Abstract
Study Objective:
The lag in creatinine-mediated diagnosis of cardiac surgery-associated acute kidney injury (AKI) may be impeding the development of renoprotection therapies. Postoperative renal resistive index (RRI) measured by transabdominal Doppler ultrasound is a promising early AKI biomarker. RRI measured intraoperatively by transesophageal echocardiography (TEE) is available even earlier but is less evaluated. Therefore, we conducted an assessment of intraoperative RRI as an AKI biomarker using previously reported post-renal insult thresholds.
Design:
Retrospective convenience sample.
Setting:
Intraoperative
Patients:
180 adult cardiac surgical patients between July 2013 and July 2014.
Intervention:
None
Measurements:
Pre- and post-cardiopulmonary bypass (CPB) RRI thresholds, measured using intraoperative TEE, exceeding 0.74 or 0.79 were used to evaluate for an association with KDIGO AKI risk using the Chi-square test. Other consensus AKI criteria (AKIN, RIFLE) were similarly evaluated. Additional t-test analyses examined the relationship of pre- and pre-to-post (delta) CPB RRI with AKI.
Main Results:
Post-CPB RRI for 99 patients included 36 and 23 with values exceeding 0.74 and 0.79, respectively. Analyses confirmed associations of both RRI thresholds with all consensus AKI definitions (0.74; KDIGO: p=0.05, AKIN: p=0.03, RIFLE: p=0.03, 0.79; KDIGO: p=0.002, AKIN: p=0.001, RIFLE: p=0.004). In contrast, pre-CPB and pre-to post-CPB RRI were not associated with AKI.
Conclusions:
RRI obtained intraoperatively in cardiac surgery patients, assessed using previously reported thresholds, is highly associated with AKI and warrants further evaluation as a promising “earliest” AKI biomarker. These significant findings suggest that RRI assessment should be included in the standard intraoperative TEE exam.
Keywords: Renal Resistive Index, Acute Kidney Injury, Cardiac Surgery, Transesophageal Echocardiography, Biomarker
Introduction
Up to 50% of cardiac surgery patients manifest an acute kidney injury (AKI) early postoperatively[1, 2]. Postoperative AKI is associated with longer intensive care unit (ICU) and hospital stays, higher readmission rates, excess mortality, and higher cost[3–6]. Existing consensus criteria commonly require 48 hours post-insult to confidently diagnose AKI due to their reliance on serum creatinine accumulation (e.g., Kidney Disease Improving Global Outcomes; KDIGO guidelines)[7, 8]. Therefore, hastening AKI diagnosis through novel biomarker discovery has become a priority in the ongoing search for kidney protection strategies to improve cardiac surgical outcomes[7, 9]. For practical purposes, a biomarker is a single characteristic that can be objectively measured and evaluated as an indicator of a pathogenic process[10]. Many early serum and urine AKI biomarkers that correlate with renal insult (e.g., neutrophil gelatinase-associated lipocalin, kidney injury molecule) have not demonstrated clinical utility sufficient to replace existing creatinine-based AKI criteria[11]. A promising early AKI biomarker has been developed from the combination of two urine biomarker proteins, tissue inhibitor of metalloproteinases 2 (TIMP-2) and insulin-like growth factor-binding protein 7 (IGFBP7)[9, 12]. Notably, limitations of this laboratory test in cardiac surgical patients include its additional cost, poor reliability in some settings (e.g., albuminuria, bilirubin, methylene blue), and current lack of availability at most clinical centers. Renal ultrasound, as a source of AKI “imaging biomarkers”, is readily available and could be routinely performed during or immediately after cardiac surgical procedures at the majority of heart centers. However, the potential of renal ultrasound to reflect and diagnose AKI remains incompletely explored.
Renal arterial blood flow abnormalities that occur coincident with AKI may reflect renal swelling and a resultant kidney compartment syndrome[13–15]. Renal resistive index (RRI), as characterized by pulsed wave Doppler (PWD) ultrasonography, is a marker of variability in intrarenal arterial pulsatility. RRI is determined by three parameters: 1) the ratio of systolic to diastolic blood pressure (pulse pressure), 2) intrarenal vascular compliance, and 3) a combination of venous pressure and interstitial pressure (renal swelling), which is of interest as an indicator of acute renal injury[15]. Because of its potential to reflect decreased compliance and increased intracapsular pressure, RRI offers a novel approach to early AKI diagnosis (Figure 1)[16]. Transabdominal RRI in critically ill[17–20], and postoperative cardiac and orthopedic surgery patients[21–23] has identified threshold values above which AKI is significantly more prevalent[17–19, 21–27]. Notably, wherein cardiac surgery routinely involves transesophageal echocardiography (TEE), an even earlier opportunity to assess RRI exists[28]. RRI determined by transabdominal ultrasonography and TEE are highly correlated[29, 30]. However, RRI during surgery as a potential “point-of-insult” AKI biomarker is sparsely studied[29, 30]. In a small intraoperative study (60 patients), Kararmaz and colleagues found higher average RRI values post-bypass (CPB) in patients who subsequently developed AKI.
Figure 1: RRI determination using TEE.
Calculation of renal resistive index (RRI) using transesophageal echocardiography. Abbreviations: Vdia-trough diastolic velocity, Vsys-peak systolic velocity (with permission [16]).
Therefore, we used two previously reported RRI thresholds, 0.74[17, 21, 31] and 0.79[18], to conduct a preliminary assessment of intraoperative post-CPB RRI as an early AKI biomarker. In addition, we assessed pre-CPB RRI and RRI change (pre-to post-CPB) as other potential AKI biomarkers.
