A national shortage of suitable cadaveric livers available for transplantation has resulted in the establishment of multiple living donor liver transplant programs. In 1989, the first successful living donor liver transplant was performed using the left lobe (1), and since then both left and right lobe donor hepatectomy (RLDH) have rapidly gained acceptance as an effective procedure for selected patients with end-stage liver disease (2).
The overall outcome of living donor liver transplantation should be evaluated from the perspectives of both donor and recipient. Because an increasing number of RLDHs are being performed (2), multiple aspects of donor outcome should be evaluated. The donor is typically a healthy individual. Therefore, minimizing postdonation complications and reducing the impact of donation on the donor quality of life is extremely important and could potentially affect future acceptance of this form of transplantation. Several studies conducted have shown that RLDH is a safe procedure for the donor with only minor complications being reported (3).
One of the major concerns of any patient undergoing a surgical procedure is postoperative pain. Poor postoperative pain control may have a significant negative impact on the quality of life of an otherwise healthy donor. After performing several living donor liver transplants in our institution, we observed that patients who underwent RLDH reported a significant amount of postoperative pain despite placement of a thoracic patient-controlled epidural analgesia (PCEA) infusion catheter for pain management. We chose to retrospectively analyze this group of patients and later compare them with patients who had undergone major hepatic resection for tumor (MHRT) in an attempt to elucidate the cause for this observation.
After obtaining IRB approval for this investigation, we retrospectively analyzed 15 adult patients who underwent RLDH and compared them with 15 adult patients who had undergone MHRT. The intraoperative anesthetic management and surgical approaches were similar in both groups. In addition, patients in both groups had a thoracic epidural catheter placed (T8–10) before induction of general anesthesia. All epidural catheters were evaluated to exclude inadvertent subarachnoid or intravascular placement by aspiration and via a test dose injection consisting of lidocaine 1.5% with epinephrine 5 μg/mL (total test dose of 3 mL). These tested epidural catheters were then connected to an infusion pump during surgery. All PCEA infusion pumps were programmed for a basal rate infusion of 6 mL/h and a patient-activated bolus of 3 mL every 15 min. All PCEA infusions were initiated within the first 2 h after induction of general anesthesia at the basal rate only without administration of either an initial dose or “clinician dose.”
The choice of PCEA infusion anesthetic solution for each patient was based on the preference of the attending anesthesiologist. In the RLDH group, 10 patients received an epidural solution consisting of 0.1% bupivacaine with fentanyl 5 μg/mL, whereas three patients received 0.125% bupivacaine without opioid and two patients received a mixture of 0.1% bupivacaine with morphine 0.05 mg/mL. In the MHRT group, 11 patients received epidural solution consisting of 0.1% bupivacaine with fentanyl 5 μg/mL, whereas four patients received a mixture of 0.1% bupivacaine with morphine 0.05 mg per mL. All epidural solutions were assumed to be equianalgesic.
All patients in both groups had standard ASA monitors placed for surgery and were monitored for radial arterial blood pressure and internal jugular central venous pressure. General anesthesia was induced by thiopental and a nondepolarizing muscle relaxant to facilitate endotracheal intubation. General anesthesia was maintained with isoflurane and a nondepolarizing muscle relaxant. IV fentanyl was dosed according to the clinical judgment of the attending anesthesiologist at the time of induction and intubation. Finally, all PCEA basal rate infusions were begun within the first 2 h after the start of surgery.
A chevron incision was used to achieve surgical exposure with the same surgical retractor system used in all cases. Intraoperatively, anesthetic and fluid management were titrated to achieve hemodynamic stability as well as for adequate surgical exposure of the liver (heart rate and arterial blood pressure within 20% of initial values and urine output >0.5 mL · kg−1 · h−1). At the conclusion of surgery, all patients were awakened and tracheally extubated in the operating room. Postoperatively, all MHRT patients were transferred to the postanesthesia care unit (PACU) and all RLDH patients were transferred to the surgical intensive care unit (SICU). Postoperative resting pain was assessed using a 0–10 visual analog pain score (VAPS) on arrival to either PACU or SICU and then subsequently at 6-h intervals up to 48 h after surgery. In patients with inadequate pain control on arrival to either the PACU or SICU, the PCEA infusion catheters were tested with 4 to 6 mL of 2% lidocaine to determine if an adequate sensory level could be obtained. Requested and delivered PCEA doses, clinician rescue doses, and the presence of side effects (including nausea or vomiting, pruritus, hypotension, lower extremity motor deficit, and level of sedation during the same 6-h intervals) were also recorded. All patients were followed postoperatively by a dedicated acute pain medicine service (APMS) with all interventions regarding postoperative pain management being prescribed by the APMS without the authors’ involvement.
