In an effort to eliminate the morbidity associated with bone graft harvest from the iliac crest, as well as reduce the incidence of nonunion, the search for the optimal bone graft substitute has intensified.1,2 Various bone graft substitutes currently on the market have been met with limited success; however, an area with considerable promise are the growth factors called bone morphogenetic proteins (BMPs). Bone morphogenetic protein-2 (BMP-2) is an osteoinductive bone growth factor and a member of the transforming growth factor-β superfamily.1,3–5 Of the 14 known types, BMP-2, -6, and -9 appear to be the most biologically potent with the ability to induce differentiation of mesenchymal stem cells into osteoblasts.6 In preclinical studies, recombinant human BMP-2 (rhBMP-2), delivered in a variety of carriers, has been shown to be an effective substitute for autogenous bone in posterolateral lumbar spine fusion, resulting in more rapid and reliable healing than that seen in control groups.1,3–16
Several centers have published their data regarding single-level anterior and single-level posterior fusions performed with rhBMP-2 in humans.17–24 Burkus et al reported a multicenter, prospective, randomized 2-year study of 279 patients who underwent anterior lumbar interbody fusions with 2 tapered threaded fusion cages for lumbar degenerative disc disease.18 The investigational group (8–12 mg at 1.5 mg/mL of rhBMP-2 on an absorbable collagen sponge) had shorter operative time, less intraoperative blood loss, higher fusion rates, and no bone graft harvest issues when compared with the control group (autogenous iliac bone). This single-level anterior study was an FDA-controlled clinical trial, which ultimately led to rhBMP-2 being released by the FDA for single-level use anteriorly with a threaded cage (LT-CAGE).
The data for rhBMP-2 used posteriorly are somewhat less clear. Boden et al20 reported an early prospective, randomized clinical study of 25 patients in which patients in the investigational group received rhBMP-2 (40 mg total with or without internal fixation) and the control group received autogenous bone (all with internal fixation). The fusion rate for the investigational group (rhBMP-2) was 100% (20 of 20 patients) while the control group (autogenous bone) was 40% (2 of 5 patients). There is an ongoing FDA trial for single-level posterolateral fusion using a compression resistant matrix (CRM) carrier combined with biphasic ceramic phosphate (BCP) granules (15% hydroxyapatite, 85% tricalcium phosphate). With this carrier, the concentration being used is 2 mg/mL with a dosage of 40 mg; 20 mg on each side of the fusion. None of those results have yet been published.
There are two reasons for considering the use of rhBMP-2 as a bone graft substitute or extender. One is to reduce or eliminate the need for harvesting extra bone, such as rib or ilium, which can add substantial morbidity to the operative procedure.25–28 The other reason is to address the issue of nonunions or pseudarthroses at levels where achieving a solid fusion can be difficult. Achieving a multilevel fusion in the teenage patient with idiopathic scoliosis is usually relatively “easy,” in contrast to an adult with spinal deformity/scoliosis. In the adult patient, the pseudarthrosis rate is substantial, especially if a long fusion (i.e., T3 to sacrum) is being attempted, particularly in patients over the age of 55 years.29 In addition, achieving an adequate supply of autogenous bone can be difficult, not only because of the length of the fusion but also because the ilium may be needed for fixation, hence leaving adequate amount of bone stock for rigid fixation is paramount. In addition, for posterior spinal fusions there are no good bone extenders. Although demineralized bone matrix substitutes have been suggested and reported in clinical trials and laboratory, there are no conclusive clinical studies verifying efficacy in adult spinal deformity fusions.30,31
To prevent postoperative nonunions/pseudarthroses in long fusions to the sacrum, most spine surgeons recommend an anterior surgical procedure to increase stability and likelihood of solid union. This anterior operation is typically either a T11 to the sacrum fusion done through a thoracoabdominal approach or a paramedian approach for fusion of the lower 2 to 4 lumbar levels. When performing a thoracoabdominal approach, a rib can be harvested for bone grafting; however, a single rib does not usually fill more than two disc spaces in the adult patient. With a paramedian approach, accessing a rib is not straightforward and, therein other sources of bone graft are required (i.e., anterior iliac crest). Use of fresh frozen or freeze dried femoral rings has gained some popularity in the anterior spine but is not universally accepted.
The study purposes and hypotheses are:
- Stand-alone (no added bone) use of rhBMP-2 in the anterior spine (concentration of 1.5 mg/mL and a dosage of 8–12 mg/mL), when combined with titanium mesh cages and protected with posterior instrumentation, would result in >90% rate of radiographic union.
