- Email: [email protected]
Comparison of Perioperative Renal Function Between Epidural and Intravenous Patient-Controlled Analgesia After Living-Donor Hepatectomy: A Retrospective Study
Comparison of Perioperative Renal Function Between Epidural and Intravenous Patient-Controlled Analgesia After Living-Donor Hepatectomy: A Retrospective Study S.Y. Hama,b, E.J. Kima,b, T.H. Kima, and B.-N. Kooa,b,* a Department of Anesthesiology and Pain Medicine, Yonsei University College of Medicine, Seoul, Republic of Korea; and bAnesthesia and Pain Research Institute, Yonsei University College of Medicine, Seoul, Republic of Korea
ABSTRACT Background. Sympathetic blockade associated with epidural analgesia was reported to be a risk factor for acute kidney injury (AKI) following liver resection. The purpose of this study was to compare the incidence of AKI after living-donor hepatectomies according to the type of patient-controlled analgesia (PCA). Methods. A total of 316 patients after living-donor hepatectomy were retrospectively analyzed; 148 patients in the epidural PCA group and 168 patients in the intravenous (IV) PCA group were evaluated. AKI was defined as an increase in serum creatinine 0.3 mg/dL, ie, 1.5-fold from the baseline, or a reduction in the urine output in the first 48 hours after surgery, based on the Acute Kidney Injury Network criteria. Logistic regression analysis was performed to identify the independent risk factors for AKI after living-donor hepatectomy. Results. Baseline characteristics were similar between the 2 groups except the age. Volumes of fluids and colloids administered intraoperatively were greater in the epidural PCA group (P < .001 and P ¼ .006, respectively). The incidence of AKI did not show significant differences between the 2 groups (8.1% vs 7.1%; P ¼ .747). In multivariate analysis, preoperative serum alanine transaminase level 50 U/L was identified as a risk factor for postoperative AKI. However, epidural PCA failed to be a risk factor for postoperative AKI. Conclusions. The type of PCA did not affect the incidence of postoperative AKI after living-donor hepatectomy. Despite significant differences in the postoperative hemodynamics, the incidence of AKI was similar between 2 groups.
E
PIDURAL analgesia is a common analgesic technique in patients undergoing living-donor hepatectomy and provides better pain management after living-donor liver donation [1]. Epidural analgesia improves bowel activity and reduces cardiovascular and pulmonary complications in the postoperative phase. However, despite these benefits, there are concerns over the increased risk of epidural hematoma, infection, urinary retention, and hypotension [2,3]. Postoperative coagulopathy and perioperative hypotension due to sympathetic blockade can be problematic, especially in patients undergoing liver resection. To decrease the amount of blood loss and blood transfusion requirements during liver resection, various cardiopulmonary interventions, including
ª 2018 Elsevier Inc. All rights reserved. 230 Park Avenue, New York, NY 10169
Transplantation Proceedings, 50, 1365e1371 (2018)
lowering of the central venous pressure (CVP), have been tried [4]. Sympathetic blockade associated with epidural analgesia can be beneficial for lowering the blood pressure. However, fluid restriction for lowering the CVP, combined with sympathetic blockade, can lead to an increased risk of AKI. Recently, epidural analgesia was reported to be an
*Address correspondence to Bon-Nyeo Koo, Department of Anesthesiology and Pain Medicine, Anesthesia and Pain Research Institute, Yonsei University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, 120-752 Seoul, Korea. E-mail: [email protected] 0041-1345/18 https://doi.org/10.1016/j.transproceed.2018.03.006
1365
1366
HAM, KIM, KIM ET AL Table 1. Patient Characteristics Characteristic
Epidural PCA (n ¼ 148)
IV PCA (n ¼ 168)
P Value
Age, y Sex, M/F Height, cm Weight, kg ASA PS, I/II Hypertension Diabetes mellitus Emergency operation Operation time, min Anesthesia time, min Major liver resection† Preop AST, IU/L Preop ALT, IU/L Preop platelets, 103/mL Intraop fluid, mL Intraop colloid, mL Intraop pRBC, mL Intraop transfusion Intraop urine output, mL Intraop blood loss, mL Hospital stay, d ICU stay, d
36.03 10.83 (19e62) 89/59 168.53 8.39 64.40 9.51 133/14 4 (2.7%) 0 7 (4.7%) 415.71 66.71 (265e600) 478.89 965.52 (330e645) 136 (92.5%) 16.86 4.89 16.89 10.18 263.95 56.48 3,017.74 944.25 (1,250e5,750) 557.69 174.99 (0e1,000) 270.00 294.96 3 (2.0%) 861.97 756.17 (100e8,100) 450.35 352.75 17.16 5.11 2.36 0.57
32.28 10.11 (16e61) 98/70 168.64 8.60 63.74 9.86 151/16 1 (0.6%) 1 (0.6%) 3 (1.8%) 384.53 64.07 (140e646) 449.27 71.80 (160e695) 161 (96.4%) 17.29 6.92 17.04 14.14 256.59 65.90 2,459.64 801.77 (200e5,400) 507.27 58.83 (500e1,000) 0 0 616.34 432.51 (80e3,660) 387.44 205.95 17.68 13.50 2.12 1.36
.002* .745 .909 .545 .986 .190 1.000 .198 <.001* <.001* .129 .537 .918 .297 <.001* .006* .45 .103 .001* .063 .638 .048*
Note. Values are presented as n (%) or mean SD (range). Abbreviations: PCA, patient-controlled anesthesia; IV, intravenous; ASA PS, American Society of Anesthesiologists Physical Status; AST, aspartate transaminase; ALT, alanine transaminase; pRBC, packed red blood cells; ICU, intensive care unit. *P < .05. † Right or left lobectomy of liver.
independent risk factor for AKI after major liver resection [5]. The incidence of AKI after liver resection was 12%-15% [6,7]. Because it is associated with postoperative morbidities such as increased rates of mechanical ventilation, renal replacement therapy, and hospital mortality [6], it is important to investigate the risk factors for AKI. In this retrospective study, we hypothesized that sympathetic blockade associated with epidural analgesia can be a risk factor for AKI following liver resection. Furthermore, to exclude the effect of the diseased liver on perioperative
kidney function, we included only donors with healthy livers to undergo liver resection. The purpose of the present study was to compare the incidence of AKI after living-donor hepatectomy under either epidural or intravenous (IV) patient-controlled analgesia (PCA).
METHODS After obtaining approval from the Institutional Review Board of Severance Hospital (4-2016-1040), we retrospectively reviewed the
Table 2. Perioperative Kidney Function Parameter
Epidural PCA (n ¼ 148)
IV PCA (n ¼ 168)
P Value
Acute kidney injury Before surgery Cr, mg/dL eGFR, mL/min/1.73 m2 Immediately after surgery Cr eGFR Postoperative day 1 Cr eGFR Urine output, mL Postoperative day 2 Cr eGFR Urine output
12 (8.1%)
12 (7.1%)
.747
0.82 0.16 (0.42e1.1) 109.61 15.21 (70.27e145)
0.77 0.16 (0.47e1.21) 115.53 12.51 (69e153)
.004* <.001*
0.79 0.19 (0.41e1.21) 113.42 14.96 (74e148)
0.71 0.18 (0.34e1.06) 119.89 12.02 (85e167)
<.001* <.001*
0.71 0.16 (0.36e1.19) 120.92 14.02 (78.26e150) 3,208.95 960.84 (870e6,180)
0.63 0.17 (0.33e1.45) 126.13 14.52 (64e167) 2,668.28 944.57 (510e5,410)
<.001* .002* <.001*
0.72 0.16 (0.39e1.15) 121.28 16.44 (77e172.1) 3,040.06 1,022.96 (980e7,500)
0.68 0.18 (0.29e1.21) 122.39 14.54 (66e158) 2,458.66 791.58 (1,080e4,990)
.034* .543 <.001*
Note. Values are presented as n (%) or mean SD (range). *P < .05.
