MRI assessment of muscle damage after posterolateral versus SuperPATH approach for total hip arthroplasty Authors: Fengping Gan, Qibiao Zhang, Haoji Luo, Linjie Li, Xinxin Lin, Hao Qin Published: 2026 ABSTRACT AbstractThe minimally invasive SuperPATH approach for total hip arthroplasty (THA) is theorized to reduce soft tissue damage, but objective evidence comparing muscle injury to the conventional posterolateral approach (PLA) is lacking. This study aimed to objectively compare postoperative periacetabular muscle damage between the two approaches using magnetic resonance imaging (MRI). Hypothesis: The SuperPATH approach demonstrates superior efficacy in protecting posterior hip muscles compared to the PLA. In this retrospective comparative cohort study, patients undergoing primary unilateral THA for osteonecrosis or osteoarthritis by a single surgeon were enrolled. Patients with hip dysplasia, BMI > 35 kg/m², history of cerebrovascular disease or cognitive dysfunction or prior ipsilateral hip surgery were excluded. The patients were divided into two groups based on the surgical approach: the SuperPAHT group and the PLA group.The primary outcome was muscle damage assessed on MRI at 12 months postoperatively, including fatty infiltration (using the modified Goutallier classification) and muscle atrophy (via cross-sectional area or maximal transverse diameter measurements). Secondary outcomes included perioperative variables, Visual Analog Scale (VAS) pain scores, and Harris Hip Scores (HHS). At 12 months, MRI revealed no significant intergroup differences in the degree of fatty infiltration or atrophy for the piriformis, obturator internus, or gluteus minimus muscles. However, the SuperPATH approach provided superior protection for the quadratus femoris, demonstrating significantly less fatty infiltration (7.3% vs. 85.4%, p < .05), preserved muscle bulk, and maintained tendon continuity in all cases (100% vs. 56.1%, p < .05) compared to PLA. Tendon discontinuity was significantly more frequent in the PLA group for the piriformis (61.0% vs. 36.7%, p = .027), obturator internus (39.0% vs. 9.8%, p = .002), and quadratus femoris (43.9% vs. 0%, p < .05) than that in the SuperPATH group.Perioperatively, the SuperPATH group had a longer operative time and greater intraoperative blood loss but a smaller incision length (p < .05). The SuperPATH group exhibited statistically better early VAS and HHS scores with lower VAS pain and higher HHS at 1 month (p < .05), but these differences were not significant at the 12-month follow-up. Two early dislocations occurred in the PLA group, with none in the SuperPATH group. The SuperPATH effectively preserves the integrity of the quadratus femoris, the obturator internus, and the gluteus minimus muscles, with minimal damage to the quadratus femoris. However, the injuries to the piriformis muscle, the obturator muscle, and the gluteus minimus are comparable to those caused by PLA.Clinical Trial Registeration: This study was registered in the Chinese Clinical Trial Register (ChiCTR2600118828).Keywords: Fatty infiltration, MRI, Total hip arthroplasty, SuperPATH approach, Posterolateral approachSubject terms: Anatomy, Diseases, Health care, Medical research Introduction Total hip arthroplasty (THA) remains one of the most successful orthopedic procedures and an effective treatment for end-stage hip disease1. The demand for THA is increasing worldwide, with patients becoming younger, more active, and seeking faster recovery2. To accelerate postoperative rehabilitation, several interventions have been introduced in recent decades, including fast-track protocols3 and minimally invasive surgical (MIS) approaches4. The proposed advantages of MIS techniques include reduced soft tissue injury, decreased blood loss, shorter hospital stay, and accelerated functional recovery5. The Supercapsular Percutaneously-Assisted Total Hip (SuperPATH) approach accesses the hip joint through the intermuscular interval between the gluteus medius and the piriformis, without dislocating the femoral head. Acetabular preparation and component implantation are performed through a percutaneous working portal, aiming to preserve the posterior joint capsule and short external rotators. Therefore, the SuperPATH technique is considered a tissue-preserving, intermuscular approach for THA. In contrast, the conventional posterolateral approach (PLA) necessitates transection of the posterior hip musculature, including the short external rotators, which is associated with an increased risk of postoperative dislocation6. Some literatures comparing peri-operative and functional outcomes between the SuperPATH and PLA approaches have yielded conflicting results. A recent randomized controlled trial (RCT) by Korytkin et al. reported that the SuperPATH approach resulted in less postoperative pain, superior early gait function, and a higher quality of life compared to the PLA7. While the SuperPATH approach may facilitate faster functional recovery and reduce surgical trauma, it has been associated with longer operative times and potentially lower implant positioning accuracy than the PLA8. Conversely, some studies have questioned its clinical superiority. Research by Khoja et al. found that the SuperPATH approach offered no clear clinical advantage over the posterior approach beyond short-term pain relief and was accompanied by a higher complication rate9. Other comparisons indicated that the SuperPATH approach may require longer operative time, result in greater blood loss and soft-tissue damage, and cause more severe early postoperative pain10. Muscle injury or degeneration is positively correlated with postoperative functional recovery. Evidence suggests that reducing fatty infiltration (FI) in lower-limb muscles, particularly the gluteus maximus and quadriceps, may