Percutaneous Treatment of Long Bone Nonunions: The Use of Autologous Bone Marrow and Allograft Bone Matrix

ABSTRACT

Sixty-six patients with 69 "stiff" nonunions (no gross motion) of long bones were entered into a prospective study. The only therapeutic intervention was the percutaneous administration of a mixture of autologous bone marrow and allograft demineralized bone matrix on an outpatient basis. Sixty-one of the percutaneous treatments (88%) resulted in union at an average of 8.1 months (range: 2 months to 3 years). This method of treating nonunions is as successful as standard iliac crest autologous bone grafting and offers the distinct advantages of decreased morbidity, reduced costs, and shorter hospital stay.

Despite continued advances in the treatment of long bone fractures, nonunion remains a rare but difficult challenge. Numerous etiologies exist, including interposition of soft tissues, distraction at the fracture site, infection, decreased vascular supply, and excessive motion.

To successfully overcome nonunions, one must address all of the potentially correctable factors. Once the goal of stable fixation has been met, attention may then be given to other local factors that contribute to delayed healing. These include the availability of recruitable healthy multipotential cells and local growth factors.

Because autologous cancellous bone grafting has been the standard, an alternative substance or method must be equally successful in achieving union, as well as providing some increased benefit, to justify its use. Percutaneous administration of autologous bone marrow and allograft demineralized bone matrix (DBM) offers the advantage of decreased morbidity associated with the classic open grafting techniques. Additional advantages include decreased cost and hospital stay, as the procedures are performed in an outpatient setting.

This article details the experience in a prospective series of patients with established long bone nonunions and stable fixation who underwent percutaneous grafting with a combination of autologous bone marrow and allograft DBM.

Materials and Methods

The following criteria were used for inclusion in this study:

• Established nonunion >6 months after injury with no evidence of progressive healing for the previous 3 months. In many patients, tomograms were necessary to adequately document the nonunion.
  
  
• No evidence of motion at the nonunion site. Preoperative work-up included appropriate roentgenographic investigation to document the presence of a stiff nonunion. Often, stress radio-graphs or intraoperative fluoroscopic demonstration of an absence of motion was necessary. Only patients with rigid nonunions were included in the study.
  
  
• No surgical or treatment intervention in the past 3 months.
  
  
• No evidence of active infection, although a history of a resolved infection was not a contraindication.
  
  
• No smoking for 6 weeks before and during the treatment period. If patients were smoking at the time of the first evaluation, they were asked to discontinue. These patients were then re-evaluated at 6 weeks with new studies to document that there had been no progressive healing. If there was a question as to the patient’s nonsmoking compliance, carboxyhemoglobin levels were obtained. Likewise, if there was a question of compliance throughout the study, these levels were again performed.
  
  
• A state of good overall health. No concurrent interfering medications such as cortisone preparations or antimetabolic drugs were allowed in the study.

Patients meeting all of the above criteria were entered into the protocol. Sixty-six patients with nonunions underwent 69 percutaneous grafting procedures. Overall, 37 men and 29 women with an average age of 42 years (range: 15-81 years) were studied.

Nonunion etiology included 27 open fractures, 30 closed fractures, and 12 postoperative nonunions (6 ankle fusions and 6 tibial or femoral osteotomies). Twenty-four of 66 patients (36%) had a previous history of infection. There were a variety of nonunion sites (N=69); the tibia was the most common bone injected (36), followed by the femur (16), failed ankle fusion (6), humerus (4), ulna (4), radius (2), and fibula (1). Of the types of nonunions at initial presentation, 22 were hypertrophic and 47 were atrophic.

Following the outpatient procedure, each participant was followed on a regular basis with clinical examinations, roentgenograms, and functional evaluations. Union was determined on the basis of clinical evaluation and strict roentgenographic criteria (Table). Roentgenograms were taken at 3, 6, and 12 weeks, and then every 2 to 3 months until a determination could be made with regard to healing. Anteroposterior, lateral, and two oblique views were taken during office visits and, in cases where bony bridging was questionable, tomograms were also obtained.

Table 1. Postoperative Radiographic and Clinical Evaluation Radiographic Evaluation Clinical Evaluation No bone formation Bone formation at injection site Callus formation Bridging callus Cortical contact (circle one): <25% 25%-50% 50%-75% 100% Pain with rest Pain with weight bearing Pain with palpation Clinical healing Union Postoperative Complications Final Physical Assesment 1. None 2. Nerve injury 3. Pain at injection site 4. Pain at donor site 5. Infection 6. Malunion 99. Other ______________ 1. Union (cortical contact/no pain) 2. Progressive healing 3. Delayed healing 4. Nonunion 99. Other ________________

In addition, patients evaluated their pain, activity, emotional acceptance, functional ability, and need for assistive devices at each office visit by completing the Musculoskeletal Tumor Society’s Functional Evaluation, which had been modified to a patient-based format.

