ORIGINAL ARTICLE

Facilitation of pain sensitization in knee osteoarthritis and persistent post-operative pain: A cross-sectional study S.T. Skou1,2, T. Graven-Nielsen1, S. Rasmussen1,2, O.H. Simonsen2, M.B. Laursen2, L. Arendt-Nielsen1 1 Center for Sensory-Motor Interaction (SMI), Department of Health Science and Technology, Faculty of Medicine, Aalborg University, Denmark 2 Orthopaedic Surgery Research Unit, Aalborg University Hospital, Denmark

Correspondence Lars Arendt-Nielsen E-mail: [email protected] Funding sources This study is supported by the Danish Rheumatism Association, the Danish National Advanced Technology Foundation, the Aase and Ejnar Danielsen’s Foundation, the Lions Club Denmark, and the Danish Council for Technology and Innovation (09-052174). None of the funders have any role in the study other than to provide funding. Conflicts of interest None declared. Accepted for publication 22 November 2013 doi:10.1002/j.1532-2149.2013.00447.x

Abstract Background: Around 20% of patients with osteoarthritis (OA) have chronic post-operative pain after total knee arthroplasty (TKA) and often undergo revision surgery with unfavourable pain outcome. This study compared sensitization in pain patients with knee OA and after revision TKA (re-TKA). Methods: Median pressure pain thresholds (PPTs) assessed from the most affected knee (localized sensitization) were used to subgroup 53 patients with OA pain and 20 patients with pain after re-TKA: group 1: OA and high-knee PPT; group 2: OA and low-knee PPT; group 3: re-TKA and high-knee PPT; group 4: re-TKA and low-knee PPT. Clinical pain intensity was assessed using a visual analogue scale (VAS). Bilateral PPTs were measured from the lower leg and forearm (spreading sensitization). Furthermore, the pain intensities evoked by 10 repeated pressure pain stimuli (temporal summation) at the knee and lower leg were assessed on an electronic VAS. Results: The mean clinical pain intensity was not significantly different between groups. The PPTs from both lower leg and forearm were significantly lower in group 4 compared to groups 1, 2, and 3 and in groups 2 and 3 compared to group 1 (p < 0.05). Temporal summations from the knee and lower leg were significantly facilitated in groups 3 and 4 compared to groups 1 and 2 (p < 0.05). Conclusions: Despite similar pain intensities, facilitated temporal summation is worse in re-TKA than in OA and patients with high local knee hyperalgesia show more prominent spreading sensitization. The study suggests that sensitization should be considered in knee OA especially before re-TKA.

1. Introduction Despite an improved understanding of pain over the past decades, the pathophysiology of osteoarthritis (OA) pain still remains poorly understood (Gwilym et al., 2008). The radiographic severity of OA is only weakly associated with the clinical severity and it is a puzzle why minor joint injuries can cause massive pain and vice versa (Davis et al., 1992; Dieppe and Lohmander, 2005; Felson, 2005). Moreover, around 20% of patients receiving a total knee arthroplasty 1024 Eur J Pain 18 (2014) 1024–1031

(TKA) experience chronic post-operative pain (Beswick et al., 2012), even though the procedure is considered effective (Carr et al., 2012). The deficient understanding of persistent post-operative pain may be due to the complexity of OA pain. Targeting one mechanism may not improve symptoms if other mechanisms are co-responsible for the pain (Suokas et al., 2012). Sensitization is a prominent mechanism in OA pain (Arendt-Nielsen et al., 2010; Arendt-Nielsen and Graven-Nielsen, 2011) and factors outside the joint © 2013 European Pain Federation - EFIC®

S.T. Skou et al.

What’s already known about this topic? • Sensitization could partly explain why 20% of patients with osteoarthritis (OA) experience chronic post-operative pain after otherwise technically successful total knee arthroplasty (TKA). • Sensitization across different stages of treatment and knee hyperalgesia is unknown. What does this study add? • Sensitization is worse after revision TKA than in OA patients and patients with high local knee hyperalgesia show more prominent spreading sensitization. • Sensitization should be considered carefully in knee OA and especially before revision TKA.

