Abstract
Background:
Acute kidney injury (AKI) is a common complication of cardiac surgery. Post-procedural AKI is a risk factor for 30-day readmission. We sought to examine the association of AKI and kidney injury biomarkers with readmission after cardiac surgery.
Methods:
Patients alive at discharge who underwent cardiac surgery from the TRIBE-AKI cohort were enrolled from six medical centers in the United States and Canada. AKI duration was defined as the total number of days AKI was present during index admission (no AKI, 1–2, 3–6, 7+ days). Preoperative and postoperative urinary levels were collected for interleukin-18, Neutrophil gelatinase-associated lipocalin, Kidney Injury Molecule-1, liver-fatty-acid-binding protein, cystatin C, microalbumin, creatinine, and albumin-to-creatinine ratio. Readmission and death events were identified through US (Medicare) and Canadian administrative databases at 30-days and 365-days after discharge.
Results:
Of 968 patients, 15.9% were readmitted or died within 30-days of discharge and 35.9% were readmitted or died within 365-days. AKI duration of 3–6 days was significantly associated with 30-day readmission or mortality (adjusted OR: 1.82% CI:1.08–3.05). Patients with AKI duration ≥7 days had increased odds of readmission or death at both 30 (adjusted OR: 2.49% CI:1.15–5.43) and 365 days (adjusted OR: 3.67% CI:1.73–7.79). Urinary biomarkers had no association with readmission and mortality.
Conclusions:
AKI duration ≥3 days, and not kidney biomarkers, was strongly associated with readmission or mortality. Readmission and mortality are potentially due to cardiovascular or hemodynamic causes rather than intrinsic injury to the kidney parenchyma.
INTRODUCTION
Acute kidney injury (AKI) after cardiac surgery occurs in up to 30% of patients and is associated with substantial morbidity and mortality.1 AKI duration may be as important as AKI severity to predict readmission or mortality among cardiac surgery patients.2,3 Prior research has demonstrated that AKI duration after cardiac surgery is directly proportional to in-hospital and long-term mortality.4 Furthermore, AKI duration was shown to provide additional predictive evidence for 5-year mortality compared to Acute Kidney Injury Network (AKIN) stages alone.4
Several urinary biomarkers have been evaluated for their ability to predict AKI after cardiac surgery. Neutrophil gelatinase-associated lipocalin (NGAL), an inflammatory biomarker, is one of the first, most robust proteins up-regulated in the kidney in response to ischemia or exposure to nephrotoxins.5 The pro-inflammatory cytokine, interleukin-18 (IL-18), is induced and activated in the proximal tubules.6 Kidney Injury Molecule-1 (KIM-1), a trans-membrane protein, is also overexpressed in the proximal tubules in ischemic and nephrotoxic AKI.7 Additionally, preoperative proteinuria has reliably predicted postoperative AKI and 30-day readmission.8 However, the relationship between AKI duration and clinical utility of urinary biomarkers with readmission has not yet been evaluated. Therefore, we sought to examine the association between (1) AKI duration and severity and (2) a panel of urinary biomarkers with readmission or mortality following cardiac surgery.
MATERIAL AND METHODS
Study Population
Patients at high risk for AKI, undergoing coronary artery bypass grafting (CABG) and/or valve surgery from the Translational Research Investigating Biomarker Endpoints-Acute Kidney Injury (TRIBE-AKI) cohort, were enrolled from six medical centers in the United States from July 2007 to December 2009.9,10 High risk was defined as one or more of the following: emergency surgery, preoperative serum creatinine (SCr) >2mg/dL, ejection fraction <35% or grade 3 or 4 left ventricular dysfunction, age >70 years, diabetes mellitus, concomitant CABG and valve surgery or repeat revascularization surgery. Excluding patients with evidence of AKI before surgery, prior kidney transplantation, preoperative SCr level > 4.5 mg/dL, or end-stage renal disease left 1,219 patients. After excluding patients who died prior to discharge (n=20) and those who could not be linked to administrative claims data to assess readmission and mortality (n=231), 968 patients comprised the final sample. Included patients provided written informed consent to participate in this study. Each participating hospital’s institutional review board approved the conduct of this investigation.
AKI Duration and Staging
AKI, defined according to AKIN criteria, is at least a 0.3 mg/dL or 50% increase in postoperative SCr compared to baseline. AKI duration was categorically defined as no AKI and AKI lasting 1–2, 3–6, and ≥7 days.11 Baseline creatinine was the last preoperative SCr obtained prior to surgery. Postoperative SCr values from routine clinical care were recorded for every patient throughout the length of stay (LOS).
