Search tips
Search criteria 


Logo of medicineHomeSearchSubmit a ManuscriptMedicine
Medicine (Baltimore). 2017 April; 96(15): e6624.
Published online 2017 April 14. doi:  10.1097/MD.0000000000006624
PMCID: PMC5403110

Incidence of persistent postoperative pain after hepatectomies with 2 regimes of perioperative analgesia containing ketamine

Paula Masgoret, MD,a,* Carmen Gomar, MD, PhD,a Beatriz Tena, MD, PhD,a Pilar Taurá, MD,a José Ríos, MSc,b and Miquel Coca, MDa
Monitoring Editor: Maria Kapritsou.


Studies designed to assess persistent postoperative pain (PPP) incidence after hepatectomies are lacking. Our aim was to assess PPP incidence 6 months after hepatectomies with intravenous (IV) or epidural (EPI) analgesia containing ketamine.

Prospective observational comparative study between 2 cohorts of patients submitted to hepatectomy. Patients received 1 of 2 analgesic regimes containing ketamine: EPI group or IV group. Visual analog scale (VAS), Neuropathic Pain Symptom Inventory (NPSI), Pain Catastrophizing Scale (PCS), and quantitative sensorial testing (QST: to determine area of hyperalgesia/allodynia) were assessed preoperatively and postoperatively at 2 h, 24 h, 7 days, 1 month, and 6 months. VAS ≥ 1 at 1 and 6 months was considered indicative of PPP and VAS > 3 was considered as not controlled pain. Side effects and complications were registered.

Forty-four patients were included: 23 in EPI group and 21 in IV group. Patients in IV group were older and had more comorbidities. No patient presented VAS > 3 at 1 or 6 months. VAS ≥ 1 at 1 and 6 months was 36.4% and 22.7%, respectively. No differences in VAS, NPSI, or PCS were found between groups. Allodynia/hyperalgesia area did not differ between groups and was infrequent and slight. Pain pressure threshold in the wound vertical component was significantly higher in EPI group after 7 days. IV group showed more cognitive side effects.

Incidence of PPP at 6 months after open hepatectomies with EPI or IV analgesia containing ketamine was lower than previously reported for other abdominal surgeries.

Ketamine influence on low PPP incidence and hyperalgesia cannot be discarded.

Keywords: hepatectomy, ketamine, neuropathic pain, persistent postoperative pain, quantitative sensorial testing

1. Introduction

Incidence of persistent postoperative pain (PPP) after abdominal surgery has been reported between 0% and 34%.[1,2] Studies specifically designed to evaluate the incidence of PPP after open hepatectomies have not been published to date. In 2 studies on quality of life after hepatic surgery, the incidence of PPP was reported between 9% and 50%.[35] The main component of PPP is neuropathic and the central sensitization after nervous injury generates secondary hyperalgesia that could intervene not only in the intensity of the perception of immediate postsurgical pain but also in chronification of pain after healing of the wound.[68] Nervous injury and intensity of acute postoperative pain are the most constant predisposing factors after abdominal surgery.[9] Relation of PPP with uncontrolled postoperative pain makes postsurgical analgesia the most accessible way to prevent PPP.

Secondary hyperalgesia has been related to PPP.[5] In conventional hepatic resection, the predominant surgical incision has a vertical supraumbilical medial component and a horizontal subcostal component. Thoracic epidural (EPI) analgesia has been considered the most effective analgesic technique for hepatic resection surgery in multiple series,[1013] although the frequent pre- and postsurgical coagulation alterations in these patients make its indication controversial.[14,15] The doses of opioids[16,17] and nonsteroidal anti-inflammatory drugs[18] for intravenous (IV) analgesia are limited in patients with hepatic disease, reason for which low-dose ketamine is frequently associated.[19]

N Metil D Aspartate receptors (NMDAr) have a key role in central sensitization causing PPP, which is manifested as secondary hyperalgesia. Ketamine is one of the few clinically available NMDAr antagonists and it is widely used in a complimentary way for postsurgical analgesia.[1923] There is clear evidence that ketamine decreases intensity of postsurgical acute pain,[19,21,23] but results regarding its effect in PPP prevention are inconsistent, although it is proven to reduce the secondary hyperalgesia area.[2426] The possibility to reach higher concentrations closer to the NMDAr activation areas in the spinal cord by its EPI administration is being studied.[24,27,28]

Great variability about PPP incidence and its response to treatments may be attributed to subjectivity in the evaluation of pain, lack of long-term follow-up, and heterogeneity in applied analgesic protocols. Our research group has designed a simplified protocol for PPP assessment that combines subjective scales with quantitative sensory tests (QST), to detect intensity, extension, and evolution of secondary hyperalgesia.[24,29]

The aim of this study was to determine PPP incidence during the first 6 months after open hepatectomies and its relationship with the hyperalgesia area and with the analgesic techniques most frequently used in our institution.

