Research Paper
Clinical, histological, and biochemical predictors of postsurgical neuropathic pain ¨ ¸ eylerc, Skander Ben Ammara, Jean-Claude Alvareze, Fabrice Gaudotd, Valeria ´ Martineza,b,*, Nurcan Uc c Claudia Sommer , Didier Bouhassirab, Dominique Fletchera,b
Abstract Surgical nerve injury sometimes leads to chronic postsurgical neuropathic pain (CPSNP). The risk factors for this condition are not well understood. We prospectively assessed 46 patients scheduled for iliac crest bone harvest, 2 days (D2) and 3 months (M3) after surgery, to determine the time course of nerve fiber degeneration and expression of the TNF-a and NGF genes in skin punch biopsies. Mechanical and thermal detection and pain thresholds were evaluated at D2 and M3, by quantitative sensory testing. Skin punch biopsies were also obtained from the thighs ipsilateral and contralateral to iliac crest bone harvest. Intraepidermal nerve fiber density (IENFD) and cutaneous TNF-a and NGF gene expression were analyzed. Forty-five volunteers matched for age, sex, skin color were examined as controls. Chronic postsurgical neuropathic pain was defined as pain in an area of hypesthesia with a positive Douleur Neuropathique 4 questionnaire score. Overall, 73% (N 5 32) of patients developed hypesthesia and 40% (N 5 13) of these patients had developed CPSNP at M3. Quantitative sensory testing results, IENFD, and skin TNF-a and NGF gene expression at D2 and M3 did not differ between patients with and without CPSNP. However, in patients with CPSNP, burning, compression, and pain provoked by brushing were correlated with IENFD at M3, suggesting a possible association between partial nerve lesions and more intense CPSNP, than with total nerve lesion. Furthermore, preoperative pain and opioid use were higher in patients who developed CPSNP than in those without CPSNP. These findings suggest that the predictors of CPSNP development are clinical rather than histological or biochemical. Keywords: Neuropathic pain, Surgery, Nerve lesion, Small nerve fiber degeneration, TNF-a, NGF
1. Introduction 28,36
Most persistent pain after surgery is neuropathic and related to nerve lesions.9 However, surgical nerve injury does not always lead to neuropathic pain,1,26,39 and the relationship between nerve lesion severity and chronic postsurgical neuropathic pain (CPSNP) development remains unclear. We previously investigated the link between nerve lesions and hyperalgesia in the development of CPSNP.37 We used the iliac crest bone harvest (ICBH) model, in which it is possible to explore the territory of nerve lesions at some distance from the surgical site. We reported that 60% of patients had hypesthesia after surgery and 23% of patients had developed neuropathic pain 3 months after surgery. Multivariate analysis identified the area of Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article. a Service d’Anesthesie ´ Reanimation ´ Chirurgicale, Hopital ˆ Raymond-Poincare, ´ APHP, Universite´ de Versailles St-Quentin, Garches, France, b INSERM U-987, Centre d’Evaluation et de Traitement de la Douleur, Hopital ˆ Ambroise Pare, ´ BoulogneBillancourt, France; Universite´ Versailles Saint-Quentin, Versailles, France, c Department of Neurology, University of Wurzburg, ¨ Wurzburg, ¨ Germany, d Service d’orthopedie, ´ Hopital ˆ Raymond-Poincare, ´ AP-HP, Universite´ de Versailles StQuentin, Garches, France, e Service de Biochimie, Hopital ˆ Raymond-Poincare, ´ Garches, France
*Corresponding author. Address: Service d’Anesthesie ´ Reanimation ´ Chirurgicale, Hopital ˆ Raymond-Poincare, ´ AP-HP, Universite´ de Versailles St-Quentin, 104 Boulevard Raymond Poincare´ 92380 Garches, France. Tel.: 33(1)47107622; fax 33(1)47107623. E-mail address: [email protected] (V. Martinez). PAIN 156 (2015) 2390–2398 © 2015 International Association for the Study of Pain http://dx.doi.org/10.1097/j.pain.0000000000000286
2390
·
V. Martinez et al. 156 (2015) 2390–2398
hyperalgesia and the neuropathic nature of early pain as 2 independent predictive factors for CPSNP. No link was found between the area or severity of mechanical hypesthesia and the development of CPSNP.2,21,37 Skin punch biopsy is a validated technique for quantifying epidermal unmyelinated small nerve fiber density with the panaxonal marker protein gene product 9.5 (PGP9.5).16,33 We used the ICBH model because, in this model, skin biopsies can be performed at some distance from the site of potential nerve damage, where fiber degeneration may predominate, as suggested in the sciatic nerve ligature animal model.7,22 Immune activation also plays a crucial role in pain induction and maintenance.14,62 Peripheral nerve injury–induced inflammatory mediators may contribute to the development of abnormalities in the sensory system, such as ectopic activity and enhanced responses to sensory stimuli.14 It has been suggested that inflammation related to Wallerian degeneration can also cause unscathed nerve fiber hyperexcitability.51,52 A number of mediators and cells are thought to be involved, including the proinflammatory cytokine tumor necrosis factor a (TNF-a) and neurotrophic growth factor (NGF), both of which are well-known mediators of neuropathic pain.14,42,47 We investigate the reasons for which only one-third of patients with nerve lesion develop CPSNP, by analyzing the contributions of skin small nerve fiber degeneration and cutaneous expression of the TNF-a and NGF genes to CPSNP. We prospectively assessed patients scheduled for ICBH, from the early postoperative period until 3 months after surgery, to determine the time course of small nerve fiber degeneration and cutaneous expression of the TNF-a and NGF genes. PAIN®
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
November 2015
·
Volume 156
·
Number 11
www.painjournalonline.com
2391
2. Methods 2.1. Patients Approval was obtained from the institutional review board (Comite´ de Protection des Personnes, Boulogne-Billancourt). We prospectively included consecutive patients undergoing ICBH surgery at our institution between January 2010 and September 2012. All patients provided written informed consent. The inclusion criteria were age $18 years and patients scheduled for orthopedic surgery with ICBH. The noninclusion criteria were patient refusal, previous ICBH, ipsilateral hip surgery, previous surgery, or chronic pain in the area of skin biopsy. The exclusion criteria were decision of the patient to withdraw from the study, postoperative complications (hematoma, infection), inability to answer questions, impossibility of examining the patient, and nonstandardized anesthesia and analgesia procedures. 2.2. Anesthesia and postoperative pain relief Patients received 100 mg hydroxyzine before surgery, which was performed under balanced general anesthesia with propofol, sufentanil, atracurium, and sevorane, together with a 1:1 mixture of nitrous oxide and oxygen. Pain immediately after surgery was controlled by 48 hours of intravenous morphine treatment, delivered by patient-controlled analgesia methods, together with 1 g acetaminophen every 6 h and 50 mg ketoprofen every 6 h over the first week. 2.3. Preoperative assessment We collected demographic data, including age, sex, body mass index, and American Society of Anesthesiology score.41 The presence of preoperative chronic pain at the principal site of surgery was recorded, including the intensity of the pain on a numerical rating scale (NRS) from 0 to 10, at rest and on movement, pain duration, and opioid consumption during the last month. 2.4. Postoperative assessment 2.4.1. Pain assessment The intensity, location, and characteristics of ICBH-related pain were assessed on day 2 (D2) after surgery and 3 months after surgery (M3). Pain intensity was evaluated on an NRS. Pain location was determined with a diagram from the Brief Pain Inventory (BPI).15 Chronic postsurgical neuropathic pain was defined as pain in an area of hypesthesia corresponding to the territory of the lateral femoral cutaneous, ilioinguinal, or cluneal nerves, 3 months after ICBH, with a positive Douleur Neuropathique 4 questionnaire score11 (Fig. 1). Patients were asked to report the intensity and location of the pain. The areas of pain, including that of maximal pain, were indicated on the body map from the BPI.15 Current, average, and maximum pain intensity over the last 24 hours were assessed in the area of maximum pain, using an NRS from the BPI.15 At 3 months, patients with neuropathic pain were interviewed with the Neuropathic Pain Symptom Inventory (NPSI), which includes questions relating to 10 symptoms rated on an NRS (burning, squeezing, pressure, electric shocks, stabbing, pain evoked by brushing, by pressure, by cold, tingling, pins, and needles).12 2.4.2. Sensory mapping of postoperative hypesthesia The methodology used has been described elsewhere.37 The area of mechanical hypesthesia was assessed at D2 and M3. We
Figure 1. Localization of the areas explored. (A) Site of surgery for iliac crest bone harvest; (B) territory innervated by the lateral femoral cutaneous nerve; (C) territory innervated by the ilioinguinal nerve; (D) territory innervated by the cuneal nerve. When no area of mechanical hypesthesia was detectable in one of these 3 territories (A, B, or C), assessments (quantitative sensory testing and skin punch biopsies) were performed on the top of the thigh.
first checked for a sensory deficit in the 3 cutaneous territories innervated by the nerves crossing the surgical site (the top of the thigh for the lateral femoral cutaneous nerve, the inguinal area for the iliohypogastric nerve, and the buttocks for the superior cluneal nerves). If a sensory deficit was found, the border of hypesthesia was determined by stimulation along 8 linear paths diverging outwards from the site of tactile hypesthesia, until the patient reported normal perception with a von Frey monofilament (60g; Bioseb, Chaville, France). The first point at which sensation was similar to that on the contralateral side was marked. The 8 demarcating points defined the territory of mechanical hypesthesia. 2.4.3. Quantitative sensory testing Bedside quantitative sensory testing (QST) was performed for the site of mechanical hypesthesia. Quantitative sensory testing was performed the day before surgery, on day 2, and 3 months after surgery. For preoperative evaluation and for patients without a detectable area of mechanical hypesthesia, QST was performed on the top of the thigh, by 2 experienced investigators (V.M. and S.B.) (Fig. 1). Similar tests were performed on healthy volunteers (V.M.). Thermal detection and pain thresholds were assessed with a Somedic Thermotest (Somedic AB, Stockholm, Sweden), using the Marstock method.19 A contact thermode of Peltier elements measuring 25 3 50 mm was applied to the skin. The baseline temperature of the thermode was set at 32˚C. The maximum and minimum temperatures were set at 50˚C for heat and 4˚C for cold pain. A thermal rate of change of 1˚C/s was used. All thresholds were calculated as the mean of 3 successive determinations. Calibrated von Frey monofilaments, 0.007 to 300g (Bioseb), were used to measure detection and pain thresholds in response to mechanical punctate stimuli, according to the method of limits. The tactile detection and pain detection thresholds were defined as the smallest force eliciting a sensation of contact or pain, respectively. The force required to bend the filaments was converted into log units. The contralateral side was used as a control.
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
2392
·
PAIN®
V. Martinez et al. 156 (2015) 2390–2398
Table 1
Preoperative characteristics of the patients and controls. Patient characteristics No. of patients Age, y Sex (men/women) BMI, kg/m2 ASA classification (I, II) Duration of preoperative pain, mo Preoperative pain intensity at rest (NRS) Preoperative pain intensity on movement (EN) Preoperative daily opioid dose (morphine equivalent) Surgery characteristics Upper limb surgery Lower limb surgery Spinal surgery Duration of surgery, min ICBH characteristics Length of the scar, cm Site of crest harvest incision (anterior/posterior) Volume (cortical), cm3
All patients
Patients with hypesthesia
Control
CPSNP
No CPSNP
46 50 (16-80) 26 (56.5%)/20 (43.5%) 25 (16-37) 41 (91%), 4 (9%) 18 (0-200) 1 (0-10) 5 (0-10) 0 (0-60)
13 48 (34-74) 4 (30%)/9 (70%) 25 (18-33) 13 (100%) 48 (0-190) 3 (0-10)* 6.5 (0-10) 10 (0-60)*
19 47 (19-67) 14 (73%)/5 (27%) 24 (16-30) 18 (94%)/1 (6%) 12 (0-120) 0 (0-6) 4.5 (0-9) 3 (0-10)
45 50 (19-79) 25 (55%)/20 (45%) 24 (19-39) NA NA NA NA NA
3 42 1 159 (50-300)
0 13 0 175 (120-230)
2 17 0 150 (60-300)
NA
NA
8.8 (3-15) 40 (87%)/6 (13%) 31 (1-320)
10 (5-13) 11 (84%)/2 (16%) 15 (3-120)
10 (5-15) 18 (94%)/1 (6%) 31 (1.5-320)
NA NA NA
Continuous variables are presented as median (range). Other variables are presented as n (%). * Significant difference between CPSNP and no CPSNP, P , 0.05. ASA, American Society of Anesthesiology Physical Status Score; BMI, body mass index; CPSNP, chronic postsurgical neuropathic pain; ICBH, iliac crest bone harvest; NA, not applicable; NRS, numerical rating scale.