Methods
After Duke University institutional review board approval (consent requirement waived), a retrospective convenience sample was constructed containing all patients aged 18 years or older that underwent cardiac surgery between July 1, 2013 and July 10, 2014 with a TEE exam that included renal artery blood flow velocity measurements. Study data were collected from the electronic health record by departmental data analysts and managed using secure, web-based Research Electronic Data Capture (REDCap) hosted at Duke University. Exclusions included patients with pre-operative renal dysfunction (defined as <60 mL/min/1.73 m2 by the CKD-EPI1 equation[32]), missing pre- or post-op serum creatinine values, and patients who did not undergo CPB. Patients were also excluded by blinded TEE-certified reviewers if their renal blood flow image quality was deemed poor (e.g., significant beat-to-beat waveform variability), or if ECG rhythm at the time of imaging was other than normal sinus rhythm (e.g., atrial fibrillation). Demographic and clinical variables recorded included age, gender, admission diagnoses (hypertension, stroke, peripheral vascular disease [PVD]), body mass index (BMI), surgical procedure, preoperative hemoglobin (most recent value recorded within 1 week prior to, but not on, the day of surgery), lowest hemoglobin (Hb) recorded intraoperatively, highest lactate recorded on the day of surgery, and the durations of aortic cross clamping and CPB.
Anesthesia was administered according to the attending anesthesiologist’s preference. Typically, general anesthesia was induced with a combination of intravenous midazolam, fentanyl, and propofol, with 0.5–1 minimal alveolar concentration isoflurane for anesthesia maintenance and vecuronium or rocuronium for muscle relaxation. CPB was via a centrifugal pump with nonpulsatile flow indexed to body surface area (2–2.4 L/min/m2), cooling as indicated by surgical procedure, and vasoactive or vasodilatory support as necessary to support mean arterial pressures (target 50–80 mmHg) at the anesthesiologist’s discretion. PCO2 and PaO2 were monitored and targeted for 35–40 mmHg and 150–250 mmHg (without temperature adjustment), respectively; hematocrit was targeted for ≥0.21.
A comprehensive TEE examination (Philips iE33, Andover, MA) with a multiplane TEE probe (Philips X7–2t) was performed as a part of routine care by a TEE-certified attending anesthesiologist and supervised cardiothoracic anesthesiology fellow prior to initiation of CPB and repeated within one hour after CPB separation. As part of the TEE study, recording of the left intra-renal artery pulse wave Doppler blood flow velocity was obtained by a method described in detail by Bandyopadhyay and colleagues[28]. RRI measurement was introduced to trainees and attending anesthesiologists in July 2013 and encouraged as a part of the intraoperative TEE for all patients.
For analysis, each left renal artery PWD image was linked to details regarding timing (pre- or post-CPB), heart rate and rhythm (from the ECG tracing), then de-identified. Clinician-generated RRI measurements were masked, and systolic and diastolic renal blood flow velocities, for calculation of RRI (Figure 1), were measured for each available cardiac cycles (up to 3 per image) to generate an average RRI value, for use in all subsequent analyses. Reviewers (blinded to pre/post-CPB timing, procedural details, and AKI status) were randomly selected from among seven investigators who were certified in perioperative TEE (National Board of Echocardiography Examination of Special Competence in Advanced Perioperative Transesophageal Echocardiography [Advanced PTEeXAM ®]).
Per institutional protocol, daily pre- and postoperative serum creatinine values were measured until hospital discharge. Preoperative creatinine was defined as the most recent value recorded within 1 week prior to, but not on, the day of surgery. Daily creatinine was defined, in the case of multiple measurements, as the first creatinine value on each date. The peak postoperative creatinine value was the highest daily creatinine value in the first 10 postoperative days. Serum creatinine was measured using the Jaffe technique (UniCel DxC 800, Beckman Coulter, Brea, CA) with a normal range of 31–76 μmol/L (0.4–1.0mg/dl) for females and 46–99 μmol/L (0.6–1.3 mg/dl) for males. Occurrence of AKI was defined by the KGIDO network criteria (AKI = KDIGO stage 1, 2, or 3) [33]. Notably, KDIGO AKI determination was based solely on serum creatinine due to confounders in the interpretation of urine output after cardiac surgery[34]. For secondary comparisons, AKI was also assessed using Acute Kidney Injury Network (AKIN)[7] and Risk, Injury, Failure, Loss of kidney function, End stage kidney disease (RIFLE)[35] criteria, and a more sensitive linear variable: pre-to peak-postoperative serum creatinine change expressed as a percentage of the preoperative creatinine value (%ΔCr) [3].
Analysis Plan
We tested the validity of two previously identified RRI thresholds (0.74 and 0.79) when applied to intraoperative post-CPB TEE-determined RRI. These thresholds were developed from cohorts of ICU[17, 18, 31] or postoperative[21] patients. The Chi-square test was used to assess the association between the proportion of patients meeting these thresholds and the proportion of patients with AKI as defined by KDIGO AKI consensus criteria (primary analysis), and AKIN and RIFLE AKI consensus criteria (secondary analyses). We additionally examined the association between AKI status (as determined by KDIGO) and RRI as a continuous parameter using the Wilcoxon rank sum test due to the non-normal distribution of RRI. These assessments were repeated for intraoperative pre-CPB TEE-determined RRI using the same thresholds.
We also examined the association between post-CPB RRI thresholds and peak rise in serum creatinine relative to baseline (%ΔCr) as a continuous parameter using the Wilcoxon rank sum test due to its non-normal distribution. Finally, we examined the association between AKI status (as defined by KDIGO) and both the absolute and relative differences in pre- and post-CPB RRI (diffRRI = [pre-CPB RRI - post-CPB RRI] and %ΔRRI = [pre-CPB RRI - post-CPB RRI]/pre-CPB RRI, respectively) using the equal variance t-test. All tests were two-sided, with p<0.05 considered significant.