2. Patients and method

2.1. Patients

The study was approved by the Research Ethics Board of our institution as an observational comparative study between 2 cohorts registered as HCB/2016/0690.

Patients over 18 years of age scheduled for conventional hepatic resection surgery according to Brisbane terminology[30] in our institution during 18 months were considered for inclusion. Major resection was defined as the 1 developed on 3 or more adjacent segments. Inclusion was confirmed the day before surgery after obtaining written consent.

Exclusion criteria were patient refusal, previous chronic pain or chronic analgesic treatment, history of drug addiction or neurological/psychiatric disorder, cirrhosis, contraindication to any drug going to be used, emergency surgery, previous upper abdominal surgery, anticipation of nonresectable disease because of extension, and prevision of nonimmediate extubation after surgery.

The day before surgery, patients were instructed about how to refer pain using the visual analog scale (VAS 0–10) and the use of an electronic patient controlled analgesia (PCA) pump. Neuropathic Pain Symptoms Inventory (NPSI),[24,31] Pain Catastrophizing Scale (PCS),[24,32] and QST (Von Frey monofilaments [VFM], electronic Von Frey [EVF], and electric toothbrush) were explained and applied as baseline measurements. Patients were instructed to report a “painful” sensation with QST when a “change in perceived stimulation was disturbing enough to make him/her wish to stop it.”

Patients were classified into 2 groups according to the type of analgesia they received: EPI group or IV group. The decision on which type of analgesia each patient would receive was made by the responsible anesthesiologist (PT) based on general criteria of contraindications for both techniques.

According to the criteria previously used by our group,[24,26] values of VAS ≥ 1 at 1 and 6 months were considered indicative of PPP and VAS > 3 was considered as not controlled pain. Any positive change in QST with respect to the preoperative baseline examination was considered as allodynia/hyperalgesia. Lack of sensitivity with all the QST was considered as anesthesia.

2.2. Anesthesia and surgical procedures

Sublingual diazepam 5 to 10 mg was administered the night before surgery and IV midazolam 0.025 mg kg−1 was given at the arrival to the operating room. In the EPI group, an EPI catheter was inserted at the vertebral interspace T7–8 before anesthesia induction. Monitoring included electrocardiogram, pulse oxymetry, capnography, invasive arterial pressure, central venous pressure, esophageal temperature, urine output, and bispectral index. Induction of anesthesia consisted of fentanyl (2–3 μg kg−1), propofol (1 mg kg−1), and cisatracurium (0.15 mg kg−1). Anesthesia was maintained with desflurane 4% to 6% in O2/air 60% (to keep BIS between 40 and 50) and cisatracurium 0.1 mg kg−1 h−1. After induction the analgesic technique was set up as foreseen, EPI group or IV group. Patients were extubated in the operating room when awake, following residual muscle block reversal with neostigmine. Patients remained monitored in the postanesthetic care unit for 24 h.

2.3. Analgesia

Analgesic regimen for both groups is described in detail in Table Table11.

Table 1

Analgesic regimen administered in both groups.

2.3.1. EPI group

Before surgical incision, an EPI bolus of lidocaine + morphine was administered. Twenty minutes before skin closure, patients received an IV bolus of morphine and the EPI PCA pump was started and kept for 5 days. PCA pump was programmed to administer 5 mL h−1, with possibility of 2 boluses of 3 mL in 1 h and lockout interval of 20 min.

2.3.2. IV group

Before surgical incision, an IV bolus of ketamine was administered and an IV infusion of morphine was started. Twenty minutes before skin closure, patients received an IV bolus of morphine and the IV PCA pump was started and kept for 5 days. PCA pump was programmed to administer 1 mL h−1, with possibility of 3 boluses of 0.5 mL in 1 h and lockout interval of 5 min.

In addition, both groups received IV acetaminophen 1 g every 8 h and subcutaneous methadone 4 mg if VAS > 3. After PCA stop (either EPI or IV) on day 5 postsurgery, acetaminophen and methadone were continued.

2.4. Outcome assessment

2.4.1. Subjective scales

VAS was evaluated the day before surgery and postoperatively at 2 h, 24 h, 7 days, 1 month, and 6 months. Patients also completed NPSI,[31] which allows identification and qualification of neuropathic symptoms reflecting spontaneous or paroxysmal pain and evoked pain, that was considered positive with scores higher than zero. PCS was also filled,[32] which measures presence of distress reactions to nociceptive stimulation. These 2 last tests were evaluated the day before surgery and at 1 and 6 months after surgery. NPSI was also completed on day 7 after surgery.

