Abstract
Introduction
The total hip arthroplasty (THA) is considered a very successful surgery for reducing pain and increasing function in individuals most commonly suffering from severe hip osteoarthritis [1–7]. Over the past 20 years, the focus of the THA has been to accelerate post-operative rehabilitation, improve functional outcomes, and minimize the extent of the scar following THA [8]. To achieve these post-operative goals, surgeons began implementing the use of minimally invasive THA procedures. The aim of the minimally invasive THA is to reduce the amount of soft tissue damage that occurs during surgery in order to decrease the recovery time and increase the rate of return to normal function following surgery [9–11]. A wide range of THA procedures fall under the minimally invasive heading, with each procedure varying in the extent of soft tissue damage that occurs during surgery. Although it appears less soft tissue damage does occur during surgery, there is conflicting evidence as to whether minimally invasive THA procedures result in a shorter recovery period or return of function [12–19].
The direct anterior (DA) approach is a specific type of minimally invasive THA procedure known as a tissue sparing approach, designed to access the hip joint without muscle incisions to minimize tissue damage [20–22]. The DA, first described by Smith Petersen in 1949, involves separation and retraction of the tensor fascia latae muscle laterally and the sartorius and rectus femoris muscles medially via the intermuscular plane on the anterior aspect of the hip [20–23]. A study by Meneghini et al. [11] analyzed the soft tissue damage that occurred during the DA in six fresh frozen cadavers. They found that during the DA, a mean of 2.62% of the gluteus medius, 8.48% of the gluteus minimus, 12.24% of the rectus femoris, and 31.32% of the tensor fascia latae muscle were damaged during the procedure. A similar study by van Oldenrijk et al. [24] however, revealed no damage to the gluteus medius in four out of five fresh frozen cadavers, but did find that tensor fascia latae sustained 35% damage during the DA.
Appropriate hip muscular strength is essential for normal ambulation. Additionally, the hip abductor muscles are vital for pelvic stabilization during the single leg stance, while the deep hip external rotators and posterior hip capsule are responsible for joint stability and resisting posterior hip joint dislocation [25, 26]. It appears that the DA results in more damage to the hip musculature than theorized, yet the damage to these muscles is still minimal compared to other minimally invasive or standard procedures [11, 24]. It has been documented that patients who underwent THA demonstrated hip muscular weakness years after surgery [27–29]. Although these patients are pain free and are much more functional than prior to the THA, the effects that the muscle weakness has on gait and function may have an impact on future joint replacement or an association with risk of falling [30–33]. Due to the different types of THA procedures commonly performed, it is important to understand how each surgery impacts the patient. Identifying and addressing areas of weakness early in the recovery process my significantly impact the patients long term quality of life. Therefore, the purpose of this study is to prospectively analyze isometric hip muscle strength and functional ability in the DA population to determine the recovery of normal function and strength.
Methods
Research design
This study involved a longitudinal prospective research design to analyze the strength and functional recovery of THA patients undergoing DA compared to a group of normal control subjects. Trendelenburg Single Leg Balance Test (SLB), Timed Up and Go Test (TUG), and hip isometric musculature strength tests were conducted on the DA patients at four different data collection sessions: 1) pre-operation, 2) 3 weeks, 3) 6 weeks, and 4) 3 months post-operatively. The same measures were also collected on a group of control subjects at four different data collection sessions: 1) initial visit, 2) 3 weeks, 3) 6 weeks, and 4) 3 months following the initial visit. Prior to participation, all participants were informed about the nature of the study and signed their respective informed consent and HIPAA release forms approved by the hospital and university’s institutional review board.
Participants
Participants included twenty (eight female, twelve male) hip osteoarthritis (OA) adult patients with a mean age of 64.2 ± 9.8 years who elected to undergo THA surgery. These participants were recruited from a pool of hip OA patients undergoing unilateral DA surgery. An additional fourteen (five female, nine male) non-OA involved control participants with a mean age of 61.7 ± 4.0 years were recruited for study participation.
Volunteers for the DA group were screened and cleared for study participation by their surgeon during their routine pre and postoperative visits. Exclusionary criteria for the study participation included any lower extremity joint replacements other than the one being replaced over the course of this study, a history of rheumatoid or inflammatory arthritis, or the inability to walk without an assistive device. An additional exclusionary criterion for the control subjects included a history of diagnosed severe OA in the lower extremity.
