- Email: [email protected]
Mannitol and Renal Dysfunction After Endovascular Aortic Aneurysm Repair Procedures: A Randomized Trial
Mannitol and Renal Dysfunction after Endovascular Aortic Aneurysm Repair Procedures: A Randomized Trial Konstantinos Kalimeris, MD, PhD,* Nikolaos Nikolakopoulos, MD,* Maria Riga, MD,* Kalliopi Christodoulaki, MD,* Konstantinos G. Moulakakis, MD, PhD,† Cleanthi Dima, MD, PhD,‡ Christos Papasideris, MD,† Tatiana Sidiropoulou, MD, PhD,* Georgia Kostopanagiotou, MD, PhD,* and Angeliki Pandazi, MD, PhD* Objective: Endovascular aortic aneurysm repair (EVAR) may result in deterioration of renal function. Mannitol has renovascular and antioxidant properties that could prove beneficial in this respect. Design: A randomized prospective study. Setting: Attikon University Hospital, single institution. Participants: Eighty-six patients undergoing elective EVAR under regional anesthesia. Methods: Patients received hydration alone (controls) or hydration plus mannitol (0.5 g/kg). Measurements and Main Results: Creatinine, serum cystatin-C, urine neutrophil-gelatinase-associated lipocalin (NGAL), albuminuria and serum urea were measured 24 hours and 72 hours after the procedure (baseline NGAL was measured in 19 randomly selected patients). Serum creatinine also was measured at the followup of the patients. Serum creatinine and cystatin-C were lower in the mannitol group at 24 hours postoperatively (creatinine, mannitol [n ¼ 43]; 1.07 ⫾ 0.26 [CI95%: 0.99-1.15] v controls [n ¼ 43]; 1.20 ⫾ 0.30 [CI95%: 1.11-1.30]), but not at 72 hours (creatinine, mannitol [n ¼ 43]; 1.13 ⫾ 0.29 [CI95%: 1.04-1.22] v controls [n ¼ 43]; 1.26 ⫾ 0.41 [CI95% 1.15 – 1.38]). Urine
D
ESPITE THE AVOIDANCE of aortic cross-clamping, hemodynamic instability, and ischemia-reperfusion in endovascular aortic aneurysm repair (EVAR) compared to open aortic aneurysm repair, trials suggested that renal function may sustain greater deterioration after EVAR than after open repair.1,2 Implicated factors include contrast-induced nephropathy (CIN), suprarenal endograft fixation, microembolization of renal arteries, and ischemia-reperfusion of the lower limbs.1,3 The pathogenesis of CIN involves renal vasospasm, reduced blood flow to the outer renal medulla, activation of the tubuloglomerular feedback mechanism and oxidative stress.2,4–7 Apart from general precautions, such as avoidance of hypoperfusion,8 adequate hydration and avoidance of nephrotoxic drugs,9 and use of low osmolarity contrast media,10 current measures to prevent renal dysfunction following EVAR in clinical practice are lacking.1 Mannitol is an osmotic diuretic with antioxidant and renovascular effects that could preserve renal function after EVAR. However, the protective potential of mannitol in CIN is not clear and a randomized controlled trial of mannitol in EVAR has not been performed to date. Instead, mannitol has been used as part of a forced diuresis protocol, so that the effects of the individual components cannot be evaluated.11,12 Moreover, dehydration due to inadequate replacement of urinary losses also could obscure any benefit. Furthermore, its protective potential has not been evaluated with the newer sensitive indices of renal injury such as cystatin-C and neutrophil gelatinase-associated lipocalin (NGAL). To this end, the authors conducted this study to compare hydration versus hydration plus mannitol for prevention of renal dysfunction after EVAR. The authors evaluated creatinine, cystatin-C, and NGAL changes as well as more traditional markers such as urea and albuminuria.
NGAL increased substantially at 24 hours without differences between groups. At followup (controls: 13 ⫾ 7 months; mannitol: 12 ⫾ 7 months), there were no differences between creatinine or creatinine clearance (creatinine: controls [n ¼ 28]; 1.15 ⫾ 0.39 [CI95% 1.02-1.29] v mannitol [n ¼ 23]; 1.05 ⫾ 0.27 [CI95%: 0.95-1.17]). The overall changes of creatinine and creatinine clearance with time were significant in controls but not in the mannitol group. The classification according to the RIFLE criteria yielded 4 patients at risk for renal injury and 2 with renal injury in the control group and 6 patients at risk with no patients with injury in the mannitol group, but the difference of renal dysfunction between the 2 groups was not statistically significant. Conclusions: Mannitol plus hydration during EVAR provides a small but significant benefit for renal function. Future preventive protocols aiming at greater restoration of renal function after EVAR could include mannitol as a useful component. & 2013 Published by Elsevier Inc. KEY WORDS: endovascular aneurysm repair (EVAR), renal function, mannitol, acute kidney injury MATERIALS AND METHODS The research protocol was approved by the local ethics committee (No. protocol 429/17-12-09), and after written informed consent, 100 patients undergoing elective EVAR surgery were enrolled in the study (155 were assessed for enrollment in the study). Exclusion criteria (a priori exclusions) included patient refusal, end-stage renal failure (hemodialysis therapy) or 1 kidney, severe heart failure (left ventricular ejection fraction o25%), emergency surgery, known allergy to local anesthetics or mannitol, impaired coagulation or other contraindication to regional anesthesia, preceding angiography, embolism and embolectomy, and known renal artery stenosis or occlusion. Randomization was performed with the method of the closed envelope without block design (opaque envelopes prepared by A.P.), and the study was not blinded. Patients received hydroxyzine, 50 mg, and ranitidine, 150 mg, orally the night before surgery. Diuretics and antihypertensive agents were omitted on the morning of surgery. In the operating room, after application of standard anesthetic monitoring and insertion of 2 largebore intravenous catheters, ranitidine, 50 mg, diphenydramine, 4 mg, and metoclopramide, 10 mg, were administered IV, and an arterial catheter was placed for direct measurement of arterial pressure. Patients received hydration with 500 mL of Ringer’s lactated solution prior to
From the *2nd Department of Anesthesiology, †Department of Vascular Surgery; and ‡2nd Department of Clinical Biochemistry, Medical School, University of Athens, Attikon Hospital, Athens, Greece. Address reprint requests to Angeliki Pandazi, MD, PhD, Attikon University Hospital, 2nd Department of Anesthesiology, 1 Rimini Street 12462, Athens, Greece. E-mail: [email protected] © 2013 Published by Elsevier Inc. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2013.08.009
Journal of Cardiothoracic and Vascular Anesthesia, Vol ], No ] (Month), 2013: pp ]]]–]]]
1
2