2.4.2. Quantitative sensory testing

All tests were performed the day before surgery (baseline) and at 7 days, 1 month, and 6 months after surgery by the same expert investigator in this evaluation (BT). All mechanical measurements were carried out around the surgical incision (horizontal and vertical component) and at the middle third of the contralateral thigh as a control skin zone.

  • – Von Frey monofilaments (VFM): Area of allodynia/hyperalgesia for punctuate mechanical stimuli around the surgical incision was tested using nylon monofilaments (Bioseb, Vitrolles, France) of constant length and increasing diameter which exert escalating pressure proportional to their diameter. In accordance with the method described by Stubhaug,[33,34] consecutive stimulation was applied with 3 monofilaments of 0.6, 2, and 10 g. The day before surgery, QST were applied in the area where the incision would be made and in the thigh. After the surgery, VFM were applied starting 10 cm away from the wound where no pain sensation was experienced, moving toward the incision in different radial axis until the patient reported a change in perception. In each axis, the first point where a “painful” or “sharper” sensation occurred was marked and the distance to the wound was measured. If no change in sensation occurred, the stimulation stopped at 0.5 cm from the incision. The area of allodynia/hyperalgesia was determined as a rectangle whose long axis was the wound length plus the distance of allodynia/hyperalgesia if present; and the short axis the distance of allodynia/hyperalgesia above and under the incision in the horizontal segment and at each side of it in the vertical component.[2,24,35] VFM were applied at vertical and horizontal components of the wound and both areas were added without repeating the overlapping zone. Figure Figure11 shows the area explored.
    Figure 1

    Von Frey monofilaments and electronic Von Frey measurements.

  • – EVF: Pressure pain threshold was tested using an EVF device (Electronic Von Frey Anesthesiometer IITC, Woodland Hills, USA)[29,36] with a constant slope of increasing pressure up to the detection of mechanical pain threshold. Threshold was defined as the lowest pressure that produced pain sensation. Measurements were performed at 6, 4, and 2 cm from the surgical incision, above and under the wound in the horizontal segment and at each side of it at the vertical component (Fig. (Fig.1).1). Each stimulus was applied 3 times and the average from the 6 measurements (3 at each side) for each segment was registered.
  • – Brush-evoked allodynia was explored using an electrical toothbrush applied to the skin area as described for VFM.
  • – Anesthesia area was calculated as described for VFM.

QST were applied in the anterior middle third of the left thigh at all times.

2.5. Side effects

Side effects were recorded during the 1st postsurgical week. Complications such as postoperative nausea/vomiting (PONV), hemodynamic side effects (new onset arrhythmia or 20% variation on median arterial pressure compared with the basal values), cognitive side effects (hallucinations, acute confusional syndrome, nightmares), visual side effects (diplopia), and sedation (Ramsay ≥ 2),[37] hospital length of stay and need for vasoactive drugs or transfusion were registered. They were considered severe if they required treatment or withdrawal of analgesic regimen.

Postsurgical complications were registered according to Dindo-Clavien Classification.[38] Analgesia requirement after discharge was also recorded.

2.6. Statistical analysis

This was an observational comparative study between 2 cohorts with the aim to describe several subjective pain scales and QST evolution in patients scheduled for major or minor conventional hepatic resection. Therefore, presence of PPP was not evaluated using just a single variable but a combination of several ones.

It was decided to recruit all patients who fulfilled inclusion criteria during 18 months. Forty-four patients were included and it was considered that this sample size would be representative of the population undergoing hepatic resection in a third-level hospital like ours. Comparisons between both groups for variables with a single measurement such as demographical data, preoperative and intrasurgical variables, side effects, PONV incidence, hospital length of stay, and vasoactive or transfusion needs were analyzed by means Mann–Whitney U and data are reported as median [interquartile range: 25 and 75 percentiles] and absolute range if they were continuous variables. For qualitative variables Fisher exact test was applied and data are reported as absolute and relative frequencies. Inferential analyses in longitudinal results were conducted using general estimated equations; these were applied using an unstructured matrix to account for within-subject correlations and an appropriated distribution for each dependent variable. These models were made to include groups, time, and their interaction, and the aim was to compare time-by-time VAS, analgesia requirement after discharge, NPSI, PCS, and QST evaluation between groups. Results of these variables are shown as mean (standard deviation) or absolute and relative frequencies for each time. Data were analyzed with statistical analysis software (SPSS version 20; IBM Corp., Armonk, NY) and the level of significance was set at the 2-sided 5% level.

3. Results

During the inclusion period, 99 patients underwent elective hepatic resection in our institution; 44 patients who met the inclusion criteria were analyzed: 23 in the EPI group and 21 in the IV group (Fig. (Fig.2).2). All patients completed follow-up (6 months).

Figure 2

Patients flow chart.

