Combining results from hip impingement and range of motion tests can increase diagnostic accuracy in patients with FAI syndrome

From 2014 to 2017, all patients referred for non-arthritic hip and groin pain (n = 156) to the Department of Orthopedics at Skåne University Hospital in Sweden were consecutively recruited and screened for eligibility. The inclusion criteria were unilateral or bilateral hip/groin pain lasting > 3 months and an age of 18–55 years. The exclusion criteria were as follows: (i) a history of hip surgery; (ii) a hip pathology (i.e., Perthes disease); (iii) verified moderate or severe osteoarthritis (OA) (Tönnis grade > 1); (iv) a history of intra-articular or peri-articular injection with corticosteroids within the last 2 months; (v) palpable hernia; (vi) low-back pain with a positive Lasègue test result with or without an MRI-verified lower back/spine pathology (i.e., spinal stenosis, disc herniation); (vii) other musculoskeletal comorbidities overriding the hip-related symptoms and dysfunction; (viii) comorbidities excluding physical activity and training; (ix) psycho-social disorders; (x) drug abuse; and (xi) an inability to understand the language of interest (any Scandinavian languages or English). Ninety-five patients were eligible, 12 of whom declined participation. Eighty-three participants were consequently recruited. After the initial clinical examination, two participants declined further participation and were thus excluded. Eighty-one participants were finally included in the study (Table 1). The participants rated their perceived pain, disability and associated problems on the Copenhagen Hip and Groin Outcome Score (HAGOS), which includes six subscales: pain, symptoms, physical function in activities of daily living (ADL), physical function in sport and recreation (Sports/rec), participation in physical activity (PA), and quality of life (QOL) (Table 1). The HAGOS has been shown to be a reliable and valid tool in the assessment of long-standing hip and groin pain in a young to middle-aged population [50]. The score for each subscale ranges from 0 to 100, where 0 indicates extreme problems and 100 indicates no problems. The participants rated their preinjury and current activity level on the Hip Sports Activity Scale (HSAS), a valid and reliable questionnaire for assessing individuals’ activity level in this patient group [38] (Table 1).

The clinical assessment of the hip included clinically relevant hip impingement tests, as described by Martin et al. [32]: the Anterior Impingement Test (AIMT), Flexion/Adduction/Internal Rotation (FADIR) test, Flexion/Abduction/External Rotation (FABER) test, Dynamic External Rotatory Impingement Test (DEXRIT), Dynamic Internal Rotatory Impingement Test (DIRIT), Posterior Rim Impingement Test (PRIMT), and passive ROM (flexion, internal rotation with 90° of hip flexion, internal rotation with a neutral hip position, external rotation with 90° of hip flexion, abduction and extension). All participants were assessed by an orthopedic surgeon (IK). To evaluate the inter-rater agreement, the first 69 participants (hips, n = 138) recruited for the study were assessed by both an orthopedic surgeon (IK) and a physical therapist (AP) in a random order. The two raters examined each patient within 1 hour. Both raters had extensive clinical experience in the assessment and treatment of people with hip and groin pain, they examined the patients independently in separate rooms, and they were blinded to the results of the other rater. Prior to the data collection, the raters practiced administering the tests together on two occasions for calibration purposes, and the criteria for a negative/positive test result were agreed upon.

The six impingement tests were performed according to the methods described by Martin et al. [32] (Fig. 1a–f). All tests were performed in a supine position. The patients were instructed to report any pain in the hip and/or groin area. The test results were categorized as either (1) negative (no pain) or (2) positive (the patient’s typical hip and/or groin pain was reproduced).

Passive flexion (Fig. 2a), internal rotation with 90° of hip flexion (Fig. 2b), external rotation with 90° of hip flexion (Fig. 2c) and abduction (Fig. 2d) were examined with the patient in a supine position. Passive extension (Fig. 2e) and internal rotation with a neutral hip position (Fig. 2f) were examined with the patient in a prone position. The patients were instructed to stay relaxed during the tests and to report any pain experienced in the hip/groin area. Each test result was categorized in a clinical manner as either (1) negative (defined as full ROM with or without pain) or (2) positive (defined as decreased ROM with or without pain).

The alpha angle and lateral center–edge angle (LCE angle) were identified and analyzed in accordance with a report by Clohisy et al. [16]. The Lauenstein (frog-leg lateral) projection was used to obtain the alpha angle, whereas the LCE angle was identified on the anteroposterior pelvic view. For seven patients, no Lauenstein projection was available, and the alpha angle was therefore not calculated for these patients. An alpha angle of ≥ 60° was used as the cutoff defining a cam morphology, and an LCE angle of ≥ 40° indicated the presence of a pincer morphology [37]. Excellent inter-rater reliability (ICC ≥ 0.84) for the alpha angle and LCE angle measurements was observed in an analysis of 67 patients from this cohort [40].

