Vascularized Fibular Grafting for Osteonecrosis of the Femoral Head

Published on 11/04/2015 by admin

Filed under Orthopaedics

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 1584 times

CHAPTER 31 Vascularized Fibular Grafting for Osteonecrosis of the Femoral Head

Introduction

The first documented report of osteonecrosis of the femoral head (ONFH) appeared more than a century ago, yet, since then, few definitive conclusions have been reached regarding its causes or treatment. Approximately 20,000 new cases are diagnosed each year, and ONFH accounts for 5% to 12% of total hip replacements performed in the United States annually. The natural history of ONFH is subchondral collapse with a squaring of the femoral head that leads to hip degeneration. Because of this natural history and the fact that ONFH has a proclivity to manifest in younger patients, most surgeons have traditionally believed that procedures aimed at preserving the native hip are preferable to joint replacement.

Conceptually, the ideal procedure would remove the necrotic bone within the femoral head and replace this void with healthy bone that is replete with a nascent vascular source to improve the likelihood of restoring a vital subchondral plate. Such a procedure would relieve the patient’s pain, preserve or restore the sphericity of the femoral head, and ultimately prevent the deterioration of the hip. Fulfilling these criteria would likely prevent the need for a second surgery (i.e., arthroplasty); this is an important objective, particularly for younger patients. We feel that the free vascularized fibular graft more closely addresses these objectives as compared with any other biologic-preserving procedure currently available; it is our preferred method for the treatment of ONFH. We have outlined our indications, contraindications, results, and surgical technique, and we have highlighted some of the critical steps that we feel are paramount to obtaining successful results with this procedure.

Basic science

There are more than 100 conditions or factors that are known to activate intravascular coagulation and to potentially cause osteonecrosis. A growing body of evidence supports the theory that ONFH in adults and Legg-Calvé-Perthes disease in childhood is related to an underlying thrombophilia or hypofibrinolysis. A recent study reported that 136 of 206 patients (66%) with osteonecrosis revealed abnormal clotting values of the tested factors. In a follow up study of patients with ONFH, activated protein C resistance was the most frequent abnormality found, with a 50% prevalence. Lipoprotein A is also thought to play a role in the development of osteonecrosis, with a reported prevalence of 32% in one study of 124 subjects with ONFH.

Aberrant microvascular anatomy has also been implicated in the development of ONFH. With the use of digital subtraction angiography, one study showed that 94% of 99 hips with osteonecrosis demonstrated abnormal hip vasculature as compared with 31% (5 out of 16) abnormal findings in the control group. Regardless of the cause, we know that the eventual pathway ends with bone death that results from inadequate perfusion.

Despite this wealth of associated conditions and known risk factors, no one factor has proven to have a direct causative effect. In fact, the ultimate development of ONFH is more likely the unfortunate nexus of multiple risk factors that individually fall short of causing osteonecrosis but that collectively tip the scales toward an intraosseous ischemic event.

We have studied the efficacy, both clinically and in the laboratory, of the free vascularized fibular graft (FVFG) for the treatment of ONFH. With the use of a canine model in which ONFH was induced with cryotherapy, we implanted into the canine femoral head core either nonvascularized bone grafts, pedicled muscular flaps, or vascularized bone grafts. The necropsy femoral head analysis at 1 year demonstrated statistically significant increased trabecular thickness only in the vascularized bone grafts.

Indications

For all patients, as adjuncts to this list of objective criteria, we include the patient’s “hip health,” which is determined by the degree of limp, the limitation of motion, and the degree and frequency of pain. For example, if we are equivocal about the benefit of the FVFG for a certain patient on the basis of radiographic analysis alone but discover that the patient has only mild groin pain and nearly full or full range of hip motion, we are inclined to offer the procedure. Conversely, if this same patient reports severe hip pain with limited hip motion and a severely antalgic gait, we are more likely to recommend arthroplasty for that hip.