In the demographic and prognostic characteristics, the only significant differences found were older age and more frequent American Society of Anesthesiologists Physical Status Classification System (ASA) III–IV class in the IV group (Table (Table22).

Table 2

Patients’ characteristics.

3.1. Subjective scales

Table Table33 shows VAS values during the study in the whole sample and in each separate group. During the 1st postoperative week, mean VAS values remained below 3 and at 1 and 6 months below 1 without differences between groups. In the 1st week, the number of patients with VAS > 3 (not controlled pain) was low and similar between groups. No patient suffered VAS > 3 at 1 or 6 months. Patients with VAS ≥ 1 at 1 month were 36.4%, and at 6 months 22.7% without significant differences between groups.

Table 3

VAS mean values and patients presenting VAS ≥ 1 and VAS > 3 throughout the study in both groups and in the whole sample.

Means of NPSI and PCS and frequencies of values >0 in the 2 groups and times are shown in Table Table4.4. No significant differences were observed between groups. Total frequency of patients which still required analgesia at 1 month was 38.6%, and 13.6% at 6 months without significant differences between groups.

Table 4

Mean values and frequencies of scores >0 of NPSI and PCS throughout the study in both groups.

3.2. QST

At baseline exploration, no patients showed positive responses with VFM or toothbrush in the thigh or in the expected incision area. Pain threshold to pressure measured with EVF before surgery in the thigh or surgical area did not differ between groups (Table (Table2).2). No patient showed changes in QST in the thigh at any time of the study.

Regarding postsurgical allodynia/mechanical hyperalgesia in the surgical incision area, no patient showed positive responses with 0.6 or 2 g VFM at any time of the study. With 10 g VFM, small and infrequent areas of allodynia/mechanical hyperalgesia were found at 7 days and 1 month without statistical differences between groups. At 6 months, no patient showed positive responses with any of the VFM. Regarding anesthesia no statistical significant differences were found between groups (Table (Table5).5). Brushed evoked allodynia was not found in any patient during the study.

Table 5

Means of areas and frequencies of results >0 of allodynia/mechanical hyperalgesia with VFM of 10 g at 7 days and 1 month and of anesthesia areas at 7 days, 1 month, and 6 months.

Pain thresholds measured with EVF are shown in Table Table6.6. No significant changes were observed between baseline and postsurgical values at any time. There were no significant differences between groups in the evolution of the horizontal component of the incision. In the vertical segment of the wound, pain thresholds were higher in the EPI group compared with the IV group at all times with significant differences at 2 cm from the incision at day 7 and 6 months and at 6 cm from the wound in the 1st month.

Table 6

Pain thresholds with EVF at 6, 4, and 2 cm from the surgical wound.

3.3. Side effects and postsurgical complications

Table Table77 shows incidence of postsurgical side effects, needs for analgesia rescue, and length of hospital stay in the whole general sample and in the 2 separate groups. Twenty-five percent of patients did not suffer postsurgical complications according to Dindo-Clavien criteria; 25% presented type I complications, 34.1% type II, 13.6% type III, and 2.3% type IV without significant differences between groups.

Table 7

Postsurgical complications, side effects incidence, needs for analgesia rescue, and length of hospital stay.

No hemodynamic alterations were detected. No significant differences were observed between groups regarding sedation or visual alterations. No patient in the EPI group showed cognitive alterations while in the IV group, 19% (4 patients) presented cognitive alterations (P = .044). Out of 4, 3 patients in the IV group who suffered cognitive side effects, presented them concomitantly with postsurgical complications type II or of Dindo-Clavien. None of the side effects was considered severe and withdrawal of analgesic treatment was not needed in any case.

PONV had a global incidence of 13.6% and need for analgesic recue with methadone 15.9%, without differences between groups. Median hospital length of stay was 10 days in both groups. There were no significant differences regarding need for vasoactive drugs or transfusion.

4. Discussion

The aim of this study was to determine PPP incidence after 6 months of open hepatectomies, given the absence of prospective data in the reviewed literature. The study was designed in order to evaluate PPP incidence according to clinical manifestations and neurosensitive changes with 2 different postsurgical analgesic regimes commonly prescribed in our institution: EPI or IV analgesia, both containing ketamine.[19] Results showed a global clinical PPP incidence of 22.7% (considering it as VAS ≥ 1) at 6 months with a median VAS < 0.33 and analgesic requirement at this time in 13.6% of cases. No patient showed not-controlled pain (considered as VAS > 3) at 6 months. There were no differences in PPP incidence assessed neither by VAS nor by NPSI or PCS between patients with EPI and IV analgesia. Allodynia/secondary hyperalgesia area assessed by QST was found in few patients and, when present, it was small sized. Despite the fact that the study has limitations that will be discussed later, it provides new information.

