Comparison of Management of AVN of Femoral Head by Decompression with TFL Muscle Pedicle Bone Graft and Fibular Graft

Comparison of Management of Femoral by and Abstract: Background: Avascular necrosis (AVN) of the femoral head usually occurs in the active age of life, and is an increasingly common cause of musculoskeletal disability. Objective: we envisage to evaluate and compare the clinic oradiological outcomes between TFL muscle pedicle bone graft and no vascularized fibular grafting after core decompression prior to hip joint involvement of AVN femoral head provides painless and mobile life. Method: January 2013 to December 2019, total 64 hips (44 patients) in the age group of 18–48 years (mean 30.36±4.64 years) were included in this prospective study. Ficat and Arlet staging system was used, Stage I (n=10 hips), Stage II (n=44 hips) and stage III (n=10) of AVN of femoral head, were included after clinical and radiological evaluation, where 32 hips (half of all stage-I, II and III) were treated with fibular graft (Group 1) and rest 32 hips were treated with TFL muscle pedicle bone graft (Group 2) after decompression of femoral head. Preoperative Harris Hip Score (HHS), visual analog score (VAS), plain radiographs, and magnetic resonance imaging (MRI) were compared with serial postoperative HHS, VAS, plain radiographs at regular interval. The average follow up was 56 months. Result: Out of 44 patients, male was 28 (63.64%) and female was 16 (36.36%). Failure of surgery was defined as progression of the disease, which was n=1,10% (1/5, 20% in group 1, 0% in group-2) in stage I, n=12/44, 27.27% (7/22, 31.82% in group-1 and 5/22, 22.73% in group-2) in stage II and n=5/10, 50% (2/5, 40% in group 1 and 3/5, 60% in group 2) in stage III disease. Median values of HHS at the end of the follow up in Group I was 80 and 76 in Group 2, compared to the preoperative HHS of 56 and 52 respectively. Overall satisfactory result was 71.88%, p value was <0.01, that is significant. In group 1 satisfactory result was 22/32 (68.75%) but 24/32 (75%) in group 2, no statistical significant difference (>0.05) between two groups. Even in early stage III disease, only 50% was effective. Conclusion: Core decompression with bone graft is effective in preserving the sphericity of the femoral head and to delay the progression in the early stages of the AVN of femoral head, (Stage I and II), fibular graft gives early stability but long term results are similar in both group.


Introduction
AVN of femoral head is a painful and disabling condition. It mostly affects the people in third to fifth decade of life [1]. AVN is mostly caused by cortiosteriod abuse, but it may also be associated with post traumatic, idiopathic, ITP, ALL, renal, hepatic and skin disorder [2]. Early clinical and radiological TFL Muscle Pedicle Bone Graft and Fibular Graft diagnosis is usually difficult but if done in time it may prevent further collapse of femoral head and is important for favorable outcome of disease [3]. The treatment of AVN is done according to staging of the disease. The ficat's grading system that is based on the plain x-ray, is the most commonly used classification in clinical practice [4]. In the inititial stages of disease the head preserving surgeries which aim at reducing intra-osseous pressure and enhancing the vascular supply of the femoral head are preferred. Among these the core decompression of the femoral head is preferred treament in ficat stage I and stage IIA diseases [5]. Core decompression (CD) reduce the intra-osseous pressure and thus decrease the pain, but it does not enhance the vascularization of necrosed head adequately, especially in an advanced stage of the disease. Therefore it is better supplemented with various bone grafting or muscle pedile bone grafting technique to enhance vascularity and healing of the lesion. There are several techniques described to enhance the vascularization of femoral head like CD with vascularised and non-vascularised bone (fibular) grafting, various types of muscle pedicle bone graft etc. Fibular (cortical) graft ensured more structural support but vascularized bone graft enhanced revascularization of necrosed head. Both the procedures have shown favorable results in stage IIA, stage IIB and stage III diseases [6]. Muscle pedicle bone graft has been advanced as a modality of treatment in Ficat et al. stage I & stage II avascular necrosis of the head of femur [7][8][9][10]. Untreated osteonecrosis of femoral head ultimately leads to collapse and degeneration of the hip joint. For adequate and proper management of AVN of femoral head, uses a combination of plain radiographs, MRI, Bone scan and clinical features to stage avascular necrosis of the femoral head. In stage 0-plain radiograph: normal, MRI: normal and clinical symptoms: nil. In stage I-plain radiograph: normal or minor osteopenia, MRI: edema, bone scan: increased uptake and clinical symptoms: pain typically in the groin. In stage II-plain radiograph: mixed osteopenia and/or sclerosis and/or subchondral cysts, without any subchondral lucency (crescent sign), MRI: geographic defect, bone scan: increased uptake, clinical symptoms: pain and stiffness. IIA-consider small central lesions, IIB-Multiple small lesion, subchondral cresent sign. In stage III-plain radiograph: crescent sign and eventual cortical collapse, MRI: same as plain radiograph and clinical symptoms: pain and stiffness +/-radiation to knee and limp. In stage IV-plain radiograph: end-stage with evidence of secondary degenerative change, MRI: same as plain radiograph, and clinical symptoms: pain and limp [9][10][11][12].
Preservation of femoral head delays the replacement procedures which have finite lifespan. Hence, femoral headpreserving procedures should be considered for the management of properly selected patients in order to delay the arthroplasty. Nevertheless, preservation of head allows for floor level activities which are required for our populations. Therefore, we envisage to analyze the clinical outcome and radiological outcomes after core decompression (CD) and compare between TFL muscle pedicle bone graft with nonvascularized fibular grafting.

