Summary
Background and objectives
Preoperative proteinuria is associated with a higher incidence of postoperative AKI. Whether the same is true for postoperative proteinuria is uncertain. This study tested the hypothesis that increased proteinuria after cardiac surgery is associated with an increased risk for AKI.
Design, setting, participants, & measurements
This prospective cohort study included 1198 adults undergoing cardiac surgery at six hospitals between July 2007 and December 2009. Albuminuria, urine albumin-to-creatinine ratio (ACR), and dipstick proteinuria were measured 0–6 hours after surgery. The primary outcome was AKI, defined as a doubling in serum creatinine or receipt of acute dialysis during the hospital stay. Analyses were adjusted for patient characteristics, including preoperative albuminuria.
Results
Compared with the lowest quintile, the highest quintile of albuminuria and highest grouping of dipstick proteinuria were associated with greatest risk for AKI (adjusted relative risks [RRs], 2.97 [95% confidence interval (CI), 1.20–6.91] and 2.46 [95% CI, 1.16–4.97], respectively). Higher ACR was not associated with AKI risk (highest quintile RR, 1.66 [95% CI, 0.68–3.90]). Of the three proteinuria measures, early postoperative albuminuria improved the prediction of AKI to the greatest degree (clinical model area under the curve, 0.75; 0.81 with albuminuria). Similar improvements with albuminuria were seen for net reclassification index (0.55; P<0.001) and integrated discrimination index (0.036; P<0.001).
Conclusions
Higher levels of proteinuria after cardiac surgery identify patients at increased risk for AKI during their hospital stay.
Introduction
AKI is a common complication of major surgery (1). Patients who develop AKI carry a much higher risk for death (2–7), and those who recover renal function remain at increased risk for death and CKD (3,8,9).
Despite multiple trials, current therapeutic approaches for AKI remain limited to supportive measures (10). One proposed reason for the failure of prior interventional trials is the inability to accurately diagnose AKI in its early stages, when the disease process may be more amenable to treatment. Currently, the diagnosis of AKI is based on a single marker of filtration, serum creatinine. Unfortunately, serum creatinine is known to have a variable delay in increase depending on renal reserve and the number of injured nephrons. There is also a need for substantial tubular injury before creatinine increases, and many factors aside from kidney injury can cause variations in the serum creatinine measurement (11–14). In response, there has been a recent movement to make the development of more sensitive and specific AKI biomarkers a top research priority (15).
When measured in patients with CKD, albuminuria predicts progression to ESRD (16,17) and a higher risk for cardiovascular disease and death (18,19). In the surgical setting, the presence of preoperative proteinuria is associated with a higher incidence of AKI during the hospital stay (20). Proteinuria levels have been shown to increase in the hours after surgery (21). Whether postoperative proteinuria levels identify patients with a higher incidence of AKI and other adverse outcomes remains uncertain. We conducted the secondary analysis of a large, prospective, multicenter cohort study to test the hypothesis that high levels of proteinuria in the 6 hours after cardiac surgery are associated with an increased risk for AKI during the hospital stay. We also considered whether the association differed across three measures of proteinuria: urine albumin concentration, urine albumin concentration standardized for urine creatinine concentration (i.e., urine albumin-to-creatinine ratio [ACR]), and urine dipstick values.
Materials and Methods
Study Population
As previously described, we prospectively enrolled adults undergoing cardiac surgery (coronary artery bypass grafting or valve surgery) who were at high risk for AKI at six academic medical centers in North America between July 2007 and December 2009 (22,23). High risk for AKI was defined by the presence of one or more of the following: emergency surgery, preoperative serum creatinine >2 mg/dl (>177 µmol/L), ejection fraction <35% or grade 3 or 4 left ventricular dysfunction, age >70 years, diabetes mellitus, concomitant coronary artery bypass grafting and valve surgery, or repeat revascularization surgery. We excluded patients with evidence of AKI before surgery, prior kidney transplantation, preoperative serum creatinine level >4.5 mg/dl (>398 µmol/L), or ESRD. Participants with multiple surgeries could be enrolled in the study only once. All participants provided written informed consent, and the study was approved by each institution’s research ethics board. The reporting of this study follows guidelines set out in the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement (24).
Sample Collection
We collected urine and plasma specimens before surgery and daily for up to 5 days after surgery. The first postoperative samples (labeled as day 1) were collected soon after admission to the intensive care unit (median, 0.25 [interquartile range (IQR), 0.1–0.5] hour after surgery), and the postoperative day 1 (labeled as day 2) samples were collected on the morning of the first postoperative day in the intensive care unit (median, 16.9 [IQR, 13.8–19.6] hours after surgery). For the first 24 postoperative hours, urine samples were collected every 6 hours. The remaining daily blood and urine samples were obtained at the time of routine morning blood collection done for clinical care. Specimen collection was stopped on postoperative day 3 in patients who had no evidence of an increase in serum creatinine. We obtained fresh urine samples from the urometer of the Foley catheter system and centrifuged the samples at 3000 g for 10 minutes to remove cellular debris. Blood was collected in EDTA tubes and was also centrifuged to separate plasma. Urine supernatant and plasma were divided into aliquots in bar-coded cryovials and stored at −80°C until biomarker measurement. No additives or protease inhibitors were added. IL-18 and neutrophil gelatinase-associated lipocalin (NGAL), markers of renal tubular injury, were measured as described elsewhere (25).
Albuminuria, Urine ACR, and Dipstick Proteinuria Measurements
All urine albumin assays were measured by immunoturbidimetry on a Siemens Dimension Plus with an HM clinical analyzer, per manufacturer’s instructions. We measured urine creatinine by the modified Jaffe reaction. Urine albumin and creatinine were measured on frozen aliquots in two separate batches (May 2010 and September 2010). The samples did not undergo any additional freeze-thaw cycles. Dipstick proteinuria was graded as negative, trace, 30–99 mg/dl (0.03–0.099 g/L), and ≥100 mg/dl (≥0.1 g/L) and was measured during the perioperative period (preoperatively, every 6 hours for the first 24 hours postoperatively, daily up to postoperative day 2 [labeled as day 3] or 4 [labeled as day 5]) in the fresh urine using Siemens Clinitek Status ID number SN48923. Personnel performing the biomarker measurements were blinded to each patient’s clinical information.