Pain after conventional hepatic resection is considered of high intensity and it is usually treated with potent analgesic techniques such as EPI. Nevertheless, PPP research after this specific type of surgery is limited. Available information about PPP after major abdominal surgery is in general controversial and PPP following hepatectomy has not been specifically addressed.

A retrospective control–case study performed on 101 patients who underwent different types of abdominal surgeries, which compared combination of epidural technique and general anesthesia with general anesthesia alone found a PPP incidence (measured thorough SF-36 questionnaire[1,39]) at 6 months of 25.7%. In 85 patients who had rectal adenocarcinoma surgery performed with different analgesia regimes, all of them including ketamine, a prospective study with telephonic control of VAS after hospital discharge reported a PPP incidence between 0% and 48% after a year.[2] On the other hand, in a prospective series on 20 patients intervened from gastrointestinal or urinary surgery, PPP (considered as VAS > 0 in movement or rest) incidence was 30%.[15]

Available data on PPP after hepatic resection surgery are scarce. There is a retrospective study on 65 liver living donors in which almost all patients received IV ketamine in antihyperalgesic doses at the anesthesia induction. It showed a PPP incidence of 9.23% assessed 3 months after surgery without specifying VAS level.[5] There are also 2 studies designed to evaluate quality of life in which PPP is indirectly evaluated. One prospective study on 21 liver living donors that describes an incidence of “wound-related symptoms” of 24% using the SF-36 form[40] and a questionnaire[4] and another on 96 patients who underwent hepatic resection and who received and returned the SF-12 formulaire[41] and a VAS by email 3 to 36 months after surgery. This last one reported a “50% of patients pain-free.”[3] However, none of these studies specified VAS level.

Data from published studies make difficult to get clear information on this topic and to compare our results with others. Previous studies show disparities regarding analgesic regimes with or without ketamine, divergence and simplicity in PPP assessment (only by subjective scales: VAS, SF-36, and SF-12 formulaires and other questionnaires administered by email), little information about PPP definition, and lack of long-term prospective follow-up.

PPP is a chronic kind of pain with neuropathic characteristics and neurosensorial alterations that range between anesthesia and allodynia/hyperalgesia. These changes are expressed around the wound (secondary hyperalgesia area), are related with neuroplasticity, and could be related, according to some authors, with PPP.[2] Assessment of neurosensorial changes before and after surgical operation should be part of a PPP evaluation together with subjective scales in order to get information about neuropathic changes generated after surgery. Nevertheless, the conventional complete neurosensitive assessment is not applicable to surgical patient or to repeated explorations. An assessment protocol previously applied by our investigation group[24,29] has been used in this study. This protocol evaluates patients from the presurgical time until 6 months after surgery combining subjective scales with semiobjective assessment through QST. It includes VAS measuring (defining presence of clinical pain at 6 months as VAS ≥ 1 and uncontrolled pain as VAS > 3), PCS, NPSI, analgesic requirement after discharge, and secondary hyperalgesia area exploration through QST in order to detect changes. Applying this standard PPP exploration procedure, assessed in a prospective and extensive way and with a longer follow-up period than other series, our study provides information on PPP incidence after hepatectomies not available in the existing literature. This protocol of PPP assessment is easy to apply repeatedly in postsurgical patients and permits comparison of data after surgery with that before it. This makes possible to follow-up pain characteristics from its beginning until its eventual chronification, allowing to correlate the intensity of acute pain with QST findings and PPP development.[5]

Our results suggest that PPP incidence (considering it as VAS ≥ 1 in context of other subjective scales and QST results) after 6 months of conventional hepatic resection with usual analgesic regimes containing ketamine is in the low range, 22.7% and that information was not reported before.

Available data about the influence of different postoperative analgesic techniques in abdominal surgery regarding PPP incidence are contradictory.[1,2] The advantages of EPI analgesia after hepatectomy regarding quality of analgesia, hospital length of stay, and postsurgical complications are not clear and it is associated with an increase in transfusion and fluids requirements.[1012,42,43] In our study, no differences in analgesic efficacy, PPP, or other perioperative variables were found in patients receiving IV or EPI analgesia regimens, both containing ketamine. Therefore, the EPI route does not seem to offer advantages in these patients who also have frequent contraindications for EPI catheter insertion.[1,10,14,44] The only difference between both analgesic techniques used was a higher incidence of cognitive alterations in the IV group. According to one previous study which used the same doses of ketamine and published by our group,[31] plasmatic levels of ketamine were similar with IV or EPI administration, therefore we attribute cognitive alterations found in IV group to the older age and more frequent comorbidities of the patients. The fact that the election of EPI or IV analgesic technique relied upon the single treating anesthesiologist could have biased the distribution of patients. In older patients with more comorbidities, IV analgesia would have been considered safer than the EPI technique. This is a study limitation because it does not allow to compare PPP incidence between groups. However, this does not invalidate the low global incidence of PPP and secondary hyperalgesia after conventional hepatic resection surgery found in this study.