Patients were considered to present hypesthesia when at least 1 of the 3 mean thresholds, heat, cold, or mechanical detection thresholds, differed from preoperative values by at least 2 SDs. 2.4.4. Skin punch biopsies Skin punch biopsies were obtained according to the European recommendations.32 Two days and 3 months after surgery, two 5-mm skin punch biopsies (apparatus from Stiefel GmbH, Offenbach, Germany) were performed under local anesthesia, for each of the patients. One was performed on the leg ipsilateral to ICBH, either in the area of hypesthesia or on the lateral thigh in the territory of the lateral femoral cutaneous nerve. The second biopsy was performed on the contralateral thigh. The control group for skin analysis consisted of 45 healthy volunteers matched for age, sex, and skin color (25 men, 20 women; median age: 52 years [range: 19-70]). All controls underwent biopsies of the skin at the top of the right lateral thigh, in the territory of the lateral femoral cutaneous nerve. Each skin specimen was cut into 2 pieces. 2.4.4.1. Intraepidermal nerve fiber density The sample for assessing intraepidermal nerve fiber density (IENFD) was fixed in 4% paraformaldehyde (PFA) (pH 7.4; 2 hours; 4˚C) and stained for PGP9.5.38,50 An Axiophot2 microscope (Zeiss, Germany) with a CCD camera (Visitron Systems, Tuchheim, Germany) and SPOT Advanced software (Windows version 4.5) were used for visualization. Nerve fibers were counted under a 403 objective, and the mean IENFD per millimeter of epidermal length was calculated according to international criteria.32
reagent (Invitrogen, Karlsruhe, Germany), and homogenized (Polytron 1600E, Lucerne, Switzerland). All samples were incubated in 200 mL chloroform (25˚C, 3 minutes) and centrifuged (12,000g, 15 minutes, 4˚C). The supernatant was mixed with 500 mL isopropanol and incubated at 25˚C for 10 minutes. It was then centrifuged (12,000g, 10 minutes, 4˚C), and the pellet was washed with 75% ethanol and recovered by centrifugation (7500g, 5 minutes, 4˚C). The samples were air-dried on ice; the pellet was dissolved in diethylpyrocarbonate-treated water and incubated in a water bath at 55˚C for 10 minutes. It was then cooled and stored frozen at 280˚C until further processing. For reverse transcription polymerase chain reaction (PCR), all PCR reagents and cyclers were purchased from Life Technologies (Carlsbad, CA). The extracted mRNA was reversetranscribed using TaqMan Reverse Transcription Reagents. The reactions were carried out with 10 mL of 10X PCR Buffer, 6.25 mL MultiScribe Reverse Transcriptase, 2 mL RNase inhibitor, 22 mL MgCl2, 20 mL dNTPs, in an ABI PRISM 7700 Cycler, under the following conditions: 25˚C for 10 minutes; 48˚C for 60 minutes; 95˚C for 5 minutes. TaqMan Universal Master Mix and 5 mL of complementary DNA were used for quantitative real-time PCR in a StepOnePlus Thermocycler, in gene-specific TaqMan Assays. We investigated the proinflammatory cytokine TNF-a and NGF. The 25 mL reaction mix contained 12.5 mL TaqMan Master Mix and 1.25 mL primer. The cycler conditions were as follows: 50˚C for 2 minutes; 95˚C for 10 minutes; 45 cycles of 95˚C for 15 seconds, and 60˚C for 1 minute. Each plate contained a negative control and a blood sample from the control whose threshold cycle (Ct) values were nearest to the calculated mean of all control blood samples for each primer, as a calibrator for blood determinations. Samples were tested in triplicate, with the exception of 18s RNA levels that were determined in duplicate. All plates were analyzed under identical conditions.
2.4.4.2. Gene expression analysis The skin sample for gene expression analysis was flash-frozen in liquid nitrogen and stored at 280˚C until processing. Skin samples were then thawed on ice, immersed in 1 mL TRIzol
2.5. Statistical analysis Because of the small sample size (ie, ,20), nonparametric tests were used to compare patients with and without pain. Spearman
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
November 2015
·
Volume 156
·
Number 11
rank correlation tests were used for correlations; continuous variables were compared in Mann–Whitney U tests. Values are presented as median (range). Values of P , 0.05 were considered significant. Data were analyzed with SPSS version 18 software (SPSS, Chicago, IL).
3. Results 3.1. Characteristics of the study population
www.painjournalonline.com
2393
detection: r 5 0.64, P 5 0.001; cold detection: r 5 20.57, P 5 0.001; mechanical detection: r 5 0.64, P 5 0.001). Finally, 32 (72.8%) of the 44 patients examined at M3 had hypesthesia. Of these patients, 13 (28%) developed CPSNP (Fig. 2). We analyzed these patients further, to identify predictors of neuropathic pain. 3.3. Comparisons between patients with and without chronic postsurgical neuropathic pain in the presence of hypesthesia
Forty-six patients and 45 controls were included. The clinical characteristics of the patients are summarized in Table 1. One patient died and one was lost to follow-up during the study. Three patients refused the second biopsy at 3 months (unsightly scar). Quantitative sensory testing was performed at all 3 time points (baseline, D2, and M3), for all patients. A skin punch biopsy was performed at D2 for 44 patients and at M3 for 40 patients. Minor complications included 1 case of local bleeding requiring sutures, 1 case of persistent hypesthesia, 1 vasovagal attack, and 1 case of delayed wound healing (ie, .2 weeks). Each skin punch biopsy was divided in 2 parts and had independent transport from the harvest to the site of analysis. Six (2.7 %) biopsies for IENFD assessment and fourteen (6.3%) biopsies for mRNA cytokine assessment were lost during transport. Twenty-six (11.7%) biopsies for IENFD assessment and thirtyfive (15.8%) biopsies for mRNA cytokine assessment were not analyzed because of insufficient quality of tissue sample, conservation, or analysis problems. Finally, 189 biopsies were available for IENFD and 182 for mRNA cytokine assessment. When combining all these issues, the biopsy was analyzable for 44 patients at D2 and for 35 patients at M3, and a complete set of data (QST 1 IENFD 1 TNF-a 1 NGF) was obtained for 26 patients at D2 and 25 patients at M3.
Quantitative sensory testing showed hypesthesia for mechanical and heat detection and an increase in pain thresholds ipsilateral to ICBH, on day 2, in both groups (with and without CPSNP). This sensory deficit persisted at M3. However, no differences in QST results were observed when comparing patients with and without CPSNP at M3 (Table 2). Intraepidermal nerve fiber density did not differ between patients (or between either group of patients) and controls at D2. At M3, IENFD on the side ipsilateral to ICBH was lower than that on the contralateral side in patients. However, IENFD did not differ between patients with and without CPSNP, at D2 and M3 (Table 2 and Fig. 3). Cutaneous TNF-a and NGF gene expression did not differ between the ipsilateral and contralateral skin samples of patients with and without CPSNP. The TNF-a gene expression of these 2 patient groups did not differ from that of healthy controls at D2 and M3 (Table 2). At M3, levels of NGF gene expression in the skin were higher in both patient groups (with and without CPSNP) than in healthy controls (P 5 0.003). In patients with CPSNP, levels of NGF gene expression in the skin at M3 were higher on the ipsilateral side than on the contralateral side (P 5 0.04).
3.2. Mechanical and thermal detection thresholds and intraepidermal nerve fiber density after surgery
At M3, eleven patients had developed pain in the territory of the lateral femoral cutaneous nerve and 2 had developed pain in the territory of the iliohypogastric nerve. Three patients had severe pain (NRS $ 6), and 10 had moderate pain (NRS 3-6). The highest average pain intensity was 4.7 6 1.9 (range: 3-8). Total NPSI score was 12.25 6 10.3 (range: 3-36). Intraepidermal nerve fiber density at M3 was positively correlated with several clinical symptoms of neuropathic pain evaluated by the NPSI score: burning (r 5 0.68, P 5 0.02, Fig. 4), compression (r 5 0.76, P 5 0.006), and pain provoked or increased by brushing the painful area (r 5 0.67, P 5 0.02). No correlation was found between skin
Quantitative sensory testing revealed hypesthesia for mechanical, heat, and cold detection in 73.9%, 65.9%, and 59.1% of patients, respectively, at D2, and in 66.7%, 70.7%, and 61% of patients, respectively, at M3. The results obtained at D2 and M3 were strongly correlated for mechanical (r 5 0.875, P 5 0.001) and heat detection thresholds (r 5 0.87, P 5 0.001) and strongly negatively correlated for cold detection thresholds (r 5 20.85, P 5 0.001). Mechanical and heat detection thresholds were positively correlated (r 5 0.55, P 5 0.001), as were mechanical detection and heat pain thresholds (r 5 0.43, P 5 0.003) and mechanical pain and heat detection (r 5 0.58, P 5 0.001) and heat pain (r 5 0.6, P 5 0.001) thresholds. A negative correlation was found between mechanical and cold detection thresholds (r 5 20.46, P 5 0.001) and between mechanical pain and cold detection (r 5 20.32, P 5 0.02) and cold pain (r 5 20.39, P 5 0.04) thresholds. Intraepidermal nerve fiber density was low in 11.4% (4/35) of patients at D2 and in 60.6% (13/33) of patients at M3. Intraepidermal nerve fiber density was correlated with changes in the response to thermal stimulation. Thus, lower fiber densities were found to be associated with higher thresholds. This correlation was observed for all thermal thresholds (ie, heat detection: r 5 20.64, P 5 0.001; heat pain: r 5 20.42, P 5 0.01; cold detection: r 5 0.70, P 5 0.001; cold pain: r 5 0.39, P 5 0.02). The difference in IENFD between D2 and M3 was correlated with the detection threshold at D2 (heat detection: r 5 20.63, P 5 0.001; cold detection: r 5 20.58, P 5 0.001; mechanical detection: r 5 0.70, P 5 0.001) and at M3 (heat
3.4. Description of chronic postsurgical neuropathic pain at M3
Figure 2. Study flowchart. CPSNP, chronic postsurgical neuropathic pain; CPSP, chronic postsurgical pain.
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
2394
·
PAIN®
V. Martinez et al. 156 (2015) 2390–2398
Table 2
Comparison between patients with and without CPSNP in the presence of hypesthesia, at 3 months. CPSNP (N 5 13) Day 2 QST at D2 Mechanical detection threshold (log mg) Mechanical pain threshold (log mg) Heat detection threshold (˚C) Heat pain threshold (˚C) Cold detection threshold (˚C) Cold pain threshold (˚C) IENFD at D2 (fibers per mm3) mRNA gene expression at D2 TNF-a NGF 3 mo QST at M3 Mechanical detection threshold (log mg) Mechanical pain threshold (log mg) Heat detection threshold (˚C) Heat pain threshold (˚C) Cold detection threshold (˚C) Cold pain threshold (˚C) IENFD at M3 (fibers per mm3) mRNA gene expression at M3 TNF-a NGF
No CPSNP (N 5 19)
Control
Ipsilateral
Contralateral
Ipsilateral
Contralateral
Volunteers
Preoperative values
4.9 (0.9)*†‡ 5.3 (0.4)*‡ 47 (5.6)*†‡ 48.3 (3.1)*†‡ 12.8 (7.9)*†‡ 9.1 (1)‡ 7.4 (4)
2.4 (0.6) 4.4 (0.6) 34.8 (1.7)† 43 (2.7) 27.2 (4.7)† 13 (6) 8.6 (4.1)
4.7 (1)*†‡ 5.3 (0.2)*‡ 47 (6)*†‡ 48.8 (2.2)*†‡ 16.6 (8.4)*†‡ 10.9 (4)*† 8.5 (5.7)
2.03 (0.6) 4.5 (0.6) 36 (1.7) 42.4 (3.8) 27 (5) 15.6 (6.5) 10.7 (4.3)
2.32 (0.5) 3.8 (0.4) 33.4 (4.8) 43.3 (3.6) 29.5 (1.5) 11.9 (8) 9.8 (4)
2 (0.5) 4.25 (0.6) 35.4 (1.9) 44.3 (3) 30.1 (0.9) 15.1 (7) Nd
1.1 (0.5) 1.22 (0.5)
1 (1.2) 1.42 (0.4)
1.9 (1.7) 1.53 (0.5)†
1.28 (0.6) 1.25 (0.3)
1.35 (0.67) 1.15 (0.38)
Nd Nd
4.7 (0.9)*†‡ 5.4 (0.3)*†‡ 44.9 (5.4)*†‡ 48.0 (2.9)*†‡ 15.9 (8.2)*‡ 10 (0)*‡ 2.2 (3.8)*†‖
2.2 (0.7) 4.5 (0.7)† 34.3 (1.1) 42 (2.6) 30 (1.5) 16 (7) 9.2 (3)
4.6 (0.9)*†‡ 5.2 (0.4)*†‡ 45.2 (4.6)*†‡ 48.2 (2.3)*†‡ 19.9 (7.7)*†‡ 12.4 (6) 2.9 (5.5)*†‖
2.1 (0.6) 4.5 (0.6) 35 (1.5) 43.1 (3.6) 29.3 (1.5) 14.1 (7.4) 9.1 (3.4)
4.8 (2.5) 3.8 (4) 33.4 (4.8) 43.3 (3.6) 29.5 (1.5) 11.9 (8) 9.8 (4)
2 (0.5) 4.25 (0.6) 35.4 (1.9) 44.3 (3) 30.1 (0.9) 15.1 (7) Nd
0.99 (0.6) 1.44 (0.5)*
0.77 (0.4) 0.9 (0.5)
1.22 (0.6) 1.47 (0.7)†
0.9 (0.5) 1.1 (0.5)
1.35 (0.67) 1.15 (0.3)
Nd Nd
Continuous variables are presented as mean (SD). No difference was observed between CPSNP and no CPSNP groups. * P , 0.05 vs contralateral. † P , 0.05 vs volunteers. ‡ P , 0.05 vs preoperative values. ‖ P , 0.05 vs M3 values. CPSNP, chronic postsurgical neuropathic pain; IENFD, intraepidermal nerve fiber density; Nd, not determined; QST, quantitative sensory testing.