In our hands, core decompression alone is effective only for patients who are less than 50 years old with a stage I or stage II central lesion of dense bone (not cystic) that involves less than 20% of the femoral head. If the procedure is performed for conditions outside of these parameters, we believe that core decompression may exacerbate the already compromised vascular status of the femoral head and ultimately hasten the progression of the condition and the subchondral collapse.

Surgical Technique

Fibular Graft Harvest

With the use of tourniquet control, a straight, lateral, 15-cm longitudinal incision is made coincident with the natural sulcus between the lateral and posterior compartments of the leg. The incision is begun at least 10 cm distal to the fibular head, and it ends at least 10 cm proximal to the lateral malleolus. The peroneal muscles are reflected in an extraperiosteal fashion off of the lateral aspect of the fibula, working from posterior to anterior and stopping when the anterior intermuscular septum is visualized (Figure 31-2).

The anterior intermuscular septum is then divided to expose the anterior musculature, which is reflected bluntly off of the fibula. At this point, the interosseous membrane is easily visualized, and the adjacent anterior musculature, with its accompanying deep peroneal nerve and anterior tibial artery, is gently swept off of the interosseous membrane and away from the fibula. With the use of a specially designed right-angle beaver blade, the interosseous membrane is divided close to its fibular attachment along the entire length of the proposed fibular graft. The posterior intermuscular septum is then divided to expose the posterior muscles: the soleus proximally and the flexor hallucis longus distally.

Fibular Osteotomy

Directly beneath the distal aspect of the flexor hallucis longus muscle, the distal pedicle of the peroneal vessels is identified, and malleable retractors are passed between this pedicle and the fibula. Diligent care is critical during the placement of the retractors to ensure the protection of the pedicle during the osteotomy. After reconfirming that the planned osteotomy is at least 10 cm proximal to the distal tip of the fibula, an oscillating saw is used to cut the fibula. Irrigation during the osteotomy is vital to prevent thermal osteonecrosis. Next, the proximal pedicle is identified deep to the soleus muscle along the posterior aspect of the fibula (Figure 31-3). It is protected, and the proximal fibular osteotomy is performed in a manner similar to that of the distal osteotomy. The fibular cuts are made 15 cm apart to ensure an adequate pedicle length. It is important when performing the proximal osteotomy to identify and protect the superficial peroneal nerve, which is exposed proximally on the deep surface of the peroneus longus muscle.

After the proximal and distal fibular osteotomies are performed, a bone clamp is placed around the fibula to allow for better control and easier rotation during the delicate pedicle dissection. Starting distally, the peroneal vessels are again identified, isolated, and divided with the use of hemostatic clips. The now-free distal pedicle is attached to the distal aspect of the fibula with a hemoclip to ensure that the peroneal vessels and any nutrient branches to the bone are not avulsed from the fibula during the remainder of the harvest. The fibula and the adjoining peroneal vessels are then dissected from the surrounding flexor hallucis longus, posterior tibialis, and soleus muscles. The fibula is elevated until it is tethered only by the proximal vascular pedicle. The tibial nerve can often be seen coursing in close proximity to the peroneal vessels at this level and should be carefully dissected away. After adequate pedicle length is established, the vessels are ligated with two large hemostatic clips and divided with scissors, and the graft is passed to a back table. The tourniquet is then deflated, the wound is copiously irrigated, and any bleeding is addressed. The leg wound is closed a short time later during the vascular anastomosis at the hip. The deep fascial layers of the leg are not closed in an effort to prevent compartment syndrome. The subcutaneous layer and the skin are closed over a drain, and the leg is wrapped in a soft, bulky dressing.

