Ò

PAIN 155 (2014) 792–800

www.elsevier.com/locate/pain

Pain referral and regional deep tissue hyperalgesia in experimental human hip pain models Masashi Izumi a,b, Kristian Kjær Petersen a, Lars Arendt-Nielsen a, Thomas Graven-Nielsen a,⇑ a Laboratory for Musculoskeletal Pain and Motor Control, Center for Sensory-Motor Interaction (SMI), Department of Health Science and Technology, Faculty of Medicine, Aalborg University, Aalborg, Denmark b Department of Orthopedic Surgery, Kochi University, Kochi, Japan

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

a r t i c l e

i n f o

Article history: Received 26 August 2013 Received in revised form 6 December 2013 Accepted 14 January 2014

Keywords: Hip pain Referred pain Deep tissue hyperalgesia Hip pain provocation tests

a b s t r a c t Hip disorder patients typically present with extensive pain referral and hyperalgesia. To better understand underlying mechanisms, an experimental hip pain model was established in which pain referrals and hyperalgesia could be studied under standardized conditions. In 16 healthy subjects, pain was induced by hypertonic saline injection into the gluteus medius tendon (GMT), adductor longus tendon (ALT), or gluteus medius muscle (GMM). Isotonic saline was injected contralaterally as control. Pain intensity was assessed on a visual analogue scale (VAS), and subjects mapped the pain distribution. Before, during, and after injections, passive hip joint pain provocation tests were completed, together with quantitative sensory testing as follows: pressure pain thresholds (PPTs), cuff algometry pain thresholds (cuff PPTs), cutaneous pin-prick sensitivity, and thermal pain thresholds. Hypertonic saline injected into the GMT resulted in higher VAS scores than hypertonic injections into the ALT and GMM (P < .05). Referred pain areas spread to larger parts of the leg after GMT and GMM injections compared with more regionalized pain pattern after ALT injections (P < .05). PPTs at the injection site were decreased after hypertonic saline injections into GMT and GMM compared with baseline, ALT injections, and isotonic saline. Cuff PPTs from the thigh were decreased after hypertonic saline injections into the ALT compared with baseline, GMT injections, and isotonic saline (P < .05). More subjects had positive joint pain provocation tests after hypertonic compared with isotonic saline injections (P < .05), indicating that this provocation test also assessed hyperalgesia in extra-articular soft tissues. The experimental models may open for better understanding of pain mechanisms associated with painful hip disorders. Ó 2014 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.

1. Introduction Refractory hip pain is a common problem. In younger patients, pain resulting from sports injury are frequent, especially in soccer players, and runners, and ballet dancers [37]. In older people, persistent hip pain is generally due to osteoarthritis (OA), and a survey showed that 14% of adults more than 60 years of age reported significant hip pain on most days over a 6-weeks period [7]. The current diagnostic approach to hip pain is often inconclusive because the source of pain around the hip joint is multifactorial and includes a variety of intra-articular pathologies, extra-articular soft tissue and tendon pathologies, and mimickers [43]. Hip ⇑ Corresponding author. Address: Laboratory for Musculoskeletal Pain and Motor Control, Center for Sensory-Motor Interaction (SMI), Department of Health Science and Technology, Faculty of Medicine, Aalborg University, Frederik Bajers Vej 7D-3, Aalborg E 9220, Denmark. Tel.: +45 9940 9832; fax: +45 9815 4008. E-mail address: [email protected] (T. Graven-Nielsen).

disorder patients typically present with extensive pain referral patterns to areas including the groin, thigh, buttock, greater trochanteric region, low back, knee, and lower leg [5,21,22,24]. Moreover, painful hip OA patients before replacement surgery show widespread pressure pain hypersensitivity [1]. A better understanding of the pain-generating hip structures, pain referral patterns, and deep-tissue sensitization may provide a better understanding of pain mechanisms in hip disorder patients. Experimental pain induced by injections of hypertonic saline into muscle [13], tendons [11,40] and ligaments [33] is a frequently used model to induce referred pain and deep tissue hyperalgesia in healthy subjects. Experimental tendon and ligament pain models are reported with higher pain intensity and longer pain duration compared with muscle pain models [11,39,44] along with localized deep tissue hyperalgesia [11,33,39,40]. Injections of hypertonic saline into the gluteus medius muscle produced an intense short-term muscle pain, localized to the lateral hip region, and reduced hip and knee joint moment during walking [15].

http://dx.doi.org/10.1016/j.pain.2014.01.008 0304-3959/Ó 2014 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.

Ò

M. Izumi et al. / PAIN 155 (2014) 792–800

Recently, injection of hypertonic saline into the long posterior sacroiliac ligament showed extensive pain referral and regional hyperalgesia [33]. However, no experimental models have been developed to mimic hip pain soft tissue disorders, including a variety of pain referral patterns with increased pain sensitivity. Hip joint pain provocation tests are important in physical examinations to detect potential sensitized structures involved in hip joint disorders. The hip positioned in flexion, abduction, and external rotation (FABER) test is the most widely used hip joint pain provocation test [29], probably demonstrating sensitization of intra-articular hip structures and the sacroiliac joint [27]. The hip positioned in flexion, adduction, and internal rotation (FADIR) test, also called the impingement test, has been used for assessment of intra-articular pathologies such as labral tears and femoroacetabular impingement [6,10]. Recent studies have, however, reported that the FABER and FADIR provocation tests are nonspecific to intra-articular hip pathology [28,29], and hence the association between positive pain provocation tests and specific extra-articular hip pathologies should be clarified. The aim of the present study was to develop an experimental model in which hip pain was induced by hypertonic saline injections into extra-articular soft tissue structures. It was hypothesized that this model could cause (1) pain referral and deep tissue hyperalgesia similar to hip pain disorders as shown previously [1,23], and (2) increased frequency of positive hip joint pain provocation tests, indicating that this test provides information about not only the intra-articular structures but also the surrounding extraarticular soft tissues. 2. Methods 2.1. Subjects Sixteen healthy subjects (8 female and 8 male) with no history of musculoskeletal or neurological problems participated in this study (age, 28 ± 5 years [mean ± SD]; body mass index, 23.6 ± 3.2 kg/m2). Subjects were given a detailed written and verbal explanation of the experimental procedures and signed an informed consent form. The study was conducted in accordance with the Declaration of Helsinki and was approved by the local ethics committee (N2012-0078) within the North Jutland Region. 2.2. Experimental protocol This experiment was randomized, single blinded, and placebo controlled, and included 3 sessions with at least a 1-week interval. All assessments were performed with subjects lying on a bench in either the supine or lateral decubitus position. The quantitative sensory testing (QST) and reactions to hip pain provocation tests were evaluated before (baseline), during, and after (post-pain) experimental hip pain induced by hypertonic saline injections into the gluteus medius tendon (GMT), adductor longus tendon (ALT), and gluteus medius muscle (GMM). Isotonic saline solution was injected into contralateral tendons or muscles as control. The post-pain state was determined at 15 minutes after the pain had subsided. In each session, the subjects received 1 hypertonic and 1 isotonic saline injection, 1 in each side, with the order of the saline type being randomized in a balanced design (left or right) and blinded (saline type) to the subject. The sequence of the 3 injection sites (GMT, ALT, and GMM) among the 3 sessions and the sequence of QST and hip provocation tests were randomized in each subject. 2.3. Experimental hip joint pain Sterile saline solution (1 mL) was injected as either hypertonic (5.8%) or isotonic (0.9%) solutions. After the skin was cleaned with