Outcome Definitions
The primary outcome was the development of AKI, defined as the receipt of acute dialysis or a doubling in serum creatinine from the baseline preoperative value during the entire hospital stay. In modern staging systems, this reflects RIFLE (Risk of renal dysfunction; Injury to the kidney; Failure of kidney function, Loss of kidney function and End-stage kidney disease) stage I (26) or Acute Kidney Injury Network (AKIN) stage 2 AKI (27). Mild AKI, defined by an increase in serum creatinine ≥0.3 mg/dl or ≥50% above the baseline preoperative value (AKIN stage 1), was examined in a supplementary analysis. All preoperative creatinine values were measured within 2 months before surgery. Pre- and postoperative serum creatinine levels were measured in the same clinical laboratory for each patient at all centers. Serum creatinine values were recorded for every patient throughout the hospital stay. Additional clinical outcomes were in-hospital mortality and lengths of in-hospital and intensive care unit stays.
Variable Definitions
We collected preoperative characteristics, operative details, and postoperative complications using definitions of the Society of Thoracic Surgeons (http://www.ctsnet.org/file/rptDataSpecifications252_1_ForVendorsPGS.pdf). We recorded whether the patient had received a cardiac catheterization within 48 hours before surgery. We estimated preoperative GFR (eGFR) using the Chronic Kidney Disease-Epidemiology Collaboration equation (28).
Statistical Analyses
We divided the population into quintiles using the first postoperative value of albuminuria and urine ACR and into four groups based on the first value of dipstick proteinuria. We assessed unadjusted linear trends by the Cochran-Armitage test for dichotomous outcomes, the Jonckheere-Terpstra test for continuous outcomes, and adjusted linear trends using contrasts in logistic or linear regression, respectively. Continuous variables were compared with two-sample t test or Wilcoxon rank-sum test and dichotomous variables with the chi-squared test or Fisher exact test. We determined the adjusted odds ratios of AKI with mixed logistic regression with random intercepts for each center. We adjusted for important covariates that predict AKI in the cardiac surgery setting (29), including patient demographic characteristics (age [<65, 65–75, 76–85, >85 years], sex, race), clinical risk factors (preoperative eGFR, hypertension, diabetes, myocardial infarction, congestive heart failure, cardiac catheterization in the last 48 hours), and operative characteristics (type of surgery, duration of cardiopulmonary bypass [CPB>120 minutes], and site as a random effect). We also adjusted for preoperative ACR (<10, 11–19, 30–299, or ≥300 mg/g) or dipstick proteinuria. To prevent overfitting in the model due to several covariates mentioned above, we also performed a separate analysis adjusting only for the Thakar score (30) and site as a random effect. We used area under the receiver-operating characteristic curve (AUC) to determine the ability of the proteinuria measures to discriminate between patients with and those without AKI. We quantified the improvement of proteinuria measures on AKI risk prediction with the continuous net reclassification index (NRI) and integrated discrimination improvement (IDI) indices (31). We compared AUCs using the test developed by DeLong et al. (32). We performed the analyses in SAS software, version 9.2 (SAS Institute, Cary, NC), and R 2.10.1 (R Foundation for Statistical Computing, Vienna, Austria).
Results
Among the 1238 patients enrolled in the study, 19 patients were excluded for various reasons and 21 patients did not have postoperative urine albumin measured, leaving 1198 patients for analysis (Figure 1). Patients were grouped into quintiles based on their degree of albuminuria in the 6 hours after surgery and into four groups based on their degree of dipstick proteinuria. When compared with the lowest quintile of albuminuria, patients in the highest quintile differed on some characteristics. Patients in the highest quintile were more likely to be male and have congestive heart failure, a lower preoperative eGFR, CPB time >120 minutes, postoperative intra-aortic balloon pump use, and more preoperative albuminuria. Unexpectedly, patients in the lowest quintile had higher rates of cardiopulmonary bypass use and cardioplegia. Patients in the highest quintile of albuminuria also had higher levels of renal tubular injury biomarkers at 0–6 hours (IL-18 and NGAL) (Table 1).
Figure 1.
Final selection of study population from patients meeting inclusion and exclusion criteria.
Table 1.
Patient characteristics by quintiles of albuminuria in the 6 hours after cardiac surgery
| Characteristic | Postoperative Albuminuria (mg/L) | P Value for Trend | ||||
|---|---|---|---|---|---|---|
| Q1 (≤5.5) n=237 | Q2 (5.6–10.4) n=244 | Q3 (10.5–20.8) n=237 | Q4 (20.9–48.8) n=241 | Q5 (≥48.9) n=239 | ||
| Mean age at time of surgery (yr) | 70.9±11.1 | 71.5±9.4 | 71.8±9.9 | 71.9±10.3 | 71.3±9.5 | 0.8 |
| Age <65 yr at time of surgery, n (%) | 55 (23) | 51 (21) | 57 (24) | 48 (20) | 51 (21) | |
| Men, n (%) | 148 (62) | 163 (67) | 161 (68) | 173 (72) | 167 (70) | 0.04 |
| White patients, n (%) | 225 (95) | 227 (93) | 225 (95) | 229 (95) | 215 (90) | 0.1 |
| Diabetes, n (%) | 93 (39) | 93 (38) | 105 (44) | 108 (45) | 101 (42) | 0.2 |
| Hypertension, n (%) | 181 (76) | 18 (77) | 188 (79) | 192 (80) | 198 (83) | 0.06 |
| Myocardial infarction, n (%) | 63 (28) | 61 (25) | 58 (25) | 59 (25) | 66 (28) | 0.9 |
| Congestive heart failure, n (%) | 53 (22) | 54 (22) | 52 (23) | 58 (24) | 86 (36) | 0.001 |
| Mean ejection fraction (%) | 54.0±12.5 | 51.6±12.8 | 52.0±12.9 | 50.2±12.4 | 49.6±13.3 | 0.0002 |