Radiographically, a patient was considered healed when cortical bridging was at least 75% of the circumference of the bone and pain at the nonunion site had resolved. Four patients who had an initial partial response with symptomatic relief were reinjected a second time in an attempt to complete the cortical bridging. The average follow-up period was 21.7 months (range: 3 months to 4.5 years).

Surgical Procedure

Figure 1. Intertrochanteric nonunion showing cannula/trocar used to access the defect under fluoroscopic guidance.

Patients were taken to the operating room where, under general or spinal anesthetic, the procedure was performed. For purposes of consistency, a hematologist/oncologist experienced in bone marrow transplantation harvested the autologous marrow from either the anterior or posterior iliac crest using standard technique.¹

After appropriate prepping and draping, the area of the nonunion was located and mapped out using fluoroscopic control. A 1.5-cm skin incision was made over each grafting area. The skin incisions were located in a manner so that the grafting material could be placed circumferentially around the nonunion site. In the majority of patients, two incisions were necessary.

A special set of percutaneous instruments (Wright Medical Technology, Inc, Arlington, Tenn) was designed to facilitate percutaneous localization, dissection, biopsy, and graft introduction around the nonunion site. A trocar was placed through a cannula into the incision and gently inserted down to bone (Figure 1). Care was taken to prevent direct injury or traction of the neurovascular structures throughout the procedure.

Generally, the recognized principles for placement of external fixation pins as reported by Paley et al² were followed. After the trocar was placed down to bone, the position was checked with fluoroscopy. The trocar was then removed leaving the cannula sheath in place.

Depending on the consistency of the tissue, various percutaneous periosteal elevators were used to develop a sub-periosteal envelope around the nonunion site. A modified curetting instrument was used to obtain tissue from the nonunion for biopsy and culture. The anatomy and consistency of the tissue dictated whether a debridement was performed. In most cases, this was not necessary, and the intramedullary space and intracortical fibrous tissue were not otherwise disturbed.

Once an adequate potential space was created by developing this subperiosteal envelope, the cannula was then positioned for intrusion of the graft material.

Aspirated autologous marrow was mixed with a DBM composite. The material was briefly mixed and inserted into a funnel device that was placed onto the cannula. A plunger was then used to insert the combined material into the space previously developed.

The goal was to have a layer of injected material surrounding the entire circumference of the bone and approximately 1-2 cm proximal and distal to the bony discontinuity. This was usually accomplished via two incisions. Care was taken to avoid excessive volume in areas where compartment syndrome or neurovascular compromise were of concern. The incision for the bone marrow harvest was closed with an adhesive strip closure. The recipient’s grafting incisions were closed with either a subcuticular absorbable suture or adhesive strips. No other procedure was performed, such as screw removal or dynamization, nor was a tourniquet used (Figure 2).

Figure 2. Ignite ICS Biocomposite (Wright Medical Technology, Inc, Arlington, Tenn) used in an intertrochanteric fracture. The preoperative film 10 months before injury (A). Intraoperative image showing the Ignite ICS Cannula (Wright Medical Technology, Inc) (B). Eight-week postoperative film showing callus formation and apparent bony union (C).

Postoperatively, patients recovered and were discharged through the out-patient department. No other changes were made in the patients’ management, immobilization, or weight bearing status. For example, if patients were weight bearing in an orthosis preoperatively, they continued using such a device postoperatively. If they were partially weight bearing preoperatively, they continued with this activity restriction. No changes in fixation or other variables were instituted subsequent to the percutaneous procedure.

Results

The average follow-up period was 21.7 months (range: 3 months to 4.5 years). A total of 61 sites (88%) healed following percutaneous treatment. Eight nonunions required a second percutaneous procedure for delayed healing, of which seven went on to achieve union. At the last follow-up, one nonunion persisted, most likely due to continued nicotine use by the patient. Six nonunions were interpreted as frank failures.