All injections were performed by the senior orthopedic surgeon (IK) under fluoroscopic guidance. The intra-articular position of the needle was confirmed by the injection of approximately 1 ml of contrast agent iohexol (Omnipaque, 180 mg/ml) prior to a blockage injection of a mixture containing 2 ml triamcinolone (Lederspan, 20 mg/ml), 4 ml ropivacaine (Narop 10 mg/ml) and 4 ml lidocaine (Xylocain 10 mg/ml). The patients were asked to score the severity of pain they experienced on a visual analog scale (VAS), from 0 (no pain) to 100 (maximal pain) mm, prior to the injection and 1, 2, and 4 hours after the injection. During this period, the patients were instructed to perform activities that would normally provoke pain to determine whether there was any improvement in the symptoms [28]. A decrease in the VAS score by 50% or more over a period of 4 h after the injection was considered to indicate a true effect [33]. The patients were categorized as either responders to the injection (≥ 50% decrease in the VAS score) or nonresponders to the injection (< 50% decrease in the VAS score). Seven patients declined the intra-articular injection, and four patients did not complete the VAS scoring after the injection.

It is recommended to use a triad of symptoms, clinical signs and radiological findings for the diagnosis of FAI syndrome, and to further confirm the hip as the source of pain, an intra-articular block injection can be performed [25]. As the aim of the present study was to evaluate the diagnostic accuracy of the clinical tests, the following criteria for the diagnosis of FAI syndrome had to be fulfilled and were then used as reference standards for evaluating the diagnostic accuracy of the clinical tests: (1) there were symptoms indicating FAI syndrome; (2) there were radiological findings showing CAM and/or pincer morphology; and (3) the patient was considered a responder to the intra-articular block injection.

Cohen’s kappa statistics were used for the inter-rater agreement analysis. The kappa values were interpreted as follows: < 0.00 = poor agreement, 0.00–0.20 = slight agreement, 0.21–0.40 = fair agreement, 0.41–0.60 = moderate agreement, 0.61–0.80 = substantial agreement and 0.81–1.00 = almost perfect agreement [31]. Absolute agreement was expressed in percents. Tests showing kappa values below 0.4 were not included in the analysis of diagnostic accuracy. The results from one rater (the orthopedic surgeon) in the clinical examination were used to determine the diagnostic accuracy. The sensitivity, specificity, positive predictive values (PPVs), and negative predictive values (NPVs) were calculated for each test using cross table analysis [3]. All calculations were performed in SPSS for Windows, V.22.0 (IBM Corp., Armonk, New York, USA). No sample size calculations were performed prior to the study due to its exploratory design. However, the number of participants was selected so that a sample size deemed “good” (n = 50–99) was met according to 4-point scale checklist for methodological quality in the consensus-based standards for the selection of health measurement instruments (COSMIN) [48].

The kappa values were greater than 0.6 for the AIMT, FADIR and FABER tests, greater than 0.4 for DEXRIT and DIRIT, and less than 0.4 for the PRIMT. The absolute agreement ranged between 80 and 83% (Table 2). For the passive hip ROM, the kappa values were greater than 0.4 for flexion, internal rotation with 90° of hip flexion, internal rotation with a neutral hip position, external rotation with 90° of hip flexion and abduction, and they were less than 0.4 for extension. The absolute agreement ranged from 78 to 91% (Table 2). Detailed inter-rater agreement data for the two raters are presented in the Appendix.

Of the 81 participants, 7 were excluded due to missing radiographs or missing Lauenstein projections, and 11 participants were excluded because they either declined the injection or failed to complete the VAS scoring after the injection. Thus, 63 participants [6 with bilateral symptoms and 57 with unilateral symptoms (hips n = 69)] were included in the analysis. The results for the criteria used as reference standards are described in Table 3. The PRIMT and passive extension data were excluded from the analysis of diagnostic accuracy due to poor inter-rater agreement.

The AIMT and FADIR test both showed a sensitivity of 80%, whereas the FABER test, DEXRIT and DIRIT had a sensitivity of no higher than 60%. The specificity ranged from 24 to 51% for all five tests. The PPV ranged from 48 to 53%, and the NPV ranged from 45 to 56% for all tests (Table 4).

The sensitivity for all ROM tests ranged from 29 to 56%. Internal rotation with a neutral hip position showed the highest specificity (94%), followed by abduction and external rotation with 90° of hip flexion (79%), flexion (68%) and internal rotation with 90° of hip flexion (63%). The PPV ranged from 62 to 83%, where the internal rotation with a neutral hip position exhibited the highest PPV value. The NPV ranged from 55 to 59% for all ROM tests (Table 4).