793

alcohol, injections were performed over 10 seconds using a 2-mL plastic syringe with a disposable needle (27 gauge). The location of each injection site was confirmed by ultrasound imaging (Acuson 128XP10, Native; Siemens Medical Solutions, Malvern, PA). The GMT is the femoral insertion of GMM at the lateral surface of the greater trochanter, which was easily located by manual palpation and its position marked on the skin. According to an anatomical study, the specific insertion of GMT was known [35], and, based on this, the injection started at a point of 2 cm distal to the tip of the greater trochanter with subjects lying in the lateral decubitus position. The needle penetrated the skin perpendicularly until bone contact with the greater trochanter, after which it was withdrawn 1 to 2 mm before injection. With subjects lying in a lateral decubitus position, GMM injections were given at the midpoint of the line between the most cranial and lateral point of the iliac crest and the tip of the greater trochanter [15]. The needle penetrated the skin and fascia perpendicularly and was inserted into the muscle belly. The depth from the skin to the muscle belly was confirmed by ultrasound before needle insertion. The ALT is the proximal tendon of adductor longus muscle inserted into the pubic bone. Injection to the ALT was performed with subjects lying in supine position. The ALT was easily palpated when subjects flexed their leg, and the foot was placed on the opposite knee (the leg position called ‘‘figure 4’’). The examiner stabilized ALT with 2 fingers and inserted the needle into ALT approximately 1 cm far from the pubic bone with the direction toward the bone. The experimental pain intensity was assessed on a 10-cm electronic visual analogue scale (VAS) with an external hand-held slider to adjust the scale. The VAS was anchored with ‘‘no pain’’ and ‘‘maximum pain’’ at 0 cm and 10 cm, respectively. The signal from the VAS was recorded after each injection until the pain had subsided (sample frequency, 0.5 Hz). The peak pain intensity (VAS peak) and area under the VAS–time curve (VAS area) were extracted. The pain duration was estimated as the difference between the last and first time that the VAS exceeded 0; in cases in which the VAS scores remained 0, the pain duration was defined as 0 seconds. After the pain had subsided, the quality of the pain was assessed by completion of an English [30] or Danish [8] version of the McGill Pain Questionnaire. The 3 most frequently selected words after hypertonic saline injection into each site were determined. Moreover, subjects were asked to mark the pain distribution by filling in a body chart. Localized pain was defined as pain felt only around the injection site, whereas referred pain was defined as pain occurring outside the injection-pain area. For more detailed analysis, the body chart was divided into 10 different areas (Fig. 1), and the occurrence of pain in the different areas was registered as follows: (1) groin, (2) greater trochanter, (3) buttock, (4) anterior thigh, (5) posterior thigh, (6) lateral thigh, (7) medial thigh, (8) knee, (9) lower leg, and (10) foot. The frequency of referred pain, the frequency of pain occurrence in each area, and the number of pain areas after hypertonic saline injections were compared among the 3 injection sites. 2.4. Quantitative sensory testing Pressure pain thresholds (PPTs), cuff algometry pain thresholds (cuff PPTs), cutaneous pin-prick pain sensitivity, and thermal pain thresholds were assessed. Each measurement was recorded 3 times in the baseline and the post-pain condition. One (PPTs and cuff PPTs) or 2 (cutaneous pin-prick pain sensitivity and thermal pain thresholds) repetitions were used during pain to ensure recordings during the short-lasting saline-induced pain. Averages of the measurements were used for analysis.

794

Ò

M. Izumi et al. / PAIN 155 (2014) 792–800

Fig. 1. Location of injection sites, outline of body areas used for the analysis of pain distribution (right), and sites for assessment of the somatosensory sensitivity (left). Plus signs indicate injection sites. Assessment sites for pressure pain thresholds (PPTs) are as follows: musculus gluteus medius (site 1), m. gluteus maximus (site 2), m. vastus lateralis (site 3), m. tensor fascia latae (site 4), m. tibialis anterior (TA), and m. extensor carpi radialis longus (Arm). The body areas are as follows: (1) groin, (2) greater trochanter, (3) buttock, (4) anterior thigh, (5) posterior thigh, (6) lateral thigh, (7) medial thigh, (8) knee, (9) lower leg, and (10) foot. Note that the injection sites, body areas, and assessment sites are illustrated separately but assessed on the same side.

