- Email: [email protected]
Improved bone formation in smokers with recombinant human bone morphogenetic protein–2
Proceedings of the NASS 16th Annual Meeting / The Spine Journal 2 (2002) 3S–44S the Gold-Coat tubing, accelerating the DNA-coated gold particles on the inside of the tubing cartridge to penetrate the target cells. The gene gun was positioned at a minimal distance from the petri dish and tissue, and a single bombardment was carried out. After 3 days, the transfection efficiency of a Lac reporter gene construct (pCMV-b-galactosidase) in the primary monolayer cultures of normal bovine NP and AF cells was assessed using an In Situ b-galactosidase staining kit. The DNA content and the total PG content were measured in the cell layer to assess metabolic activity. PG synthesis was also measured using [35S]-sulfate labeling, followed by rapid filtration, and was compared between the OP-1–transfected (pW24) and the control (vacant vector) groups. Statistical analyses were performed by one-way analysis of variance with Fisher’s PLSD test as a post hoc test. Results: The gene transfer of b-galactosidase was performed to probe the efficiency of transfection in two different cell sources. Analysis of X-gal staining demonstrated an efficiency of 14.2% in normal NP cells and 8.2% in AF cells. The DNA content and rate of PG synthesis in the three cell types did not differ significantly when the pCMV-b-gal–transfected and nontreated groups were compared. This suggested that the gene gun procedure does not have a significant adverse effect on cell metabolism. To study whether gene transfection can alter the metabolism of IVD cells, the human OP-1 gene was transfected using a pW24 vector. On day 3 after transfection, there were no significant differences in the DNA content. A small increase in the PG accumulation was observed in the OP-1-groups. However, there were no significant differences in the PG content of any cell group. On the other hand, in the OP-1–transfected group, the rate of PG synthesis was significantly higher in all cell types (AF [124%]; p.05 and NP [144%] cells [p.01]). NP cells were more responsive than AF cells to the transfection of the OP-1 gene. Discussion: The results of this study reveal, for the first time, that transfection of the OP-1 gene by a gene gun system to both AF and NP cells in vitro can alter the metabolism of these cells. Both the efficiency and the metabolic studies provide evidence that the NP cells may be the best target for transfection. Although it remains to be proven that OP-1 production was enhanced after gene transfer, this study suggests that gene therapy with the OP-1 expression vector can be a useful method for inducing the regeneration of IVD and articular cartilage tissues. Additional studies are now ongoing to determine if transfection of the OP-1 gene into the IVD and articular cartilage tissues can be achieved and if it can influence the metabolism of those tissues. The effect of osteogenic protein-1 instrumented and noninstrumented lumbar fusion in rabbits Louis G. Jenis, MD, Boston, MA, USA; Donna Wheeler, PhD, Fort Collins, CO, USA; Steven Parazin, MD, Raymond J. Connolly, PhD, Hubert Wolfe, MD, Boston, MA, USA Introduction: The goal of lumbar spinal fusion is to attain solid arthrodesis. Nonunion remains a significant challenge and may lead to added morbidity and requirements for additional surgeries. The goal of the study is to evaluate the effect of osteogenic protein-1 (OP-1: also known as bone morphogenetic protein–7 [BMP-7]) in a rabbit instrumented and noninstrumented lumbar fusion model as a bone graft substitute. The current study was designed to determine whether the effect of OP-1 could be enhanced in a more rigid environment. Methods: The outcome of intertransverse process fusion with OP-1 versus iliac crest autograft in the presence and absence of rigid fixation was studied. Animal care research committee approval was obtained. Forty-eight adult male New Zealand white rabbits were divided into four groups of 12 animals: 1) control animals: in situ posterolateral arthrodesis L5–L6 using autogenous bone graft; 2) fixation group: posterolateral arthrodesis L5–L6 with autogenous bone graft and interspinous wiring supplemented with polymethylmethacrylate (PMMA) to simulate rigid fixation; 3) OP-1 group: undergo in situ posterolateral arthrodesis L5–L6 using OP-1 and 4) combined OP-1 and rigid fixation. Animals were sacrificed at 3 and 12 weeks after surgery and lumbar spines harvested. Each specimen was evaluated with computed tomographic axial, including reconstructed sagittal and coronal, images to assess fusion. Computed tomographic data were analyzed for
