The role of lumbar lordosis and pelvic sagittal balance in femoroacetabular impingement

Patients and Methods

From November 2015 to April 2017, 30 subjects were prospectively enrolled to participate in this Institutional Review Board (IRB)-approved study. To be included, participants had to be healthy, active males between the ages of 18 and 40 years. Those with osteoarthritis of the hip or spine, or a history of congenital deformity, trauma, or surgery of the hip or spine were excluded, as were those with a history of lumbar pain, pincer morphology (lateral centre edge angle > 40°, retroverted acetabulum), acetabular dysplasia (lateral centre edge angle < 20°), an ongoing claim for worker’s compensation, or a limb length discrepancy > 2 cm.

Subjects were recruited and categorized into one of the following groups: subjects with neither hip pain nor radiographic evidence of FAI (non-FAI group); subjects with no pain in the hip but with radiographic signs of FAI (asymptomatic FAI group); or those with pain in the hip and radiographic signs of FAI (symptomatic FAI group). Symptomatic patients were recruited from the senior author’s clinic (ASR). Subjects in the asymptomatic and non-FAI group were volunteers. All participants gave a detailed history of any symptomatology they may have, were examined to assess the range of movement (ROM) and any signs of impingement, and underwent radiographic imaging. Subjects were assigned to the non-FAI cohort if there was no clinical or radiographic evidence of FAI (defined as those with an alpha angle < 50° and internal rotation (IR) > 15°).¹¹ Subjects demonstrating clinical or radiological evidence of FAI (alpha angle > 60°, IR < 15°), but who had no pain from the hip during normal activities or during provocative testing were assigned to the asymptomatic FAI group. Subjects in the symptomatic FAI group demonstrated both clinic and radiographic evidence of FAI (alpha angle > 60°, IR < 15°) with positive examination findings and pain in activities of daily living.

Ten subjects were recruited for each of the three cohorts. However, after radiographic imaging was acquired, ten subjects did not meet inclusion criteria. Four patients with symptomatic FAI were excluded as their alpha angle was too small; five asymptomatic subjects were excluded as they did not have sufficient internal rotation; and one asymptomatic subject was excluded for reporting pain in the hip during provocative testing. The final study group contained 20 subjects, six in the non-FAI group, seven in the asymptomatic FAI group, and seven in the symptomatic FAI group. All subjects had a morphologically normal, asymptomatic contralateral hip. The mean age for the study cohort was 31 years old (22 to 41), with no difference in age between groups (p = 0.07, analysis of variance, ANOVA).

Imaging evaluation

Plain radiographs (anteroposterior (AP) pelvis and Dunn lateral view) were obtained of each subject. Alpha angle was evaluated independently in a standardized method by two fellowship-trained sports surgeons (RRF, MAT), who were blinded to each other’s measurements. Measurements were taken from digital radiographs using the tools within our institution’s Picture Archiving and Communication System (Sectra PACS; Sectra Imtec AB, Linköping, Sweden). The alpha angle was measured on the Dunn view by measuring the angle between a line drawn from the centre of the femoral head bisecting the femoral neck, and a line from the centre of the femoral neck intersecting the point at the head-neck junction. (Fig. 1). FAI was defined as a combination of an alpha angle > 60° on the Dunn lateral and internal rotation < 15° on examination.¹²

Fig. 1

Radiograph showing an example of alpha measurement on Dunn lateral imaging.

In addition to plain radiographs, each participant also underwent low-dose, biplanar 2D standing and sitting stereoradiograph from the thoracolumbar junction (T12-L1) to the femur using the EOS Imaging System (EOS Imaging Inc., Paris, France). Standing and sitting position were standardized for all subjects. Subjects were imaged with their femora aligned approximately parallel to the floor to assume 90° of apparent hip flexion and allow measurement of the degree of true flexion of the spine and hip. This relaxed seated position has been previously used in THA studies.^6,8 Lumbar lordosis, proximal femoral angle, and sacral slope were measured digitally for both the standing and sitting EOS scans according to a previously described method (Fig. 2).⁸ Lumbar lordosis was measured as the angle between the superior endplate of L1 and the superior endplate of S1. The proximal femur angle was measured between a vertical line and the line defined by the anterior cortex of the most visible femur. Sacral slope was measured as the angle between a line parallel to the superior endplate of S1 and a line parallel to the horizontal border of the radiograph. Sacral slope angle was used as a surrogate for pelvic alignment in sitting and standing positions.

Fig. 2

Radiographs showing sacral slope, proximal femoral angle, and lumbar lordosis measurements for: a 24-year-old male patient without femoroacetabular impingement (non-FAI) a) standing and b) sitting (52° spine flexion, 57° femoroacetabular flexion); and a 33-year-old male patient with symptomatic FAI c) standing and d) sitting (7° spine flexion, 80° femoroacetabular flexion).

Measurements of flexion of the spine and hip were derived from the above measurements. Flexion of the spine was calculated as the change in lumbar lordosis between standing and sitting positions. Flexion of the hip joint was calculated by adding the change in position of the proximal femur to the change in position of the pelvis (using the sacral slope angle), always subtracting sitting position values from standing position values. For example, to calculate flexion of the hip joint in the subject without FAI shown in Figure 2, we added the difference in proximal femur angle (89° - 1° = ٨٨°) to the change in sacral slope (8° - 39° = -31°), to find that he had 57°of flexion at the hip.