Results
After exclusions, a total of 243 TEE Doppler renal blood flow images from 180 cardiac surgical patients were suitable for RRI determination (144 pre-CPB, 99 post-CPB; Figure 2), including 63 patients with both waveforms. Among the 99 patients with post-CPB images, patient and procedural characteristics were similar to those in other cardiac surgical cohorts (Table 1). In this group, there was an overall 36% (36/99) rate of KDIGO AKI, and 37% (37/99) with RRI values greater than 0.74. KDIGO AKI rates were statistically higher in patients with RRI values exceeding 0.74, compared to those at or below (50 vs. 30%; p=0.05; Table 2). KDIGO AKI rates were also significantly greater in the 23 (23%) patients with RRI values exceeding 0.79, compared to those at or below (65 vs. 29%; p=0.002; Table 2). We detected similar associations with other consensus AKI criteria for both the 0.74 threshold (AKIN: 50 vs. 29%; p=0.03, RIFLE: 44 vs. 22%; p=0.03; Table 3) and 0.79 threshold (AKIN: 65 vs. 28%; p=0.001, RIFLE: 55 vs. 23%; p=0.004; Table 3). While similar trends were observed for both RRI thresholds in association with the %ΔCr, this only met statistical significance for the 0.79 threshold (Table 3).
Figure 2: STROBE flow-chart.
STROBE study population flow-chart. Abbreviations: AKI – acute kidney injury, CPB – cardiopulmonary bypass, GFR – glomerular filtration rate, RRI – renal resistive index.
Table 1.
Patient and procedural characteristics, and perioperative variables for adult cardiac surgery patients with post-CPB RRI imaging.
| No AKI (n=62) | AKI (n=37) | Total (n=99) | |
|---|---|---|---|
| Patient Characteristics | |||
| Age (years) (Q1,Q3) | 60.3 (46.9, 69.6) | 63.0 (59.6, 68.5) | 62.6 (50.9, 69.6) |
| Female gender n (%) | 24 (39%) | 14 (38%) | 38 (38%) |
| Body Mass Index (kg/m2) (Q1,Q3) | 29.0 (26.0, 33.1) | 29.5 (27.3, 33.8) | 29.1 (26.2, 33.6) |
| History of hypertension | 27 (44%) | 24 (65%) | 51 (52%) |
| History of diabetes | 12 (19%) | 12( 32%) | 24 (24%) |
| History of stroke | 1 (2%) | 2 (5%) | 3 (3%) |
| History of peripheral vascular disease | 3 (5%) | 0 (0%) | 3 (3%) |
| Procedure Characteristics | |||
| Procedure type | |||
| CABG only | 16 (26%) | 11 (30%) | 27 (27%) |
| Valve only | 25 (40%) | 15 (41%) | 40 (40%) |
| Aortic only | 4 (6%) | 1 (3%) | 5 (5%) |
| Mixed | 15 (23%) | 9 (24%) | 24 (24%) |
| Others | 2 (3%) | 1 (3%) | 3 (3%) |
| Duration of AoX Clamp (min) (Q1,Q3) | 103.5 (70.0, 135.0) | 95.0 (63.0, 127.0) | 97.0 (71.0, 132.0) |
| Duration of CPB (min) (Q1,Q3) | 161.5 (126.0, 219.0) | 146.0 (110.0, 183.0) | 161.0 (124.0, 215.0) |
| Perioperative Variables | |||
| Preoperative hemoglobin (g/dL) (Q1,Q3) | 13.6 (12.7, 14.7) | 12.6 (11.3, 14.2) | 13.3 (12.1, 14.6) |
| Lowest intraoperative hemoglobin (g/dL) (Q1,Q3) | 9.0 (7.8, 10.0) | 8.3 (7.9, 10.2) | 8.7 (7.9, 10.0) |
| Highest serum lactate level, day of surgery (mg/dL) (Q1,Q3) | 2.2 (1.8, 3.6) | 2.8 (2.1, 3.3) | 2.5 (1.9, 3.5) |
| Left ventricular ejection fraction (% patients) (intraoperative Pre-CPB) | |||
| Missing | 15 | 8 | 23 |
| 15–25% | 1 (2%) | 0 (0%) | 1 (1%) |
| 25–35% | 0 (0%) | 0 (0%) | 0 (0%) |
| 35–45% | 6 (12%) | 3 (10%) | 9 (12%) |
| 45–55% | 14 (30%) | 6 (21%) | 20 (26%) |
| >55% | 26 (55%) | 20 (69%) | 46 (61%) |
Abbreviations: AKI – acute kidney injury defined by consensus KDIGO criteria (see text), CABG – coronary artery bypass graft surgery, AoX clamp – aortic crossclamp, CPB – cardiopulmonary bypass. Continuous variables are presented as medians with Q1 and Q3 interquartile ranges.
Table 2.
The association of post-CPB TEE RRI with KDIGO (see text) AKI following cardiac surgery in 99 adult patients.
| No AKI n=62 (%) | AKI n=37 (%) | NPV and PPV | p value | |
|---|---|---|---|---|
| RRI post-CPB >0.74 (n=36) | 18 (50.0%) | 18 (50.0%) | 0.70 (NPV) | 0.050 |
| RRI post-CPB <=0.74 (n=63) | 44 (69.8%) | 19 (30.2%) | 0.50 (PPV) | |
| Specificity/Sensitivity | 0.71 (Sp) | 0.49 (Sn) | ||
| RRI post-CPB >0.79 (n=23) | 8 (37.8%) | 15 (65.2%) | 0.71 (NPV) | 0.002 |
| RRI post-CPB <=0.79 (n=76) | 54 (71.1%) | 22 (29.0%) | 0.65 (PPV) | |
| Specificity/Sensitivity | 0.87 (Sp) | 0.41 (Sn) | ||
Abbreviations: AKI – acute kidney injury, CPB – cardiopulmonary bypass, NPV – negative predictive value, PPV – positive predictive value, RRI – renal resistive index, Sn – sensitivity, Sp – specificity, TEE – transesophageal echocardiography.