improve postoperative mobility11. In THA patients, increased FI in the surgical-side gluteus medius and non-surgical-side quadriceps is associated with decreased walking speed at two weeks postoperatively12. Furthermore, gluteus medius tendon degeneration and FI are determinants of persistent limping and a positive Trendelenburg sign after THA13.Magnetic resonance imaging (MRI) provides excellent soft-tissue contrast, is less influenced by operator variability, and allows for clear visualization of periarticular musculature. Consequently, MRI is a reliable tool for objectively assessing the invasiveness of different surgical approaches and evaluating muscle damage following joint arthroplasty14. To the authors’ best knowledge, there are no literature reports using MRI to compare hip muscle damage after THA between the SuperPATH and PLA approaches. Therefore, we conducted a retrospective comparative cohort study to evaluate the differences in the degree of muscle damage between these two approaches.The study hypothesis was that the SuperPATH approach causes less damage to the muscles surrounding the hip (the piriformis, the obturator internus and the quadratus femoris) compared to the PLA approach. Methods Patient selection This retrospective comparative cohort study analyzed clinical data from patients diagnosed with osteonecrosis of the femoral head (ONFH) or hip osteoarthritis (OA) at Guigang People’s Hospital between January 2022 and June 2024, who underwent primary unilateral total hip arthroplasty performed by a single senior physician. A total of 347 consecutive patients underwent THA during the study period. After applying the inclusion and exclusion criteria, 265 patients were excluded: 94 did not undergo both preoperative and postoperative pelvic MRI, 56 were > 75 years of age, 48 had BMI > 35 kg/m², 32 had acetabular dysplasia, 18 had previous surgical intervention on the affected hip, 12 had severe mobility impairment or joint ankylosis, and 5 had history of cerebrovascular disease or cognitive dysfunction. MRI examinations were not performed routinely for all THA patients in our institution. Instead, preoperative and 12-month postoperative pelvic MRI scans were performed according to a predefined research protocol approved by the institutional ethics committee. All eligible patients were prospectively invited to participate in the MRI substudy. The primary reasons for missing MRI data (94 excluded patients) were patient refusal due to financial constraints (MRI is not fully covered by insurance for research purposes in our region), claustrophobia (n = 4), scheduling conflicts, or inability to return for the 12-month follow-up appointment (n = 11). No patients were excluded based on clinical symptoms, surgeon discretion, or postoperative complications.The remaining 82 patients were divided into two groups based on the surgical approach: the SuperPATH group (n = 41) and the PLA group (n = 41). A patient flow diagram was presented in Fig. 1. This study was approved by the Medical Ethics Committee of Guigang People’s Hospital (GYLLPJ-20210813-21) in accordance with the Declaration of Helsinki and was registered in the Chinese Clinical Trial Register (ChiCTR2600118828). Written informed consent was obtained from all participants. CONSORT flow chart. Sample size calculation The sample size was calculated a priori using PASS software (Version 15.0, NCSS, LLC, Kaysville, UT, USA). The primary outcome was the change in CSA of the obturator internus muscle from pre-operative to post-operative assessment. Based on the reported data from a comparable study by Kyrill Rykov et al.15, the mean CSA was − 4.36 cm² for the PLA group and − 2.99 cm² for the Direct anterior approach (DAA) group (used as a proxy for the SuperPATH approach in our calculation). The estimated common standard deviation for the cross-sectional area(CSA) was 2.07 cm². To detect a mean difference of 1.37 cm² in CSA between the two surgical approaches with a two-sided significance level (α) of 0.05 and a statistical power (1-β) of 80%, a minimum of 37 patients per group (74 in total) was required. Accounting for a potential data loss of approximately 10% in this retrospective study and aiming for a more robust power of 90%, we targeted a sample size of 41 patients per group (82 in total). A retrospective review of our institutional database confirmed the availability of sufficient eligible cases within the specified study period. Surgical technique All procedures were performed by the same surgical team. The lead surgeon had extensive experience with the PLA and was beyond the learning curve for the SuperPATH approach (> 200 cases). Anesthesia was either combined spinal-epidural or general endotracheal. Uncemented acetabular and femoral components were used in all cases. SuperPATH Approach: The technique was performed as described by Chow et al.16. The patient was positioned laterally with the operative hip flexed at 45° and internally rotated 10°–15°. A 6–8 cm incision was made from the tip of the greater trochanter proximally along the femoral shaft axis. After splitting the gluteus maximus fibers, the interval between the gluteus medius and piriformis was developed. The capsule was incised longitudinally. If necessary, the piriformis tendon was released from its trochanteric insertion. Femoral preparation proceeded through the piriformis fossa. A percutaneous working portal was established for acetabular reaming and cup insertion. The capsule was sutured, the piriformis tendon was reattached if released, and the wound was closed in layers. Posterolateral Approach: A standard 15 cm incision was made centered slightly posterior to the greater trochanter. The fascia lata and gluteus maximus were split. The short external rotators (piriformis, obturator internus/gemelli, quadratus