ACUTE KIDNEY INJURY AFTER LIVING-DONOR HEPATECTOMY
1367
Table 3. Long-Term Follow-up Record Epidural PCA (n ¼ 148)
IV PCA (n ¼ 168)
P Value
10.25 8.28 (1.8e70) 0.73 0.16 (0.36e1.13) 118.90 14.70 (76e172.1) 37.74 18.29 53.75 33.94 (7e244) 259.00 78.11 10.33 7.73
7.45 3.05 (1.9e17.9) 0.66 0.14 (0.37e1.03) 124.52 12.89 (90e157) 41.85 22.32 43.77 31.21 (7e237) 246.89 78.26 15.26 32.45
<.001* <.001* .001 .090 .009* .194 .564
10.96 2.92 0.79 0.16 (0.45e1.2) 112.19 13.89 (76e144) 24.74 12.74 21.29 16.19 224.32 74.56 5.74 7.98
10.38 2.98 0.72 0.15 (0.37e1.12) 119.28 11.60 (86e153) 24.51 12.08 17.90 15.57 217.55 54.29 37.00 54.71
.096 <.001* <.001* .873 .065 .365 .054
14.21 19.77 0.83 0.16 (0.48e1.2) 109.16 14.34 (68.84e137) 24.20 25.30 21.06 15.27 202.47 53.48
13.10 11.21 0.76 0.16 (0.38e1.3) 116.34 12.75 (72e157) 21.00 6.99 18.88 10.98 204.98 55.97
.565 <.001* <.001* .177 .177 .713
13.24 10.04 0.85 0.16 (0.45e1.3) 108.68 15.74 (63.27e146.28) 20.24 7.70 17.56 9.60 197.88 45.99
11.97 2.97 0.78 0.15 (0.44e1.05) 114.58 13.07 (78e151) 19.29 4.67 15.98 6.67 205.39 55.00
.203 .002* .003* .271 .154 .279
12.40 3.39 0.85 0.16 (0.45e1.13) 108.69 15.31 19.72 7.19 19.31 27.41 195.70 44.74 (93e300)
11.81 2.79 0.77 0.16 (0.47e1.07) 114.00 13.87 21.09 11.56 14.50 4.18 220.42 63.63 (120e456)
.384 .029* .096 .457 .327 .027*
Measure
Discharge BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL CRP, mg/L 1 month BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL CRP, mg/L 3 months BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL 6 months BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL 12 months BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL
Note. Values are presented as mean SD (range). *P < .05.
medical records of patients who underwent living-donor hepatectomy from August 2006 to June 2015. Three hundred eighteen cases of living-donor hepatectomy were found. Patients with perioperative discontinuation of PCA were excluded. The primary end point of this study was to compare the incidence of AKI after living-donor hepatectomy. The secondary end points were identifying the risk factors of AKI after living-donor hepatectomy, perioperative kidney function (serum creatinine level, estimated glomerular filtration rate [eGFR]), postoperative pain score, use of rescue analgesics, perioperative hemodynamics, and incidence of complications related to PCA. All patients underwent preoperative evaluation, including history taking, physical examination, and laboratory tests, and the related data were collected based on the electronic medical records. Standardized general anesthesia was administered to all patients. The patients were premedicated with the use of glycopyrrolate (0.1 mg) intravenously immediately before induction. Anesthesia was induced with the use of propofol (2 mg/kg), desflurane, or sevoflurane and remifentanil. Either rocuronium (0.6 mg/kg) or
cisatracurium (0.15 mg/kg) was used to facilitate endotracheal intubation. The anesthesia was maintained by means of continuous infusion of remifentanil (0.05e2 mg/kg/min) and 1.0 minimum alveolar concentration of desflurane or sevoflurane. Electrocardiography, pulse oximetry, invasive arterial blood pressure, and central venous pressure were monitored throughout the operation. The radial artery was cannulated to monitor the blood pressure. A central venous catheter was placed via the internal jugular vein to monitor CVP and administer vasoactive drugs and fluid. CVP was maintained at w4e5 mm Hg by restricting fluid administration and adjusting the depth of anesthesia. At the end of the surgery, neuromuscular blockade was reversed with the use of 0.2 mg glycopyrrolate and 1 mg neostigmine. Ramosetron (0.3 mg) or ondansetron (4 mg) was given for antiemesis. After extubating the trachea, the patient was transferred to the post-anesthesia care unit. Intravenous PCA consisted of fentanyl (10e20 mg/kg), administered at a basal rate of 2 mL/h, bolus dose of 0.5 mL with a lockout time of 15 minutes, and provided with the PCA infusion device for 48 hours after surgery.
1368
HAM, KIM, KIM ET AL Table 4. Logistic Regression Analysis to Find Possible Risk Factors for Developing Acute Kidney Injury Univariate Factor
Age Preop ALT 50 Intraop T/F Intraop colloids Intraop catecholamines Epidural PCA Preop Cr
OR (95% CI)
0.968 8.272 11.219 0.529 0.565 1.104 0.532
(0.924e1.015) (1.214e56.378) (0.826e152.359) (0.189e1.476) (0.113e2.819) (0.415e2.939) (0.063e4.496)
Multivariate P Value
.177 .031* .069 .224 .486 .842 .562
OR (95% CI)
P Value
6.810 (1.178e39.354) 6.810 (0.593e78.205)
.032* .123
Abbreviations: OR, odds ratio; CI, confidence interval; others as in Table 1. *P < .05.
In the epidural PCA group, an epidural catheter was inserted at the thoracic level (around T6-T10) and advanced 3 cm upward. After a test dose of 3 mL lidocaine (2%) with epinephrine (1:200 000), 5 mL lidocaine (2%) was administered before surgery, via the epidural route, according to the institute protocol. Ropivacaine (0.15%-0.2%) and fentanyl (5e10 mg/mL) at a basal rate of 2 mL/h, bolus dose of 0.5 mL with a lockout time of 15 minutes, were provided with the PCA infusion device for 48 hours after surgery. Preoperative data including sex, age at the time of the surgery, height, weight, and medical history were recorded. Intraoperative data included surgical and anesthetic times, fluid input, blood loss, urine output, and transfusion of blood products. As for the postoperative data, fluid input, transfusion of blood products, blood loss, urine output, serum creatinine level, and eGFR were collected until the 2nd postoperative day. To compare the efficacy and effectiveness of the different types of PCA, postoperative pain score, use of additional rescue analgesics, and the incidence of complications related to PCA were recorded. Pain was assessed by an 11-point numeric pain score (0 ¼ no pain; 10 ¼ worst pain) by an assigned nurse. Rescue analgesics were a combination of acetaminophen and tramadol, acetaminophen, nonsteroidal antiinflammatory drugs (NSAIDs), IV tramadol, or IV pethidine. Long-term laboratory data, such as serum creatinine level, eGFR, aspartate transaminase, alanine transaminase (ALT), platelets, and C-reactive protein, were also recorded for a year after surgery. AKI was defined as an absolute increase in serum creatinine 0.3 mg/dL, ie, 1.5-fold from the baseline, or a reduction in urine output within a 48-hour period, based on Acute Kidney Injury Network (AKIN) criteria [8].
Statistical Analysis Statistical analysis was performed with the use of IBM SPSS Statistics 23 (SPSS, Chicago, Illinois). Categoric variables were expressed as n (%) and compared between groups with the use of Fisher exact test or chi-square test. Continuous variables were expressed as mean SD and compared between the groups by means of the 2-sample test or Mann-Whitney U test. A linear mixed model was used to compare the tendency of mean values of serum creatinine level between the 2 groups measured for a year after surgery. Logistic regression analysis was constructed to investigate independent risk factors for AKI after living-donor hepatectomy. To evaluate the possible risk factors for AKI, a univariate logistic regression analysis was performed. For the multivariate model, a stepwise selection method was used to select the variables that showed statistical significance in the univariate analysis. Because age at the time of the surgery was significantly different between the 2
groups, multiple linear regression analysis was performed to decide if it was a confounder. Statistical significance was accepted for P < .05.
