We performed a retrospective study of all of the patients with a completely dislocated hip joint without pseudo- articulation between the femoral head and iliac bone who underwent THA between October 1999 and September 2001. In total, we performed THA operations on six hips in six patients without subtrochanteric femoral shortening osteotomy (Fig. 1). In same the period, three hips in three patients received THA combined with subtrochanteric femoral shortening osteotomy according in accordance with the procedure described by Sonohata [12] (Fig. 2). These three hips were used as controls.

Fig. (1). Total hip arthroplasty without femoral shortening osteotomy in patients with a completely dislocated hip. a, before the operation. b, after the operation.

Nine hips in nine patients (7 females, 2 males) were included in this study. The patients had no history of nerve trauma, neurological disease, or pelvis or femoral osteotomy. The average age of the patients at the time of the operation was 58 years (range, 49-66 years). The average duration of follow-up monitoring was 135 months (range, 18-1213 months) (Table 1).

The indications for surgery were severe hip pain and/or considerable difficulty walking and performing daily activities. All of the operations were performed by a single surgeon. The decision to perform femoral shortening osteotomy was made by the operator. Six patients received only spinal anesthesia, one patient received only epidural anesthesia, and two patients received spinal and epidural anesthesia, based on the decision of the anesthesia team.

All of the operations were performed using the posterolateral approach with a cementless femoral component (PerFix-HA femoral component; Kyocera, Kyoto, Japan) with a 22- or 28-mm alumina or zirconia ball and an AMS-HA acetabular shell with an ABS (alumina ceramic inlay mechanically fixed to a polyethylene liner) or an AMS (polyethylene liner) liner. Whether or not the operator checked the tension of the sciatic nerve under direct vision is unclear.

A routine radiographic examination including anteroposterior and frog-leg lateral radiographs was performed. On the anteroposterior radiographs, a line was drawn through the teardrops, and the perpendicular vertical distance of the apex of the lessor trochanter from this reference line was compared to identify leg length discrepancy and leg lengthening [13]. The acetabular components were evaluated at the most recent follow-up for evidence of migration, in accordance with the method of Carlsson and Gentz [14]. The bone-metal interface was evaluated at the most recent follow-up to identify the presence and progression of radiolucent lines in the three zones described by DeLee and Charnley [15]. The femoral component was evaluated for changes in the position, subsidence, and radiolucency in the seven zones described by Gruen et al. [16]. The stability of the femoral component was assessed as bone-ingrown fixation, stable fixation, or unstable fixation in accordance with the fixation/stability score described by Engh et al. [17]. The orientation of the component was classified as valgus, slightly valgus, neutral, slightly varus, or varus, in accordance with the classifications of Christie et al. [18]. Slightly valgus or slightly varus alignment was used to describe a femoral stem with less than 5° of malalignment with respect to the neutral axis of the femoral canal. Varus or valgus alignment was used to describe a femoral stem that was oriented ≥5° beyond the neutral position.

The height of the hip center was measured as the perpendicular distance from the inter-teardrop line. A high hip center was defined as a hip with a center located >35 mm proximal to the inter-teardrop line [19].

All of the hips were evaluated using the Japanese Orthopaedic Association (JOA) Hip Score. The JOA score consisted of four categories, with a maximum total score of 100 points: pain (40 points), range of motion (20 points), walking ability (20 points), and activities of daily living (20 points).

The mean JOA scores at the preoperative and final follow-up examinations and the preoperative and postoperative leg length discrepancy were compared using the paired t-test. The JOA score was statistically evaluated in patients with over five years of follow-up. All of the statistical analyses were performed using the software program SPSS for Windows (Version 22; IBM Corp, Armonk, NY, USA). P values <0.05 were considered to indicate statistical significance.

The average duration of follow-up was 135 months (range, 18-1213 months). The average JOA hip score for the patients improved from an average of 54 (range, 22-76) preoperatively to 77 (range, 60-99) postoperatively (p<0.05) (Table 2).