| Cardiac catheterization in last 48 hr, n (%) | 15 (6) | 22 (9) | 11 (5) | 11 (5) | 12 (17) | 0.14 |
| Median preoperative urine albumin (mg/L) | 7.5 (3.8–22.7) | 10 (6–21.2) | 11.5 (5.5–32.8) | 15.5 (7.1–50.9) | 24.8 (9.0–133.7) | <0.001 |
| Renal function | ||||||
| Median preoperative serum creatinine (mg/dl) | 1.0 (0.8–1.1) | 1.0 (0.9–1.2) | 1.0 (0.9–1.2) | 1.1 (0.9–1.3) | 1.1 (0.9–1.3) | <0.001 |
| Mean preoperative eGFR (ml/min per 1.73 m2) | 70.3±19.0 | 67.7±18.8 | 68.7±18.6 | 65.1±20.2 | 65.0±19.9 | <0.0010 |
| Preoperative eGFR, n (%) | ||||||
| ≥90 ml/min per 1.73 m2 | 35 (15) | 25 (10) | 33 (14) | 30 (12) | 20 (8.4) | |
| 60–89 ml/min per 1.73 m2 | 130 (55) | 133 (55) | 134 (57) | 117 (49) | 128 (54) | |
| 30–59 ml/min per 1.73 m2 | 66 (28) | 82 (34) | 65 (27) | 85 (35) | 78 (33) | |
| 15–29 ml/min per 1.73 m2 | 6 (2.5) | 4 (1.6) | 5 (2.1) | 9 (3.7) | 13 (5.4) | |
| Preoperative medications, n (%) | ||||||
| β-blockers | 159 (74) | 163 (75) | 164 (77) | 158 (72) | 159 (72) | 0.5 |
| ACE inhibitors | 95 (44) | 97 (45) | 101 (48) | 99 (45) | 109 (49) | 0.3 |
| ARBs | 33 (15) | 63 (29) | 45 (21) | 48 (22) | 41 (19) | 0.9 |
| Aspirin | 158 (73) | 165 (76) | 157 (74) | 155 (71) | 168 (76) | 0.9 |
| Statins | 156 (72) | 177 (82) | 165 (78) | 147 (67) | 155 (70) | 0.06 |
| Operative characteristics | ||||||
| Status of procedure, n (%) | 0.3 | |||||
| Elective | 194 (82) | 191 (78) | 185 (78) | 197 (82) | 181 (76) | |
| Urgent or emergent | 43 (18) | 53 (22) | 52 (22) | 44 (18) | 58 (24) | |
| Incidence, n (%) | 0.2 | |||||
| First CV surgery | 212 (89) | 215 (88) | 210 (89) | 209 (87) | 205 (86) | |
| Reoperation CV surgery | 25 (11) | 29 (12) | 27 (11) | 32 (13) | 34 (14) | |
| Type of surgery, n (%) | 0.2 | |||||
| CABG | 113 (48) | 123 (50) | 127 (54) | 109 (45) | 105 (44) | |
| Valve | 67 (28) | 72 (30) | 60 (25) | 79 (33) | 70 (29) | |
| CABG and valve | 57 (24) | 49 (20) | 50 (21) | 53 (22) | 64 (27) | |
| CPB use, n (%) | <0.001 | |||||
| Full | 217 (92) | 221 (91) | 211 (89) | 205 (85) | 189 (79) | |
| None | 14 (5.9) | 15 (6.2) | 21 (10) | 30 (12) | 40 (17) | |
| Combination | 6 (2.5) | 8 (3.3) | 2 (0.8) | 6 (2.5) | 10 (4.2) | |
| CPB time > 120 min, n (%)a | 72 (30) | 72 (30) | 78 (33) | 102 (42) | 113 (47) | <0.001 |
| Mean cross-clamp time (min) | 78.0±38.8 | 74.2±33.8 | 74.8±44.1 | 74.1±46.8 | 87.9±53.6 | 0.07 |
| Cardioplegia, n (%) | 218 (92) | 228 (93) | 211 (89) | 207 (86) | 195 (82) | <0.001 |
| Postoperative IABP, n (%) | 7 (3.0) | 8 (3.3) | 8 (3.4) | 9 (3.8) | 22 (9.2) | 0.002 |
| Median injury biomarkers | ||||||
| 0–6 hr urine IL-18 (pg/ml) | 4.0 (1.8–8.9) | 5.9 (2.7–13.9) | 12.0 (5.4–32.1) | 25.7 (9.3–78.8) | 79.2 (17.4–478.0) | <0.001 |
| 0–6 hr urine NGAL (ng/ml) | 3.2 (1.4–7.1) | 5.3 (2.6–11.3) | 10.5 (5.0–37.0) | 23.5 (9.5–162.8) | 92.3 (15.1–799.1) | <0.001 |
| 0–6 hr plasma NGAL (ng/ml) | 149 (103–233) | 182 (121–259) | 186 (125–270) | 210 (129–293) | 203 (129–300) | <0.001 |
Means are expressed with SDs. Medians are expressed with interquartile ranges. Q = quintile; eGFR, estimated GFR; ACE, angiotensin-converting enzyme; ARB, angiotensin-receptor blocker; CV, cardiovascular; CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass; IABP, intra-aortic balloon pump; NGAL, neutrophil gelatinase-associated lipocalin.
Perfusion time is reported for the patients who had CPB
Outcomes
During the hospital stay, 56 (4.7%) patients developed AKI, defined as the receipt of acute dialysis or a doubling of the serum creatinine from the baseline preoperative value. The median time to AKI was 3 days after surgery (IQR, 2–3.5 days). Of the patients who developed AKI, 32% developed AKI by day 2 and 79% by day 4 (Supplemental Table 1). Twenty-nine patients (12%) in the highest quintile of albuminuria 0–6 hours after surgery developed AKI, compared with six patients (2.5%) in the lowest quintile of albuminuria (Table 2). Higher levels of postoperative albuminuria were statistically associated with the greatest risk for AKI (adjusted relative risk [RR], 2.97 [95% confidence interval (CI), 1.20–6.91], for the highest compared with the lowest quintile). Higher levels of postoperative dipstick proteinuria were associated with a greater risk for AKI (adjusted RR, 2.46 [95% CI, 1.16–4.97], for the highest compared with the lowest grouping). Higher ACR was not statistically associated with AKI risk (adjusted RR, 1.66 [95% CI, 0.68–3.90], for the highest compared with the lowest quintile) (Table 3). Higher levels of all three proteinuria measures were associated with a higher incidence of death or dialysis (composite outcome) during the hospital stay. After adjustment, none of the proteinuria measures were statistically associated with the length of intensive care unit stay. After adjustment, only the highest grouping of dipstick proteinuria was statistically associated with a greater length of hospital stay (Table 3).
Table 2.
Patient outcomes by postoperative albuminuria
| Outcome | Postoperative Albuminuria (mg/L) | P Value for Trend | ||||
|---|---|---|---|---|---|---|
| Q1 (≤5.5) n=237 | Q2 (5.6–10.4) n=244 | Q3 (10.5–20.8) n=237 | Q4 (20.9–48.8) n=241 | Q5 (≥48.9) n=239 | ||
| Stage II or higher AKI, n (%)a | 6 (2.5) | 2 (0.8) | 12 (5.1) | 7 (2.9) | 29 (12) | <0.001 |
| stage I or higher AKI, n (%)a | 63 (27) | 69 (28) | 77 (32) | 85 (35) | 120 (50) | <0.001 |
| Median length of ICU stay (d) | 2 (1–3) | 2 (1–3) | 2 (1–3) | 2 (1–3) | 2 (1–4) | 0.6 |
| Median length of hospital stay (d) | 6 (5–8) | 6 (5–7) | 6 (5–8) | 6 (5–9) | 7 (5–10) | <0.001 |
| Dialysis or mortality, n (%) | 1 (0.4) | 3 (1.2) | 6 (2.5) | 5 (2.1) | 13 (5.4) | 0.0004 |
| Dialysis, n (%) | 0 (0) | 0 (0) | 4 (1.7) | 1 (0.4) | 11 (4.6) | <0.001 |
| Mortality, n (%) | 1 (0.4) | 3 (1.2) | 4 (1.7) | 4 (1.7) | 6 (2.5) | 0.06 |
Medians are expressed with interquartile ranges. Q, quintile.