2.4.1. Pressure algometry A handheld algometer (Somedic, Hörby, Sweden) mounted with a 1-cm2 probe (covered by a disposable latex sheath) was used to record the PPT at 8 different locations on the body, 6 on the injection side and 2 on the contralateral side (Fig. 1). The assessment sites on the injection side were as follows: (1) musculus gluteus medius, 3 cm proximal to the tip of the greater trochanter; (2) musculus gluteus maximus, 3 cm posterior to the posterior edge of the greater trochanter; (3) m. vastus lateralis, 3 cm distal to the distal edge of the the greater trochanter; (4) m. tensor fascia latae, 3 cm anterior to the anterior edge of the greater trochanter; (5) m. tibialis anterior (TA), 5 cm distal to the tibial tuberosity; (6) m. extensor carpi radialis longus (arm), 5 cm distal to the lateral epicondyle of the humerus. In addition, sites 2 and site 4 were selected for measurement on the contralateral side. An interval of minimum 20 seconds was maintained among the PPT assessments. The PPT was defined to the subject as ‘‘the time point at which the pressure sensation changed into pain.’’ Pressure was increased gradually at a rate of 30 kPa/s until the pain threshold was reached and the subject pressed a button. 2.4.2. Cuff pressure algometry The experimental cuff algometry (NociTech, Denmark) setup consisted of a double-chamber, 13-cm wide tourniquet cuff (a silicone high-pressure cuff, separated lengthwise into 2 equal-size chambers, VBM Medizintechnik GmbH, Sulz, Germany) and a computer-controlled air compressor. The cuff was connected to the compressor and wrapped around the thigh as proximally as possible. The cuff was automatically inflated at a rate of 1 kPa/s until a maximum pressure limit of 100 kPa had been reached. One chamber was used for this assessment, and the subject was instructed to press a hand-held release button when the pressure sensation changed into pain. 2.4.3. Cutaneous pin-prick pain sensitivity A weight-calibrated pinprick device (Aalborg University, Aalborg, Denmark) was used to apply punctate stimulation. The skin (assessment sites 2 and 4) was stimulated by standardized

application of a 12.8 g pin (0.6-mm tip diameter tip), and the subjects scored the pin-prick sensation on a VAS from 0 to 10 cm, where 0 indicated ‘‘no sensation,’’ 5 cm indicated ‘‘pain threshold,’’ and 10 cm indicated ‘‘maximum pain.’’ 2.4.4. Thermal pain sensitivity A 3  3-cm (9-cm2) contact thermode (Medoc Advanced Medical Systems, Ramat Yishai, Israel) was used to apply thermal stimulation. Heat and cold pain thresholds were assessed on site 4. Each stimulus began at 32 °C from which the temperature increased or decreased by 1 °C/s. The temperature could vary between 0 °C and 55 °C during stimulation. The subject was instructed to push a hand-held stop button when the heat or cold sensation changed into pain. 2.5. Hip joint pain provocation tests A clinically trained experimenter applied 2 passive pain provocation tests for the hip joint. Both tests were performed with the subjects lying in supine position. The FABER test was performed with the testing leg flexed and the foot placed on the opposite knee; the passive motions applied were flexion, abduction, and external rotation at the hip joint. The examiner then slowly pressed down on the tested knee, lowering the leg into further abduction [25]. The FADIR test was performed with the testing hip and knee flexed at 90°. Then, the examiner adducted and internally rotated the hip [6]. At baseline the subject was asked whether any pain was experienced in the hip when the tests were performed. In the presence of experimental pain, the subject was asked whether the tests increased the hip pain caused by the injection of saline. To evaluate positive hip pain provocation tests, the pain areas of interest were restricted to the groin, buttock, and greater trochanter area, previously classified as the ‘‘hip region’’ [32,41]. In clinical settings this definition is often used to interpret the provocation tests [26,29]. Both maneuvers were cautiously performed with keeping the flexion angle of testing hip and knee joint, speed of the passive leg movement, and applied force similar to those of the baseline tests.

Ò

795

M. Izumi et al. / PAIN 155 (2014) 792–800

2.6. Statistical analysis Normally distributed data are presented as mean and standard error of the mean (SEM), and other data as median and interquartile range (0.25–0.75). The VAS peak, VAS area, pain duration, and the number of pain areas did not pass the Kolmogorov–Smirnov test for normal distribution and were analyzed with the Friedman test and Wilcoxon signed rank test. In the Friedman test, data from all different injections (6) were included. The PPT, cuff PPT, cutaneous pin-prick pain sensitivity, and thermal pain threshold passed the Kolmogorov–Smirnov test and were analyzed with a mixed-model analysis of variance (ANOVA). Repeated factors were ‘‘injection site’’ (GMT, ALT, and GMM), ‘‘saline type’’ (isotonic or hypertonic), and ‘‘time’’ (baseline, during pain, and post pain). Patient gender, saline sequence, test sequence (QST before provocation test or vice versa), injection site sequence, and injection side were set as independent factors. The PPT and cutaneous pin-prick pain sensitivity were analyzed for each assessment site. For comparison of the sensitivity of QST modalities to detect hip hyperalgesia, percent change of threshold (PPT, cuff PPT, and heat pain threshold) or VAS score (cutaneous pin-prick sensitivity) after hypertonic saline injection was calculated for each injection site. The Wilcoxon signed rank test with Bonferroni correction (WB) or the Neuman–Keuls (NK) test was used for post hoc comparisons incorporating correction for the multiple comparisons when the Friedman test or ANOVA showed significant factors or interactions. The frequency of referred pain, the frequency of pain occurrence in each area, and the response to hip pain provocation tests were analyzed with Fisher’s exact test. A value of P 6 .05 was considered significant. 3. Results 3.1. Experimental hip pain intensity The VAS peak, VAS area, and pain duration were significantly increased after hypertonic saline injections compared with isotonic saline injections (Friedman: P < .01; WB: P < .01) (Table 1). Hypertonic saline injected into the GMT caused higher VAS peak than injections into the ALT and GMM (WB: P < .05), and increased the VAS area compared with injections into the ALT (WB: P < .05). 3.2. Experimental hip pain distribution and quality Superimposed body charts of pain drawings illustrated widespread pain areas after the hypertonic saline injections, whereas isotonic saline caused mainly localized pain around the injection site (Fig. 2). Referred pain was perceived by 81% and 69% of subjects after hypertonic saline injections into the GMT and GMM, respectively, which was significantly more compared with ALT injections, from which 19% of subjects experienced referred pain (Fisher: P < .05).