Table 3.
Summary of post-CPB RRI threshold results for patients without vs. with AKI as defined by AKIN[7], RIFLE[35], or %ΔCr criteria. RIFLE data missing for 6 patients.
| AKIN yes n=36/99 | p value1 | RIFLE yes n=28/93 | p value1 | %ΔCr median (Q1,Q3) | p value2 | |
|---|---|---|---|---|---|---|
| RRI post-CPB>0.74 (n=36) | 18/36 (50.0%) | 0.03 | 15/34 (44.1%) | 0.03 | 25% (0,71) | 0.11 |
| RRI post-CPB<=0.74 (n=63) | 18/63 (28.6%) | 13/59 (22.0%) | 17% (0,33) | |||
| RRI post-CPB>0.79 9 (n=23) | 15/23 (65.2%) | 0.001 | 12/22 (54.6%) | 0.004 | 38% (11,80) | 0.005 |
| RRI post-CPB<=0.79 (n=76) | 21/76 (27.6%) | 16/71 (22.5%) | 17% (0,33) | |||
Chi-Square
Wilcoxon
Abbreviations: AKI – acute kidney injury, CPB – cardiopulmonary bypass, RRI – renal resistive index, %ΔCr – peak postoperative rise in serum creatinine relative to baseline.
We performed additional secondary RRI analyses for correlation with KDIGO AKI. These included pre-CPB RRI, and the change in RRI from pre-to post-CPB in the same patient (absolute change - diffRRI; relative change - %ΔRRI). Of the 144 patients with pre-CPB renal imaging, 64 sustained KDIGO AKI (44.4%); RRI values exceeded 0.74 and 0.79 in 43 (30%) and 19 (13%), respectively. Pre-CPB RRI was not associated with KDIGO AKI for either threshold (0.74 and 0.79; p=0.49 and p=0.44, respectively; Table 4). In the 63 patients with both pre- and post-CPB renal imaging, 23 sustained KDIGO AKI (36.5%); analyses of RRI change also found no correlation between absolute of relative RRI change and AKI (diffRRI: 0.04 for AKI, 0.04 for no AKI, p=0.52; %ΔRRI: 5% for AKI, 5% for no AKI, p=0.27; Table 4). Additionally, in analyses of RRI as a continuous variable, neither the pre- nor post-CPB RRI was significantly different between patients with and without KDIGO AKI (for 99 post-CPB patients, RRI was 0.74 (0.68, 0.81) [median (Q1,Q3)] for those with AKI vs. 0.70 (0.64, 0.76) without AKI) (Table 4).
Table 4.
The association of pre-CPB RRI, and pre-post-CPB RRI difference with KDIGO AKI following cardiac surgery in adult patients.
| No AKI n=80 (55.6%) | AKI n=64 (44.4%) | p value | |
|---|---|---|---|
| RRI pre-CPB>0.74 (n=43/144) | 22 (51.2%) | 21 (48.8%) | 0.491 |
| RRI pre-CPB>0.79 (n=19/144) | 9 (47.4%) | 10 (52.6%) | 0.441 |
| n=40 (63.5%) | n=23 (36.5%) | ||
| pre-post Diff RRI (n=63) | 0.04 (−0.04, 0.07) | 0.04 (−0.01, 0.13) | 0.292 |
| pre-post %ΔRRI (n=63) | 5% (−6%, 13%) | 5% (−1%, 20%) | 0.272 |
| median (Q1,Q3) | median (Q1,Q3) | ||
| RRI post-CPB (median, Q1,Q3) (n=99) | 0.70 (0.64, 0.76) | 0.74 (0.68, 0.81) | 0.072 |
| RRI pre-CPB (median, Q1,Q3) (n=144) | 0.70 (0.64, 0.75) | 0.69 (0.64, 0.76) | 0.522 |
Chi-Square
Wilcoxon.
Abbreviations: AKI – acute kidney injury, CPB – cardiopulmonary bypass, RRI – renal resistive index, DiffRRI – absolute difference between pre- and post-CPB RRI, %ΔRRI – relative difference between pre- and post-CPB RRI.
Discussion
We confirmed significant associations between two previously described RRI thresholds when applied to intraoperative post-CPB RRI measurements (0.74 and 0.79), with subsequent development of KDIGO AKI following cardiac surgery. Notably, these previously described early AKI biomarkers were developed in critically ill[17, 18, 31] and postoperative cardiac surgery[21] patient cohorts, respectively, but in our study were applied to assessment of intraoperative RRI values determined within 1 hour of separation from CPB using standard TEE imaging. Secondary analyses, using other AKI consensus criteria (AKIN and RIFLE), revealed similar results. Non-significant secondary analyses included associations of AKI with pre-CPB RRI, pre-to post-CPB RRI change (absolute or relative), and linear approaches to RRI analysis. Collectively these observations support inclusion of RRI assessment as a part of routine intraoperative TEE studies and highlight the importance of continued investigation to best characterize the utility of intraoperative RRI as an early (earliest) post-renal insult AKI biomarker in cardiac surgery patients.
Surgical and critical care unit AKI biomarker studies investigating renal arterial pulsatility have previously described sequential RRI determinations at time points that involve a delay from several hours to a full day following the likely occurrence of renal insult (Table 5). Although many describe strong associations of elevated RRI with AKI development and characterize the strongest “signal,” including optimal timing and RRI threshold, very few of these studies address an alternate question, whether sufficient renal blood flow abnormalities are present immediately following renal insult to avoid these delays and reflect AKI risk sooner. Notably, previous animal studies indicate that renal perfusion is highly abnormal immediately after separation from CPB[36], and it is plausible that RRI elevation may also manifest immediately in the setting of AKI. This is important since, when ultimately developed, effective renoprotection interventions will presumably be more successful when applied immediately following injury[37]. In this context, our findings, representing the largest intraoperative cardiac surgical cohort to date, support and extend observations by Regolisti and colleagues, who found correlations between AKI and RRI elevation at the end of surgery in 60 patients[30], and Kararmaz and colleagues[29, 30], who found correlations of AKI with RRI elevation within one hour following CPB in 60 patients.