any other intervention. Afterwards, piperacillin-tazobactam, 4.5 g, and vancomycin, 1 g (or less, according to patient’s creatinine clearance) (hourly infusion in 250 mL normal saline) infusions were instituted. Following standard procedures, the epidural space was approached in the sitting position in the lumbar region (L2-L4) with a Tuohy 18G needle, and an epidural catheter was placed either without or following a spinal injection with a 27G needle through the epidural catheter, according to the anesthesiologist’s preference. The spinal injection included ropivacaine (Naropeine, Astra Zeneca, London, UK), 7.5 mg/ mL, 1.8-2.2 mL and fentanyl, 20-30 μg, according to the patient’s height. In the case of the epidural without spinal anesthesia, after a test dose of 3 mL of lidocaine, 20 mg/mL through the catheter, fentanyl 50 μg, and 10-15 mL of ropivacaine, 7.5 mg/mL, were administered, aiming for a sensory block up to T8-T10. If additional epidural doses were needed during the surgery, 3-4 mL of 7.5 mg/mL of ropivacaine were administered. Upon surgical incision, 0.5 g/kg of mannitol 20% (Mannitol, VIOSER S.A., Trikala, Greece) was administered in the mannitol group within 15 minutes. Further fluid management (after the initial 500 mL of lactated Ringer’s solution) included the additional infusion of at least 2 mL/kg/h of lactated Ringer’s solution and then further replacement of blood and urine losses. Colloids generally were administered when blood loss ensued (6% hydroxyethyl starch 130/0.4 solution [Voluvens, Fresenius-Kabi, Bad Homburg, Germany]) with a general rule of crystalloids:colloids 3:1, as generally is applied in the authors’ institution. Red blood cells were administered if necessary to preserve blood hematocrit 428%. Mean arterial pressure was maintained within 80-120% of baseline values with infusion of phenylephrine or nitroglycerine if necessary. If urine output dropped below 0.5 mL/kg/h for 2 hours, furosemide, 5-20 mg, was given IV. Normal saline, sodium bicarbonate, or nonsteroidal anti-inflammatory drugs were avoided. Postoperatively, patients were monitored in the postanesthesia care unit for at least 6 hours with 1-2 mL/kg/h hydration and the same goals as above. Intraoperative recordings included heart rate, mean arterial pressure, arterial pH and hematocrit, hourly diuresis, administered fluids, total volume of administered contrast media, type of contrast media, hypotension (systolic arterial pressure o90 mmHg) or hypertension (systolic arterial pressure 4150 mmHg) episodes with duration 410 minutes, the use of vasoconstrictors (phenylephrine or ephedrine) or vasodilators (nitroglycerine), type of endovascular stent (infrarenal or suprarenal), and duration of surgery (incision to last suture). In addition, major postoperative complications (hemorrhage, acute renal failure, reoperation) were recorded. The contrast media used was either iodixanol, 270 mg/mL (Visipaques, GE Healthcare Biosciences, Princeton, NJ) or iobitridol, 300 mg/mL (Xenetixs, Guerbet, Aulnay-sous-Bois, France). Peripheral blood and urine spot samples were collected just before the surgical incision (baseline) and at 24 and 72 hours after the end of the operation for measurement of serum creatinine, serum urea, urinary creatinine, and urinary urea (spot urine samples at 72 h were collected from spontaneous diuresis in sterile urine boxes). Blood samples were centrifuged at 2,500 rpm for 10 minutes, and the supernatants were stored at –801C for further analysis. Serum creatinine also was measured at the time of followup. Neutrophil gelatinase-associated lipocalin (NGAL) was measured in urine by ELISA (HyCult Biotechnology, Uden, Netherlands), according to manufacturer’s instructions. NGAL was measured in all patients at 24 and 72 hours and in 19 randomly selected patients at baseline (only for the NGAL measurement), and the values were corrected to urine creatinine. Cystatin-C was measured in serum with a multiplex bead-based assay at 24 and 72 hours, using Milliplex Human Kidney Toxicity Panel 2 (HKTX2-38K, Millipore Corporation, Billerica, MA), according to the manufacturer’s instructions. Serum samples were diluted 1:2,000 with assay buffer. The filter plate was prewashed with 200 μl of
KALIMERIS ET AL
assay buffer on a plate shaker for 10 minutes at room temperature. Assay buffer was removed by vacuum, and then 25 μl of standards, controls, assay buffer and diluted serum sample were added into the appropriate wells, and 25 μl of the beads were added to each well. The plate was incubated with agitation overnight (16-18 h) at 41C. Liquid was removed by vacuum. After 2 washes, 50 μl of detection antibodies were added and incubated for 1 hour at room temperature followed by the addition of 50 μl of streptavidin-phycoerythrin and incubation for 30 minutes. Finally, 100 μl of sheath fluid were added to each well and the beads were resuspended for five minutes on a plate shaker. The analysis was performed on the Luminex 200TM platform (Luminex Corporation, Austin, TX). The samples were batch analyzed for NGAL and cystatin-C at the end of the study period. The primary outcome in the authors’ study was serum creatinine values. Creatinine clearance was calculated according to CockroftGault formula (creatinine clearance = (140 – age in years) (body weight in kg) / (72 Screat in mg/dL) ( 0.85 for women) and fractional excretion of urea was calculated according to formula: Feurea % = (Uurea Screat) / (Surea Ucreat), where Uurea = urine urea concentration (mg/dL), Screat = serum creatinine concentration (mg/dL), Surea = serum urea concentration (mg/dL), Ucreat = urine creatinine concentration (mg/dL). Albuminuria was evaluated with the corrected albumin-to-creatinine ratio in urine. To evaluate the postoperative renal dysfunction in the first 72 hours after EVAR, the Risk-Injury-FailureLoss-Endstage (RIFLE) criteria of creatinine and estimated GFR alterations were used.13 The urine output criteria were not used since, in the authors’ institution, oliguria and anuria are triggers for therapeutic interventions sooner than 6 hours and also because urine catheters are removed after 24 hours or less after EVAR. Quantitative data were checked for normality of distribution with the Kolmogorov-Smirnov test. The minimum necessary number (total n ¼ 84) of patients to assure a 0.80 power for creatinine values at 72 hours was calculated from an internal pilot study based on the authors’ first 10 patients (Cohen’s d ¼ 0.31), who underwent exactly the same protocol as the other patients and subsequently were included in the final analysis without calculating any penalty for the power of the study. Normal distributed variables were analyzed with 1-way analysis of variance for between-groups comparisons and with repeated-measures one-way analysis of variance for within-groups comparisons. Non-normal distributed variables were analyzed with the Kruskal-Wallis and MannWhitney U tests for between-groups comparisons and with the Friedman’s test for within-groups differences. Qualitative data were analyzed with cross-tabulation. Logistic regression for prognostic factors for renal risk/injury was carried out with the enter method and multicolinearity was checked. Correlations were checked with the Spearman’s coefficient of correlation. The software used were SPSS v.15 and the G*Power v.3.1.5. RESULTS
The protocol was abandoned in 11 patients due to conversion to general anaesthesia (n ¼ 4), conversion of EVAR to open surgery (n ¼ 2), persisting macroscopic hematuria due to urethral injury (n ¼ 1), postoperative control of bleeding in the angiography suite (n ¼ 2), and departure from protocol (n ¼ 2) (without any analysis of the data). Additionally, 3 patients were discharged earlier than 72 h and were excluded due to missing data (exclusion of all data). In the final analysis, 86 patients were included. Data from followup measurements could be collected for 51 patients. Age, sex, body mass index, ASA class, and time of followup of patients did not differ significantly between the 2 groups. Regarding coexisting diseases, arterial hypertension and diabetes mellitus were more frequent in the control group.
3
MANNITOL AND RENAL DYSFUNCTION
Table 1. Demographic Data
n Sex (M/F) Age (years) Body mass index (kg/m2) ASA class (I/II/III) Current smoker Arterial hypertension Coronary artery disease Diabetes mellitus Obesity (BMI 430 kg/m2) Hyperlipidemia Diuretics Beta-blockers Statins Calcium channel blockers ACEI or ARA Time of followup (months)
Control
Mannitol
p
43 43/0 72 ⫾ 7 28.2 ⫾ 4.0 0 / 13 / 30 34.9% 95.3% 48.8% 25.6% 34.9% 67.4% 44.2% 39.5% 65.1% 39.5% 74.4% 13.3 ⫾ 7.0
43 41/2 72 ⫾ 8 27.7 ⫾ 3.7 1 / 22 / 20 34.9% 74.4% 34.9% 7.0% 25.6% 55.8% 30.2% 53.5% 58.1% 37.2% 37.2% 11.8 ⫾ 6.5
0.152 0.836 0.586 0.070 1.000 0.007 0.190 0.019 0.348 0.268 0.181 0.195 0.506 0.825 0.001 0.454
Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARA, angiotensin-receptor antagonist; ASA, American Society of Anesthesiologists; BMI, body mass index.