Direct anterior THA surgical approach
All DA THAs were performed by the same board certified fellowship trained orthopedic surgeon. The DA, as performed in this study, utilized a hana® hip and knee arthroplasty table (Mizuho OSI, Union City, CA, USA) with the patient in the supine position. This procedure used the intermuscular plane on the anterior aspect of the body between the sartorius and rectus femoris medially and the tensor fascia latae laterally [21, 34]. A 6–10 cm skin incision was made approximately 3 cm lateral and 3 cm distal to the anterior superior iliac spine [34–36]. The fascia was incised and the muscles were retracted exposing the anterior hip joint capsule. Once exposed, the joint capsule was incised and preserved and the femoral neck was cut in situ and the femoral head removed. After component implantation, the muscular fascia was sutured and the skin incision was closed.
Data collection
The data collection sessions for all participants were conducted in the university’s Human Performance and Gait Laboratory. Pre operation trials for the DA participants were conducted within two week prior to surgery. The first post operation data collection session occurred at an average of 3.22 ( ± 0.55) weeks, the second at 6.5 ( ± 0.51) weeks, and the third at 13.25 ( ± 1.31) weeks. The control participants’ data collection sessions occurred at an average of 2.7 ( ± 0.46), 5.9 ( ± 0.92) and 14.5 ( ± 2.03) weeks following the initial visit. Upon reporting to the laboratory, height, weight, age, and body mass index were obtained (Table 1). Additionally, the Harris Hip Score (HHS) was taken at each data collection session (Table 1) [37]. The HHS assess pain and functional ability on a 100 point scale [37]. Participants then completed the Trendelenburg SLB Test, the TUG, and hip maximal isometric strength trials. All tests were performed barefoot.
Trendelenburg Single Leg Balance (SLB)
The Trendelenburg SLB test is commonly used to assess hip abductor function and the procedure described by Hardcastle et al. [38] was utilized in this study [39]. Participants were instructed to place their hands on their hips and raise their non-THA leg off the ground until the hip was flexed at 30 degrees. Once balance was achieved, the participant was asked to raise the non-stance side of the pelvis as high as possible while maintaining a vertical balance position. Participants were instructed to hold that position for 30 seconds. Time was recorded using a hand held stopwatch from the time the participant established the elevated hip position until balance was lost or 30 seconds was reached. Three trials were taken on each leg and the maximum time for each leg was used for analysis [38, 39].
Timed-Up-and-Go Test (TUG)
The TUG is a test commonly used to assess physical mobility in the elderly population [40]. During the TUG, participants were instructed to sit in the chair (base height of 45 cm) with their back against the backrest, their arms relaxed on the armrests, and their feet shoulder width apart on the ground. On a “go” command, the participant rose from the chair, walked 3 m, turned around, walked back to the chair, and returned to a seated position. Time was recorded with a handheld stopwatch from the “go” command to the time the participants back returned to the backrest of the chair [40]. Three trials were recorded and the best score was used in analysis.
Handheld Dynamometer
Handheld Dynamometers (HHD) are commonly used in the clinical setting because of their ability to quantify strength quickly and simply, however there has been uncertainty over their validity and reliability [41]. The muscle group being tested, the strength of the patient and the strength of the examiner may all have an effect on the accuracy of the measurement [41–43]. Researchers have found however, that hand-held dynamometers are valid and reliable for assessing hip strength in the older adult population [41–43]
Participants completed a series of isometric hip musculature strength measures (flexion, extension, abduction, adduction, internal, and external rotation) using a MicroFET hand held dynamometer (HHD) (Hoggan Health Industries, West Jordan, UT, USA). Participants were instructed to hold each contraction for three-seconds, reaching maximal contraction at three seconds [44]. Following a submaximal familiarization trial, three maximal trials were collected, with 60 seconds of rest between trials [45–47]. All measurements were conducted bilaterally and the maximum score for each movement was used foranalysis.
Statistical analysis
The results of the Trendelenburg SLB and the TUG tests were recorded in seconds and were used for within- and between-subject analysis. The isometric strength outcomes were recorded bilaterally in newtons, and were normalized to body weight. Hip muscle weakness was estimated using within-subject or between-subject comparisons [48]. Within-subject comparisons for all isometric strength were analyzed by calculating the asymmetry or strength deficit between the affected (A) and non-affected (NA) legs [(A –NA)/NA×100] [49]. Between-subject comparisons were made where the involved leg for the DA patients and the average of both legs of the control group were used in analysis [48, 50].
