 14.13±22.08 22.13±40.68 11.68±10.35 0.013* 
Total (N = 557)AKI (N = 134)No AKI (N = 423)p value
Age, years 66.4±11.8 70.4±9.96 65.11±12.06 0.000* 
Male, n (%) 393 (70.6) 99 (73.9) 294 (69.5) 0.333 
BMI, kg/m2 25.85±4.06 26.24±3.88 25.74±4.13 0.468 
Preoperative creatinine, mg/dL 0.93±0.27 1.02±0.35 0.9±0.23 0.000* 
Baseline eGFR, mL/min/1.73m2 83.44±14.72 80.32±15.66 84.39±14.29 0.005* 
Comorbidities 
 Hypertension, n (%) 175 (31.4) 42 (31.3) 133 (31.4) 0.983 
 Coronary artery disease, n (%) 123 (22.1) 28 (20.9) 95 (22.5) 0.292 
 Diabetes, n (%) 40 (7.2) 14 (10.4) 26 (6.1) 0.012* 
 Peripheral artery disease, n (%) 25 (4.5) 18 (5.2) 7 (4.3) 0.308 
 COPD, n (%) 15 (2.7) 4 (3.0) 11 (2.6) 0.753 
 Previous heart surgery, n (%) 33 (5.9) 9 (6.7) 24 (5.7) 0.314 
 LVEF, % 58.85±11.32 55.98±13.73 59.66±10.5 0.112 
CPB time, min 127.94±58.22 134.60±68.57 126.26±55.39 0.392 
EuroSCORE II 7.31±2.48 6.01±2.53 0.002* 
Cleveland Clinic Score 4.29±2.04 3.38±1.20 0.004* 
 Stage 1 (scores 0–2), n (%) 26 (19.4%) 94 (22.2%) <0.001* 
 Stage 2 (scores 3–5), n (%) 79 (59%) 315 (74.5%) 
 Stage 3 (scores 6–8), n (%) 20 (14.9%) 14 (3.3%) 
 Stage 4 (scores 9–13), n (%) 9 (6.7%)  
SRI score 2.10±1.07 1.53±0.78 0.001* 
 Stage 0 (scores 0–1), n (%) 44 (32.8) 245 (57.9) <0.001* 
 Stage 1 (scores 2–3), n (%) 80 (59.7) 171 (40.4) 
 Stage 2 (scores 4–5), n (%) 10 (7.5) 7 (1.7) 
 Stage 3 (scores 6–8)  
LOIS, days 3.89±7.93 6.32±7.68 3.27±7.9 0.022* 
LOS, days 14.13±22.08 22.13±40.68 11.68±10.35 0.013* 

AKI, acute kidney injury; BMI, body mass index; eGFR, estimated glomerular filtration rate; COPD, chronic obstructive pulmonary disease; LVEF, left ventricular ejection fraction; CPB, cardiopulmonary bypass; EuroSCORE, European System for Cardiac Operative Risk Evaluation; SRI, simplified renal index; ICU, intensive care unit; LOIS, length of ICU stay; LOS, length of hospital stay. *p < 0.05.

The percentage of patients with hypertension, coronary artery disease, peripheral artery disease, chronic obstructive pulmonary disease, and previous heart surgery was similar between the groups with and without AKI (p > 0.05). The time of CPB did not significantly differ between the two groups (134.60 vs. 126.26 min; p > 0.05). In addition, patients with AKI stayed longer in the ICU (6.32 vs. 3.27 days, p < 0.05) and hospital (22.13 vs. 11.68 days, p < 0.05).

EuroSCORE II is a cardiac risk model for predicting mortality after cardiac surgery. Patients with AKI had higher EuroSCORE II than those without (7.31 vs. 6.01, p < 0.05). The Cleveland Clinic Score and SRI predict kidney failure and the need for kidney replacement therapy (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000538031). The Cleveland score in the AKI group was significantly higher than that in the no AKI group (4.29 vs. 3.38, p < 0.05). Between these two groups, the number of patients in stage 2 (Cleveland Clinic Score, 3–5) was the highest. Similarly, patients with AKI had a higher SRI score than those without (2.10 vs. 1.53, p < 0.05). Additionally, among patients without AKI, the majority were in stage 0 (SRI score, 0–1), whereas among patients with AKI, the majority were in stage 1 (SRI score, 2–3) (Table 1).

Occurrence of CSA-AKI Defined by KDIGO and Its Subgroups

In total, 134 (24.06%) patients developed AKI of any stage: 101 (18.1%) had AKI stage 1 and 33 (5.9%) had moderate or severe AKI. Within 24 h after cardiac surgery, 84 (15.08%) patients had developed AKI defined by KDIGO: 72 (85.71%) were classified as stage 1 and 12 (14.29%) as stages 2 and 3. Of these patients, 82 patients had AKI defined by creatinine and only 5 patients defined by urinary output.