Stage I AKI defined by a 50% or 0.3 mg/dl increase in serum creatinine. Stage II AKI defined by a doubling in serum creatinine.
Table 3.
Association of postoperative proteinuria measures with outcomes
| Quintile (cut points) | Primary Outcome: AKIa | Other Outcomes | ||||||
|---|---|---|---|---|---|---|---|---|
| AKI* Cases (%) | Unadjusted RR (95% CI) | Adjusted RR Model 1 (95% CI)b | Adjusted RR Model 2 (95% CI) c | Adjusted RR Model 3 (95% CI) d | In-Hospital Death or Dialysis (%) | Median Length of Stay in ICU (IQR) (d) | Median Length of Stay in Hospital (IQR) (d) | |
| Postoperative albumin | ||||||||
| Q1 (0–5.5 mg/L), n=237 | 2.5 | 1 | 1 | 1 | 1 | 0.4 | 2 (1–3) | 6 (5–8) |
| Q2 (5.6–10.4 mg/L), n=244 | 0.8 | 0.32 (0.07–1.59) | 0.28 (0.05–1.38) | 0.25 (0.05–1.26) | 0.29 (0.06–1.44) | 1.2 | 2 (1–3) | 6 (5–7) |
| Q3 (10.5–20.8 mg/L), n=237 | 5.1 | 2.00 (0.76–5.24) | 1.65 (0.61–4.26) | 1.37 (0.49–3.67) | 1.84 (0.69–4.65) | 2.5 | 2 (1–3) | 6 (5–8) |
| Q4 (20.9–48.8 mg/L), n=241 | 2.9 | 1.15 (0.39–3.36) | 0.78 (0.25–2.36) | 0.57 (0.17–1.83) | 0.92 (0.31–2.70) | 2.1 | 3 (1–3) | 6 (5–9) |
| Q5 (48.9–2009.7 mg/L), n=239 | 12.1 | 4.79 (2.03–11.33) | 3.37 (1.38–7.63) | 2.97 (1.20–6.91) | 3.85 (1.63–8.42) | 5.4 | 2 (1–4) | 7 (5–10) |
| Unadjusted P for trend | <0.001 | 0.0004 | 0.58 | <0.001 | ||||
| Adjusted P for trendb | 0.0002 | 0.01 | 0.18 | 0.05 | ||||
| Adjusted P for trendc | 0.001 | 0.02 | 0.35 | 0.21 | ||||
| Adjusted P for trendd | <0.001 | |||||||
| Postoperative urine albumin-to-creatinine ratio | ||||||||
| Q1 (0–23 mg/g), n=239 | 3.4 | 1 | 1 | 1 | 1 | 0.8 | 2 (1–3) | 6 (5–7) |
| Q2 (24–53 mg/g), n=240 | 4.6 | 1.37 (0.56–3.34) | 1.19 (0.47–2.91) | 1.09 (0.43–2.71) | 1.18 (0.47–2.84) | 1.3 | 2 (1–3) | 6 (5–8) |
| Q3 (54–101 mg/g), n=240 | 2.5 | 0.75 (0.26–2.12) | 0.74 (0.25–2.11) | 0.65 (0.22–1.89) | 0.69 (0.24–1.95) | 1.7 | 2 (1–3) | 6 (5–8) |
| Q4 (102–189 mg/g), n=240 | 5.0 | 1.49 (0.62–3.59) | 1.27 (0.50–3.08) | 0.97 (0.36–2.49) | 1.32 (0.54–3.13) | 2.5 | 2 (1–3) | 6 (5–9) |
| Q5 (190–42,3813 mg/g), n=239 | 8.0 | 2.37 (1.06–5.32) | 2.01 (0.85–4.51) | 1.66 (0.68–3.90) | 2.05 (0.90–4.44) | 5.4 | 2 (1–4) | 7 (5–10) |
| Unadjusted P for trend | 0.03 | 0.0007 | <0.001 | <0.001 | ||||
| Adjusted P for trendb | 0.09 | 0.01 | 0.58 | 0.18 | ||||
| Adjusted P for trendc | 0.31 | 0.02 | 0.52 | 0.19 | ||||
| Adjusted P for trendd | 0.07 | |||||||
| Postoperative dipstick proteinuria | ||||||||
| Negative, n=577 | 3.3 | 1 | 1 | 1 | 1 | 1.9 | 2 (1–3) | 6 (5–8) |
| Trace, n=249 | 2.8 | 0.85 (0.36–2.00) | 0.69 (0.28–1.66) | 0.75 (0.30–1.83) | 0.78 (0.33–1.84) | 0.8 | 2 (1–3) | 6 (5–8) |
| 30+, n=203 | 5.9 | 1.80 (0.89–3.63) | 1.23 (0.57–2.58) | 1.38 (0.63–2.94) | 1.48 (0.71–3.00) | 3.0 | 2 (1–3) | 7 (5–9) |
| 100++, 300+++, or ≥2000, n=133 | 12.0 | 3.65 (1.93–6.91) | 2.25 (1.09–4.47) | 2.46 (1.16–4.97) | 2.94 (1.51–5.49) | 6.0 | 2 (1–4) | 7 (6–10) |
| Unadjusted P for trend | <0.001 | 0.01 | 0.05 | <0.001 | ||||
| Adjusted P for trendb | 0.002 | 0.03 | 0.12 | 0.009 | ||||
| Adjusted P for trendcc | 0.001 | 0.04 | 0.16 | 0.04 | ||||
| Adjusted P for trendd | <0.001 | |||||||
RR, relative risk; CI, confidence interval; ICU, intensive care unit; IQR, interquartile range; Q, quintile.
AKI defined as a doubling in serum creatinine.
Clinical model includes age (<65, 65–75, 76–85, >85 years), sex, white race, estimated GFR, diabetes, hypertension, myocardial infarction, congestive heart failure, cardiac catheterization in the last 48 hours, type of surgery, cardiopulmonary bypass time >120 minutes, and site as random effect.
In addition to above, includes preoperative albuminuria for the categories of postoperative albumin and ACR or preoperative dipstick for the categories of postoperative dipstick proteinuria.
Thakar score (30), site adjusted as random effect.