RESULTS
During the 10-year study period, a total of 318 cases of livingdonor hepatectomy were identified. One patient was excluded because of postoperative discontinuation of epidural PCA, and another was excluded because of cancellation of the donation. As a result, 148 patients in the epidural PCA group and 168 patients in the IV PCA group were evaluated. All of the patients could be categorized as having American Society of Anesthesiologists Physical Status (ASA PS) I or II. There were no significant differences between the 2 groups in the baseline preoperative characteristics, except for their ages at the time of the surgery. Patients in the epidural PCA group were older, although all of the patients were <65 years of age. Moreover, the operating time and anesthesia time were significantly longer in the epidural PCA group. Volumes of fluid and colloid administered during surgery were greater in the epidural PCA group, and the duration of intensive care unit stay was longer in the epidural PCA group (Table 1). There was no mortality or morbidity throughout the 1st postoperative year in both groups. Table 2 presents the perioperative data related to kidney function for both groups. The incidence of AKI did not show any significant differences between the epidural and IV PCA groups (8.1% vs 7.1%, respectively; P ¼ .747). Perioperative creatinine level was significantly higher in the epidural PCA group than in the IV PCA group. However, despite the statistically significant difference of serum creatinine level between the groups, serum creatinine levels throughout the perioperative period were within the normal range. Furthermore, serum creatinine level was consistently higher for the 1st postoperative year in the epidural PCA group with statistical significance (Table 3); however, linear mixed-model analysis revealed that the tendency of the mean value of creatinine level over time was similar between the 2 groups (P ¼ .135). The amount of urine output was greater in the epidural PCA group throughout the postoperative 48-hour period. None of the patients required continuous renal placement therapy. In both univariate logistic regression analysis and stepwise backward multivariate analysis, only preoperative ALT 50
ACUTE KIDNEY INJURY AFTER LIVING-DONOR HEPATECTOMY
Fig 1. Postoperative hemodynamics. Values are expressed as mean SD. Abbreviations: PCA, patient-controlled anesthesia; IV, intravenous; SBP, systolic blood pressure; DBP, diastolic blood pressure. *P < .05.
U/L was identified as a risk factor for postoperative AKI. However, the analyses failed to demonstrate that epidural PCA was a risk factor for postoperative AKI (Table 4). Figure 1 shows the postoperative hemodynamic data. In the immediate postoperative period, both systolic and diastolic blood pressures were lower in the epidural PCA group, and the heart rate was higher in the IV PCA group for 3 postoperative days. The incidence of postoperative nausea was higher in the IV PCA group (12.2% vs 29.9%; P < .001; Table 5). The highest pain score measured by the 11-point numeric pain score was higher in the epidural PCA group (6.61 2.00 vs 5.07 1.71; P < .001). Moreover, PCA failure rate, which is defined as the highest visual analog scale score being 8 in the 2 days after surgery, was higher in the epidural PCA group (21.6% vs 8.9%; P < .001). DISCUSSION
In this retrospective study, we compared the incidence of AKI and perioperative renal function between patients receiving epidural and IV PCA after living-donor
1369
hepatectomy. Despite differences in the perioperative fluid intake, urine output, blood pressure, heart rate, and postoperative serum creatinine level, the incidence of AKI was similar between the 2 groups. Although the type of PCA might affect the perioperative renal function and hemodynamics, it did not have an effect on the occurrence of AKI in patients undergoing living-donor hepatectomy. It is very challenging to determine how to control postoperative pain in patients undergoing living-donor hepatectomy. Despite its superior pain management after living-donor liver donation [1], epidural analgesia, because of its associated sympathetic blockade and hypotension, might be associated with a risk of renal dysfunction. Recently, epidural PCA was reported to be an independent risk factor for AKI after major liver resection [5]. Lowering the CVP to decrease the amount of blood loss and blood transfusion requirements in liver resection [4] can be an additional contributive factor for renal dysfunction, in addition to sympathetic blockade associated with epidural analgesia. Epidural PCA might influence perioperative kidney function by means of sympathetic blockade. A greater amount of fluid was administered in the epidural PCA group during surgery, and the urine output was greater in the epidural PCA group during the 48 hours after surgery. This finding is in accordance with a report that suggested postoperative hypotension to be the reason for greater fluid requirement [9]. Epidural PCA can be associated with hypotension due to sympathetic blockade; the current study revealed that immediate postoperative hypotension is more common in the epidural PCA group than in the IV PCA group. This finding can be related to greater fluid requirement in the epidural PCA group. Furthermore, heart rates were faster in the IV PCA group during the 3 postoperative days in this study. However, hypotension resulting from sympathetic blockade was not associated with an increased incidence of AKI in healthy living donors undergoing hepatectomy. Contrary to a previous study [5], the present study did not identify epidural PCA as a risk factor for postoperative AKI. However, there were several differences between our studies. First, although the previous study compared epidural PCA and no PCA, we compared epidural PCA and IV PCA. In addition, a higher concentration of ropivacaine (0.33%) was used in the previous study than in ours (0.15%0.2%), which resulted in more profound sensory and Table 5. Adverse Events Related to PCA Event
Epidural PCA (n ¼ 148)
IV PCA (n ¼ 168)
P Value
Nausea Vomiting Pruritus Dizziness VAS
18 (12.2%) 3 (2.0%) 16 (10.9%) 7 (4.8%) 6.61 2.00
50 (29.9%) 4 (2.4%) 20 (12.0%) 8 (4.8%) 5.07 1.71
<.001* 1.000 .762 .991 <.001*
Note. Values are presented as n (%) or mean SD.; Abbreviations: VAS, visual analog scale; others as in Table 1. *P < .05.
1370
sympathetic blocks. Second, the present study enrolled healthy patients undergoing living-donor hepatectomy. Because the donors were ASA PS I or II and with healthy livers, we could exclude the effect of diseased liver on the perioperative kidney function. The incidence of AKI was 8.1% in the epidural PCA group and 7.1% in the IV PCA group in the present study, compared with 12%-15% in the previous study [6,7]. This low incidence might be due to enrollment of healthy patients. Third, the previous study [5] had selection bias, ie, major liver resections were included in the epidural PCA group (57.3% vs 40%); in our study, there was no difference regarding extent of resection (Table 1). Thus, despite the significant differences in the hemodynamics and fluid requirement following sympathetic blockade associated with epidural analgesia, sympathetic blockade itself does not affect the incidence of AKI in healthy patients undergoing living-donor hepatectomy. However, even though creatinine level was within the normal limits and none of the patients were diagnosed as AKI by the AKIN criteria or required renal replacement therapy, it is prudent to keep in mind that epidural PCA can possess possible risk of renal dysfunction. Because living-donor hepatectomy is intended for healthy patients, it is very important to predict the postoperative morbidity and identify the risk factors. Several possible risk factors of AKI were identified in previous studies conducted on liver surgery [5,6]. Preoperative eGFR, preoperative hypertension, and intraoperative red blood cell transfusion were previously identified as possible risk factors for AKI after liver resection surgery [6]. Moreover, another study found that epidural anesthesia, age, blood transfusion, extent of liver resection, and preoperative kidney function were strong predictors of AKI [5]. Different types of surgery may have different risk factors for AKI. Body mass index, number of comorbidities, and preoperative creatinine concentrations were independent predictors for AKI in patients undergoing esophageal surgery [10]. In lung resection surgery, hypertension, peripheral vascular disease, preoperative use of angiotensin II receptor blocker, and intraoperative hydroxyethyl starch administration were risk factors for developing AKI [11]. Another study conducted in patients with major orthopedic surgery revealed perioperative diuretics, nonsteroidal antiinflammatory drugs, and angiotensin-converting enzyme inhibitor or angiotensin receptor blocker to be risk factors for postoperative AKI [12]. This discrepancy may be due to different surgical characteristics. Because liver resection surgery can be associated with massive bleeding, hypovolemia due to bleeding or aforementioned fluid restrictive surgery can lead to AKI. The fact that blood transfusion was not a risk factor for AKI in other surgeries, but in liver resection surgery, disproves this. However, living-donor hepatectomy is usually accompanied with less bleeding than liver resection surgery. Furthermore, even though the patients undergo the same type of surgery, there is a difference in the presence of liver disease. Based on our results, preoperative serum ALT level 50 U/L was identified as an independent risk factor for developing AKI in patients undergoing living-donor hepatectomy.