Table 5.
Studies of the association of RRI with AKI in Perioperative and Critically Ill Patient Populations.
| AKI Setting | Sample Size | AKI Definition | AKI Rate (%) | Optimal RRI Threshold | U/S Method:Timing |
|---|---|---|---|---|---|
| Perioperative | |||||
| Cardiothoracic Surgery with CPB | |||||
| Bossard et al. 2011 (mixed cardiac, elective) [21] | 65 | creatinine rise ≥30% within 96h : y/n | 28 | >0.74 (ICU arrival) | TA: preop, ICU arrival |
| Guinot et al. 2013 (mixed) [47] | 82 | RIFLE: none or transient (≤3d) /persistent (>3d) | 25 | >0.73 (ICU arrival) | TA: preop, ICU arrival, +6h, +1d |
| Hertzberg et al. 2017 (mixed, elective) [48] | 96 | AKIN: rates in patients with/without RRI >0.70 | 28 | x | TA: 24h preop |
| Kararmaz et al 2014 (mixed, elective) [29] | 60 | KDIGO : y/n | 23 | >0.72 (post-CPB) | TA/TEE: anesthesia induction; TEE only: post-CPB; TA only: end of surgery |
| Regolisti et al. 2017 (mixed cardiac) [30] | 60 | creatinine increase ≥0.3 mg/dl vs. baseline within 72h : y/n | 38 | >0.71 (end of surgery) | TA/TEE: anesthesia induction, end of surgery; TA only: +4h, +24h postop |
| Sinning et al. 2013 (TAVR) [49] | 132 | VARC-2: y/n | 24 | >0.85 (4h postop) | TA: preop, +4h, +1d, +2d, +3d, +7d |
| Wu et al. 2017 (type A aortic dissection) [26] | 62 | AKIN: none or transient (<3d) / persistent (≥3d) | 66 | ≥0.725 (ICU arrival) | TA: preop, ICU arrival, +6h, +1d, +2d |
| Orthopedic Surgery | |||||
| Marty et al. 2015 (Major joint replacement) [23] | 50 | AKIN: y/n | 32 | >0.705 (PACU) | TA: preop, PACU |
| Marty et al. 2016 (Hip fracture) [50] | 48 | AKIN: y/n | 60 | >0.705 (PACU) | TA: preop, PACU |
| Critically Ill | |||||
| Surgical or Mixed ICU | |||||
| Haitsma Mulier et al 2018 [31] | 99 | KDIGO: none or stages 1–2 / 2–3 | 49 | >0.74 (stage 2/3) | TA: within 24h of ICU arrival |
| Song et al. 2018 [51] | 124 | KDIGO: y/n n patients with/without RRI>0.695 | 42 | x | TA: within 3h of ICU arrival for sepsis |
| Medical ICU | |||||
| Darmon et al. 2011 [18] | 51 | AKIN: none or transient (≤3d) / persistent (>3d) | 69 | >0.795 (persistent) | TA: post-intubation |
| Lerolle et al. 2006 [17] | 35 | RIFLE: none or R / I or F | 51 | >0.74 | TA: within 24h of ICU vasopressor initiation |
Abbreviations: AKI – acute kidney injury, AKIN – Acute Kidney Injury Network, CPB – cardiopulmonary bypass, KDIGO - Kidney Disease Improving Global Outcomes, PACU – post-anesthesia care unit, RIFLE - Risk, Injury, Failure, Loss of kidney function, End stage kidney disease, RRI – renal resistive index, TA – transabdominal, TEE – transesophageal echocardiography, VARC-2 - Valve Academic Research Consortium-2.
While the retrospective design of our study did not allow for protocol-driven patient management (e.g., regulated use of nephrotoxic medication, vasoactive agents, etc.) or creatinine determinations timed relative to renal insult, all patients participated in the standardized postoperative care map at the reference institution, that includes daily postoperative serum creatinine determination. Although the RRI image sample represents a minority of the total patients during the study period, and the 424 eligible renal blood flow measurements emerged subsequent to exclusion of 227 images due to poor image quality, these samples reflect the introductory period for RRI determination at the reference institution. Nonetheless, as this low enrollment reflects the introduction of routine intraoperative RRI determination to our practice, the study sample represents the largest intraoperative cohort to date. Renal blood flow measurement was initially overseen by two board certified echocardiographers and disseminated to all adult cardiothoracic anesthesiology faculty and fellows. Over the course of the study period, an increasing proportion of patients had renal blood flow measurements. It is impossible to determine whether factors such as difficulty in obtaining images were shared with certain patient characteristics or risk factors that could have introduced bias into the study.