Hypertonic saline injection into the GMT and GMM caused similar frequencies of pain in areas of the greater trochanter, buttock, posterior thigh, lateral thigh, knee, and lower leg, which were greater than with isotonic saline injection (Fisher: (P < .05) (Table 2). In contrast, ALT injection caused significantly more regionalized pain distribution with higher frequency of pain in the groin and medial thigh compared with GMT and GMM injection (Fisher: P < .05). The number of pain areas was significantly higher after hypertonic saline injection compared with isotonic saline in all sites (Friedman: P < .01; WB: P < .05), and hypertonic saline injection into GMT and GMM caused significantly more pain areas compared with ALT injection (WB: P < .05). The 3 most common words describing the quality of pain after hypertonic saline injections were intense (69% of subjects), exhausting (50%), and spreading (44%) in the GMT; pressing (44%), heavy (31%), and hot (31%) in the ALT; and spreading (44%), intense (38%), and pressing (38%) in the GMM. 3.3. Pressure algometry The PPT values at baseline were as follows: site 1: 442 ± 18 kPa; site 2: 565 ± 21 kPa; site 3: 462 ± 23 kPa; site 4: 577 ± 25 kPa; TA: 562 ± 25 kPa; and arm: 369 ± 20 kPa; these were significantly higher on site 2, site 4, and the TA compared with site 1, site 3, and the arm (ANOVA: F5 = 29.9; P < .0001, NK: P < .0001). A significant interaction between injection site, saline, and time was found on the site 1, site 2, and site 4 (Fig. 3; ANOVA: F4 > 2.6; P < .05) in the ipsilateral hip. Gender, saline sequence, QST sequence, injection site sequence, and injection side did not have a significant impact on the PPT values. Hypertonic saline injection into the GMT caused significantly lower PPTs on site 1 during pain compared with baseline (NK: P < .03), isotonic saline injection (NK: P < .0002), and hypertonic saline injection into the ALT (NK: P 6.7; P < .004). Post hoc testing showed a significant increase during pain after hypertonic saline injection compared with baseline (NK: P < .0002) and isotonic saline injection (NK: P < .002).

Table 1 Median (interquartile range) VAS peak, VAS area, and pain duration following isotonic and hypertonic saline injections into GMT, ALT, and GMM (n = 16). GMT

VAS peak (cm) VAS area (cm s) Pain duration (s)

ALT

GMM

Isotonic

Hypertonic

Isotonic

Hypertonic

Isotonic

Hypertonic

1.9 (0.2–5.1) 99 (0–583) 120 (0–265)

9.1 (7.6–9.9)⁄# 3319 (2231–4280)⁄§ 691 (524–864)⁄

1.0 (0.1–2.3) 34 (0–173) 88 (8–139)

7.4 (5.9–8.5)⁄ 1219 (889–2334)⁄ 408 (319–621)⁄

1.4 (0.6–2.6) 95 (36–236) 136 (61–236)

6.8 (4.3–8.1)⁄ 1685 (1049–3337)⁄ 537 (467–672)⁄

ALT = adductor longus tendon; GMM = gluteus medius muscle; GMT = gluteus medius tendon. Significantly increased compared with isotonic saline (⁄Wilcoxon with Bonferroni correction: P < .01), ALT and GMM injections (#Wilcoxon with Bonferroni correction: P < .05), and ALT injections (§Wilcoxon with Bonferroni correction: P < .05).

796

Ò

M. Izumi et al. / PAIN 155 (2014) 792–800

Fig. 2. Superimposed body chart pain drawings (n = 16) after saline injections into gluteus medius tendon (GMT; A), adductor longus tendon (ALT; B), and gluteus medius muscle (GMM; C). Pain referral patterns after isotonic saline (left) and hypertonic saline (right) injections are illustrated. The side for hypertonic and isotonic saline was randomized, but in this figure, hypertonic saline is illustrated on the right side and vice versa for isotonic saline.

3.4. Cuff pressure algometry

3.6. Thermal pain sensitivity

The mean cuff PPT values at baseline was 31.6 ± 1.1 kPa. There was a significant interaction among injection site, saline, and time (Fig. 4; ANOVA: F4 = 2.9; P < .04). Gender, saline sequence, QST sequence, injection site sequence, and injection side were not a significant independent factor on cuff PPT values. Cuff PPTs were decreased after hypertonic saline injection into the ALT compared with baseline (NK: P < .04), isotonic saline injection (NK: P < .03) and hypertonic saline into the GMT (NK: P < .0003).

The cold pain threshold was mostly not detected within the temperature limit of the apparatus (0 °C) and only 2 subjects in the session of GMT injection reported pain in the 3 different assessment periods during cold stimulation. Consequently, the data were not statistically analyzed. The mean heat pain threshold values at baseline were 45.8 °C ± 0.5 °C, and no significant difference was found among injection sites, saline types, and time (data not presented).

3.5. Cutaneous pin-prick pain sensitivity

3.7. Comparison of QST modalities

The mean baseline VAS scores after punctuate pin-prick stimulation for sites 2 and 4 were 2.7 ± 0.1 cm and 2.6 ± 0.1 cm, respectively. There was no significant interaction between injection site, saline, and time for each assessment site (data not presented).

Regarding PPT and cutaneous pin-prick pain sensitivity, the most sensitive site was taken for the analysis. The PPT on the site 1 after GMT and GMM injections, and the cuff PPT after ALT injection, could detect hyperalgesia induced by hypertonic saline.