17S
four specimens within each treatment group and at each time interval. Fusion was classified by grade: 0) pseudarthrosis; 1) traversing bone lacking bridging and 2) solid fusion. In addition, decalcified histologic analysis of the intertransverse process region was performed with emphasis on cellular activity and bone formation. Gene expression within the fusion mass was performed by RT-polymerase chain reaction techniques. Type 1 collagen, osteocalcin, osteopontin and alkaline phosphatase activity was evaluated. Results: At 3 weeks, autograft within fixation specimens showed 0% grade 2, 100% grade 1 and 0% grade 0 fusion by CT imaging. OP-1– treated animals at 3 weeks showed 0% grade 2, 50% grade 1 and 50% grade 0 fusion. At this early time interval radiographic bone formation was noted in half of the rabbits, although complete arthrodesis was not evident. At 3 months, autograft animals showed 0% grade 2, 75% grade 1 and 25% grade 0 fusion. OP-1 animals showed 100% grade 2 solid fusions by this late time interval. Autograft and OP-1 treated animals with fixation at 3 weeks revealed identical 25% grade 2, 75% grade 1 and 0% grade 0 fusion. Three–month specimens of autograft-fixation exhibited 50% grade 2 and 50% grade 1 fusion, whereas OP-1 fixation achieved 100% grade 2 fusion. Histologic and gene expression results will also be presented. Discussion: This study demonstrates that OP-1 is an effective osteoinductive agent in lumbar intertransverse process fusion at both early and late time intervals. Spinal rigid fixation may potentiate the effect of OP-1 as evidenced by one of four animals showing arthrodesis by 3 weeks. OP-1 is effective as a bone graft substitute, as evidenced by radiographic, histologic and molecular data. Improved bone formation in smokers with recombinant human bone morphogenetic protein–2 Harvinder S. Sandhu, MD, New York, NY, USA; Rick B. Delamarter, MD, Linda E.A. Kanim, MA, Santa Monica, CA, USA; Frank P. Cammisa, Jr., MD, New York, NY, USA; Aleksandar Curcin, MD, Baltimore, MD, USA; Edward N. Hanley, Jr., MD, Charlotte, NC, USA; Jeffrey S. Fischgrund, MD, Southfield, MI, USA; Alexandre Valentin-Opran, MD, Cambridge, MA, USA; Thomas Lang, MD, San Francisco, CA, USA Introduction: Smoking clearly inhibits bone fusion in spinal surgery patients. Recombinant human bone morphogenetic protein (rhBMP)-2 has been shown to increase bone formation. Can rhBMP-2 reverse the deleterious effects of smoking? Purpose: To assess bone formation after implantation of varying concentrations of rhBMP-2 in attempted reconstruction of the iliac autogenous bone donor site in patients undergoing cervical fusion with and without a reported history of smoking. Methods: A four-center, patient-blinded, randomized controlled study comparing reconstruction of the iliac crest donor site with three doses of rhBMP-2/ Absorbable Collagen Sponge (ACS) to buffer/ACS or standard methods of closure (iliac defect left empty) was conducted. Forty patients undergoing a one- or two-level cervical fusion procedure requiring autogenous iliac crest bone grafting were informed of the study, enrolled and randomized before surgery (between 1998 and 1999). The iliac crest harvest technique was standardized. Donor site was left empty (n8), reconstructed with buffer/ACS (n8) or with rhBMP-2/ACS 0.43 mg/ml (n8), 0.75 mg/ml (n8) or 1.50 mg/ml (n8). At the time of enrollment, patients reported recent smoking history (ever and current smokers). Osteoinduction was monitored at baseline and 24 weeks on computed tomography (CT) and three-dimensional reconstructions. “Bridging bone” was considered as none (0), small to moderate (1) or extensive (2) filling in of the defect. Multivariate analysis techniques were used. Results: Follow-up evaluations were completed by all patients (CT available on 39 patients). Mean age was 42 years (range, 26 to 33 years); 65% of patients were men and 35% women. Thirteen patients reported a history of smoking, and 26 patients had no recent smoking history. Rates of smoking were comparable among all treatment groups. Bridging bone occurred more frequently in patients without a smoking history than for those with a history across all treatments except for 1.5 mg/ml rhBMP-2/ACS (highest dose). After 1.5 mg/dl rhBMP-2 administration, there was no difference in osteoinduction, bridging bone in the defect site, between patients with and without a history of smoking (overall model logistic, p.05).