Although RRI shows promise as an early AKI biomarker, a stepwise approach to biomarker validation must underpin its development as an AKI biomarker if further renoprotection advances are to be made. At this early stage, examination for proof of concept for RRI (or any other variable) as an AKI biomarker is most appropriately conducted through univariate assessments rather than including RRI as a contributing variable among other risk factors in multivariable assessment of AKI risk; hence, no multivariable analyses were conducted in the current study. However, with respect for the concerns outlined above, it is interesting to speculate on limitations of and next steps in evaluating intraoperative post-CPB RRI as potentially the earliest AKI biomarker yet available. Many AKI biomarkers have struggled to separate “signal” from unrelated background variability (i.e., “noise”). Similarly, although we have found that “raw” post-CPB RRI is associated with subsequent AKI, both RRI thresholds tested had rather weak diagnostic capability in terms of Sn/Sp and PPV/NPV (Table 2). In this regard, we have previously identified confounding factors such as aortic insufficiency that affect RRI elevation without a clear relation to AKI, contributing to background variability, particularly in the pre-CPB group when significant aortic insufficiency has not yet been corrected [38]. Because pulse pressure and venous congestion (systemic hemodynamics) are also major parameters of the RRI, there are a number of potential confounders of RRI as a valid biomarker of AKI risk, particularly in the context of cardiac surgery where extreme perturbations of hemodynamics can occur. Related to these, renal artery atherosclerosis[39], elevated pulse pressure[40] (increased arterial stiffness[41], increased aortic pulse wave velocity[42], increased peak left ventricular outflow tract velocities[43], left ventricular hypertrophy and/or degree of diagnosed hypertension[44]), venous congestion (e.g. high portal venous flow pulsatility)[45], and potentially blood viscosity (e.g., hematocrit)[46] may all impact the RRI. The fact that a moderately predictive signal is present even without controlling for these additional variables supports the view that RRI has promise as a readily available biomarker, but also suggests that further refinement would be useful. A priori adjustment for such potential confounders offers the possibility of strengthening RRI as an early AKI biomarker and highlights the need for further studies with larger sample sizes.
In summary, we present compelling evidence that intraoperative RRI elevation determined post-CPB by TEE warrants further study as currently the earliest available candidate AKI biomarker in cardiac surgery patients. Furthermore, we believe that routine RRI assessment should be included in the standard intraoperative TEE examination. Regardless of the consensus AKI definition used, and using two previously reported RRI AKI thresholds, we noted significant associations of intraoperative post-CPB RRI elevation with subsequent development of AKI in a cohort of 99 adult cardiac surgery patients. Future investigations are required to confirm these observations in other cohorts but also to better characterize the optimal intraoperative RRI threshold for AKI risk, including a more structured approach to adjustment for potential confounding factors such as aortic insufficiency.
Highlights:
Post-cardiopulmonary bypass renal resistive index correlates with acute kidney injury
RRI thresholds >0.74 and >0.79 are both associated with major consensus AKI criteria
Pre-cardiopulmonary bypass RRI does not correlate with cardiac surgery AKI
Pre- to post-bypass RRI change does not correlate with cardiac surgery AKI
Acknowledgements
Role of Funding Sources
Neither funding source had any role in study design; in the collection, analysis and interpretation of data; in the writing of the report; nor in the decision to submit the article for publication.
Funding
Research reported in this publication was supported by the National Institutes of Health under Award Numbers T32GM008600 and TL1TR001116. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
List of Abbreviations
- %ΔCr
pre-to peak-postoperative serum creatinine change, as a percentage of the preoperative creatinine value
- %ΔRRI
Relative pre-to post-CPB RRI change, as a percentage of the preoperative RRI value ([pre-CPB RRI - post-CPB RRI]/pre-CPB RRI)
- Advanced PTEeXAM
Examination of Special Competence in Advanced Perioperative Transesophageal Echocardiography
- AKI
Acute kidney injury
- AKIN
Acute Kidney Injury Network
- BMI
Body mass index
- CKD-EPI
Chronic Kidney Disease Epidemiology Collaboration
- CPB
Cardiopulmonary bypass
- diffRRI
Absolute pre-to post-CPB RRI change (pre-CPB RRI - post-CPB RRI)
- ECG
Electrocardiogram
- eGFR
Estimated glomerular filtration rate
- Hb
Hemoglobin
- ICU
Intensive care unit
- IGFBP7
Insulin-like growth factor-binding protein 7
- KDIGO
Kidney Disease Improving Global Outcomes
- NPV
Negative predictive value
- PaCO2
Partial pressure of arterial carbon dioxide
- PaO2
Partial pressure of arterial oxygen
- PPV
Positive predictive value
- PVD
Peripheral vascular disease
- PWD
Pulsed wave Doppler
- REDCap
Research Electronic Data Capture
- RIFLE
Risk, Injury, Failure, Loss of kidney function, End stage kidney disease
- RRI
Renal resistive index
- TEE
Transesophageal echocardiography
- TIMP-2
Tissue inhibitor of metalloproteinases 2
- Sn
Sensitivity
- Sp
Specificity
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Declarations:
Ethics approval and consent to participate
This study is approved by the Duke University Institutional Review Board on protocol# Pro00056123; patient consent was not deemed necessary.
Consent for publication
Not applicable
Declaration of interests
The authors declare that they have no interests to declare.
Availability of data and material
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
- [1].Arnaoutakis GJ, Bihorac A, Martin TD, Hess PJ Jr., Klodell CT, Ejaz AA, et al. RIFLE criteria for acute kidney injury in aortic arch surgery. The Journal of thoracic and cardiovascular surgery. 2007;134:1554–60; discussion 60–1. [DOI] [PubMed] [Google Scholar]
- [2].Haase M, Bellomo R, Matalanis G, Calzavacca P, Dragun D, Haase-Fielitz A. A comparison of the RIFLE and Acute Kidney Injury Network classifications for cardiac surgery-associated acute kidney injury: a prospective cohort study. The Journal of thoracic and cardiovascular surgery. 2009;138:1370–6. [DOI] [PubMed] [Google Scholar]
- [3].Brown JR, Cochran RP, Dacey LJ, Ross CS, Kunzelman KS, Dunton RF, et al. Perioperative increases in serum creatinine are predictive of increased 90-day mortality after coronary artery bypass graft surgery. Circulation. 2006;114:I409–13. [DOI] [PubMed] [Google Scholar]
- [4].Brown JR, Parikh CR, Ross CS, Kramer RS, Magnus PC, Chaisson K, 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;97:111–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].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. 2008;23:1970–4. [DOI] [PubMed] [Google Scholar]
- [6].Hobson CE, Yavas S, Segal MS, Schold JD, Tribble CG, Layon AJ, et al. Acute Kidney Injury Is Associated With Increased Long-Term Mortality After Cardiothoracic Surgery. Circulation. 2009;119:2444–53. [DOI] [PubMed] [Google Scholar]
- [7].Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, et al. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Critical care. 2007;11:R31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Group KDIGOKAKIW. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney Int. Suppl. 2012:1–138.