Ò

797

M. Izumi et al. / PAIN 155 (2014) 792–800

Table 2 Frequency of pain occurrence (number of subjects and percentages in parenthesis) in each body area and number of pain areas after isotonic and hypertonic saline injections into GMT, ALT, and GMM (n = 16). GMT

Groin Greater Trochanter Buttock Anterior Thigh Posterior Thigh Lateral Thigh Medial Thigh Knee Lower Leg Foot No. of pain areas

ALT

GMM

Isotonic

Hypertonic

Isotonic

Hypertonic

Isotonic

Hypertonic

0 (0) 14 (88) 1 (6) 1 (6) 1 (6) 2 (13) 0 (0) 0 (0) 1 (6) 0 (0) 1.0 (1.0–1.5)

0 (0)# 16 (100)# 10 (63)⁄# 3 (19) 7 (44)⁄# 14 (88)⁄# 1 (6)# 9 (56)⁄# 11 (69)⁄# 4 (25) 5.5 (3.0–6.0)⁄#

13 (81) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1.0 (1.0–1.0)

16 (100) 1 (6) 0 (0) 2 (13) 0 (0) 3 (19) 9 (56)⁄ 0 (0) 0 (0) 0 (0) 2.0 (1.0–3.0)⁄

0 (0) 14 (88) 2 (13) 1 (6) 1 (6) 1 (6) 0 (0) 0 (0) 0 (0) 0 (0) 1.0 (1.0–1.0)

1 (6)# 16 (100)# 9 (56)⁄# 1 (6) 5 (31) 8 (50)⁄ 1 (6)# 6 (38)⁄# 9 (56)⁄# 1 (6) 4.0 (2.5–4.0)⁄#

Data are numbers of subjects with percentages in parentheses unless otherwise specified. ALT = adductor longus tendon; GMM = gluteus medius muscle; GMT = gluteus medius tendon. Frequency of pain occurrence; significantly increased compared with isotonic saline (⁄Fisher: P < .05), significantly different from the ALT injections (#Fisher: P < .05). Number of pain areas; significantly increased compared with isotonic saline (⁄Wilcoxon with Bonferroni correction: P < .05), significantly different from the ALT injections (#Wilcoxon with Bonferroni correction: P < .05).

Fig. 3. Mean ( ± SEM, n = 16) pressure pain thresholds (PPTs) recorded at the 8 assessment sites (6 on injection side) before, during, and after the injection of hypertonic (A) and isotonic saline (B) into the gluteus medius tendon (GMT; filled circle), adductor longus tendon (ALT; open circle), and gluteus medius muscle (GMM; triangle). Although statistical analysis was performed on raw values, the data in this figure are normalized to baseline value and are indicated as percentage changes. Significantly different from baseline (Neuman–Keuls [NK]: ⁄P < .05), ALT injection (#NK: P < .05), and isotonic saline (§NK: P < .05). B denotes baseline; D, during pain; P, post pain.

Neither cutaneous pin-prick pain sensitivity on site 2 nor heat pain threshold was as sensitive as PPT or cuff PPT at the 3 injection sites (Table 3).

saline injections (Fisher: P < .05), except for the ALT injections, which did not show a significantly higher frequency of positive FADIR test results compared with isotonic saline.

3.8. Hip joint pain provocation tests

4. Discussion

No participants experienced hip pain by the passive pain provocation tests at baseline. The number and percentage of subjects who showed positive pain provocation tests after isotonic/hypertonic saline injections into the GMT, ALT, and GMM are shown in Table 4. More subjects had positive FABER test results during the pain condition after hypertonic compared with isotonic saline injections into all sites (Fisher: P < .05). The FADIR test also provoked additional pain after hypertonic compared with isotonic

This is the first study to describe a human experimental model mimicking hip pain. Hypertonic saline injected into extra-articular hip tendon or muscle evoked consistent patterns of pain referral, deep tissue hyperalgesia, and positive pain provocation tests. The experimentally evoked pain intensity and distribution, frequency of referred pain, and manifestations of deep tissue hyperalgesia were different among anatomic structures, which helps to elucidate pain mechanisms in hip disorders.

Ò

798

M. Izumi et al. / PAIN 155 (2014) 792–800

Fig. 4. Mean (±SEM, n = 16) cuff pressure pain thresholds (cuff PPTs) recorded at the ipsilateral thigh before, during, and after the injection of hypertonic (A) and isotonic (B) saline into the gluteus medius tendon (GMT; filled circle), adductor longus tendon (ALT; open circle), and gluteus medius muscle (GMM; triangle). Although statistical analysis was performed on raw values, the data in this figure are normalized to baseline value and indicated as percentage changes. Significantly increased compared with baseline (⁄Neuman–Keuls [NK]: P < .01). Significantly decreased compared with GMT injection (#NK: P < .01) and isotonic saline (§NK: P < .05).

Table 3 Percentage of baseline pain thresholds (PPT, cuff PPT, heat pain threshold) or VAS scores (cutaneous pin-prick sensitivity) after hypertonic saline injections into GMT, ALT, and GMM. GMT PPT (site 1) Cuff PPT Pin-prick (site 2) Heat pain threshold

*

89.2 ± 6.3 118.6 ± 8.3 106.7 ± 10.7 101.9 ± 0.7

ALT

GMM

111.7 ± 5.7 86.8 ± 5.8* 109.9 ± 6.8 101.8 ± 0.8

85.0 ± 6.8* 107.3 ± 4.6 107.2 ± 11.6 100.4 ± 0.7

Data are mean ± SEM (n = 16). No change from baseline is equivalent to 100%. Values of less than 100% indicate the modalities detecting hip hyperalgesia. ALT = adductor longus tendon; GMM = gluteus medius muscle; GMT = gluteus medius tendon; PPT = pressure pain threshold; VAS = visual analogue scale. * Significantly different from baseline (further details in Results).