Within 48 h after cardiac surgery, 106 (19.03%) patients had an AKI diagnosis by KDIGO and 41 (38.68%) were newly diagnosed. A total of 100 patients had AKI defined by the creatinine level and 10 by urinary output. Nine patients developed AKI 3–7 days postoperatively (Table 2).

Table 2.

Occurrence of AKI at different times after cardiac surgery, defined by KDIGO and its subgroups

24 h48 h
no AKIAKI stage 1AKI stage 2–3no AKIAKI stage 1AKI stage 2–3
AKI creatinine, n (%) 475 (85.28) 70 (12.57) 12 (2.15) 457 (82.05) 73 (13.11) 27 (4.85) 
AKI urine output, n (%) 552 (99.10) 5 (0.90) 0 (0) 547 (98.20) 8 (1.43) 2 (0.37) 
KDIGO, n (%) 473 (84.92) 72 (12.93) 12 (2.15) 451 (80.97) 79 (14.18) 27 (4.85) 
24 h48 h
no AKIAKI stage 1AKI stage 2–3no AKIAKI stage 1AKI stage 2–3
AKI creatinine, n (%) 475 (85.28) 70 (12.57) 12 (2.15) 457 (82.05) 73 (13.11) 27 (4.85) 
AKI urine output, n (%) 552 (99.10) 5 (0.90) 0 (0) 547 (98.20) 8 (1.43) 2 (0.37) 
KDIGO, n (%) 473 (84.92) 72 (12.93) 12 (2.15) 451 (80.97) 79 (14.18) 27 (4.85) 

AKI, acute kidney injury; KDIGO, Kidney Disease Improving Global Outcomes.

Prediction of the AKI Development

Individual values of urinary TIMP-2·IGFBP7 were based on the occurrence of AKI or not. Patients with AKI had significantly higher TIMP-2·IGFBP7 values than those without AKI (0.17 ± 0.01 versus 0.51 ± 0.08 [ng/mL]2/1,000, p < 0.0001) (Fig. 2). The AUC for urine TIMP-2·IGFBP7 in predicting the AKI for 6 h after cardiac surgery was 0.66 (95% confidence interval 0.60–0.72, p < 0.05) (Fig. 3a). The resulting sensitivity and specificity were 44.0% and 83.9%, respectively, for a calculated threshold TIMP-2·IGFBP7 value of 0.265 (ng/mL)2/1,000. Youden’s index was 0.28, the PPV was 46.5%, and NPV was 82.6%. During the 6-h measurement, the AUC was 0.70 (95% CI 0.60–0.80, p < 0.05) to predict stage 2 and 3 AKI (Fig. 3b). The sensitivity and specificity were 80.0% and 57.6%, respectively, for a threshold TIMP-2·IGFBP7 value of 0.28 (ng/mL)2/1,000. The PPV and NPV were 15.3% and 96.8%, respectively.

Fig. 2.

Individual values of urinary TIMP-2·IGFBP7 according to the occurrence of AKI or not. Patients with AKI had significantly higher TIMP-2·IGFBP7 values than those without AKI (0.17 ± 0.01 vs. 0.51 ± 0.08 (ng/mL)2/1,000, p < 0.0001).

Fig. 2.

Individual values of urinary TIMP-2·IGFBP7 according to the occurrence of AKI or not. Patients with AKI had significantly higher TIMP-2·IGFBP7 values than those without AKI (0.17 ± 0.01 vs. 0.51 ± 0.08 (ng/mL)2/1,000, p < 0.0001).

Close modal
Fig. 3.

The ROC curves to predict AKI using TIMP-2·IGFBP7 results. a The AUC for urine TIMP-2·IGFBP7 in predicting all AKI during 6 h after cardiac surgery was 0.66 (95% CI: 0.60–0.72, p < 0.05). The resulting sensitivity and specificity were 44.0% and 83.9%, respectively, for a calculated threshold TIMP-2·IGFBP7 value of 0.265 (ng/mL)2/1,000. b The AUC was 0.70 (95% CI 0.60–0.80, p < 0.05) to predict stage 2 and 3 AKI. The sensitivity and specificity were 80.0% and 57.6%, respectively, for a threshold TIMP-2·IGFBP7 value of 0.28 (ng/mL)2/1,000. ROC, receiver operating characteristic.

Fig. 3.

The ROC curves to predict AKI using TIMP-2·IGFBP7 results. a The AUC for urine TIMP-2·IGFBP7 in predicting all AKI during 6 h after cardiac surgery was 0.66 (95% CI: 0.60–0.72, p < 0.05). The resulting sensitivity and specificity were 44.0% and 83.9%, respectively, for a calculated threshold TIMP-2·IGFBP7 value of 0.265 (ng/mL)2/1,000. b The AUC was 0.70 (95% CI 0.60–0.80, p < 0.05) to predict stage 2 and 3 AKI. The sensitivity and specificity were 80.0% and 57.6%, respectively, for a threshold TIMP-2·IGFBP7 value of 0.28 (ng/mL)2/1,000. ROC, receiver operating characteristic.