18S
Proceedings of the NASS 16th Annual Meeting / The Spine Journal 2 (2002) 3S–44S
Conclusion: Although bone formation was less for patients with a recent history of smoking at lower doses of rhBMP-2, these differences were not observed at the dose of 1.5 mg/ml rhBMP-2. Treatment with the highest concentration of rhBMP-2 may overcome the nicotine effect on osteoinduction. Despite these very early and preliminary results, smoking cessation before a fusion procedure is crucial. IGF-I and TGF-1 application by a poly-(D,L-lactide)–coated cage promotes fusion in the sheep cervical spine Frank Kandziora, MD, N.P. Hass, Gerhard Schmidmaier, Georg Schollmeier, Herman Bail, Robert Pflugmacher, MD, Thomas Gorke, Martin Wagner, Thomas Mittlmeier, PhD, Michael Raschke, Berlin, Germany Study design: Growth factors such as bone morphogenetic protein (BMP)-2 have proven to promote spine fusion and to overcome the disadvantages of an autologous bone graft. The optimum growth factor to promote spinal fusion, as well as the optimum method to deliver such growth factors, is still a matter of discussion. A sheep cervical spine interbody fusion model was used to determine the effect of IGF-I and TGF-1 application by a biodegradable poly-(D,L-lactide) (PDLLA)–coated cage. The purpose of this study was to determine the safety and efficacy of a new PDLLA carrier system and to evaluate the safety and efficacy of combined IGF-I and TGF-1 application in a sheep cervical spine model. Methods: Thirty-two sheep underwent C3–4 discectomy and fusion: Group 1: autologous tricortical iliac crest bone graft (n8); Group 2: titanium cage (n8); Group 3: titanium cage coated with a PDLLA carrier (n8); Group 4: titanium cage coated with a PDLLA carrier including IGF-I (5% w/w) and TGF-1 (1% w/w) (n8). Blood samples, body weight and temperature were analyzed. Radiographic scans were performed pre- and postoperatively and after 1, 2, 4, 8 and 12 weeks, respectively. At the same time points, disc space height, intervertebral angle and lordosis angle were measured. After 12 weeks, animals were killed and fusion sites were evaluated using functional radiographic views in flexion and extension. Quantitative computed tomographic scans were performed to assess bone mineral density, bone mineral content and bony callus volume. Biomechanical testing was performed in flexion, extension, axial rotation and lateral bending. Stiffness, ranges of motion, neutral and elastic zone were determined. Histomorphological and histomorphometrical analysis was performed, and polychrome sequential labeling was used to determine the time frame of new bone formation. Results: There were no differences between the groups concerning blood counts, body weight and temperature. Over a 12-week period, the cage Groups 2, 3 and 4 showed significantly higher values for intervertebral angle compared with the bone graft. Functional radiographic assessment revealed significantly lower residual flexion/extension movement in Group 4 than in any other group. The PDLLA-coated cages with IGF-I and TGF-1 showed significantly highest values for bone mineral density, bone mineral content and bony callus volume. Average stiffness in rotation and bending was significantly higher, and range of motion, neutral and elastic zone in rotation were significantly lower in Group 4 than in any other group. Histomorphometrical evaluation showed a more progressed bone matrix formation in PDLLAcoated cages with IGF-I and TGF-1 than in any other group. Polychrome sequential labeling showed accelerated intervertebral fusion in Group 4. Conclusion: PDLLA coating of cervical spine interbody fusion cages as a delivery system for growth factors was effective and safe. The PDLLA coating showed no adverse effects. IGF-I and TGF-1 application by a PDLLA-coated interbody cage significantly increased results of interbody fusion in a sheep cervical spine model without adverse side effects. In this new combination (implant plus PDLLA plus growth factors) the cage represents a “real fusion” cage, because it does not only serve as a mechanical device for spine fixation but also as a local drug delivery system. A retrospective review of early outcomes of balloon kyphoplasty Steven R. Garfin, MD, San Diego, CA, USA Background context: An estimated 275,000 vertebral body compression fractures (VCFs) secondary to osteopenia, and refractory to medical man-