Methods
This prospective longitudinal observational study was carried out in the department of orthopaedics, BSMMU, from January 2012 to December 2019. Ethical clearance was obtained from the Ethical Committee of the institution. Consequative sampling technique was used. Adults of age 18-48 years with Stage I, II and early III of Ficat and Arlet classification were included for the study. Early stage III indicates, ROM of hip is good but mild (early) cortical collapse. Hip diseases (AVN) with local site infection, patients with traumatic etiology, psychiatric patients, patients with comorbidities and Ficat-Arlet advanced stage III-IV were excluded from the study. If one hip was in upto early stage III that was included but if another hip involment was in advanced III or IV excluded that hip but not the patients. The diagnosis of AVN was made based on the clinical and radiological investigations, that is, X-ray and MR. We analysed 48 patients in 64 hips of AVN of femoral head, equally distributed in two groups, Group I considered as non vascular fibular graft group and Group II considered as TFL muscle pedicle bone graft group. Core decompression with removal of maximum necrosed bones from the head in all cases under fluroscope. In group 1, non vascular fibular grafting and in group 2, TFL muscle pedicle bone grafing were done. A detailed history regarding pain and visual analog score (VAS), drug history of steroid intake, smoking and alcohol intake, and history suggestive of coagulation disorders were taken. General examination and systemic examination in detail were done. Local examination of both the hip joints for gait abnormality and restriction of movements were assessed. Blood investigations such as complete blood count and peripheral smear, liver function tests, kidney function tests, random blood sugar, and coagulation profile were done. X-rays of the pelvis with bilateral hip of anteroposterior view and frog-leg lateral view was obtained. We used Ficat and Arlet classification system for staging. Pre operative MRI was also done in all cases up to stage II disease. Baseline HHS and VAS were recorded. Total 64 hips were equally divided in two groups. Same number 32 and same stages of hips (stage I-5, II-22 & III-5) were included in both group 1 (Fibula) and group 2 (TFL muscle pedicle). CD was done in all cases, after that fibular cortical graft was inserted in group 1 and TFL muscle pedicle grafting in group 2.

Group 1 (DC with Fibular Grafting)
The surgeries were done under spinal anesthesia, in supine position on a fracture table and under image intensifier. The lateral femoral cortex was exposed through a short linear incision. Fascia lata and vastuslateralis muscle was divided in line with their fibers and identify the flare of the greater trochanter. A small drill hole was placed in the mid-lateral cortex and entry point was just proximal to the level of lesser trochanter to avoid stress fractures. Small guidewire was inserted through this hole into the center of the lesion in the femoral head. With the help of the triple reamer (8 mm) of dynamic hip screw, which was slowly advanced into the lesion and necrosed bone was removed and sample was sent for biopsy. Autograft of fibula about 8-10 cm was obtained, proximal end made tapered and distal end was made rounded. The appropriate length of the graft was measured from the reverse measuring device. Graft was passed through the lateral cortex to place up into the head with the proximal end just underneath the subchondral bone within 5 mm from the outer cortex, in the area of the lesion. In addition cancellous bone graft was put if there was larse lesion. Closure was done in layers.