Regarding medications, the 2 groups differed in the use of angiotensin-converting enzyme inhibitors and angiotensinreceptor antagonists, which were more common in the control group (Table 1). Intraoperatively, the type of anesthesia, type of stent, volume of contrast media, type of contrast media, administered fluids, transfusion of blood units, and duration of surgery did not differ between groups (Table 2). Although urine output was higher in the mannitol group, the difference did not reach statistical significance. However, no patient required administration of furosemide in the mannitol group in contrast to the control group (p ¼ 0.041). The percentage of patients who received colloids did not differ between the 2 groups (81.4% in controls v 88.4% in mannitol group, p ¼ 0.366). Hypotensive and hypertensive episodes, the use of vasoconstrictors and vasodilators, and major postoperative complications (hemorrhage) also did not differ between groups. No patient died or required hemodialysis or other major intervention postoperatively. Regarding the primary outcome of the study, serum creatinine increased in the control group postoperatively, but in the mannitol group there was no statistically significant change (Fig 1A). At 24 h after the operation, serum creatinine levels were significantly lower in the mannitol compared to the control group (1.07 ⫾ 0.26 v 1.20 ⫾ 0.30, respectively). Creatinine clearance decreased in the control group both at 24 h and 72 h; whereas in the mannitol group, there was no statistically significant change from baseline (Fig 1B). However, there were no direct differences between the 2 groups at any time point. According to the RIFLE classification of renal dysfunction, 4 patients (9.3%) were diagnosed at renal risk and 2 (4.7%) with renal injury in the control group; whereas in the mannitol group, 6 patients (14%) were at risk but none had renal injury. The difference between the 2 groups was not statistically significant (p ¼ 0.301). Urinary levels of NGAL followed a similar course in the 2 groups, peaking at 24 h and subsequently decreasing at 72 h
without differences between the 2 groups (Fig 2A). Serum cystatin-C levels were significantly lower in the mannitol group at 24 h (controls: 2.6 ⫾ 0.9 v mannitol: 2.2 ⫾ 0.8) (Fig 2B). Serum cystatin-C levels at 24 and 72 hours correlated closely to postoperative serum creatinine levels (r ¼ 0.558, p o 0.001), and creatinine clearance (r ¼ 0.468, p o 0.001). No correlation was found between serum creatinine and corrected urine NGAL concentrations. The albumin-to-creatinine ratio of urine increased significantly in both groups postoperatively without differences between the 2 groups (data not shown). Serum urea levels also decreased significantly without differences between the groups (data not shown). Fractional excretion of urea did not change significantly in either group, and there were no differences between the groups at any time point (data not shown). Multivariate analysis with logistic regression (Table 3) revealed that diabetes mellitus, arterial hypertension, the use of ACEI or ARA, type of anesthesia, volume of crystalloids or colloids administered, urine output, and intraoperative use of diuretics were not significant predictors of renal risk or injury after EVAR. The logistic model was able to explain 67% of the variability in renal risk/injury. DISCUSSION
In this first study to evaluate the renal effects of mannitol after EVAR, mannitol improved postoperative creatinine and cystatin-C values, indicating a small but significant clinical benefit in preventing postoperative creatinine clearance decrease. Not surprisingly, the traditional and less sensitive renal markers of serum urea, albuminuria, and fractional excretion of urea failed to identify a benefit by mannitol. Mannitol has useful properties that could account for the improved renal function after EVAR. First, mannitol stimulates prostaglandin-I2 synthesis and increases renal blood flow, especially in ischemic kidneys but, more importantly, also reverses established renal vasoconstriction, a fundamental Table 2. Intraoperative Data Control
Mannitol
p
Anesthesia (ED/CSEA) 17/26 20/23 0.514 Endovascular stent fixation 29/14 29/14 1.000 (infrarenal/suprarenal) Surgery duration (min) 133 ⫾ 36 128 ⫾ 40 0.583 Crystalloids (mL) 1651 ⫾ 422 1808 ⫾ 463 0.104 Colloids (mL) 500 (0-1000) 500 (0-1000) 0.476 Contrast volume (mL) 150 (70-600) 150 (70-350) 0.598 Iodixanol/Iobitridol 27/16 34/9 0.096 Urine output (mL) 677 ⫾ 363 845 ⫾ 454 0.062 Blood transfusion (%) 9.3 7.0 0.693 Patients requiring diuretics (%) 9.3 0 0.041 Hypotensive episodes (%) 4.7 4.7 1.000 Hypertensive episodes (%) 0 0 – Patients requiring 14.0 9.3 0.501 vasodilators (%) Patients requiring 32.6 34.9 0.820 vasoconstrictors (%) Postoperative major 7.0 7.0 1.000 complications (%) Abbreviations: CSEA, combined spinal-epidural anaesthesia; ED, epidural anesthesia.
4
KALIMERIS ET AL
Fig 1. Mean values of serum creatinine (A) and creatinine clearance (B) during the observation period. *p o 0.05 between the 2 groups at the depicted time point; †p o 0.05 over time in the control group; error bars, standard deviation.
mechanism of contrast-induced nephropathy.14–16 In addition, mannitol decreased CIN in nondiabetic patients with renal insufficiency undergoing cardiac catheterization.17 Second, mannitol can scavenge free radicals released by contrast media and by ischemia-reperfusion of the lower limbs, which also ensues during EVAR.1,18 The role of oxidative damage by contrast-media is emphasized by the attenuation of CIN through reactive-oxygen species inhibition.6,7 In addition, mannitol as an osmotic diuretic maintains a high urine flow, inhibiting formation of tubular casts and swelling of endothelial cells.19 This also leads to the secondary benefit of avoiding furosemide, as was shown in the present study by the completely negated need for furosemide when mannitol was administered. A recent meta-analysis showed that furosemide at least doubled the risk of contrast-induced nephropathy, possibly by reversing blood flow within the kidney and rendering the vulnerable outer medulla ischemic.12 The authors’ results seemed to confirm this suggestion, since the omission of loop diuretics from the hydration protocol was a plausible cause of the revealed benefit of mannitol, in contrast to previous studies.11,12 Despite the early favorable outcomes with mannitol, these were not maintained after 24 hours, and there were a number of reasons that could account for this. First, renal dysfunction after EVAR also is induced by mechanisms that cannot be addressed
by mannitol. Such mechanisms are renal microembolizations (newly acquired renal infarcts are reported in 18% of patients after EVAR), bilateral renal artery stenosis, and the CT scans before discharge of the patient but also during followup.1,3 Additional factors that could have obscured a sustained benefit of mannitol at followup also included aortitis and ureteric obstruction as well as poor compliance with medications.1,20 Moreover, the effect of mannitol on diuresis gradually could have contracted the extracellular space, factitiously obscuring any greater benefit in the serum markers. Although the authors used a low-dose regimen due to the reports of renal injury after high doses of mannitol, the possibility of an even better outcome after higher doses (eg, 1-1.5 g/kg, but o200 g/day) or extended use in the postoperative period also could be examined.15,21–24 Although postoperative renal function showed some improvement with mannitol, there was no difference in the incidence of renal dysfunction according to the RIFLE criteria. The RIFLE classification is considered the most accurate in detecting and classifying renal dysfunction and also correlates closely to mortality.25 However, renal injury and frank renal failure are rather uncommon following EVAR (as shown by the 2.3% incidence of renal injury in controls and the absence of renal failure), and, therefore, a very large sample size would have been needed to detect a difference with the RIFLE
Fig 2. Urine NGAL (A) and serum cystatin-c (B) concentrations during the observation period. *p o 0.05 between the 2 groups at the depicted time point; †p o 0.01 for changes over time in each group; error bars, standard deviation.
5
MANNITOL AND RENAL DYSFUNCTION
Table 3. Logistic Regression Analysis for Renal Risk or Injury after EVAR Odds ratio (95% CI)
Age 470 years Male sex Hypertension Diabetes mellitus Statins ACEI or ARA Combined Spinal/Epidural Anesthesia Suprarenal stent fixation Surgery duration Crystalloids Colloids Contrast volume Iobitridol/ Iodixanol Intraoperative use of diuretics Urine output
2.33 0.16 4.09 0.13 0.54 0.59 3.12 0.86 0.99 0.99 1.00 1.00 1.52 4.43 1.00
(0.42-13.1) (0.01-4.43) (0.38-44.2) (0.01-1.71) (0.12-2.49) (0.14-2.55) (0.70-13.9) (0.17-4.39) (0.96-1.02) (0.99-1.00) (0.99-1.00) (0.99-1.01) (0.29-8.13) (0.18-112.6) (0.99-1.00)
p
0.335 0.278 0.245 0.122 0.432 0.482 0.136 0.860 0.487 0.441 0.512 0.965 0.622 0.367 0.858
Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARA, angiotensin receptor antagonist; CI, confidence interval; EVAR, endovascular aortic aneurysm repair.