Patient demographic data including age, gender, height, and weight, as well as intraoperative data including duration of surgery, estimated blood loss, and total administered IV fluids and opioids were collected. The cumulative total of all nonepidural (IV or oral) analgesic medications and the dosages administered during the postoperative period were collected as well.
Demographic data between the two patient groups, as well as all intraoperative data, were analyzed using the nonparametric Mann-Whitney U-test when data were not normally distributed and Student’s t-test when data were normally distributed. Summary statistics are presented as the mean with standard deviations for normally distributed data and as median with quartiles for data not normally distributed.
The two patient groups were compared for the PCEA doses requested and delivered using repeated-measures analysis of variance after log transformation. The Mann-Whitney U-test was used to compare the total bupivacaine dose (mg) administered between groups via PCEA infusion, and Student’s t-test was used to compare the total volume of PCEA solution administered between groups. Both pain scores, as well as side effects, including sedation scores resulting from PCEA infusion use, were evaluated using generalized estimating equations with a cumulative logit model adjusting for repeated measures. The correlation between the two non-normally distributed continuous variables was assessed using Spearman correlation coefficients with 95% confidence intervals (CI).
The RLDH group, on average, had a significantly longer surgery time and larger estimated blood loss than did the MHRT group (Table 1). No significant difference regarding the PCEA dose requested or the cumulative doses delivered between groups (P = 0.34 and P = 0.81, respectively) was observed (Figs. 1 and 2). In addition, there were no differences between groups for total local anesthetic dose or volume delivered (including patient demand doses and clinician doses) over the observed 48-h postoperative period (Table 2). In both groups, there were several patients (n = 4 in the RLDH group, n = 5 in the MHRT group, respectively) with inadequate pain control after emergence from general anesthesia and on arrival to either the PACU or SICU. In these patients, the epidural catheter was tested with 4 to 6 mL of 2% lidocaine to determine if an adequate sensory level could be obtained. Two patients in each group had upward adjustments in their PCEA basal infusion rate. However, neither the use of lidocaine test doses nor the upward adjustments of the PCEA basal infusion rate had a statistically significant effect on the average total volume of epidural solution administered between groups.
Overall, the RLDH group had a significantly greater chance of having more pain than did the MHRT group (P = 0.034, Fig. 3). An odds ratio of 2.76 (1.12–6.82, 95% CI) indicates that the incidence of higher pain scores were 2.76 times more likely in the RLDH group than in MHRT patients. There was no significant difference in the incidence of side effects such as nausea or vomiting, pruritus, hypotension, lower extremity motor deficit or sedation between groups. A weak correlation was found between postoperative VAPS scores and PCEA doses requested or delivered and total volume of PCEA solution administered in either group (Tables 3 and 4 and Fig. 4).
Peak VAPS scores occurred at 24 and 42 h postoperatively in the RLDH and MHRT groups, respectively. The most postoperative requested and delivered PCEA doses occurred at 18 and 36 h, respectively, in the RLDH group and at 24 and 6 h, respectively, in the MHRT group.
Neither IV nor oral opioids were administered to any patient in the MHRT group during the postoperative study period. However, in the RLDH group, four patients received IV fentanyl with the total cumulative dose being only 150 μg or less for each of these patients within the first 48 h postoperatively. No oral opioids were administered to RLDH patients.
Living related liver donation provides for the expansion of the potential pool of allografts available for transplantation (2). It is important that the overall success of this procedure is determined by the outcome of both the donor and the recipient. Although RLDH is considered safe, with infrequent major intraoperative and postoperative complications, one should evaluate the medical, psychological, and financial impact of donation as well as the resulting donor quality of life (3). Significant postoperative pain after RLDH can cause physical and psychological distress in an otherwise healthy group of patients. Living liver donation, however, as shown by Trotter et al. (3), did not affect the donor’s relationship with the recipient, and all patients stated they would donate again if necessary. In addition, the authors observed that postoperative pain was more intense than expected in the donor group and more severe when compared with the recipient patient group.
In this study, we found that despite the use of thoracic PCEA infusion catheters, RLDH patients still experienced significant postoperative pain. By comparing postoperative pain in RLDH patients with patients who underwent MHRT, we conclude there may be additional factors (beyond the significantly longer duration of surgery observed) predisposing this patient population to increased postoperative pain perception. As elucidated by Trotter et al. (3), patients who underwent RLDH had no abdominal symptoms or discomfort preoperatively, thereby making postoperative pain more noticeable. In contrast, some patients undergoing MHRT may already have associated abdominal pain preoperatively. Therefore, postoperative pain, although present, may be an improvement compared with their preoperative presentation. However, in this study, we did not specifically investigate the presence of preoperative pain in the MHRT group.