agement, occur in the United States each year. The deformity caused by VCFs is associated with chronic pain, impaired physical functioning, reduced quality of life, restrictive airway disease and increased mortality. This is a retrospective chart review of outcomes of inflatable bone tamp reduction and internal fixation using polymethylmethacrylate (PMMA). Methods: Vertebral body height was measured from X-rays from all patients treated by the authors (n451). The point of greatest height loss was measured, as was that at a comparable location in the closest normal vertebral body. The height of the fractured vertebral body was divided by the height of the normal vertebral body to calculate the percent estimated height before treatment. After treatment, the same location was measured in the treated and nearby vertebral body, to calculate the estimated percent height after treatment. The results were stratified by duration of symptoms at diagnosis, before or after 3 months. The percent lost height restored was determined by: (percent estimated height before fracture minus percent estimated height after fracture)/(1 percent estimate height after treatment). Reviewing the use of pain medication assessed reduction in pain. Institutional variations in these determinations were compared with the other centers performing kyphoplasty. The potential for inducing adjacent level fracture was assayed by determining the number of adjacent level fractures before treatment and then the number of new fractures that were adjacent to previously treated fractures. Results: A total of 2,194 fractures were treated (T43, T525, T648, T7106, T8114, T9117, T10102, T11174, T12333, L1365, L2249, L3241, L4183, L585, L61) in 1,439 patients in the United States from September 1998 through November 2000. Seventy-five percent were women. The average age was 73.5 years (range, 19 to 99 years). In fractures where the duration of symptoms at the initial visit was under 3 months, the average fractured vertebral body height was 71% before treatment and 92% after treatment (p.002). In fractures over 3 months, the average fractured vertebral body height was 74% before treatment and 84% after treatment (p.05). The average lost height restored for all fractures was 57%. Ninety percent of patients returned to prefracture pain medication (most often nonsteroidal anti-inflammatory drugs for arthritic type complaints) levels by 2 weeks. Major complications included three postoperative thoracic-level parapareses related to instrument insertion through the medial wall seventh pedicle with cord injuries. One patient had kyphoplasty performed while on anticoagulants and developed paraparesis and an epidural hematoma, which recovered completely after surgical decompression. There was one cement embolus without breathing consequences and a postoperative ileus caused by retroperitoneal bleeding in a patient on warfarin sodium (Coumadin; DuPont Pharmaceuticals Company, Wilmington, DE, USA). The serious adverse event rate is 0.2% per fracture and 0.4% per patient. Unrelated serious adverse events perioperatively included two myocardial infarctions related to overhydration, during the procedure, and two non–procedure-related perioperative deaths (cancer and coronary artery disease related). In patients with more than one fracture before their first kyphoplasty procedures, 63% of fractures not yet treated were adjacent. In 46 patients who returned for additional treatment of 139 fractures, 30% of the new fractures were adjacent to previously treated fractures. The results were nearly identical at each center, independent of physician. Conclusion: Kyphoplasty is a safe and effective treatment option for patients with painful VCF secondary to osteopenia. Height restoration was most improved if treatment occurred before 3 months from fracture, but some height restoration was obtained after 3 months. Comparing adjacent fractures in the untreated patients with the 2% of treated patients returning for additional kyphoplasty procedures does not indicate that kyphoplasty increases the rate of adjacent fracture. There was a remarkable consistency of treatment outcomes in the multicenter study. The four neurologic complications occurred in the first 100 fractures. Technique adjustments were made, and none have occurred since. (PMMA is approved by the Food and Drug Administration for long bones and joints. It is off-label for use in the spine.)