Preparation of the Fibular Graft

On a back table, the artery and the two veins of the fibular pedicle are delineated from one another and separated with the use of microscissors and jewelers’ forceps. All three vessels are then irrigated with a heparin-impregnated lactated Ringer’s solution and visually inspected for any major leaks. Neither vein will fill over the entire length of the graft because of the valves, but the artery should insufflate throughout its entire course during the injection of the heparin solution. Some oozing from the attached muscle and the periosteum is anticipated and considered normal; however, we consider any leaks from the main vessels that form a stream to be major and worthy of repair with either 8-0 suture or micro hemostatic clips. Such attention to detail is imperative to minimize the risk of the patient’s developing a vascular steal and thus to ensure adequate endosteal blood flow after the anastomosis. The vein with the better size match is chosen as the recipient, whereas the other vein is ligated with a hemostatic clip. The diameter of the fibula is reported to the hip surgeon to determine the endpoint for the core reaming of the hip; this is discussed in detail later in this chapter. Next, the proximal pedicle is reflected in a subperiosteal fashion from the fibula until a nutrient vessel is seen entering the cortex (Figure 31-4). The length of the pedicle at this point should be approximately 4 cm to 5 cm. The proximal fibula is then cut with an oscillating saw at the level of the most proximal nutrient vessel, with the pedicle being protected during the osteotomy. Again, copious irrigation should accompany all osteotomies to prevent thermal necrosis. After the exact length of fibula required has been determined by the preparation of the proximal femur (see Operative Procedure on the Hip Section, p. 254), this length is measured and marked on the fibular graft. The distal pedicle and a small cuff of evaginated periosteum are secured to the distal extent of the fibula with a 4–0 absorbable suture to prevent the stripping of the pedicle and the periosteum during insertion into the femoral core (Figure 31-5).

Operative Procedure on the Hip

Certain anatomic landmarks are marked to assist with the placement and design of the surgical incision (Figure 31-6). The lateral aspect of the femur is approached through an interval between the tensor fascia lata and the gluteus medius. The vastus lateralis is then encountered, and the donor vessels (i.e., the ascending branch of the lateral femoral circumflex artery and two veins) are identified as they lie between the rectus femoris and the vastus intermedius. After the vessels are identified, the origin of the vastus lateralis is reflected sharply from the vastus ridge and then in a posterior direction for approximately 5 cm. The origin of the vastus intermedius is then carefully detached with a right-angle clamp and knife from its anterior position on the proximal femur. A specially designed four-quadrant retractor is introduced to provide better visibility for the dissection of the donor vessels.

Preparation of the Femoral Head

A C-arm fluoroscope is draped with a sterile sleeve and then positioned over the patient’s hip region like an arch (Figure 31-7); this provides for the obtaining of anteroposterior and frog-leg lateral views of the proximal femur with relative ease. Starting no lower than the middle of the lesser trochanter and at the junction of the middle and posterior third of the lateral femur, a 3-mm guide pin is inserted under fluoroscopic control into the center of the necrotic nidus within the femoral head. Pin position must be checked on both anteroposterior and lateral views. The pin must not only be positioned within the center of the necrotic bone, but it must also be spaced appropriately between the cortices of the femoral neck to allow for the passage of a large reamer (Figure 31-8). With correct pin placement confirmed by fluoroscopy, sequential reaming ensues over the guide pin; the procedure starts with a 10-mm reamer, the reamers are then increased in size, and the increasing then stops at the measured diameter of the harvested fibula. The reaming should extend to within 3 mm to 5 mm of the femoral head subchondral plate. This portion of the reaming is best performed with the use of live fluoroscopic guidance. Necrotic bone removed during the reaming process is discarded, whereas healthy appearing bone is saved for later grafting (Figure 31-9). Additional bone is captured with a filtered suction tip (KAM Super Sucker, Anspach, Palm Beach Gardens, FL) during the reaming process, when bone slurry is expressed from the core. Bone from this process is emptied onto a surgical sponge, dried, and fashioned by the scrub nurse into rectangular “bullets” to be used later for grafting. After the final straight reamer is passed, the guide pin is removed, and a special ball-tip reamer is introduced into the femoral core. With the use of fluoroscopic control, additional necrotic bone is excavated from the femoral head to create a bulbous cavity, usually in the anterior and superior quadrants. This step is performed with a water-soluble radiographic contrast medium injected into the femoral core to assess the adequacy and amount of the necrotic bone removed (Figure 31-10, A and B).