In both patient groups, the methods for surgical exposure were similar. However, overall RLDH patient surgical time was significantly longer than that observed in MHRT patients. It is possible that longer durations of tissue retraction and surgical manipulation may result in modulation of the neuroendocrine system, thereby causing more postoperative pain. The central nervous system responds to surgical stimulation in dynamic fashion with noxious stimuli sensitizing central neural structures involved in pain perception. Neuroplasticity (altered central nervous system sensory processing) is responsible for changes in pain perception after nociceptive stimulation (4). A prolonged duration of surgical stimulation can produce central sensitization and increases patient sensitivity to noxious stimuli over an expanded receptive field (i.e., hyperalgesia). In addition, peripheral nerve sensitization may also influence postoperative pain perception. In this case, prolonged tissue retraction can sensitize nociceptors, resulting in hyperalgesia at the site of injury as well as in surrounding nontraumatized tissue. Chemical mediators from damaged tissue, such as leukotrienes, bradykinins, histamine, and arachidonic acid metabolites, and increased activity of the sympathetic nervous system are potential modulators for this phenomenon (5).
In this study, PCEA infusions were activated intraoperatively after surgical incision in all patients in both groups. In the RLDH group, all patients underwent acute normovolemic hemodilution after induction of general anesthesia. Once finished, the PCEA basal rate infusions were started. In the MHRT group, all PCEA infusions were begun once the major portion of the hepatic resection was completed. Further prospective study is needed to examine whether application of the somewhat controversial concept of preemptive analgesia through administration of local anesthetic with or without opioid into the epidural space before skin incision would indeed decrease the effects of neuromodulation and plasticity on postoperative pain (6,7) and improve postoperative pain control. As demonstrated by Katz et al. (6) preincisional administration of a local anesthetic with fentanyl into the epidural space decreased postoperative morphine PCA consumption as well as postoperative hyperalgesia. The effect on postoperative pain was evident with standardized movement pain scores with preemptive administration of epidural local anesthetic decreasing the pain score. However, the authors stressed that psychological and mental health factors may also play a significant role in postoperative pain perception. In contrast, only pain scores at rest were evaluated and analyzed in this retrospective study.
Despite similar surgical techniques between patient groups, the peak VAPS scores occurred at 24 and 42 hours postoperatively in the RLDH and MHRT groups, respectively. Wilder-Smith et al. (8) demonstrated a biphasic nature of postoperative neuroplasticity. In the first 24 hours postoperatively, the predominant neuroplastic response was inhibition, whereas in the later postoperative period an excitation response with a reduction in pain threshold was observed. Interestingly, the authors found a decreased pain threshold at sites not directly affected by surgery, suggesting supraspinal neuroplasticity. In the present study, RLDH group patients experienced peak VAPS scores earlier in the postoperative course, indicating a possible greater influence from supraspinal excitory neuroplastic modulation.
All donor patients were healthy individuals without previous exposure to major surgical procedures. The pain tolerance in a patient group such as this may be less than in those patients previously exposed to procedures, such as patients with hepatic tumors. In addition, the donors all underwent RLDH for altruistic reasons without any medical indications. After completion of surgery and on awakening, the altruistic motivation may be suppressed as a result of the common sequelae of having undergone an extensive surgical procedure. Thus, donors were not benefiting medically from RLDH, but were, on the contrary, being temporarily debilitated as a result of part of a healthy organ being removed. This may make these patients more vulnerable to pain exaggeration. Diaz et al. (9) reported postoperative pain as being more than anticipated in 44% of donor patients despite the universal use of epidural analgesia and PCA. The same authors also made the interesting observation that female donors with a surgical history had more postoperative pain complaints than did male donors with a surgical history (9). Psychosocial variables including extent of social support, mental health status, degree of optimism, coping style, and mood have all been shown to influence the pain experience (6) and may have influenced the results of this investigation. Thus, the conscious pain experience is the result of chain processing occurring in the central nervous system, making the initial nociceptive event one of many contributing elements in a multifactorial pain experience (8). Unfortunately, because of the retrospective nature of this study, we were not able to obtain this additional information in the preoperative period.
We were surprised to find that patient postoperative VAPS scores had only a weak correlation with either the number of PCEA doses requested or delivered in either group. We believe this observation can be at least partially explained as a result of inadequate preoperative patient education regarding the use of PCEA catheter infusions. In addition, patients in both groups received similar postoperative nursing care. This included similar nursing competencies in assessing postoperative pain in patients with PCEA infusion catheters, as well as similar nursing staff to patient ratios in the PACU, SICU, and on the ward for both patient study groups. Thus, a difference in postoperative nursing care is an unlikely reason for the differences in postoperative pain observed.