TNF-a and NGF gene expression and clinical symptoms or IENFD. Interestingly, patients developing CPSNP had higher preoperative pain ratings at rest, at the principal site of orthopedic surgery, than patients without CPSNP (median score of 3 on an 11-point NRS, range: 0-10, vs a median score of 0, range: 0-6, P 5 0.0001) and used larger amounts of preoperative opioids daily dose (median of 10 mg of morphine equivalent, range: 0-60, vs median of 3 mg of morphine equivalent, range: 0-10, P 5 0.02). Patients developing CPSNP had also higher pain ratings in recovery room (median score of 4.4 on an 11-point NRS, range: 2-7.5, vs a median score of 2.2, range: 0-7, P 5 0.003).
Figure 3. Correlation between IENFD and intensity of burning pain sensations at 3 months, in patients with CPSNP. CPSNP, chronic postsurgical neuropathic pain; IENFD, intraepidermal nerve fiber density; NRS, numerical rating scale.
4. Discussion 4.1. Main findings Neither IENFD nor changes in the expression of the TNF-a and NGF genes in the skin were associated with the development of CPSNP in patients with hypesthesia after ICBH. By contrast, high preoperative pain intensity and opioid use were associated with CPSNP. In patients with pain, CPSNP severity was inversely correlated with small fiber loss after ICBH.
4.2. Type of nerve lesion after surgery Nerve lesions are frequent after ICBH; hypesthesia and a decrease in IENFD are observed in 72.8% and 60% of patients, respectively. All fibers are affected by the nerve lesion. Detection and pain thresholds for both mechanical and thermal stimuli exploring large myelinated fibers, small myelinated Ad fibers, and unmyelinated C fibers were found to have increased at D2 and were almost stable at M3. The decrease in IENFD was observed only at M3 and was correlated with the changes in the response to thermal stimulation observed at M3. This delay in the appearance of decreases in nerve fiber density has already been reported in an animal nerve entrapment model.24 The correlation between functional impairment and a decrease in IENFD has also been observed in other neuropathies.16,23,48,59,61 We found that early impairment of nerve function, as revealed by QST at D2, was associated with a decrease in IENFD reduction at M3 and nerve lesion severity. Surgical trauma seems to induce homogeneous nerve fiber damage. We therefore suggest that a mechanical evaluation of hypesthesia might be sufficient to predict the
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
November 2015
·
Volume 156
·
Number 11
www.painjournalonline.com
2395
Figure 4. Comparison of IENFD 2 days and 3 months after ICBH. CPSNP, chronic postsurgical neuropathic pain; IENFD, intraepidermal nerve fiber density; ICBH, iliac crest bone harvest; no CPSNP, no chronic postsurgical neuropathic pain (mean 6 SD).
severity of the nerve lesion. This postsurgical nerve lesion model clearly differs from other degenerative neuropathies in which progressive small nerve fiber degeneration is the main mechanism. In conclusion, our data suggest a homogeneous lesion of nerve fiber subgroups after ICBH, involving large myelinated fibers, small myelinated Ad fibers, and unmyelinated C fibers. Mechanical sensory testing can be used for early detection and quantification of nerve lesions, at D2, whereas histological analysis can be used to confirm the lesion at M3. 4.3. Nerve lesions are not predictive of neuropathic pain after iliac crest bone harvest Although 60% to 73% of patients had signs of a nerve lesion, such as hypesthesia or a decrease in IENFD, only 40% of these patients developed CPSNP. The severity of the nerve lesion, as indicated by the intensity of hypesthesia or the magnitude of the decrease in IENFD, was not predictive of CPSNP. These findings confirm previous results in different surgical settings, in which no specific QST profile was identified for patients with CPSNP.1,21,39 In addition, our results indicate that IENFD is not predictive of CPSNP either. 4.4. Partial nerve lesion leads to more severe chronic postsurgical neuropathic pain Partial nerve lesion was associated with more severe CPSNP. Indeed, IENFD at M3 was positively correlated with pain intensity at M3 and with qualitative items of spontaneous pain, such as burning sensations. Our results are different from those obtained in Guillain–Barre´ syndrome,46 sarcoidosis,6 HIV, or diabetic neuropathies, in which inverse correlations with nerve fiber density have been reported.53,59,65 However, similar results have been reported in patients with postherpetic neuralgia, in which the decrease in the number of unmyelinated nerve fibers in the skin has been found to be inversely correlated with allodynia severity, suggesting that the presence of intact unmyelinated fibers is important for the induction of allodynia.29 These results
oppose the progressive degeneration of small nerves with a single-incident surgical lesion of small and large fibers. Our results are consistent with the hypothesis that neuroactive substances released along the length of the degenerating fibers of the partial lesion can activate uninjured fibers, generating neuropathic pain.47,51,52,62 This mechanism has been tested in animal models, using entrapment, stretching, or constriction of the sciatic nerve to develop partial nerve lesions and related neuropathic pain behavior in experimental animals; pure nerve transection does not always lead to similar symptoms.5,10,45 Our findings are consistent with those of clinical studies in other surgical models. In inguinal hernia repair, a combined analysis of 3 studies suggested that neurectomy less frequently leads to CPSNP than nerve preservation.3,27 In mandibular osteotomy, electrophysiological findings indicated that axotomy was less predictive of CPSNP than partial axonal lesions.25 These results suggest that partial nerve damage is associated with a higher risk of severe CPSNP than total nerve section. 4.5. Skin TNF-a gene expression During nerve degeneration, proinflammatory cytokines are produced by various cells, including the Schwann cells.29,47 Our results for patients undergoing ICBH do not confirm those of previous clinical studies suggesting an increase in the expression of proinflammatory cytokine genes after trauma30 or in peripheral neuropathies.55 Such an increase in proinflammatory cytokine expression has been observed at various sites, including the blood,56 cerebrospinal fluid, and the sural nerve.35 This increase has been confirmed by different techniques measuring cytokine expression directly by immunochemical methods or by assessing gene expression. An increase in TNFa expression has been reported in various peripheral neuropathies17,35 and in complex regional pain syndrome.34 An increase in TNF-a expression has also been observed bilaterally34 and can persist for months.30 Such an increase has been reported in painful neuropathies but not in painless neuropathies.17,56 We found no increase in skin TNF-a gene expression in
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
2396
·
PAIN®
V. Martinez et al. 156 (2015) 2390–2398
the acute and chronic postoperative phases. We also found no differences in the expression of this gene between the 2 sides of the body or between patients with and without pain. These findings are similar to those reported for patients with postherpetic neuralgia57 and fibromyalgia syndrome.54 There are several possible reasons for the lack of change in TNF-a gene expression observed here. These reasons include the much shorter observation period (only 3 months) in our study than in previous studies on cutaneous TNF-a expression, in which follow-up continued over a period of several years. Differences in the physiopathology of pain after surgery (ie, acute cutting) and in diseases with a chronic pattern of development (ie, neuropathies) must also be taken out, although proinflammatory cytokines have been implicated in different types of traumatic and inflammatory neuropathies.15 Methodological limitations, such as the small sample size and unstandardized timing of skin punch biopsies, may also have influenced the results. Finally, our study focused on TNF-a, and we cannot rule out the possibility of changes in the expression of genes encoding other proinflammatory and anti-inflammatory cytokines. 4.6. Skin NGF gene expression Neurotrophic factors are crucial for nerve integrity and several studies have shown that NGF levels, for example, are high in nerve specimens from patients with peripheral neuropathies.63 Cutaneous NGF expression has been investigated in only a few studies to date. Levels of NGF expression are low in the skin samples of patients with diabetic neuropathy or neuropathies of other etiologies,4,55 and these previous studies found no correlation between NGF expression in patients with and without pain. Our findings conflict with these previous reports. We observed an increase in NGF expression at 3 months after surgery, and NGF gene expression was not associated with the development of CPSNP in patients with hypesthesia. These findings are consistent with those of animal studies showing an increase in skin NGF gene expression after chronic constriction injury in rats.42 It is possible that excessive cutaneous NGF expression, together with a decrease in innervation density, causes nociceptor sensitization through phenotypic changes, eventually leading to chronic pain.18 We observed no difference in NGF gene expression in the skin between patients with and without hypesthesia. Our results therefore suggest that NGF is not suitable for use as a marker of nerve lesions (results not shown). 4.7. Clinical predictors of the development of chronic postsurgical neuropathic pain Interestingly, although psychophysical, histological, and molecular factors were not associated with the development of CPSNP, we observed that, in cases of nerve lesion after ICBH, the severity of preoperative at the surgery site and the use of higher doses of opioids before surgery were associated with the development of CPSNP. We also found that the severity of early pain was also predictive of chronic postsurgical pain (CPSP) as previously reported in our surgical model.37 The presence and duration of preoperative pain is predictive of CPSP in various surgical models.28 Both pain at the site of surgery8,31,40,43,44 and pain at a distant location have been identified as predictive factors for CPSP.13,20,44,64 As previously reported for spinal surgery, we observed that the intensity of preoperative pain was associated with a poor
outcome in surgical treatment for lumbar radiculopathy.60 We suggest that preoperative nervous system sensitization may increase the risk of developing CPSP after surgery. The use of higher doses of opioids before surgery was also associated with CPSP. It is unclear whether this association is due to the severity of preoperative pain or the potential impact of opioids on this sensitization. There are experimental data suggesting that higher doses of opioids may lead to hyperalgesia and a potential increase in acute postsurgical pain, leading to CPSP.49 One clinical study has reported that the preoperative use of opioids might modify the postoperative pain trajectory, magnifying the impact of preoperative pain on the incidence of CPSP after gynecologic surgery.58 Our results support the notion that preoperative clinical factors may favor not only CPSP but also, more specifically, the development of neuropathic pain after a nerve lesion.
Conflict of interest statement The authors have no conflicts of interest to declare. This study was supported by a grant from the Institut UPSA de la Douleur, the APICIL Foundation, and CNP Assurances. Article history: Received 8 May 2015 Received in revised form 24 June 2015 Accepted 25 June 2015 Available online 7 July 2015
References [1] Aasvang EK, Brandsborg B, Christensen B, Jensen TS, Kehlet H. Neurophysiological characterization of postherniotomy pain. PAIN 2008; 137:173–81. [2] Aasvang E, Kehlet H. Chronic postoperative pain: the case of inguinal herniorrhaphy. Br J Anaesth 2005;95:69–76. [3] Alfieri S, Amid PK, Campanelli G, Izard G, Kehlet H, Wijsmuller AR, Di Miceli D, Doglietto GB. International guidelines for prevention and management of post-operative chronic pain following inguinal hernia surgery. Hernia 2011;15:239–49. [4] Anand P, Terenghi G, Warner G, Kopelman P, Williams-Chestnut RE, Sinicropi DV. The role of endogenous nerve growth factor in human diabetic neuropathy. Nat Med 1996;2:703–7. [5] Attal N, Filliatreau G, Perrot S, Jazat F, Di Giamberardino L, Guilbaud G. Behavioural pain-related disorders and contribution of the saphenous nerve in crush and chronic constriction injury of the rat sciatic nerve. PAIN 1994;59:301–12. [6] Bakkers M, Merkies IS, Lauria G, Devigili G, Penza P, Lombardi R, Hermans MC, van Nes SI, De Baets M, Faber CG. Intraepidermal nerve fiber density and its application in sarcoidosis. Neurology 2009;73:1142–8. [7] Basbaum AI, Gautron M, Jazat F, Mayes M, Guilbaud G. The spectrum of fiber loss in a model of neuropathic pain in the rat: an electron microscopic study. PAIN 1991;47:359–67. [8] Bates T, Ebbs SR, Harrison M, A’Hern RP. Influence of cholecystectomy on symptoms. Br J Surg 1991;78:964–7. [9] Benedetti F, Amanzio M, Casadio C, Filosso PL, Molinatti M, Oliaro A, Pischedda F, Maggi G. Postoperative pain and superficial abdominal reflexes after posterolateral thoracotomy. Ann Thorac Surg 1997;64:207–10. [10] Bennett GJ, Xie YK. A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. PAIN 1988;33: 87–107. [11] Bouhassira D, Attal N, Alchaar H, Boureau F, Brochet B, Bruxelle J, Cunin G, Fermanian J, Ginies P, Grun-Overdyking A, Jafari-Schluep H, LanteriMinet M, Laurent B, Mick G, Serrie A, Valade D, Vicaut E. Comparison of pain syndromes associated with nervous or somatic lesions and development of a new neuropathic pain diagnostic questionnaire (DN4). PAIN 2005;114:29–36. [12] Bouhassira D, Attal N, Fermanian J, Alchaar H, Gautron M, Masquelier E, Rostaing S, Lanteri-Minet M, Collin E, Grisart J, Boureau F. Development and validation of the Neuropathic Pain Symptom Inventory. PAIN 2004; 108:248–57.