femoris) were transected at their femoral insertions using electrocautery. The capsule was incised in a T-shape, the hip was dislocated, and a femoral neck osteotomy was performed. After acetabular and femoral preparation, the components were implanted. The hip was reduced, and stability was assessed. The transected external rotators were repaired through bone tunnels. No surgical drain was used in either group. All patients received prophylactic intravenous cefuroxime preoperatively and thromboprophylaxis with low-molecular-weight heparin or rivaroxaban postoperatively. Weight-bearing was initiated on postoperative day 1 with crutches or a walker, progressing to full weight-bearing as tolerated after 2–4 weeks. Data collection MRI assessment MRI (Siemens 3.0T MAGNETOM Skyra magnetic resonance imaging system, Germany) was performed before the surgery and 12 months after surgery using a 3.0 Tesla scanner with 3.5 mm slice thickness. Protocol consisted of metal artifact reduction sequences coronal STIR and T2W through the whole pelvis.MRI scans were analyzed by 2 observers (LJ.L. and HJ.L.). One was a orthopedic surgeon, and the other was a musculoskeletal radiologist. Both of them were not involved in the surgery and unaware of the surgical approach.Measurements of the muscle surface and FI were performed on t2-tse-dixon-tra-WARP-in sequence images. Muscle FI was assessed using the Goutallier grading system with 5 grades from 0 (no FI) to 4: Grade 0 (normal), 1 (minimal streaks), 2 (fat < muscle), 3 (fat = muscle), 4 (fat > muscle). For the analysis of muscle FI, the Goutallier classification was dichotomized into two grades: a Goutallier grade of 0 or 1 was considered as no or mild FI and a grade ≥ 2 as moderate or severe FI15.Muscle surface measurement was quantified by measuring CSA for the psoas, gluteus medius and minimus, and the maximal transverse diameter(MTD) for the piriformis, obturator internus, and quadratus femoris due to their irregular morphology. Freehand region of interest muscle surface of the psoas muscle was measured at the anterior superior iliac spine level. Quadratus femoris was measured at the level of the intertrochanteric ridge. Piriformis muscle was measured at at the level of 15 mm from the superior margin of the acetabulum, and gluteus minimus, medius at at the level of 20 mm from the superior margin of the acetabulum. Measurement of the obturator internus muscle at the level of the lacrimal sac.(Fig. 2). The locations for measuring the muscle the CSA, MTD and FI in the muscles surrounding the hip joint were evaluated using the t2-tse-dixon-tra-WARP-in sequence. (A) The psoas was measured at the level of the anterior superior iliac spine. (B)The piriformis muscle was measured at a level 15 mm from the superior margin of the acetabulum. (C) The gluteus minimus and medius were measured at a level 20 mm from the superior margin of the acetabulum. (D) The obturator internus was measured at the level of the lacrimal sac. (E) The quadratus femoris was measured at the level of the intertrochanteric ridge. To evaluate the reliability of CSA, MTD, and FI measurements, test-retest reliability was analyzed. Interobserver and intraobserver reliabilities were assessed using 30 randomly selected MRI scans (15 from each group). Two independent observers (L.J. Li and F. Zheng) who were blinded to clinical data and surgical approach independently measured CSA/MTD and graded FI on two separate occasions, with a minimum interval of 4 weeks between readings. Intraclass correlation coefficients (ICC) with 95% confidence intervals were calculated for continuous variables (CSA and MTD), and weighted Kappa coefficients (κ) with 95% confidence intervals were calculated for categorical variables (Goutallier grades). ICC values were interpreted as follows: <0.5 = poor, 0.5–0.75 = moderate, 0.75–0.9 = good, > 0.90 = excellent. Kappa values were interpreted as: <0 = poor, 0.01–0.20 = slight, 0.21–0.40 = fair, 0.41–0.60 = moderate, 0.61–0.80 = substantial, 0.81-1.00 = almost perfect. Perioperative outcomes Operative time (skin incision to closure), intraoperative blood loss, total calculated blood loss (using Gross’s formula), and length of hospital stay (LOS) were recorded. Serum creatine kinase (CK) and lactate dehydrogenase (LDH) levels were measured preoperatively and on postoperative day 3. Functional outcomes were assessed using the Harris Hip Score (HHS) preoperatively and at the 12-month follow-up. Pain was evaluated with the Visual Analog Scale (VAS) preoperatively, and at 3 days, 1 month, and 1 year postoperatively. All complications were documented. Statistical analysis Data were analyzed using SPSS version 22.0 (IBM Corp.). Continuous variables are presented as mean ± standard deviation or median (interquartile range), and categorical variables as frequencies (%). Normality was assessed using the Shapiro-Wilk test. Student t-test was used for normally distributed continuous clinical parameters, otherwise the Mann-Whitney U-test was used. Two-way mixed ANOVA was performed to assess postoperative differences in muscle FI, CSA and MTD between the SuperPATH and PLA.Between-group comparisons for continuous variables were made using the independent samples t-test or Mann-Whitney U test. Categorical variables were compared using the Chi-square or Fisher’s exact test. Within-group changes from preoperative to postoperative values were analyzed using paired t-tests or Wilcoxon signed-rank tests. A two-tailed p-value < 0.05 was considered statistically significant. Results The demographic data of the patients before surgery are shown in Table 1.There were no statistically significant differences between the two groups regarding demographic characteristics, including gender, age, BMI, diagnosis, ASA classification, or comorbidities . Table 