Next, with the use of a large curette, cancellous bone is taken from the greater trochanteric region, with the use of the proximal femoral core as an access point. Some of the cancellous graft is then placed into the femoral head cavity with DeBakey forceps and impacted with the use of a custom-made cancellous bone impaction instrument. This instrument is particularly helpful for elevating the sunken subchondral floor in cases of femoral head collapse. At its inserted end, the device has several windows through which additional cancellous bone is extruded with the help of a specialized drill bit, thus filling voids in the subchondral bone (Figure 31-11). With the impactor fully inserted, the length of the fibular graft is determined by reading the circumferential markings (units in millimeters) on the impactor’s side. After the cancellous bone is inserted, the bone “bullets” created from the reamings are inserted and extruded likewise into the femoral head cavities. Finally, contrast material is reinjected to confirm the adequate filling of the femoral subchondral voids (Figure 31-12, A and B). The femoral head is now ready to receive the fibular graft.

Vessel Anastomosis

The four-quadrant hip retractor is replaced to optimize the exposure of the harvested vessels (i.e., the ascending branch of the lateral femoral circumflex artery and the two veins). Attention is initially directed toward the venous anastomosis. We have continued to enjoy success with our venous anastomoses with the use of a coupling device (Microvascular Anastomotic Coupler system, Medical Companies Alliance, Homewood, AL). The device comes in sizes that range from 1.0 mm to 3.5 mm in 0.5-mm increments; however we use either the 2.5 -mm or 3.0 -mm coupler for the majority of cases. After this, the microscope is brought into the surgical field, a blue microsurgical suction mat (Micromat, PMT, Chanhasen, MN) is placed as a backdrop, and the arterial anastomosis is completed with the use of an 8-0 or 9-0 black nylon monofilament suture with a 100-μm needle (Sharpoint, Pearsalls Limited, Taunton, Somerset TAI, IRY, UK) (Figure 31-13). After the anastomosis, the repair site is observed for any leaks. We will often place a small local fat graft over a minor leak, whereas larger leaks require additional suturing. Next, the exposed end of the fibular graft is observed for endosteal bleeding. Bleeding from the endosteal vessels is seen within 5 minutes of completing the anastomosis in more than 90% of cases. For those cases in which flow is absent after 5 minutes, we recommend several steps. First, we check the patient’s blood pressure and core body temperature. If either or both are low, elevation will often produce adequate flow. In addition, we irrigate the medullary canal of the exposed fibula with papavarin and heparin. If this does not improve the flow, we recheck the anastomosis site for leaks and perform a patency test on both sides of the anastomosis. If flow is sluggish or absent on the femoral side of the anastomosis, we trace the artery back to its origin from the lateral femoral circumflex artery. The ascending branch can be either kinked or tethered anywhere along this course. If this is the case, the removal of the aggravating tissue (most often the vastus intermedius) will often allow for adequate flow. If poor flow is localized to the fibular pedicle side, we check the vein that was not used for the venous anastomosis, because often this vein can have a small leak that induces spasm throughout the vessels or that causes a vascular steal. Any leaks are repaired with either 8-0 suture or micro hemoclips. Rarely if ever does the graft have to be removed entirely for size adjustment if accurate measurements have been made during the preparation of the core and of the fibular graft. If all else fails, the anastomosis is redone, and an interpositional vein graft is used if there is tension.

The vastus lateralis and the intermedius muscles are not reattached during closure for fear of constricting the vascular pedicle. The tensor fascia lata and the iliotibial band are closed over a drain. The subcutaneous tissue and the skin are closed in the same manner as the leg wound.