- In posterior spinal fusions, rhBMP-2 (2 mg/mL) combined with BCP (biphasic calcium phosphate granules) with or without local bone graft, but without any autogenous iliac bone graft, would result in >90% rate of radiographic union.
- In rhBMP-2 “compassionate-use” cases for posterior spinal fusion (2 mg/mL, most commonly 40 mg/level impregnated CRM with BCP granules), a radiographic fusion would exist in >90% of patients. In this circumstance, there is no use of local bone graft or harvested bone graft (no iliac or rib harvesting). The “compassionate use” consists of patients who had prior surgeries with harvesting of iliac and rib bone graft such that additional harvesting for bone graft was not feasible. Also, many of these patients carried substantial comorbidities; therefore, a case-by-case application was made to the FDA to use a slightly different product on a multilevel basis. All “compassionate use” cases were individually IRB and FDA approved. A higher dosage and concentration of rhBMP-2 and a different carrier (CRM) were used than in all the other patients in this series.
Materials and Methods
Since December 1999, we have been using rhBMP-2 for adult spinal deformity fusions in off-label applications. To date, the two senior surgeons have used the product either in the anterior and/or posterior spine in 241 adults, through December 1, 2004. The product has been used anteriorly in 61 patients, posteriorly in 136 patients, and both anterior and posterior in 44 patients. Anterior spine fusions (n = 105) were typically done through paramedian approach at one, two, three, or four levels along with the use of titanium mesh cages at each operative level. In this circumstance, the rhBMP-2 (InFUSE) was used at a concentration of 1.5 mg/mL and a total of 8 to 12 mg/level. Posterior applications (n = 180) have largely fallen into in three separate categories. The first application is a one- or two-level fusion with posterior instrumentation in which rhBMP-2 (2 mg/mL) was used along with the collagen sponge and BCP granules and locally harvested bone (no autogenous iliac crest or allograft bone graft).2 The second circumstance has been that of a longer fusion in which the rhBMP-2 product is being used to supplement or to extend the bone graft. In this circumstance, a combination of local bone graft, iliac bone graft, or fresh frozen femoral head is being used along with the rhBMP-2. A “kit” of rhBMP-2/InFUSE may contain anywhere from 4 to 12 mg. A small kit is 4.2 mg, medium kit is 8.4 mg, and a large kit is 12 mg. The third circumstance is that of a posterior spinal fusion labeled as a “compassionate case.” This is a situation in which an individual case application has been made to the FDA to use the rhBMP-2 for multiple levels posteriorly in which it is judged that there is not a reasonable opportunity to harvest iliac or rib bone graft. In these “compassionate-use” cases, the rhBMP-2 concentration is 2 mg/mL with 40 mg typically used at each level (20 mg on each side); the carrier is a CRM sponge and BCP granules. Detailed breakdown of the 241 cases performed through December 2004 at our institution by procedure type are anterior 105 patients, posterior with rhBMP-2 only 15 patients, posterior with rhBMP-2 and local bone 32 patients, posterior with rhBMP-2 and local bone and iliac bone 68 patients, and posterior with rhBMP-2 and local bone, iliac bone, and fresh frozen femoral head in 65 patients (285 total patient samples in 241 patients, 44 patients used anteriorly and posteriorly). The rest of this manuscript will focus on subgroups of patients who have a minimum 12-month follow-up. This study is focusing on three groups of patients, as mentioned in the study purpose/hypothesis. Group 1 are the patients that had anterior application of rhBMP-2 (Case 1; Figure 1A, B). Group 2 are the posterior cases in which rhBMP-2 was used either as a stand-alone or with local bone graft only (Case 1; Figure 1A–E). Group 3 are the compassionate-use posterior cases (Case 2; Figure 2A, B). For the purpose of this analysis, patients were excluded from analysis if autogenous rib bone or iliac bone was used in combination with the rhBMP-2; only locally harvested bone from the posterior elements was used for Group 2 patients. No local bone graft was used for Group 3 patients. Minimum follow-up was 12 months after surgery. A total of 70 patients were studied. There were a total of 95 patient samples as 25 of the patients were done circumferentially with rhBMP-2. All of the anterior patient samples/patients studied had a posterior instrumented fusion.