Close modal

In logistic regression modeling, we analyzed the relationship of preoperative and intraoperative variables with AKI within 7 days postoperatively. TIMP-2·IGFBP7 significantly improved risk prediction when added to the eight-parameter clinical model for the primary endpoint. The OR for TIMP-2·IGFBP7 was 4.524 (95% CI: 1.127–18.158, p < 0.05), for SRI score was 3.085 (95% CI: 1.635–5.716, p < 0.001), and that for age was 1.077 (95% CI: 1.021–1.136, p < 0.05). No other significant associations could be observed, neither the baseline estimated glomerular filtration rate, diabetes, hypertension, CPB time, EuroSCORE II, or Cleveland Clinic Score (Table 3).

Table 3.

Logistic regression risk models for TIMP-2·IGFBP7 and clinical covariates

VariableReference risk modelNew risk model (addition of NC to reference risk model)
odds ratiop valueodds ratio1p value
Age 1.069 (1.015–1.125) 0.005* 1.077 (1.021–1.136) 0.007* 
Baseline eGFR 0.535 (0.042–6.875) 0.897 0.897 (0.064–12.489) 0.882 
Diabetes 0.660 (0.181–2.400) 0.927 0.811 (0.218–3.011) 0.946 
Hypertension 0.684 (0.201–2.327) 0.582 0.630 (0.179–2.216) 0.523 
CPB time 1.009 (0.999–1.018) 0.139 1.007 (0.997–1.016) 0.234 
EuroSCORE II 0.790 (0.573–1.090) 0.337 0.799 (0.576–1.109) 0.295 
Cleveland Clinic Score 1.338 (0.603–2.967) 0.948 1.246 (0.553–2.803) 0.969 
SRI score 3.289 (1.797–6.018) <0.001* 3.058 (1.636–5.716) <0.001* 
TIMP2·IGFBP7 Not included in the model 4.524 (1.127–18.158) 0.033* 
VariableReference risk modelNew risk model (addition of NC to reference risk model)
odds ratiop valueodds ratio1p value
Age 1.069 (1.015–1.125) 0.005* 1.077 (1.021–1.136) 0.007* 
Baseline eGFR 0.535 (0.042–6.875) 0.897 0.897 (0.064–12.489) 0.882 
Diabetes 0.660 (0.181–2.400) 0.927 0.811 (0.218–3.011) 0.946 
Hypertension 0.684 (0.201–2.327) 0.582 0.630 (0.179–2.216) 0.523 
CPB time 1.009 (0.999–1.018) 0.139 1.007 (0.997–1.016) 0.234 
EuroSCORE II 0.790 (0.573–1.090) 0.337 0.799 (0.576–1.109) 0.295 
Cleveland Clinic Score 1.338 (0.603–2.967) 0.948 1.246 (0.553–2.803) 0.969 
SRI score 3.289 (1.797–6.018) <0.001* 3.058 (1.636–5.716) <0.001* 
TIMP2·IGFBP7 Not included in the model 4.524 (1.127–18.158) 0.033* 

eGFR, estimated glomerular filtration rate; CPB, cardiopulmonary bypass; EuroSCORE, European System for Cardiac Operative Risk Evaluation; SRI, simplified renal index. 1Adding TIMP2·IGFBP7 improves the model significantly (p = 0.001, likelihood ratio test). *p < 0.05.

Prediction of the Composite Endpoint

In the entire cohort, 27 (4.84%) patients died and 16 (2.87%) patients required the RRT duration of hospital stay. A total of 33 patients reached the composite endpoint of the RRT initiation or ICU mortality, of whom 24 patients had TIMP-2·IGFBP7 of >0.265 (ng/mL)2/1,000, while the remaining 9 patients had TIMP-2·IGFBP7 of ≤0.265 (ng/mL)2/1,000 (Table 4).

Table 4.

The percentage of patients who reached the composite endpoint according to [TIMP-2]·[IGFBP7] categories

[TIMP-2]·[IGFBP7][TIMP-2]·[IGFBP7]p value
>0.265 (ng/mL)2/1,000≤0.265 (ng/mL)2/1,000
Total 127 430  
RRT, n (%) 13 (10.24) 3 (0.70) <0.001* 
Death, n (%) 20 (15.75) 7 (1.63) <0.001* 
Composite endpoint, n (%) 24 (18.9) 9 (2.09) <0.001* 
[TIMP-2]·[IGFBP7][TIMP-2]·[IGFBP7]p value
>0.265 (ng/mL)2/1,000≤0.265 (ng/mL)2/1,000
Total 127 430  
RRT, n (%) 13 (10.24) 3 (0.70) <0.001* 
Death, n (%) 20 (15.75) 7 (1.63) <0.001* 
Composite endpoint, n (%) 24 (18.9) 9 (2.09) <0.001* 

RRT, renal replacement therapy. *p < 0.05.