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
November 2015
·
Volume 156
·
Number 11
[13] Brandsborg B, Nikolajsen L, Hansen CT, Kehlet H, Jensen TS. Risk factors for chronic pain after hysterectomy: a nationwide questionnaire and database study. Anesthesiology 2007;106:1003–12. [14] Calvo M, Dawes JM, Bennett DL. The role of the immune system in the generation of neuropathic pain. Lancet Neurol 2012;11:629–42. [15] Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain Inventory. Ann Acad Med Singapore 1994;23:129–38. [16] Devigili G, Tugnoli V, Penza P, Camozzi F, Lombardi R, Melli G, Broglio L, Granieri E, Lauria G. The diagnostic criteria for small fibre neuropathy: from symptoms to neuropathology. Brain 2008;131: 1912–25. [17] Empl M, Renaud S, Erne B, Fuhr P, Straube A, Schaeren-Wiemers N, Steck AJ. TNF-alpha expression in painful and nonpainful neuropathies. Neurology 2001;56:1371–7. [18] Evans LJ, Loescher AR, Boissonade FM, Whawell SA, Robinson PP, Andrew D. Temporal mismatch between pain behaviour, skin nerve growth factor and intra-epidermal nerve fibre density in trigeminal neuropathic pain. BMC Neurosci 2014;15:1. [19] Fruhstorfer H, Lindblom U, Schmidt WC. Method for quantitative estimation of thermal thresholds in patients. J Neurol Neurosurg Psychiatry 1976;39:1071–5. [20] Gartner R, Jensen MB, Nielsen J, Ewertz M, Kroman N, Kehlet H. Prevalence of and factors associated with persistent pain following breast cancer surgery. JAMA 2009;302:1985–92. [21] Gottrup H, Andersen J, Arendt-Nielsen L, Jensen TS. Psychophysical examination in patients with post-mastectomy pain. PAIN 2000;87: 275–84. [22] Guilbaud G, Gautron M, Jazat F, Ratinahirana H, Hassig R, Hauw JJ. Time course of degeneration and regeneration of myelinated nerve fibres following chronic loose ligatures of the rat sciatic nerve: can nerve lesions be linked to the abnormal pain-related behaviours? PAIN 1993;53: 147–58. [23] Herrmann DN, McDermott MP, Sowden JE, Henderson D, Messing S, Cruttenden K, Schifitto G. Is skin biopsy a predictor of transition to symptomatic HIV neuropathy? A longitudinal study. Neurology 2006;66: 857–61. [24] Hsieh CH, Jeng SF, Lu TH, Chen YC, Hsieh MW, Chen SS. Loss of small fibers in entrapment neuropathy and their regeneration after surgical decompression in a rat model. J Neurotrauma 2007;24:1658–66. [25] Jaaskelainen SK, Teerijoki-Oksa T, Forssell H. Neurophysiologic and quantitative sensory testing in the diagnosis of trigeminal neuropathy and neuropathic pain. PAIN 2005;117:349–57. [26] Jaaskelainen SK, Teerijoki-Oksa T, Virtanen A, Tenovuo O, Forssell H. Sensory regeneration following intraoperatively verified trigeminal nerve injury. Neurology 2004;62:1951–7. [27] Johner A, Faulds J, Wiseman SM. Planned ilioinguinal nerve excision for prevention of chronic pain after inguinal hernia repair: a meta-analysis. Surgery 2011;150:534–41. [28] Kehlet H, Jensen TS, Woolf CJ. Persistent postsurgical pain: risk factors and prevention. Lancet 2006;367:1618–25. [29] Koike H, Iijima M, Mori K, Yamamoto M, Hattori N, Watanabe H, Tanaka F, Doyu M, Sobue G. Neuropathic pain correlates with myelinated fibre loss and cytokine profile in POEMS syndrome. J Neurol Neurosurg Psychiatry 2008;79:1171–9. [30] Kramer HH, Eberle T, Uceyler N, Wagner I, Klonschinsky T, Muller LP, Sommer C, Birklein F. TNF-alpha in CRPS and “normal” trauma—significant differences between tissue and serum. PAIN 2011; 152:285–90. [31] Kroner K, Krebs B, Skov J, Jorgensen HS. Immediate and long-term phantom breast syndrome after mastectomy: incidence, clinical characteristics and relationship to pre-mastectomy breast pain. PAIN 1989;36:327–34. [32] Lauria G, Cornblath DR, Johansson O, McArthur JC, Mellgren SI, Nolano M, Rosenberg N, Sommer C. EFNS guidelines on the use of skin biopsy in the diagnosis of peripheral neuropathy. Eur J Neurol 2005;12: 747–58. [33] Lauria G, Hsieh ST, Johansson O, Kennedy WR, Leger JM, Mellgren SI, Nolano M, Merkies IS, Polydefkis M, Smith AG, Sommer C, Valls-Sole J. European Federation of Neurological Societies/Peripheral Nerve Society Guideline on the use of skin biopsy in the diagnosis of small fiber neuropathy. Report of a joint task force of the European Federation of Neurological Societies and the Peripheral Nerve Society. Eur J Neurol 2010;17:903–12, e44–9. [34] Lenz M, Uceyler N, Frettloh J, Hoffken O, Krumova EK, Lissek S, Reinersmann A, Sommer C, Stude P, Waaga-Gasser AM, Tegenthoff M, Maier C. Local cytokine changes in complex regional pain syndrome type I (CRPS I) resolve after 6 months. PAIN 2013;154: 2142–9.
www.painjournalonline.com
2397
[35] Lindenlaub T, Sommer C. Cytokines in sural nerve biopsies from inflammatory and non-inflammatory neuropathies. Acta neuropathologica 2003;105:593–602. [36] Macrae WA. Chronic post-surgical pain: 10 years on. Br J Anaesth 2008; 101:77–86. [37] Martinez V, Ben Ammar S, Judet T, Bouhassira D, Chauvin M, Fletcher D. Risk factors predictive of chronic postsurgical neuropathic pain: the value of the iliac crest bone harvest model. PAIN 2012;153:1478–83. [38] McCarthy BG, Hsieh ST, Stocks A, Hauer P, Macko C, Cornblath DR, Griffin JW, McArthur JC. Cutaneous innervation in sensory neuropathies: evaluation by skin biopsy. Neurology 1995;45:1848–55. [39] Mikkelsen T, Werner MU, Lassen B, Kehlet H. Pain and sensory dysfunction 6 to 12 months after inguinal herniotomy. Anesth Analg 2004;99:146–51. [40] Nikolajsen L, Ilkjaer S, Kroner K, Christensen JH, Jensen TS. The influence of preamputation pain on postamputation stump and phantom pain. PAIN 1997;72:393–405. [41] Owens WD, Felts JA, Spitznagel EL Jr. ASA physical status classifications: a study of consistency of ratings. Anesthesiology 1978; 49:239–43. [42] Peleshok JC, Ribeiro-da-Silva A. Neurotrophic factor changes in the rat thick skin following chronic constriction injury of the sciatic nerve. Mol Pain 2012;8:1. [43] Perttunen K, Tasmuth T, Kalso E. Chronic pain after thoracic surgery: a follow-up study. Acta Anaesthesiol Scand 1999;43:563–7. [44] Puolakka PA, Rorarius MG, Roviola M, Puolakka TJ, Nordhausen K, Lindgren L. Persistent pain following knee arthroplasty. Eur J Anaesthesiol 2010;27:455–60. [45] Robinson LR. Traumatic injury to peripheral nerves. Muscle Nerve 2000; 23:863–73. [46] Ruts L, van Doorn PA, Lombardi R, Haasdijk ED, Penza P, Tulen JH, Hempel RJ, van den Meiracker AH, Lauria G. Unmyelinated and myelinated skin nerve damage in Guillain-Barre syndrome: correlation with pain and recovery. PAIN 2012;153:399–409. [47] Schafers M, Geis C, Svensson CI, Luo ZD, Sommer C. Selective increase of tumour necrosis factor-alpha in injured and spared myelinated primary afferents after chronic constrictive injury of rat sciatic nerve. Eur J Neurosci 2003;17:791–804. [48] Schuning J, Scherens A, Haussleiter IS, Schwenkreis P, Krumova EK, Richter H, Maier C. Sensory changes and loss of intraepidermal nerve fibers in painful unilateral nerve injury. Clin J Pain 2009;25:683–90. [49] Simonnet G. Acute tolerance to opioids: methodological, theoretical and clinical implications. PAIN 2009;142:3–4. [50] Sommer C. Skin biopsy as a diagnostic tool. Curr Opin Neurol 2008;21: 563–8. [51] Sommer C, Galbraith JA, Heckman HM, Myers RR. Pathology of experimental compression neuropathy producing hyperesthesia. J Neuropathol Exp Neurol 1993;52:223–33. [52] Sommer C, Lalonde A, Heckman HM, Rodriguez M, Myers RR. Quantitative neuropathology of a focal nerve injury causing hyperalgesia. J Neuropathol Exp Neurol 1995;54:635–43. [53] Sorensen L, Molyneaux L, Yue DK. The relationship among pain, sensory loss, and small nerve fibers in diabetes. Diabetes Care 2006;29:883–7. [54] Uceyler N, Kewenig S, Kafke W, Kittel-Schneider S, Sommer C. Skin cytokine expression in patients with fibromyalgia syndrome is not different from controls. BMC Neurol 2014;14:185. [55] Uceyler N, Riediger N, Kafke W, Sommer C. Differential gene expression of cytokines and neurotrophic factors in nerve and skin of patients with peripheral neuropathies. J Neurol 2015;262:203–12. [56] Uceyler N, Rogausch JP, Toyka KV, Sommer C. Differential expression of cytokines in painful and painless neuropathies. Neurology 2007;69:42–9. [57] Uceyler N, Valet M, Kafke W, Tolle TR, Sommer C. Local and systemic cytokine expression in patients with postherpetic neuralgia. PLoS One 2014;9:e105269. [58] VanDenKerkhof EG, Hopman WM, Goldstein DH, Wilson RA, Towheed TE, Lam M, Harrison MB, Reitsma ML, Johnston SL, Medd JD, Gilron I. Impact of perioperative pain intensity, pain qualities, and opioid use on chronic pain after surgery: a prospective cohort study. Reg Anesth Pain Med 2012;37:19–27. [59] Vlckova-Moravcova E, Bednarik J, Dusek L, Toyka KV, Sommer C. Diagnostic validity of epidermal nerve fiber densities in painful sensory neuropathies. Muscle Nerve 2008;37:50–60. [60] Voorhies RM, Jiang X, Thomas N. Predicting outcome in the surgical treatment of lumbar radiculopathy using the Pain Drawing Score, McGill Short Form Pain Questionnaire, and risk factors including psychosocial issues and axial joint pain. Spine J 2007;7:516–24. [61] Walk D, Wendelschafer-Crabb G, Davey C, Kennedy WR. Concordance between epidermal nerve fiber density and sensory examination in
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
2398
·
V. Martinez et al. 156 (2015) 2390–2398
patients with symptoms of idiopathic small fiber neuropathy. J Neurol Sci 2007;255:23–6. [62] Watkins LR, Maier SF. Beyond neurons: evidence that immune and glial cells contribute to pathological pain states. Physiol Rev 2002;82: 981–1011. [63] Yamamoto M, Ito Y, Mitsuma N, Hattori N, Sobue G. Pain-related differential expression of NGF, GDNF, IL-6, and their receptors in human vasculitic neuropathies. Intern Med 2003;42:1100–3.
PAIN®
[64] Yunker AC, Ritch JM, Robinson EF, Golish CT. Incidence and risk factors for chronic pelvic pain after hysteroscopic sterilization. J Minim Invasive Gynecol 2015;22:390–4. [65] Zhou L, Kitch DW, Evans SR, Hauer P, Raman S, Ebenezer GJ, Gerschenson M, Marra CM, Valcour V, Diaz-Arrastia R, Goodkin K, Millar L, Shriver S, Asmuth DM, Clifford DB, Simpson DM, McArthur JC. Correlates of epidermal nerve fiber densities in HIV-associated distal sensory polyneuropathy. Neurology 2007;68:2113–19.