The importance of secondary hyperalgesia area in PPP prediction and its possible reduction with ketamine is a topic of great interest because it would allow an early detection of patients that could more probably develop PPP and to act upon it. Although the secondary hyperalgesia area that we have found in our study was small and in few patients, we cannot discard that ketamine could have had an influence in the low incidence of PPP. Up to now, it has not been reliably proved that ketamine reduces PPP incidence, despite the fact that it diminishes the opioid requirements in postoperative period[20,23] and the size of secondary hyperalgesia area.[2] In the literature, doses and routes of administration of ketamine as antihyperalgesic in the perioperative period are very variable and generally given during less time than in our study, in which all patients received it, either IV or EPI, during surgery and for the first 5 postsurgical days. The doses administered in the present study have demonstrated to achieve plasmatic levels within the recommended antihyperalgesic range.[24]

The exploration of the secondary hyperalgesia area with QST has been made in different surgical incisions, but not after hepatectomies.[2] Conventional hepatic resection wound is different to lineal incisions like those made for thoracotomies or other laparotomies[2,31] because it has an “L” shape with a vertical and a horizontal components. The hyperalgesia area that has to be evaluated should include the 2 components, which coincide in the angle zone, and although we have not found references to this type of calculation, it is considered the best way to do it (Fig. (Fig.11).

Among the QST applied in our study, we considered that small size VFM would explore mechanical allodynia (Aβ fibers). The thicker VFM of 10 g in this case could reach to explore Aδ and C fibers that would correspond to mechanical hyperalgesia, although this threshold has not been clearly established. The only response suggestive of secondary hyperalgesia found, was with 10 g VFM, but it was discrete and infrequent. We cannot discard that if we had applied thicker VFM or a “pin prick” test, more marked changes would have been detected. This simplified QST exploration protocol used could have been a study limitation and after discussing this point we have included in our protocol thicker VFM as well as temperature exploration which would give information about changes in Aδ and C fibers and would not complicate the exploration too much.

Regarding pain threshold with EFV, changes found throughout the study have been also very discrete; however, an interesting finding was that the vertical component, but not the horizontal one showed differences between groups. In the EPI group, EFV threshold was higher compared with the IV one. In spite of this, we have not found data in the literature about QST in complex wounds like that of hepatectomy. We speculate that perhaps because the vertical component of the incision injures more nerves with the corresponding increase in the neuropathic component,[45,46] the EPI block containing ketamine could be more effective to prevent secondary hyperalgesia. These differences were not evident with any of the subjective scales.

The lack of correlation between QST and subjective pain expression is frequently found in the literature,[24,33,34,47] which has made that some authors doubt about the relationship between the hyperalgesia area and PPP.[39,45] When this study was designed, we expected both PPP and QST changes to be higher, but their low incidence did not allow us to assess their relationship. The small size of the sample is a limitation of this study. Nevertheless, even though our hospital is a reference center for hepatic and transplantation surgery, to recruit our sample took a long time.

In PPP development, the concept of “diffuse noxious inhibitory control” (DNIC) is gaining more importance.[8] For this reason, we included in our protocol the assessment of a distant area from the wound as the thigh. Although we did not find any changes in this area, we believe that it is worth considering because it is a very simple exploration that allows discarding central influences.

In conclusion, PPP incidence after open hepatectomies with EPI or IV analgesia containing ketamine was 22% at 6 months after surgery. Secondary hyperalgesia responses were infrequent and small sized. Immediate clinical application of our results is that EPI analgesia does not offer advantages in front of IV analgesia in these particular patients with potential coagulation alterations. On the other hand, it seems advisable to include ketamine in the analgesic regimes, considering the low incidence of adverse effects and PPP found in this study. Regarding research in PPP, the various responses to the QST found in the vertical component of the incisions demonstrate the importance of establishing exploration protocols in different types of wounds.


The authors would like to thank all the hepatic surgery and anesthesiology team of the Hospital Clinic Barcelona for their help with data acquisition. The authors also thank the nurses involved in patient care for their dedication and collaboration in the study.


Abbreviations: EFV = electronic Von Frey, EPI = epidural, IV = intravenous, NMDAr = N Metil D Aspartate receptors, NPSI = Neuropathic Pain Symptom Inventory, PCA = patient controlled analgesia, PCS = Pain Catastrophizing Scale, PONV = postoperative nausea/vomiting, PPP = persistent postoperative pain, QST = quantitative sensorial testing, VAS = visual analog scale, VFM = Von Frey monofilaments.

The authors have no funding and conflicts of interest to disclose.