HAM, KIM, KIM ET AL
However, a previous study demonstrated that preoperative serum ALT level was not a risk factor for postoperative AKI after major hepatectomy [5]. Whereas previous studies enrolled patients undergoing liver resection, the present study differs in that it was aimed at healthy patients without liver disease. Therefore, preoperative serum ALT level that was previously not known to be a risk factor could be an independent risk factor in healthy donors. Furthermore, it is possible that epidural analgesia failed to be a risk factor for AKI in patients after living-donor hepatectomy because of a different patient population. In patients undergoing living-donor hepatectomy, epidural PCA is commonly used [1]. However, studies comparing the effect of IV and epidural PCA in patients undergoing living-donor hepatectomy are scarce. Mann et al compared the safety and effectiveness on postoperative pain of epidural and IV PCA in elderly patients undergoing major abdominal surgery [13]. They concluded that both techniques of PCA were effective but that epidural PCA provided better analgesia. Other studies have reported better analgesic effect of epidural PCA compared with systemic opioids [14]. However, in studies on patients with liver resection, w20% of patients experienced inadequate postoperative pain control. In addition, the authors explained the causes of epidural failure to be inadequate block and dislodged or leaking catheter [9]. In accordance with this result, PCA failure rate, which is defined as the highest VAS score being 8 during the 2 days after surgery, was higher in the epidural PCA group in the present study: 21.6%. Despite previous studies demonstrating better analgesia with epidural PCA [13,14], the higher failure rate of the epidural technique can be a serious concern. Previous studies reported the PCA failure rates to be 8% and 18% for IV and epidural PCA, respectively [9,15]. However, the higher failure rate and higher VAS score in the present study may be associated with targeting young and healthy patients. Despite the conflicting result of age-related differences of pain perception [16], several studies have suggested that younger patients are more sensitive to pain. One study demonstrated that younger patients perceived more intense pain sensation during the transrectal ultrasoundeguided biopsy procedure [17]. Another study suggested that pressure nociception decreased with age [18]. From this point of view, it is much more complicated to control pain in patients undergoing living-donor hepatectomy. We could not find any significant difference in the occurrence of AKI between the 2 groups, even though we re-analyzed the occurrence of AKI after excluding the cases of PCA failure. The incidence of adverse events can be a significant consideration when selecting the method of PCA. The incidence of nausea was higher in the IV PCA group. The results of previous studies conducted to compare the various methods of PCA regarding the incidence of nausea were inconsistent. In a study comparing the efficacy of IV and epidural PCA, the incidence of nausea was higher on the 3rd postoperative day in the epidural PCA group [1].
ACUTE KIDNEY INJURY AFTER LIVING-DONOR HEPATECTOMY
Another study demonstrated that oral intake without nausea occurred significantly faster in the epidural PCA group [13]. Furthermore, faster return of normal bowel function was noted [19]. From these results, it seems that epidural PCA has a favorable effect on the physiology of the gastrointestinal tract. There were several limitations of the present study. First, because the study was a retrospective review, the regimen of PCA was not uniform. Second, ages at the time of surgery were different between the 2 groups. However, age was not a significant variable associated with the incidence of AKI in a multiple linear regression analysis when statistically adjusted, and all of the patients were <65 years old. In addition, because of the retrospective study design, we could obtain information only from electronic medical records. Therefore, the factors that can affect the results of the measurement of the variables were not consistent. CONCLUSION
This study demonstrated that epidural PCA did not affect the incidence of postoperative AKI in living liver donors, even though there were significant differences in the postoperative hemodynamics, fluid balances, and serum creatinine level between the 2 groups. REFERENCES [1] Clarke H, Chandy T, Srinivas C, et al. Epidural analgesia provides better pain management after live liver donation: a retrospective study. Liver Transpl 2011;17:315e23. [2] Cook TM, Counsell D, Wildsmith JA. Major complications of central neuraxial block: report on the Third National Audit Project of the Royal College of Anaesthetists. Br J Anaesth 2009;102:179e90. [3] Tzimas P, Prout J, Papadopoulos G, et al. Epidural anaesthesia and analgesia for liver resection. Anaesthesia 2013;68: 628e35. [4] Gurusamy KS, Li J, Vaughan J, et al. Cardiopulmonary interventions to decrease blood loss and blood transfusion requirements for liver resection. Cochrane Database Syst Rev 2012;(5):CD007338.
1371 [5] Kambakamba P, Slankamenac K, Tschuor C, et al. Epidural analgesia and perioperative kidney function after major liver resection. Br J Surg 2015;102:805e12. [6] Tomozawa A, Ishikawa S, Shiota N, et al. Perioperative risk factors for acute kidney injury after liver resection surgery: an historical cohort study. Can J Anaesth 2015;62:753e61. [7] Slankamenac K, Breitenstein S, Held U, et al. Development and validation of a prediction score for postoperative acute renal failure following liver resection. Ann Surg 2009;250:720e8. [8] Mehta RL, Kellum JA, Shah SV, et al. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007;11:R31. [9] Revie EJ, Massie LJ, McNally SJ, et al. Effectiveness of epidural analgesia following open liver resection. HPB (Oxford) 2011;13:206e11. [10] Konda P, Ai D, Guerra CE, et al. Identification of risk factors associated with postoperative acute kidney injury after esophagectomy for esophageal cancer. J Cardiothorac Vasc Anesth 2017;31:474e81. [11] Ishikawa S, Griesdale DE, Lohser J. Acute kidney injury after lung resection surgery: incidence and perioperative risk factors. Anesth Analg 2012;114:1256e62. [12] Ying T, Chan S, Lane S, et al. Acute kidney injury postmajor orthopaedic surgery: a single-centre case-control study. Nephrology (Carlton) 2018;23:126e32. [13] Mann C, Pouzeratte Y, Boccara G, et al. Comparison of intravenous or epidural patient-controlled analgesia in the elderly after major abdominal surgery. Anesthesiology 2000;92:433e41. [14] Liu S, Carpenter RL, Neal JM. Epidural anesthesia and analgesia. Their role in postoperative outcome. Anesthesiology 1995;82:1474e506. [15] Chhibber A, Dziak J, Kolano J, et al. Anesthesia care for adult live donor hepatectomy: our experiences with 100 cases. Liver Transpl 2007;13:537e42. [16] Kivrak Y, Kose-Ozlece H, Ustundag MF, et al. Pain perception: predictive value of sex, depression, anxiety, somatosensory amplification, obesity, and age. Neuropsychiatr Dis Treat 2016;12:1913e8. [17] Dell’Atti L, Borea PA, Russo GR. Age: “a natural anesthetic” in pain perception during the transrectal ultrasound-guided prostate biopsy procedure. Urologia 2011;78:257e61. [18] Pickering G, Jourdan D, Eschalier A, et al. Impact of age, gender and cognitive functioning on pain perception. Gerontology 2002;48:112e8. [19] Rawlinson A, Kitchingham N, Hart C, et al. Mechanisms of reducing postoperative pain, nausea and vomiting: a systematic review of current techniques. Evid Based Med 2012;17:75e80.