The average time from the date of injury to the injection was 21 months (range: 6 months to 16 years). The average time to union after the bone marrow and DBM injection was 8.1 months (range: 2 months to 3 years), which correlates with previous marrow injection studies^3-5 as well as open iliac crest grafting results.^6,7

There were no perioperative complications that could be attributed to the percutaneous harvesting and grafting technique. Most patients had only moderate discomfort at their donor iliac crest site, as well as their recipient site, which generally resolved in a few days. Several patients developed subcutaneous, hard bony masses several centimeters in diameter at the injection site. These areas were first evident 3-6 weeks post injection and resolved without intervention over several months.

Case 1. A 55-year-old man with malignant fibrocystic sarcoma of the right proximal tibia. The patient was treated with wide local resection with replacement of the resected tibia with a cemented allograft intercalary construct. The patient has union at the proximal junction site 8 months postoperative, but he does not have bridging bone at the distal site. At this time, the patient was treated percutaneously with a mixture of demineralized bone matrix and autologous marrow (A). Anteroposterior roentgenogram 2 months after the percutaneous treatment shows mature bridging callus (B). Anteroposterior roentgenogram 8 months after percutaneous treatment shows mature remodeled callus in the area of the graft placement (C). At this point in treatment, the patient was asymptomatic for weight bearing and resumed normal activities. Case 2. A 52-year-old woman with an atrophic humerus fracture secondary to an automobile accident. The patient was treated for her nonunion with an intramedullary rod and autologous iliac crest graft (A). The patient still had a painful nonunion 6 months after surgery. The patient has new callus bridging the nonunion site 2 months after percutaneous treatment with autologous marrow and demineralized bone matrix (B). She was asymptomatic and resumed normal activities at this time. The patient had corticalization of the nonunion site 5 months after percutaneous injection.

Discussion

Autologous bone grafting for nonunion remains the standard by which all other techniques are compared. Autologous bone potentially contributes three vital local components: osteoconduction, osteoinduction, and osteogenic cells.^8-11 To be successful, an alternative technique must also provide these osteogenic components. More recently, numerous substitutes to autologous bone grafting have been reported with varying degrees of success.^12-19

Connolly et al^3,4,20 and Healey et al²¹ have demonstrated that percutaneous injections of autologous bone marrow can successfully treat between 78% and 95% of nonunions. Tiedeman et al⁵ and Wilkins and Stringer²² have published data on the use of DBM in the treatment of osseous defects. As reported by Connolly,³ perhaps one of the more important facets of this percutaneous procedure is that it encourages early treatment of fractures when healing problems are anticipated or becoming evident.

In this study, an attempt was made through the inclusion criteria to eliminate as many variables as possible in this group of patients. Basically healthy, nonsmoking patients with rigid or stiff nonunions were entered. Overall, no changes in fixation, activity, or weight bearing status were instituted. The only intervention was the percutaneous procedure and its short perioperative influences. However, variabilities in the osteoinductive activity of DBM are known to exist.

The ability of bone marrow to form bone was first demonstrated in the late 1860s by Goujon.²³ Since then, a variety of laboratory and animal studies have demonstrated the presence of osteogenic precursor cells in marrow capable of bone formation.^{9,10,14,24,25}

The clinical use of autologous bone marrow has been shown to also be effective in the treatment of fracture nonunions.^3,4,20,21 Despite this success, marrow lacks certain properties, such as readily available growth factors, that are important elements involved in efficient bone healing.

Demineralized bone grafts were also used in the late 1800s by Senn²⁶ for reconstruction of osseous defects. Numerous studies have investigated the use of this material in a variety of situations. Animal studies have demonstrated the reliable use of demineralized grafts to promote osteogenesis and bridge segmental osseous defects.^27-30

A clinical study performed by Urist and Dawson³¹ in 1981 also proved successful using this material. They demonstrated that DBM provides both osteoinduction (the phenotypic conversion of multipotential cells into osteoblasts) and osteoconduction (the scaffolding for bone ingrowth).^11,31,32 As already stated, bone marrow provides osteogenic cells that have the potential for differentiation. Therefore, the combination of autogenous bone marrow and allograft DBM theoretically provides the necessary elements to aid in successful bone healing by significantly changing the local milieu.

Numerous animal studies have confirmed the synergistic effect of these two materials,^32-34 although few clinical studies have been performed. Tiedeman et al⁵ demonstrated successful use of a composite of DBM and bone marrow in nonunions, intended fusion sites, acute fractures with comminution, and to fill cavitary lesions.

Similarly, our current study also revealed excellent results with these materials, with an overall union rate of 88%. These figures, as well as the time to achieve union, are comparable to those for the standard open techniques using cancellous iliac crest bone graft.