Based on the TIMP-2·IGFBP7 categories, the percentage of patients that reached the composite endpoint in the TIMP-2·IGFBP7 of >0.265 (ng/mL)2/1,000 group was significantly higher (18.90% vs. 2.09%, p < 0.001) than that of ≤0.265 (ng/mL)2/1,000 group. The AUC was 0.84 (95% CI: 0.71–0.98, p < 0.001) for the prediction of composite endpoints using TIMP-2·IGFBP7 results after cardiac surgery (Fig. 4).

Fig. 4.

The ROC curves for the prediction of composite endpoint after cardiac surgery using TIMP-2·IGFBP7 results. The AUC was 0.84 (95% CI: 0.71–0.98, p < 0.001) for the prediction of composite endpoints. ROC, receiver operating characteristic.

Fig. 4.

The ROC curves for the prediction of composite endpoint after cardiac surgery using TIMP-2·IGFBP7 results. The AUC was 0.84 (95% CI: 0.71–0.98, p < 0.001) for the prediction of composite endpoints. ROC, receiver operating characteristic.

Close modal

CSA-AKI is a common and serious complication [2] with an incidence rate of up to 83% and is associated with a more complicated clinical course and increased mortality [16‒18]. In our cohort of adult patients who underwent cardiac surgery, the incidence of early AKI at any stage within 48 h was 19.03%, and moderate or severe AKI was 5.7%. AKI developed more frequently in older patients with diabetes or higher baseline sCr levels. The hospitalization time of patients with CSA-AKI in the ICU and at the hospital has doubled. Patients with AKI had a higher EuroSCORE II, Cleveland Clinic Score, and SRI score than those without AKI. This means that patients with AKI have a higher risk of death, kidney failure, or kidney replacement therapy.

Currently, AKI is defined by an increase in sCr levels and/or a decline in urine volume, which is not suitable for the early recognition of AKI and early implementation of strategies to prevent the development of severe kidney damage. Although new promising biomarkers (e.g., cystatin C and NGAL) are found, they are validated of limited relevance for AKI prediction [19, 20]. TIMP-2 and IFGBP7, which are involved in the tubular G1 cell-cycle arrest, earliest phases of cellular stress, and tubular injury [6], have recently been suggested as promising tools for the early detection of AKI [21‒23].

These biomarkers are not affected by other acute or chronic comorbidities, including chronic kidney disease [24]. In our study, the AUC was 0.66 for urine TIMP-2·IGFBP7 after cardiac surgery in predicting all AKI and 0.70 in predicting stage 2 and 3 AKI. The optimal cutoff value observed in our population was 0.265(ng/mL)2/1,000, which is slightly lower than that of the previous studies, probably because all patients with AKI at any stage 6 h after cardiac surgery were included, not only moderate to severe AKI in the first 12 h. We also found that TIMP-2·IGFBP7, SRI score, and age were associated with AKI within 7 days postoperatively. Generally, the AKI incidence might occur after cardiac surgery as a consequence of CPB utilization [25]. In our study, the time of CPB was not significantly different between AKI and no AKI groups.

Studies have shown different performances of urinary TIMP-2·IGFBP7 in predicting AKI after cardiac surgery [8, 11, 26‒29]. There may be some reasons for this. First, the time interval between the onset of AKI and study enrollment was variable. To evaluate the earliest change of NephroCheck after cardiac surgery, Wetz et al. [26] found that TIMP-2·IGFBP7 decreased at the end of surgery and then increased at the next measurement point 4 h after CPB and increased further on the first postoperative day. At earlier time points, no significant difference in the TIMP-2·IGFBP7 concentration was found. Wang et al. [30] also found that the combination of urinary TIMP-2 and IGFBP7 4 h after postoperative ICU admission identifies patients at risk for developing AKI. As the respective half-life of the biomarkers is short, some studies might have missed earlier increases in urinary levels of NephroCheck. Thus, in our study, urinary TIMP-2·IGFBP7 concentrations were collected 6–12 h after cardiac surgery. Second, some study population was small with rare AKI occurrence [11]. Moreover, some studies analyzed the predicting capability of NephroCheck in all patients with AKI, not just those with severe AKI [27]. These may lead to different results. Lastly, Wetz et al. [26] also reported that the predictive performance of the NephroCheck as measured by the AUROC was not good to predict AKI based on creatinine only. Using only creatinine, including AKI stage 1, or expanding the observation period can also result in an AKI underestimation.