Improvement in Risk Prediction
The AUCs for albuminuria and ACR are shown in Figure 2. Constructing an AUC for dipstick proteinuria was not possible because this is not a continuous measure; however, sensitivities and specificities for each grouping are outlined in Figure 2. Early postoperative albuminuria improved the clinical prediction of AKI over clinical models that considered pre- and intraoperative factors (clinical model AUC, 0.75; improved to 0.81 with the addition of albuminuria; P=0.006). Similar improvements with albuminuria were seen in IDI (0.036; P<0.001) and NRI (0.55; P<0.001). ACR resulted in some improvement of the clinical prediction of AKI, but this was not statistically significant (clinical model AUC, 0.75; improved to 0.77 with the addition of ACR; P=0.07). Similarly, IDI and NRI were not statistically significant for ACR (IDI, 0.0035, P=0.34; NRI, 0.19, P=0.18). The NRI for dipstick proteinuria was statistically significant (IDI, 0.0074, P=0.21; NRI, 0.28, P=0.04) (Table 4).
Figure 2.
Receiver-operating characteristic curve of AKI with first measure of postoperative proteinuria. AKI was defined as the receipt of acute dialysis or the doubling of serum creatinine during the hospital stay. Urine albumin-to-creatinine ratio: to convert from mg/g to mg/mmol, multiply by 0.113. ACR, albumin-to-creatinine ratio; AUC, area under the receiver-operating characteristic curve; LR+, positive likelihood ratio; LR−, negative likelihood ratio; NPV, negative predictive value; PPV, positive predictive value.
Table 4.
Area under the receiver-operating characteristic curve, continuous net reclassification index, and integrated discrimination improvement of clinical model with postoperative proteinuria measures
| Variable | AUC | IDIa | NRIa | |||||
|---|---|---|---|---|---|---|---|---|
| Biomarker | Biomarker + Clinical Modelb | Clinical Modelb | P Valuec | IDI | P Value | NRI | P Value | |
| Postoperative urine albumin | 0.70±0.04 | 0.81±0.03 | 0.75±0.04 | 0.006 | 0.036±0.0089 | <0.001 | 0.55±0.14 | <0.001 |
| Postoperative ACR | 0.59±0.04 | 0.77±0.03 | 0.07 | 0.0035±0.0037 | 0.34 | 0.19±0.14 | 0.18 | |
| Postoperative dipstick | 0.77±0.04 | 0.74±0.04 | 0.07 | 0.0074±0.0059 | 0.21 | 0.28±0.14 | 0.04 | |
Unless otherwise noted, values are expressed as the mean ± SEM. AUC, area under receiver-operating characteristic curve; IDI, integrated discrimination improvement; NRI, net reclassification index.
IDI and continuous NRI quantify the improvement of the biomarkers on predicting the risk for AKI. In comparison of the clinical model to the biomarker plus the clinical model, continuous NRI considers an improvement in reclassification as any increase in the model-based predicted probabilities after the addition of the biomarker for a patient with AKI or a decrease in probabilities for a patient without AKI. Similarly, a worse reclassification occurs if a patient with AKI has a decrease in probabilities or if a patient without AKI has an increase in probabilities. Overall, NRI is the difference in the proportion of improvements in reclassification and the proportion of worse reclassifications.
Clinical model includes age (<65, 65–75, 76–85, >85 years), sex, white race, estimated GFR, diabetes, hypertension, myocardial infarction, congestive heart failure, cardiac catheterization in last 48 hours, type of surgery, cardiopulmonary bypass time >120 minutes, preoperative albuminuria (<10, 10–29, 30–299, >300 mg/g), or preoperative dipstick proteinuria (negative, trace, 30–99, >100 mg/dl), and site.
P value for AUC of biomarker plus clinical model compared with clinical model.
Trends in Albuminuria and Creatinine Over Time
Patients with postoperative AKI had higher levels of albuminuria and a higher ACR before surgery and on all postoperative measures compared with patients who did not develop AKI. Albuminuria levels increased after surgery in patients who developed AKI, with median postoperative albuminuria values about two-fold higher than preoperative values. Among patients who developed AKI, the urine albumin and ACR both increased before a measurable increase in the serum creatinine. Among those who did and did not develop AKI, urine creatinine was very similar before surgery but showed no particular pattern in the early postoperative period (Figure 3, Supplemental Figure 1, and Supplemental Table 2).
Figure 3.
Trend of albuminuria and urine creatinine over time. AKI is defined as receipt of acute dialysis or a doubling in serum creatinine during the hospital stay. Median time to AKI was 3 days (interquartile range, 2–3.5 days). Day 1 is the day of surgery, with time 0 representing the point when the patient arrived in the postoperative intensive care unit. Median values for urine albumin and urine creatinine are presented.
Supplementary Analyses
The results were no different when we performed a separate analysis adjusting only for the Thakar score (30) and site as a random effect (Table 3). We considered a secondary outcome of mild AKI, defined as a 0.3-mg/dl or 50% increase in serum creatinine. The results for quintile analyses and AUCs were similar, whereas the results for IDI and NRI were slightly different (data shown in Supplemental Tables 4–6). We also considered a secondary analysis in which individuals were analyzed on the basis of CPB use. Individuals in the highest quintile of albuminuria done on CPB had a higher risk for AKI compared with those in the lowest quintile. Unfortunately, the number of individuals who had cardiac surgery performed off cardiopulmonary bypass was too low to allow us to comment on the results (Supplemental Table 6).
Patients were divided into quintiles based on the degree of change in albuminuria (postoperative compared with preoperative values). Patients in the highest quintile (greatest observed increase in albuminuria) had a higher risk for AKI (adjusted P for trend=0.04) (Table 5 and Supplemental Table 7).
Table 5.
Association of change in albuminuria with outcomes
| Change in Postoperative Albuminuria from Day 1, 0–6 hr from Before Surgery | Primary Outcome: AKI | Other Outcomes | ||||
|---|---|---|---|---|---|---|
| AKIa Cases (%) | Unadjusted RR (95% CI) | Adjusted RR (95% CI) | In-Hospital Death or Dialysis (%) | Median Length of Stay in ICU (IQR) (d) | Median Length of Stay in Hospital (IQR) (d) | |
| Q1 (−1641 to −29), n=228 | 4.4 | 1 | 1 | 2.6 | 2 (1–4) | 6 (5–8.5) |
| Q2 (−28 to −3), n=228 | 2.6 | 0.6 (0.22–1.62) | 0.6 (0.21–1.65) | 1.3 | 2 (1–3) | 6 (5–8) |
| Q3 (−2 to 5), n=230 | 2.6 | 0.59 (0.22–1.61) | 0.6 (0.21–1.67) | 1.7 | 2 (1–3) | 6 (5–8) |
| Q4 (6–22), n=227 | 4.9 | 1.1 (0.48–2.55) | 1.08 (0.44–2.51) | 2.6 | 2 (1–3) | 6 (5–8) |
| Q5 (23–1989), n=228 | 9.2 | 2.1 (1.01–4.36) | 1.65 (0.75–3.48) | 3.5 | 2 (1–3) | 7 (5–10) |
| Unadjusted P for trend | 0.008 | 0.33 | 0.04 | 0.41 | ||
| Adjusted P for trendb | 0.04 | 0.53 | 0.69 | 0.12 | ||
RR, relative risk; CI, confidence interval; ICU, intensive care unit; IQR, interquartile range; Q, quintile.
