Ten-year results of mesenchymal stromal cell transplantation augmented with vascularised bone grafts for advanced osteonecrosis of the femoral head

Abstract

Background

A prospective, open-label clinical trial, in which transplantation of cultured autologous bone marrow-derived multipotent mesenchymal stromal cells in combination with vascularised bone grafts for the treatment of post-collapse extensive osteonecrosis of the femoral head in ten patients, was conducted previously. The aim of this study was to assess the 10-year clinical and radiographic results of that study.

Methods

Patients were evaluated for radiographic progression of osteonecrosis of the femoral head using anteroposterior radiographs at 10 years postoperatively. Clinical score and hip function, including the timed up and go test, were also estimated.

Results

Osteoarthritic changes in the affected hip were found in five of the ten patients, two of whom had undergone total hip arthroplasty at 7 and 9 years postoperatively. Five of the six cases (83.3%) in which pre-operative femoral head collapse was less than 3 mm, had no further collapse. On the other hand, all four cases in which pre-operative femoral head collapse was ≥3 mm, showed osteoarthritic changes within 10 years. The average clinical score significantly improved postoperatively and was maintained at 10 years.

Conclusions

Considering that eight of 10 post-collapse cases could avoid total hip arthroplasty conversion with good clinical results for 10 years and five of 6 post-collapse cases (collapse <3 mm) could avoid further collapse and osteoarthritic changes for 10 years, mesenchymal stromal cell transplantation in combination with vascularised bone grafts could be an effective treatment for post-collapse osteonecrosis of the femoral head.

Introduction

Osteonecrosis of the femoral head (ONFH) is a devastating disease affecting young and middle-aged patients, and it often progresses to femoral head collapse and osteoarthritis of the hip joint.^1,2 The precise pathological mechanism of ONFH remains unknown, while cessation of blood supply to the femoral head is considered to be a trigger, which causes necrosis of bone-forming cells.³ In several cases, bone tissue without bone-forming cells gradually loses its mechanical properties and eventually collapses. To avoid the progression of femoral head collapse, various hip-preserving surgeries, including core decompression, several types of osteotomies, and non-vascularised or vascularised bone grafting with or without adjunctive cell therapy have been developed, and many promising clinical results have been reported for ONFH at the pre-collapse stage.^2,4, 5, 6, 7, 8, 9 However, as a recent review article indicated, none of the hip-preserving surgeries are advocated for a collapsed femoral head, and the only recommended treatment with good evidence is total hip arthroplasty (THA),^2,10 which may pose a risk of revision surgery and other major complications, especially in young patients.¹¹ Most patients with ONFH are asymptomatic at the pre-collapse stage, and eventually, many of them are diagnosed at the post-collapse stage, as previously reported.^12,13 In this context, the development of an effective hip preservation surgery for ONFH in the post-collapse stage is desired.

A prospective, open-label clinical trial, in which transplantation of cultured autologous bone marrow-derived multipotent mesenchymal stromal cells (MSCs) in combination with vascularised bone grafts for the treatment of ONFH at the post-collapse stage, was previously conducted, and the 2-year clinical results were reported.¹⁴ The aim of this study was to assess the 10-year clinical and radiographic results of this clinical trial.

Materials and methods

The original study was designed as a prospective, open-label clinical trial conducted at a university hospital and registered in the Clinical Trials Registry. This 10-year follow-up study and the original study were approved by the Ethics Committee of our university and were conducted according to the Declaration of Helsinki. All study participants provided informed consent.

Patient eligibility

Inclusion criteria were patients between 20 and 50 years of age and the presence of necrotic stage 3A or 3B post-collapse stage without joint-space narrowing (3A, collapse <3 mm; 3B, collapse of ≥3 mm) according to the radiographic stage system proposed by the Japanese Investigation Committee (JIC),¹⁵ which is a modified version of the system proposed by the Association Research Circulation Osseous (ARCO) Committee.¹⁶ Exclusion criteria were described in a previous study.¹⁴