Progressive pain sensitization in knee OA

OA and in patients after re-TKA with chronic postoperative pain across different levels of knee pain sensitization, but with similar clinical pain intensities. Such findings may clarify the importance of clinical knee pain intensity and knee pain sensitization for the spreading sensitization and facilitated temporal summation. The aim of the present study was therefore to compare sensitization (spreading sensitization, facilitated temporal summation) in knee OA patients and in patients suffering from chronic post-operative pain after one or more re-TKA(s).

2. Methods 2.1 Material

(such as sensitization and periarticular structures) seem important for the maintenance of pain (Bajaj et al., 2001; Arendt-Nielsen et al., 2010; GravenNielsen et al., 2012; Skou et al., 2013a). Before and after TKA, and after revision TKA (re-TKA), hyperalgesia to pressure stimulation has been found at the painful joint and at remote extra-segmental sites (Kosek and Ordeberg, 2000; Bajaj et al., 2001; Imamura et al., 2008; Arendt-Nielsen et al., 2010; Lee et al., 2011; Graven-Nielsen et al., 2012; Finan et al., 2013; Skou et al., 2013a,b). Hyperalgesia assessed locally at the affected OA knee may reflect sensitization of peripheral neuronal and central neuronal pain mechanisms whereas hyperalgesia extra-segmentally to the affected OA knee represents more generalized spreading sensitization (Arendt-Nielsen and GravenNielsen, 2011). Facilitated temporal summation, another centrally mediated indicator of sensitization, has been demonstrated in painful OA before TKA and after re-TKA if the pain is still present (Arendt-Nielsen et al., 2010; Skou et al., 2013a). A previous study in knee OA indicated an association between sensitization and disease progression, reflected as facilitated temporal summation with higher clinical pain intensities and longer pain durations (Arendt-Nielsen et al., 2010). Furthermore, the extent of hyperalgesia (Arendt-Nielsen et al., 2010; Graven-Nielsen et al., 2012; Skou et al., 2013a,b) and temporal summation (Arendt-Nielsen et al., 2010; Skou et al., 2013a,b) are related to higher clinical pain intensities. Thus, OA disease progression seems better associated with pain and sensitization than with the actual joint destruction assessed by radiological scorings (Arendt-Nielsen et al., 2010). However, no studies have compared spreading sensitization and temporal summation in patients with © 2013 European Pain Federation - EFIC®

Seventy-three pain patients previously contacted regarding enrolment (some participating) in quantitative sensory testing (QST) studies focusing on differences between patients and pain-free control subjects (Arendt-Nielsen et al., 2010; Skou et al., 2013a) participated. Raw data from a subset of patients published previously (Arendt-Nielsen et al., 2010; Skou et al., 2013a) were included and reanalysed according to the new protocol. Fifty-three had knee OA pain and 20 had undergone at least one re-TKA still suffering from chronic post-operative pain. The patients were divided into four groups according to the degree of their knee pressure pain threshold (PPT) assessed from the most affected knee. Demographics and clinical characteristics are shown in Table 1. The patients were asked to refrain from using any analgesics 24 h before the QST session. The study was approved by the local ethics committee of the North Denmark Region (N-20100050) and conducted in accordance with the Helsinki Declaration. Both oral and written information were provided to the patients, and written consent was obtained from all patients.

2.2 Protocol and questionnaire Prior to the QST, the patients completed a short questionnaire on demographics and clinical characteristics including questions on duration of knee pain and peak clinical pain intensity in the affected knee in the previous 24 h measured on a 100-mm visual analogue scale (VAS) with the endpoint descriptors of ‘no pain’ and ‘maximal pain’, respectively. The patients rested in a comfortable recumbent position during the QST and were carefully instructed in the QST methods and made familiar with the procedures.

2.3 Subgrouping of patients The pressure pain sensitivity from the peripatellar region of the most affected knee (localized sensitization/local knee hyperalgesia) was used to subgroup the patients. PPTs from

Eur J Pain 18 (2014) 1024–1031

1025

Progressive pain sensitization in knee OA

S.T. Skou et al.