criteria.26 Although the looser criteria of contrast-induced nephropathy occasionally have been used, they have not been validated in clinical circumstances in which other causes of renal injury coexisted, such as in EVAR.4 Nevertheless, a study with a larger sample size examining if there is a benefit in renal risk/injury after EVAR with mannitol would be the most helpful because the clinical significance of the rather small and temporary reduction of creatinine and cystatin-C values is unclear. Cystatin-C is a cysteine protease inhibitor whose levels reflect glomerular filtration more accurately than creatinine since it not affected by age, muscle mass, or infection. In addition, it has proved more sensitive in detecting CIN and that its sensitivity peaks at 12-24 hours.27,28 Similarly, urinary NGAL is a protein of the tubular epithelium that has evolved into one of the earliest and most sensitive markers of renal injury, both in the setting of surgery and in CIN.28,29 However, the accuracy of NGAL peaks at 2-4 hours after surgery and
then subsequently wanes, which could account for the lack of difference between the groups.28 A limitation of this study was that controls had higher frequency of diabetes mellitus, hypertension, and angiotensinconverting enzyme inhibitors/angiotensin-receptor antagonists intake, which could have influenced the results. Although diabetes mellitus and hypertension are recognized factors in the development of chronic renal failure, they seem to be unrelated to renal function alterations after surgery or after contrast media.30–32 In a recent study of 102 patients who underwent EVAR, hypertension and diabetes mellitus also were insignificant in predicting postoperative creatinine rise.33 In addition, the multivariate analysis revealed that diabetes mellitus and hypertension were not significant prognostic factors for renal risk/injury in the authors’ patients. The same applied for the administered amount of crystalloids and colloids and also for urine output, which were, therefore, unlikely reasons for the better outcome in the mannitol group. Finally, angiotensin-converting enzyme inhibitors have been proposed both as beneficial and deleterious for CIN, but in a recent study, their preoperative withholding did not have any influence on CIN.34,35 In addition, they do not impair the renal vasodilatory effects of mannitol.16 Therefore, although the difference between the groups should be taken into account in the interpretation of the authors’ results, there is no proof that it had significant effect in renal outcomes. In conclusion, the authors found that mannitol plus hydration offered a rather small but significant protection against renal dysfunction 24 hours after EVAR compared to hydration alone, indicating a small but statistically significant benefit in renal function markers beyond that of hydration. Considering the paucity of readily applicable measures to attenuate renal dysfunction and also the simplicity and inexpensiveness of mannitol, preventive protocols including mannitol should be examined for their efficacy. A larger study looking at the incidences of renal risk and injury is required, as well as the evaluation of larger doses of mannitol or extension of its administration postoperatively.
REFERENCES 1. Walsh SR, Tang TY, Boyle JR: Renal consequences of endovascular abdominal aortic aneurysm repair. J Endovasc Ther 15: 73-82, 2008 2. Alsac JM, Zarins CK, Heikkinen MA, et al: The impact of aortic endografts on renal function. J Vasc Surg 41:926-930, 2005 3. Harris JR, Fan CM, Geller SC, et al: Renal perfusion defects after endovascular repair of abdominal aortic aneurysms. J Vasc Interv Radiol 4:329-333, 2003 4. Wong GT, Irwin MG: Contrast-induced nephropathy. Br J Anaesth 99:474-483, 2007 5. Allie DE, Lirtzman MD, Wyatt CH, et al: Targeted renal therapy and contrast-induced nephropathy during endovascular abdominal aortic aneurysm repair: Results of a feasibility pilot trial. J Endovasc Ther 14:520-527, 2007 6. Persson PB, Hansell P, Liss P: Pathophysiology of contrast medium-induced nephropathy. Kidney Int 68:14-22, 2005 7. Katholi RE, Woods WT Jr, Taylor GJ, et al: Oxygen free radicals and contrast nephropathy. Am J Kidney Dis 32:64-71, 1998
8. Wahlberg E, DiMuzio PJ, Stoney RJ: Aortic clamping during elective operations for infrarenal disease: The influence of clamping time on renal function. J Vasc Surg 36:13-18, 2002 9. Erley C: Concomitant drugs with exposure to contrast media. Kidney Int 69:S20-S24, 2006 10. Haveman JW, Gansevoort RT, Bongaerts AH, et al: Low incidence of nephropathy in surgical ICU patients receiving intravenous contrast: A retrospective analysis. Intensive Care Med 32:1199-1205, 2006 11. Solomon R, Werner C, Mann D, et al: Effects of saline, mannitol, and furosemide to prevent acute decreases in renal function induced by radiocontrast agents. N Engl J Med 331:1416-1420, 1994 12. Majumdar SR, Kjellstrand CM, Tymchak WJ, et al: Forced euvolemic diuresis with mannitol and furosemide for prevention of contrast-induced nephropathy in patients with CKD undergoing coronary angiography: A randomized controlled trial. Am J Kidney Dis 54: 602-609, 2009 13. Cruz DN, Ricci Z, Ronco C: Clinical review: RIFLE and AKIN– time for reappraisal. Crit Care 13:211, 2009
6
14. Behnia R, Koushanpour E, Brunner EA: Effects of hyperosmotic mannitol infusion on hemodynamics of dog kidney. Anesth Analg 82: 902-908, 1996 15. Better OS, Rubinstein I, Winaver JM, et al: Mannitol therapy revisited (1940-1997). Kidney Int 52:886-894, 1997 16. Johnston PA, Bernard DB, Perrin NS, et al: Prostaglandins mediate the vasodilatory effect of mannitol in the hypoperfused rat kidney. J Clin Invest 68:127-133, 1981 17. Weisberg LS, Kurnik PB, Kurnik BR: Risk of radiocontrast nephropathy in patients with and without diabetes mellitus. Kidney Int 45:259-265, 1994 18. Moore NN, Lapsley M, Norden AG, et al: Does N-acetylcysteine prevent contrast-induced nephropathy during endovascular AAA repair? A randomized controlled pilot study. J Endovasc Ther 13: 660-666, 2006 19. Lameire N, Vanholder R: Pathophysiologic features and prevention of human and experimental acute tubular necrosis. J Am Soc Nephrol 17:S20-S32, 2001 20. Vallabhaneni SR, McWilliams RG, Anbarasu A, et al: Perianeurysmal fibrosis: A relative contraindication to endovascular repair. Eur J Vasc Endovasc Surg 21:567-568, 2001 21. Dorman HR, Sondheimer JH, Chadnapaphomhai P: Mannitol induced acute renal failure. Medicine 69:153-159, 1990 22. Nicholson ML, Baker DM, Hopkinson BR, et al: Randomised controlled trial of the effect of Mannitol on renal reperfusion injury during aortic aneurysm surgery. Br J Surg 83:1230-1233, 1996 23. Wijnen MHWA, Vader HL, Van Den Wall Bake AWL, et al: Can renal dysfunction after infra-renal aortic aneurysm repair be modified by multi-antioxidant supplementation? J Cardiovasc Surg 43: 483-488, 2002 24. van Valenberg PL, Hoitsma AJ, Tiggeler RG, et al: Mannitol as an indispensable constituent of an intraoperative hydration protocol for the prevention of acute renal failure after renal cadaveric transplantation. Transplantation 44:784-788, 1987
KALIMERIS ET AL