1Demographic of the study cohorts.CharacteristicSuperPATH (n = 41)PLA (n = 41)P-ValueAge (years)a59.2 ± 9.156.9 ± 11.20.320Gender (Male/Female)b22 / 1920 / 210.659BMI (kg/m²)a24.0 ± 3.724.5 ± 4.00.496ASA Grade (I/II)b18 / 2320 / 210.658Comorbiditiesb20180.658Diagnosis (ON/OA)b24 / 1728 / 130.359BMI body mass index; ASA American Society of Anesthesiologists; ON osteonecrosis; OA osteoarthritis; PLA posterolateral approach; SuperPATH Supercapsular Percutaneously-Assisted Total Hip.aData presented as mean ± standard deviation; bData presented as number of patients. Demographic of the study cohorts. BMI body mass index; ASA American Society of Anesthesiologists; ON osteonecrosis; OA osteoarthritis; PLA posterolateral approach; SuperPATH Supercapsular Percutaneously-Assisted Total Hip. aData presented as mean ± standard deviation; bData presented as number of patients. Reliability of MRI measurements Interobserver and intraobserver reliability for MRI measurements are presented in Table 2. For CSA and MTD measurements, interobserver ICC values ranged from 0.85 to 0.92, indicating good to excellent reliability. Intraobserver ICC values ranged from 0.79 to 0.93 for observer 1 and 0.87 to 0.94 for observer 2. For Goutallier grading of fatty infiltration, interobserver weighted Kappa values ranged from 0.87 to 0.94, indicating substantial to almost perfect agreement. Intraobserver weighted Kappa values ranged from 0.76 to 0.92 for observer 1 and 0.88 to 0.94 for observer 2. These results confirm that our MRI-based measurements have good to excellent reliability. Table 2Interobserver reliablity and intraobserver reliability.M-CSA/MTDFIInterobserver ReliabilityIntraobserver ReliabilityInterobserver ReliabilityIntraobserver ReliabilityObserver1Observer2Observer1Observe2G.med0.850.920.870.870.920.94G.mini0.920.930.900.930.760.90Piriformis0.880.920.940.910.860.91OI0.850.880.860.900.840.88QF0.870.790.870.940.890.90Interobserver and intraobserver reliabilities for M-CSA/MTD calculated using intraclass correlation coefficient.Interobserver and intraobserver reliabilities for FI calculated using Kappa test.M-CSA muscle cross-sectional area; FI fatty infiltration; Gmed gluteus medius; Gmini gluteus minimus; OI obturator internus; QF quadratus femoris. Interobserver reliablity and intraobserver reliability. Interobserver and intraobserver reliabilities for M-CSA/MTD calculated using intraclass correlation coefficient.Interobserver and intraobserver reliabilities for FI calculated using Kappa test. M-CSA muscle cross-sectional area; FI fatty infiltration; Gmed gluteus medius; Gmini gluteus minimus; OI obturator internus; QF quadratus femoris. Fatty infiltration The muscle FI is shown in Table 3.Compared with preoperative levels, both groups showed significantly increased muscle FI in the piriformis, the obturator internus, and the gluteus minimus muscles at 1 year postoperatively.At 1 year postoperatively, muscle FI was observed in 20 cases (48.8%) in the SuperPATH group and 25 cases (61.0%) in the PLA group for the piriformis muscle. In the obturator internus muscle, muscle FI was detected in 30 cases (73.2%) in the SuperPATH group and 36 cases (87.8%) in the PLA group. In the gluteus minimus muscle, muscle FI was found in 32 cases (78.0%) in the SuperPATH group and 28 cases (68.3%) in the PLA group.There was no significant difference between the two groups in the FI of piriformis muscle, obturator internus muscle and gluteus minimus muscle at 1 year postoperatively. Table 3Comparison of postoperative muscle fat infiltration in accordance with the dichotomized Goutallier Score.MusclesSuperPATH (n = 41)PLA (n = 41)P- valuePiriformisYesNoYesNo.Preoperative1 (2.4%)40 (97.6%)2 (4.9%)39 (95.1%).1y-postoperative20(48.8%)21 (51.2%)25(61.0%)16(39%)0.286.P-value <0.05 <0.05 Obturator Internus.Preoperative0 (0%)41 (100%)1 (2.4%)40 (97.6%).1y-postoperative30 (73.2%)11 (26.8%)36(87.8%)5(12.2%)0.125.P-value <0.05 <0.05 Quadratus Femoris.Preoperative0 0%(0%)41 (100%)1 (2.4%)40 (97.6%).1y-postoperative3 (7.3%)38 (92.7%)35(85.4%)6(14.6%) <0.05 .P-value0.250 <0.05 Gluteus Medius.Preoperative1 (2.4%)40 (97.6%)0 (0%)41 (100%).1y-postoperative3 (7.3%)38 (92.7%)4(9.8%)37(90.2%)1.000.P-value0.5000.125Gluteus Minimus24 (96.0).Preoperative2 (4.9%)39 (95.1%)1 (2.4%)40 (97.6%).1y-postoperative32 (78.0%)9 (22.0%)28(68.3%)13(31.7%)0.476.P-value <0.05 <0.05 Psoas.Preoperative1(2.4%)41(97.6%)2(4.9%)39(95.1%).1y-postoperative1(2.4%)41(97.6%)2(4.9%)39(95.1%)1.000.P-value1.0000.494Data are presented as N [%].SuperPATH, Supercapsular Percutaneously-Assisted Total Hip, PLA posterolateral approach.aA Goutallier grade of 0 or 1 was considered as “No” and a grade 2–4 as “Yes”. Comparison of postoperative muscle fat infiltration in accordance with the dichotomized Goutallier Score. Data are presented as N [%]. SuperPATH, Supercapsular Percutaneously-Assisted Total Hip, PLA posterolateral approach. aA Goutallier grade of 0 or 1 was considered as “No” and a grade 2–4 as “Yes”. However, compared with preoperative levels, the FI in the gluteus medius muscle slightly increased in both groups, but no statistically significant difference was observed. There was no significant difference between the two groups in the FII of gluteus medius muscle at 1 year postoperatively.The FI in the psoas muscle showed no change. The muscle FI in the quadratus femoris was significantly higher in the PLA group than in the SuperPATH group, with 35 cases (85.4%) in the PLA group versus 3 cases (7.3%) in the SuperPATH group. Muscle atrophy and tendon integrity The MTD or CAS of the muscles surrounding the hip joint is shown in Table 4.One year postoperatively, significant muscle volume decreasing was observed in the piriformis, obturator internus, and gluteus minimus muscles in both groups, with a marked reduction in muscle MTD or