Routine postoperative follow-up at yearly intervals included clinical and radiographic evaluation. Plain anteroposterior and lateral standing radiographs were obtained for all patients over the length of the fusion. In some instances, oblique radiographs were additionally used. Spiral CT scans (with sagittal and coronal reformatting) were obtained as clinically indicated to better visualize the fusion masses if there was uncertainty on the plain radiographs.32 The assessment of fusion mass was based on a combination of plain radiographs and CT scans. Group 1 (21 patients, 46%) had spiral CT scans, Group 2 (15 patients, 37%) had spiral CT scans, and in Group 3 (2 patients, 25%) had spiral CT scans. The grading systems used have been previously reported by Eck et al33 (Table 1) for the anterior fusions and Lenke et al (Table 2) for the posterior fusions.34 Both grading systems (Eck et al and Lenke et al) use a grading system from 1 to 4 based on fusion mass appearance and implant subsidence/failure. For the purposes of this analysis, Grades 1 and 2 were considered “fused” at the level of surgery and Grades 3 and 4 were considered “not fused.” Two surgeons not involved in the operative procedure analyzed and graded all plain radiographs and CT scans. The mean score of the two reviewers was calculated and used for the analyses. A mean score from 1 to 2.4 was classified as “fused” and a score from 2.5 to 4 was classified as “not fused.” All data were analyzed using Statistical Analysis System (SAS, version 8.02) to manage patient data and to calculate descriptive statistics, including various percentages. Radiographic evaluation included fusion grading and assessment of postoperative pseudarthrosis, such as loss of correction or implant failure as observed to be either screw breakage, hook pullout, or rod breakage.
Routine patient demographic data such as age, gender, history of previous spine surgery and type, presence of pseudarthroses at operative site were analyzed. Postoperative complications such as wound infections (deep and superficial), wound hematomas were also recorded. Seventy patients satisfied the criteria for inclusion. There were 56 females and 14 males whose mean age at the time of surgery was 55.4 years (21–80 years). Mean follow-up was 17.9 months after surgery (12–60 months). Breakdown by surgical procedure for the 70 patients was as follows: 21 patients had anterior spine fusion with rhBMP-2 (protected with posterior fusion/posterior segmental spinal instrumentation, but no rhBMP-2 used posteriorly), 16 Group 2 patients had posterior fusion only (no anterior spinal fusion performed), 25 had anterior and posterior fusion with rhBMP-2 on the same day or in a staged fashion, and 8 Group 3 patients were categorized as “compassionate-use” posterior fusion cases (no anterior spinal fusion performed). Overall, 43 patients (61%) underwent previous spine surgery. The patient’s main diagnosis for the three groups and further demographic data on these groups are listed on Table 3.
Operative data for the three groups is listed on Table 4. Mean levels fused were 2.3 for Group 1, 2.9 levels for Group 2, and 6.5 levels for Group 3. Mean rhBMP-2 used for each group was as follows: Group 1, 10.8 mg; Group 2, 13.7 mg; and Group 3, 28.6 mg. Of the 43 patients who underwent previous spine surgery, a total of 23 (53%) were documented with postsurgical pseudarthroses, which were detected on preoperative imaging studies or at the time of surgery.
In Group 1 (anterior cases), a total of 93 intervertebral levels underwent attempted fusion (Table 5). At the time of review of imaging studies at all levels, there was evidence of new bone formation within the intervertebral space. Using the grading system of Eck et al,33 90 levels (96%) were graded as fused and only 4 levels deemed “not fused” (all with fusion grades of 2.5). Breakdown of anterior fusion by intervertebral level is outlined in Table 5. Statistical analyses failed to indicate any association between fusion grades and level fused, gender, age, amount of rhBMP-2 used, or the presence of a pseudarthrosis.
In Group 2 (posterior cases), a total of 118 intervertebral levels underwent attempted fusion (Table 6). New fusion mass bone could be identified at all levels involved in the posterior fusion. Posterior grading of fusion mass was more difficult due the presence of the radio-opaque BCP granules. Using the grading system of Lenke et al,34 110 levels (93%) were graded as “fused” and only 8 levels were graded as “not fused” (all with fusion grades of 2.5). Breakdown of the posterior fusion group by intervertebral level is outlined in Table 6. Statistical analyses failed to identify any association between fusion grades and age, gender, level fused, amount of rhBMP-2, or the presence of pseudarthrosis. In addition, 25 of the 41 patients underwent simultaneous or staged anterior fusion within the region of the posterior fusion. Theoretically, the presence of an anterior fusion should increase the posterior fusion grade for that respective level. However, because of the high posterior fusion rate, no statistically significant difference could be detected between the fusion grades for the posterior-only levels and those levels having an anterior and posterior fusion.