ABSTRACT Background. Sympathetic blockade associated with epidural analgesia was reported to be a risk factor for acute kidney injury (AKI) following liver resection. The purpose of this study was to compare the incidence of AKI after living-donor hepatectomies according to the type of patient-controlled analgesia (PCA). Methods. A total of 316 patients after living-donor hepatectomy were retrospectively analyzed; 148 patients in the epidural PCA group and 168 patients in the intravenous (IV) PCA group were evaluated. AKI was defined as an increase in serum creatinine 0.3 mg/dL, ie, 1.5-fold from the baseline, or a reduction in the urine output in the first 48 hours after surgery, based on the Acute Kidney Injury Network criteria. Logistic regression analysis was performed to identify the independent risk factors for AKI after living-donor hepatectomy. Results. Baseline characteristics were similar between the 2 groups except the age. Volumes of fluids and colloids administered intraoperatively were greater in the epidural PCA group (P < .001 and P ¼ .006, respectively). The incidence of AKI did not show significant differences between the 2 groups (8.1% vs 7.1%; P ¼ .747). In multivariate analysis, preoperative serum alanine transaminase level 50 U/L was identified as a risk factor for postoperative AKI. However, epidural PCA failed to be a risk factor for postoperative AKI. Conclusions. The type of PCA did not affect the incidence of postoperative AKI after living-donor hepatectomy. Despite significant differences in the postoperative hemodynamics, the incidence of AKI was similar between 2 groups.
E
PIDURAL analgesia is a common analgesic technique in patients undergoing living-donor hepatectomy and provides better pain management after living-donor liver donation [1]. Epidural analgesia improves bowel activity and reduces cardiovascular and pulmonary complications in the postoperative phase. However, despite these benefits, there are concerns over the increased risk of epidural hematoma, infection, urinary retention, and hypotension [2,3]. Postoperative coagulopathy and perioperative hypotension due to sympathetic blockade can be problematic, especially in patients undergoing liver resection. To decrease the amount of blood loss and blood transfusion requirements during liver resection, various cardiopulmonary interventions, including
ª 2018 Elsevier Inc. All rights reserved. 230 Park Avenue, New York, NY 10169
Transplantation Proceedings, 50, 1365e1371 (2018)
lowering of the central venous pressure (CVP), have been tried [4]. Sympathetic blockade associated with epidural analgesia can be beneficial for lowering the blood pressure. However, fluid restriction for lowering the CVP, combined with sympathetic blockade, can lead to an increased risk of AKI. Recently, epidural analgesia was reported to be an
*Address correspondence to Bon-Nyeo Koo, Department of Anesthesiology and Pain Medicine, Anesthesia and Pain Research Institute, Yonsei University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, 120-752 Seoul, Korea. E-mail: [email protected] 0041-1345/18 https://doi.org/10.1016/j.transproceed.2018.03.006
1365
1366
HAM, KIM, KIM ET AL Table 1. Patient Characteristics Characteristic
Epidural PCA (n ¼ 148)
IV PCA (n ¼ 168)
P Value
Age, y Sex, M/F Height, cm Weight, kg ASA PS, I/II Hypertension Diabetes mellitus Emergency operation Operation time, min Anesthesia time, min Major liver resection† Preop AST, IU/L Preop ALT, IU/L Preop platelets, 103/mL Intraop fluid, mL Intraop colloid, mL Intraop pRBC, mL Intraop transfusion Intraop urine output, mL Intraop blood loss, mL Hospital stay, d ICU stay, d
36.03 10.83 (19e62) 89/59 168.53 8.39 64.40 9.51 133/14 4 (2.7%) 0 7 (4.7%) 415.71 66.71 (265e600) 478.89 965.52 (330e645) 136 (92.5%) 16.86 4.89 16.89 10.18 263.95 56.48 3,017.74 944.25 (1,250e5,750) 557.69 174.99 (0e1,000) 270.00 294.96 3 (2.0%) 861.97 756.17 (100e8,100) 450.35 352.75 17.16 5.11 2.36 0.57
32.28 10.11 (16e61) 98/70 168.64 8.60 63.74 9.86 151/16 1 (0.6%) 1 (0.6%) 3 (1.8%) 384.53 64.07 (140e646) 449.27 71.80 (160e695) 161 (96.4%) 17.29 6.92 17.04 14.14 256.59 65.90 2,459.64 801.77 (200e5,400) 507.27 58.83 (500e1,000) 0 0 616.34 432.51 (80e3,660) 387.44 205.95 17.68 13.50 2.12 1.36
.002* .745 .909 .545 .986 .190 1.000 .198 <.001* <.001* .129 .537 .918 .297 <.001* .006* .45 .103 .001* .063 .638 .048*
Note. Values are presented as n (%) or mean SD (range). Abbreviations: PCA, patient-controlled anesthesia; IV, intravenous; ASA PS, American Society of Anesthesiologists Physical Status; AST, aspartate transaminase; ALT, alanine transaminase; pRBC, packed red blood cells; ICU, intensive care unit. *P < .05. † Right or left lobectomy of liver.
independent risk factor for AKI after major liver resection [5]. The incidence of AKI after liver resection was 12%-15% [6,7]. Because it is associated with postoperative morbidities such as increased rates of mechanical ventilation, renal replacement therapy, and hospital mortality [6], it is important to investigate the risk factors for AKI. In this retrospective study, we hypothesized that sympathetic blockade associated with epidural analgesia can be a risk factor for AKI following liver resection. Furthermore, to exclude the effect of the diseased liver on perioperative
kidney function, we included only donors with healthy livers to undergo liver resection. The purpose of the present study was to compare the incidence of AKI after living-donor hepatectomy under either epidural or intravenous (IV) patient-controlled analgesia (PCA).
METHODS After obtaining approval from the Institutional Review Board of Severance Hospital (4-2016-1040), we retrospectively reviewed the
Table 2. Perioperative Kidney Function Parameter
Epidural PCA (n ¼ 148)
IV PCA (n ¼ 168)
P Value
Acute kidney injury Before surgery Cr, mg/dL eGFR, mL/min/1.73 m2 Immediately after surgery Cr eGFR Postoperative day 1 Cr eGFR Urine output, mL Postoperative day 2 Cr eGFR Urine output
12 (8.1%)
12 (7.1%)
.747
0.82 0.16 (0.42e1.1) 109.61 15.21 (70.27e145)
0.77 0.16 (0.47e1.21) 115.53 12.51 (69e153)
.004* <.001*
0.79 0.19 (0.41e1.21) 113.42 14.96 (74e148)
0.71 0.18 (0.34e1.06) 119.89 12.02 (85e167)
<.001* <.001*
0.71 0.16 (0.36e1.19) 120.92 14.02 (78.26e150) 3,208.95 960.84 (870e6,180)
0.63 0.17 (0.33e1.45) 126.13 14.52 (64e167) 2,668.28 944.57 (510e5,410)
<.001* .002* <.001*
0.72 0.16 (0.39e1.15) 121.28 16.44 (77e172.1) 3,040.06 1,022.96 (980e7,500)
0.68 0.18 (0.29e1.21) 122.39 14.54 (66e158) 2,458.66 791.58 (1,080e4,990)
.034* .543 <.001*
Note. Values are presented as n (%) or mean SD (range). *P < .05.