CSA. No muscle atrophy was noted in the gluteus medius, while the CSA of the psoas increased in both group. The MTD of the quadratus femoris muscle increased from an average of 1.55 cm pre - operatively to 1.58 cm post - operatively in the SuperPATH group. In contrast, it decreased from an average of 1.58 cm pre - operatively to 1.40 cm post - operatively in the PLA group. Table 4Comparison of Muscle Atrophy.MuscleSuperPATH (n = 41)PLA(n = 41)P- valuePiriformisPreoperatie1.99 ± 0.361.89 ± 0.32post-operative1.75 ± 0.341.69 ± 0.290.605P-value < 0.05 < 0.05 Obturator InternusPreoperatie2.11 ± 0.292.28 ± 0.41post-operative1.85 ± 0.382.00 ± 0.760.254P-value (Intra) < 0.05 < 0.05 Gluteus MinimusPreoperatie14.44 ± 2.8614.97 ± 2.62post-operative13.31 ± 2.6013.52 ± 2.600.715P-value < 0.05 < 0.05 Gluteus MediusPreoperatie24.86 ± 2.5525.58 ± 2.94post-operative24.93 ± 2.3525.83 ± 3.020.134P-value0.6170.111Quadratus FemorisPreoperatie1.55 ± 0.341.58 ± 0.34post-operative1.58 ± 0.321.40 ± 0.30 0.008 P-value < 0.05 < 0.05 PsoasPreoperatie11.06 ± 2.1511.89 ± 2.27post-operative12.99 ± 1.8913.67 ± 1.980.220P-value < 0.05 < 0.05 The CSA of the psoas major, gluteus minimus, and gluteus medius was measured, while the MTDof the piriformis, internal obturator, and quadratus lumborum was measured. Data are presented as Mean ± SD.PLA posterolateral approach; SuperPATH Supercapsular Percutaneously-Assisted Total Hip. CSA cross-sectional area; MTD maximal transverse diameter. Comparison of Muscle Atrophy. The CSA of the psoas major, gluteus minimus, and gluteus medius was measured, while the MTDof the piriformis, internal obturator, and quadratus lumborum was measured. Data are presented as Mean ± SD. PLA posterolateral approach; SuperPATH Supercapsular Percutaneously-Assisted Total Hip. CSA cross-sectional area; MTD maximal transverse diameter. A representative case for SuperPATH is a 52-year-old male with Association Research Circulation Osseous (ARCO) 4 grade due to right ONFH. After right THA through a SuperPATH was performed, the gluteus minimus, the obturator internus, and the piriformis showed a decrease in CSA or MTD and an increase in FI.The psoas, quadratus femoris and the gluteus medius showed an increase in CSA or MTD (Fig. 3). Preoperative and 12-month postoperative 2-weighted MRI axial scans of the bilateral hip joint. (A,C,E,G,I) and (B,D,F,H,J) show the preoperative and 12-month postoperative images, respectively. The representative case in the Super’PATH group was a 52-year-old male with right femoral head necrosis who underwent total hip arthroplasty via the Super’PATH approach. The changes in CSA and MTD values before and after surgery were as follows: A and B showed the changes in the CSA of the psoas, from 8.02 cm² before surgery to 9.73 cm² after surgery. C and D showed the changes in the CAS of the gluteus minimus and gluteus medius, with the gluteus minimus decreasing from 10.79 cm² before surgery to 10.35 cm² after surgery, and the gluteus medius decreasing from 21.67 cm² before surgery to 23.21 cm² after surgery. E and F showed the changes in the MTD of the piriformis muscle before and after surgery, from 1.48 cm before surgery to 1.35 cm after surgery. G and H showed the changes in the MTD of the obturator internus before and after surgery, from 1.99 cm before surgery to 1.38 cm after surgery. I and J showed the changes in the MTD of the quadratus femoris before and after surgery, from 1.35 cm before surgery to 1.42 cm after surgery. A representative case for PLA is a 70-year-old male with ARCO 4 grade due to right ONFH. After right THA through a PLA was performed, the gluteus minimus, the obturator internus, the piriformis, the quadratus femoris and the gluteus medius showed a decrease in CSA or MTD and an increase in FI.The psoas showed an increase in CSA (Fig. 4). Preoperative and 12-month postoperative 2-weighted MRI axial scans of the bilateral hip joint. (A,C,E,G,I) and (B,D,F,H,J) show the preoperative and 12-month postoperative images, respectively. The representative case in the Super’PATH group was a 70-year-old male with right femoral head necrosis who underwent total hip arthroplasty via the Super’PATH approach. The changes in CSA and MTD values before and after surgery were as follows: A and B showed the changes in the CSA of the psoas, from 4.69 cm² before surgery to 4.83 cm² after surgery. C and D showed the changes in the CAS of the gluteus minimus and gluteus medius, with the gluteus minimus decreasing from 10.98 cm² before surgery to 10.83 cm² after surgery, and the gluteus medius decreasing from 23.02 cm² before surgery to 20.87 cm² after surgery. E and F showed the changes in the MTD of the piriformis muscle before and after surgery, from 1.24 cm before surgery to 0.91 cm after surgery. G and H showed the changes in the MTD of the obturator internus before and after surgery, from 1.99 cm before surgery to 1.38 cm after surgery. I and J showed the changes in the MTD of the quadratus femoris before and after surgery, from 1.35 cm before surgery to 1.42 cm after surgery. Tendon integrity is shown in Table 5.Tendon discontinuity was significantly more frequent in the PLA group for the piriformis (61.0% vs. 36.7%, p = .027), obturator internus (39.0% vs. 9.8%, p = .002), and quadratus femoris (43.9% vs. 0%, p < .001) than that in the SuperPATH group. No significant difference was found for the gluteus minimus. Table 5Comparison of postoperative Muscle-Tendon Continuity.MuscleSuperPATH(n = 41)PLA (n = 41)P-valueYesNoYesNoPiriformis26(63.4%)15(36.7%)16 (39.0%)25(61.0%) 0.027 Obturator Internus37(90.2%)4(9.8%)25(61.0%)16(39.0%) 0.002 Quadratus Femoris41(100%)0(0%)23(56.1%)18(43.9%) < 0.001 Gluteus Minimus38(92.7%)3(7.3%)39(95.1%)2(4.9%)1.000Date are presented as N[%].PLA, posterolateral approach; SuperPATH, Supercapsular Percutaneously-Assisted