25. Uchino S, Bellomo R, Goldsmith D, et al: An assessment of the RIFLE criteria for acute renal failure in hospitalized patients. Crit Care Med 34:1913-1917, 2006 26. Wald R, Waikar SS, Liangos O, et al: Acute renal failure after endovascular vs open repair of abdominal aortic aneurysm. J Vasc Surg 43:460-466, 2006 27. Kimmel M, Butscheid M, Brenner S, et al: Improved estimation of glomerular filtration rate by serum cystatin C in preventing contrast induced nephropathy by N-acetylcysteine or zinc–preliminary results. Nephrol Dial Transplant 23:1241-1245, 2008 28. Moore E, Bellomo R, Nichol A: Biomarkers of acute kidney injury in anesthesia, intensive care and major surgery: from the bench to clinical research to clinical practice. Minerva Anestesiol 76: 425-440, 2010 29. Ling W, Zhaohui N, Ben H, et al: Urinary IL-18 and NGAL as early predictive biomarkers in contrast-induced nephropathy after coronary angiography. Nephron Clin Pract 108:c176-c181, 2008 30. Abelha FJ, Botelho M, Fernandes V, et al: Determinants of postoperative acute kidney injury. Crit Care 13:R79, 2009 31. Cochran ST, Wong WS, Roe DJ: Predicting angiographyinduced acute renal function impairment: Clinical risk model. AJR Am J Roentgenol 141:1027-1033, 1983 32. Meguid El Nahas A, Bello AK: Chronic kidney disease: The global challenge. Lancet 365:331-340, 2005 33. Brooks CE, Middleton A, Dhillon R, et al: Predictors of creatinine rise post-endovascular abdominal aortic aneurysm repair. ANZ J Surg 81:827-830, 2011 34. Li X, Li T, Cong H: Is angiotensin-converting enzyme inhibitor a contraindication for contrast-induced nephropathy prophylaxis? A review about its paradox. Cardiovasc Therg 30:273-276, 2011 35. Rosenstock JL, Bruno R, Kim JK, et al: The effect of withdrawal of ACE inhibitors or angiotensin receptor blockers prior to coronary angiography on the incidence of contrast-induced nephropathy. Int Urol Nephrol 40:749-755, 2008
D
ESPITE THE AVOIDANCE of aortic cross-clamping, hemodynamic instability, and ischemia-reperfusion in endovascular aortic aneurysm repair (EVAR) compared to open aortic aneurysm repair, trials suggested that renal function may sustain greater deterioration after EVAR than after open repair.1,2 Implicated factors include contrast-induced nephropathy (CIN), suprarenal endograft fixation, microembolization of renal arteries, and ischemia-reperfusion of the lower limbs.1,3 The pathogenesis of CIN involves renal vasospasm, reduced blood flow to the outer renal medulla, activation of the tubuloglomerular feedback mechanism and oxidative stress.2,4–7 Apart from general precautions, such as avoidance of hypoperfusion,8 adequate hydration and avoidance of nephrotoxic drugs,9 and use of low osmolarity contrast media,10 current measures to prevent renal dysfunction following EVAR in clinical practice are lacking.1 Mannitol is an osmotic diuretic with antioxidant and renovascular effects that could preserve renal function after EVAR. However, the protective potential of mannitol in CIN is not clear and a randomized controlled trial of mannitol in EVAR has not been performed to date. Instead, mannitol has been used as part of a forced diuresis protocol, so that the effects of the individual components cannot be evaluated.11,12 Moreover, dehydration due to inadequate replacement of urinary losses also could obscure any benefit. Furthermore, its protective potential has not been evaluated with the newer sensitive indices of renal injury such as cystatin-C and neutrophil gelatinase-associated lipocalin (NGAL). To this end, the authors conducted this study to compare hydration versus hydration plus mannitol for prevention of renal dysfunction after EVAR. The authors evaluated creatinine, cystatin-C, and NGAL changes as well as more traditional markers such as urea and albuminuria.
NGAL increased substantially at 24 hours without differences between groups. At followup (controls: 13 ⫾ 7 months; mannitol: 12 ⫾ 7 months), there were no differences between creatinine or creatinine clearance (creatinine: controls [n ¼ 28]; 1.15 ⫾ 0.39 [CI95% 1.02-1.29] v mannitol [n ¼ 23]; 1.05 ⫾ 0.27 [CI95%: 0.95-1.17]). The overall changes of creatinine and creatinine clearance with time were significant in controls but not in the mannitol group. The classification according to the RIFLE criteria yielded 4 patients at risk for renal injury and 2 with renal injury in the control group and 6 patients at risk with no patients with injury in the mannitol group, but the difference of renal dysfunction between the 2 groups was not statistically significant. Conclusions: Mannitol plus hydration during EVAR provides a small but significant benefit for renal function. Future preventive protocols aiming at greater restoration of renal function after EVAR could include mannitol as a useful component. & 2013 Published by Elsevier Inc. KEY WORDS: endovascular aneurysm repair (EVAR), renal function, mannitol, acute kidney injury MATERIALS AND METHODS The research protocol was approved by the local ethics committee (No. protocol 429/17-12-09), and after written informed consent, 100 patients undergoing elective EVAR surgery were enrolled in the study (155 were assessed for enrollment in the study). Exclusion criteria (a priori exclusions) included patient refusal, end-stage renal failure (hemodialysis therapy) or 1 kidney, severe heart failure (left ventricular ejection fraction o25%), emergency surgery, known allergy to local anesthetics or mannitol, impaired coagulation or other contraindication to regional anesthesia, preceding angiography, embolism and embolectomy, and known renal artery stenosis or occlusion. Randomization was performed with the method of the closed envelope without block design (opaque envelopes prepared by A.P.), and the study was not blinded. Patients received hydroxyzine, 50 mg, and ranitidine, 150 mg, orally the night before surgery. Diuretics and antihypertensive agents were omitted on the morning of surgery. In the operating room, after application of standard anesthetic monitoring and insertion of 2 largebore intravenous catheters, ranitidine, 50 mg, diphenydramine, 4 mg, and metoclopramide, 10 mg, were administered IV, and an arterial catheter was placed for direct measurement of arterial pressure. Patients received hydration with 500 mL of Ringer’s lactated solution prior to
From the *2nd Department of Anesthesiology, †Department of Vascular Surgery; and ‡2nd Department of Clinical Biochemistry, Medical School, University of Athens, Attikon Hospital, Athens, Greece. Address reprint requests to Angeliki Pandazi, MD, PhD, Attikon University Hospital, 2nd Department of Anesthesiology, 1 Rimini Street 12462, Athens, Greece. E-mail: [email protected] © 2013 Published by Elsevier Inc. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2013.08.009
Journal of Cardiothoracic and Vascular Anesthesia, Vol ], No ] (Month), 2013: pp ]]]–]]]
1
2
any other intervention. Afterwards, piperacillin-tazobactam, 4.5 g, and vancomycin, 1 g (or less, according to patient’s creatinine clearance) (hourly infusion in 250 mL normal saline) infusions were instituted. Following standard procedures, the epidural space was approached in the sitting position in the lumbar region (L2-L4) with a Tuohy 18G needle, and an epidural catheter was placed either without or following a spinal injection with a 27G needle through the epidural catheter, according to the anesthesiologist’s preference. The spinal injection included ropivacaine (Naropeine, Astra Zeneca, London, UK), 7.5 mg/ mL, 1.8-2.2 mL and fentanyl, 20-30 μg, according to the patient’s height. In the case of the epidural without spinal anesthesia, after a test dose of 3 mL of lidocaine, 20 mg/mL through the catheter, fentanyl 50 μg, and 10-15 mL of ropivacaine, 7.5 mg/mL, were administered, aiming for a sensory block up to T8-T10. If additional epidural doses were needed during the surgery, 3-4 mL of 7.5 mg/mL of ropivacaine were administered. Upon surgical incision, 0.5 g/kg of mannitol 20% (Mannitol, VIOSER S.A., Trikala, Greece) was administered in the mannitol group within 15 minutes. Further fluid management (after the initial 500 mL of lactated Ringer’s solution) included the additional infusion of at