This study has several other limitations to consider when designing possible future investigations to explore this topic. Because of the retrospective nature of the data collection, several variables, such as the initiation, activation, and anesthetic solution to be administered via PCEA infusion, were not controlled for. Also, no protocol was used for rescue dosing of epidural catheters if inadequate pain control was encountered. In several instances, patients received intermittent “clinician doses” of epidural anesthetic solution in addition to the preprogrammed patient-activated boluses. However, only minor adjustments were made to either the PCEA infusion catheter demand doses or basal infusion rates in those patients. In addition, the possible under-use of supplemental IV opioids should also be addressed in any future study, as well as more frequent (less than every 6 hours) postoperative pain assessments that may prompt more expeditious therapeutic intervention. The use of epidural catheters in hepatic surgery is controversial. Yong et al. (10) found no significant differences in living donor satisfaction regarding postoperative pain control in those receiving either IV PCA opioids or epidural analgesia. Also, the authors cautioned that postoperative coagulopathy after partial hepatectomy may interfere with the safe removal of the epidural catheter. However, no complications were reported in this particular study. In contrast, in a the review of 22 donor left lateral hepatectomies by Choudhry et al. (11) the authors observed superior postoperative pain relief with thoracic epidural analgesia without any reported complications. Borromeo et al. (12) described the observed postoperative perturbation of serum coagulation tests in patients after right hepatectomy, including increases in both prothrombin and activated partial thromboplastin times, with a concomitant decrease in the serum platelet count. In our experience, the resultant abnormal coagulation profile related to RLDH or MHRT typically normalizes 2–3 days after surgery and has not affected the management of any PCEA infusion catheter. However, postoperative vigilance after major hepatic surgery is paramount with close neurologic monitoring warranted.
In this study, RLDH patients experienced significantly more postoperative pain as compared retrospectively with patients who underwent MHRT. As a result of the longer surgical duration for donor hepatectomy, neuroplasticity may play a role in exaggerated postoperative pain perception along with various psychological factors, although this is unknown at this time. In addition, further prospective studies are needed to determine if the application of preemptive analgesia in these patients may offer some advantage for postoperative pain control. Although advocates for the use of PCEA infusion catheters in these patients, we believe these individuals require more extensive preoperative teaching on the use of these devices. In addition, the concept of PCEA infusion catheter dosing should be continually reinforced by those directly caring for these patients. Despite several recognized limitations of this study, we believe that the observed increased pain after RLDH, as compared with that seen after MHRT, is a real and likely multifactorial phenomenon that requires further prospective investigation.
1. Strong RW, Lynch SV, Ong TH, et al. Successful liver transplantation from a living donor to her son. N Engl J Med 1990;322:1505–7.
2. Trotter JF, Wachs M, Everson GT, Kam I. Adult-to-adult transplantation of the right hepatic lobe from a living donor. N Engl J Med 2002;346:1074–82.
3. Trotter JF, Talamantes M, McClure M, et al. Right hepatic lobe donation for living donor liver transplantation: impact on donor quality of life. Liver Transpl 2001;7:485–93.
4. Melzack R, Coderre TJ, Katz J, Vaccarino AL. Central neuroplasticity and pathological pain. Ann N Y Acad Sci 2001;933:157–74.
5. Bush DJ. Pre-emptive analgesia. BMJ 1993;306:285–6.
6. Katz J, Cohen L, Schmid R, et al. Postoperative morphine use and hyperalgesia are reduced by preoperative but not intraoperative epidural analgesia: implications for preemptive analgesia and the prevention of central sensitization. Anesthesiology 2003;98:1449–60.
7. Rockemann MG, Seeling W, Bischof C, et al. Prophylactic use of epidural mepivacaine/morphine, systemic diclofenac, and metamizole reduces postoperative morphine consumption after major abdominal surgery. Anesthesiology 1996;84:1027–34.
8. Wilder-Smith OH, Tassonyi E, Crul BJ, Arendt-Nielsen L. Quantitative sensory testing and human surgery: effects of analgesic management on postoperative neuroplasticity. Anesthesiology 2003;98:1214–22.
9. Diaz GC, Renz JF, Mudge C, et al. Donor health assessment after living-donor liver transplantation. Ann Surg 2002;236:120–6.
10. Yong BH, Tsui SL, Leung CC, et al. Management of postoperative analgesia in living liver donors. Transplant Proc 2000;32:2110.
11. Choudhry DK, Schwartz RE, Stayer SA, et al. Anesthetic management of living liver donors. Can J Anaesth 1999;46:788–91.
12. Borromeo CJ, Stix MS, Lally A, Pomfret EA. Epidural catheter and increased prothrombin time after right lobe hepatectomy for living donor transplantation. Anesth Analg 2000;91:1139–41.