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
Clinical, histological, and biochemical predictors of postsurgical neuropathic pain ¨ ¸ eylerc, Skander Ben Ammara, Jean-Claude Alvareze, Fabrice Gaudotd, Valeria ´ Martineza,b,*, Nurcan Uc c Claudia Sommer , Didier Bouhassirab, Dominique Fletchera,b
Abstract Surgical nerve injury sometimes leads to chronic postsurgical neuropathic pain (CPSNP). The risk factors for this condition are not well understood. We prospectively assessed 46 patients scheduled for iliac crest bone harvest, 2 days (D2) and 3 months (M3) after surgery, to determine the time course of nerve fiber degeneration and expression of the TNF-a and NGF genes in skin punch biopsies. Mechanical and thermal detection and pain thresholds were evaluated at D2 and M3, by quantitative sensory testing. Skin punch biopsies were also obtained from the thighs ipsilateral and contralateral to iliac crest bone harvest. Intraepidermal nerve fiber density (IENFD) and cutaneous TNF-a and NGF gene expression were analyzed. Forty-five volunteers matched for age, sex, skin color were examined as controls. Chronic postsurgical neuropathic pain was defined as pain in an area of hypesthesia with a positive Douleur Neuropathique 4 questionnaire score. Overall, 73% (N 5 32) of patients developed hypesthesia and 40% (N 5 13) of these patients had developed CPSNP at M3. Quantitative sensory testing results, IENFD, and skin TNF-a and NGF gene expression at D2 and M3 did not differ between patients with and without CPSNP. However, in patients with CPSNP, burning, compression, and pain provoked by brushing were correlated with IENFD at M3, suggesting a possible association between partial nerve lesions and more intense CPSNP, than with total nerve lesion. Furthermore, preoperative pain and opioid use were higher in patients who developed CPSNP than in those without CPSNP. These findings suggest that the predictors of CPSNP development are clinical rather than histological or biochemical. Keywords: Neuropathic pain, Surgery, Nerve lesion, Small nerve fiber degeneration, TNF-a, NGF
1. Introduction 28,36
Most persistent pain after surgery is neuropathic and related to nerve lesions.9 However, surgical nerve injury does not always lead to neuropathic pain,1,26,39 and the relationship between nerve lesion severity and chronic postsurgical neuropathic pain (CPSNP) development remains unclear. We previously investigated the link between nerve lesions and hyperalgesia in the development of CPSNP.37 We used the iliac crest bone harvest (ICBH) model, in which it is possible to explore the territory of nerve lesions at some distance from the surgical site. We reported that 60% of patients had hypesthesia after surgery and 23% of patients had developed neuropathic pain 3 months after surgery. Multivariate analysis identified the area of Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article. a Service d’Anesthesie ´ Reanimation ´ Chirurgicale, Hopital ˆ Raymond-Poincare, ´ APHP, Universite´ de Versailles St-Quentin, Garches, France, b INSERM U-987, Centre d’Evaluation et de Traitement de la Douleur, Hopital ˆ Ambroise Pare, ´ BoulogneBillancourt, France; Universite´ Versailles Saint-Quentin, Versailles, France, c Department of Neurology, University of Wurzburg, ¨ Wurzburg, ¨ Germany, d Service d’orthopedie, ´ Hopital ˆ Raymond-Poincare, ´ AP-HP, Universite´ de Versailles StQuentin, Garches, France, e Service de Biochimie, Hopital ˆ Raymond-Poincare, ´ Garches, France
*Corresponding author. Address: Service d’Anesthesie ´ Reanimation ´ Chirurgicale, Hopital ˆ Raymond-Poincare, ´ AP-HP, Universite´ de Versailles St-Quentin, 104 Boulevard Raymond Poincare´ 92380 Garches, France. Tel.: 33(1)47107622; fax 33(1)47107623. E-mail address: [email protected] (V. Martinez). PAIN 156 (2015) 2390–2398 © 2015 International Association for the Study of Pain http://dx.doi.org/10.1097/j.pain.0000000000000286
2390
·
V. Martinez et al. 156 (2015) 2390–2398
hyperalgesia and the neuropathic nature of early pain as 2 independent predictive factors for CPSNP. No link was found between the area or severity of mechanical hypesthesia and the development of CPSNP.2,21,37 Skin punch biopsy is a validated technique for quantifying epidermal unmyelinated small nerve fiber density with the panaxonal marker protein gene product 9.5 (PGP9.5).16,33 We used the ICBH model because, in this model, skin biopsies can be performed at some distance from the site of potential nerve damage, where fiber degeneration may predominate, as suggested in the sciatic nerve ligature animal model.7,22 Immune activation also plays a crucial role in pain induction and maintenance.14,62 Peripheral nerve injury–induced inflammatory mediators may contribute to the development of abnormalities in the sensory system, such as ectopic activity and enhanced responses to sensory stimuli.14 It has been suggested that inflammation related to Wallerian degeneration can also cause unscathed nerve fiber hyperexcitability.51,52 A number of mediators and cells are thought to be involved, including the proinflammatory cytokine tumor necrosis factor a (TNF-a) and neurotrophic growth factor (NGF), both of which are well-known mediators of neuropathic pain.14,42,47 We investigate the reasons for which only one-third of patients with nerve lesion develop CPSNP, by analyzing the contributions of skin small nerve fiber degeneration and cutaneous expression of the TNF-a and NGF genes to CPSNP. We prospectively assessed patients scheduled for ICBH, from the early postoperative period until 3 months after surgery, to determine the time course of small nerve fiber degeneration and cutaneous expression of the TNF-a and NGF genes. PAIN®
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
November 2015
·
Volume 156
·
Number 11
www.painjournalonline.com
2391
2. Methods 2.1. Patients Approval was obtained from the institutional review board (Comite´ de Protection des Personnes, Boulogne-Billancourt). We prospectively included consecutive patients undergoing ICBH surgery at our institution between January 2010 and September 2012. All patients provided written informed consent. The inclusion criteria were age $18 years and patients scheduled for orthopedic surgery with ICBH. The noninclusion criteria were patient refusal, previous ICBH, ipsilateral hip surgery, previous surgery, or chronic pain in the area of skin biopsy. The exclusion criteria were decision of the patient to withdraw from the study, postoperative complications (hematoma, infection), inability to answer questions, impossibility of examining the patient, and nonstandardized anesthesia and analgesia procedures. 2.2. Anesthesia and postoperative pain relief Patients received 100 mg hydroxyzine before surgery, which was performed under balanced general anesthesia with propofol, sufentanil, atracurium, and sevorane, together with a 1:1 mixture of nitrous oxide and oxygen. Pain immediately after surgery was controlled by 48 hours of intravenous morphine treatment, delivered by patient-controlled analgesia methods, together with 1 g acetaminophen every 6 h and 50 mg ketoprofen every 6 h over the first week. 2.3. Preoperative assessment We collected demographic data, including age, sex, body mass index, and American Society of Anesthesiology score.41 The presence of preoperative chronic pain at the principal site of surgery was recorded, including the intensity of the pain on a numerical rating scale (NRS) from 0 to 10, at rest and on movement, pain duration, and opioid consumption during the last month. 2.4. Postoperative assessment 2.4.1. Pain assessment The intensity, location, and characteristics of ICBH-related pain were assessed on day 2 (D2) after surgery and 3 months after surgery (M3). Pain intensity was evaluated on an NRS. Pain location was determined with a diagram from the Brief Pain Inventory (BPI).15 Chronic postsurgical neuropathic pain was defined as pain in an area of hypesthesia corresponding to the territory of the lateral femoral cutaneous, ilioinguinal, or cluneal nerves, 3 months after ICBH, with a positive Douleur Neuropathique 4 questionnaire score11 (Fig. 1). Patients were asked to report the intensity and location of the pain. The areas of pain, including that of maximal pain, were indicated on the body map from the BPI.15 Current, average, and maximum pain intensity over the last 24 hours were assessed in the area of maximum pain, using an NRS from the BPI.15 At 3 months, patients with neuropathic pain were interviewed with the Neuropathic Pain Symptom Inventory (NPSI), which includes questions relating to 10 symptoms rated on an NRS (burning, squeezing, pressure, electric shocks, stabbing, pain evoked by brushing, by pressure, by cold, tingling, pins, and needles).12 2.4.2. Sensory mapping of postoperative hypesthesia The methodology used has been described elsewhere.37 The area of mechanical hypesthesia was assessed at D2 and M3. We
Figure 1. Localization of the areas explored. (A) Site of surgery for iliac crest bone harvest; (B) territory innervated by the lateral femoral cutaneous nerve; (C) territory innervated by the ilioinguinal nerve; (D) territory innervated by the cuneal nerve. When no area of mechanical hypesthesia was detectable in one of these 3 territories (A, B, or C), assessments (quantitative sensory testing and skin punch biopsies) were performed on the top of the thigh.
first checked for a sensory deficit in the 3 cutaneous territories innervated by the nerves crossing the surgical site (the top of the thigh for the lateral femoral cutaneous nerve, the inguinal area for the iliohypogastric nerve, and the buttocks for the superior cluneal nerves). If a sensory deficit was found, the border of hypesthesia was determined by stimulation along 8 linear paths diverging outwards from the site of tactile hypesthesia, until the patient reported normal perception with a von Frey monofilament (60g; Bioseb, Chaville, France). The first point at which sensation was similar to that on the contralateral side was marked. The 8 demarcating points defined the territory of mechanical hypesthesia. 2.4.3. Quantitative sensory testing Bedside quantitative sensory testing (QST) was performed for the site of mechanical hypesthesia. Quantitative sensory testing was performed the day before surgery, on day 2, and 3 months after surgery. For preoperative evaluation and for patients without a detectable area of mechanical hypesthesia, QST was performed on the top of the thigh, by 2 experienced investigators (V.M. and S.B.) (Fig. 1). Similar tests were performed on healthy volunteers (V.M.). Thermal detection and pain thresholds were assessed with a Somedic Thermotest (Somedic AB, Stockholm, Sweden), using the Marstock method.19 A contact thermode of Peltier elements measuring 25 3 50 mm was applied to the skin. The baseline temperature of the thermode was set at 32˚C. The maximum and minimum temperatures were set at 50˚C for heat and 4˚C for cold pain. A thermal rate of change of 1˚C/s was used. All thresholds were calculated as the mean of 3 successive determinations. Calibrated von Frey monofilaments, 0.007 to 300g (Bioseb), were used to measure detection and pain thresholds in response to mechanical punctate stimuli, according to the method of limits. The tactile detection and pain detection thresholds were defined as the smallest force eliciting a sensation of contact or pain, respectively. The force required to bend the filaments was converted into log units. The contralateral side was used as a control.
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
2392
·
PAIN®
V. Martinez et al. 156 (2015) 2390–2398
Table 1
Preoperative characteristics of the patients and controls. Patient characteristics No. of patients Age, y Sex (men/women) BMI, kg/m2 ASA classification (I, II) Duration of preoperative pain, mo Preoperative pain intensity at rest (NRS) Preoperative pain intensity on movement (EN) Preoperative daily opioid dose (morphine equivalent) Surgery characteristics Upper limb surgery Lower limb surgery Spinal surgery Duration of surgery, min ICBH characteristics Length of the scar, cm Site of crest harvest incision (anterior/posterior) Volume (cortical), cm3
All patients
Patients with hypesthesia
Control
CPSNP
No CPSNP
46 50 (16-80) 26 (56.5%)/20 (43.5%) 25 (16-37) 41 (91%), 4 (9%) 18 (0-200) 1 (0-10) 5 (0-10) 0 (0-60)
13 48 (34-74) 4 (30%)/9 (70%) 25 (18-33) 13 (100%) 48 (0-190) 3 (0-10)* 6.5 (0-10) 10 (0-60)*
19 47 (19-67) 14 (73%)/5 (27%) 24 (16-30) 18 (94%)/1 (6%) 12 (0-120) 0 (0-6) 4.5 (0-9) 3 (0-10)
45 50 (19-79) 25 (55%)/20 (45%) 24 (19-39) NA NA NA NA NA
3 42 1 159 (50-300)
0 13 0 175 (120-230)
2 17 0 150 (60-300)
NA
NA
8.8 (3-15) 40 (87%)/6 (13%) 31 (1-320)
10 (5-13) 11 (84%)/2 (16%) 15 (3-120)
10 (5-15) 18 (94%)/1 (6%) 31 (1.5-320)
NA NA NA
Continuous variables are presented as median (range). Other variables are presented as n (%). * Significant difference between CPSNP and no CPSNP, P , 0.05. ASA, American Society of Anesthesiology Physical Status Score; BMI, body mass index; CPSNP, chronic postsurgical neuropathic pain; ICBH, iliac crest bone harvest; NA, not applicable; NRS, numerical rating scale.