[1] Bouman EA, Theunissen M, Bons SA, et al. Reduced incidence of chronic postsurgical pain after epidural analgesia for abdominal surgery. Pain Pract 2013;14:76–85. [PubMed]
[2] Lavand’homme P, De Kock M, Waterloos H. Intraoperative epidural analgesia combined with ketamine provides effective preventive analgesia in patients undergoing major digestive surgery. Anesthesiology 2005;103:813–20. [PubMed]
[3] Bruns H, Krätschmer K, Hinz U, et al. Quality of life after curative liver resection: a single center analysis. World J Gastroenterol 2010;16:2388–95. [PMC free article] [PubMed]
[4] Ishizaki M, Kaibori M, Matsui K, et al. Change in donor quality of life after living donor liver transplantation surgery: a single-institution experience. Transplant Proc 2012;44:344–6. [PubMed]
[5] Bonnet A, Lavand’homme P, France MN, et al. Postoperative pain trajectories to identify risk of chronic postsurgical pain in living donors for liver transplantation. Ann Fr Anesth Reanim 2012;31:945–9. [PubMed]
[6] Woolf CJ, Mannion RJ. Neuropathic pain: aetiology, symptoms, mechanisms, and management. Lancet 1999;353:1959–64. [PubMed]
[7] Lavand’homme P, De Kock M. The use of intraoperative epidural or spinal analgesia modulates postoperative hyperalgesia and reduces residual pain after major abdominal surgery. Acta Anaesthesiol Belg 2006;57:373–9. [PubMed]
[8] Wilder-Smith OH, Schreyer T, Scheffer GJ, et al. Patients with chronic pain after abdominal surgery show less preoperative endogenous pain inhibition and more postoperative hyperalgesia: a pilot study. J Pain Palliat Care Pharmacother 2010;24:119–28. [PubMed]
[9] Kehlet H, Jensen TS, Woolf CJ. Persistent postsurgical pain: risk factors and prevention. Lancet 2006;367:1618–25. [PubMed]
[10] Revie EJ, McKeown DW, Wilson JA, et al. Randomized clinical trial of local infiltration plus patient-controlled opiate analgesia vs. epidural analgesia following liver resection surgery. HPB 2012;14:611–8. [PubMed]
[11] Tzimas P, Prout J, Papadopoulos G, et al. Epidural anaesthesia and analgesia for liver resection. Anaesthesia 2013;68:628–35. [PubMed]
[12] Kelliher L, Jones C, Dickinson M, et al. Epidural anaesthesia and analgesia for liver resection. Anaesthesia 2013;68:975–6. [PubMed]
[13] Clarke H, Chandy T, Srinivas C, et al. Epidural analgesia provides better pain management after live liver donation: a retrospective study. Liver Transplant 2011;17:315–23. [PubMed]
[14] Morisaki H, Doi J, Ochiai R, et al. Epidural hematoma after epidural anesthesia in a patient with hepatic cirrhosis. Anesth Analg 1995;80:1033–5. [PubMed]
[15] Bezeaud A, Denninger MH, Dondero F, et al. Hypercoagulability after partial liver resection. Thromb Haemost 2007;98:1252–6. [PubMed]
[16] Hazell AS, Butterworth RF. Hepatic encephalopathy: an update of pathophysiologic mechanisms. Proc Soc Exp Biol Med 1999;222:99–112. [PubMed]
[17] Mazoit JX, Sandouk P, Zetlaoui P, et al. Pharmacokinetics of unchanged morphine in normal and cirrhotic subjects. Anesth Analg 1987;66:293–8. [PubMed]
[18] Rudnick DA, Perlmutter DH, Muglia LJ. Prostaglandins are required for CREB activation and cellular proliferation during liver regeneration. Proc Natl Acad Sci USA 2001;98:8885–90. [PubMed]
[19] Taurá P, Fuster J, Blasi A, et al. Postoperative pain relief after hepatic resection in cirrhotic patients: the efficacy of a single small dose of ketamine plus morphine epidurally. Anesth Analg 2003;96:475–80. [PubMed]
[20] Bell RF, Dahl JB, Moore R, et al. Perioperative ketamine for acute postoperative pain. Cochrane Database Syst Rev 2015;7:CD004603. [PubMed]
[21] Gottschalk A, Freitag M, Steinacker E, et al. Pre-incisional epidural ropivacaine, sufentanil, clonidine, and (S)+-ketamine does not provide pre-emptive analgesia in patients undergoing major pancreatic surgery. Br J Anaesth 2008;100:36–41. [PubMed]
[22] Silva EP, Sakata RK, Garcia JB, et al. Evaluation of preemptive effect of epidural S(+)-ketamine for hysterectomy: plasmatic concentrations of interleukins. Rev Bras Anestesiol 2012;62:3–9. [PubMed]