ACUTE KIDNEY INJURY AFTER LIVING-DONOR HEPATECTOMY
1367
Table 3. Long-Term Follow-up Record Epidural PCA (n ¼ 148)
IV PCA (n ¼ 168)
P Value
10.25 8.28 (1.8e70) 0.73 0.16 (0.36e1.13) 118.90 14.70 (76e172.1) 37.74 18.29 53.75 33.94 (7e244) 259.00 78.11 10.33 7.73
7.45 3.05 (1.9e17.9) 0.66 0.14 (0.37e1.03) 124.52 12.89 (90e157) 41.85 22.32 43.77 31.21 (7e237) 246.89 78.26 15.26 32.45
<.001* <.001* .001 .090 .009* .194 .564
10.96 2.92 0.79 0.16 (0.45e1.2) 112.19 13.89 (76e144) 24.74 12.74 21.29 16.19 224.32 74.56 5.74 7.98
10.38 2.98 0.72 0.15 (0.37e1.12) 119.28 11.60 (86e153) 24.51 12.08 17.90 15.57 217.55 54.29 37.00 54.71
.096 <.001* <.001* .873 .065 .365 .054
14.21 19.77 0.83 0.16 (0.48e1.2) 109.16 14.34 (68.84e137) 24.20 25.30 21.06 15.27 202.47 53.48
13.10 11.21 0.76 0.16 (0.38e1.3) 116.34 12.75 (72e157) 21.00 6.99 18.88 10.98 204.98 55.97
.565 <.001* <.001* .177 .177 .713
13.24 10.04 0.85 0.16 (0.45e1.3) 108.68 15.74 (63.27e146.28) 20.24 7.70 17.56 9.60 197.88 45.99
11.97 2.97 0.78 0.15 (0.44e1.05) 114.58 13.07 (78e151) 19.29 4.67 15.98 6.67 205.39 55.00
.203 .002* .003* .271 .154 .279
12.40 3.39 0.85 0.16 (0.45e1.13) 108.69 15.31 19.72 7.19 19.31 27.41 195.70 44.74 (93e300)
11.81 2.79 0.77 0.16 (0.47e1.07) 114.00 13.87 21.09 11.56 14.50 4.18 220.42 63.63 (120e456)
.384 .029* .096 .457 .327 .027*
Measure
Discharge BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL CRP, mg/L 1 month BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL CRP, mg/L 3 months BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL 6 months BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL 12 months BUN, mg/dL Cr, mg/dL eGFR, mL/min/1.73m2 AST, IU/L ALT, IU/L Platelets, 103/uL
Note. Values are presented as mean SD (range). *P < .05.
medical records of patients who underwent living-donor hepatectomy from August 2006 to June 2015. Three hundred eighteen cases of living-donor hepatectomy were found. Patients with perioperative discontinuation of PCA were excluded. The primary end point of this study was to compare the incidence of AKI after living-donor hepatectomy. The secondary end points were identifying the risk factors of AKI after living-donor hepatectomy, perioperative kidney function (serum creatinine level, estimated glomerular filtration rate [eGFR]), postoperative pain score, use of rescue analgesics, perioperative hemodynamics, and incidence of complications related to PCA. All patients underwent preoperative evaluation, including history taking, physical examination, and laboratory tests, and the related data were collected based on the electronic medical records. Standardized general anesthesia was administered to all patients. The patients were premedicated with the use of glycopyrrolate (0.1 mg) intravenously immediately before induction. Anesthesia was induced with the use of propofol (2 mg/kg), desflurane, or sevoflurane and remifentanil. Either rocuronium (0.6 mg/kg) or
cisatracurium (0.15 mg/kg) was used to facilitate endotracheal intubation. The anesthesia was maintained by means of continuous infusion of remifentanil (0.05e2 mg/kg/min) and 1.0 minimum alveolar concentration of desflurane or sevoflurane. Electrocardiography, pulse oximetry, invasive arterial blood pressure, and central venous pressure were monitored throughout the operation. The radial artery was cannulated to monitor the blood pressure. A central venous catheter was placed via the internal jugular vein to monitor CVP and administer vasoactive drugs and fluid. CVP was maintained at w4e5 mm Hg by restricting fluid administration and adjusting the depth of anesthesia. At the end of the surgery, neuromuscular blockade was reversed with the use of 0.2 mg glycopyrrolate and 1 mg neostigmine. Ramosetron (0.3 mg) or ondansetron (4 mg) was given for antiemesis. After extubating the trachea, the patient was transferred to the post-anesthesia care unit. Intravenous PCA consisted of fentanyl (10e20 mg/kg), administered at a basal rate of 2 mL/h, bolus dose of 0.5 mL with a lockout time of 15 minutes, and provided with the PCA infusion device for 48 hours after surgery.
1368
HAM, KIM, KIM ET AL Table 4. Logistic Regression Analysis to Find Possible Risk Factors for Developing Acute Kidney Injury Univariate Factor
Age Preop ALT 50 Intraop T/F Intraop colloids Intraop catecholamines Epidural PCA Preop Cr
OR (95% CI)
0.968 8.272 11.219 0.529 0.565 1.104 0.532
(0.924e1.015) (1.214e56.378) (0.826e152.359) (0.189e1.476) (0.113e2.819) (0.415e2.939) (0.063e4.496)
Multivariate P Value
.177 .031* .069 .224 .486 .842 .562
OR (95% CI)
P Value
6.810 (1.178e39.354) 6.810 (0.593e78.205)
.032* .123
Abbreviations: OR, odds ratio; CI, confidence interval; others as in Table 1. *P < .05.
In the epidural PCA group, an epidural catheter was inserted at the thoracic level (around T6-T10) and advanced 3 cm upward. After a test dose of 3 mL lidocaine (2%) with epinephrine (1:200 000), 5 mL lidocaine (2%) was administered before surgery, via the epidural route, according to the institute protocol. Ropivacaine (0.15%-0.2%) and fentanyl (5e10 mg/mL) at a basal rate of 2 mL/h, bolus dose of 0.5 mL with a lockout time of 15 minutes, were provided with the PCA infusion device for 48 hours after surgery. Preoperative data including sex, age at the time of the surgery, height, weight, and medical history were recorded. Intraoperative data included surgical and anesthetic times, fluid input, blood loss, urine output, and transfusion of blood products. As for the postoperative data, fluid input, transfusion of blood products, blood loss, urine output, serum creatinine level, and eGFR were collected until the 2nd postoperative day. To compare the efficacy and effectiveness of the different types of PCA, postoperative pain score, use of additional rescue analgesics, and the incidence of complications related to PCA were recorded. Pain was assessed by an 11-point numeric pain score (0 ¼ no pain; 10 ¼ worst pain) by an assigned nurse. Rescue analgesics were a combination of acetaminophen and tramadol, acetaminophen, nonsteroidal antiinflammatory drugs (NSAIDs), IV tramadol, or IV pethidine. Long-term laboratory data, such as serum creatinine level, eGFR, aspartate transaminase, alanine transaminase (ALT), platelets, and C-reactive protein, were also recorded for a year after surgery. AKI was defined as an absolute increase in serum creatinine 0.3 mg/dL, ie, 1.5-fold from the baseline, or a reduction in urine output within a 48-hour period, based on Acute Kidney Injury Network (AKIN) criteria [8].
Statistical Analysis Statistical analysis was performed with the use of IBM SPSS Statistics 23 (SPSS, Chicago, Illinois). Categoric variables were expressed as n (%) and compared between groups with the use of Fisher exact test or chi-square test. Continuous variables were expressed as mean SD and compared between the groups by means of the 2-sample test or Mann-Whitney U test. A linear mixed model was used to compare the tendency of mean values of serum creatinine level between the 2 groups measured for a year after surgery. Logistic regression analysis was constructed to investigate independent risk factors for AKI after living-donor hepatectomy. To evaluate the possible risk factors for AKI, a univariate logistic regression analysis was performed. For the multivariate model, a stepwise selection method was used to select the variables that showed statistical significance in the univariate analysis. Because age at the time of the surgery was significantly different between the 2
groups, multiple linear regression analysis was performed to decide if it was a confounder. Statistical significance was accepted for P < .05.