Urinary Biomarkers
Urine specimens were collected preoperatively, and then daily for three postoperative days. For the first 72 hours after surgery, urine samples were collected upon arrival to the intensive care unit (ICU) and then every six hours. The following nine urinary biomarkers were analyzed: IL-18, NGAL, KIM-1, liver-fatty acid binding protein (L-FABP), cystatin, albumin, microalbumin, creatinine, and albumin-to-creatinine ratio (ACR). Details of biomarker assays and coefficients of variation are described in prior studies.10,12–14
Outcomes
The primary and secondary outcomes were readmission or death within 30-days (n=154) and 365-days (n=348) of discharge, respectively, following cardiac surgery. For TRIBE-AKI participants living in the United States, vital status was obtained by calling participants’ homes, reviewing hospital records and searching the National Death Index. For Canadian participants, we called participants’ homes and utilized data from the Registered Persons Database from the Institute for Clinical Evaluative Sciences (ICES). Readmission, defined as overnight admission to an acute care facility, excluded emergency department visits, rehabilitation visits and patient transfers. For participants living in the United States, readmission data was obtained though linkages with Center for Medicare and Medicaid Services databases. These datasets were linked using unique, encoded identifiers and analyzed at the ICES.
Statistical Analysis
Baseline patient and disease characteristics and clinical outcomes were summarized by percentage and mean (± standard deviation). We used χ2 tests and Wilcoxon rank-sum tests to assess similarities between categories of 30-day and 365-day readmission or mortality. We examined relationships between indicators of AKI (including AKI severity, duration, and urinary biomarkers for AKI) and risk of readmission or death within 30-days or 365-days after discharge from cardiac surgery. Urinary biomarkers were log-transformed and not corrected for urinary creatinine.
Logistic regression models were used to examine relationships between AKI severity, duration, urinary biomarker levels and 30-day and 365-day readmission or mortality, adjusted for the Society of Thoracic Surgeon (STS) score, which predicts postoperative dialysis following cardiac surgery15 (Model 1). Model 1 variables include preoperative SCr, age, surgery type, diabetes, recent myocardial infarction, race, reoperation, New York Heart Association class, and cardiogenic shock. STS includes chronic lung disease15 which was not included in TRIBE-AKI; we excluded it with the assumption that no patients had the disease. A second multivariable model (Model 2) which includes the STS score and delta SCr also adjusted for change in peak SCr from preoperative. Area under the curve was calculated to evaluate the discriminatory power of the adjusted logistic regression models for the primary and secondary outcomes. To determine if AKI, defined by SCr, influences relationships between biomarkers and readmission, an interaction variable between AKI and biomarker levels was included in each model as a secondary analysis. In order to represent a homogeneous cohort, an interaction term was included in our models between exposure (AKI stage or duration) and surgery type. All analyses were performed using SAS 9.3 (SAS Institute, Inc, Cary, NC) software.
RESULTS
30-day Readmission or Mortality
Of 968 cardiac surgery patients who met eligibility criteria, 15.9% (n=154) were readmitted (n=146) or died (n=8) within 30-days of live discharge. Patients who experienced readmission or death within 30-days (Table 1) were less likely to be white (p=0.02) and more likely to have congestive heart failure (CHF) (p<0.01) and longer cross-clamp times (p=0.03).