Table 1 Demographics and clinical characteristics of patients (n = 73). Demographic variable or clinical characteristic mean ± SEM or fractions

Group 1 (n = 26)

Group 2 (n = 27)

Group 3 (n = 10)

Group 4 (n = 10)

Age (years) Gender (female/male) Body mass index (kg/m2) Peak clinical pain in previous 24 h (mm) Duration of knee pain (months) PPT knee (kPa)

64.1 ± 1.5 10/16 28.9 ± 1.1 52.1 ± 5.7 86.6 ± 14.1 702.7 ± 43.7

61.4 ± 1.6 15/12 28.3 ± 0.7 62.4 ± 5.0 89.1 ± 13.8 331.5 ± 19.2

61.4 ± 3.1 7/3 29.3 ± 1.9 58.0 ± 7.5 152.2 ± 24.1 227.2 ± 20.1

61.5 ± 1.8 7/3 32.1 ± 1.5 65.6 ± 7.0 181.8 ± 39.1 130.9 ± 6.0

PPTs measured using a handheld pressure algometer in the peripatellar region of the affected knee. The patients were grouped according to sensitivity at the most affected knee determined using the median PPTs from eight test sites in the peripatellar region. Group 1 (n = 26): knee OA pain and least knee pain sensitivity. Group 2 (n = 27): knee OA pain and high-knee pain sensitivity. Group 3 (n = 10): pain after re-TKA and low-knee pain sensitivity. Group 4 (n = 10): pain after re-TKA and high-knee pain sensitivity. OA, osteoarthritis; PPT, pressure pain threshold; re-TKA, revision total knee arthroplasty.

the peripatellar region were assessed using a handheld pressure algometer (Algometer Type II, Somedic AB, Hörby, Sweden). Pressure was applied perpendicular to the skin (30 kPa/s) with a 1 cm2 probe until the patient felt the pressure as pain and pressed a stop button attached to the handheld algometer after which the pressure was released. This defined the PPT. The average PPT for each patient was calculated from PPTs measured twice from eight sites in the peripatellar region: (1) 2 cm distal to the inferior medial edge of patella; (2) 2 cm distal to the inferior lateral edge of patella; (3) 3 cm lateral to the midpoint on the lateral edge of patella; (4) 2 cm proximal to the superior lateral edge of patella; (5) 2 cm proximal to the superior edge of patella; (6) 2 cm proximal to the superior medial edge of patella; (7) 3 cm medial to the midpoint on the medial edge of patella; and (8) at centre of patella (Arendt-Nielsen et al., 2010; Graven-Nielsen et al., 2012). In the OA group and re-TKA group, respectively, the median peripatellar PPT value for each group was used to subdivide into four groups based on the degree of sensitization: group 1: OA patients with peripatellar PPTs higher than the median PPT based on all OA patients; group 2: OA patients with peripatellar PPTs equal to or lower than the median PPT based on all OA patients; group 3: re-TKA with peripatellar PPTs higher than the median PPT based on all re-TKA patients; group 4: re-TKA with peripatellar PPTs lower than the median PPT based on all re-TKA patients. The median PPT was chosen as the cut-off point, since this divides the groups in equally sized subgroups with distinguishable degrees of local knee hyperalgesia.

2.4 Outcome measurements Spreading sensitization was assessed by handheld pressure algometry at the tibialis anterior (TA) muscle and the extensor carpi radialis longus muscle (forearm) while temporal summation to repeated pressure pain stimuli was assessed at the TA muscle and the knee using a computer-controlled pressure algometer and an electronic VAS (Aalborg University, Aalborg, Denmark) (Graven-Nielsen et al., 2004). These procedures were performed bilaterally.

1026 Eur J Pain 18 (2014) 1024–1031

2.4.1 Spreading sensitization to pressure pain stimulation PPTs were measured from the TA muscle (5 cm distal to the tibial tuberosity) and from the forearm (5 cm distal to the lateral epicondyle of humerus) (Arendt-Nielsen et al., 2010; Graven-Nielsen et al., 2012). The PPT was measured twice at each site and the averages were used for further analysis.

2.4.2 Temporal summation of pressure pain The mechanical pressure stimuli were applied perpendicular to the skin surface using a circular aluminium footplate with a 1 cm2 padded contact surface fixed to the tip of the piston. Using recordings of the actual force, the pressure stimulation was feedback controlled. The PPT was found by increasing the pressure until the patient defined the pressure as pain. At the level of the PPT, 10 sequential pressure stimuli were applied to the most sensitive site in the peripatellar region and to the TA muscle (1 s duration and 1 s inter-stimulus interval). Between the individual pressure stimuli, skin contact was kept by applying a constant force of 0.1 kg, which did not evoke pain. During the sequential stimulation, the patients rated their pain intensity continuously on an electronic VAS where 0 cm indicated ‘no pain’ and 10 cm indicated ‘maximal pain’. The VAS signal for each stimulus was sampled by a computer at 200 Hz. The mean VAS scores during 1 s after each stimulus were extracted and normalized by subtraction of the mean VAS score from the first stimulation. The sum of normalized VAS scores of two series of stimulations from each site was applied in the statistics (VAS sum; possible range 0–90).