RESULTS
During the 10-year study period, a total of 318 cases of livingdonor hepatectomy were identified. One patient was excluded because of postoperative discontinuation of epidural PCA, and another was excluded because of cancellation of the donation. As a result, 148 patients in the epidural PCA group and 168 patients in the IV PCA group were evaluated. All of the patients could be categorized as having American Society of Anesthesiologists Physical Status (ASA PS) I or II. There were no significant differences between the 2 groups in the baseline preoperative characteristics, except for their ages at the time of the surgery. Patients in the epidural PCA group were older, although all of the patients were <65 years of age. Moreover, the operating time and anesthesia time were significantly longer in the epidural PCA group. Volumes of fluid and colloid administered during surgery were greater in the epidural PCA group, and the duration of intensive care unit stay was longer in the epidural PCA group (Table 1). There was no mortality or morbidity throughout the 1st postoperative year in both groups. Table 2 presents the perioperative data related to kidney function for both groups. The incidence of AKI did not show any significant differences between the epidural and IV PCA groups (8.1% vs 7.1%, respectively; P ¼ .747). Perioperative creatinine level was significantly higher in the epidural PCA group than in the IV PCA group. However, despite the statistically significant difference of serum creatinine level between the groups, serum creatinine levels throughout the perioperative period were within the normal range. Furthermore, serum creatinine level was consistently higher for the 1st postoperative year in the epidural PCA group with statistical significance (Table 3); however, linear mixed-model analysis revealed that the tendency of the mean value of creatinine level over time was similar between the 2 groups (P ¼ .135). The amount of urine output was greater in the epidural PCA group throughout the postoperative 48-hour period. None of the patients required continuous renal placement therapy. In both univariate logistic regression analysis and stepwise backward multivariate analysis, only preoperative ALT 50
ACUTE KIDNEY INJURY AFTER LIVING-DONOR HEPATECTOMY
Fig 1. Postoperative hemodynamics. Values are expressed as mean SD. Abbreviations: PCA, patient-controlled anesthesia; IV, intravenous; SBP, systolic blood pressure; DBP, diastolic blood pressure. *P < .05.
U/L was identified as a risk factor for postoperative AKI. However, the analyses failed to demonstrate that epidural PCA was a risk factor for postoperative AKI (Table 4). Figure 1 shows the postoperative hemodynamic data. In the immediate postoperative period, both systolic and diastolic blood pressures were lower in the epidural PCA group, and the heart rate was higher in the IV PCA group for 3 postoperative days. The incidence of postoperative nausea was higher in the IV PCA group (12.2% vs 29.9%; P < .001; Table 5). The highest pain score measured by the 11-point numeric pain score was higher in the epidural PCA group (6.61 2.00 vs 5.07 1.71; P < .001). Moreover, PCA failure rate, which is defined as the highest visual analog scale score being 8 in the 2 days after surgery, was higher in the epidural PCA group (21.6% vs 8.9%; P < .001). DISCUSSION
In this retrospective study, we compared the incidence of AKI and perioperative renal function between patients receiving epidural and IV PCA after living-donor
1369
hepatectomy. Despite differences in the perioperative fluid intake, urine output, blood pressure, heart rate, and postoperative serum creatinine level, the incidence of AKI was similar between the 2 groups. Although the type of PCA might affect the perioperative renal function and hemodynamics, it did not have an effect on the occurrence of AKI in patients undergoing living-donor hepatectomy. It is very challenging to determine how to control postoperative pain in patients undergoing living-donor hepatectomy. Despite its superior pain management after living-donor liver donation [1], epidural analgesia, because of its associated sympathetic blockade and hypotension, might be associated with a risk of renal dysfunction. Recently, epidural PCA was reported to be an independent risk factor for AKI after major liver resection [5]. Lowering the CVP to decrease the amount of blood loss and blood transfusion requirements in liver resection [4] can be an additional contributive factor for renal dysfunction, in addition to sympathetic blockade associated with epidural analgesia. Epidural PCA might influence perioperative kidney function by means of sympathetic blockade. A greater amount of fluid was administered in the epidural PCA group during surgery, and the urine output was greater in the epidural PCA group during the 48 hours after surgery. This finding is in accordance with a report that suggested postoperative hypotension to be the reason for greater fluid requirement [9]. Epidural PCA can be associated with hypotension due to sympathetic blockade; the current study revealed that immediate postoperative hypotension is more common in the epidural PCA group than in the IV PCA group. This finding can be related to greater fluid requirement in the epidural PCA group. Furthermore, heart rates were faster in the IV PCA group during the 3 postoperative days in this study. However, hypotension resulting from sympathetic blockade was not associated with an increased incidence of AKI in healthy living donors undergoing hepatectomy. Contrary to a previous study [5], the present study did not identify epidural PCA as a risk factor for postoperative AKI. However, there were several differences between our studies. First, although the previous study compared epidural PCA and no PCA, we compared epidural PCA and IV PCA. In addition, a higher concentration of ropivacaine (0.33%) was used in the previous study than in ours (0.15%0.2%), which resulted in more profound sensory and Table 5. Adverse Events Related to PCA Event
Epidural PCA (n ¼ 148)
IV PCA (n ¼ 168)
P Value
Nausea Vomiting Pruritus Dizziness VAS
18 (12.2%) 3 (2.0%) 16 (10.9%) 7 (4.8%) 6.61 2.00
50 (29.9%) 4 (2.4%) 20 (12.0%) 8 (4.8%) 5.07 1.71
<.001* 1.000 .762 .991 <.001*
Note. Values are presented as n (%) or mean SD.; Abbreviations: VAS, visual analog scale; others as in Table 1. *P < .05.
1370
sympathetic blocks. Second, the present study enrolled healthy patients undergoing living-donor hepatectomy. Because the donors were ASA PS I or II and with healthy livers, we could exclude the effect of diseased liver on the perioperative kidney function. The incidence of AKI was 8.1% in the epidural PCA group and 7.1% in the IV PCA group in the present study, compared with 12%-15% in the previous study [6,7]. This low incidence might be due to enrollment of healthy patients. Third, the previous study [5] had selection bias, ie, major liver resections were included in the epidural PCA group (57.3% vs 40%); in our study, there was no difference regarding extent of resection (Table 1). Thus, despite the significant differences in the hemodynamics and fluid requirement following sympathetic blockade associated with epidural analgesia, sympathetic blockade itself does not affect the incidence of AKI in healthy patients undergoing living-donor hepatectomy. However, even though creatinine level was within the normal limits and none of the patients were diagnosed as AKI by the AKIN criteria or required renal replacement therapy, it is prudent to keep in mind that epidural PCA can possess possible risk of renal dysfunction. Because living-donor hepatectomy is intended for healthy patients, it is very important to predict the postoperative morbidity and identify the risk factors. Several possible risk factors of AKI were identified in previous studies conducted on liver surgery [5,6]. Preoperative eGFR, preoperative hypertension, and intraoperative red blood cell transfusion were previously identified as possible risk factors for AKI after liver resection surgery [6]. Moreover, another study found that epidural anesthesia, age, blood transfusion, extent of liver resection, and preoperative kidney function were strong predictors of AKI [5]. Different types of surgery may have different risk factors for AKI. Body mass index, number of comorbidities, and preoperative creatinine concentrations were independent predictors for AKI in patients undergoing esophageal surgery [10]. In lung resection surgery, hypertension, peripheral vascular disease, preoperative use of angiotensin II receptor blocker, and intraoperative hydroxyethyl starch administration were risk factors for developing AKI [11]. Another study conducted in patients with major orthopedic surgery revealed perioperative diuretics, nonsteroidal antiinflammatory drugs, and angiotensin-converting enzyme inhibitor or angiotensin receptor blocker to be risk factors for postoperative AKI [12]. This discrepancy may be due to different surgical characteristics. Because liver resection surgery can be associated with massive bleeding, hypovolemia due to bleeding or aforementioned fluid restrictive surgery can lead to AKI. The fact that blood transfusion was not a risk factor for AKI in other surgeries, but in liver resection surgery, disproves this. However, living-donor hepatectomy is usually accompanied with less bleeding than liver resection surgery. Furthermore, even though the patients undergo the same type of surgery, there is a difference in the presence of liver disease. Based on our results, preoperative serum ALT level 50 U/L was identified as an independent risk factor for developing AKI in patients undergoing living-donor hepatectomy.