Although open graft harvesting and implantation is thought to be a relatively simple procedure, it has not been without complication. Donor site problems include infection, bleeding, hematoma, chronic pain/painful scar formation, fracture, hernias, sensory loss, and occasional gait disturbances.^6,35-40

Aside from these potential complications, additional morbidity can occur with opening the recipient nonunion site. Possible further devascularization in this area can result in increased risks of infection and additional delay in fracture healing.⁴¹ As demonstrated in Connolly’s work, as well as this study, percutaneous harvesting and graft injection are associated with greatly reduced morbidity compared with the classic open grafting technique. In addition, advantages such as decreased cost and hospital stay have also been observed.

One must include, however, a discussion of the potential of disease transmission when allogeneic material is used. Patients must be counseled and informed of these risks prior to any surgical procedure. Strict screening programs have been instituted for nearly all tissue banks, especially those affiliated and approved by the American Association of Tissue Banks.⁴²

Conclusion

Current data offer encouraging results for the use of percutaneous autologous bone marrow and allograft DBM composite for the treatment of rigid nonunions. Although this technique does not promote more rapid healing than classic open surgical grafting, it offers the benefits of decreased morbidity, reduced costs, and shorter hospital stays. With the successful treatment of nonunions now established, the potential benefits associated with acute fracture management remain to be studied.