2.5 Statistical analysis Confirmed by visual inspection of Q-Q plots data were normally distributed. A one-way analysis of variance (ANOVA) was used to evaluate peak clinical pain intensity in the previous 24 h, PPT and temporal summation data with group (1–4) as a factor. Due to unequal sample size and unequal

© 2013 European Pain Federation - EFIC®

S.T. Skou et al.

variance in the groups, the adjusted F statistic Brown– Forsythe test was applied for PPT and temporal summation. Games-Howell was used as a post hoc test in cases of a significant ANOVA except for peak clinical pain intensity in the previous 24 h, where Tukey–Kramer was applied due to equal variance but unequal sample size. Pearson’s product– moment correlations were applied to assess the association between peak clinical pain intensity in the previous 24 h and duration of knee pain and the QST parameters measured bilaterally. p-values less than 0.05 were considered significant. Data are presented as mean values and standard error of the mean (SEM) or fractions. All analyses were carried out using IBM SPSS Statistics (Version 20; IBM Corporation, Armonk, NY, USA).

3. Results 3.1 Demographics and clinical characteristics of subgroups Based on the knee pressure pain sensitivity, four groups of patients were established with 26, 27, 10 and 10 patients in groups 1–4, respectively (Table 1). The VAS score of the peak clinical pain intensity was not significantly different between groups (ANOVA: F(3, 72) = 0.95, p > 0.4). As expected, the PPTs from the affected knee in groups 1–4 were significantly different due to the subgrouping (ANOVA: F(3, 40.4) = 83.3, p < 0.001; Games-Howell: p < 0.01; Table 1).

3.2 Spreading sensitization to pressure pain stimulation PPTs from the TA muscle in group 4 were significantly lower (more spreading sensitization) compared to TA muscle PPTs in groups 1, 2 and 3; the TA muscle PPTs in groups 2 and 3 were significantly lower than the TA muscle PPTs in group 1 (ANOVA: F(3, 81.0) = 63.3, p < 0.001; Games-Howell: p < 0.05; Fig. 1). PPTs from the forearm in group 4 were significantly lower (more spreading sensitization) compared to the forearm PPTs in groups 1, 2 and 3; the forearm PPTs in groups 2 and 3 were significantly lower than the same PPTs in group 1 (ANOVA: F(3, 78.6) = 45.3, p < 0.001; Games-Howell: p < 0.05; Fig. 1).

3.3 Temporal summation of pressure pain Temporal summation of pressure-induced pain on the knee (Fig. 2A) and TA muscle (Fig. 2B) was clearly demonstrated in all four groups and especially pronounced in groups 3 and 4. The VAS sum extracted after repeated knee pressure pain stimulation was [mean (SEM)] 9.5 (1.9), 9.1 (1.9), 17.8 (2.9) and 23.2 © 2013 European Pain Federation - EFIC®

Progressive pain sensitization in knee OA

Figure 1 Pressure pain thresholds (PPTs). Mean (±SEM). PPTs assessed at the tibialis anterior (TA) muscle and at the extensor carpi radialis longus muscle (forearm) using a handheld pressure algometer. Group 1 (n = 26): knee OA pain and least knee pain sensitivity. Group 2 (n = 27): knee OA pain and high-knee pain sensitivity. Group 3 (n = 10): pain after revision total knee arthroplasty (re-TKA) and low-knee pain sensitivity. Group 4 (n = 10): pain after re-TKA and high-knee pain sensitivity. Significantly lower PPTs were found in group 4 compared to PPTs in groups 1–3 (*p < 0.05) and in groups 2 and 3 compared to group 1 (#p < 0.05) at both sites (forearm and TA muscle).