Table 1:
Patient Characteristics
| 30-day Readmission or death | 365-day Readmission or death | ||||
|---|---|---|---|---|---|
| Characteristic | Overall | Yes | No | Yes | No |
| N | 968 | 154 | 814 | 348 | 620 |
| Demographics | |||||
| Age at time of surgery, mean±SD | 73.39±8.28 | 73.76±9.02 | 73.32±8.14 | 73.94±8.87 | 73.09±7.93 |
| Men | 69.2% | 68.2% | 69.4% | 66.1% | 71.0% |
| White | 96.2% | 92.9% | 96.8% | 95.7% | 96.5% |
| Medical History time of surgery | |||||
| Diabetes | 37.8% | 41.6% | 37.1% | 42.0% | 35.5% |
| Hypertension | 79.8% | 74.0% | 80.8% | 77.6% | 81.0% |
| CHF | 23.6% | 32.5% | 21.9% | 31.0% | 19.4% |
| LVEF <40% | 9.6% | 13.0% | 9.0% | 11.8% | 8.4% |
| Aortic Occlusion | 78.4% | 78.8% | 78.3% | 75.5% | 79.9% |
| Cardioplegia | 88.6% | 87.0% | 88.9% | 86.4% | 89.8% |
| IABP | 3.3% | 4.8% | 3.0% | 2.7% | 4.5% |
| Previous myocardial infarction | 25.2% | 24.7% | 25.3% | 25.0% | 25.3% |
| eGFR mL/min per 1.73m2, mean±SD | 66.50±18.36 | 65.73±17.76 | 66.64±18.48 | 64.42±19.48 | 67.66±17.61 |
| eGFR >60 | 65.1% | 62.3% | 65.6% | 58.9% | 68.5% |
| eGFR 30–60 | 31.9% | 34.4% | 31.4% | 36.5% | 29.4% |
| eGFR <30 | 3.0% | 3.2% | 2.9% | 4.6% | 2.1% |
| SCr(mg/dL), median [IQR] | 1 (1–1) | 1 (1–1) | 1 (1–1) | 1 (1–1) | 1 (1–1) |
| Surgical Characteristics | |||||
| Elective Surgery | 18.8% | 22.7% | 18.1% | 22.7% | 16.6% |
| Surgery | |||||
| CABG AND Valve | 22.0% | 33.8% | 19.8% | 25.6% | 20.0% |
| Off-pump | 10.0% | 10.4% | 10.0% | 11.2% | 9.4% |
| Re-do surgery | 1.4% | 1.3% | 1.5% | 1.7% | 1.3% |
| Perfusion time (minutes), mean±SD | 109.74±56.22 | 115.97±57.83 | 108.56±55.87 | 112.55±59.62 | 108.18±54.22 |
| Cross-clamp time (minutes), mean±SD | 75.17±42.69 | 82.33±45.56 | 73.82±42.03 | 77.67±45.53 | 73.78±41.01 |
| Diseased Coronary Vessels (#) | |||||
| None | 24.9% | 20.1% | 25.8% | 26.1% | 24.2% |
| One | 14.5% | 17.5% | 13.9% | 15.8% | 13.7% |
| Two | 19.9% | 31.2% | 17.8% | 33.3% | 44.0% |
| Three | 40.2% | 29.2% | 42.3% | 23.6% | 17.9% |
| Postoperative Complications | |||||
| Clinical AKI | |||||
| Delta SCr ≥100% | 4.2% | 5.8% | 3.9% | 6.3% | 3.1% |
| Acute dialysis | 0.9% | * | 1.0% | * | * |
| Oliguria in first day, | 1.3% | * | 1.1% | 2.0% | 1.0% |
| Non-renal complications (#), | |||||
| 0 | 59.5% | 54.5% | 60.4% | 56.9% | 61.0% |
| 1–2 | 31.7% | 32.5% | 31.6% | 30.7% | 32.3% |
| >2 | 8.8% | 13.0% | 8.0% | 12.4% | 6.8% |
| Ventilator >48 hours, | 3.9% | 3.9% | 3.9% | 4.0% | 3.9% |
| ICU LOS, median [IQR] | 2 (1–3) | 2 (1–4) | 2 (1–3) | 2 (1–4) | 2 (1–3) |
| Hospital LOS, median [IQR] | 6 (5–9) | 7 (6–10) | 6 (5–8) | 7 (5–10) | 6 (5–8) |
Did not meet suppression requirements
Compared to patients who did not experience AKI, patients with AKIN Stage 1 had 1.40-fold higher odds of readmission or mortality at 30-days (95% CI:0.97–2.02, Table 2). Patients with AKIN Stage 2 had increased odds of readmission or mortality at 30-days of 2.37 (95% CI:0.94–5.95). AKIN Stage 3 was not associated with the primary outcome. AKI duration of 1–2 days was not associated with 30-day readmission or mortality, however AKI durations of 3–6 days and 7+ days were associated with an increased odds ratio of 1.82 (95% CI:1.08–3.05) and 2.49 (95% CI:1.15–5.43) respectively, compared to patients who did not experience AKI (Table 2). The interaction between AKI exposure and surgery type was not significant (Tables 2&3), and we did not find an association between preoperative urinary biomarkers and the primary outcome (Table 3). Apart from ACR, we did not observe a relationship between postoperative biomarkers and the primary outcome. Peak levels of ACR (adjusted OR: 1.22; 95% CI:1.03–1.45) were significantly associated with 30-day readmission or mortality in unadjusted analyses. After adjustment, these relationships were no longer statistically significant (Table 4).