The mean outer diameter of the acetabular components was 45 mm (range, 42-46 mm). Five AMS-HA acetabular shells with an ABS (alumina ceramic inlay mechanically fixed to a polyethylene liner) liner were used with a 28-mm alumina ball. Four AMS-HA acetabular shells with an AMS (polyethylene liner) liner were used with a 22-mm zirconia ball. The average number of screws was 3.5 (3 in 6 hips, 4 in 1 hip, 5 in 2 hip) for each of the acetabular cups. The average distal diameter of the stems was 9.8 mm (9 mm in 3 hips, 10 mm in 5 hips, 11 mm in 1 hip). A bone graft to the acetabulum was performed in four hips (block in three hips, morselized bone in one hip). A bone graft to the femur was performed in three hips at the site of subtrochanteric osteotomy. Liner and ball revision surgery for the ABS acetabular component failure was performed for all ABS liners [20]. Stem revision surgery for sciatic nerve palsy was performed seven days after primary THA in one hip. Cup revision surgery for cup loosening was performed 30 months after primary THA in 1 hip. Two hips had an intraoperative greater trochanteric fracture that necessitated additional wiring.

The average horizontal distance of the center of the femoral head was 31 mm (range, 25-42 mm), and the average height of the hip center was 25 mm (range, 11-38 mm). One hip (Case 1) was considered to have a high hip center.

The average leg length discrepancy of the patients improved from an average of 5.6 mm (range, 3.0-7.5) preoperatively to 1.0 mm (range, 0-3.0) postoperatively (p<0.01). The average increase in the leg length was 4.8 mm (range, 3.0-6.5) (Table 3).

All of the stems were in the neutral position. Two acetabular components migrated (Cases 1 and 7); while the Case 1 did not require revision surgery, Case 7 require did. The other components showed no evidence of migration or loosening. No femoral or acetabular components displayed radiolucency. There were no cases of subsidence in the femoral component. Fixation was achieved in all of the femoral components with an optimal interface at the latest follow-up. One case (Case 1) required femoral revision surgery to achieve relaxation of the sciatic nerve to treat sciatic nerve palsy.

There were no cases of infection. During the operation, two greater trochanteric fractures occurred, and additional wiring was performed. Dislocation occurred in five hips (Cases 1, 4, 7, 8, and 9) within two weeks after THA. Case 4 required open reduction.

Fig. (2). Total hip arthroplasty with femoral shortening osteotomy in patients with a completely dislocated hip. a, before the operation. b, after the operation.

There were 4 cases of neurologic abnormalities. In Cases 2 and 3, a physical examination of the motor and sensory functions along the femoral and sciatic nerves revealed no changes between the preoperative values and the most recent follow-up. In Case 1, the patient experienced numbness in her right lower legs and foot one day after THA. The muscle strength in her right ankle and toes was rated as grade 0 (maximum 5). One week after THA, stem revision was performed to reduce the tension of the soft tissue around her right hip joint. At the last follow up, she had light paresthesia in her right lower leg; the sensory disturbance in the dorsal foot and plantar foot was classified as grade 8 (maximum 10) and grade 2, respectively. Her muscle strength improved to grade 1 in dorsal flexion and grade 2 in plantar flexion. In Case 4, the patient experienced hypesthesia in her left plantar foot and severe pain on the lateral side of her left thigh, which was induced by knee extension one day after THA. Fortunately, she had no motor dysfunction. At the last follow up, she complained of mild sensory disturbance and pain. Case 5 experienced numbness in the right proper sensory area of the deep peroneal nerve and sensory disturbance in the dorsal foot and plantar foot of grade 5-6 (of 10) and grade 2-3, respectively. The muscle strengths of her right extensor hallucis longus, tibialis anterior, and gastrocnemius muscle were grade 0 (of 5), grade 1, and grade 2, respectively. At the last follow up, the sensory disturbance in the right proper sensory area of her deep peroneal nerve improved to grade 3 (of 10); however, there was no remarkable recovery in her dorsal or plantar foot. The muscle strength of her right extensor halluces longus, tibialis anterior and gastrocnemius muscle improved to grade 2 (of 5), 3, and 4, respectively. In Case 6, the patient experienced paresthesia in the right sensory area of the femoral and sciatic nerves; however, the muscle strength in her right lower limbs was not disturbed. Her neurologic abnormality completely recovered within two months after surgery (Table 4).