Results

Spinopelvic parameters are shown in Table I. Subjects with FAI, whether symptomatic or asymptomatic, had significantly higher sacral slope angles while sitting compared with subjects without FAI (p = 0.01, ANOVA), meaning that those with FAI sat with more anterior pelvic tilt (Fig. 2). Symptomatic subjects with FAI exhibited less flexion of the spine (mean 22° (sd 12°) vs mean 35° (sd 8°); p = 0.04, ANOVA) than those without FAI, with a compensatory increase in flexion at the hip joint (mean 72° (sd 6°) vs mean 62° (sd 8°); p = 0.02, ANOVA). For the group as a whole, there is an inverse linear relationship between flexion at the hip and at the spine (R² = 0.48, Fig. 3).

Fig. 3

Chart showing the relationship between femoroacetabular and spine flexion by cohort.

Discussion

In this study, we compared dynamic spinopelvic parameters in symptomatic and asymptomatic individuals with radiographic evidence of FAI as well as normal controls. Subjects with symptomatic FAI had a mean of 13° less spine flexion and 10° more hip flexion versus those without FAI. Subjects with asymptomatic FAI, however, demonstrated no significant difference in flexion of the spine or hip joint compared with controls. These findings suggest that individuals with the morphological features of FAI may remain asymptomatic if the spine is sufficiently mobile to provide compensatory flexion to avoid excess flexion at the hip joint. Without this compensatory lumbar flexion, deep flexion of the hip joint causes cam lesions to engage, precipitating pain seen in FAI. There may be a role for rehabilitation strategies in FAI to focus on improving the mobility of both the hip joint and the lumbar spine.

The current study confirms and extends the findings reported in other studies that proposed an interaction between the kinematics of the spine and the hip.^5,12,13 To our knowledge, this is the first study to investigate the role of spinopelvic kinematics in the setting of FAI. Our findings are similar to those described by Esposito et al⁵ in their study of patients following THA. Those with concomitant degenerative disc disease of the lumbar spine had altered flexion at the hip joint particularly during changes in functional position. We examined similar variables in a younger population of patients with FAI and found that the degree of mobility of the lumbar spine influenced the degree of symptoms in FAI. It is worth noting that subjects included in our cohort did not have radiographic evidence of lumbar spine disease, and therefore the origin of decreased lumbar spine excursion in these symptomatic patients is unclear.

The findings of our study have several important clinical applications. Our results suggest that rehabilitation strategies can be applied to improving flexion of the lumbar spine in patients with symptomatic FAI. We hypothesize that a certain amount of spinopelvic flexion is required to perform functional activities such as sitting, tying of shoelaces, or ascent of stairs. In patients with limited flexion at the lumbar spine, the hip joint is recruited to a greater extent to provide the requisite spinopelvic flexion for functional activities. Other studies have shown that patients with a large anterosuperior cam deformity may impinge throughout the arc of flexion, even during activities such as walking where flexion of the hip only reaches 40°.¹³ We suggest that mobility of the lumbar spine should be considered and evaluated in the investigation of patients with hip pain and possible FAI. As such, patients with limited mobility of the lumbar spine may be identified and directed toward targeted physiotherapy, in addition to traditional hip- and core-based treatments. Fusion of the lumbar spine may be expected to lead to an increase in the degree of flexion required at the hip joint, and may precipitate or exacerbate symptoms in hitherto asymptomatic patients with radiographic evidence of FAI; likewise, patients who present with new symptoms of FAI on the background of a previous fusion of the lumbar spine may have limited benefit from rehabilitative strategies.

Our study has limitations. We did not perform a formal power calculation and we had a small sample size, limiting our ability to detect significant differences between groups. Our sample size was limited by our ability to recruit control subjects, particularly asymptomatic subjects with radiographic evidence of FAI. In spite of our limited sample size, we did find strongly significant relationships, demonstrating the strength of association among the variables under investigation. Additionally, we confined our data set to male subjects. We did this to remove the potential confounding effect of gender, but it may mean that our findings are not generalizable to females. There is an opportunity in future studies to investigate the role of demographic factors such as age, gender, and ethnicity on parameters relating to spinopelvic flexion. We also focused primarily on cam-type FAI and we did not include subgroups with acetabular pathology. Again, this decision was made to preserve homogeneity of the cohort but patients with pincer-type FAI may be examined in future studies. Finally, the images we acquired were dependent on how subjects were positioned in the imaging system. We attempted to standardize this between subjects and this may not replicate how the subjects sit comfortably during their daily activities.

Spinopelvic mechanics appear to have an impact on symptoms in patients with FAI. When lumbar mobility is limited, this leads to an increase in the amount of flexion required at the hip joint, leading to symptomatic engagement of the cam lesion. Patients with symptomatic FAI demonstrate limited compensatory flexion of the lumbar spine, and therefore increased flexion angles at the hip joint during standing to sitting versus those without FAI or with asymptomatic FAI. Spinopelvic mechanics and lumbar mobility may be a target for physiotherapy in patients without degenerative lumbar disease.

Twitter

Follow R. R. Fader @DrRRFader