(4.2) in groups 1–4, respectively. VAS sum at the knee was significantly higher (facilitated temporal summation) in groups 3 and 4 compared to groups 1 and 2 (ANOVA: F(3, 72.3) = 10.7, p < 0.001; Games-Howell: p < 0.05). The VAS sum from repeated pressure stimulation on the TA muscle was [mean (SEM)] 10.5 (2.4), 7.8 (1.5), 21.3 (4.5) and 22.5 (4.1) in groups 1–4, respectively. VAS sum at the TA muscle was significantly higher in groups 3 and 4 compared to the VAS sum in groups 1 and 2 (ANOVA: F(3, 72.7) = 11.3, p < 0.001; Games-Howell: p < 0.05).

3.4 Pressure pain sensitivity, temporal summation and OA pain severity Including all patients, significant correlations were found between peak clinical pain intensity in the affected knee and PPTs from the knee, and between duration of knee pain and PPTs from the knee and the forearm (p < 0.05; Table 2). Likewise, the duration of knee pain and VAS sum (reflecting temporal summation of pain) at the most sensitive site in the peripatellar region and at the TA muscle was significantly correlated (p < 0.001; Table 2).

4. Discussion This study is the first to compare spreading of pressure pain sensitization and temporal summation in patients with painful knee OA and patients suffering from Eur J Pain 18 (2014) 1024–1031

1027

Progressive pain sensitization in knee OA

S.T. Skou et al.

Figure 2 Temporal summation of pressure-induced pain. Mean (±SEM). VAS scores after 10 pressure pain stimulations (temporal summation) at the most sensitive site in the peripatellar region (A) and at the tibialis anterior muscle (B). VAS scores were normalized by subtraction of the VAS scores from the first stimulation. Group 1 (n = 26): knee OA pain and least knee pain sensitivity. Group 2 (n = 27): knee OA pain and high-knee pain sensitivity. Group 3 (n = 10): pain after revision total knee arthroplasty (re-TKA) and low-knee pain sensitivity. Group 4 (n = 10): pain after re-TKA and high-knee pain sensitivity. Groups 3 and 4 had significantly higher VAS sum (see Table 2 for explanation of VAS sum) than groups 1 and 2 for both the peripatellar region and the tibialis anterior muscle (*p < 0.05).

chronic post-operative pain after one or more re-TKA(s). Even though the groups had similar clinical pain intensities, temporal summation was more pronounced in patients with chronic post-operative re-TKA pain compared to OA patients. The same was found for spreading sensitization when comparing re-TKA patients with high local knee hyperalgesia to OA patients with high local knee hyperalgesia and re-TKA patients with low local knee hyperalgesia to OA patients with low local knee hyperalgesia. Furthermore, the spreading sensitization was more pronounced when comparing patients with high local knee hyperalgesia (peripatellar PPTs lower than the median PPT) to patients with low local knee hyperalgesia (peripatellar PPTs higher than the median PPT) within the OA and re-TKA patients, respectively. Clinically it is well known that patients who are still suffering from pain after one or more TKA(s) develop more and more complex pain manifestations. The findings of the present study have important clinical implications as: (1) repeated surgeries in OA patients should be considered very cautiously particularly in patients with pronounced local knee hyperalgesia and (2) management strategies should address the sensitization phenomenon that seems to develop in postoperative chronic pain patients.

4.1 Sensitization at different stages of OA The pain sensitivity distant from the affected joint was more pronounced in patients who had chronic postoperative pain after re-TKA compared to OA pain patients in response to mechanical stimuli. This has previously been demonstrated when comparing hip (Kosek and Ordeberg, 2000), knee (Arendt-Nielsen et al., 2010; Graven-Nielsen et al., 2012) and hand

Table 2 Correlation coefficients in the pain patients (n = 73; Pearson’s product–moment correlations). Pain assessment parameters

Mean ± SEM

Correlation

Peak clinical pain in previous 24 h

Duration of knee pain

PPT knee (kPa)

446.7 ± 31.6

PPT forearm (kPa)

289.0 ± 19.8

R p-value R p-value R p-value R p-value

−0.19 0.02 −0.16 0.05 0.11 0.21 0.07 0.41

−0.25

Facilitation of pain sensitization in knee osteoarthritis and persistent post-operative pain: a cross-sectional study.

Around 20% of patients with osteoarthritis (OA) have chronic post-operative pain after total knee arthroplasty (TKA) and often undergo revision surger...
228KB Sizes 0 Downloads 0 Views