Patients were considered to present hypesthesia when at least 1 of the 3 mean thresholds, heat, cold, or mechanical detection thresholds, differed from preoperative values by at least 2 SDs. 2.4.4. Skin punch biopsies Skin punch biopsies were obtained according to the European recommendations.32 Two days and 3 months after surgery, two 5-mm skin punch biopsies (apparatus from Stiefel GmbH, Offenbach, Germany) were performed under local anesthesia, for each of the patients. One was performed on the leg ipsilateral to ICBH, either in the area of hypesthesia or on the lateral thigh in the territory of the lateral femoral cutaneous nerve. The second biopsy was performed on the contralateral thigh. The control group for skin analysis consisted of 45 healthy volunteers matched for age, sex, and skin color (25 men, 20 women; median age: 52 years [range: 19-70]). All controls underwent biopsies of the skin at the top of the right lateral thigh, in the territory of the lateral femoral cutaneous nerve. Each skin specimen was cut into 2 pieces. 2.4.4.1. Intraepidermal nerve fiber density The sample for assessing intraepidermal nerve fiber density (IENFD) was fixed in 4% paraformaldehyde (PFA) (pH 7.4; 2 hours; 4˚C) and stained for PGP9.5.38,50 An Axiophot2 microscope (Zeiss, Germany) with a CCD camera (Visitron Systems, Tuchheim, Germany) and SPOT Advanced software (Windows version 4.5) were used for visualization. Nerve fibers were counted under a 403 objective, and the mean IENFD per millimeter of epidermal length was calculated according to international criteria.32
reagent (Invitrogen, Karlsruhe, Germany), and homogenized (Polytron 1600E, Lucerne, Switzerland). All samples were incubated in 200 mL chloroform (25˚C, 3 minutes) and centrifuged (12,000g, 15 minutes, 4˚C). The supernatant was mixed with 500 mL isopropanol and incubated at 25˚C for 10 minutes. It was then centrifuged (12,000g, 10 minutes, 4˚C), and the pellet was washed with 75% ethanol and recovered by centrifugation (7500g, 5 minutes, 4˚C). The samples were air-dried on ice; the pellet was dissolved in diethylpyrocarbonate-treated water and incubated in a water bath at 55˚C for 10 minutes. It was then cooled and stored frozen at 280˚C until further processing. For reverse transcription polymerase chain reaction (PCR), all PCR reagents and cyclers were purchased from Life Technologies (Carlsbad, CA). The extracted mRNA was reversetranscribed using TaqMan Reverse Transcription Reagents. The reactions were carried out with 10 mL of 10X PCR Buffer, 6.25 mL MultiScribe Reverse Transcriptase, 2 mL RNase inhibitor, 22 mL MgCl2, 20 mL dNTPs, in an ABI PRISM 7700 Cycler, under the following conditions: 25˚C for 10 minutes; 48˚C for 60 minutes; 95˚C for 5 minutes. TaqMan Universal Master Mix and 5 mL of complementary DNA were used for quantitative real-time PCR in a StepOnePlus Thermocycler, in gene-specific TaqMan Assays. We investigated the proinflammatory cytokine TNF-a and NGF. The 25 mL reaction mix contained 12.5 mL TaqMan Master Mix and 1.25 mL primer. The cycler conditions were as follows: 50˚C for 2 minutes; 95˚C for 10 minutes; 45 cycles of 95˚C for 15 seconds, and 60˚C for 1 minute. Each plate contained a negative control and a blood sample from the control whose threshold cycle (Ct) values were nearest to the calculated mean of all control blood samples for each primer, as a calibrator for blood determinations. Samples were tested in triplicate, with the exception of 18s RNA levels that were determined in duplicate. All plates were analyzed under identical conditions.
2.4.4.2. Gene expression analysis The skin sample for gene expression analysis was flash-frozen in liquid nitrogen and stored at 280˚C until processing. Skin samples were then thawed on ice, immersed in 1 mL TRIzol
2.5. Statistical analysis Because of the small sample size (ie, ,20), nonparametric tests were used to compare patients with and without pain. Spearman
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
November 2015
·
Volume 156
·
Number 11
rank correlation tests were used for correlations; continuous variables were compared in Mann–Whitney U tests. Values are presented as median (range). Values of P , 0.05 were considered significant. Data were analyzed with SPSS version 18 software (SPSS, Chicago, IL).
3. Results 3.1. Characteristics of the study population
www.painjournalonline.com
2393
detection: r 5 0.64, P 5 0.001; cold detection: r 5 20.57, P 5 0.001; mechanical detection: r 5 0.64, P 5 0.001). Finally, 32 (72.8%) of the 44 patients examined at M3 had hypesthesia. Of these patients, 13 (28%) developed CPSNP (Fig. 2). We analyzed these patients further, to identify predictors of neuropathic pain. 3.3. Comparisons between patients with and without chronic postsurgical neuropathic pain in the presence of hypesthesia
Forty-six patients and 45 controls were included. The clinical characteristics of the patients are summarized in Table 1. One patient died and one was lost to follow-up during the study. Three patients refused the second biopsy at 3 months (unsightly scar). Quantitative sensory testing was performed at all 3 time points (baseline, D2, and M3), for all patients. A skin punch biopsy was performed at D2 for 44 patients and at M3 for 40 patients. Minor complications included 1 case of local bleeding requiring sutures, 1 case of persistent hypesthesia, 1 vasovagal attack, and 1 case of delayed wound healing (ie, .2 weeks). Each skin punch biopsy was divided in 2 parts and had independent transport from the harvest to the site of analysis. Six (2.7 %) biopsies for IENFD assessment and fourteen (6.3%) biopsies for mRNA cytokine assessment were lost during transport. Twenty-six (11.7%) biopsies for IENFD assessment and thirtyfive (15.8%) biopsies for mRNA cytokine assessment were not analyzed because of insufficient quality of tissue sample, conservation, or analysis problems. Finally, 189 biopsies were available for IENFD and 182 for mRNA cytokine assessment. When combining all these issues, the biopsy was analyzable for 44 patients at D2 and for 35 patients at M3, and a complete set of data (QST 1 IENFD 1 TNF-a 1 NGF) was obtained for 26 patients at D2 and 25 patients at M3.
Quantitative sensory testing showed hypesthesia for mechanical and heat detection and an increase in pain thresholds ipsilateral to ICBH, on day 2, in both groups (with and without CPSNP). This sensory deficit persisted at M3. However, no differences in QST results were observed when comparing patients with and without CPSNP at M3 (Table 2). Intraepidermal nerve fiber density did not differ between patients (or between either group of patients) and controls at D2. At M3, IENFD on the side ipsilateral to ICBH was lower than that on the contralateral side in patients. However, IENFD did not differ between patients with and without CPSNP, at D2 and M3 (Table 2 and Fig. 3). Cutaneous TNF-a and NGF gene expression did not differ between the ipsilateral and contralateral skin samples of patients with and without CPSNP. The TNF-a gene expression of these 2 patient groups did not differ from that of healthy controls at D2 and M3 (Table 2). At M3, levels of NGF gene expression in the skin were higher in both patient groups (with and without CPSNP) than in healthy controls (P 5 0.003). In patients with CPSNP, levels of NGF gene expression in the skin at M3 were higher on the ipsilateral side than on the contralateral side (P 5 0.04).
3.2. Mechanical and thermal detection thresholds and intraepidermal nerve fiber density after surgery
At M3, eleven patients had developed pain in the territory of the lateral femoral cutaneous nerve and 2 had developed pain in the territory of the iliohypogastric nerve. Three patients had severe pain (NRS $ 6), and 10 had moderate pain (NRS 3-6). The highest average pain intensity was 4.7 6 1.9 (range: 3-8). Total NPSI score was 12.25 6 10.3 (range: 3-36). Intraepidermal nerve fiber density at M3 was positively correlated with several clinical symptoms of neuropathic pain evaluated by the NPSI score: burning (r 5 0.68, P 5 0.02, Fig. 4), compression (r 5 0.76, P 5 0.006), and pain provoked or increased by brushing the painful area (r 5 0.67, P 5 0.02). No correlation was found between skin
Quantitative sensory testing revealed hypesthesia for mechanical, heat, and cold detection in 73.9%, 65.9%, and 59.1% of patients, respectively, at D2, and in 66.7%, 70.7%, and 61% of patients, respectively, at M3. The results obtained at D2 and M3 were strongly correlated for mechanical (r 5 0.875, P 5 0.001) and heat detection thresholds (r 5 0.87, P 5 0.001) and strongly negatively correlated for cold detection thresholds (r 5 20.85, P 5 0.001). Mechanical and heat detection thresholds were positively correlated (r 5 0.55, P 5 0.001), as were mechanical detection and heat pain thresholds (r 5 0.43, P 5 0.003) and mechanical pain and heat detection (r 5 0.58, P 5 0.001) and heat pain (r 5 0.6, P 5 0.001) thresholds. A negative correlation was found between mechanical and cold detection thresholds (r 5 20.46, P 5 0.001) and between mechanical pain and cold detection (r 5 20.32, P 5 0.02) and cold pain (r 5 20.39, P 5 0.04) thresholds. Intraepidermal nerve fiber density was low in 11.4% (4/35) of patients at D2 and in 60.6% (13/33) of patients at M3. Intraepidermal nerve fiber density was correlated with changes in the response to thermal stimulation. Thus, lower fiber densities were found to be associated with higher thresholds. This correlation was observed for all thermal thresholds (ie, heat detection: r 5 20.64, P 5 0.001; heat pain: r 5 20.42, P 5 0.01; cold detection: r 5 0.70, P 5 0.001; cold pain: r 5 0.39, P 5 0.02). The difference in IENFD between D2 and M3 was correlated with the detection threshold at D2 (heat detection: r 5 20.63, P 5 0.001; cold detection: r 5 20.58, P 5 0.001; mechanical detection: r 5 0.70, P 5 0.001) and at M3 (heat
3.4. Description of chronic postsurgical neuropathic pain at M3
Figure 2. Study flowchart. CPSNP, chronic postsurgical neuropathic pain; CPSP, chronic postsurgical pain.
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
2394
·
PAIN®
V. Martinez et al. 156 (2015) 2390–2398
Table 2
Comparison between patients with and without CPSNP in the presence of hypesthesia, at 3 months. CPSNP (N 5 13) Day 2 QST at D2 Mechanical detection threshold (log mg) Mechanical pain threshold (log mg) Heat detection threshold (˚C) Heat pain threshold (˚C) Cold detection threshold (˚C) Cold pain threshold (˚C) IENFD at D2 (fibers per mm3) mRNA gene expression at D2 TNF-a NGF 3 mo QST at M3 Mechanical detection threshold (log mg) Mechanical pain threshold (log mg) Heat detection threshold (˚C) Heat pain threshold (˚C) Cold detection threshold (˚C) Cold pain threshold (˚C) IENFD at M3 (fibers per mm3) mRNA gene expression at M3 TNF-a NGF
No CPSNP (N 5 19)
Control
Ipsilateral
Contralateral
Ipsilateral
Contralateral
Volunteers
Preoperative values
4.9 (0.9)*†‡ 5.3 (0.4)*‡ 47 (5.6)*†‡ 48.3 (3.1)*†‡ 12.8 (7.9)*†‡ 9.1 (1)‡ 7.4 (4)
2.4 (0.6) 4.4 (0.6) 34.8 (1.7)† 43 (2.7) 27.2 (4.7)† 13 (6) 8.6 (4.1)
4.7 (1)*†‡ 5.3 (0.2)*‡ 47 (6)*†‡ 48.8 (2.2)*†‡ 16.6 (8.4)*†‡ 10.9 (4)*† 8.5 (5.7)
2.03 (0.6) 4.5 (0.6) 36 (1.7) 42.4 (3.8) 27 (5) 15.6 (6.5) 10.7 (4.3)
2.32 (0.5) 3.8 (0.4) 33.4 (4.8) 43.3 (3.6) 29.5 (1.5) 11.9 (8) 9.8 (4)
2 (0.5) 4.25 (0.6) 35.4 (1.9) 44.3 (3) 30.1 (0.9) 15.1 (7) Nd
1.1 (0.5) 1.22 (0.5)
1 (1.2) 1.42 (0.4)
1.9 (1.7) 1.53 (0.5)†
1.28 (0.6) 1.25 (0.3)
1.35 (0.67) 1.15 (0.38)
Nd Nd
4.7 (0.9)*†‡ 5.4 (0.3)*†‡ 44.9 (5.4)*†‡ 48.0 (2.9)*†‡ 15.9 (8.2)*‡ 10 (0)*‡ 2.2 (3.8)*†‖
2.2 (0.7) 4.5 (0.7)† 34.3 (1.1) 42 (2.6) 30 (1.5) 16 (7) 9.2 (3)
4.6 (0.9)*†‡ 5.2 (0.4)*†‡ 45.2 (4.6)*†‡ 48.2 (2.3)*†‡ 19.9 (7.7)*†‡ 12.4 (6) 2.9 (5.5)*†‖
2.1 (0.6) 4.5 (0.6) 35 (1.5) 43.1 (3.6) 29.3 (1.5) 14.1 (7.4) 9.1 (3.4)
4.8 (2.5) 3.8 (4) 33.4 (4.8) 43.3 (3.6) 29.5 (1.5) 11.9 (8) 9.8 (4)
2 (0.5) 4.25 (0.6) 35.4 (1.9) 44.3 (3) 30.1 (0.9) 15.1 (7) Nd
0.99 (0.6) 1.44 (0.5)*
0.77 (0.4) 0.9 (0.5)
1.22 (0.6) 1.47 (0.7)†
0.9 (0.5) 1.1 (0.5)
1.35 (0.67) 1.15 (0.3)
Nd Nd
Continuous variables are presented as mean (SD). No difference was observed between CPSNP and no CPSNP groups. * P , 0.05 vs contralateral. † P , 0.05 vs volunteers. ‡ P , 0.05 vs preoperative values. ‖ P , 0.05 vs M3 values. CPSNP, chronic postsurgical neuropathic pain; IENFD, intraepidermal nerve fiber density; Nd, not determined; QST, quantitative sensory testing.