AKI defined as ≥100% increase in serum creatinine or dialysis during entire hospitalization.
Clinical model includes age (<65, 65–75, 76–85, >85 years), sex, white, estimated GFR, diabetes, hypertension, myocardial infarction, congestive heart failure, cardiac catheterization in the last 48 hours, type of surgery, cardiopulmonary bypass time >120 minutes, and site as random effect.
Discussion
The detection of proteinuria (albuminuria and dipstick) is done using readily available tests in clinical care. Our data suggest that postoperative proteinuria can serve as a biomarker of renal injury and thus can be used as a tool to diagnose or predict AKI early in the setting of cardiac surgery.
Albuminuria is commonly accepted as a marker of chronic renal injury among diabetic patients (33,34) and is a predictor of progression to ESRD in patients with CKD (16,17). Its presence is also associated with a greater long-term risk for death and cardiovascular disease (18,19). More recently, proteinuria has become recognized as an important risk factor for the development of AKI (20,35–38). A recent study by Huang et al. found that individuals with mild and heavy proteinuria (defined by urinary dipstick) before cardiac surgery had an increased risk for postoperative AKI (20). Their study, however did not attempt to quantify proteinuria beyond dipstick, nor did they comment on the role of postoperative proteinuria. To our knowledge, only four very small studies (largest, n=69) have commented on the role of postoperative proteinuria as a predictor of AKI (21,39–41).
Our findings are also supported by those of animal studies (42–44). Ware et al. demonstrated that urine albumin increases as early as 4 hours after intrinsic renal injury but not after hemodynamically mediated AKI or obstructive nephropathy in animal models. These results were corroborated using urine samples from patients with and without AKI (43). It is proposed that tubular injury leads to decreased reabsorption of filtered albumin by the proximal tubule. Renal injury may also lead to increased expression of the gene encoding albumin in the renal cortex (43).
In our study, individuals with the highest level of postoperative albuminuria or dipstick proteinuria had the greatest risk for AKI. The top quintile of albuminuria and grouping of dipstick proteinuria showed modest specificities and positive likelihood ratios for a diagnosis of AKI, with the negative likelihood ratios being less impressive (Figure 2). As a result, low or absent levels of albuminuria or proteinuria make a diagnosis of AKI less likely but cannot reliably rule it out. AKI still occurred in 3% of individuals in the lowest quintile of albuminuria or grouping of dipstick proteinuria.
Postoperative albuminuria also significantly improved the clinical prediction of AKI above clinical models alone. Although the clinical prediction noted with dipstick proteinuria and ACR improved somewhat, this finding was not statistically significant. Despite its known utility in other settings, a higher ACR was not statistically associated with AKI risk. The poor performance of ACR in the context of cardiac surgery may be explained by variations in the urine creatinine excretion within and between individuals, which could be especially prominent when renal function is not in a steady state (45). As seen in Figure 3, urine creatinine went from a median preoperative value of 89.6 mg/dl (AKI) and 85.6 mg/dl (non-AKI) to a value of 37.5 mg/dl (AKI) and 23.1 mg/dl (non-AKI) at 0–6 hours after surgery, then increasing thereafter to a peak value of 142 mg/dl (AKI) and 122 mg/dl (non-AKI) at day 3. The trend of urine creatinine over time and among individuals with and without AKI is clearly very different from that of urine albumin. Our study also found that albuminuria increased within hours after surgery among individuals who developed AKI, well before any increase in the serum creatinine value.
We have recently published a study, using the same study population, that examined the performance of the AKI biomarkers urine IL-18, urine NGAL, and plasma NGAL in the setting of cardiac surgery. Of note, a measure as accessible and simple as albuminuria performed better than IL-18 and NGAL (AUC, 0.76, P=0.03, for IL-18 plus clinical model; AUC, 0.73, P=0.12, for urine NGAL plus clinical model; AUC, 0.75, P=0.01, for plasma NGAL plus clinical model) (in Parikh et al. [22]). In our study, postoperative albuminuria gave an AUC of 0.81 and an NRI of 0.55. We believe that this result is not only statistically significant but also clinically significant because it means that 50% of patients will be correctly reclassified (AKI versus non-AKI) after surgery. This could lead to improved patient management, discharge planning, and overall resource utilization.
Our study has several strengths. It used prospective complete specimen collection and was performed under standardized conditions in consecutive patients undergoing cardiac surgery across multiple centers in the United States and Canada. AKI was defined according to modern AKI staging systems and as an endpoint useful for future therapeutic trials (RIFLE [Injury] [26] or AKIN stage 2 [27]). This AKI endpoint, which requires doubling of serum creatinine, is also less prone to prerenal azotemia and fluctuations in baseline creatinine.
Although a linear relationship between albuminuria and risk for AKI was not readily observed, a larger sample size may be needed to reduce the 95% CIs of the estimates in order to discern between an underlying relationship characterized by a threshold level versus a linear effect. An important limitation of our study, as well as most other studies examining biomarkers of renal injury, is the use of serum creatinine as the “gold standard” for evidence of AKI. Serum creatinine itself has been shown to be a nonspecific marker of true renal injury. When nephrotoxic agents are used to induce AKI in animals and humans, the association between elevations in serum creatinine and biopsy-proven kidney injury has been found to be modest (46). Several novel biomarkers were actually found to be superior to serum creatinine when histologic evidence of kidney injury was used as the reference standard (47).
Another limitation is that we cannot comment on the mechanism of increase in albuminuria. Evidence from animal studies suggests that albuminuria is causally related to tubular injury. However, it is possible that the inflammatory response after cardiac surgery rather than tubular injury itself causes an increase in albuminuria. For this reason, the use of albuminuria as a biomarker in other models of AKI warrants further study. In addition, because most of our study participants were white, the generalizability of our results to persons of other racial groups needs to be determined.