Kyphoplasty in the treatment of vertebral fractures secondary to multiple myeloma Sean Dudeney, MD, Mohamed Hussein, MD, Isador H. Lieberman, MD, Cleveland, OH, USA
17S
four specimens within each treatment group and at each time interval. Fusion was classified by grade: 0) pseudarthrosis; 1) traversing bone lacking bridging and 2) solid fusion. In addition, decalcified histologic analysis of the intertransverse process region was performed with emphasis on cellular activity and bone formation. Gene expression within the fusion mass was performed by RT-polymerase chain reaction techniques. Type 1 collagen, osteocalcin, osteopontin and alkaline phosphatase activity was evaluated. Results: At 3 weeks, autograft within fixation specimens showed 0% grade 2, 100% grade 1 and 0% grade 0 fusion by CT imaging. OP-1– treated animals at 3 weeks showed 0% grade 2, 50% grade 1 and 50% grade 0 fusion. At this early time interval radiographic bone formation was noted in half of the rabbits, although complete arthrodesis was not evident. At 3 months, autograft animals showed 0% grade 2, 75% grade 1 and 25% grade 0 fusion. OP-1 animals showed 100% grade 2 solid fusions by this late time interval. Autograft and OP-1 treated animals with fixation at 3 weeks revealed identical 25% grade 2, 75% grade 1 and 0% grade 0 fusion. Three–month specimens of autograft-fixation exhibited 50% grade 2 and 50% grade 1 fusion, whereas OP-1 fixation achieved 100% grade 2 fusion. Histologic and gene expression results will also be presented. Discussion: This study demonstrates that OP-1 is an effective osteoinductive agent in lumbar intertransverse process fusion at both early and late time intervals. Spinal rigid fixation may potentiate the effect of OP-1 as evidenced by one of four animals showing arthrodesis by 3 weeks. OP-1 is effective as a bone graft substitute, as evidenced by radiographic, histologic and molecular data. Improved bone formation in smokers with recombinant human bone morphogenetic protein–2 Harvinder S. Sandhu, MD, New York, NY, USA; Rick B. Delamarter, MD, Linda E.A. Kanim, MA, Santa Monica, CA, USA; Frank P. Cammisa, Jr., MD, New York, NY, USA; Aleksandar Curcin, MD, Baltimore, MD, USA; Edward N. Hanley, Jr., MD, Charlotte, NC, USA; Jeffrey S. Fischgrund, MD, Southfield, MI, USA; Alexandre Valentin-Opran, MD, Cambridge, MA, USA; Thomas Lang, MD, San Francisco, CA, USA Introduction: Smoking clearly inhibits bone fusion in spinal surgery patients. Recombinant human bone morphogenetic protein (rhBMP)-2 has been shown to increase bone formation. Can rhBMP-2 reverse the deleterious effects of smoking? Purpose: To assess bone formation after implantation of varying concentrations of rhBMP-2 in attempted reconstruction of the iliac autogenous bone donor site in patients undergoing cervical fusion with and without a reported history of smoking. Methods: A four-center, patient-blinded, randomized controlled study comparing reconstruction of the iliac crest donor site with three doses of rhBMP-2/ Absorbable Collagen Sponge (ACS) to buffer/ACS or standard methods of closure (iliac defect left empty) was conducted. Forty patients undergoing a one- or two-level cervical fusion procedure requiring autogenous iliac crest bone grafting were informed of the study, enrolled and randomized before surgery (between 1998 and 1999). The iliac crest harvest technique was standardized. Donor site was left empty (n8), reconstructed with buffer/ACS (n8) or with rhBMP-2/ACS 0.43 mg/ml (n8), 0.75 mg/ml (n8) or 1.50 mg/ml (n8). At the time of enrollment, patients reported recent smoking history (ever and current smokers). Osteoinduction was monitored at baseline and 24 weeks on computed tomography (CT) and three-dimensional reconstructions. “Bridging bone” was considered as none (0), small to moderate (1) or extensive (2) filling in of the defect. Multivariate analysis techniques were used. Results: Follow-up evaluations were completed by all patients (CT available on 39 patients). Mean age was 42 years (range, 26 to 33 years); 65% of patients were men and 35% women. Thirteen patients reported a history of smoking, and 26 patients had no recent smoking history. Rates of smoking were comparable among all treatment groups. Bridging bone occurred more frequently in patients without a smoking history than for those with a history across all treatments except for 1.5 mg/ml rhBMP-2/ACS (highest dose). After 1.5 mg/dl rhBMP-2 administration, there was no difference in osteoinduction, bridging bone in the defect site, between patients with and without a history of smoking (overall model logistic, p.05).