4.1. Experimental hip pain arising from different extra-articular structures Hypertonic saline injection into the muscle and tendon around hip joint induced short-term, high-intensity, deep-tissue pain. This has been well documented for saline-induced pain in structures other than the hip [9,11,19,42]. Among the 3 injection sites, the GMT injection caused higher VAS peak than injections into the ALT and GMM, and increased VAS area to a greater degree than injection into the ALT. This injection-site variation is probably related to a different sensitivity to hypertonic saline between tendon-bone junction (GMT), tendon (ALT), and muscle belly (GMM) sites. Gibson et al. [11] reported higher pain intensity after hypertonic saline injected into the tendon-bone junction of the tibialis anterior muscle compared with its tendon and muscle belly site. Injecting the tendon-bone junction probably stimulates periosteal tissue, which is known to be more sensitive than muscle belly [46], thus contributing to the higher pain intensity at this site. Slater et al. [40] demonstrated higher pain intensity after hypertonic saline injection into the tendoachilles compared with the common extensor tendon. A variation in nociceptor density and innervation

patterns of individual tendons may contribute to different tendon pain profiles. The different patterns of pain referral between the GMT/GMM and ALT after hypertonic saline injection may be explained partially by their myotomes. The gluteus medius is innervated by superior gluteal nerve deriving from L4–S1 spinal segments [20,34], whereas the adductor longus is innervated by obturator nerve from L2–L4 segments [20,34]. ALT injections may therefore stimulate dorsal horn neurons at the L2–L4 spinal segmental level, which induced more localized pain in the groin and thigh, whereas GMT/GMM injection may excite dorsal horn neurons in more caudal segments (L4–S1), which is associated with pain distributed to the greater trochanter area, buttock, knee, and lower leg. The most common pain distribution in patients with hip osteoarthritis and osteonecrosis was reported as the groin followed by the buttock, thigh, greater trochanter area, and knee [16,21,22,32]. In labral tear patients, the groin and greater trochanter area were the most common locations of referred pain [2,5]. According to the results of a cadaveric study, the human anterior hip joint capsule was innervated by the articular branch of obturator nerve and femoral nerve [3] and was thus innervated by afferent nerves from L2–L4 spinal segments. The posterior capsule is innervated by the sciatic nerve and superior gluteal nerve [3] which are derived from the L4–S1 spinal segments. Nakajima et al. [31] reported that the hip joint was innervated primarily from L2, L3, and L4 dosal root ganglion in rats. Based on these studies, it is possible that direct stimulation of nociceptive afferents by hypertonic saline injection into GMT, GMM (innervated by superior gluteal nerve: L4–S1), and ALT (innervated by obturator nerve: L2–L4) could reach the spinal segments similar to the afferents innervating the hip joint and thus show pain referrals similar to those in hip disorder patients with intra-articular pathology. The words describing the quality of pain after hypertonic saline injections into GMT and GMM were in accordance with, or similar to, the results of previous studies on muscle pain and tendon pain [13,40]. ‘‘Spreading’’ was extensively used after GMT and GMM

Table 4 Frequency of positive hip joint pain provocation tests after isotonic and hypertonic saline injections into GMT, ALT, and GMM (n = 16). GMT

FABER FADIR

ALT

GMM

Isotonic

Hypertonic

Isotonic

Hypertonic

Isotonic

Hypertonic

2 (12.5) 1 (6.3)

13 (81.3)* 9 (56.3)*

0 (0) 1 (6.3)

12 (75)* 4 (25)

2 (12.5) 0 (0)

9 (56.3)* 7 (43.8)*

Data are number of subjects with percentages in parentheses. ALT = adductor longus tendon; GMM = gluteus medius muscle; GMT = gluteus medius tendon; FABER = flexion, abduction, and external rotation of the hip joint; FADIR = flexion, adduction, and internal rotation of the hip joint. * Significantly higher than after isotonic saline (Fisher: P < .05).

Ò

M. Izumi et al. / PAIN 155 (2014) 792–800

injection but not after ALT injection, which might reflect the difference of pain referral patterns. Thirty-one percent of subjects chose ‘‘hot,’’ which may describe the relatively superficial sensation after ALT injection. 4.2. Hip hyperalgesia The regional deep tissue hyperalgesia is consistent with previous studies injecting hypertonic saline into muscle [14,18] and tendon [11,40]. In hip OA patients, Kosek et al. [23] reported increased sensitivity to pressure pain in the most painful area around the hip. Recently, Aranda-Villalobos et al. [1] described widespread pressure pain hyperalgesia before hip replacement surgery. The pressure hypersensitivity was normalized after successful total hip replacement [1,23]. Therefore, pressure pain sensitivity is a clinically reliable marker for monitoring sensitization of patients with hip osteoarthritis, which can be mimicked by the present experimental models. The PPTs on sites 2 and 4 were increased after hypertonic saline injected into ALT. In addition, hypertonic saline injection increased PPTs compared with baseline and isotonic saline injection on the arm and contralateral site 2 (control sites). These findings are consistent with other studies demonstrating saline-induced pressure hypoalgesia [11,40]. In the present study, ALT injections caused pain mainly in groin and medial thigh which were away from the PPT assessment sites located in the lateral hip. These data reflect a possible role of conditioned pain modulation whereby inhibitory mechanisms modulate the competitive balance between nociceptive and non-nociceptive sensory inputs [48]. This study represents the first use of cuff algometry on the thigh to assess hyperalgesia in hip pain conditions. Interestingly, cuff PPT was decreased after hypertonic saline injection into the ALT, whereas it was increased after GMT injection, which showed opposite results of hyperalgesia and hypoalgesia compared with pressure algometry findings. This suggests that cuff algometry on the thigh possibly reflects deep tissue hyperalgesia around the groin and medial thigh, where pressure algometry was not used in the present study. Therefore, a combined assessment using pressure algometry and cuff algometry might be helpful to detect deep tissue hyperalgesia in hip disorder patients and to locate potential sensitized structures (ie, GMT vs ALT). Compared with cutaneous pin-prick sensitivity and thermal pain sensitivity, pressure algometry and cuff pressure algometry demonstrated larger changes under the given conditions by the hypertonic saline injections, which indicates that PPTs and cuff PPTs are more useful modalities to detect hip hyperalgesia. 4.3. Hip joint pain provocation tests The specificity of pain provocation tests for predicting intraarticular hip pathology were previously documented as only 18% to 25% by the FABER test and 10% to 17% by the FADIR test, respectively [28,29]. This low specificity may be explained by involvement of extrinsic causes such as sacroiliac joint and lower lumbar facet joints [17]. However, extra-articular hip pathology may also result in a large number of false-positive findings. In the current study, pain arising from the specific extra-articular structures significantly increased the frequency of positive pain provocation tests. In particular, more than three-quarters of subjects showed positive FABER test after hypertonic saline injection into GMT and ALT, indicating that pain and hyperalgesia of both tendons are highly associated with the outcome of this test. In a recent experimental study of sacroiliac joint pain induced by hypertonic saline, Palsson et al. [33] demonstrated a positive correlation between the number of positive sacroiliac joint pain provocation tests and pressure pain sensitivity around the