In Group 3 (posterior “compassionate-use” cases), the 8 patients had a total of 52 intervertebral levels undergo fusion. Seven of these patients had previous spinal fusion surgery with 6 of them having pseudarthroses before surgery. Using the grading system of Lenke et al,34 52 levels (100%) were graded as “fused.” No statistical analyses were performed on the “compassionate-use” cases because of small cohort size. To date, there has been no evidence of implant failure in any of the 49 patients in the posterior fusion Groups 2 and 3.
Overall, three postoperative complications were documented in the 70 patients that may be associated with the use of rhBMP-2; 1 in Group 1 (anterior), 2 in Group 2 (posterior), and none in Group 3 (posterior “compassionate-use” cases). The anterior complication was a superficial wound dehiscence, which was allowed to granulate closed without long-term problems. The two complications in Group 2 were: 1 deep wound hematoma and 1 deep wound infection. The deep wound hematoma occurred early in the rhBMP-2 experience at our medical center. Because of concerns of evacuating the rhBMP-2 from the fusion bed via deep subfascial drains, a decision was made to not use deep drains at the fusion bed. A sterile deep wound hematoma developed in the third patient in which rhBMP-2 was used, necessitating a single operative drainage procedure in the operating room. The deep wound infection required two irrigation and debridement procedures and intravenous antibiotics for complete resolution of the deep space infection. No long-term sequelae have resulted from these three documented complications. Despite the concerns with rhBMP-2 inducing an exaggerated inflammatory response, specifically in the postoperative period after anterior cervical fusion surgery, no patient in this series had wound or systemic complications that could be attributable to the use of rhBMP-2 (Riew KD, personal communication).
Plain radiographs are the mainstay of postoperative fusion assessment for the spine surgeon; however, in a significant number of patients, the presence of metallic spinal implants and the complex 3-dimensional anatomy of the spine precludes an adequate radiographic assessment. In this situation, spiral CT scans, with sagittal and coronal reconstructions, can permit better visualization of the fusion mass. Shah et al reported on 156 lumbar interbody fusions using titanium interbody cages, filled with autogenous bone only, in 153 patients.32 Plain radiographs could identify bony bridging trabeculation inside the cages in only 4% and outside of the cage in only 8% of patients. High-quality CT scans (thin-slice, 1–3 mm) permitted better visualization of bridging trabeculation inside the cages in 95% and outside of the cages in 90% of patients. Overall, kappa values for interobserver reliability was similar for CT scans and plain radiographs; kappas of 0.85 and 0.82 and kappas 0.74 and 0.86, respectively. They concluded high-quality CT scans were a superior imaging modality for the evaluation of anterior interbody fusions with titanium cages in the lumbar spine. As documented in Table 7, the use of spiral CT scans improved the fusion grades in this study, on average, for both anterior and posterior fusions. The mean fusion grades improved from 1.4 to 1.2 for Group 1 (anterior) and from 1.6 to 1.4 for Group 2 (posterior).
Intuitively, one would think that assessing an anterior fusion would be somewhat easier than a posterior fusion. With a posterior fusion the posterior implants, and possibly anterior implants, and the BCP granules interfere with seeing the “true fusion.” Anteriorly, the titanium mesh cages pose somewhat of an obstacle, as previously documented in past studies by Eck et al.35 On plain radiographs, one of the main findings we try to identify is new ossification within the intervertebral space, but outside of the titanium mesh cage. Although the spiral CT assessment does improve our ability to “see” the fusion, we have some concern that the spiral CT scans may overstate the extent of the fusion.
Even with the use of both plain radiographs and spiral CT scans, fusion assessment is less than ideal. Current fusion assessment classification systems are only based on plain radiographs and have no biomechanical correlation to the adequacy of fusion strength.33,34 As such, the interobserver and intraobserver reliability of these systems is likely significant. Of a total of 263 fusion levels in this study (170 posterior and 93 anterior), only 12 levels were graded as “not fused.” Interestingly, all 12 levels had fusion scores of 2.5, which based on the methodology of this study puts them into the “not fused” category. Secondary review of the imaging studies for these 12 levels, after the data analysis, highlighted that most of these fusions had less than adequate imaging studies making fusion assessment less than optimal. Of the 526 fusion level assessments (263 for each of the two reviewers), no level received a fusion grade of 4, which is an obvious pseudarthrosis. Nonetheless, we would agree that a “solid fusion” cannot be guaranteed in an adult patient until 5-year follow-up is complete.