There has been a recent movement to find more sensitive and specific biomarkers of AKI. Given that AKI is a significant condition that occurs commonly in cardiac surgery (1) and general hospital settings (9, 48), we need inexpensive, accessible methods of diagnosis. We also need methods of early detection so that therapy may be instituted when the disease process may be more amenable to treatment and recovery. AKI treatment trials conducted thus far in humans have potentially failed because of the lack of an available method for the early detection of AKI (10). Promising future interventional trials may use biomarkers, as opposed to serum creatinine alone, to allow for earlier and more specific detection of AKI and therefore a longer therapeutic window in which to intervene. Although further study is needed before proteinuria can be used in clinical settings as a diagnostic tool for AKI, our results demonstrate it to be a promising biomarker. If future studies support our results, early postoperative proteinuria, alone or in conjunction with other biomarkers, such as IL-18 and NGAL, could be used to identify a population that is at high risk for AKI. This may facilitate selective enrollment in future trials that test therapeutic interventions for AKI.
Disclosures
None.
Supplementary Material
Acknowledgments
The research in this article was supported by the American Heart Association Clinical Development award and grant R01HL-085757 from the National Heart, Lung, and Blood Institute. The study was also supported by CTSA grant UL1 RR024139 from the National Center for Research Resources.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
This article contains supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.13421211/-/DCSupplemental.
See related editorial, “We Can Diagnose AKI “Early”,” on pages 1741–1742.
References
- 1.Rosner MH, Okusa MD: Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol 1: 19–32, 2006 [DOI] [PubMed] [Google Scholar]
- 2.Zanardo G, Michielon P, Paccagnella A, Rosi P, Caló M, Salandin V, Da Ros A, Michieletto F, Simini G: Acute renal failure in the patient undergoing cardiac operation. Prevalence, mortality rate, and main risk factors. J Thorac Cardiovasc Surg 107: 1489–1495, 1994 [PubMed] [Google Scholar]
- 3.Loef BG, Epema AH, Smilde TD, Henning RH, Ebels T, Navis G, Stegeman CA: Immediate postoperative renal function deterioration in cardiac surgical patients predicts in-hospital mortality and long-term survival. J Am Soc Nephrol 16: 195–200, 2005 [DOI] [PubMed] [Google Scholar]
- 4.Dasta JF, Kane-Gill SL, Durtschi AJ, Pathak DS, Kellum JA: Costs and outcomes of acute kidney injury (AKI) following cardiac surgery. Nephrol Dial Transplant 23: 1970–1974, 2008 [DOI] [PubMed] [Google Scholar]
- 5.Conlon PJ, Stafford-Smith M, White WD, Newman MF, King S, Winn MP, Landolfo K: Acute renal failure following cardiac surgery. Nephrol Dial Transplant 14: 1158–1162, 1999 [DOI] [PubMed] [Google Scholar]
- 6.Chertow GM, Levy EM, Hammermeister KE, Grover F, Daley J: Independent association between acute renal failure and mortality following cardiac surgery. Am J Med 104: 343–348, 1998 [DOI] [PubMed] [Google Scholar]
- 7.Bahar I, Akgul A, Ozatik MA, Vural KM, Demirbag AE, Boran M, Tasdemir O: Acute renal failure following open heart surgery: Risk factors and prognosis. Perfusion 20: 317–322, 2005 [DOI] [PubMed] [Google Scholar]
- 8.Hobson CE, Yavas S, Segal MS, Schold JD, Tribble CG, Layon AJ, Bihorac A: Acute kidney injury is associated with increased long-term mortality after cardiothoracic surgery. Circulation 119: 2444–2453, 2009 [DOI] [PubMed] [Google Scholar]
- 9.Coca SG, Yusuf B, Shlipak MG, Garg AX, Parikh CR: Long-term risk of mortality and other adverse outcomes after acute kidney injury: A systematic review and meta-analysis. Am J Kidney Dis 53: 961–973, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Park M, Coca SG, Nigwekar SU, Garg AX, Garwood S, Parikh CR: Prevention and treatment of acute kidney injury in patients undergoing cardiac surgery: A systematic review. Am J Nephrol 31: 408–418, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Waikar SS, Bonventre JV: Creatinine kinetics and the definition of acute kidney injury. J Am Soc Nephrol 20: 672–679, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Siew ED, Ware LB, Ikizler TA: Biological markers of acute kidney injury. J Am Soc Nephrol 22: 810–820, 2011 [DOI] [PubMed] [Google Scholar]
- 13.Doi K, Yuen PS, Eisner C, Hu X, Leelahavanichkul A, Schnermann J, Star RA: Reduced production of creatinine limits its use as marker of kidney injury in sepsis. J Am Soc Nephrol 20: 1217–1221, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Devarajan P: Emerging urinary biomarkers in the diagnosis of acute kidney injury. Expert Opin Med Diagn 2: 387–398, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.American Society of NephrologyAmerican Society of Nephrology renal research report, J Am Soc Nephrol 16: 1886–1903, 2005 [DOI] [PubMed] [Google Scholar]
- 16.Ohashi Y, Sakai K, Tanaka Y, Mizuiri S, Aikawa A: Reappraisal of proteinuria and estimated GFR to predict progression to ESRD or death for hospitalized chronic kidney disease patients. Ren Fail 33: 31–39, 2011 [DOI] [PubMed] [Google Scholar]
- 17.Landray MJ, Emberson JR, Blackwell L, Dasgupta T, Zakeri R, Morgan MD, Ferro CJ, Vickery S, Ayrton P, Nair D, Dalton RN, Lamb EJ, Baigent C, Townend JN, Wheeler DC: Prediction of ESRD and death among people with CKD: The Chronic Renal Impairment in Birmingham (CRIB) prospective cohort study. Am J Kidney Dis 56: 1082–1094, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Glassock RJ: Is the presence of microalbuminuria a relevant marker of kidney disease? Curr Hypertens Rep 12: 364–368, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gansevoort RT, Nauta FL, Bakker SJ: Albuminuria: All you need to predict outcomes in chronic kidney disease? Curr Opin Nephrol Hypertens 19: 513–518, 2010 [DOI] [PubMed] [Google Scholar]