799

injection site, indicating that saline-induced regional hyperalgesia plays a role in the outcome of sacroiliac joint pain provocation tests. In the present study, the high frequency of positive FABER tests may also be related to regional hyperalgesia, although hypertonic saline injection into ALT did not increase the frequency of positive FADIR test significantly compared with isotonic saline. A possible explanation for this finding is that the FADIR test applies less stretching force than the FABER test to the structures of groin and medial thigh, where the ALT hypertonic saline injection caused pain and hyperalgesia. 4.4. Clinical relevance The outcomes of the current study have 3 important clinical implications. First, the nociceptive inputs from the extra-articular hip structures are highly important clinically because they reflects the pain manifestations in patients with typical extra-articular hip disorders, such as greater trochanteric pain syndrome [47] and adductor tendinopathy [45]. Because of the similarities between the clinical pain distribution in the above conditions [4,12] and the experimental pain distribution, the present study validates, for the first time, that these extra-articular structures have a nociceptive capacity, as previously assumed based on blocking (anesthesia) studies of the same tissue [36,38]. Second, an optimal QST paradigm to detect hyperalgesia around the hip joint was developed under standardized experimental conditions. This could be the rationale for applying the QST in future clinical studies evaluating hip pain patients. Third, the experimental models demonstrate pain distributions similar to those of intra-articular hip disorder patients [2,5,16,21,22,24,32] without any nociceptive input from the hip joint. Thus, in terms of the referred pain patterns, a similar central mechanism may be involved in extra- and intraarticular hip pathologies, which, however, may challenge the diagnostic value of referred pain to separate the various clinical hip pain conditions. 4.5. Conclusion In this study, a novel experimental human hip pain model was developed. Injections of hypertonic saline into extra-articular soft tissue structures evoked defined patterns of pain referrals, regional deep tissue hyperalgesia, and positive provocation tests similar to what is clinically observed in hip pain disorders. Moreover, these data indicate that the hip provocation tests do not only reflect the intra-articular pain sensitivity but that the extra-articular structures are highly important for the outcome. The present study provides new insights into the pain mechanism and the structures generating specific pain patterns in hip disorder patients. Conflict of interest statement The authors have no conflict of interest to declare. Acknowledgement This study was supported by the Danish Rheumatism Association. References [1] Aranda-Villalobos P, Fernández-de-Las-Peñas C, Navarro-Espigares JL, Hernández-Torres E, Villalobos M, Arendt-Nielsen L, Arroyo-Morales M. Normalization of widespread pressure pain hypersensitivity after total hip replacement in patients with hip osteoarthritis is associated with clinical and functional improvements. Arthritis Rheum 2013;65:1262–70. [2] Arnold DR, Keene JS, Blankenbaker DG, Desmet AA. Hip pain referral patterns in patients with labral tears: analysis based on intra-articular anesthetic

800

[3] [4] [5]

[6] [7]

[8]

[9] [10]

[11] [12] [13] [14]

[15]

[16]

[17] [18] [19] [20] [21] [22] [23]

[24] [25] [26] [27]

Ò

M. Izumi et al. / PAIN 155 (2014) 792–800 injections, hip arthroscopy, and a new pain ‘‘circle’’ diagram. Phys Sportsmed 2011;39:29–35. Birnbaum K, Prescher A, Hessler S, Heller KD. The sensory innervation of the hip joint—an anatomical study. Surg Radiol Anat 1997;19:371–5. Braun P, Jensen S. Hip pain—a focus on the sporting population. Aust Fam Physician 2007;36:410–3. Burnett RS, Della Rocca GJ, Prather H, Curry M, Maloney WJ, Clohisy JC. Clinical presentation of patients with tears of the acetabular labrum. J Bone Joint Surg Am 2006;88:1448–57. Byrd JW. Hip arthroscopy: patient assessment and indications. Instr Course Lect 2003;52:711–9. Christmas C, Crespo CJ, Franckowiak SC, Bathon JM, Bartlett SJ, Andersen RE. How common is hip pain among older adults? Results from the Third National Health and Nutrition Examination Survey. J Fam Pract 2002;51:345–8. Drewes AM, Helweg-Larsen S, Petersen P, Brennum J, Andreasen A, Poulsen LH, Jensen TS. McGill Pain Questionnaire translated into Danish: experimental and clinical findings. Clin J Pain 1993;9:80–7. Feinstein B, Langton JN, Jameson RM, Schiller F. Experiments on pain referred from deep somatic tissues. J Bone Joint Surg Am 1954;36-A:981–97. Ganz R, Parvizi J, Beck M, Leunig M, Nötzli H, Siebenrock KA. Femoroacetabular impingement: a cause for osteoarthritis of the hip. Clin Orthop Relat Res 2003:112–20. Gibson W, Arendt-Nielsen L, Graven-Nielsen T. Referred pain and hyperalgesia in human tendon and muscle belly tissue. PAINÒ 2006;120:113–23. Gordon EJ. Trochanteric bursitis and tendinitis. Clin Orthop 1961;20:193–202. Graven-Nielsen T. Fundamentals of muscle pain, referred pain, and deep tissue hyperalgesia. Scand J Rheumatol Suppl 2006;122:1–43. Graven-Nielsen T, Fenger-Grøn LS, Svensson P, Steengaard-Pedersen K, Arendt-Nielsen L, Staehelin Jensen T. Quantification of deep and superficial sensibility in saline-induced muscle pain—a psychophysical study. Somatosens Mot Res 1998;15:46–53. Henriksen M, Aaboe J, Simonsen EB, Alkjaer T, Bliddal H. Experimentally reduced hip abductor function during walking: implications for knee joint loads. J Biomech 2009;42:1236–40. Hsieh PH, Chang Y, Chen DW, Lee MS, Shih HN, Ueng SW. Pain distribution and response to total hip arthroplasty: a prospective observational study in 113 patients with end-stage hip disease. J Orthop Sci 2012;17:213–8. Porterfield JA, DeRosa C. Mechanical low back pain: perspectives in functional anatomy. Philadelphia: W.B. Saunders; 1991. Jensen K, Norup M. Experimental pain in human temporal muscle induced by hypertonic saline, potassium and acidity. Cephalalgia 1992;12:101–6. Kellegren JH. Observations on referred pain arising from muscle. Clin Sci 1938;3:175–90. Kass JS, Chiou-Tan FY, Harrell JS, Zhang H, Taber KH. Sectional neuroanatomy of the pelvic floor. J Comput Assist Tomogr 2010;34:473–7. Khan AM, McLoughlin E, Giannakas K, Hutchinson C, Andrew JG. Hip osteoarthritis: where is the pain? Ann R Coll Surg Engl 2004;86:119–21. Khan NQ, Woolson ST. Referral patterns of hip pain in patients undergoing total hip replacement. Orthopedics 1998;21:123–6. Kosek E, Ordeberg G. Abnormalities of somatosensory perception in patients with painful osteoarthritis normalize following successful treatment. Eur J Pain 2000;4:229–38. Lesher JM, Dreyfuss P, Hager N, Kaplan M, Furman M. Hip joint pain referral patterns: a descriptive study. Pain Med 2008;9:22–5. Magee D. Orthopedic physical assessment. Philadelphia, PA: W.B. Saunders; 1992. Margo K, Drezner J, Motzkin D. Evaluation and management of hip pain: an algorithmic approach. J Fam Pract 2003;52:607–17. Martin RL, Enseki KR, Draovitch P, Trapuzzano T, Philippon MJ. Acetabular labral tears of the hip: examination and diagnostic challenges. J Orthop Sports Phys Ther 2006;36:503–15.