Group 2 (DC with TFL Muscle Pedicle Bone Graft)
The patient was positioned supine and small sand bag behind the buttock at the affected side and incision was made as described by smith-peterson anterior approach [13]. The groove between the sartorius and tensor fascia lata muscle and the deep fascia overlying the TFL was identified. An incision was then made along the direction of muscle fibre of TFL, 2.5 cm behind the anterior edge of the muscle. The isolated anterior fibre of TFL was elevated and its deeper fibres found intermingled with those of the underlying gluteus-minimus were sectioned. A titanium made 4mm diameter X 45mm cancellous lag was inserted, then centering this screw a cone shape segment of iliac crest 2.0 cm X 1.5cm and 2.0 depth was osteotomised, 1.5 cm bebind the ASIS and retracted down keeping its attachment to the anterior fiber of TFL. The muscle pedicle bone graft was prepared, gets its blood supply from the superior gluteal artery and ascending branch of lateral circumflex femoral artery. Bleeding from the raw surface of the muscle pedicle bone graft was observed. Next the anterior capsule of the hip was opened, using an inverted T incision. Multiple drill holes were made over the anterior surface at head neck junction, 2 cm medial to the lateral cortex, and a small window 1.5X1.5 cm was made to accommodate TFL muscle bone graft without tension and through curettage was done and removed all the necrosed bone of the head especialy weight bearing zone of the head. Free cancellous bone graft was harvested from iliac crest and packed the deepar part of the cavity of diseased head. Then the bony portion of muscle pedicle bone graft was shaved according to contour of the slot, one drill hole was made centering the window and perforate the posterior cortex, 1cm lateral to the medial cortex. The muscle pedicle bone graft is impacted inside the cortical window and fixed with one 4mm screw and finally anchored the overlying fascial sleeve to the surrounding capsule and soft tissue. The cut margin of the capsule and gluteus minimums muscle were then repaired to secure the graft. The wound was closed in layers over the suction drain.

Post Operative Follow up
Check X-ray was obtained on postoperative day 1 and VAS was recorded. In postoperative period, absolute nonweight bearing regimen was followed with traction of 2 kg throughout the day for first 1 week and during night times for next 2 weeks. Active assisted range of motion exercises were started after 1 week and passive range of exercise was started on the 2 nd weeks on ward. Patient was discharged after 5 days. Partial weight bearing was allowed after 6 weeks and full weight bearing was permitted after 12 weeks. Clinico-radiological follow-up was done at 1 week, 2 weeks, 4 weeks, 12 weeks, 6th month, and subsequently every 6 months. One week after discharge, patient was evaluated for surgical site infection and VAS score was recorded. Serial Xrays were obtained at the interval of 2 weeks, 4 weeks, 12 weeks and 6 months and at every 6 monthly, which was compared with the preoperative plain radiographs and MRI for signs of revascularization and preservation of sphericity of the femoral head. Functional outcome was assessed with the comparison of the serial VAS scores and Harris Hip Scores (HSSs). The average follow up was 56 months (range 12-120 months).

Outcome Was Evaluated with Post Operative X-ray, VAS and Subjective Harris Hip Score [12]
Outcome, Score and patients satisfaction was graded by following: Excellent: Hip score: 91-100 (patient was very satisfied). Good: Hip score: 81-90 (patient was satisfied). Fair: Hip score 71-80 (patient was satisfied but not up to the mark).
Poor o: Hip score < 70 (patient was not satisfied).
2. Return to normal density of pre-operative rarefied areas, restoration of normal traebecular pattern.
3. Disappearance of the crescent sign 4. Healing of the cystic areas and of the fracture line.

Statistical Analysis
The information collected on the questionnaires were entered into Microsoft Excel and after data cleaning, the data were transferred and further analyzed using SPSS software version 22. These outcome variables were found to be skewed, hence median values were considered and statistical analysis was done using nonparametric test. The preoperative HSSs and VAS scores were compared with that of HSSs and VAS scores at serial intervals of follow up and final follow up at least 6 months after operation. Statistical analysis was carried out by unpaired students t-test, Chi-square test and Ztest. Level of significance was considered significant when P-value was 0.05 or less.