Since 2017, NephroCheck is performed routinely upon ICU admission in our institution, and several studies have been conducted. We found that measurement of TIMP-2·IGFBP7 on ICU admission not only has a good performance in predicting the probability of AKI in the first 4 days in the emergency ICU [31] but also can serve to identify AKI patients at increased risk for adverse outcomes in the emergency ICU [32]. Simultaneously, the accuracy of TIMP2·IGFBP7 in predicting the risk of AKI in the first 7 days after ICU admission has significant variability when the reason for ICU admission is reviewed [33]. A more recent study found that a single urinary TIMP2·IGFBP7 test can identify trauma patients early who are in danger of developing AKI compared with the current methods based on the sCr measurement [34].

TIMP-2 and IGFBP7 have been validated as biomarkers for the prediction of moderate to severe AKI risk over the next 12 h in multiple cohorts. Elevated urinary TIMP-2·IGFBP7 levels of >0.3 (ng/mL)2/1,000 are considered positive, whereas >2(ng/mL)2/1,000 levels are a sign of the highest risk for AKI [22]. Many investigators have suggested implementing biomarkers in the evaluation of AKI, and the ADQI-23 consensus recommendations emphasized the need for biomarkers to add prognostic information to the existing function-based AKI staging [35]. Existing evidence suggested that among patients who have already developed AKI, an elevated TIMP-2·IGFBP7 level may identify patients at higher risk of severe AKI, death, or need for kidney replacement therapy compared with patients without elevated biomarkers [32, 36]. To test whether higher urinary TIMP-2·IGFBP7 levels are associated with lower survival among patients with the same functional stage of AKI, Luca Molinari studied critically ill patients with septic shock from the Protocolized Care for Early Septic Shock trial [37]. Among patients who developed AKI within 24 h postenrollment, a urinary TIMP-2·IGFBP7 level of >2.0 (ng/mL)2/1,000 was associated with a greater mortality risk at 30 days when AKI stages were defined based on kidney function [38]. The findings indicate that integrating cell-cycle arrest biomarkers into a new staging framework for AKI may improve the discrimination for survival. The BRAVA study is a multicenter study to evaluate the utility of NephroCheck for predicting the AKI development and short-term mortality in the emergency department (ED) [39]. They found that NephroCheck can predict both AKI development and short-term mortality in at-risk ED patients and would be a useful biomarker for early ruling-in or ruling-out of AKI in the ED. In our study, the percentage of patients who reached the composite endpoint of the RRT initiation or ICU mortality in TIMP-2·IGFBP7 of >0.265 (ng/mL)2/1,000 group was significantly higher than that of the TIMP-2·IGFBP7 of ≤0.265 (ng/mL)2/1,000 group.

Faeq et al. [40] demonstrated that an increased urinary TIMP-2·IGFBP7 level after cardiac surgery was associated with the loss of kidney functional reserve at 3 months even in the absence of AKI. Thus, crude measures, such as death or dialysis, may not help detect subclinical injury; however, measuring the kidney functional reserve may help reveal it. It is also possible that in certain conditions (e.g., loss of muscle mass), creatinine levels may fail to accurately reflect kidney function. Faeq and the IRRIV-AKI study group recently assessed the effects of a preoperative high-oral protein load on postcardiac surgery renal function and used experimental models to elucidate mechanisms by which protein might stimulate the kidney-protective effects [41]. They concluded that a preoperative high-protein diet may represent a simple and cost-efficient treatment for preserving long-term GFR. The probable mechanisms of action by which protein loading may induce a kidney-protective response might include cell-cycle inhibition of the renal tubular epithelial cells.

Our study has several limitations. First, this is a single-center study and did not identify the risk factors for CSA-AKI. Second, patients after an on-pump cardiac surgery may develop immense volume imbalances and urinary dilution. Thus, the TIMP-2·IGFBP7 results should be correlated with urinary dilution parameters. However, the study did not describe the perioperative hemodynamic variables. Third, we are limited to TIMP-2·IGFBP7 measures and did not consider other potential exposures, such as drugs or contrast media, which could have influenced the AKI development. Finally, due to low occurrence, the occurrence of severe AKI and the number of patients requiring RRT or death were limited.

Our study confirms the predictive value of NephroCheck for CSA-AKI. Measurement of TIMP-2·IGFBP7 can help identify CSA-AKI and consequent adverse outcomes, and the ideal calculated cutoff value is 0.265 (ng/mL)2/1,000 for all AKI patients. Future prospective studies should be conducted in this population to evaluate whether the use of urinary TIMP-2·IGFBP7, as a tool for AKI risk assessment, can result in a lower incidence of CSA-AKI and a reduction in mortality.

The authors are grateful to Ghada Ankawi, Gregorio Aramid Romero González, Alejandra Molano Trivino, Ana De Castro, and all the fellows of the International Renal Research Institute of Vicenza who helped in the study.