References

1. Ilizarov GA, ed. Transosseous Osteosynthesis: Theoretical and Clinical Aspects of the Regeneration and Growth of Tissue. New York: Springer-Berlag; 1992:120-124.
2. Paley D, Catagni MA, Argnani F, Villa A, Benedetti GB, Cattaneo R. Ilizarov treatment of tibial nonunions with bone loss. Clin Orthop. 1989; 241:146-165.
3. Connolly JF, Guse R, Tiedeman J, Dehne R. Autologous marrow injection for delayed unions of the tibia: a preliminary report. J Orthop Trauma. 1989; 3:276-382.
4. Connolly JF, Guse R, Tiedeman J, Dehne R. Autologous marrow injection as a substitute for operative grafting of tibial nonunions. Clin Orthop. 1991; 266:259-270.
5. Tiedeman JJ, Garvin KL, Kile TA, Connolly JF. The role of a composite, demineralized bone matrix and bone marrow in the treatment of osseous defects. Orthopedics. 1995; 18:1153-1158.
6. Gershuni DH, Pinsker R. Bone grafting for nonunion of fractures of the tibia: a critical review. J Trauma. 1982; 22:43-49.
7. Phemister DB. Treatment of un-united fractures by onlay bone grafts without breaking down of the fibrous nonunion. J Bone Joint Surg Am. 1947; 29:946.
8. Brown KL, Cruess RL. Bone and cartilage transplantation in orthopaedic surgery. A review. J Bone Joint Surg Am. 1982; 64:270-279.
9. Burwell RG. Studies in the transplantation of bone. J Bone Joint Surg Br. 1964; 46:110.
10. Chase SW, Herndon CH. The fate of autogenous and homogenous bone grafts. J Bone Joint Surg Am. 1955; 37:809.
11. Urist MR. Practical applications of basic research on bone graft physiology. Instr Course Lect. 1976; 25:1-26.
12. Bucholz RW, Carlton A, Holmes R. Interporous hydroxyapatite as a bone graft substitute in tibial plateau fractures. Clin Orthop. 1989; 240:53-62.
13. Bucholz RW, Carlton A, Holmes RE. Hydroxyapatite and tricalcium phosphate bone graft substitutes. Orthop Clin North Am. 1987; 18:323-334.
14. Burwell RG. The function of bone marrow in the incorporation of a bone graft. Clin Orthop. 1985; 200:125-141.
15. Cornell CN, Lane JM, Chapman M, et al. Multicenter trial of Collagraft as bone graft substitute. J Orthop Trauma. 1991; 5:1-8.
16. Holmes RE, Bucholz RW, Mooney V. Porous hydroxyapatite as a bone graft substitute in diaphyseal defects: a histometric study. J Orthop Res. 1987; 5:114-121.
17. Light M, Kanat IO. The possible use of coralline hydroxyapatite as a bone implant. J Foot Surg. 1991; 30:472-476.
18. Moore DC, Chapman MW, Manske D. The evaluation of a biphasic calcium phosphate ceramic for use in grafting long-bone diaphyseal defects. J Orthop Res. 1987; 5:356-365.
19. Sartoris DJ, Holmes RE, Resnick D. Coralline hydroxyapatite bone graft substitutes: radiographic evaluation. J Foot Surg. 1992; 31:301-313.
20. Connolly JF. Injectable bone marrow preparations to stimulate osteogenic repair. Clin Orthop. 1995; 313:8-18.
21. Healey JH, Zimmerman PA, McDonnell JM, Lane JM. Percutaneous bone marrow grafting of delayed union and nonunion in cancer patients. Clin Orthop. 1990; 256:280-285.
22. Wilkins RM, Stringer EA. Demineralized cortical bone powder: use in grafting space-occupying lesions of bone. Orthop Inter Ed. 1994; 2:71.
23. Goujon E. Recherches experimentales sue les proprieties. J L Anat. 1869; 6:399.
24. Boyne PJ. Autogenous cancellous bone and marrow transplants. Clin Orthop. 1970; 73:199-209.
25. Nade S, Burwell RG. Decalcified bone as a substrate for osteogenesis. An appraisal of the interrelation of bone and marrow in combined grafts. J Bone Joint Surg Br. 1977; 59:189-196.
26. Senn N. On the healing of aseptic bone cavities by implantation of antiseptic decalcified bone. Am J Med Sci. 1889; 98:219.
27. Aspenberg P, Andolf E. Bone induction by fetal and adult human bone matrix in athymic rats. Acta Orthop Scand. 1989; 60:195-199.
28. Bolander ME, Balian G. The use of demineralized bone matrix in the repair of segmental defects. Augmentation with extracted matrix proteins and a comparison with autologous grafts. J Bone Joint Surg Am. 1986; 68:1264-1274.
29. Einhorn TA, Lane JM, Burstein AH, Kopman CR, Vigorita VJ. The healing of segmental bone defects induced by demineralized bone matrix. A radiographic and biomechanical study. J Bone Joint Surg Am. 1984; 66:274-279.
30. Tuli SM, Singh AD. The osteoninductive property of decalcified bone matrix. An experimental study. J Bone Joint Surg Br. 1978; 60:116-123.
31. Urist MR, Dawson E. Intertransverse process fusion with the aid of chemosterilized autolyzed antigen-extracted allogeneic (AAA) bone. Clin Orthop. 1981; 154:97-113.
32. Lindholm TS, Urist MR. A quantitative analysis of new bone formation by induction in compositive grafts of bone marrow and bone matrix. Clin Orthop. 1980; 150:288-300.
33. Green E, Hinton C, Triffitt JT. The effect of decalcified bone matrix on the osteogenic potential of bone marrow. Clin Orthop. 1986; 205:292-298.
34. Tiedeman JJ, Connolly JF, Strates BS, Lippiello L. Treatment of nonunion by percutaneous injection of bone marrow and demineralized bone matrix. An experimental study in dogs. Clin Orthop. 1991; 268:294-302.
35. Arrington ED, Smith WJ, Chambers HG, Bucknell AL, Davino NA. Complications of iliac crest bone graft harvesting. Clin Orthop. 1996; 329:300-309.
36. Challis JH, Lyttle JA, Stuart AE. Strangulated lumbar hernia and volvulus following removal of iliac crest bone graft. Acta Orthop Scand. 1975; 46:230-233.
37. Guha SC, Poole MD. Stress fracture of the iliac bone with subfascial femoral neuropathy: unusual complications at a bone graft donor site: case report. Br J Plast Surg. 1983; 36:305-306.
38. Lane JM, Sandhu HS. Current approaches to experimental bone grafting. Orthop Clin North Am. 1987; 18:213-225.
39. Summers BN, Eisenstein SM. Donor site pain from the ilium. A complication of lumbar spine fusion. J Bone Joint Surg Br. 1989; 71:677-680.
40. Younger EM, Chapman MW. Morbidity at bone graft donor sites. J Orthop Trauma. 1989; 3:192-195.
41. Connolly JF. Common avoidable problems in nonunions. Clin Orthop. 1985; 194:226-235.
42. Buck BE, Malinin TI, Brown MD. Bone transplantation and human immunodeficiency virus. An estimate of risk of acquired immunodeficiency syndrome (AIDS). Clin Orthop. 1989; 240:129-136.

Authors

From *Denver Orthopedic Specialists, PC, and the Institute for Limb Preservation; †the University of Colorado School of Medicine; and ‡Hematology/Oncology Associates, Presbyterian-St. Luke’s Medical Center, Denver, Colo.