Results
A total of 64 hips (n=44 patients) were identified as being eligible for the study. The average age of the patients was 30.36±4.64 years (range 18-48 years). 28 (63.64%) males with 42 hips (65.63%) and 16 (36.36%) female patients with 22 hips (34.38%) were included in the study and the sex ratio being 1.75. Out of 44 patients, unilateral hip involved patient was 24 (54.55%) and bilateral hip involved patients were 20 (45.45%), but out of which 10 patients (50%) presented with symptoms pertaining to only one hip, the other hip was in stage I. On the other hand, out of 64 hips, 36 hips had involvement of the right side (56.25%), and left hip involvement was 28 (43.75%). 24 patients (54.54%) had unilateral involvement in study group, but including all hip AVN (all stage III and IV), bilateral involvement was 80% of cases. Out of 64 nontraumatic AVN of femoral head, cortico steroid induced was 48 (75%) hips, idiopathic was 12 (18.75%), ITP related was 2 (3.13%), and alcohol induced was 2 (3.13%). According to Ficat and Aret's staging, out of 64 hips, stage-I was 10 (15.63%), all cases were detected by screening MRI, other hip was stage II or III, stage-II was 44 (68.75%), and early stage III was 10 (15.63%) hips. Average duration of surgery was 90 (range 70-120) minutes in group 1 and 70 (range 50-90) minutes in group 2. Average duration of follow up was 56 (range 12 to 120) months. Clinical evaluation was done by modified Harris hip scoreans VAS. Excellent outcome was 17 (26.56%) hips, good was 21 (32.81%) hips, fair was 8 (12.50%) hips and poor was 18 (28.13%) hips. Excellent, good and fair results were grouped as satisfactory, on the other hand poor result was grouped as unsatisfactory. Overall Satisfactory outcome was 46/64 (71.88%) and unsatisfactory outcome was 18 (28.12%) hips, Radiological progression was not seen in 46 out of 64 hips (71.88%), whereas 18 hips (28.12%) were demonstrated signs of progression and collapse, that was 10 (31.25%) in group-1 and 8 (25%) in group-2. Failure of surgery was defined as progression of the disease, which was n=1/10, 10% (1/5, 20% in group 1, 0% in group-2) in stage-I, n=12/44, 27.27% (7/22, 31.82% in group-1 and 5/22, 22.73% in group-2) of stage II and n=5/10, 50% (2/5, 40% in group 1 and 3/5, 60% in group 2) of stage III disease. Median values of HHS at the end of the follow up in Group I was 80 and 76 in Group 2, compared to the preoperative HHS of 56 and 52 respectively. Median values of VAS at the end of the follow up was 2.82, compared with preoperative 5.85. In stage I, preoperative mean VAS was 2, compared with 1 at 6 month follow up, in stage II disease, preoperative mean VAS was 6, compared with postoperatily, it reduce to 2.75 at 6 months but in stage III disease, preoperative score was 9 and post operative score was 5. Overall satisfactory result was 71.88%, p value was <0.001 by Z test, that is significant. In group 1 satisfactory result was 22/32 (68.75%) and 24/32 (75%) in group 2, no statistic significant difference (>0.05) between two groups. In group 1, 80%, 68.18% and 60% of stage I, II and III respectively were satisfactory results but in group 2, 100%, 77.27% and 40% were satisfactory results. The correlation between the Ficat's stages, Harris Hip score and VAS have shown that majority of hips were showing excellent and good score belonged to stage I and II, 36/54 (66.67%) but no excellent result in stage III disease in both groups, only 20% was good result in stage III. 5/10 (50%) of hip was poor result in stage III AVN, 2/5 (40%) in group 1 and 3/6 (60%) in group 2. The scores difference between pre and postoperatively in stage I and II were also statistically significant but comparison between two groups was not significant, on the other hand in stage III disease, difference between pre and post operative results were not significant and inter-group difference also not significant.  Unilateral hip involvement was 24 (54.55%) and bilateral hip involved patients were 20 (45.45%), On the other hand, out of 64 hips, 36 hips had involvement of the right side (56.25%), and left hip involvement was 28 (43.75%). According to cause, cortico steroid induced was 48 (75%) hips, idiopathic was 12 (18.75%), ITP related was 2 (3.13%), and alcohol induced was 2 (3.13%).   In age group of below 30 years, excellent, good and fair results were 80%, in 31-40 years it was 73.33% but only 57.87% in age group of above 40 years. In stage I AVN, pre operative mean HHS was 77 and post operative mean 93.5, on the other hans VAS was 2and 1.1 respectively, but in stage II disease, preoperative mean HHS was 59.5 and postoperative score was 78. On the other hand VAS was 6.1 and 2.58 respectively. In stage III AVN, preoperative mean HHS and VAS was 41 and 8.25 consecutively, and postoperative score was 64 and 5 respectively. Overall, median values of HHS at the end of the follow up was 78.23, in compare with preoperative score was 59.34 and VAS score was 2.72, compare with 5.80 successively. P value between preoperative and postoperative by Z test, is <0.001 that is significant. P valu between group 1 and 2, by Z test, is >0.05 that is not significant. P valu between preoperative and postoperative in Ficat-Arlet stage I and II, by Z test, is <0.05 that is significant, but in stage III is not significant, >0.05.