The study was approved by the Institutional Ethics Committee of San Bortolo Hospital, Vicenza, Italy (Comitato Etico Provinciale aULSS 8 Vicenza), with protocol number 03/17. The clinical investigation was conducted according to the principles expressed in the Declaration of Helsinki.

The authors have no conflict of interest to declare.

This study was supported by the Science and Technology Commission of Shanghai Municipality (14DZ2260200, the project of Shanghai Key Laboratory of Kidney and Blood Purification).

Claudio Ronco, Gengru Jiang, and Yun Xie contributed to study design; Qin Guo and Bo Yang contributed to data collection; Pan Xie and Jingxiao Zhang contributed to patient enrollment; Jingxiao Zhang contributed to laboratory work; Qin Guo and Wei Lu contributed to statistical analysis; Yun Xie, Qin Guo, and Wei Lu contributed to manuscript writing and editing. Bo Yang, Claudio Ronco, and Gengru Jiang read and approved the final manuscript.

Additional Information

Yun Xie and Qin Guo contributed equally to this work.

The authors confirm that the data supporting the findings of this study are available within the article.

1.
De Loor
J
,
Herck
I
,
Francois
K
,
Van Wesemael
A
,
Nuytinck
L
,
Meyer
E
, et al
.
Diagnosis of cardiac surgery-associated acute kidney injury: differential roles of creatinine, chitinase 3-like protein 1 and neutrophil gelatinase-associated lipocalin: a prospective cohort study
.
Ann Intensive Care
.
2017
;
7
(
1
):
24
.
2.
Wang
Y
,
Bellomo
R
.
Cardiac surgery-associated acute kidney injury: risk factors, pathophysiology and treatment
.
Nat Rev Nephrol
.
2017
;
13
(
11
):
697
711
.
3.
Chew
STH
,
Hwang
NC
.
Acute kidney injury after cardiac surgery: a narrative review of the literature
.
J Cardiothorac Vasc Anesth
.
2019
;
33
(
4
):
1122
38
.
4.
U.S. Food and Drug Administration News Release
. FDA allows marketing of the first test to assess risk of developing acute kidney injury.
2014
.
5.
Hoste
EA
,
McCullough
PA
,
Kashani
K
,
Chawla
LS
,
Joannidis
M
,
Shaw
AD
, et al
.
Derivation and validation of cutoffs for clinical use of cell cycle arrest biomarkers
.
Nephrol Dial Transplant
.
2014
;
29
(
11
):
2054
61
.
6.
Kashani
K
,
Al-Khafaji
A
,
Ardiles
T
,
Artigas
A
,
Bagshaw
SM
,
Bell
M
, et al
.
Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury
.
Crit Care
.
2013
;
17
(
1
):
R25
.
7.
Meersch
M
,
Schmidt
C
,
Van Aken
H
,
Martens
S
,
Rossaint
J
,
Singbartl
K
, et al
.
Urinary TIMP-2 and IGFBP7 as early biomarkers of acute kidney injury and renal recovery following cardiac surgery
.
PLoS One
.
2014
;
9
(
3
):
e93460
.
8.
Tai
Q
,
Yi
H
,
Wei
X
,
Xie
W
,
Zeng
O
,
Zheng
D
, et al
.
The accuracy of urinary TIMP-2 and IGFBP7 for the diagnosis of cardiac surgery-associated acute kidney injury: a systematic review and meta-analysis
.
J Intensive Care Med
.
2020
;
35
(
10
):
1013
25
.
9.
Meersch
M
,
Schmidt
C
,
Hoffmeier
A
,
Van Aken
H
,
Wempe
C
,
Gerss
J
, et al
.
Prevention of cardiac surgery-associated AKI by implementing the KDIGO guidelines in high risk patients identified by biomarkers: the PrevAKI randomized controlled trial
.
Intensive Care Med
.
2017
;
43
(
11
):
1551
61
.
10.
Engelman
DT
,
Crisafi
C
,
Germain
M
,
Greco
B
,
Nathanson
BH
,
Engelman
RM
, et al
.
Using urinary biomarkers to reduce acute kidney injury following cardiac surgery
.
J Thorac Cardiovasc Surg
.
2020
;
160
(
5
):
1235
46.e2
.
11.
Vandenberghe
W
,
Van Laethem
L
,
Herck
I
,
Peperstraete
H
,
Schaubroeck
H
,
Zarbock
A
, et al
.