Despite the shortcomings of fusion assessment, the results anteriorly have been encouraging and persuasive enough that whenever we do multiple level anterior fusions our standard now is to use 8 to 12 mg of rhBMP-2, 1.5 mg/mL, and the titanium cage. Higher dosages of rhBMP-2 do not appear to be necessary to attain a fusion anteriorly. We have stopped harvesting either rib or anterior iliac bone for anterior fusions. One important point to note is that all of the anterior fusion levels in this study were part of a circumferential fusion; hence posterior spinal implants were present at all levels.
Unlike the anterior fusions, it is very difficult to ascertain what the “ideal” dosage and concentration is when rhBMP-2 is used posteriorly. The rhBMP-2 carrier is not entirely straightforward but should be a form of collagen sponge with a volume and bulk along the lines of BCP granules (15% hydroxyapatite, 85% tricalcium phosphate, Mastergraft). Because of the high rate of posterior fusion (95%), we could not identify a minimum dosage necessary to achieve a solid fusion. Definitive comment about the dosage and concentration, therefore, cannot be made from this study, but it appears that the product is definitely useful in dosages that approximate 40 mg/level with a concentration between 2.0 and 2.5 mg/mL, as demonstrated in the 100% fusion rate of the 8 “compassionate-use” cases. Our impression has been that the product is not as useful in lower dosages and concentrations. In addition, of the 41 patients in Group 2, 25 of them had concomitant anterior fusions either performed the same day or in a staged fashion. We could not identify any improvement in the fusion mass grade at those levels with anterior surgery, but we would anticipate that, with greater patient numbers and longer follow-up, this might become evident.
The high rate of apparent spine fusion at minimum 12-month follow-up in this study, in both the anterior and posterior spine, supports the use of rhBMP-2 in adults with spinal deformity in primary and revision surgery. No complications, local or systemic, could be definitively attributable to the use of rhBMP-2. Further investigation is necessary to identify the optimal rhBMP-2 dosages and concentration, the most appropriate rhBMP-2 carrier, the best adjunctive bone graft or bone “extender,” and the host factors that could impact the effectiveness of the rhBMP-2 at the time of fusion.
- Fusion assessments were performed on a consecutive series of adult spinal deformity patients in whom human recombinant bone morphogenetic protein 2 (rhBMP-2) was used.
- Patient groups were: 1) anterior with titanium mesh cages (no bone graft), 2) posteriorly with biphasic ceramic phosphate granules (15% hydroxyapatite, 85% tricalcium phosphate), with or without local bone graft, but no iliac or other harvested bone graft, and 3) compassionate use rhBMP-2 with a compression resistant matrix carrier, with no bone graft.
- Plain radiographs were used for fusion assessment in all patients. Spiral CT scans were additionally obtained to better assess the fusion mass as needed. A previously published standardized grading system was used.
- Fusion rates were high (93%–100%). There were no substantial complications.
1.Boden SF, Schimandle JH. Biologic enhancement of spinal fusion. Spine
2.Akamaru T, Suh D, Boden SD, et al. Simple carrier matrix modifications can enhance delivery of recombinant human bone morphogenetic protein-2 for posterolateral spine fusion. Spine
3.Aspenberg P, Turek T. BMP-2 for intramuscular bone induction: effect in squirrel monkeys is dependent on implantation site. Acta Orthop Scand
4.Urist MR. Bone formation by autoinduction. Science
5.Urist MR, Strates BS. Bone morphogenetic protein. J Dent Res Suppl
6.Cheng H, Jiang W, Phillips FM, et al. Osteogenic activity of the fourteen types of human bone morphogenetic proteins (BMP). J Bone Joint Surg Am
7.Wozney JM. Overview of bone morphogenetic protein. Spine
8.Boden SD, Moskovitz PA, Morone MA, et al. Video-assisted lateral inter-transverse process arthrodesis: validation of a new minimally invasive lumbar spinal fusion technique in the rabbit and nonhuman primate (rhesus) models. Spine