Discussion
AVN of femoral head is a major musculo-skeletal problem [12,14]. If it is left untreated it may progress to secondary osteoarthritis of hip joints. The natural progression of the disease is inevitable. Therefore early diagnosis & treatment of AVN may result in favorable outcome of the disease. Medical management of AVN is usually not satisfactory, although some studies have shown the efficacy of bisphosphonate in the early AVN with small necrotic lesion [15]. The advanced stage of AVN is only satisfactory option remains is total hip arthroplasty [16]. The etiology of AVN has been to be multifactorial and may be associated with various risk factors. Those without any obvious risk factor are considered as idiopathic AVNof femoral head [17]. Alcoholism (consuming up to 400ml pr week) has been reported as a major risk factor for AVN of femoral head. Exposure to alcohol consumption and corticosteroid account for 90% of all cuses of AVN has been rported [18,19]. The classification or staging system of AVN of femoral head is important to plan its treatment. The most commonly used classification in clinical practice is Ficat's classification system. We have used Ficat's classification since it is simple, easy to use and has good reproducibility. It grades the disease process into four stages [11,20]. The aim of treatment in early stages of AVN of femoral head is to perform the head preserving procedure by reducing the intraosseos pressure and enhancing the vascular supply of femoral head. Only core-decompression of femoral head was reported with a promising result, initially by Hungrford [21]. The effectiveness of core decompression has always been a controversy in the literature. The long term results have shown that only core decompression gives only temporary relief of pain but does not prevent the progression of the collapse of femoral head. Buckely et al. have reprted good outcomes with the use of cortico-cancellous bone graft along with core decompression in Ficat' stage I & IIA of AVN of femoral head [22]. The permeation of granulation tissues into the drilled area growing from the adjacent MPBG helps in revascularization of the necrotic area, resulting in long lasting pain relief [9,11]. Various free vascularised grafts for treatment of AVN of femoral head to achieve adequate vascularization in the necrosed femoral head, Urbanaik et al [23] have used fibula (with peroneal vascular pedicle). They have shown good clinical& radiological outcome with free vascularisedbone graft in Ficat stages IIB and III also. However these procedure are technically demanding, tedious and time consuming and not feasible bilaterally in one sitting [23,24]. To overcome the short-comings of vascularised grafts, various types of muscle pedicle bone grafting procedure (after core decompression) have been described. These procedures have shown favorable outcome in early stage of the disease with revascularization of the femoral head and preventation of collapse. They described muscle pedicle graft technique include Meyer's procedure using, quadratusfemoris muscle pedicle graft [25], Baksi's procedure usesd tensor fascia lata graft along with iliac crest bone graft [9] and Sartorius muscle pedicle graft by Lie et al. [26] after decompression of necrosed femoral head with multiple drill holes. The shortcoming of graft slippage has been reported. Tensor fascia lata muscle pedicle bone graft is seen to be more vascular than Sartorius graft. The study allowed sufficient decompression by adequate removal of necrotic subchondral bone of the head of the femur at weight bearing area and fixation of graft with one 4mmX40-50mm cancellous lag screw, this prevent slippage of the graft from neck window [9,27]. In this study, pre-collapse stage of osteonecrosis and early collaps with good range of hip movment cases, TFL muscle-pedicle bone graft being cortico-cancellous provided a better strut effect to the sub chondral area and prevented their collapse more efficiently than those obtained by quadratus femoris [25], Sartorius [26] or gluteus medius MPBG having mostly spongy bones [9]. On the other group cortical fibular graft was provided strong support to the disease subchondral bone [8]. The newer technique describing the use of autogenous stem cells from iliac crest and platelet rich plasma for the necrosed area of the femoral head have shown very good result in early stage-I &IIA of the disease but results in advanced stage of disease are unsatisfactory. In our comparative study shown, 80% of Ficat stage I, 68.18% of stage II and 60% of stage III was satisfactory result in core decompression with fibular graft group, on the other hand 100% of ficat stage I, 77.27% of stage II and 40% of stage III shown satisfactory result in core decompression and TFLM graft group. Both the groups are almost similar results but fibular cortical graft was relatively given more structural support and prevent collapse but oprative time and complications were more in fibular graft group. Overall, satisfactory results shown in Ficat and Arlet stage I & II in core decompression and bone graft, either with fibula or muscle pedicle bone graft but even in early stage III, unsatisfactory results was almost 50%. Although THA has been shown to give excellent results in late stage of AVN (advanced-III and IV), but due to the limited life span of these prosthesis become a great problem in younger patient. Core decompression and TFL muscle pedicle bone grafting or non vascular fibular graft reduced intra osseous tension and to achieve early revascularization of the ischaemic femoral head that relieved hip pain, improve hip function, also prevent collapse of femoral head and able to achieve satisfactory out come. Both the procedures have almost the equal result.