Prediction of cardiac surgery associated - acute kidney injury (CSA-AKI) by healthcare professionals and urine cell cycle arrest AKI biomarkers [TIMP-2]*[IGFBP7]: a single center prospective study (the PREDICTAKI trial)
.
J Crit Care
.
2022
;
67
:
108
17
.
12.
Pickkers
P
,
Darmon
M
,
Hoste
E
,
Joannidis
M
,
Legrand
M
,
Ostermann
M
, et al
.
Acute kidney injury in the critically ill: an updated review on pathophysiology and management
.
Intensive Care Med
.
2021
;
47
(
8
):
835
50
.
13.
Thakar
CV
,
Arrigain
S
,
Worley
S
,
Yared
JP
,
Paganini
EP
.
A clinical score to predict acute renal failure after cardiac surgery
.
J Am Soc Nephrol
.
2005
;
16
(
1
):
162
8
.
14.
Wijeysundera
DN
,
Karkouti
K
,
Dupuis
JY
,
Rao
V
,
Chan
CT
,
Granton
JT
, et al
.
Derivation and validation of a simplified predictive index for renal replacement therapy after cardiac surgery
.
JAMA
.
2007
;
297
(
16
):
1801
9
.
15.
Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group
.
KDIGO clinical practice guideline for acute kidney injury
.
Kidney Int
.
2012
(
Suppl 2
):
1
138
.
16.
Silva
TFD
,
Silva
KRDC
,
Nepomuceno
CM
,
Corrêa
CSM
,
Godoy
JPM
,
Santos
ATLD
, et al
.
Incidence of acute kidney injury post cardiac surgery: a comparison of the AKIN and KDIGO criteria
.
Braz J Anesthesiol
.
2021
;
71
(
5
):
511
6
.
17.
Mishra
PK
,
Luckraz
H
,
Nandi
J
,
Nevill
A
,
Giri
R
,
Panayiotou
A
, et al
.
Long-term quality of life postacute kidney injury in cardiac surgery patients
.
Ann Card Anaesth
.
2018
;
21
(
1
):
41
5
.
18.
McIlroy
DR
.
Predicting acute kidney injury after cardiac surgery: much work still to be done
.
Br J Anaesth
.
2021
;
127
(
6
):
825
8
.
19.
Schrezenmeier
EV
,
Barasch
J
,
Budde
K
,
Westhoff
T
,
Schmidt-Ott
KM
.
Biomarkers in acute kidney injury: pathophysiological basis and clinical performance
.
Acta Physiol
.
2017
;
219
(
3
):
554
72
.
20.
Kashani
K
,
Cheungpasitporn
W
,
Ronco
C
.
Biomarkers of acute kidney injury: the pathway from discovery to clinical adoption
.
Clin Chem Lab Med
.
2017
;
55
(
8
):
1074
89
.
21.
Luthra
A
.
[TIMP-2]*[IGFBP7] for predicting early AKI
.
Anaesth Crit Care Pain Med
.
2019
;
38
(
6
):
677
.
22.
Fan
W
,
Ankawi
G
,
Zhang
J
,
Digvijay
K
,
Giavarina
D
,
Yin
Y
, et al
.
Current understanding and future directions in the application of TIMP-2 and IGFBP7 in AKI clinical practice
.
Clin Chem Lab Med
.
2019
;
57
(
5
):
567
76
.
23.
Adler
C
,
Heller
T
,
Schregel
F
,
Hagmann
H
,
Hellmich
M
,
Adler
J
, et al
.
TIMP-2/IGFBP7 predicts acute kidney injury in out-of-hospital cardiac arrest survivors
.
Crit Care
.
2018
;
22
(
1
):
126
.
24.
Heung
M
,
Ortega
LM
,
Chawla
LS
,
Wunderink
RG
,
Self
WH
,
Koyner
JL
, et al
.
Common chronic conditions do not affect performance of cell cycle arrest biomarkers for risk stratification of acute kidney injury
.
Nephrol Dial Transplant
.
2016
;
31
(
10
):
1633
40
.
25.
Permeisari
D
.
Future insights of pharmacological prevention for AKI post cardiopulmonary bypass surgery (based on PK/PD approach)
.
Front Pharmacol
.
2022
;
13
:
975641
.
26.
Wetz
AJ
,
Richardt
EM
,
Wand
S
,
Kunze
N
,
Schotola
H
,
Quintel
M
, et al
.
Quantification of urinary TIMP-2 and IGFBP-7: an adequate diagnostic test to predict acute kidney injury after cardiac surgery
.
Crit Care
.
2015
;
19
(
1
):
3
.
27.
Lee
NM
,
Deriy
L
,
Petersen
TR
,
Shah
VO
,
Hutchens
MP
,
Gerstein
NS
.
Impact of isolyte versus 0.9% saline on postoperative event of acute kidney injury assayed by urinary [TIMP-2] × [IGFBP7] in patients undergoing cardiac surgery
.
J Cardiothorac Vasc Anesth
.
2019
;
33
(
2
):
348
56
.
28.
Gist
KM
,
Goldstein
SL
,
Wrona
J
,
Alten
JA
,
Basu
RK
,
Cooper
DS
, et al