TNF-a and NGF gene expression and clinical symptoms or IENFD. Interestingly, patients developing CPSNP had higher preoperative pain ratings at rest, at the principal site of orthopedic surgery, than patients without CPSNP (median score of 3 on an 11-point NRS, range: 0-10, vs a median score of 0, range: 0-6, P 5 0.0001) and used larger amounts of preoperative opioids daily dose (median of 10 mg of morphine equivalent, range: 0-60, vs median of 3 mg of morphine equivalent, range: 0-10, P 5 0.02). Patients developing CPSNP had also higher pain ratings in recovery room (median score of 4.4 on an 11-point NRS, range: 2-7.5, vs a median score of 2.2, range: 0-7, P 5 0.003).
Figure 3. Correlation between IENFD and intensity of burning pain sensations at 3 months, in patients with CPSNP. CPSNP, chronic postsurgical neuropathic pain; IENFD, intraepidermal nerve fiber density; NRS, numerical rating scale.
4. Discussion 4.1. Main findings Neither IENFD nor changes in the expression of the TNF-a and NGF genes in the skin were associated with the development of CPSNP in patients with hypesthesia after ICBH. By contrast, high preoperative pain intensity and opioid use were associated with CPSNP. In patients with pain, CPSNP severity was inversely correlated with small fiber loss after ICBH.
4.2. Type of nerve lesion after surgery Nerve lesions are frequent after ICBH; hypesthesia and a decrease in IENFD are observed in 72.8% and 60% of patients, respectively. All fibers are affected by the nerve lesion. Detection and pain thresholds for both mechanical and thermal stimuli exploring large myelinated fibers, small myelinated Ad fibers, and unmyelinated C fibers were found to have increased at D2 and were almost stable at M3. The decrease in IENFD was observed only at M3 and was correlated with the changes in the response to thermal stimulation observed at M3. This delay in the appearance of decreases in nerve fiber density has already been reported in an animal nerve entrapment model.24 The correlation between functional impairment and a decrease in IENFD has also been observed in other neuropathies.16,23,48,59,61 We found that early impairment of nerve function, as revealed by QST at D2, was associated with a decrease in IENFD reduction at M3 and nerve lesion severity. Surgical trauma seems to induce homogeneous nerve fiber damage. We therefore suggest that a mechanical evaluation of hypesthesia might be sufficient to predict the
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
November 2015
·
Volume 156
·
Number 11
www.painjournalonline.com
2395
Figure 4. Comparison of IENFD 2 days and 3 months after ICBH. CPSNP, chronic postsurgical neuropathic pain; IENFD, intraepidermal nerve fiber density; ICBH, iliac crest bone harvest; no CPSNP, no chronic postsurgical neuropathic pain (mean 6 SD).
severity of the nerve lesion. This postsurgical nerve lesion model clearly differs from other degenerative neuropathies in which progressive small nerve fiber degeneration is the main mechanism. In conclusion, our data suggest a homogeneous lesion of nerve fiber subgroups after ICBH, involving large myelinated fibers, small myelinated Ad fibers, and unmyelinated C fibers. Mechanical sensory testing can be used for early detection and quantification of nerve lesions, at D2, whereas histological analysis can be used to confirm the lesion at M3. 4.3. Nerve lesions are not predictive of neuropathic pain after iliac crest bone harvest Although 60% to 73% of patients had signs of a nerve lesion, such as hypesthesia or a decrease in IENFD, only 40% of these patients developed CPSNP. The severity of the nerve lesion, as indicated by the intensity of hypesthesia or the magnitude of the decrease in IENFD, was not predictive of CPSNP. These findings confirm previous results in different surgical settings, in which no specific QST profile was identified for patients with CPSNP.1,21,39 In addition, our results indicate that IENFD is not predictive of CPSNP either. 4.4. Partial nerve lesion leads to more severe chronic postsurgical neuropathic pain Partial nerve lesion was associated with more severe CPSNP. Indeed, IENFD at M3 was positively correlated with pain intensity at M3 and with qualitative items of spontaneous pain, such as burning sensations. Our results are different from those obtained in Guillain–Barre´ syndrome,46 sarcoidosis,6 HIV, or diabetic neuropathies, in which inverse correlations with nerve fiber density have been reported.53,59,65 However, similar results have been reported in patients with postherpetic neuralgia, in which the decrease in the number of unmyelinated nerve fibers in the skin has been found to be inversely correlated with allodynia severity, suggesting that the presence of intact unmyelinated fibers is important for the induction of allodynia.29 These results
oppose the progressive degeneration of small nerves with a single-incident surgical lesion of small and large fibers. Our results are consistent with the hypothesis that neuroactive substances released along the length of the degenerating fibers of the partial lesion can activate uninjured fibers, generating neuropathic pain.47,51,52,62 This mechanism has been tested in animal models, using entrapment, stretching, or constriction of the sciatic nerve to develop partial nerve lesions and related neuropathic pain behavior in experimental animals; pure nerve transection does not always lead to similar symptoms.5,10,45 Our findings are consistent with those of clinical studies in other surgical models. In inguinal hernia repair, a combined analysis of 3 studies suggested that neurectomy less frequently leads to CPSNP than nerve preservation.3,27 In mandibular osteotomy, electrophysiological findings indicated that axotomy was less predictive of CPSNP than partial axonal lesions.25 These results suggest that partial nerve damage is associated with a higher risk of severe CPSNP than total nerve section. 4.5. Skin TNF-a gene expression During nerve degeneration, proinflammatory cytokines are produced by various cells, including the Schwann cells.29,47 Our results for patients undergoing ICBH do not confirm those of previous clinical studies suggesting an increase in the expression of proinflammatory cytokine genes after trauma30 or in peripheral neuropathies.55 Such an increase in proinflammatory cytokine expression has been observed at various sites, including the blood,56 cerebrospinal fluid, and the sural nerve.35 This increase has been confirmed by different techniques measuring cytokine expression directly by immunochemical methods or by assessing gene expression. An increase in TNFa expression has been reported in various peripheral neuropathies17,35 and in complex regional pain syndrome.34 An increase in TNF-a expression has also been observed bilaterally34 and can persist for months.30 Such an increase has been reported in painful neuropathies but not in painless neuropathies.17,56 We found no increase in skin TNF-a gene expression in
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
2396
·
PAIN®
V. Martinez et al. 156 (2015) 2390–2398
the acute and chronic postoperative phases. We also found no differences in the expression of this gene between the 2 sides of the body or between patients with and without pain. These findings are similar to those reported for patients with postherpetic neuralgia57 and fibromyalgia syndrome.54 There are several possible reasons for the lack of change in TNF-a gene expression observed here. These reasons include the much shorter observation period (only 3 months) in our study than in previous studies on cutaneous TNF-a expression, in which follow-up continued over a period of several years. Differences in the physiopathology of pain after surgery (ie, acute cutting) and in diseases with a chronic pattern of development (ie, neuropathies) must also be taken out, although proinflammatory cytokines have been implicated in different types of traumatic and inflammatory neuropathies.15 Methodological limitations, such as the small sample size and unstandardized timing of skin punch biopsies, may also have influenced the results. Finally, our study focused on TNF-a, and we cannot rule out the possibility of changes in the expression of genes encoding other proinflammatory and anti-inflammatory cytokines. 4.6. Skin NGF gene expression Neurotrophic factors are crucial for nerve integrity and several studies have shown that NGF levels, for example, are high in nerve specimens from patients with peripheral neuropathies.63 Cutaneous NGF expression has been investigated in only a few studies to date. Levels of NGF expression are low in the skin samples of patients with diabetic neuropathy or neuropathies of other etiologies,4,55 and these previous studies found no correlation between NGF expression in patients with and without pain. Our findings conflict with these previous reports. We observed an increase in NGF expression at 3 months after surgery, and NGF gene expression was not associated with the development of CPSNP in patients with hypesthesia. These findings are consistent with those of animal studies showing an increase in skin NGF gene expression after chronic constriction injury in rats.42 It is possible that excessive cutaneous NGF expression, together with a decrease in innervation density, causes nociceptor sensitization through phenotypic changes, eventually leading to chronic pain.18 We observed no difference in NGF gene expression in the skin between patients with and without hypesthesia. Our results therefore suggest that NGF is not suitable for use as a marker of nerve lesions (results not shown). 4.7. Clinical predictors of the development of chronic postsurgical neuropathic pain Interestingly, although psychophysical, histological, and molecular factors were not associated with the development of CPSNP, we observed that, in cases of nerve lesion after ICBH, the severity of preoperative at the surgery site and the use of higher doses of opioids before surgery were associated with the development of CPSNP. We also found that the severity of early pain was also predictive of chronic postsurgical pain (CPSP) as previously reported in our surgical model.37 The presence and duration of preoperative pain is predictive of CPSP in various surgical models.28 Both pain at the site of surgery8,31,40,43,44 and pain at a distant location have been identified as predictive factors for CPSP.13,20,44,64 As previously reported for spinal surgery, we observed that the intensity of preoperative pain was associated with a poor
outcome in surgical treatment for lumbar radiculopathy.60 We suggest that preoperative nervous system sensitization may increase the risk of developing CPSP after surgery. The use of higher doses of opioids before surgery was also associated with CPSP. It is unclear whether this association is due to the severity of preoperative pain or the potential impact of opioids on this sensitization. There are experimental data suggesting that higher doses of opioids may lead to hyperalgesia and a potential increase in acute postsurgical pain, leading to CPSP.49 One clinical study has reported that the preoperative use of opioids might modify the postoperative pain trajectory, magnifying the impact of preoperative pain on the incidence of CPSP after gynecologic surgery.58 Our results support the notion that preoperative clinical factors may favor not only CPSP but also, more specifically, the development of neuropathic pain after a nerve lesion.
Conflict of interest statement The authors have no conflicts of interest to declare. This study was supported by a grant from the Institut UPSA de la Douleur, the APICIL Foundation, and CNP Assurances. Article history: Received 8 May 2015 Received in revised form 24 June 2015 Accepted 25 June 2015 Available online 7 July 2015
References [1] Aasvang EK, Brandsborg B, Christensen B, Jensen TS, Kehlet H. Neurophysiological characterization of postherniotomy pain. PAIN 2008; 137:173–81. [2] Aasvang E, Kehlet H. Chronic postoperative pain: the case of inguinal herniorrhaphy. Br J Anaesth 2005;95:69–76. [3] Alfieri S, Amid PK, Campanelli G, Izard G, Kehlet H, Wijsmuller AR, Di Miceli D, Doglietto GB. International guidelines for prevention and management of post-operative chronic pain following inguinal hernia surgery. Hernia 2011;15:239–49. [4] Anand P, Terenghi G, Warner G, Kopelman P, Williams-Chestnut RE, Sinicropi DV. The role of endogenous nerve growth factor in human diabetic neuropathy. Nat Med 1996;2:703–7. [5] Attal N, Filliatreau G, Perrot S, Jazat F, Di Giamberardino L, Guilbaud G. Behavioural pain-related disorders and contribution of the saphenous nerve in crush and chronic constriction injury of the rat sciatic nerve. PAIN 1994;59:301–12. [6] Bakkers M, Merkies IS, Lauria G, Devigili G, Penza P, Lombardi R, Hermans MC, van Nes SI, De Baets M, Faber CG. Intraepidermal nerve fiber density and its application in sarcoidosis. Neurology 2009;73:1142–8. [7] Basbaum AI, Gautron M, Jazat F, Mayes M, Guilbaud G. The spectrum of fiber loss in a model of neuropathic pain in the rat: an electron microscopic study. PAIN 1991;47:359–67. [8] Bates T, Ebbs SR, Harrison M, A’Hern RP. Influence of cholecystectomy on symptoms. Br J Surg 1991;78:964–7. [9] Benedetti F, Amanzio M, Casadio C, Filosso PL, Molinatti M, Oliaro A, Pischedda F, Maggi G. Postoperative pain and superficial abdominal reflexes after posterolateral thoracotomy. Ann Thorac Surg 1997;64:207–10. [10] Bennett GJ, Xie YK. A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. PAIN 1988;33: 87–107. [11] Bouhassira D, Attal N, Alchaar H, Boureau F, Brochet B, Bruxelle J, Cunin G, Fermanian J, Ginies P, Grun-Overdyking A, Jafari-Schluep H, LanteriMinet M, Laurent B, Mick G, Serrie A, Valade D, Vicaut E. Comparison of pain syndromes associated with nervous or somatic lesions and development of a new neuropathic pain diagnostic questionnaire (DN4). PAIN 2005;114:29–36. [12] Bouhassira D, Attal N, Fermanian J, Alchaar H, Gautron M, Masquelier E, Rostaing S, Lanteri-Minet M, Collin E, Grisart J, Boureau F. Development and validation of the Neuropathic Pain Symptom Inventory. PAIN 2004; 108:248–57.