18S
Proceedings of the NASS 16th Annual Meeting / The Spine Journal 2 (2002) 3S–44S
Conclusion: Although bone formation was less for patients with a recent history of smoking at lower doses of rhBMP-2, these differences were not observed at the dose of 1.5 mg/ml rhBMP-2. Treatment with the highest concentration of rhBMP-2 may overcome the nicotine effect on osteoinduction. Despite these very early and preliminary results, smoking cessation before a fusion procedure is crucial. IGF-I and TGF-1 application by a poly-(D,L-lactide)–coated cage promotes fusion in the sheep cervical spine Frank Kandziora, MD, N.P. Hass, Gerhard Schmidmaier, Georg Schollmeier, Herman Bail, Robert Pflugmacher, MD, Thomas Gorke, Martin Wagner, Thomas Mittlmeier, PhD, Michael Raschke, Berlin, Germany Study design: Growth factors such as bone morphogenetic protein (BMP)-2 have proven to promote spine fusion and to overcome the disadvantages of an autologous bone graft. The optimum growth factor to promote spinal fusion, as well as the optimum method to deliver such growth factors, is still a matter of discussion. A sheep cervical spine interbody fusion model was used to determine the effect of IGF-I and TGF-1 application by a biodegradable poly-(D,L-lactide) (PDLLA)–coated cage. The purpose of this study was to determine the safety and efficacy of a new PDLLA carrier system and to evaluate the safety and efficacy of combined IGF-I and TGF-1 application in a sheep cervical spine model. Methods: Thirty-two sheep underwent C3–4 discectomy and fusion: Group 1: autologous tricortical iliac crest bone graft (n8); Group 2: titanium cage (n8); Group 3: titanium cage coated with a PDLLA carrier (n8); Group 4: titanium cage coated with a PDLLA carrier including IGF-I (5% w/w) and TGF-1 (1% w/w) (n8). Blood samples, body weight and temperature were analyzed. Radiographic scans were performed pre- and postoperatively and after 1, 2, 4, 8 and 12 weeks, respectively. At the same time points, disc space height, intervertebral angle and lordosis angle were measured. After 12 weeks, animals were killed and fusion sites were evaluated using functional radiographic views in flexion and extension. Quantitative computed tomographic scans were performed to assess bone mineral density, bone mineral content and bony callus volume. Biomechanical testing was performed in flexion, extension, axial rotation and lateral bending. Stiffness, ranges of motion, neutral and elastic zone were determined. Histomorphological and histomorphometrical analysis was performed, and polychrome sequential labeling was used to determine the time frame of new bone formation. Results: There were no differences between the groups concerning blood counts, body weight and temperature. Over a 12-week period, the cage Groups 2, 3 and 4 showed significantly higher values for intervertebral angle compared with the bone graft. Functional radiographic assessment revealed significantly lower residual flexion/extension movement in Group 4 than in any other group. The PDLLA-coated cages with IGF-I and TGF-1 showed significantly highest values for bone mineral density, bone mineral content and bony callus volume. Average stiffness in rotation and bending was significantly higher, and range of motion, neutral and elastic zone in rotation were significantly lower in Group 4 than in any other group. Histomorphometrical evaluation showed a more progressed bone matrix formation in PDLLAcoated cages with IGF-I and TGF-1 than in any other group. Polychrome sequential labeling showed accelerated intervertebral fusion in Group 4. Conclusion: PDLLA coating of cervical spine interbody fusion cages as a delivery system for growth factors was effective and safe. The PDLLA coating showed no adverse effects. IGF-I and TGF-1 application by a PDLLA-coated interbody cage significantly increased results of interbody fusion in a sheep cervical spine model without adverse side effects. In this new combination (implant plus PDLLA plus growth factors) the cage represents a “real fusion” cage, because it does not only serve as a mechanical device for spine fixation but also as a local drug delivery system. A retrospective review of early outcomes of balloon kyphoplasty Steven R. Garfin, MD, San Diego, CA, USA Background context: An estimated 275,000 vertebral body compression fractures (VCFs) secondary to osteopenia, and refractory to medical man-