Table 2:
Logistic Regression Models for AKI Stage and Duration
| 30-day Readmission or Death | 365-day Readmission or Death | |||||
|---|---|---|---|---|---|---|
| OR (95% CI) Unadjusted | OR (95% CI) Adjusted for STS Score | Interaction p-value for exposure & surgery type | OR (95% CI) Unadjusted | OR (95% CI) Adjusted for STS Score | Interaction p-value for exposure & surgery type | |
| AKI Stage | ||||||
| No AKI (n=620) | referent | referent | 0.54 | referent | referent | 0.60 |
| Stage 1 (n=308) | 1.47 (1.02–2.12) | 1.40 (0.97–2.02) | 1.59 (1.20–2.11) | 1.54 (1.16–2.04) | ||
| Stage 2 (n=25) | 2.42 (0.98–5.95) | 2.37 (0.94–5.95) | 2.73 (1.22–6.13) | 2.73 (1.21–6.19) | ||
| Stage 3 (n=15) | 0.96 (0.21–4.31) | 0.71 (0.15–3.25) | 1.43 (0.50–4.08) | 1.15 (0.40–3.31) | ||
| AKI Duration | ||||||
| No AKI (n=620) | referent | referent | 0.27 | referent | referent | 0.37 |
| 1–2 days (n=215) | 1.13 (0.73–1.74) | 1.11 (0.72–1.72) | 1.40 (1.02–1.94) | 1.39 (1.01–1.93) | ||
| 3–6 days (n=100) | 1.96 (1.18–3.27) | 1.82 (1.08–3.05) | 1.69 (1.10–2.59) | 1.60 (1.03–2.46) | ||
| 7+ days (n=33) | 3.11 (1.45–6.63) | 2.49 (1.15–5.43) | 4.29 (2.04–9.02) | 3.67 (1.73–7.79) | ||
Table 3:
Logistic Regression Models for Preoperative Urine Biomarkers
| 30-day Readmission or Death | 365-day Readmission or Death | |||||
|---|---|---|---|---|---|---|
| Log-transformed Urine Biomarker | OR (95% CI) Unadjusted | OR (95% CI) Adjusted for STS Score | Interaction p-value for exposure & surgery type | OR (95% CI) Unadjusted | OR (95% CI) Adjusted for STS Score | Interaction p-value for exposure & surgery type |
| IL-18 | 0.97 (0.85–1.11) | 1.00 (0.87–1.14) | 0.95 | 0.93 (0.84–1.03) | 0.95 (0.85–1.05) | 0.44 |
| NGAL | 0.95 (0.81–1.10) | 0.93 (0.80–1.08) | 0.83 | 1.02 (0.91–1.14) | 1.00 (0.90–1.12) | 0.13 |
| KIM-1 | 1.11 (0.96–1.29) | 1.12 (0.96–1.30) | 0.41 | 1.05 (0.94–1.17) | 1.05 (0.94–1.18) | 0.84 |
| L-FABP | 1.07 (0.94–1.21) | 1.07 (0.94–1.21) | 0.55 | 1.09 (0.99–1.19) | 1.08 (0.99–1.19) | 0.39 |
| Microalbumin | 1.08 (0.97–1.22) | 1.06 (0.95–1.19) | 0.61 | 1.11 (1.02–1.21) | 1.09 (1.00–1.19) | 0.39 |
| ACR | 1.07 (0.96–1.20) | 1.04 (0.93–1.17) | 0.38 | 1.13 (1.03–1.23) | 1.10 (1.01–1.21) | 0.36 |
| Cystatin C | 0.99 (0.77–1.27) | 1.00 (0.77–1.29) | 0.34 | 0.94 (0.78–1.13) | 0.94 (0.78–1.14) | 0.15 |
Table 4:
Logistic Regression Model Results for Postoperative Urine Biomarkers
| 30-day Readmission or Death | 365-day Readmission or Death | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Log-transformed Urine Biomarker | Time point | OR (95% CI) Unadjusted | OR (95% CI) Adjusted Model 1 | OR (95% CI) Adjusted Model 2 | Interaction P-value for exposure & surgery type | OR (95% CI) Unadjusted | OR (95% CI) Adjusted Model 1 | OR (95% CI) Adjusted Model 2 | Interaction P-value for exposure & surgery type |
| IL-18 | Day 1 0–6 Hours | 1.00 (0.91–1.09) | 1.00 (0.91–1.09) | 1.00 (0.91–1.10) | 0.58 | 1.02 (0.95–1.10) | 1.02 (0.95–1.10) | 1.06 (0.98–1.14) | 0.27 |