least 2 mL/kg/h of lactated Ringer’s solution and then further replacement of blood and urine losses. Colloids generally were administered when blood loss ensued (6% hydroxyethyl starch 130/0.4 solution [Voluvens, Fresenius-Kabi, Bad Homburg, Germany]) with a general rule of crystalloids:colloids 3:1, as generally is applied in the authors’ institution. Red blood cells were administered if necessary to preserve blood hematocrit 428%. Mean arterial pressure was maintained within 80-120% of baseline values with infusion of phenylephrine or nitroglycerine if necessary. If urine output dropped below 0.5 mL/kg/h for 2 hours, furosemide, 5-20 mg, was given IV. Normal saline, sodium bicarbonate, or nonsteroidal anti-inflammatory drugs were avoided. Postoperatively, patients were monitored in the postanesthesia care unit for at least 6 hours with 1-2 mL/kg/h hydration and the same goals as above. Intraoperative recordings included heart rate, mean arterial pressure, arterial pH and hematocrit, hourly diuresis, administered fluids, total volume of administered contrast media, type of contrast media, hypotension (systolic arterial pressure o90 mmHg) or hypertension (systolic arterial pressure 4150 mmHg) episodes with duration 410 minutes, the use of vasoconstrictors (phenylephrine or ephedrine) or vasodilators (nitroglycerine), type of endovascular stent (infrarenal or suprarenal), and duration of surgery (incision to last suture). In addition, major postoperative complications (hemorrhage, acute renal failure, reoperation) were recorded. The contrast media used was either iodixanol, 270 mg/mL (Visipaques, GE Healthcare Biosciences, Princeton, NJ) or iobitridol, 300 mg/mL (Xenetixs, Guerbet, Aulnay-sous-Bois, France). Peripheral blood and urine spot samples were collected just before the surgical incision (baseline) and at 24 and 72 hours after the end of the operation for measurement of serum creatinine, serum urea, urinary creatinine, and urinary urea (spot urine samples at 72 h were collected from spontaneous diuresis in sterile urine boxes). Blood samples were centrifuged at 2,500 rpm for 10 minutes, and the supernatants were stored at –801C for further analysis. Serum creatinine also was measured at the time of followup. Neutrophil gelatinase-associated lipocalin (NGAL) was measured in urine by ELISA (HyCult Biotechnology, Uden, Netherlands), according to manufacturer’s instructions. NGAL was measured in all patients at 24 and 72 hours and in 19 randomly selected patients at baseline (only for the NGAL measurement), and the values were corrected to urine creatinine. Cystatin-C was measured in serum with a multiplex bead-based assay at 24 and 72 hours, using Milliplex Human Kidney Toxicity Panel 2 (HKTX2-38K, Millipore Corporation, Billerica, MA), according to the manufacturer’s instructions. Serum samples were diluted 1:2,000 with assay buffer. The filter plate was prewashed with 200 μl of
KALIMERIS ET AL
assay buffer on a plate shaker for 10 minutes at room temperature. Assay buffer was removed by vacuum, and then 25 μl of standards, controls, assay buffer and diluted serum sample were added into the appropriate wells, and 25 μl of the beads were added to each well. The plate was incubated with agitation overnight (16-18 h) at 41C. Liquid was removed by vacuum. After 2 washes, 50 μl of detection antibodies were added and incubated for 1 hour at room temperature followed by the addition of 50 μl of streptavidin-phycoerythrin and incubation for 30 minutes. Finally, 100 μl of sheath fluid were added to each well and the beads were resuspended for five minutes on a plate shaker. The analysis was performed on the Luminex 200TM platform (Luminex Corporation, Austin, TX). The samples were batch analyzed for NGAL and cystatin-C at the end of the study period. The primary outcome in the authors’ study was serum creatinine values. Creatinine clearance was calculated according to CockroftGault formula (creatinine clearance = (140 – age in years) (body weight in kg) / (72 Screat in mg/dL) ( 0.85 for women) and fractional excretion of urea was calculated according to formula: Feurea % = (Uurea Screat) / (Surea Ucreat), where Uurea = urine urea concentration (mg/dL), Screat = serum creatinine concentration (mg/dL), Surea = serum urea concentration (mg/dL), Ucreat = urine creatinine concentration (mg/dL). Albuminuria was evaluated with the corrected albumin-to-creatinine ratio in urine. To evaluate the postoperative renal dysfunction in the first 72 hours after EVAR, the Risk-Injury-FailureLoss-Endstage (RIFLE) criteria of creatinine and estimated GFR alterations were used.13 The urine output criteria were not used since, in the authors’ institution, oliguria and anuria are triggers for therapeutic interventions sooner than 6 hours and also because urine catheters are removed after 24 hours or less after EVAR. Quantitative data were checked for normality of distribution with the Kolmogorov-Smirnov test. The minimum necessary number (total n ¼ 84) of patients to assure a 0.80 power for creatinine values at 72 hours was calculated from an internal pilot study based on the authors’ first 10 patients (Cohen’s d ¼ 0.31), who underwent exactly the same protocol as the other patients and subsequently were included in the final analysis without calculating any penalty for the power of the study. Normal distributed variables were analyzed with 1-way analysis of variance for between-groups comparisons and with repeated-measures one-way analysis of variance for within-groups comparisons. Non-normal distributed variables were analyzed with the Kruskal-Wallis and MannWhitney U tests for between-groups comparisons and with the Friedman’s test for within-groups differences. Qualitative data were analyzed with cross-tabulation. Logistic regression for prognostic factors for renal risk/injury was carried out with the enter method and multicolinearity was checked. Correlations were checked with the Spearman’s coefficient of correlation. The software used were SPSS v.15 and the G*Power v.3.1.5. RESULTS
The protocol was abandoned in 11 patients due to conversion to general anaesthesia (n ¼ 4), conversion of EVAR to open surgery (n ¼ 2), persisting macroscopic hematuria due to urethral injury (n ¼ 1), postoperative control of bleeding in the angiography suite (n ¼ 2), and departure from protocol (n ¼ 2) (without any analysis of the data). Additionally, 3 patients were discharged earlier than 72 h and were excluded due to missing data (exclusion of all data). In the final analysis, 86 patients were included. Data from followup measurements could be collected for 51 patients. Age, sex, body mass index, ASA class, and time of followup of patients did not differ significantly between the 2 groups. Regarding coexisting diseases, arterial hypertension and diabetes mellitus were more frequent in the control group.
3
MANNITOL AND RENAL DYSFUNCTION
Table 1. Demographic Data
n Sex (M/F) Age (years) Body mass index (kg/m2) ASA class (I/II/III) Current smoker Arterial hypertension Coronary artery disease Diabetes mellitus Obesity (BMI 430 kg/m2) Hyperlipidemia Diuretics Beta-blockers Statins Calcium channel blockers ACEI or ARA Time of followup (months)
Control
Mannitol
p
43 43/0 72 ⫾ 7 28.2 ⫾ 4.0 0 / 13 / 30 34.9% 95.3% 48.8% 25.6% 34.9% 67.4% 44.2% 39.5% 65.1% 39.5% 74.4% 13.3 ⫾ 7.0
43 41/2 72 ⫾ 8 27.7 ⫾ 3.7 1 / 22 / 20 34.9% 74.4% 34.9% 7.0% 25.6% 55.8% 30.2% 53.5% 58.1% 37.2% 37.2% 11.8 ⫾ 6.5
0.152 0.836 0.586 0.070 1.000 0.007 0.190 0.019 0.348 0.268 0.181 0.195 0.506 0.825 0.001 0.454
Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARA, angiotensin-receptor antagonist; ASA, American Society of Anesthesiologists; BMI, body mass index.