Treatment

Preparation of autologous bone marrow-derived MSCs and transplantation surgery has been described in detail previously.¹⁴ Briefly, mononuclear cells containing MSCs derived from bone marrow aspiration were cultured and proliferated for approximately 2 weeks in the cell processing centre and thawed until transplantation. The differentiation potential and cytogenetics of MSCs were confirmed before transplantation. In the transplantation surgery, a cortical window (1.5 × 4 cm) was made at the anterior aspect of the femoral neck through a modified Smith-Peterson approach. Necrotic areas, including subchondral lesions, were extensively curetted through a bone trough from the cortical window, under fluoroscopy and endoscopy. Next, the tricortical iliac bone (1.5 × 2 × 5 cm) with a vascular pedicle (deep circumflex iliac vessels) and iliac cancellous bone chips were harvested. Then MSCs (0.5–1.5 × 10⁸ cells) pre-mixed with β-tricalcium phosphate granules (Osferion; Olympus Terumo Biomaterials Co., Tokyo, Japan) and the iliac cancellous bone chips were compacted into the cavity. Lastly, the vascularised tricortical iliac bone was grafted into the bone trough. Patients were kept non-weight-bearing for 6 weeks and allowed partial weight-bearing for another 6 weeks, before full weight-bearing started at 12 weeks after surgery.

Methods of evaluation

The primary outcome was the radiographic progression of ONFH on anteroposterior radiographs at 10 years after treatment. Progression was evaluated according to the radiographic stage established by the JIC, with modifications to the ARCO staging system.

Necrotic lesions were evaluated using the radiographic classification proposed by the JIC, in which type C1 and type C2 lesions are defined as lesions occupying more than the medial two-thirds of the weight-bearing portion. Type C2 indicates that the lesion extends laterally to the acetabular edge, while type C1 lesion does not extend beyond the acetabular edge.

Objective and patient-reported clinical outcomes were evaluated using the Japanese Orthopaedic Association (JOA) score.¹⁷ Functional assessment was performed pre-operatively and at 1 and 10 years after surgery. Passive hip flexion, extension, abduction, and external rotation angles were measured using universal goniometry. The thigh circumference was measured 10 cm from the proximal border of the patella. In the timed up and go (TUG) test, the time (in seconds) required for a participant to stand from an armless chair (chair seat height, 45 cm), walk 3 m, turn, walk back to the chair, and sit down was measured.

Statistical analysis

Continuous variables are described as mean ± standard deviation. All statistical analyses were carried out using JMP IN (Version 15, SAS Institute Inc, Cary, NC). For follow-up assessment of changes in JOA score and hip function data over time, one-way repeated measures ANOVA with post-hoc Tukey-Kramer test was performed. P values of less than .05 were considered significant.

Results

Ten participants were enrolled in this study between November 2007 and June 2009. All of them were men with an average age of 31.7 years (range, 20–48 years), and bilateral hips were affected. Four of the ten patients had a history of steroid use. Patient demographics are shown in Table 1. No serious events were observed during the establishment and expansion of MSCs, such as bacterial contamination and the transformation of MSCs. Two patients (Cases 3 and 7) underwent ipsilateral THA at 7 and 9 years after the transplantation surgery. At the final follow-up, no patient was waiting for THA conversion.

Table 1.

Patient demographics and radiological results.

					JIC staging systemb
					Tx side			Non-Tx side
Case	Age	PI	SU	Typea	Pre-Tx	2Y	10Y	Pre-Tx	10y
1	27	Nephritis	Y	C2	3B	3B	4	2	2
2	23	Cushing syndrome	Y	C2	3A	3A	3A	2	2
3	48	Meningioma	N	C2	3B	4	THA	2	2
4	20	Hepatitis	Y	C1	3A	3A	3A	2	2
5	35	None	N	C2	3A	3A	3A	THA	THA
6	28	None	N	C2	3A	3A	4	2	2
7	39	Leukaemia	Y	C2	3B	4	THA	2	THA
8	26	None	N	C2	3B	3B	4	4	4
9	33	None	N	C2	3A	3A	3A	1	1
10	38	None	N	C2	3A	3A	3A	3A	THA

JIC, Japanese Investigation Committee; PI, past illness; SU, steroid use; Tx, treatment; Y, yes; N, no.

^aRadiographic classification proposed by JIC.

^bModified version of the radiographic staging system proposed by the Association Research Circulation Osseous (ARCO) Committee.