Descriptive statistics including means and standard deviations were generated for all demographic characteristics and for the HHS. A Shapiro-Wilk test for normal distribution was conducted on each dependent variable and determined the isometric strength data to be normally distributed, but the Trendelenburg SLB test and the TUG were not normally distributed. Individual two-way analyses of variance with repeated measures were conducted on the individual hip isometric muscle strength values to identify within subject changes and between group differences. Freidman’s two-way analysis of variance was used to identify within subject changes and the Kurskal-Wallis test was used to determine between group differences for the Trendelenburg SLB test and the TUG. Post hoc analysis was performed on significant main effects. All statistical analysis was completed using SPSS v19 (IBM Corporation, Armonk, NY, USA). The significance level was set at p≤0.05.
Results
Harris Hip Score
Means and standard deviations were calculated for the HHS in the DA group (Table 2). According to Nilsdotter et al. [51] the higher the score, the less amount of dysfunction. A score less than 70 is considered poor, between 70–80 is fair, between 80–90 is good, and between 90–100 is excellent [51]. At pre operation, no participants recorded a score of excellent. By three weeks post 10%, six weeks post 50%, and three months post 80% of the DA group recorded scores of excellent.
Functional tests
There was a significant main effect for time for the Trendelenburg SLB test (p = 0.05), however post hoc testing revealed no between group differences at any data collection point. There was also a significant main effect for time for the TUG (p = 0.003) (Fig. 1). Post hoc testing revealed significant improvement in TUG between pre operation and six weeks post operation (p = 0.02) and between three and six weeks post operation (p = 0.03) (Table 2, Fig. 2).
Max isometric muscle strength
At the pre operation, three and six weeks post operation hip flexion, extension, abduction, internal rotation, and external rotation were significantly weaker than the control group. At the three month post operation data collection session, only hip external rotation remained significantly weaker in the DA group compared to the control group (p = 0.04; DA: 0.86 ± 0.54 N/BW, Control: 1.22 ± 0.37 N/BW) (Fig. 3). There were no longer significant differences observed in the other muscle groups (Table 3).
Analysis of within subject data over time revealed a decrease in all strength variables between pre-operation and three weeks post operation, with hip flexion (p < 0.00), extension (p = 0.02), and external rotation (p < 0.00) being significantly weaker at three weeks post than at pre operation. From the three to six week post operation data collection sessions, hip extension (p = 0.01) and adduction (p < 0.00) demonstrated a significant increase in strength. From the six week to three-month data collection sessions, there were no statistical increases in max isometric hip strength in any strength variable (Table 3).
Asymmetry
At pre operation, three and six weeks post operation, hip flexion, adduction, and internal rotation were significantly more asymmetric compared to the control groups asymmetry scores, indicating the weakness of affected leg compared to the non-affected leg. Additionally, at the three and six week post operation time points, hip abduction also showed significantly greater asymmetry (3 weeks: p < 0.00, DA: –26.41 ± 19.3, Control: 3.47 ± 21.21; and 6 weeks: p = 0.04, DA: –16.97 ± 17.08, Control: –5.14 ± 14.45). At the three months post operation data collection period, only hip flexion (p = 0.03, DA: –1.11 ± 2.04, Control: 0.55 ± 2.24) asymmetry was significantly greater when compared to the control group (Fig. 4). The other hip asymmetry scores were no longer different from the control group at this post operation time point (Table 4).
Analysis of within-subject data at pre operation compared to three weeks post operation revealed a statistically significant asymmetry for hip flexion (p = 0.03) and external rotation (p = 0.04). Between the three and six week post operation periods, hip flexion showed significant decrease in asymmetry (p = 0.05). Between the six week and three month post operation data collection sessions, both hip flexion (p = 0.05) and hip abduction (p < 0.00) showed significant decreases in asymmetry.