Regarding medications, the 2 groups differed in the use of angiotensin-converting enzyme inhibitors and angiotensinreceptor antagonists, which were more common in the control group (Table 1). Intraoperatively, the type of anesthesia, type of stent, volume of contrast media, type of contrast media, administered fluids, transfusion of blood units, and duration of surgery did not differ between groups (Table 2). Although urine output was higher in the mannitol group, the difference did not reach statistical significance. However, no patient required administration of furosemide in the mannitol group in contrast to the control group (p ¼ 0.041). The percentage of patients who received colloids did not differ between the 2 groups (81.4% in controls v 88.4% in mannitol group, p ¼ 0.366). Hypotensive and hypertensive episodes, the use of vasoconstrictors and vasodilators, and major postoperative complications (hemorrhage) also did not differ between groups. No patient died or required hemodialysis or other major intervention postoperatively. Regarding the primary outcome of the study, serum creatinine increased in the control group postoperatively, but in the mannitol group there was no statistically significant change (Fig 1A). At 24 h after the operation, serum creatinine levels were significantly lower in the mannitol compared to the control group (1.07 ⫾ 0.26 v 1.20 ⫾ 0.30, respectively). Creatinine clearance decreased in the control group both at 24 h and 72 h; whereas in the mannitol group, there was no statistically significant change from baseline (Fig 1B). However, there were no direct differences between the 2 groups at any time point. According to the RIFLE classification of renal dysfunction, 4 patients (9.3%) were diagnosed at renal risk and 2 (4.7%) with renal injury in the control group; whereas in the mannitol group, 6 patients (14%) were at risk but none had renal injury. The difference between the 2 groups was not statistically significant (p ¼ 0.301). Urinary levels of NGAL followed a similar course in the 2 groups, peaking at 24 h and subsequently decreasing at 72 h
without differences between the 2 groups (Fig 2A). Serum cystatin-C levels were significantly lower in the mannitol group at 24 h (controls: 2.6 ⫾ 0.9 v mannitol: 2.2 ⫾ 0.8) (Fig 2B). Serum cystatin-C levels at 24 and 72 hours correlated closely to postoperative serum creatinine levels (r ¼ 0.558, p o 0.001), and creatinine clearance (r ¼ 0.468, p o 0.001). No correlation was found between serum creatinine and corrected urine NGAL concentrations. The albumin-to-creatinine ratio of urine increased significantly in both groups postoperatively without differences between the 2 groups (data not shown). Serum urea levels also decreased significantly without differences between the groups (data not shown). Fractional excretion of urea did not change significantly in either group, and there were no differences between the groups at any time point (data not shown). Multivariate analysis with logistic regression (Table 3) revealed that diabetes mellitus, arterial hypertension, the use of ACEI or ARA, type of anesthesia, volume of crystalloids or colloids administered, urine output, and intraoperative use of diuretics were not significant predictors of renal risk or injury after EVAR. The logistic model was able to explain 67% of the variability in renal risk/injury. DISCUSSION
In this first study to evaluate the renal effects of mannitol after EVAR, mannitol improved postoperative creatinine and cystatin-C values, indicating a small but significant clinical benefit in preventing postoperative creatinine clearance decrease. Not surprisingly, the traditional and less sensitive renal markers of serum urea, albuminuria, and fractional excretion of urea failed to identify a benefit by mannitol. Mannitol has useful properties that could account for the improved renal function after EVAR. First, mannitol stimulates prostaglandin-I2 synthesis and increases renal blood flow, especially in ischemic kidneys but, more importantly, also reverses established renal vasoconstriction, a fundamental Table 2. Intraoperative Data Control
Mannitol
p
Anesthesia (ED/CSEA) 17/26 20/23 0.514 Endovascular stent fixation 29/14 29/14 1.000 (infrarenal/suprarenal) Surgery duration (min) 133 ⫾ 36 128 ⫾ 40 0.583 Crystalloids (mL) 1651 ⫾ 422 1808 ⫾ 463 0.104 Colloids (mL) 500 (0-1000) 500 (0-1000) 0.476 Contrast volume (mL) 150 (70-600) 150 (70-350) 0.598 Iodixanol/Iobitridol 27/16 34/9 0.096 Urine output (mL) 677 ⫾ 363 845 ⫾ 454 0.062 Blood transfusion (%) 9.3 7.0 0.693 Patients requiring diuretics (%) 9.3 0 0.041 Hypotensive episodes (%) 4.7 4.7 1.000 Hypertensive episodes (%) 0 0 – Patients requiring 14.0 9.3 0.501 vasodilators (%) Patients requiring 32.6 34.9 0.820 vasoconstrictors (%) Postoperative major 7.0 7.0 1.000 complications (%) Abbreviations: CSEA, combined spinal-epidural anaesthesia; ED, epidural anesthesia.
4
KALIMERIS ET AL
Fig 1. Mean values of serum creatinine (A) and creatinine clearance (B) during the observation period. *p o 0.05 between the 2 groups at the depicted time point; †p o 0.05 over time in the control group; error bars, standard deviation.
mechanism of contrast-induced nephropathy.14–16 In addition, mannitol decreased CIN in nondiabetic patients with renal insufficiency undergoing cardiac catheterization.17 Second, mannitol can scavenge free radicals released by contrast media and by ischemia-reperfusion of the lower limbs, which also ensues during EVAR.1,18 The role of oxidative damage by contrast-media is emphasized by the attenuation of CIN through reactive-oxygen species inhibition.6,7 In addition, mannitol as an osmotic diuretic maintains a high urine flow, inhibiting formation of tubular casts and swelling of endothelial cells.19 This also leads to the secondary benefit of avoiding furosemide, as was shown in the present study by the completely negated need for furosemide when mannitol was administered. A recent meta-analysis showed that furosemide at least doubled the risk of contrast-induced nephropathy, possibly by reversing blood flow within the kidney and rendering the vulnerable outer medulla ischemic.12 The authors’ results seemed to confirm this suggestion, since the omission of loop diuretics from the hydration protocol was a plausible cause of the revealed benefit of mannitol, in contrast to previous studies.11,12 Despite the early favorable outcomes with mannitol, these were not maintained after 24 hours, and there were a number of reasons that could account for this. First, renal dysfunction after EVAR also is induced by mechanisms that cannot be addressed
by mannitol. Such mechanisms are renal microembolizations (newly acquired renal infarcts are reported in 18% of patients after EVAR), bilateral renal artery stenosis, and the CT scans before discharge of the patient but also during followup.1,3 Additional factors that could have obscured a sustained benefit of mannitol at followup also included aortitis and ureteric obstruction as well as poor compliance with medications.1,20 Moreover, the effect of mannitol on diuresis gradually could have contracted the extracellular space, factitiously obscuring any greater benefit in the serum markers. Although the authors used a low-dose regimen due to the reports of renal injury after high doses of mannitol, the possibility of an even better outcome after higher doses (eg, 1-1.5 g/kg, but o200 g/day) or extended use in the postoperative period also could be examined.15,21–24 Although postoperative renal function showed some improvement with mannitol, there was no difference in the incidence of renal dysfunction according to the RIFLE criteria. The RIFLE classification is considered the most accurate in detecting and classifying renal dysfunction and also correlates closely to mortality.25 However, renal injury and frank renal failure are rather uncommon following EVAR (as shown by the 2.3% incidence of renal injury in controls and the absence of renal failure), and, therefore, a very large sample size would have been needed to detect a difference with the RIFLE
Fig 2. Urine NGAL (A) and serum cystatin-c (B) concentrations during the observation period. *p o 0.05 between the 2 groups at the depicted time point; †p o 0.01 for changes over time in each group; error bars, standard deviation.
5
MANNITOL AND RENAL DYSFUNCTION
Table 3. Logistic Regression Analysis for Renal Risk or Injury after EVAR Odds ratio (95% CI)
Age 470 years Male sex Hypertension Diabetes mellitus Statins ACEI or ARA Combined Spinal/Epidural Anesthesia Suprarenal stent fixation Surgery duration Crystalloids Colloids Contrast volume Iobitridol/ Iodixanol Intraoperative use of diuretics Urine output
2.33 0.16 4.09 0.13 0.54 0.59 3.12 0.86 0.99 0.99 1.00 1.00 1.52 4.43 1.00
(0.42-13.1) (0.01-4.43) (0.38-44.2) (0.01-1.71) (0.12-2.49) (0.14-2.55) (0.70-13.9) (0.17-4.39) (0.96-1.02) (0.99-1.00) (0.99-1.00) (0.99-1.01) (0.29-8.13) (0.18-112.6) (0.99-1.00)
p
0.335 0.278 0.245 0.122 0.432 0.482 0.136 0.860 0.487 0.441 0.512 0.965 0.622 0.367 0.858
Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARA, angiotensin receptor antagonist; CI, confidence interval; EVAR, endovascular aortic aneurysm repair.