.
Kinetics of the cell cycle arrest biomarkers (TIMP-2*IGFBP-7) for prediction of acute kidney injury in infants after cardiac surgery
.
Pediatr Nephrol
.
2017
;
32
(
9
):
1611
9
.
29.
Pilarczyk
K
,
Edayadiyil-Dudasova
M
,
Wendt
D
,
Demircioglu
E
,
Benedik
J
,
Dohle
DS
, et al
.
Urinary [TIMP-2]*[IGFBP7] for early prediction of acute kidney injury after coronary artery bypass surgery
.
Ann Intensive Care
.
2015
;
5
(
1
):
50
.
30.
Wang
Y
,
Zou
Z
,
Jin
J
,
Teng
J
,
Xu
J
,
Shen
B
, et al
.
Urinary TIMP-2 and IGFBP7 for the prediction of acute kidney injury following cardiac surgery
.
BMC Nephrol
.
2017
;
18
(
1
):
177
.
31.
Di Leo
L
,
Nalesso
F
,
Garzotto
F
,
Xie
Y
,
Yang
B
,
Virzì
GM
, et al
.
Predicting acute kidney injury in intensive care unit patients: the role of tissue inhibitor of metalloproteinases-2 and insulin-like growth factor-binding protein-7 biomarkers
.
Blood Purif
.
2018
;
45
(
1–3
):
270
7
.
32.
Xie
Y
,
Ankawi
G
,
Yang
B
,
Garzotto
F
,
Passannante
A
,
Breglia
A
, et al
.
Tissue inhibitor metalloproteinase-2 (TIMP-2) • IGF-binding protein-7 (IGFBP7) levels are associated with adverse outcomes in patients in the intensive care unit with acute kidney injury
.
Kidney Int
.
2019
;
95
(
6
):
1486
93
.
33.
Yang
B
,
Xie
Y
,
Garzotto
F
,
Ankawi
G
,
Passannante
A
,
Brendolan
A
, et al
.
Influence of patients’ clinical features at intensive care unit admission on performance of cell cycle arrest biomarkers in predicting acute kidney injury
.
Clin Chem Lab Med
.
2020
;
59
(
2
):
333
42
.
34.
Golino
G
,
Greco
M
,
Rigobello
A
,
Danzi
V
,
De Cal
M
,
Malchiorna
N
, et al
.
Incidence of acute kidney injury in polytrauma patients and predictive performance of TIMP2 × IGFBP7 biomarkers for early identification of acute kidney injury
.
Diagnostics
.
2022
;
12
(
10
):
2481
.
35.
Ostermann
M
,
Zarbock
A
,
Goldstein
S
,
Kashani
K
,
Macedo
E
,
Murugan
R
, et al
.
Recommendations on acute kidney injury biomarkers from the acute disease quality initiative consensus conference: a consensus statement
.
JAMA Netw Open
.
2020
;
3
(
10
):
e2019209
.
36.
Joannidis
M
,
Forni
LG
,
Haase
M
,
Koyner
J
,
Shi
J
,
Kashani
K
, et al
.
Use of cell cycle arrest biomarkers in conjunction with classical markers of acute kidney injury
.
Crit Care Med
.
2019
;
47
(
10
):
e820
6
.
37.
ProCESS Investigators
;
Yealy
DM
,
Kellum
JA
,
Huang
DT
,
Barnato
AE
,
Weissfeld
LA
, et al
.
A randomized trial of protocol-based care for early septic shock
.
N Engl J Med
.
2014
;
370
(
18
):
1683
93
.
38.
Molinari
L
,
Del Rio-Pertuz
G
,
Smith
A
,
Landsittel
DP
,
Singbartl
K
,
Palevsky
PM
, et al
.
Utility of biomarkers for sepsis-associated acute kidney injury staging
.
JAMA Netw Open
.
2022
;
5
(
5
):
e2212709
.
39.
Yang
HS
,
Hur
M
,
Lee
KR
,
Kim
H
,
Kim
HY
,
Kim
JW
, et al
.
Biomarker rule-in or rule-out in patients with acute diseases for validation of acute kidney injury in the emergency department (BRAVA): a multicenter study evaluating urinary TIMP-2/IGFBP7
.
Ann Lab Med
.
2022
;
42
(
2
):
178
87
.
40.
Husain-Syed
F
,
Ferrari
F
,
Sharma
A
,
Hinna Danesi
T
,
Bezerra
P
,
Lopez-Giacoman
S
, et al
.
Persistent decrease of renal functional reserve in patients after cardiac surgery-associated acute kidney injury despite clinical recovery
.
Nephrol Dial Transplant
.
2019
;
34
(
2
):
308
17
.
41.
Husain-Syed
F
,
Emlet
DR
,
Wilhelm
J
,
Danesi
TH
,
Ferrari
F
,
Bezerra
P
, et al
.
Effects of preoperative high-oral protein loading on short- and long-term renal outcomes following cardiac surgery: a cohort study
.
J Transl Med
.
2022
;
20
(
1
):
204
.