Total Hip. “Yes” indicates preserved tendon continuity; “No” indicates tendon discontinuity. Comparison of postoperative Muscle-Tendon Continuity. Date are presented as N[%].PLA, posterolateral approach; SuperPATH, Supercapsular Percutaneously-Assisted Total Hip. “Yes” indicates preserved tendon continuity; “No” indicates tendon discontinuity. Perioperative and clinical outcomes Perioperative data and clinical outcomes are presented in Table 6.The SuperPATH group had a significantly longer operative time and greater intraoperative blood loss but a smaller incision length compared to the PLA group (p < .05). Total blood loss and LOS were similar. Serum CK and LDH levels increased significantly on postoperative day 1 in both groups, with no intergroup differences.Both groups showed significant improvements in VAS and HHS postoperatively. At the 1-month follow-up, the SuperPATH group had significantly lower VAS scores and higher HHS than the PLA group (p < .05). However, these differences were not sustained, and their clinical importance was not formally evaluated against MCID thresholds.These differences were no significant at the 12-month assessment. Two early dislocations occurred in the PLA group; none occurred in the SuperPATH group.The surgical incisions of both groups healed as grade A. No thrombosis, peri - prosthetic infection, or peri - prosthetic fracture was found. Table 6Comparison of perioperative outcomes.VariableSuperPATH (n = 41)PLA(n = 41)P -valueOperative time (min)75.5 ± 6.663.8 ± 5.5 < 0.001 LOS (d)5.8 ± 1.05.7 ± 1.20.620Incision length (cm)7.98 ± 0.914.4 ± 1.6 < 0.001 Intraoperative blood loss (mL)263.0 ± 35.8239.8 ± 37.0 0.005 Total blood loss (mL)624.5 ± 77.8614.0 ± 77.60.546CK (U/L)Preoperative77.9 ± 32.386.4 ± 31.00.228Postoperative416.1 ± 146.1462.5 ± 201.60.237LDH (U/L)Preoperative170.4 ± 34.2166.0 ± 39.20.589Postoperative614.3 ± 165.0554.8 ± 169.50.111VASPreoperative6.0 ± 0.85.8 ± 1.00.4041 month postoperatively1.3 ± 0.61.7 ± 0.6 0.003 12 months postoperatively0 (0, 0)0 (0, 1)0.195HHSPreoperative46.9 ± 8.145.2 ± 9.90.4641 month postoperatively83.7 ± 2.476.0 ± 3.2 < 0.001 12 months postoperatively91.1 ± 2.792.3 ± 3.30.743Data are presented as Mean ± SD.PLA, posterolateral approach; SuperPATH, Supercapsular Percutaneously-Assisted Total Hip. LOS, length of hospital stay; CK, Creatine kinase; LDH, lactic dehydrogenase; VAS, Visual Analog Scale; HHS, Harris hip score. Comparison of perioperative outcomes. Data are presented as Mean ± SD. PLA, posterolateral approach; SuperPATH, Supercapsular Percutaneously-Assisted Total Hip. LOS, length of hospital stay; CK, Creatine kinase; LDH, lactic dehydrogenase; VAS, Visual Analog Scale; HHS, Harris hip score. Discussion Our findings demonstrate that the muscle-sparing benefit of the SuperPATH approach, as assessed by MRI at 12 months, is largely confined to the quadratus femoris. For this muscle, SuperPATH resulted in significantly less fatty infiltration, preserved muscle bulk, and intact tendon continuity in all cases, compared to the PLA group. In contrast, for the piriformis, obturator internus, and gluteus minimus, both approaches caused comparable degrees of fatty infiltration and atrophy, with no significant intergroup differences. Thus, our hypothesis that SuperPATH causes less damage to all posterior hip muscles was only partially supported. The success of THA depends not only on long-term implant survivorship but also on the extent of perioperative soft-tissue damage and the quality of functional recovery. While intermuscular MIS approaches are theoretically less traumatic, systematic reviews have found their overall revision rates, dislocation risks, and mid-term functional outcomes to be equivalent to those of conventional approaches17. The purported muscle-sparing benefit of MIS techniques thus remains partially theoretical and requires objective validation. Our findings provide novel insights into the differential effects of SuperPATH and PLA on periarticular muscle integrity and early functional recovery. The preservation of muscle-tendon continuity emerged as a key distinguishing feature of the SuperPATH approach. Specifically, the SuperPATH group demonstrated significantly higher rates of preserved tendon continuity for the piriformis (63.4% vs. 39.0%, P = .027), obturator internus (90.2% vs. 61.0%, P = .002), and quadratus femoris (100% vs. 56.1%, P < .001) compared to the PLA group (Table 4). This aligns with the fundamental design of the SuperPATH technique, which utilizes an intermuscular and internervous plane between the gluteus medius and piriformis, thereby avoiding the need for transection of the external rotators that is typically required in the PLA17. The quadratus femoris, in particular, is frequently injured during PLA due to its close proximity to the posterior acetabulum18, and its preservation in the SuperPATH group is noteworthy as this muscle plays a crucial role in hip stability and preventing posterior dislocation19,20. The near-complete preservation of tendon continuity in the SuperPATH group strongly supports the muscle-sparing potential of this approach at a structural level, which had previously been inferred from clinical observations but lacked detailed radiological confirmation. In terms of muscle atrophy, as measured by MTD and CSA, both groups exhibited significant intra-group reductions in several muscles postoperatively, consistent with the known phenomenon of disuse atrophy following THA11,21. However, inter-group comparisons revealed nuanced differences. For the quadratus femoris MTD, the SuperPATH group showed a slight, non-significant increase from preoperative to postoperative measurements (1.55 ± 0.34 to 1.58 ± 0.32 cm, P = .007), whereas the PLA group experienced a significant decrease (1.58 ± 0.34 to 1.40 ± 0.30 cm, P < .05), resulting in a significant inter-group difference in postoperative MTD (P = .008). This finding is particularly striking when considered alongside the 100% tendon continuity preservation in the SuperPATH group for this muscle, suggesting that maintaining structural integrity may mitigate atrophy, possibly by preserving muscle activation patterns and reducing denervation injury. In contrast, the piriformis and obturator internus displayed significant intra-group atrophy in both cohorts, with no significant inter-group differences in postoperative measurements. This may be attributed to the fact that while SuperPATH avoids transection of these muscles, they may still experience retraction, edema, or iatrogenic irritation during the approach, leading to some degree of functional impairment and subsequent atrophy22,23. The gluteus medius and minimus, which are critical for hip abduction and gait stability24, showed minimal and non-significant changes in CSA in both groups, with no inter-group differences. This is reassuring, as both approaches aim to preserve the abductor mechanism, and our data confirm that neither results in substantial atrophy of these key muscles, which is consistent with previous reports25. The observed differences in muscle-tendon continuity and quadratus femoris atrophy likely contributed to the significant advantages favoring the SuperPATH group at the 1-month follow-up. At this time point, the SuperPATH group had significantly lower VAS scores (1.3 ± 0.6 vs. 1.7 ± 0.6, P = 0.003) and higher HHS (83.7 ± 2.4 vs. 74.4 ± 11.2, P < 0.001). However, it is important to note that both VAS and HHS are composite measures substantially influenced by pain perception, and the differences observed at 1 month may primarily reflect reduced postoperative pain rather than superior functional capacity perse26. The VAS directly measures pain intensity, while the HHS assigns 44 of 100 possible points to pain (44%), with the remaining points distributed across function (47%), range of motion (5%), and absence of deformity (4%). Thus, the higher HHS in the SuperPATH group at 1 month is largely driven by lower pain scores26.The mechanisms underlying this early pain reduction may include diminished surgical stress and inflammation due to preserved posterior soft tissues, which mitigates both local tissue sensitization and central pain amplification27. Additionally, preservation of the quadratus femoris and external rotator tendons may reduce pain-generating forces during early mobilization28. Importantly, when interpreting these early differences as ‘early functional recovery,’ we acknowledge that the term ‘functional’ may be imprecise. The HHS function component (walking and stair climbing) improved in the SuperPATH group at 1 month, but this improvement may be confounded by pain relief, as patients with less pain are able to ambulate more readily. Importantly, the two dislocations observed in the PLA group, compared to none in the SuperPATH group, may be related to the higher rate of tendon discontinuity in the external rotators (piriformis and obturator internus) in the PLA cohort, as these muscles contribute to hip stability by preventing excessive internal rotation29. This is supported by imaging studies linking tendon discontinuity to compromised posterior stability30, and aligns with the design rationale of the SuperPATH approach to preserve these stabilizers16.While the sample size is small, this trend warrants consideration, as dislocation remains a significant complication following THA.Nevertheless, it is important to interpret these early differences with caution. First, we did not formally assess whether the observed 1-month differences in VAS or HHS exceeded the minimal clinically important difference (MCID) for THA patients; therefore, their clinical relevance remains uncertain. Second, and more fundamentally, these early functional advantages were not sustained at 12 months, at which point both groups achieved excellent and comparable scores. This convergence indicates that the MRI-detected structural preservation of the quadratus femoris and external rotator tendons, while objectively measurable, did not translate into longer-term superior patient-reported outcomes in this cohort. However, it is important to acknowledge the trade-offs associated with the SuperPATH approach. The SuperPATH approach was associated with significant perioperative disadvantages, including a longer operative time (75.5 vs. 63.8 min, P < .001) and greater intraoperative blood loss (263.0 vs. 239.8 mL, P = .005).These findings are consistent with the learning curve associated with MIS techniques, which often require more precise dissection and instrumentation, leading to longer operative time initially31. The increased intraoperative blood loss may be due to the longer operative time or more meticulous soft-tissue dissection required in the intermuscular plane or the use of specific retractors in the SuperPATH approach10. Nevertheless, total blood loss and LOS were similar between groups, suggesting that the increased intraoperative blood loss in the SuperPATH group was not clinically significant enough to affect transfusion requirements or hospital stay. The elevated serum CK and LDH levels in both groups postoperatively indicate muscle damage, but the lack of inter-group differences suggests that the overall extent of muscle injury, as reflected by these markers, is comparable between the two approaches, despite the differences in muscle-tendon continuity. This may be because CK and LDH are released from muscle fibers due to various factors including retraction, ischemia, and inflammation, not solely from transection32. Notably, the early functional differences between the groups diminished by the 12-month follow-up, with both groups achieving