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
November 2015
·
Volume 156
·
Number 11
[13] Brandsborg B, Nikolajsen L, Hansen CT, Kehlet H, Jensen TS. Risk factors for chronic pain after hysterectomy: a nationwide questionnaire and database study. Anesthesiology 2007;106:1003–12. [14] Calvo M, Dawes JM, Bennett DL. The role of the immune system in the generation of neuropathic pain. Lancet Neurol 2012;11:629–42. [15] Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain Inventory. Ann Acad Med Singapore 1994;23:129–38. [16] Devigili G, Tugnoli V, Penza P, Camozzi F, Lombardi R, Melli G, Broglio L, Granieri E, Lauria G. The diagnostic criteria for small fibre neuropathy: from symptoms to neuropathology. Brain 2008;131: 1912–25. [17] Empl M, Renaud S, Erne B, Fuhr P, Straube A, Schaeren-Wiemers N, Steck AJ. TNF-alpha expression in painful and nonpainful neuropathies. Neurology 2001;56:1371–7. [18] Evans LJ, Loescher AR, Boissonade FM, Whawell SA, Robinson PP, Andrew D. Temporal mismatch between pain behaviour, skin nerve growth factor and intra-epidermal nerve fibre density in trigeminal neuropathic pain. BMC Neurosci 2014;15:1. [19] Fruhstorfer H, Lindblom U, Schmidt WC. Method for quantitative estimation of thermal thresholds in patients. J Neurol Neurosurg Psychiatry 1976;39:1071–5. [20] Gartner R, Jensen MB, Nielsen J, Ewertz M, Kroman N, Kehlet H. Prevalence of and factors associated with persistent pain following breast cancer surgery. JAMA 2009;302:1985–92. [21] Gottrup H, Andersen J, Arendt-Nielsen L, Jensen TS. Psychophysical examination in patients with post-mastectomy pain. PAIN 2000;87: 275–84. [22] Guilbaud G, Gautron M, Jazat F, Ratinahirana H, Hassig R, Hauw JJ. Time course of degeneration and regeneration of myelinated nerve fibres following chronic loose ligatures of the rat sciatic nerve: can nerve lesions be linked to the abnormal pain-related behaviours? PAIN 1993;53: 147–58. [23] Herrmann DN, McDermott MP, Sowden JE, Henderson D, Messing S, Cruttenden K, Schifitto G. Is skin biopsy a predictor of transition to symptomatic HIV neuropathy? A longitudinal study. Neurology 2006;66: 857–61. [24] Hsieh CH, Jeng SF, Lu TH, Chen YC, Hsieh MW, Chen SS. Loss of small fibers in entrapment neuropathy and their regeneration after surgical decompression in a rat model. J Neurotrauma 2007;24:1658–66. [25] Jaaskelainen SK, Teerijoki-Oksa T, Forssell H. Neurophysiologic and quantitative sensory testing in the diagnosis of trigeminal neuropathy and neuropathic pain. PAIN 2005;117:349–57. [26] Jaaskelainen SK, Teerijoki-Oksa T, Virtanen A, Tenovuo O, Forssell H. Sensory regeneration following intraoperatively verified trigeminal nerve injury. Neurology 2004;62:1951–7. [27] Johner A, Faulds J, Wiseman SM. Planned ilioinguinal nerve excision for prevention of chronic pain after inguinal hernia repair: a meta-analysis. Surgery 2011;150:534–41. [28] Kehlet H, Jensen TS, Woolf CJ. Persistent postsurgical pain: risk factors and prevention. Lancet 2006;367:1618–25. [29] Koike H, Iijima M, Mori K, Yamamoto M, Hattori N, Watanabe H, Tanaka F, Doyu M, Sobue G. Neuropathic pain correlates with myelinated fibre loss and cytokine profile in POEMS syndrome. J Neurol Neurosurg Psychiatry 2008;79:1171–9. [30] Kramer HH, Eberle T, Uceyler N, Wagner I, Klonschinsky T, Muller LP, Sommer C, Birklein F. TNF-alpha in CRPS and “normal” trauma—significant differences between tissue and serum. PAIN 2011; 152:285–90. [31] Kroner K, Krebs B, Skov J, Jorgensen HS. Immediate and long-term phantom breast syndrome after mastectomy: incidence, clinical characteristics and relationship to pre-mastectomy breast pain. PAIN 1989;36:327–34. [32] Lauria G, Cornblath DR, Johansson O, McArthur JC, Mellgren SI, Nolano M, Rosenberg N, Sommer C. EFNS guidelines on the use of skin biopsy in the diagnosis of peripheral neuropathy. Eur J Neurol 2005;12: 747–58. [33] Lauria G, Hsieh ST, Johansson O, Kennedy WR, Leger JM, Mellgren SI, Nolano M, Merkies IS, Polydefkis M, Smith AG, Sommer C, Valls-Sole J. European Federation of Neurological Societies/Peripheral Nerve Society Guideline on the use of skin biopsy in the diagnosis of small fiber neuropathy. Report of a joint task force of the European Federation of Neurological Societies and the Peripheral Nerve Society. Eur J Neurol 2010;17:903–12, e44–9. [34] Lenz M, Uceyler N, Frettloh J, Hoffken O, Krumova EK, Lissek S, Reinersmann A, Sommer C, Stude P, Waaga-Gasser AM, Tegenthoff M, Maier C. Local cytokine changes in complex regional pain syndrome type I (CRPS I) resolve after 6 months. PAIN 2013;154: 2142–9.
www.painjournalonline.com
2397
[35] Lindenlaub T, Sommer C. Cytokines in sural nerve biopsies from inflammatory and non-inflammatory neuropathies. Acta neuropathologica 2003;105:593–602. [36] Macrae WA. Chronic post-surgical pain: 10 years on. Br J Anaesth 2008; 101:77–86. [37] Martinez V, Ben Ammar S, Judet T, Bouhassira D, Chauvin M, Fletcher D. Risk factors predictive of chronic postsurgical neuropathic pain: the value of the iliac crest bone harvest model. PAIN 2012;153:1478–83. [38] McCarthy BG, Hsieh ST, Stocks A, Hauer P, Macko C, Cornblath DR, Griffin JW, McArthur JC. Cutaneous innervation in sensory neuropathies: evaluation by skin biopsy. Neurology 1995;45:1848–55. [39] Mikkelsen T, Werner MU, Lassen B, Kehlet H. Pain and sensory dysfunction 6 to 12 months after inguinal herniotomy. Anesth Analg 2004;99:146–51. [40] Nikolajsen L, Ilkjaer S, Kroner K, Christensen JH, Jensen TS. The influence of preamputation pain on postamputation stump and phantom pain. PAIN 1997;72:393–405. [41] Owens WD, Felts JA, Spitznagel EL Jr. ASA physical status classifications: a study of consistency of ratings. Anesthesiology 1978; 49:239–43. [42] Peleshok JC, Ribeiro-da-Silva A. Neurotrophic factor changes in the rat thick skin following chronic constriction injury of the sciatic nerve. Mol Pain 2012;8:1. [43] Perttunen K, Tasmuth T, Kalso E. Chronic pain after thoracic surgery: a follow-up study. Acta Anaesthesiol Scand 1999;43:563–7. [44] Puolakka PA, Rorarius MG, Roviola M, Puolakka TJ, Nordhausen K, Lindgren L. Persistent pain following knee arthroplasty. Eur J Anaesthesiol 2010;27:455–60. [45] Robinson LR. Traumatic injury to peripheral nerves. Muscle Nerve 2000; 23:863–73. [46] Ruts L, van Doorn PA, Lombardi R, Haasdijk ED, Penza P, Tulen JH, Hempel RJ, van den Meiracker AH, Lauria G. Unmyelinated and myelinated skin nerve damage in Guillain-Barre syndrome: correlation with pain and recovery. PAIN 2012;153:399–409. [47] Schafers M, Geis C, Svensson CI, Luo ZD, Sommer C. Selective increase of tumour necrosis factor-alpha in injured and spared myelinated primary afferents after chronic constrictive injury of rat sciatic nerve. Eur J Neurosci 2003;17:791–804. [48] Schuning J, Scherens A, Haussleiter IS, Schwenkreis P, Krumova EK, Richter H, Maier C. Sensory changes and loss of intraepidermal nerve fibers in painful unilateral nerve injury. Clin J Pain 2009;25:683–90. [49] Simonnet G. Acute tolerance to opioids: methodological, theoretical and clinical implications. PAIN 2009;142:3–4. [50] Sommer C. Skin biopsy as a diagnostic tool. Curr Opin Neurol 2008;21: 563–8. [51] Sommer C, Galbraith JA, Heckman HM, Myers RR. Pathology of experimental compression neuropathy producing hyperesthesia. J Neuropathol Exp Neurol 1993;52:223–33. [52] Sommer C, Lalonde A, Heckman HM, Rodriguez M, Myers RR. Quantitative neuropathology of a focal nerve injury causing hyperalgesia. J Neuropathol Exp Neurol 1995;54:635–43. [53] Sorensen L, Molyneaux L, Yue DK. The relationship among pain, sensory loss, and small nerve fibers in diabetes. Diabetes Care 2006;29:883–7. [54] Uceyler N, Kewenig S, Kafke W, Kittel-Schneider S, Sommer C. Skin cytokine expression in patients with fibromyalgia syndrome is not different from controls. BMC Neurol 2014;14:185. [55] Uceyler N, Riediger N, Kafke W, Sommer C. Differential gene expression of cytokines and neurotrophic factors in nerve and skin of patients with peripheral neuropathies. J Neurol 2015;262:203–12. [56] Uceyler N, Rogausch JP, Toyka KV, Sommer C. Differential expression of cytokines in painful and painless neuropathies. Neurology 2007;69:42–9. [57] Uceyler N, Valet M, Kafke W, Tolle TR, Sommer C. Local and systemic cytokine expression in patients with postherpetic neuralgia. PLoS One 2014;9:e105269. [58] VanDenKerkhof EG, Hopman WM, Goldstein DH, Wilson RA, Towheed TE, Lam M, Harrison MB, Reitsma ML, Johnston SL, Medd JD, Gilron I. Impact of perioperative pain intensity, pain qualities, and opioid use on chronic pain after surgery: a prospective cohort study. Reg Anesth Pain Med 2012;37:19–27. [59] Vlckova-Moravcova E, Bednarik J, Dusek L, Toyka KV, Sommer C. Diagnostic validity of epidermal nerve fiber densities in painful sensory neuropathies. Muscle Nerve 2008;37:50–60. [60] Voorhies RM, Jiang X, Thomas N. Predicting outcome in the surgical treatment of lumbar radiculopathy using the Pain Drawing Score, McGill Short Form Pain Questionnaire, and risk factors including psychosocial issues and axial joint pain. Spine J 2007;7:516–24. [61] Walk D, Wendelschafer-Crabb G, Davey C, Kennedy WR. Concordance between epidermal nerve fiber density and sensory examination in
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
2398
·
V. Martinez et al. 156 (2015) 2390–2398
patients with symptoms of idiopathic small fiber neuropathy. J Neurol Sci 2007;255:23–6. [62] Watkins LR, Maier SF. Beyond neurons: evidence that immune and glial cells contribute to pathological pain states. Physiol Rev 2002;82: 981–1011. [63] Yamamoto M, Ito Y, Mitsuma N, Hattori N, Sobue G. Pain-related differential expression of NGF, GDNF, IL-6, and their receptors in human vasculitic neuropathies. Intern Med 2003;42:1100–3.
PAIN®
[64] Yunker AC, Ritch JM, Robinson EF, Golish CT. Incidence and risk factors for chronic pelvic pain after hysteroscopic sterilization. J Minim Invasive Gynecol 2015;22:390–4. [65] Zhou L, Kitch DW, Evans SR, Hauer P, Raman S, Ebenezer GJ, Gerschenson M, Marra CM, Valcour V, Diaz-Arrastia R, Goodkin K, Millar L, Shriver S, Asmuth DM, Clifford DB, Simpson DM, McArthur JC. Correlates of epidermal nerve fiber densities in HIV-associated distal sensory polyneuropathy. Neurology 2007;68:2113–19.
Copyright Ó 2015 by the International Association for the Study of Pain. Unauthorized reproduction of this article is prohibited.