Postoperative rehabilitation

Postoperatively, all patients are placed on an intravenous infusion of dextran for 3 days and then transitioned to aspirin and Persantine daily, which is continued for 6 weeks. The operative drains, the Foley catheter, and the epidural catheter are removed on the second postoperative day, and physical therapy is also initiated on that day. The average hospital stay is 3 days. Patients do not bear weight on the operative side for 6 weeks, after which time progressive weight bearing is permitted. Full weight bearing is achieved by 5 to 6 months. We recommend aquatherapy and stationary bicycling at the 6-week postoperative mark. Patients are encouraged to begin early active and passive motion of the toes and ankle, with special emphasis on the passive stretching of the great toe. The great toe is susceptible to a flexion contracture as a result of the scarring of the flexor hallucis longus muscle. Follow up radiographic and clinical examinations are performed at 3 months, 6 months, and yearly thereafter. Unrestricted activity is allowed at 1 year after surgery (Tables 31-1 and 31-2).

Table 31–1 Results and Outcomes for Patients with Avascular Necrosis of the Femoral Head Treated with Free Vascularized Fibular Grafting

Type No. of Cases No. of Revisions
Idiopathic 492 107 (22%)
Steroids 726 123 (17%)
Alcohol 380 77 (20%)
Trauma 274 53 (17%)
Perthes 25 2 (8%)
Other 37 4 (11%)
SCFE 32 1 (3%)
Pregnancy 34 2 (6%)

SCFE, Slipped capital femoral epiphysis.

Complications

Less than 1% of our patients have required a blood transfusion, and we have had only four infections occur during the entire duration that we have performed the procedure.

A contracture of the great toe has occurred in our series in 3% of cases. We have performed Z-lengthenings of the flexor hallucis longus in those patients whose contractures were profound enough to either impede the normal gait or cause painful pressure on the tip of the toe.

Subtrochanteric femur fractures occur at a rate of approximately 1%. Since limiting the harvest of cancellous graft to only the greater trochanter (and not also harvesting from the lesser trochanter), our incidence of subtrochanteric fractures has decreased from 2% to 1%. The majority of these fractures occur 6 to 8 weeks postoperatively, which is typically when patients begin to feel better and place excessive weight on the leg, with a torsional force being applied (i.e., standing with weight on the leg and turning the torso).

Irritation of the lateral cutaneous branch of the superficial peroneal nerve occurs to some degree in less than 10% of patients. In the majority of these patients, the condition resolves completely by the 6-month postoperative visit.

Ankle pain is reported in up to 5% of patients, but, with few exceptions, this resolves by the 1-year postoperative mark. Uchiyama reported that ankle stability is maintained if at least 6 cm of distal fibula remains. We are careful to always leave 10 cm of distal fibula from the graft harvest.

Some patients experience trochanteric bursitis, which typically manifests between 3 and 12 months postoperatively. This seems to be related to a prominent Kirschner wire or to heterotopic bone being prominent at the trochanteric flare. The majority of patients respond to local steroid and lidocaine injections, but a select few with recalcitrant symptoms ultimately require surgery to remove the pin or the heterotopic bone.

Annotated references and suggested readings

Aldridge J.M.3rd, Berend K.R., Gunneson E.E., Urbaniak J.R. Free vascularized fibular grafting for the treatment of postcollapse osteonecrosis of the femoral head. Surgical technique. J Bone Joint Surg Am.. 2004;86-A(suppl 1):87-101.

This article is part of a series of surgical technique guides. It describes in thorough detail the surgical steps to take when performing the free vascularized fibular graft procedure for the treatment of femoral head osteonecrosis..

Aldridge J.M., Urbaniak J.R. Avascular necrosis of the femoral head: role of vascularized bone grafts. Orthop Clin North Am.. 2007;38(1):13-22.

This article presents the history, development, and results of the various techniques of vascularized bone grafting for the treatment of osteonecrosis of the femoral head. The results of treating more than 2800 patients who had femoral head osteonecrosis with the use of a vascularized fibular graft by way of an intraosseous approach are summarized, and certain pearls and pitfalls regarding the treatment of femoral head osteonecrosis with the use of a free vascularized fibular graft are highlighted..

Kawate K., Yajima H., Sugimoto K., et al. Indications for free vascularized fibular grafting for the treatment of osteonecrosis of the femoral head. BMC Musculoskelet Disord.. 2007;8:78.