Radiological progression

Pre-operative necrotic lesions, which were examined radiographically, occupied more than the medial two-thirds of the weight-bearing portion, and extended laterally beyond the acetabular edge in nine out of ten patients (Type C2 in Table 1). At the 10-year follow-up, osteoarthritic changes of the affected hip were found in five of the ten patients, two of whom had undergone THA. Five of the six cases (83.3%) in which pre-operative femoral head collapse was <3 mm (stage 3A) showed no further collapse at the 10-year follow-up (Fig. 1, Fig. 2). On the other hand, all four cases in which the pre-operative femoral head collapse was ≥3 mm (stage 3B) showed osteoarthritic changes (stage 4) within the 10 years of follow-up (Figs. 1 and 3). Regarding contralateral ONFH, two patients experienced further collapse of the femoral head and subsequently underwent THA.

Fig. 1.

Transition of radiographic stages. THA, total hip arthroplasty.

Fig. 2.

Representative case: 38-year-old male (Case 10). (a) Preoperative anteroposterior radiograph. (b) Preoperative coronal computed tomography image. (c) Preoperative coronal image by T1-weighted magnetic resonance imaging. (d) Anteroposterior radiograph 10 years after surgery. No apparent osteoarthritic change is observed.

Fig. 3.

Representative case: 27-year-old male (Case 1). (a) Preoperative anteroposterior radiograph. (b) Preoperative coronal computed tomography image. (c) Preoperative coronal image by T1-weighted magnetic resonance imaging. (d) Anteroposterior radiograph at 10 years after surgery. Osteoarthritic changes are evident.

Clinical score and hip function

The JOA score improved from 65.6 ± 25.5 pre-operatively to 90.2 ± 14.2 at 1-year follow-up (N = 10), and 91.9 ± 5.8 at 10-year follow-up (N = 8; those without ipsilateral THA). There were significant differences between 10-year and pre-Tx (p = 0.012) and between 1-year and pre-Tx (p = 0.013).

In the affected hip range of motion (ROM), flexion, abduction, and external rotation improved after treatment and were maintained until 10-year follow-up, while the differences among each time point were not significant. In contrast, the time required for TUG significantly decreased at 1 and 10 years postoperatively (Table 2). The mean values, excluding two ipsilateral THA cases at 10-year follow-up, were equivalent to those including them (Table 2).

Table 2.

Comparison of the affected hip ROM, thigh circumference, TUG between pre-Tx, one year after Tx, and 10 year after Tx (N = 10).

	Pre-Tx	1 Year	10 Year
Flexion (°)	96.5 ± 12.9	106.0 ± 11.0	105.5 ± 9.6 (108.8 ± 7.4)
Extension (°)	17.0 ± 4.2	15.0 ± 4.7	16.0 ± 4.6 (15.6 ± 5.0)
Abduction (°)	29.0 ± 10.5	33.0 ± 6.7	33.5 ± 10.1 (33.1 ± 11.0)
External rotation (°)	42.0 ± 16.0	47.0 ± 12.3	45.0 ± 6.7 (45.6 ± 7.3)
Thigh circumference (cm)	43.8 ± 4.0	43.1 ± 4.5	43.4 ± 4.7 (43.1 ± 5.0)
TUGa (s)	6.8 ± 1.2	5.9 ± 1.2	5.6 ± 1.0 (5.4 ± 1.1)

ROM, range of motion; TUG, timed up and go; Tx, treatment.

Values are presented as mean ± standard deviation. The values in parentheses are those without ipsilateral THA (n = 8).

^aSignificant difference between 10-year and pre-Tx (p = 0.0006) and between 1-year and pre-Tx (p = 0.003).

Discussion

This prospective clinical trial was performed to evaluate the safety and efficacy of MSC transplantation. No serious adverse events associated with the surgery were observed perioperatively or during the follow-up period, except for disease progression of ONFH. The ten-year clinical outcomes including and excluding THA conversion cases were equally good. Considering the ten-year radiographic results in which five of the ten hips progressed to stage 4 osteoarthritis and the surgical invasiveness of this procedure, we could not insist on the efficacy of MSC transplantation augmented with vascularised bone grafts for post-collapse ONFH. Nevertheless, it should be noted that eight of 10 post-collapse cases could avoid THA conversion with good clinical results for 10 years. Five of the six hips with extensive ONFH (stage 3A; collapse less than 3 mm) experienced no further collapse and did not progress to stage 4 within 10 years. They showed excellent functional outcomes 10 years postoperatively. Subchondral collapse of the necrotic lesion could be repaired with revascularisation by MSCs transplantation and vascularised iliac bone graft, and a collapse of <3 mm presumably did not seriously affect the surface morphology of the femoral head. On the other hand, all four hips with extensive ONFH (stage 3B; collapse of ≥3 mm) progressed to the osteoarthritic stage (stage 4) within 10 years. Bone grafts with MSC transplantation could not modify the uneven articular surface of the femoral head, which might inevitably lead to osteoarthritic changes.