Discussion
This study was conducted to assess isometric hip muscle strength and functional ability in the DA population to determine the recovery of normal function and strength following the DA THA. At pre-operation, all hip isometric hip measures except for hip adduction were significantly lower than the control group. Additionally, hip flexion, adduction, and internal rotation were significantly more asymmetrical than the control group, with the affected leg being significantly weaker than the non-affected leg. However, by three months post operation, only hip external rotation remained significantly weaker in the DA group and hip flexion remained significantly more asymmetric.
The significant asymmetric difference in hip flexion at three months post operation may be the result of the DA procedure. The DA is a tissue sparing procedure designed to minimize damage to soft tissue structures during surgery. During the DA, the sartorius and rectus femoris are retracted medially while the tensor fascia latae is retracted laterally. Although these muscles are not incised during the procedure, authors have found that damage still occurs to these structures, reporting a range of 18.31–58.48% damage to the tensor fascia latae muscular area during the DA in cadavers, and a mean of 12.24% damage to the muscular area of the rectus femoris [11, 24]. The damage to these hip flexors during surgery may have resulted in the decreased force production and asymmetry observed in this study.
This strength discrepancies recorded at pre-operation may be the result of disuse atrophy and/or pain. Hip OA is a disease that can progress over years and commonly causes pain in the hip joint due to degeneration of the bone and articular cartilage [52]. Patients who are suffering from hip OA tend to live with this pain until it becomes too severe to continue with activities of daily living and THA is the only option [52]. To minimize the pain at the hip joint, patients often develop an antalgic gait or limp. These gait alterations minimize pain around the hip joint by decreasing the force that the hip musculature has to exert to maintain pelvic stabilization [53]. Over time, the hip musculature can become atrophied due to the lack of forceful contraction when weight bearing [54]. This disuse atrophy and the pain experienced from muscle contraction around the affected hip may be the reason for the significant differences in hip isometric strength and asymmetry between groups at pre-operation. It may also help explain why the hip external rotators remained significantly weaker in the DA group at three months post operation, even though they were not incised during the surgical procedure.
Hip abduction strength has historically been the primary concern following THA, where it has been documented to be significantly weaker than control groups up to ten years post operation [29]. The hip abductor musculature is responsible for maintaining pelvic stabilization during single leg stance of gait [53, 54]. Many THA procedures access the hip joint through an area where the least dissection of the gluteus medius or minimus is necessary [55]. During the DA the gluteus medius and minimus are not required to be incised to access the hip joint and should not be damaged during this surgery [21]. Research conducted on cadaver supports this idea that there is limited damage to the hip abductors during DA [11, 24]. In the current study, by six weeks post operation, hip abduction strength surpassed the pre operation values indicating that pain or disuse atrophy may have been a factor for the initial weakness. When looking at the asymmetry values for hip abduction, the affected leg of the DA participants was significantly weaker at three and six weeks post operation when compared to controls, however there was significant improvement between six weeks and three months, and was no longer statistically asymmetrical at three months post operation. It is difficult to determine if the decreased hip abductor strength is a result of disuse atrophy pre-surgery, pain from OA or post-surgical pain, or a result of muscle damage from the DA. Regardless of the cause, the hip abductor strength asymmetry was restored similar to that of control group by three months postoperation.
There were no significant differences found between groups for the Trendelenburg SLB or the TUG. Although not statistically significant, the DA group demonstrated continual improvement in scores throughout the three-month data collection period. The lack of difference especially at the three week post operation period may indicate that the damage that occurs during surgery may not have a great effect on functional ability in the DA population.
The primary limitation of this study is that the post-operative rehabilitation was not standardized or monitored. All DA participants were advised to attend post operation physical therapy; however this was not controlled in this study. This study also did not control for hip implant type nor measure implant placement. Instead of matched controls, this study used a group with a similar age and anthropometric measures. Finally, the last follow-up period was three months post operation. This early post operation follow-up period may not have been long enough after surgery to determine normal function.
Rehabilitation following THA has shown to improve functional outcomes in patients; however there seems to be a lack of consistency in rehabilitation protocol [56]. This may be a result of the varying types of THA procedures performed. The results of this study indicate that in addition traditional rehabilitation techniques, emphasis on improving hip flexion and external rotation strength may be valuable in regaining normal strength and function in the DA THA population. In conclusion, by three months post operation, DA patients regained normal strength in all ranges except for hip external rotation, and no longer had asymmetric strength differences between their affected and unaffected legs in any ranges except for hip flexion.
Conflict of interest
None.