9.David SM, Murakami T, Tabor OB, et al. Lumbar spinal fusion using recombinant human bone morphogenetic protein-2 (rhBMP-2): a randomized, blinded and controlled study. Transactions of the International Society for Study of the Lumbar Spine
10.Fischgrund JS, James SB, Chabot MC, et al. Augmentation of autograft using rhBMP-2 and different carrier media in the canine spine fusion model. J Spinal Disord
11.Hollinger EH, Trawick RH, Boden SD, et al. Morphology of the lumbar inter-transverse process fusion mass in the rabbit model: a comparison between two bone graft materialshBMP-2 and autograft. J Spinal Disord
12.Martin GJ, Boden SD, Morone MA, et al. Posterolateral intertransverse process spinal fusion arthrodesis with rhBMP-2 in a non-human primate: important lessons learned regarding dose, carrier, and safety. J Spinal Disord
13.Sandhu HS, Kanim LEA, Kabo JM, et al. Effective doses of recombinant bone morphogenetic protein-2 in experimental spinal fusion. Spine
14.Sandhu HS, Kanim LE, Kabo JM, et al. Evaluation of rhBMP-2 with an OPLA carrier in a canine posterolateral (transverse process) spinal fusion model. Spine
15.Schimandle JH, Boden SD, Hutton WC. Experimental spinal fusion with recombinant human bone morphogenetic protein-2 (rhBMP-2). Spine
16.Damien CJ, Grob D, Boden SD, et al. Purified bovine BMP extract and collagen for spine arthrodesis: preclinical safety and efficacy. Spine
17.Boden SD, Zdeblick TA, Sandhu HS, et al. The use of rhBMP-2 in interbody fusion cages. Spine
18.Burkus JK, Transfeldt EE, Kitchel SH, et al. Clinical and radiographic outcomes of anterior lumbar interbody fusion using recombinant human bone morphogenetic protein-2. Spine
19.Boden SD, Schimandle JH, Hutton WC. Volvo Award Winner in Basic Sciences. The use of an osteoinductive growth factor for lumbar spinal fusion: II. Study of dose, carrier, and species. Spine
20.Boden SD, Kang J, Sandhu H, et al. Use of recombinant human bone morphogenetic protein-2 to achieve posterolateral lumbar spine fusion in humans. Spine
21.Boden SD, Zdeblick TA, Sandhu HS, et al. The use of rhBMP-2 in interbody fusion cages: definite evidence of osteoinduction in humans: a preclinical report. Spine
22.Burkus JK, Gornet MF, Dickman CA, et al. Anterior lumbar interbody fusion using rhBMP-2 with tapered interbody cages. J Spinal Disord
23.Burkus JK, Heim SE, Gornet MF, et al. Is INFUSE bone graft superior to autograft bone? An integrated analysis of clinical trials using the LT-CAGE lumbar tapered fusion device. J Spinal Disord
24.Burkus JK, Dorchak JD, Sanders DL. Radiographic assessment of interbody fusion using recombinant human bone morphogenetic protein type 2. Spine
25.Arrington ED, Smith WJ, Chambers HG, et al. Complications of iliac crest bone graft harvesting. Clin Orthop
26.Banwart JC, Asher MA, Hassanein RS. Iliac crest bone graft harvest donor site morbidity: a statistical evaluation. Spine
27.Fernyhough JC, Schimandle JH, Weigel MC, et al. Chronic donor site pain complicating bone graft harvesting from the posterior iliac crest for spinal fusion. Spine
28.Boden SD. Overview of the biology of lumbar spine fusion and principles for selecting a bone graft substitute. Spine
29.Kim YJ, Bridwell KH, Lenke LG, et al. Pseudarthrosis in long adult spinal deformity instrumentation and fusions: risk factor and clinical outcome analysis of 232 cases. Spine
30.Peterson B, Whang PG, Iglesiasa R, et al. Osteoinductivity of commercially available demineralized bone matrix: preparations in a spine fusion model. J Bone Joint Surg Am
31.Cammisa FP, Lowery G, Garfin SR, et al. Two-year fusion rate equivalency between Grafton DBM gel and autograft in posterolateral spine fusion. Spine
32.Shah RR, Mohammed S, Saifuddin A, et al. Comparison of plain radiographs with CT scan to evaluate interbody fusion following the use of titanium interbody cages and transpedicular instrumentation. Eur Spine J
33.Eck KR, Lenke LG, Bridwell KH, et al. Radiographic assessment of anterior titanium mesh cages. J Spinal Disord
34.Lenke LG, Bridwell KH, Bullis D, et al. Results of in situ fusion for isthmic spondylolisthesis. J Spinal Disord
35.Eck KR, Bridwell KH, Ungacta FF, et al. Analysis of titanium mesh cages in adults with minimum two-year follow-up. Spine