| Peak | 0.90 (0.79–1.04) | 0.92 (0.81–1.06) | 0.90 (0.78–1.03) | 0.98 | 0.94 (0.85–1.04) | 0.96 (0.86–1.06) | 0.93 (0.84–1.04) | 0.49 | |
| NGAL | Day 1 0–6 Hours | 0.99 (0.91–1.08) | 0.98 (0.90–1.07) | 0.98 (0.90–1.07) | 0.79 | 0.99 (0.92–1.05) | 0.98 (0.92–1.04) | 0.99 (0.93–1.06) | 0.64 |
| Peak | 0.95 (0.85–1.07) | 0.95 (0.84–1.07) | 0.93 (0.83–1.05) | 0.31 | 1.00 (0.92–1.10) | 1.00 (0.91–1.09) | 0.99 (0.90–1.08) | 0.46 | |
| KIM-1 | Day 1 0–6 Hours | 1.07 (0.93–1.24) | 1.07 (0.93–1.24) | 1.08 (0.93–1.26) | 0.64 | 1.08 (0.97–1.21) | 1.08 (0.97–1.21) | 1.12 (1.00–1.26) | 0.09 |
| Peak | 1.01 (0.82–1.23) | 1.04 (0.85–1.28) | 1.02 (0.83–1.25) | 0.85 | 0.96 (0.82–1.11) | 0.98 (0.84–1.15) | 0.96 (0.82–1.12) | 0.75 | |
| L-FABP | Day 1 0–6 Hours | 1.03 (0.95–1.12) | 1.01 (0.93–1.10) | 1.02 (0.93–1.11) | 0.17 | 1.04 (0.97–1.10) | 1.02 (0.96–1.09) | 1.04 (0.97–1.11) | 0.39 |
| Peak | 0.99 (0.89–1.11) | 0.98 (0.87–1.10) | 0.97 (0.86–1.08) | 0.0098 | 1.04 (0.95–1.13) | 1.03 (0.94–1.12) | 1.02 (0.93–1.11) | 0.15 | |
| Microalbumin | Day 1 0–6 Hours | 1.01 (0.89–1.15) | 0.99 (0.87–1.13) | 1.00 (0.87–1.14) | 0.77 | 1.06 (0.96–1.17) | 1.04 (0.94–1.15) | 1.07 (0.96–1.18) | 0.85 |
| Peak | 1.04 (0.88–1.23) | 1.02 (0.86–1.20) | 1.00 (0.84–1.18) | 0.0154 | 1.10 (0.97–1.25) | 1.08 (0.95–1.23) | 1.07 (0.94–1.21) | 0.12 | |
| ACR | Day 1 0–6 Hours | 1.08 (0.94–1.25) | 1.03 (0.89–1.19) | 1.03 (0.89–1.20) | 0.93 | 1.11 (1.00–1.24) | 1.07 (0.96–1.20) | 1.08 (0.96–1.20) | 0.19 |
| Peak | 1.22 (1.03–1.45) | 1.15 (0.97–1.37) | 1.14 (0.96–1.36) | 0.22 | 1.24 (1.09–1.42) | 1.19 (1.04–1.36) | 1.18 (1.03–1.35) | 0.47 | |
| Cystatin C | Day 1 0–6 Hours | 0.92 (0.73–1.16) | 0.96 (0.76–1.20) | 0.95 (0.76–1.20) | 0.34 | 0.89 (0.74–1.06) | 0.90 (0.76–1.08) | 0.92 (0.77–1.11) | 0.26 |
| Peak | 0.86 (0.70–1.05) | 0.89 (0.73–1.09) | 0.88 (0.72–1.08) | 0.27 | 0.85 (0.73–0.99) | 0.88 (0.75–1.02) | 0.87 (0.74–1.01) | 0.69 | |
A significant interaction effect was observed for AKI on readmission for preoperative urine L-FABP (p<0.001) and microalbumin (p=0.04). These biomarkers showed an increased odds of readmission of 1.07(p<0.001) and 1.08 (p=0.04) respectively for those who experienced AKI. A significant interaction effect was also observed for AKI on readmission for postoperative urine ACR (p=0.04) as well as Day 1 (p<0.001) and peak L-FABP (p=0.02). Peak urine L-FABP showed an increased odds of readmission for those who experienced AKI of 1.05 while Day 1 L-FABP and peak ARC showed a decreased odds of readmission of 0.98 and 0.95 respectively.
365-day Readmission or Mortality
For the secondary outcome, 35.9% (n=348) of cardiac surgery patients were readmitted (n=331) or died (n=17) within 365-days of live discharge (Table 1). These patients were more likely to have diabetes (p=0.05), CHF (p<0.01), lower estimated glomerular filtration rate (eGFR) (p<0.01) and longer median hospital LOS (p<0.01).