[28] Martin RL, Irrgang JJ, Sekiya JK. The diagnostic accuracy of a clinical examination in determining intra-articular hip pain for potential hip arthroscopy candidates. Arthroscopy 2008;24:1013–8. [29] Maslowski E, Sullivan W, Forster Harwood J, Gonzalez P, Kaufman M, Vidal A, Akuthota V. The diagnostic validity of hip provocation maneuvers to detect intra-articular hip pathology. PM R 2010;2:174–81. [30] Melzack R, Torgerson WS. On the language of pain. Anesthesiology 1971;34:50–9. [31] Nakajima T, Ohtori S, Inoue G, Koshi T, Yamamoto S, Nakamura J, Takahashi K, Harada Y. The characteristics of dorsal-root ganglia and sensory innervation of the hip in rats. J Bone Joint Surg Br 2008;90:254–7. [32] Nakamura J, Oinuma K, Ohtori S, Watanabe A, Shigemura T, Sasho T, Saito M, Suzuki M, Takahashi K, Kishida S. Distribution of hip pain in osteoarthritis patients secondary to developmental dysplasia of the hip. Mod Rheumatol 2013;23:119–24. [33] Palsson TS, Graven-Nielsen T. Experimental pelvic pain facilitates pain provocation tests and causes regional hyperalgesia. PAINÒ 2012;153:2233–40. [34] Patni P, Hayman LA, Duncan G, Chiou-Tan F, Taber KH. Sectional neuroanatomy of the lower limb I: lower back and hip. J Comput Assist Tomogr 2001;25:656–60. [35] Robertson WJ, Gardner MJ, Barker JU, Boraiah S, Lorich DG, Kelly BT. Anatomy and dimensions of the gluteus medius tendon insertion. Arthroscopy 2008;24:130–6. [36] Schilders E, Bismil Q, Robinson P, O’Connor PJ, Gibbon WW, Talbot JC. Adductor-related groin pain in competitive athletes. Role of adductor enthesis, magnetic resonance imaging, and entheseal pubic cleft injections. J Bone Joint Surg Am 2007;89:2173–8. [37] Scopp JM, Moorman CT. The assessment of athletic hip injury. Clin Sports Med 2001;20:647–59. [38] Shbeeb MI, O’Duffy JD, Michet CJ, O’Fallon WM, Matteson EL. Evaluation of glucocorticosteroid injection for the treatment of trochanteric bursitis. J Rheumatol 1996;23:2104–6. [39] Slater H, Arendt-Nielsen L, Wright A, Graven-Nielsen T. Experimental deep tissue pain in wrist extensors—a model of lateral epicondylalgia. Eur J Pain 2003;7:277–88. [40] Slater H, Gibson W, Graven-Nielsen T. Sensory responses to mechanically and chemically induced tendon pain in healthy subjects. Eur J Pain 2011;15:146–52. [41] Street J, Lenehan B, Flavin R, Beale E, Murray P. Do pain referral patterns determine patient outcome after total hip arthroplasty? Acta Orthop Belg 2005;71:540–7. [42] Graven-Nielsen T, Arendt-Nilesen L, Svensson P, Jensen TS. Experimental muscle pain: a quantitative study of local and referred pain in humans following injection of hypertonic saline. J Musculoskele Pain 1997;5:49–69. [43] Tibor LM, Sekiya JK. Differential diagnosis of pain around the hip joint. Arthroscopy 2008;24:1407–21. [44] Tsao H, Tucker KJ, Coppieters MW, Hodges PW. Experimentally induced low back pain from hypertonic saline injections into lumbar interspinous ligament and erector spinae muscle. PAINÒ 2010;150:167–72. [45] Valent A, Frizziero A, Bressan S, Zanella E, Giannotti E, Masiero S. Insertional tendinopathy of the adductors and rectus abdominis in athletes: a review. Muscles Ligaments Tendons J 2012;2:142–8. [46] Inman VT, Saunders JB. Referred pain from skeletal structures. J Nerv Ment Dis 1944;99:660–7. [47] Williams BS, Cohen SP. Greater trochanteric pain syndrome: a review of anatomy, diagnosis and treatment. Anesth Analg 2009;108:1662–70. [48] Yarnitsky D. Conditioned pain modulation (the diffuse noxious inhibitory control-like effect): its relevance for acute and chronic pain states. Curr Opin Anaesthesiol 2010;23:611–5.