criteria.26 Although the looser criteria of contrast-induced nephropathy occasionally have been used, they have not been validated in clinical circumstances in which other causes of renal injury coexisted, such as in EVAR.4 Nevertheless, a study with a larger sample size examining if there is a benefit in renal risk/injury after EVAR with mannitol would be the most helpful because the clinical significance of the rather small and temporary reduction of creatinine and cystatin-C values is unclear. Cystatin-C is a cysteine protease inhibitor whose levels reflect glomerular filtration more accurately than creatinine since it not affected by age, muscle mass, or infection. In addition, it has proved more sensitive in detecting CIN and that its sensitivity peaks at 12-24 hours.27,28 Similarly, urinary NGAL is a protein of the tubular epithelium that has evolved into one of the earliest and most sensitive markers of renal injury, both in the setting of surgery and in CIN.28,29 However, the accuracy of NGAL peaks at 2-4 hours after surgery and
then subsequently wanes, which could account for the lack of difference between the groups.28 A limitation of this study was that controls had higher frequency of diabetes mellitus, hypertension, and angiotensinconverting enzyme inhibitors/angiotensin-receptor antagonists intake, which could have influenced the results. Although diabetes mellitus and hypertension are recognized factors in the development of chronic renal failure, they seem to be unrelated to renal function alterations after surgery or after contrast media.30–32 In a recent study of 102 patients who underwent EVAR, hypertension and diabetes mellitus also were insignificant in predicting postoperative creatinine rise.33 In addition, the multivariate analysis revealed that diabetes mellitus and hypertension were not significant prognostic factors for renal risk/injury in the authors’ patients. The same applied for the administered amount of crystalloids and colloids and also for urine output, which were, therefore, unlikely reasons for the better outcome in the mannitol group. Finally, angiotensin-converting enzyme inhibitors have been proposed both as beneficial and deleterious for CIN, but in a recent study, their preoperative withholding did not have any influence on CIN.34,35 In addition, they do not impair the renal vasodilatory effects of mannitol.16 Therefore, although the difference between the groups should be taken into account in the interpretation of the authors’ results, there is no proof that it had significant effect in renal outcomes. In conclusion, the authors found that mannitol plus hydration offered a rather small but significant protection against renal dysfunction 24 hours after EVAR compared to hydration alone, indicating a small but statistically significant benefit in renal function markers beyond that of hydration. Considering the paucity of readily applicable measures to attenuate renal dysfunction and also the simplicity and inexpensiveness of mannitol, preventive protocols including mannitol should be examined for their efficacy. A larger study looking at the incidences of renal risk and injury is required, as well as the evaluation of larger doses of mannitol or extension of its administration postoperatively.
REFERENCES 1. Walsh SR, Tang TY, Boyle JR: Renal consequences of endovascular abdominal aortic aneurysm repair. J Endovasc Ther 15: 73-82, 2008 2. Alsac JM, Zarins CK, Heikkinen MA, et al: The impact of aortic endografts on renal function. J Vasc Surg 41:926-930, 2005 3. Harris JR, Fan CM, Geller SC, et al: Renal perfusion defects after endovascular repair of abdominal aortic aneurysms. J Vasc Interv Radiol 4:329-333, 2003 4. Wong GT, Irwin MG: Contrast-induced nephropathy. Br J Anaesth 99:474-483, 2007 5. Allie DE, Lirtzman MD, Wyatt CH, et al: Targeted renal therapy and contrast-induced nephropathy during endovascular abdominal aortic aneurysm repair: Results of a feasibility pilot trial. J Endovasc Ther 14:520-527, 2007 6. Persson PB, Hansell P, Liss P: Pathophysiology of contrast medium-induced nephropathy. Kidney Int 68:14-22, 2005 7. Katholi RE, Woods WT Jr, Taylor GJ, et al: Oxygen free radicals and contrast nephropathy. Am J Kidney Dis 32:64-71, 1998
8. Wahlberg E, DiMuzio PJ, Stoney RJ: Aortic clamping during elective operations for infrarenal disease: The influence of clamping time on renal function. J Vasc Surg 36:13-18, 2002 9. Erley C: Concomitant drugs with exposure to contrast media. Kidney Int 69:S20-S24, 2006 10. Haveman JW, Gansevoort RT, Bongaerts AH, et al: Low incidence of nephropathy in surgical ICU patients receiving intravenous contrast: A retrospective analysis. Intensive Care Med 32:1199-1205, 2006 11. Solomon R, Werner C, Mann D, et al: Effects of saline, mannitol, and furosemide to prevent acute decreases in renal function induced by radiocontrast agents. N Engl J Med 331:1416-1420, 1994 12. Majumdar SR, Kjellstrand CM, Tymchak WJ, et al: Forced euvolemic diuresis with mannitol and furosemide for prevention of contrast-induced nephropathy in patients with CKD undergoing coronary angiography: A randomized controlled trial. Am J Kidney Dis 54: 602-609, 2009 13. Cruz DN, Ricci Z, Ronco C: Clinical review: RIFLE and AKIN– time for reappraisal. Crit Care 13:211, 2009
6
14. Behnia R, Koushanpour E, Brunner EA: Effects of hyperosmotic mannitol infusion on hemodynamics of dog kidney. Anesth Analg 82: 902-908, 1996 15. Better OS, Rubinstein I, Winaver JM, et al: Mannitol therapy revisited (1940-1997). Kidney Int 52:886-894, 1997 16. Johnston PA, Bernard DB, Perrin NS, et al: Prostaglandins mediate the vasodilatory effect of mannitol in the hypoperfused rat kidney. J Clin Invest 68:127-133, 1981 17. Weisberg LS, Kurnik PB, Kurnik BR: Risk of radiocontrast nephropathy in patients with and without diabetes mellitus. Kidney Int 45:259-265, 1994 18. Moore NN, Lapsley M, Norden AG, et al: Does N-acetylcysteine prevent contrast-induced nephropathy during endovascular AAA repair? A randomized controlled pilot study. J Endovasc Ther 13: 660-666, 2006 19. Lameire N, Vanholder R: Pathophysiologic features and prevention of human and experimental acute tubular necrosis. J Am Soc Nephrol 17:S20-S32, 2001 20. Vallabhaneni SR, McWilliams RG, Anbarasu A, et al: Perianeurysmal fibrosis: A relative contraindication to endovascular repair. Eur J Vasc Endovasc Surg 21:567-568, 2001 21. Dorman HR, Sondheimer JH, Chadnapaphomhai P: Mannitol induced acute renal failure. Medicine 69:153-159, 1990 22. Nicholson ML, Baker DM, Hopkinson BR, et al: Randomised controlled trial of the effect of Mannitol on renal reperfusion injury during aortic aneurysm surgery. Br J Surg 83:1230-1233, 1996 23. Wijnen MHWA, Vader HL, Van Den Wall Bake AWL, et al: Can renal dysfunction after infra-renal aortic aneurysm repair be modified by multi-antioxidant supplementation? J Cardiovasc Surg 43: 483-488, 2002 24. van Valenberg PL, Hoitsma AJ, Tiggeler RG, et al: Mannitol as an indispensable constituent of an intraoperative hydration protocol for the prevention of acute renal failure after renal cadaveric transplantation. Transplantation 44:784-788, 1987
KALIMERIS ET AL
25. Uchino S, Bellomo R, Goldsmith D, et al: An assessment of the RIFLE criteria for acute renal failure in hospitalized patients. Crit Care Med 34:1913-1917, 2006 26. Wald R, Waikar SS, Liangos O, et al: Acute renal failure after endovascular vs open repair of abdominal aortic aneurysm. J Vasc Surg 43:460-466, 2006 27. Kimmel M, Butscheid M, Brenner S, et al: Improved estimation of glomerular filtration rate by serum cystatin C in preventing contrast induced nephropathy by N-acetylcysteine or zinc–preliminary results. Nephrol Dial Transplant 23:1241-1245, 2008 28. Moore E, Bellomo R, Nichol A: Biomarkers of acute kidney injury in anesthesia, intensive care and major surgery: from the bench to clinical research to clinical practice. Minerva Anestesiol 76: 425-440, 2010 29. Ling W, Zhaohui N, Ben H, et al: Urinary IL-18 and NGAL as early predictive biomarkers in contrast-induced nephropathy after coronary angiography. Nephron Clin Pract 108:c176-c181, 2008 30. Abelha FJ, Botelho M, Fernandes V, et al: Determinants of postoperative acute kidney injury. Crit Care 13:R79, 2009 31. Cochran ST, Wong WS, Roe DJ: Predicting angiographyinduced acute renal function impairment: Clinical risk model. AJR Am J Roentgenol 141:1027-1033, 1983 32. Meguid El Nahas A, Bello AK: Chronic kidney disease: The global challenge. Lancet 365:331-340, 2005 33. Brooks CE, Middleton A, Dhillon R, et al: Predictors of creatinine rise post-endovascular abdominal aortic aneurysm repair. ANZ J Surg 81:827-830, 2011 34. Li X, Li T, Cong H: Is angiotensin-converting enzyme inhibitor a contraindication for contrast-induced nephropathy prophylaxis? A review about its paradox. Cardiovasc Therg 30:273-276, 2011 35. Rosenstock JL, Bruno R, Kim JK, et al: The effect of withdrawal of ACE inhibitors or angiotensin receptor blockers prior to coronary angiography on the incidence of contrast-induced nephropathy. Int Urol Nephrol 40:749-755, 2008