- [9].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. Critical care. 2013;17:R25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Biomarkers and surrogate endpoints: preferred definitions and conceptual framework. Clinical pharmacology and therapeutics. 2001;69:89–95. [DOI] [PubMed] [Google Scholar]
- [11].Parikh CR, Coca SG, Thiessen-Philbrook H, Shlipak MG, Koyner JL, Wang Z, et al. Postoperative biomarkers predict acute kidney injury and poor outcomes after adult cardiac surgery. Journal of the American Society of Nephrology : JASN. 2011;22:1748–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Cummings JJ, Shaw AD, Shi J, Lopez MG, O’Neal JB, Billings FTt. Intraoperative prediction of cardiac surgery-associated acute kidney injury using urinary biomarkers of cell cycle arrest. J Thorac Cardiovasc Surg. 2019;157:1545–53.e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Herrler T, Tischer A, Meyer A, Feiler S, Guba M, Nowak S, et al. The intrinsic renal compartment syndrome: new perspectives in kidney transplantation. Transplantation. 2010;89:40–6. [DOI] [PubMed] [Google Scholar]
- [14].Cruces P, Salas C, Lillo P, Salomon T, Lillo F, Hurtado DE. The renal compartment: a hydraulic view. Intensive Care Med Exp. 2014;2:26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].O’Neill WC. Renal resistive index: a case of mistaken identity. Hypertension. 2014;64:915–7. [DOI] [PubMed] [Google Scholar]
- [16].Andrew BY, Andrew EY, Cherry AD, Hauck JN, Nicoara A, Pieper CF, et al. Intraoperative Renal Resistive Index as an Acute Kidney Injury Biomarker: Development and Validation of an Automated Analysis Algorithm. J Cardiothorac Vasc Anesth. 2018. [DOI] [PMC free article] [PubMed]
- [17].Lerolle N, Guerot E, Faisy C, Bornstain C, Diehl JL, Fagon JY. Renal failure in septic shock: predictive value of Doppler-based renal arterial resistive index. Intensive care medicine. 2006;32:1553–9. [DOI] [PubMed] [Google Scholar]
- [18].Darmon M, Schortgen F, Vargas F, Liazydi A, Schlemmer B, Brun-Buisson C, et al. Diagnostic accuracy of Doppler renal resistive index for reversibility of acute kidney injury in critically ill patients. Intensive care medicine. 2011;37:68–76. [DOI] [PubMed] [Google Scholar]
- [19].Dewitte A, Coquin J, Meyssignac B, Joannes-Boyau O, Fleureau C, Roze H, et al. Doppler resistive index to reflect regulation of renal vascular tone during sepsis and acute kidney injury. Critical care. 2012;16:R165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Ninet S, Schnell D, Dewitte A, Zeni F, Meziani F, Darmon M. Doppler-based renal resistive index for prediction of renal dysfunction reversibility: A systematic review and meta-analysis. Journal of critical care. 2015;30:629–35. [DOI] [PubMed] [Google Scholar]
- [21].Bossard G, Bourgoin P, Corbeau JJ, Huntzinger J, Beydon L. Early detection of postoperative acute kidney injury by Doppler renal resistive index in cardiac surgery with cardiopulmonary bypass. Br J Anaesth. 2011;107:891–8. [DOI] [PubMed] [Google Scholar]
- [22].Guinot PG, Bernard E, Abou Arab O, Badoux L, Diouf M, Zogheib E, et al. Doppler-Based Renal Resistive Index Can Assess Progression of Acute Kidney Injury in Patients Undergoing Cardiac Surgery. J Cardiothorac Vasc Anesth. 2013. [DOI] [PubMed]
- [23].Marty P, Szatjnic S, Ferre F, Conil JM, Mayeur N, Fourcade O, et al. Doppler renal resistive index for early detection of acute kidney injury after major orthopaedic surgery: a prospective observational study. Eur J Anaesthesiol. 2015;32:37–43. [DOI] [PubMed] [Google Scholar]
- [24].Annual Data Report of the US Organ Procurement and Transplantation Network (OPTN) and the Scientific Registry of Transplant Recipients (SRTR). Preface. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons. 2013;13 Suppl 1:1–7. [DOI] [PubMed] [Google Scholar]
- [25].Schnell D, Deruddre S, Harrois A, Pottecher J, Cosson C, Adoui N, et al. Renal resistive index better predicts the occurrence of acute kidney injury than cystatin C. Shock. 2012;38:592–7. [DOI] [PubMed] [Google Scholar]
- [26].Wu HB, Qin H, Ma WG, Zhao HL, Zheng J, Li JR, et al. Can Renal Resistive Index Predict Acute Kidney Injury After Acute Type A Aortic Dissection Repair? The Annals of thoracic surgery. 2017. [DOI] [PubMed]
- [27].Marty P, Ferre F, Labaste F, Jacques L, Luzi A, Conil JM, et al. The Doppler renal resistive index for early detection of acute kidney injury after hip fracture. Anaesth Crit Care Pain Med. 2016. [DOI] [PubMed]
- [28].Bandyopadhyay S, Kumar Das R, Paul A, Sundar Bhunia K, Roy D. A transesophageal echocardiography technique to locate the kidney and monitor renal perfusion. Anesth Analg. 2013;116:549–54. [DOI] [PubMed] [Google Scholar]
- [29].Kararmaz A, Kemal Arslantas M, Cinel I. Renal Resistive Index Measurement by Transesophageal Echocardiography: Comparison With Translumbar Ultrasonography and Relation to Acute Kidney Injury. J Cardiothorac Vasc Anesth. 2015;29:875–80. [DOI] [PubMed] [Google Scholar]
- [30].Regolisti G, Maggiore U, Cademartiri C, Belli L, Gherli T, Cabassi A, et al. Renal resistive index by transesophageal and transparietal echo-doppler imaging for the prediction of acute kidney injury in patients undergoing major heart surgery. Journal of nephrology. 2016. [DOI] [PubMed]
- [31].Haitsma Mulier JLG, Rozemeijer S, Rottgering JG, Spoelstra-de Man AME, Elbers PWG, Tuinman PR, et al. Renal resistive index as an early predictor and discriminator of acute kidney injury in critically ill patients; A prospective observational cohort study. PloS one. 2018;13:e0197967. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Zanocco JA, Nishida SK, Passos MT, Pereira AR, Silva MS, Pereira AB, et al. Race Adjustment for Estimating Glomerular Filtration Rate Is Not Always Necessary. Nephron Extra. 2012;2:293–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [33].National Guideline C. KDIGO clinical practice guideline for acute kidney injury. 2012.
