Is Prior Nonoperative or Operative Treatment of Dysplasia of the Hip Associated With Poorer Results of Periacetabular Osteotomy?

Abstract

Background: Understanding the implications of either nonoperative or operative treatment of developmental dysplasia of the hip (DDH) performed before periacetabular osteotomy (PAO) is critical to counseling patients and their families. There are limited studies, however, on PAO for the treatment of residual DDH after surgical intervention during childhood, and even less information about PAO after prior nonoperative treatment.

Questions/purposes: We analyzed patients who had undergone PAO for DDH and asked: Did patients with prior childhood treatment (either operative or nonoperative) (1) improve less in modified Harris hip score (mHHS), 12-item International Hip Outcome Tool (iHOT-12) score, or WOMAC score; (2) demonstrate more severe preoperative deformities; and (3) receive less complete radiographic correction and have more frequent complications than did patients whose hips had not undergone prior treatment? We also asked: (4) Were there subgroup differences among patients with DDH treated nonoperatively versus operatively before PAO in these same functional and radiographic parameters?

Methods: Between January 2011 and December 2020, a total of 90 PAOs were performed in 82 patients who had prior surgical or nonsurgical treatment. Of those, 3 patients (3 hips) with neuromuscular diseases were excluded, 4 patients (5 hips) were excluded for having received treatment after childhood, 7 hips that had undergone bilateral PAOs were excluded, and another 4 patients (4 hips) were lost to follow-up before the minimum study period of 2 years, leaving 71 patients (71 hips) for analysis (the previous treatment group). Among these, 32 patients had a history of previous surgery (the previous surgery group), and 39 patients had prior nonsurgical treatment (such as a Pavlik harness, closed reduction, spica casting) (the previous nonoperative group). During the same period, 1109 PAOs were performed in 956 patients who had no history of previous hip treatment. Following a 1:2 ratio, 142 patients (142 hips) were selected as the control group by matching for age (within 2 years difference), year of surgery (same year), and follow-up time (within 1-year difference). The patient characteristics for both the previous treatment group and the control group exhibited comparability, with mean ± SD follow-up durations of 49 ± 23 months and 48 ± 19 months, respectively. Within the previous 5 years, 3 patients (8%) in the previous nonoperative group, 4 patients (13%) in the previous surgery group, and 15 patients (11%) in the control group had not attended follow-up visits. We compared hip function and radiographic results between the two groups and performed a subgroup analysis between the previous surgery group and the previous nonoperative group. Hip function was assessed using the mHHS questionnaire, the WOMAC, and the iHOT-12 with attention to the minimum clinically important differences of these tools. The threshold values for clinically important improvement were 9.6 points, 13 points, and 16.1 points for the mHHS, iHOT-12, and WOMAC, respectively. Radiographic measurements included the lateral center-edge angle (LCEA), anterior center-edge angle (ACEA), Tönnis angle, acetabulum-head index, and acetabular wall index. We also evaluated Tönnis osteoarthritis grade and femoral head deformity. Occurrences of adverse radiographic events such as posterior column fracture, nonunion, stress fractures, insufficient coverage or overcoverage, acetabular protrusion, and progression of osteoarthritis were recorded.

Results: We found no clinically important differences in magnitude of improvement between the previous treatment group and the control group in terms of mHHS (mean ± SD 10 ± 12 versus 12 ± 12; p = 0.36), iHOT-12 (25 ± 18 versus 26 ± 19; p = 0.51), or WOMAC score (12 ± 12 versus 15 ± 19; p = 0.17). Preoperative deformity in the previous treatment group was more severe than in the control group (mean ± SD LCEA -1° ± 9° versus 5° ± 8°; ACEA -8° ± 18° versus 1° ± 14°; Tönnis angle 31° ± 7° versus 27° ± 7°; acetabulum-head index 56% ± 13% versus 61% ± 8%; all p < 0.001). In the previous treatment group, a higher percentage of patients exhibited flattening or irregularity of the femoral head compared with the control group (52% versus 9%; p < 0.001), and there was also a higher proportion of patients with Tönnis grade 1 or above (51% versus 42%; p < 0.001). Although there were still differences in LCEA, ACEA, and Tönnis angle between the two groups at the last follow-up, the differences were small, and the mean values were within the normal range. The previous treatment group had a higher risk of intraoperative posterior column fracture (14% and 5%; p = 0.02), insufficient acetabular coverage (20% and 8%; p = 0.01), and progression of osteoarthritis (17% and 8%; p = 0.04) compared with the control group. Subgroup analysis revealed no clinically important differences in magnitude of improvement between the previous surgery group and the previous nonoperative group in terms of mHHS (10 ± 14 versus 10 ± 11; p = 0.91), iHOT-12 (22 ± 21 versus 27 ± 14; p = 0.26), or WOMAC score (12 ± 14 versus 12 ± 11; p = 0.94). Apart from a higher proportion of patients who presented with arthritis (72% versus 34%; p = 0.01) and a smaller anterior wall index (11% ± 11% versus 20% ± 12%; p = 0.01) in the previous surgery group, all other preoperative radiographic parameters were consistent between the two groups. Additionally, the previous surgery group had a higher frequency of arthritis progression (28% versus 8%; p = 0.02), while the frequencies of other complications were similar between the two groups. Specifically, the frequencies of pubic ramus nonunion (22% versus 21%; p = 0.89), intraoperative posterior column fracture (19% versus 10%; p = 0.50), and insufficient acetabular coverage (25% versus 15%; p = 0.31) were high in both groups.

Conclusion: We found no clinically important difference in the magnitude of improvement between patients who had childhood treatment and those who did not, but patients who had prior childhood treatment were more likely to experience serious complications, and radiographic correction in those patients was less complete. As in the case of patients who have had prior operative treatments, it is crucial not to overlook the unexpectedly severe deformity of residual DDH after previous nonoperative treatment and complications following PAO. Surgeons and patients alike should be aware of the potential for worse radiographic outcomes or an increased risk of complications when prior operative or nonoperative treatment has preceded PAO. Future studies might investigate optimal management strategies for this specific group of patients to improve outcomes and reduce complications.

Level of evidence: Level III, therapeutic study.