agement, occur in the United States each year. The deformity caused by VCFs is associated with chronic pain, impaired physical functioning, reduced quality of life, restrictive airway disease and increased mortality. This is a retrospective chart review of outcomes of inflatable bone tamp reduction and internal fixation using polymethylmethacrylate (PMMA). Methods: Vertebral body height was measured from X-rays from all patients treated by the authors (n451). The point of greatest height loss was measured, as was that at a comparable location in the closest normal vertebral body. The height of the fractured vertebral body was divided by the height of the normal vertebral body to calculate the percent estimated height before treatment. After treatment, the same location was measured in the treated and nearby vertebral body, to calculate the estimated percent height after treatment. The results were stratified by duration of symptoms at diagnosis, before or after 3 months. The percent lost height restored was determined by: (percent estimated height before fracture minus percent estimated height after fracture)/(1 percent estimate height after treatment). Reviewing the use of pain medication assessed reduction in pain. Institutional variations in these determinations were compared with the other centers performing kyphoplasty. The potential for inducing adjacent level fracture was assayed by determining the number of adjacent level fractures before treatment and then the number of new fractures that were adjacent to previously treated fractures. Results: A total of 2,194 fractures were treated (T43, T525, T648, T7106, T8114, T9117, T10102, T11174, T12333, L1365, L2249, L3241, L4183, L585, L61) in 1,439 patients in the United States from September 1998 through November 2000. Seventy-five percent were women. The average age was 73.5 years (range, 19 to 99 years). In fractures where the duration of symptoms at the initial visit was under 3 months, the average fractured vertebral body height was 71% before treatment and 92% after treatment (p.002). In fractures over 3 months, the average fractured vertebral body height was 74% before treatment and 84% after treatment (p.05). The average lost height restored for all fractures was 57%. Ninety percent of patients returned to prefracture pain medication (most often nonsteroidal anti-inflammatory drugs for arthritic type complaints) levels by 2 weeks. Major complications included three postoperative thoracic-level parapareses related to instrument insertion through the medial wall seventh pedicle with cord injuries. One patient had kyphoplasty performed while on anticoagulants and developed paraparesis and an epidural hematoma, which recovered completely after surgical decompression. There was one cement embolus without breathing consequences and a postoperative ileus caused by retroperitoneal bleeding in a patient on warfarin sodium (Coumadin; DuPont Pharmaceuticals Company, Wilmington, DE, USA). The serious adverse event rate is 0.2% per fracture and 0.4% per patient. Unrelated serious adverse events perioperatively included two myocardial infarctions related to overhydration, during the procedure, and two non–procedure-related perioperative deaths (cancer and coronary artery disease related). In patients with more than one fracture before their first kyphoplasty procedures, 63% of fractures not yet treated were adjacent. In 46 patients who returned for additional treatment of 139 fractures, 30% of the new fractures were adjacent to previously treated fractures. The results were nearly identical at each center, independent of physician. Conclusion: Kyphoplasty is a safe and effective treatment option for patients with painful VCF secondary to osteopenia. Height restoration was most improved if treatment occurred before 3 months from fracture, but some height restoration was obtained after 3 months. Comparing adjacent fractures in the untreated patients with the 2% of treated patients returning for additional kyphoplasty procedures does not indicate that kyphoplasty increases the rate of adjacent fracture. There was a remarkable consistency of treatment outcomes in the multicenter study. The four neurologic complications occurred in the first 100 fractures. Technique adjustments were made, and none have occurred since. (PMMA is approved by the Food and Drug Administration for long bones and joints. It is off-label for use in the spine.)
Kyphoplasty in the treatment of vertebral fractures secondary to multiple myeloma Sean Dudeney, MD, Mohamed Hussein, MD, Isador H. Lieberman, MD, Cleveland, OH, USA