Only one patient (Case 6), in which stage 3A ONFH progressed to stage 4 within 10 years, continued to perform heavy labour activities postoperatively, which might be a risk factor for the osteoarthritic change.

Cell therapies using bone marrow cells were reported mainly in combination with core decompression for pre-collapse ONFH, and those for post-collapse ONFH have rarely been reported.⁹ Mao et al. reported in a recent review article that stem cell therapy was safe and effective for pre-collapse ONFH, and there was a beneficial effect for patients under 40 years of age.⁹ It was also stressed that the quantity of stem cells at a magnitude of 10⁸ had better effects on disease progression events.⁹ In this series, the two THA conversions occurred in Case 3 (age, 48 years) and Case 7 (age, 39 years). The stored and transplanted number of MSCs was an average of 1.1 × 10⁸, as reported in a previous study [14]. The present results are consistent with those of other studies, although all the cases had post-collapse ONFH.^7,9,18

Long-term results of cell therapy for ONFH have rarely been reported.^6,8,19 Hernigou et al. reported on a thirty-year follow-up prospective randomised study comparing cell therapy with simultaneous contralateral decompression for ONFH and concluded that core decompression with bone marrow injection improved outcome; however, it was conducted only for pre-collapse ONFH.⁶ Tomaru et al. reported ten-year results of concentrated autologous bone marrow aspirate transplantation for ONFH, and half of the post-collapse cases (stage 3 by the JIC classification) experienced THA conversion within 10 years.⁸ However, there is a paucity of detailed data on the progression of ONFH. In this context, our study provided detailed analyses of the long-term outcome of cell therapy for post-collapse ONFH, despite the limited number of cases.

Long-term follow-up studies on vascularised bone grafts for ONFH are also limited.²⁰ The type of vascularised graft varied among the studies, including the fibula, iliac bone, and greater trochanter. In particular, long-term results have rarely been reported on iliac bone grafts vascularised with deep circumflex iliac arteries.²¹ The radiographic and clinical results were poor for post-collapse and extensive ONFH.²¹ Cell therapy with vascularised iliac bone graft for ONFH has not been reported, except in one previous study,¹⁴ and this is the first report of long-term results. Bone marrow mononuclear cell implantation with non-vascularised iliac bone graft was only reported by Wang et al., but the follow-up period was only for 2 years.²²

This study has several limitations. It was conducted as a very small-scale, single-group feasibility study, and large-scale studies are mandatory to demonstrate the efficacy of this treatment method. Furthermore case-control studies should be considered to demonstrate the efficacy of MSC transplantation without vascularised bone grafts. Annual follow-up beyond 2 years postoperatively was partially omitted before the 10-year follow-up. Therefore, clinical and radiographic time courses cannot be described in detail. Nevertheless, this study revealed 10-year follow-up results for each patient and provided an opportunity to observe the long-term results of an unprecedented treatment for post-collapse ONFH.

Recently, excellent clinical results of THA for ONFH have been reported, and promising results of simple regenerative therapy using fibroblast growth factor have also been reported.^23,24 These surgical treatments are less invasive and require less hospitalisation than that using the present surgical method. Considering five of six post-collapse ONFH cases (collapse <3 mm) could prevent further collapse and osteoarthritic changes for 10 years, MSC transplantation in combination with vascularised bone grafts could be an effective treatment for selected cases of post-collapse ONFH, although further modifications should be considered to reduce the surgical invasiveness of this treatment.

Funding

This research received no external funding.

Author contributions

Koji Goto: Conceptualization, data curation, writing, Tomoki Aoyama: Conceptualization, data curation, Junya Toguchida: Supervision, Yutaka Kuroda: review & editing, Toshiyuki Kawai: review & editing, Yaichiro Okuzu: review & editing, Shuichi Matsuda: Validation. Conceptualization, data curation, K.G., T.A.; writing, K.G.; Supervision, J.T.; review & editing, Y.K., T.K., Y.O.; Validation, S.M.

Declaration of competing interest

All the authors have no conflict of interest.