[23] Laskowski K, Stirling A, McKay WP, et al. A systematic review of intravenous ketamine for postoperative analgesia. Can J Anaesth 2011;58:911–23. [PubMed]
[24] Tena B, Gomar C, Rios J. Perioperative epidural or intravenous ketamine does not improve the effectiveness of thoracic epidural analgesia for acute and chronic pain after thoracotomy. Clin J Pain 2014;30:490–500. [PubMed]
[25] De Kock M, Lavand’homme P, Waterloos H. “Balanced analgesia” in the perioperative period: is there a place for ketamine? Pain 2001;92:373–80. [PubMed]
[26] Ryu H-G, Lee C-J, Kim Y-T, et al. Preemptive low-dose epidural ketamine for preventing chronic postthoracotomy pain: a prospective, double-blinded, randomized, clinical trial. Clin J Pain 2011;27:304–8. [PubMed]
[27] Elia N, Tramèr MR. Ketamine and postoperative pain—a quantitative systematic review of randomized trials. Pain 2005;113:61–70. [PubMed]
[28] Sethi M, Sethi N, Jain P, et al. Role of epidural ketamine for postoperative analgesia after upper abdominal surgery. Indian J Anaesth 2011;55:141–5. [PMC free article] [PubMed]
[29] Tena B, Escobar B, Arguis MJ, et al. Reproducibility of electronic von Frey and von Frey monofilaments testing. Clin J Pain 2012;28:318–23. [PubMed]
[30] Belgihiti J, Clavien PA, Gadzijev E, et al. The Brisbane 2000 terminology of liver anatomy and resections. HPB 2000;2:333–9.
[31] Bouhassira D, Attal N, Fermanian J, et al. Development and validation of the Neuropathic Pain Symptom Inventory. Pain 2004;108:248–57. [PubMed]
[32] Sullivan MJL, Bishop SR, Pivik J, et al. The Pain Catastrophizing Scale: development and validation. Psychol Assess 1995;7:524–32.
[33] Wilder-Smith OHG, Tassonyi E, Crul BJP, et al. Quantitative sensory testing and human surgery: effects of analgesic management on postoperative neuroplasticity. Anesthesiology 2003;98:1214–22. [PubMed]
[34] Stubhaug A, Breivik H, Eide PK, et al. Mapping of punctuate hyperalgesia around a surgical incision demonstrates that ketamine is a powerful suppressor of central sensitization to pain following surgery. Acta Anaesthesiol Scand 1997;41:1124–32. [PubMed]
[35] Salengros JC, Huybrechts I, Ducart A, et al. Different anesthetic techniques associated with different incidences of chronic post-thoracotomy pain: low-dose remifentanil plus presurgical epidural analgesia is preferable to high-dose remifentanil with postsurgical epidural analgesia. J Cardiothorac Vasc Anesth 2010;24:608–16. [PubMed]
[36] KuKanich B, Lascelles BD, Papich MG. Assessment of a von Frey device for evaluation of the antinociceptive effects of morphine and its application in pharmacodynamic modeling of morphine in dogs. Am J Vet Res 2005;66:1616–22. [PubMed]
[37] Ramsay MA, Savege TM, Simpson BR, et al. Controlled sedation with alphaxalone-alphadolone. Br Med J 1974;2:656–9. [PMC free article] [PubMed]
[38] Dindo D, Demartines N, Clavien PA. Classification of surgical complications. Ann Surg 2004;240:205–13. [PubMed]
[39] Perkins FM, Kehlet H. Chronic pain as an outcome of surgery. A review of predictive factors. Anesthesiology 2000;93:1123–33. [PubMed]
[40] Ware JE, Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30:473–83. [PubMed]
[41] Ware JJ, Kosinski MM, Keller SSD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996;34:220–33. [PubMed]
[42] Page A, Rostad B, Staley CA, et al. Epidural analgesia in hepatic resection. J Am Coll Surg 2008;206:1184–92. [PubMed]
[43] Revie EJ, Massie LJ, McNally SJ, et al. Effectiveness of epidural analgesia following open liver resection. HPB (Oxford) 2011;13:206–11. [PubMed]
[44] Wulf H. Epidural anaesthesia and spinal haematoma. Can J Anaesth 1996;43:1260–71. [PubMed]
[45] Brennan T, Kehlet H. Preventive analgesia to reduce wound hyperalgesia and persistent postsurgical pain: not an easy path. Anesthesiology 2005;103:681–3. [PubMed]
[46] 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. [PubMed]
[47] Dirks J, Møiniche S, Hilsted KL, et al. Mechanisms of postoperative pain: clinical indications for a contribution of central neuronal sensitization. Anesthesiology 2002;97:1591–6. [PubMed]

Articles from Medicine are provided here courtesy of Wolters Kluwer Health