- [34].McIlroy DR, Argenziano M, Farkas D, Umann T, Sladen RN. Incorporating oliguria into the diagnostic criteria for acute kidney injury after on-pump cardiac surgery: impact on incidence and outcomes. J Cardiothorac Vasc Anesth. 2013;27:1145–52. [DOI] [PubMed] [Google Scholar]
- [35].Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P, the Aw. 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. Critical Care. 2004;8:R204–R12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [36].Stafford-Smith M, Grocott HP. Renal medullary hypoxia during experimental cardiopulmonary bypass: a pilot study. Perfusion. 2005;20:53–8. [DOI] [PubMed] [Google Scholar]
- [37].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 medicine. 2017;43:1551–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [38].Andrew BY, Cherry AD, Hauck JN, Nicoara A, Maxwell CD, Konoske RM, et al. The Association of Aortic Valve Pathology With Renal Resistive Index as a Kidney Injury Biomarker. The Annals of thoracic surgery. 2018;106:107–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [39].Stefan G, Capusa C, Stancu S, Petrescu L, Nedelcu ED, Andreiana I, et al. Abdominal aortic calcification and renal resistive index in patients with chronic kidney disease: is there a connection? Journal of nephrology. 2014;27:173–9. [DOI] [PubMed] [Google Scholar]
- [40].Tublin ME, Tessler FN, Murphy ME. Correlation between renal vascular resistance, pulse pressure, and the resistive index in isolated perfused rabbit kidneys. Radiology. 1999;213:258–64. [DOI] [PubMed] [Google Scholar]
- [41].Calabia J, Torguet P, Garcia I, Martin N, Mate G, Marin A, et al. The relationship between renal resistive index, arterial stiffness, and atherosclerotic burden: the link between macrocirculation and microcirculation. Journal of clinical hypertension. 2014;16:186–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [42].Geraci G, Mule G, Geraci C, Mogavero M, D’Ignoto F, Morreale M, et al. Association of renal resistive index with aortic pulse wave velocity in hypertensive patients. European journal of preventive cardiology. 2014. [DOI] [PubMed]
- [43].Kuznetsova T, Cauwenberghs N, Knez J, Thijs L, Liu YP, Gu YM, et al. Doppler Indexes of Left Ventricular Systolic and Diastolic Flow and Central Pulse Pressure in Relation to Renal Resistive Index. American journal of hypertension. 2014. [DOI] [PubMed]
- [44].Tedesco MA, Natale F, Mocerino R, Tassinario G, Calabro R. Renal resistive index and cardiovascular organ damage in a large population of hypertensive patients. Journal of human hypertension. 2007;21:291–6. [DOI] [PubMed] [Google Scholar]
- [45].Eljaiek R, Cavayas YA, Rodrigue E, Desjardins G, Lamarche Y, Toupin F, et al. High postoperative portal venous flow pulsatility indicates right ventricular dysfunction and predicts complications in cardiac surgery patients. Br J Anaesth. 2019;122:206–14. [DOI] [PubMed] [Google Scholar]
- [46].Hitsumoto T. Relationship Between Hemorheology Assessed Using Microchannel Array Flow Analyzer and Kidney Function in Hypertensive Patients. Cardiol Res. 2017;8:147–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [47].Guinot PG, Bernard E, Abou Arab O, Badoux L, Diouf M, Zogheib E, et al. Doppler-based renal resistive index can assess progression of acute kidney injury in patients undergoing cardiac surgery. J Cardiothorac Vasc Anesth. 2013;27:890–6. [DOI] [PubMed] [Google Scholar]
- [48].Hertzberg D, Ceder SL, Sartipy U, Lund K, Holzmann MJ. Preoperative Renal Resistive Index Predicts Risk of Acute Kidney Injury in Patients Undergoing Cardiac Surgery. J Cardiothorac Vasc Anesth. 2017;31:847–52. [DOI] [PubMed] [Google Scholar]
- [49].Sinning JM, Adenauer V, Scheer AC, Lema Cachiguango SJ, Ghanem A, Hammerstingl C, et al. Doppler-based renal resistance index for the detection of acute kidney injury and the non-invasive evaluation of paravalvular aortic regurgitation after transcatheter aortic valve implantation. EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology. 2014;9:1309–16. [DOI] [PubMed] [Google Scholar]
- [50].Marty P, Ferre F, Labaste F, Jacques L, Luzi A, Conil JM, et al. The Doppler renal resistive index for early detection of acute kidney injury after hip fracture. Anaesth Crit Care Pain Med. 2016;35:377–82. [DOI] [PubMed] [Google Scholar]
- [51].Song J, Wu W, He Y, Lin S, Zhu D, Zhong M. Value of the combination of renal resistance index and central venous pressure in the early prediction of sepsis-induced acute kidney injury. Journal of critical care. 2018;45:204–8. [DOI] [PubMed] [Google Scholar]