After adjustment, AKIN Stage 1 (OR: 1.54; 95% CI:1.16–2.04) and AKIN Stage 2 (OR 2.73; 95% CI:1.21–6.19) were associated with readmission or mortality at 365-days of follow-up. AKI duration of 1–2 days was associated with readmission or mortality at 365-days (OR: 1.39; 95% CI:1.01–1.93). Additionally, AKI duration ≥7 days was associated with 365-day readmission or mortality (OR: 3.67; 95% CI:1.73–7.79) (Table 2). In unadjusted analyses, preoperative urine microalbumin (OR: 1.11; 95% CI:1.02–1.21) and ACR (OR: 1.13; 95% CI:1.03–1.23, Table 3) were significantly associated with readmission or death within 365-days of discharge but were no longer statistically significant after adjustment. The interaction term between exposure and surgery type was not significant (Tables 2&3). Postoperative ACR levels collected within 6 hours of ICU admission (OR: 1.11; 95% CI:1.00–1.24) and at peak (OR: 1.24; 95% CI:1.09–1.42) were significantly associated with readmission or mortality at 365-days in unadjusted analyses (Table 4). After adjustment, these relationships remained similar, but were no longer statistically significant.
DISCUSSION
In patients undergoing cardiac surgery, postoperative AKI duration was significantly associated with readmission or mortality at 30 and 365-days. Compared to no AKI, patients with AKI lasting 7 or more days had 3.25 and 4.55 greater odds of experiencing readmission or mortality at 30 and 365-days, respectively. Our findings suggest statistically significant associations between AKI duration, but not urinary biomarkers, with 30-day and 365-day readmission and mortality following cardiac surgery. We are the first to report on a lack of association of a panel of novel urinary biomarkers with up to 365-day readmission and mortality. Other measures of AKI severity demonstrate a strong statistically significant association with this outcome. Efforts to reduce AKI duration after cardiac surgery may have profound short and long-term implications for patients, providers and payers.
Despite prior research demonstrating an association between urinary biomarkers for AKI and adverse patient outcomes16, readmission events may not be as strongly correlated.
Immense attention has been placed on improving early detection of AKI and implementing quality improvement efforts to mitigate known risk factors for AKI.10,17,18 Increasing AKI severity, defined by AKIN Stages, is correlated with 30-day readmission and long-term mortality after cardiac surgery.19 However, AKI duration may be as important a risk factor for readmission and mortality. Previously, we demonstrated that AKI severity was significantly associated with 5-year readmission or mortality in 1,690 patients undergoing isolated CABG surgery.20 Additionally, we recently observed strong relationships between AKI severity and duration with 30-day and 365-day readmission or mortality. Compared to AKI severity and duration, we found the association between urinary biomarkers and readmission or mortality to be weak and largely non-significant. This finding may suggest that creatinine-based definitions of AKI may be more strongly associated with readmission events after cardiac surgery than urinary biomarkers.
The primary strength of our study is the use of the TRIBE-AKI cohort, designed to investigate the relationship between novel biomarkers and the detection of early AKI.21 This was a large prospective cohort comprised of patients from multiple healthcare institutions across the U.S. and Canada, which increases the generalizability of our results. Supplemental analyses allow the reader to make a comparison between the effect sizes of the relationship between urinary biomarkers versus creatinine-based definitions of AKI and readmission or mortality. Methods used to collect and process urinary specimens were standardized across clinical sites and validated assay were used to measure all biomarkers.
Recently, other novel biomarkers have demonstrated an association with AKI and readmission following cardiac surgery22–25 Interleukin-1 receptor, serum soluble ST2, is significantly associated with AKI following cardiac surgery.24 Galectin-3, an inflammatory biomarker, enhances the predictability of AKI when measured preoperatively.23 By integrating these and other cardiac biomarkers, including N-terminal pro-brain natriuretic peptide, cystatin C, interleukin-6, and interleukin-10 researchers have increased the power of risk prediction models for readmission and mortality.22,26 These models, however, may not be as robust against variability in region, complexity in patient comorbidity, or variation in study design.22 The presently analyzed suite of urinary biomarkers measured by TRIBE-AKI seemingly lacks the predictive power of those above.
Limitations of this study should be considered. Given that this study reviewed patients at high risk for AKI from cardiac surgery, results may not be generalizable to other patient populations. Uncollected protocol deviations and additional interoperative parameters, if measured, would allow for analysis of variations in procedure as markers for readmission and mortality. Other important techniques which predict kidney injury and severity, such as the Doppler Renal Resistive Index,27 were also not included. Furthermore, this analysis is restricted to urinary biomarkers, and although recent studies have revealed connections between novel biomarkers and AKI, these were not measured or were not urinary biomarkers.25,28 Biomarker levels were measured from spot sampling rather than 24-hour collections. However, spot sampling the absolute concentration of biomarkers may be just as effective as 24-hour collection in predicting adverse outcomes. 29 Additionally, there is the potential that AKI duration is acting as a surrogate marker for LOS or injury severity. We attempted to correct for this by controlling for AKIN stage.