HAM, KIM, KIM ET AL
However, a previous study demonstrated that preoperative serum ALT level was not a risk factor for postoperative AKI after major hepatectomy [5]. Whereas previous studies enrolled patients undergoing liver resection, the present study differs in that it was aimed at healthy patients without liver disease. Therefore, preoperative serum ALT level that was previously not known to be a risk factor could be an independent risk factor in healthy donors. Furthermore, it is possible that epidural analgesia failed to be a risk factor for AKI in patients after living-donor hepatectomy because of a different patient population. In patients undergoing living-donor hepatectomy, epidural PCA is commonly used [1]. However, studies comparing the effect of IV and epidural PCA in patients undergoing living-donor hepatectomy are scarce. Mann et al compared the safety and effectiveness on postoperative pain of epidural and IV PCA in elderly patients undergoing major abdominal surgery [13]. They concluded that both techniques of PCA were effective but that epidural PCA provided better analgesia. Other studies have reported better analgesic effect of epidural PCA compared with systemic opioids [14]. However, in studies on patients with liver resection, w20% of patients experienced inadequate postoperative pain control. In addition, the authors explained the causes of epidural failure to be inadequate block and dislodged or leaking catheter [9]. In accordance with this result, PCA failure rate, which is defined as the highest VAS score being 8 during the 2 days after surgery, was higher in the epidural PCA group in the present study: 21.6%. Despite previous studies demonstrating better analgesia with epidural PCA [13,14], the higher failure rate of the epidural technique can be a serious concern. Previous studies reported the PCA failure rates to be 8% and 18% for IV and epidural PCA, respectively [9,15]. However, the higher failure rate and higher VAS score in the present study may be associated with targeting young and healthy patients. Despite the conflicting result of age-related differences of pain perception [16], several studies have suggested that younger patients are more sensitive to pain. One study demonstrated that younger patients perceived more intense pain sensation during the transrectal ultrasoundeguided biopsy procedure [17]. Another study suggested that pressure nociception decreased with age [18]. From this point of view, it is much more complicated to control pain in patients undergoing living-donor hepatectomy. We could not find any significant difference in the occurrence of AKI between the 2 groups, even though we re-analyzed the occurrence of AKI after excluding the cases of PCA failure. The incidence of adverse events can be a significant consideration when selecting the method of PCA. The incidence of nausea was higher in the IV PCA group. The results of previous studies conducted to compare the various methods of PCA regarding the incidence of nausea were inconsistent. In a study comparing the efficacy of IV and epidural PCA, the incidence of nausea was higher on the 3rd postoperative day in the epidural PCA group [1].
ACUTE KIDNEY INJURY AFTER LIVING-DONOR HEPATECTOMY
Another study demonstrated that oral intake without nausea occurred significantly faster in the epidural PCA group [13]. Furthermore, faster return of normal bowel function was noted [19]. From these results, it seems that epidural PCA has a favorable effect on the physiology of the gastrointestinal tract. There were several limitations of the present study. First, because the study was a retrospective review, the regimen of PCA was not uniform. Second, ages at the time of surgery were different between the 2 groups. However, age was not a significant variable associated with the incidence of AKI in a multiple linear regression analysis when statistically adjusted, and all of the patients were <65 years old. In addition, because of the retrospective study design, we could obtain information only from electronic medical records. Therefore, the factors that can affect the results of the measurement of the variables were not consistent. CONCLUSION
This study demonstrated that epidural PCA did not affect the incidence of postoperative AKI in living liver donors, even though there were significant differences in the postoperative hemodynamics, fluid balances, and serum creatinine level between the 2 groups. REFERENCES [1] Clarke H, Chandy T, Srinivas C, et al. Epidural analgesia provides better pain management after live liver donation: a retrospective study. Liver Transpl 2011;17:315e23. [2] Cook TM, Counsell D, Wildsmith JA. Major complications of central neuraxial block: report on the Third National Audit Project of the Royal College of Anaesthetists. Br J Anaesth 2009;102:179e90. [3] Tzimas P, Prout J, Papadopoulos G, et al. Epidural anaesthesia and analgesia for liver resection. Anaesthesia 2013;68: 628e35. [4] Gurusamy KS, Li J, Vaughan J, et al. Cardiopulmonary interventions to decrease blood loss and blood transfusion requirements for liver resection. Cochrane Database Syst Rev 2012;(5):CD007338.
1371 [5] Kambakamba P, Slankamenac K, Tschuor C, et al. Epidural analgesia and perioperative kidney function after major liver resection. Br J Surg 2015;102:805e12. [6] Tomozawa A, Ishikawa S, Shiota N, et al. Perioperative risk factors for acute kidney injury after liver resection surgery: an historical cohort study. Can J Anaesth 2015;62:753e61. [7] Slankamenac K, Breitenstein S, Held U, et al. Development and validation of a prediction score for postoperative acute renal failure following liver resection. Ann Surg 2009;250:720e8. [8] Mehta RL, Kellum JA, Shah SV, et al. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care 2007;11:R31. [9] Revie EJ, Massie LJ, McNally SJ, et al. Effectiveness of epidural analgesia following open liver resection. HPB (Oxford) 2011;13:206e11. [10] Konda P, Ai D, Guerra CE, et al. Identification of risk factors associated with postoperative acute kidney injury after esophagectomy for esophageal cancer. J Cardiothorac Vasc Anesth 2017;31:474e81. [11] Ishikawa S, Griesdale DE, Lohser J. Acute kidney injury after lung resection surgery: incidence and perioperative risk factors. Anesth Analg 2012;114:1256e62. [12] Ying T, Chan S, Lane S, et al. Acute kidney injury postmajor orthopaedic surgery: a single-centre case-control study. Nephrology (Carlton) 2018;23:126e32. [13] Mann C, Pouzeratte Y, Boccara G, et al. Comparison of intravenous or epidural patient-controlled analgesia in the elderly after major abdominal surgery. Anesthesiology 2000;92:433e41. [14] Liu S, Carpenter RL, Neal JM. Epidural anesthesia and analgesia. Their role in postoperative outcome. Anesthesiology 1995;82:1474e506. [15] Chhibber A, Dziak J, Kolano J, et al. Anesthesia care for adult live donor hepatectomy: our experiences with 100 cases. Liver Transpl 2007;13:537e42. [16] Kivrak Y, Kose-Ozlece H, Ustundag MF, et al. Pain perception: predictive value of sex, depression, anxiety, somatosensory amplification, obesity, and age. Neuropsychiatr Dis Treat 2016;12:1913e8. [17] Dell’Atti L, Borea PA, Russo GR. Age: “a natural anesthetic” in pain perception during the transrectal ultrasound-guided prostate biopsy procedure. Urologia 2011;78:257e61. [18] Pickering G, Jourdan D, Eschalier A, et al. Impact of age, gender and cognitive functioning on pain perception. Gerontology 2002;48:112e8. [19] Rawlinson A, Kitchingham N, Hart C, et al. Mechanisms of reducing postoperative pain, nausea and vomiting: a systematic review of current techniques. Evid Based Med 2012;17:75e80.