This study prospectively tracked 71 hips in 60 patients for an average of 7 years. Radiographs, Harris Hip Scores, and survivorship were evaluated. Overall survivorship for the group was 83% at 7 years. The authors delineate certain preoperative factors that portend an inferior clinical outcome..

Kim S.Y., Kim Y.G., Kim P.T., Ihn J.C., Cho B.C., Koo K.H. Vascularized compared with nonvascularized fibular grafts for large osteonecrotic lesions of the femoral head. J Bone Joint Surg Am.. 2005;87(9):2012-2018.

This prospective case-control study compares hips treated with either a vascularized fibular graft or with a nonvascularized fibular graft, with a mean duration of follow up of 4 years. The rates of radiographic progression and collapse were significantly lower and the mean dome depression was significantly less in the group that was treated with a vascularized fibular graft as compared with the group treated with a nonvascularized graft. The authors conclude that vascularized fibular grafting was associated with better clinical results and that it was more effective than nonvascularized fibular grafting for the prevention of collapse of the femoral head in a matched population with a Steinberg stage IIC or larger osteonecrotic lesion..

Marciniak D., Furey C., Shaffer J.W. Osteonecrosis of the femoral head. A study of 101 hips treated with vascularized fibular grafting. J Bone Joint Surg Am.. 2005 Ap;Volume 87(4):742-747.

Plakseychuk A.Y., Kim S.Y., Park B.C., Varitimidis S.E., Rubash H.E., Sotereanos D.G. Vascularized compared with nonvascularized fibular grafting for the treatment of osteonecrosis of the femoral head. J Bone Joint Surg Am.. 2003;85-A(4):589-596.

This is a retrospective case-control study with a large number of patients that compared the clinical and radiographic results between two groups that were treated with either a vascularized fibular graft or a nonvascularized fibular graft. The results of this study strongly suggest that vascularized fibular grafting is associated with better clinical and radiographic results..

Sotereanos D.G., Plakseychuk A.Y., Rubash H.E. Free vascularized fibula grafting for the treatment of osteonecrosis of the femoral head. Clinical Orthopaedics & Related Research (344); 1997 No:243-256.

Urbaniak J.R., Coogan P.F., Gunneson E.B., Nunley J.A. Treatment of osteonecrosis of the femoral head with free vascularized fibular grafting: a long-term follow-up study of one hundred and three hips. J Bone Joint Surg.. 1995;77A:681-694.

This level IV study retrospectively evaluated the outcomes of 103 hips that were treated with free vascularized fibular grafting. Harris Hip Scores and SF-12 forms are the clinical measurements that were used, whereas radiographs were evaluated for changes in the femoral head and joint space. The overall success rate for all patients, regardless of their preoperative stage, was 80% at a 5-year follow up..

Urbaniak J.R., Jones J.P. Osteonecrosis: etiology, diagnosis, and treatment. AAOS publication, 1997.

This textbook is the only comprehensive text that addresses osteonecrosis of the human skeleton. All facets of the disease process are discussed in detail..

Vail T.P., Urbaniak J.R. Donor-site morbidity with use of vascularized autogenous fibular grafts. J Bone Joint Surg Am. 1996 Fe;Volume 78(2):204-211.

Zhang C., Zeng B., Xu Z., et al. Treatment of femoral head necrosis with free vascularized fibula grafting: a preliminary report. Microsurgery. 2005;25(4):305-309.

This retrospective study reported short-term results (16 months) for 56 hips with the diagnosis of osteonecrosis of the femoral head. All but 3 femoral heads demonstrated on radiographs to have either no change (25%) or improvement (69.6%). Harris Hip Scores improved for all patients..

www.dukehealth.org/FVFG.

This is a thirty-page pdf file found online at Duke University. It addresses the many questions and challenges that the over 3000 patients treated by the two authors have had over the years. It is a comprehensive layman’s review of the procedure..