Urinary biomarkers may be better proxies for renal damage compared to renal function. Due to renal reserve, SCr levels may not rise and eGFR may not drop until more than half of renal function has been lost.30 Concurrently, levels of urinary biomarkers may significantly increase indicating severe kidney damage. However, it is the immediate loss of renal function and associated complications (e.g. hypertension, anemia, azotemia) that has been repeatedly shown to predict readmission and mortality, especially in heart failure.31
Conclusion
Preoperative and postoperative urinary biomarkers were not associated with readmission or mortality at either 30-days or 365-days after adjustment. Creatinine-based definitions of AKI were more strongly associated with readmission or death at 30 or 365-days, respectively, and may better identify cardiac surgery patients at high risk for readmission.
ACKNOWLEDGEMENTS:
This study was supported by R01HL119664 and R01HL085757, K24DK090203, and P30 DK079310–07. Dr Parikh is a member of NIH-sponsored U01DK082185. Dr Coca was supported in part by U01DK106926. Dr Garg was supported by Dr. Adam Linton, Chair in Kidney Health Analytics and Canadian Institutes of Health Research.
Common Abbreviations
- AKI(N)
Acute Kidney Injury (Network)
- CABG
Coronary Artery Bypass Grafting
- ICES
Institute for Clinical Evaluative Sciences
- ICU
Intensive Care Unit
- IL-18
Interleukin-18
- KIM-1
Kidney Injury Molecule-1
- L-FABP
Liver-fatty-acid-binding protein
- LOS
Length of Stay
- NGAL
Neutrophil gelatinase-associated lipocalin
- SCr
Serum Creatinine
- STS
Society of Thoracic Surgeons (score)
- TRIBE-AKI
Translational Research Investigating Biomarker Endpoints-Acute Kidney Injury
Footnotes
DISCLOSURES:
Dr Coca received consulting fees from Goldfinch Bio and Janssen Pharmaceuticals and is on the scientific advisory board of pulseData LLC. All other authors declare no conflicting financial interests with respect to the research, authorship, and/or publication of this article.
REFERENCES
- 1.Rosner MH, Okusa MD. Acute Kidney Injury Associated with Cardiac Surgery. Clinical Journal of the American Society of Nephrology 2006;1(1):19–32. [DOI] [PubMed] [Google Scholar]
- 2.Brown JR, Parikh CR, Ross CS, et al. Impact of perioperative acute kidney injury as a severity index for thirty-day readmission after cardiac surgery. The Annals of thoracic surgery 2014;97(1):111–117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Machado MN, Nakazone MA, Maia LN. Prognostic value of acute kidney injury after cardiac surgery according to kidney disease: improving global outcomes definition and staging (KDIGO) criteria. PloS one 2014;9(5):e98028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Brown JR, Kramer RS, Coca SG, Parikh CR. Duration of acute kidney injury impacts long-term survival after cardiac surgery. The Annals of thoracic surgery 2010;90(4):1142–1148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Devarajan P Neutrophil gelatinase-associated lipocalin (NGAL): A new marker of kidney disease. Scandinavian journal of clinical and laboratory investigation Supplementum 2008;241:89–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Parikh CR, Mishra J, Thiessen-Philbrook H, et al. Urinary IL-18 is an early predictive biomarker of acute kidney injury after cardiac surgery. Kidney international 2006;70(1):199–203. [DOI] [PubMed] [Google Scholar]
- 7.Han WK, Bailly V, Abichandani R, Thadhani R, Bonventre JV. Kidney Injury Molecule-1 (KIM-1): a novel biomarker for human renal proximal tubule injury. Kidney international 2002;62(1):237–244. [DOI] [PubMed] [Google Scholar]
- 8.Wahl TS, Graham LA, Morris MS, et al. Association Between Preoperative Proteinuria and Postoperative Acute Kidney Injury and Readmission. JAMA Surg 2018;153(9):e182009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Shlipak MG, Coca SG, Wang Z, et al. Presurgical serum cystatin C and risk of acute kidney injury after cardiac surgery. Am J Kidney Dis 2011;58(3):366–373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Parikh CR, Coca SG, Thiessen-Philbrook H, et al. Postoperative biomarkers predict acute kidney injury and poor outcomes after adult cardiac surgery. Journal of the American Society of Nephrology : JASN 2011;22(9):1748–1757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mehta RL, Kellum JA, Shah SV, et al. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Critical care (London, England) 2007;11(2):R31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Parikh CR, Thiessen-Philbrook H, Garg AX, et al. Performance of kidney injury molecule-1 and liver fatty acid-binding protein and combined biomarkers of AKI after cardiac surgery. Clin J Am Soc Nephrol 2013;8(7):1079–1088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Koyner JL, Garg AX, Shlipak MG, et al. Urinary cystatin C and acute kidney injury after cardiac surgery. Am J Kidney Dis 2013;61(5):730–738. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Molnar AO, Parikh CR, Sint K, et al. Association of postoperative proteinuria with AKI after cardiac surgery among patients at high risk. Clinical Journal of the American Society of Nephrology : CJASN 2012;7(11):1749–1760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chang CH, Fu CM, Yang CH, et al. Society of Thoracic Surgeons score predicts kidney injury in patients not undergoing bypass surgery. The Annals of thoracic surgery 2015;99(1):123–129. [DOI] [PubMed] [Google Scholar]
- 16.Bonventre JV. Kidney injury molecule-1 (KIM-1): a urinary biomarker and much more. Nephrology Dialysis Transplantation 2009;24(11):3265–3268. [DOI] [PubMed] [Google Scholar]
- 17.Coca SG, Parikh CR. Urinary Biomarkers for Acute Kidney Injury: Perspectives on Translation. Clinical Journal of the American Society of Nephrology 2008;3(2):481–490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Parikh CR, Abraham E, Ancukiewicz M, Edelstein CL. Urine IL-18 is an early diagnostic marker for acute kidney injury and predicts mortality in the intensive care unit. Journal of the American Society of Nephrology : JASN 2005;16(10):3046–3052. [DOI] [PubMed] [Google Scholar]
- 19.Bjessmo S Increased rehospitalization rate after coronary bypass operation for acute coronary syndrome: a prospective study in 200 patients. The Annals of thoracic surgery 2009;88(4):1148–1152. [DOI] [PubMed] [Google Scholar]
- 20.Brown JR, Hisey WM, Marshall EJ, et al. Acute Kidney Injury Severity and Long-Term Readmission and Mortality After Cardiac Surgery. The Annals of thoracic surgery 2016;102(5):1482–1489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Parikh CR, Moledina DG, Coca SG, Thiessen-Philbrook HR, Garg AX. Application of new acute kidney injury biomarkers in human randomized controlled trials. Kidney international 2016;89(6):1372–1379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Brown JR, Jacobs JP, Alam SS, et al. Utility of Biomarkers to Improve Prediction of Readmission or Mortality After Cardiac Surgery. The Annals of thoracic surgery 2018;106(5):1294–1301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Wyler von Ballmoos M, Likosky DS, Rezaee M, et al. Elevated preoperative Galectin-3 is associated with acute kidney injury after cardiac surgery. BMC nephrology 2018;19(1):280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lobdell KW, Parker DM, Likosky DS, et al. Preoperative serum ST2 level predicts acute kidney injury after adult cardiac surgery. J Thorac Cardiovasc Surg 2018;156(3):1114–1123 e1112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Parikh CR, Mansour SG. Perspective on Clinical Application of Biomarkers in AKI. Journal of the American Society of Nephrology : JASN 2017;28(6):1677–1685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Polineni S, Parker DM, Alam SS, et al. Predictive Ability of Novel Cardiac Biomarkers ST2, Galectin-3, and NT-ProBNP Before Cardiac Surgery. J Am Heart Assoc 2018;7(14). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Boddi M, Bonizzoli M, Chiostri M, et al. Renal Resistive Index and mortality in critical patients with acute kidney injury. Eur J Clin Invest 2016;46(3):242–251. [DOI] [PubMed] [Google Scholar]
- 28.Shahian DM, Grover FL. Biomarkers and Risk Models in Cardiac Surgery. Circulation 2014;130(12):932. [DOI] [PubMed] [Google Scholar]
- 29.Ralib AM, Pickering JW, Shaw GM, et al. Test characteristics of urinary biomarkers depend on quantitation method in acute kidney injury. Journal of the American Society of Nephrology : JASN 2012;23(2):322–333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.de Geus HRH, Betjes MG, Bakker J. Biomarkers for the prediction of acute kidney injury: a narrative review on current status and future challenges. Clinical Kidney Journal 2012;5(2):102–108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Zaya M, Phan A, Schwarz ER. Predictors of re-hospitalization in patients with chronic heart failure. World Journal of Cardiology 2012;4(2):23–30. [DOI] [PMC free article] [PubMed] [Google Scholar]