- 20.Huang TM, Wu VC, Young GH, Lin YF, Shiao CC, Wu PC, Li WY, Yu HY, Hu FC, Lin JW, Chen YS, Lin YH, Wang SS, Hsu RB, Chang FC, Chou NK, Chu TS, Yeh YC, Tsai PR, Huang JW, Lin SL, Chen YM, Ko WJ, Wu KD, National Taiwan University Hospital Study Group of Acute Renal Failure : Preoperative proteinuria predicts adverse renal outcomes after coronary artery bypass grafting. J Am Soc Nephrol 22: 156–163, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Devarajan P, Krawczeski CD, Nguyen MT, Kathman T, Wang Z, Parikh CR: Proteomic identification of early biomarkers of acute kidney injury after cardiac surgery in children. Am J Kidney Dis 56: 632–642, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Parikh CR, Coca SG, Thiessen-Philbrook H, Shlipak MG, Koyner JL, Wang Z, Edelstein CL, Devarajan P, Patel UD, Zappitelli M, Krawczeski CD, Passik CS, Swaminathan M, Garg AX, TRIBE-AKI Consortium : Postoperative biomarkers predict acute kidney injury and poor outcomes after adult cardiac surgery. J Am Soc Nephrol 22: 1748–1757, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Shlipak MG, Coca SG, Wang Z, Devarajan P, Koyner JL, Patel UD, Thiessen-Philbrook H, Garg AX, Parikh CR, TRIBE-AKI Consortium : Presurgical serum cystatin C and risk of acute kidney injury after cardiac surgery. Am J Kidney Dis 58: 366–373, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, STROBE Initiative : The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. J Clin Epidemiol 61: 344–349, 2008 [DOI] [PubMed] [Google Scholar]
- 25.Parikh CR, Coca SG, Thiessen-Philbrook H, Shlipak MG, Koyner JL, Wang Z, Edelstein CL, Devarajan P, Patel UD, Zappitelli M, Krawczeski CD, Passik CS, Swaminathan M, Garg AX: Postoperative biomarkers predict acute kidney injury and poor outcomes after adult cardiac surgery. J Am Soc Nephrol 22: 1748–1757, 2011 [DOI] [PMC free article] [PubMed]
- 26.Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P, Acute Dialysis Quality Initiative workgroup : Acute renal failure - definition, outcome measures, animal models, fluid therapy and information technology needs: The Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 8: R204–R212, 2004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A, Acute Kidney Injury Network : Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 11: R31, 2007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF, 3rd, Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, Coresh J, CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) : A new equation to estimate glomerular filtration rate. Ann Intern Med 150: 604–612, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mehta RH, Grab JD, O’Brien SM, Bridges CR, Gammie JS, Haan CK, Ferguson TB, Peterson ED, Society of Thoracic Surgeons National Cardiac Surgery Database Investigators : Bedside tool for predicting the risk of postoperative dialysis in patients undergoing cardiac surgery. Circulation 114: 2208–2216, quiz 2208, 2006 [DOI] [PubMed] [Google Scholar]
- 30.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 16: 162–168, 2005 [DOI] [PubMed] [Google Scholar]
- 31.Pencina MJ, D’Agostino RB, Sr, Steyerberg EW: Extensions of net reclassification improvement calculations to measure usefulness of new biomarkers. Stat Med 30: 11–21, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.DeLong ER, DeLong DM, Clarke-Pearson DL: Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44: 837–845, 1988 [PubMed] [Google Scholar]
- 33.Ruilope L, Izzo J, Haller H, Waeber B, Oparil S, Weber M, Bakris G, Sowers J: Prevention of microalbuminuria in patients with type 2 diabetes: What do we know? J Clin Hypertens (Greenwich) 12: 422–430, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.de Zeeuw D: Albuminuria: a target for treatment of type 2 diabetic nephropathy. Semin Nephrol 27: 172–181, 2007 [DOI] [PubMed] [Google Scholar]
- 35.James MT, Hemmelgarn BR, Wiebe N, Pannu N, Manns BJ, Klarenbach SW, Tonelli M, Alberta Kidney Disease Network : Glomerular filtration rate, proteinuria, and the incidence and consequences of acute kidney injury: A cohort study. Lancet 376: 2096–2103, 2010 [DOI] [PubMed] [Google Scholar]
- 36.Hsu RK, Hsu CY: Proteinuria and reduced glomerular filtration rate as risk factors for acute kidney injury. Curr Opin Nephrol Hypertens 20: 211–217, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Hsu CY, Ordoñez JD, Chertow GM, Fan D, McCulloch CE, Go AS: The risk of acute renal failure in patients with chronic kidney disease. Kidney Int 74: 101–107, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Grams ME, Astor BC, Bash LD, Matsushita K, Wang Y, Coresh J: Albuminuria and estimated glomerular filtration rate independently associate with acute kidney injury. J Am Soc Nephrol 21: 1757–1764, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Tallgren M, Niemi T, Pöyhiä R, Raininko E, Railo M, Salmenperä M, Lepäntalo M, Hynninen M: Acute renal injury and dysfunction following elective abdominal aortic surgery. Eur J Vasc Endovasc Surg 33: 550–555, 2007 [DOI] [PubMed] [Google Scholar]
- 40.Dittrich S, Kurschat K, Dähnert I, Vogel M, Müller C, Alexi-Meskishvili V, Lange PE: Renal function after cardiopulmonary bypass surgery in cyanotic congenital heart disease. Int J Cardiol 73: 173–179, 2000 [DOI] [PubMed] [Google Scholar]
- 41.Aiba M, Murakami A, Funami M, Takaba T: [A study of glomerular damage due to extracorporeal circulation—meaning and usefulness of measuring of urinary microalbumin]. Nippon Kyobu Geka Gakkai Zasshi 39: 68–75, 1991 [PubMed] [Google Scholar]
- 42.Yu Y, Jin H, Holder D, Ozer JS, Villarreal S, Shughrue P, Shi S, Figueroa DJ, Clouse H, Su M, Muniappa N, Troth SP, Bailey W, Seng J, Aslamkhan AG, Thudium D, Sistare FD, Gerhold DL: Urinary biomarkers trefoil factor 3 and albumin enable early detection of kidney tubular injury. Nat Biotechnol 28: 470–477, 2010 [DOI] [PubMed] [Google Scholar]
- 43.Ware LB, Johnson AC, Zager RA: Renal cortical albumin gene induction and urinary albumin excretion in response to acute kidney injury. Am J Physiol Renal Physiol 300: F628–F638, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Bolisetty S, Agarwal A: Urine albumin as a biomarker in acute kidney injury. Am J Physiol Renal Physiol 300: F626–F627, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Garde AH, Hansen AM, Kristiansen J, Knudsen LE: Comparison of uncertainties related to standardization of urine samples with volume and creatinine concentration. Ann Occup Hyg 48: 171–179, 2004 [DOI] [PubMed] [Google Scholar]
- 46.Hotchkiss RS, Swanson PE, Freeman BD, Tinsley KW, Cobb JP, Matuschak GM, Buchman TG, Karl IE: Apoptotic cell death in patients with sepsis, shock, and multiple organ dysfunction. Crit Care Med 27: 1230–1251, 1999 [DOI] [PubMed] [Google Scholar]
- 47.Bonventre JV, Vaidya VS, Schmouder R, Feig P, Dieterle F: Next-generation biomarkers for detecting kidney toxicity. Nat Biotechnol 28: 436–440, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Hoste EA, Kellum JA, Katz NM, Rosner MH, Haase M, Ronco C: Epidemiology of acute kidney injury. Contrib Nephrol 165: 1–8, 2010 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.