excellent and comparable VAS and HHS scores. This phenomenon aligns with the broader literature on MIS for THA. Systematic reviews and RCTs have documented that while MIS techniques can lead to superior patient-reported outcomes in the first few months postoperatively, these differences typically diminish and become statistically non-significant by 6 to 12 months33,34. This convergence of long-term outcomes is likely attributable to the completion of soft-tissue healing and neuromuscular adaptation after comprehensive rehabilitation, allowing patients from both groups to achieve similar functional plateaus21. Several limitations of this study should be considered. First, the sample size is relatively small (41 patients per group), which may limit the power to detect small but clinically meaningful differences, particularly in rare outcomes like dislocation. Second, We did not formally assess whether the observed 1‑month differences in VAS or HHS exceeded the minimal clinically important difference (MCID). Without such benchmarking, the clinical relevance of these early statistical differences remains uncertain. Future prospective trials with predefined MCID thresholds are needed to determine whether the short‑term advantages of SuperPATH are clinically meaningful. Third, we used MTD and CSA as surrogate measures for muscle atrophy, which, while widely used, do not directly assess muscle strength or functional capacity35. Future studies incorporating quantitative muscle MRI (e.g., T2 mapping to evaluate edema and fat content, diffusion tensor imaging to assess microstructural integrity) or dynamometric testing to obtain objective strength data could provide more detailed insights into muscle quality and function, moving beyond pure morphometry towards a comprehensive structure-function assessment.Fourth, Our study did not quantitatively link tendon continuity to functional scores or dislocation risk beyond descriptive observation. The small number of dislocations (n = 2) precludes any definitive conclusion. Future prospective studies with larger sample sizes and dedicated power for dislocation or functional outcomes are needed to establish such relationships.Fifth, the study was conducted at a single center with surgeons experienced in both techniques, which may limit generalizability to settings with less experienced surgeons, particularly for the technically demanding SuperPATH approach. Sixth, despite no significant differences in baseline demographic characteristics between groups, the retrospective, surgeon-selected design introduces potential selection bias. The indications for choosing SuperPATH versus PLA were not strictly protocolized but were based on surgeon preference and evolving experience over the study period. Several important confounders were not adjusted for, including preoperative muscle mass, preoperative activity level, acetabular bone morphology, severity of femoral head deformity (e.g., collapse extent in ONFH), and surgical difficulty (e.g., obesity, scar tissue). These factors may influence both muscle recovery and functional outcomes. Therefore, we have revised our language to avoid overstating the control of selection bias and now acknowledge this as an inherent limitation of the study design.Finally, while we controlled for baseline characteristics, there may be unmeasured confounding factors influencing muscle recovery, such as variations in rehabilitation protocols or patient compliance. Conclusion In conclusion, SuperPATH effectively preserves the integrity of the quadratus femoris, the obturator internus, and the gluteus minimus muscles, with minimal damage to the quadratus femoris. However, the injuries to the piriformis muscle, the obturator muscle, and the gluteus minimus are comparable to those caused by the posterolateral approach.Importantly, these structural differences did not translate into sustained functional advantages at 12 months, suggesting that the long-term clinical impact of muscle preservation via SuperPATH is limited in this patient population. Abbreviations Total hip arthroplasty Minimally invasive surgical Supercapsular Percutaneously-Assisted Total Hip Posterolateral approach Fatty infiltration Magnetic resonance imaging Osteonecrosis of the femoral head osteoarthritis Body mass index Direct anterior approach Cross-sectional area Maximal transverse diameter Length of hospital stay Serum creatine kinase Lactate dehydrogenase Harris Hip Score Visual Analog Scale Association Research Circulation Osseous Author contributions Fengping gan: Conception and design of the study. Qibiao Zhang: Material preparation, data collection and analysis. Haoji Luo: Collect and analyze magnetic resonance data. Linjie Li: Collect and analyze magnetic resonance data. Xinxin Lin: Statistical Analysis. Hao Qin: First draft of the manuscript. Funding Received funding from the Guangxi Natural Science Foundation (2023GXNSFAA026507). Data availability The datasets generated and/or analyzed during the current study are not publicly available due to patient privacy, but are available from the correspondingauthor on reasonable request. Declarations Competing interests The authors declare no competing interests. Ethics approval and consent to participate Ethical approval for this study was obtained from Medical Ethics Committee of Guigang City People’s Hospital (Approval No. GYLLPJ-20210813-21). Informed consent was obtained from all individual participants included in the study. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Fengping Gan and Qibiao Zhang contributed equally to this work. References Associated Data Data Availability Statement The datasets generated and/or analyzed during the current study are not publicly available due to patient privacy, but are available from the correspondingauthor on reasonable request.