An inception cohort study of patients in a military clinic treated for lower back pain with lumbar fusion and SIGNAFUSE® with a systematic review of the literature

Background: Use of synthetic bone graft substitutes for spinal fusion has increased sharply over the past 20 years. SIGNAFUSE® is one such synthetic graft material that provides an osteostimulatory effect for spinal fusion. Because clinical trials are not required for commercialization of synthetic bone graft substitutes in the United States, fusion rates attained using SIGNAFUSE for lumbar fusion are not well documented. The goal of the current study is to determine the rate of spinal fusion in a military clinic following lumbar fusion surgery augmented with SIGNAFUSE. Methods: We report a retrospective chart review of 8 patients who received lumbar spinal fusion surgery augmented with SIGNAFUSE. All patients were assessed by computed tomographic (CT) imaging at least 1-year post-surgery to determine whether bony fusion had occurred. We also systematically reviewed literature sources that report fusion rate following spinal fusion surgery, for broader context. Results: An average of 1.6 spinal levels were treated with SIGNAFUSE-loaded interbody cages. All patients had stabilization hardware via pedicle screws or integrated cage fixation. Seven of 8 patients successfully fused, for an overall fusion rate of 87.5% (95% confidence interval: 47.4% to 99.7%). Systematic review of 26 recent publications that included 1,126 patients treated with synthetic bone graft showed that the overall fusion rate in the literature is 84.4%. Conclusions: Fusion was achieved in 87.5% of patients treated with SIGNAFUSE. This is comparable to the fusion rate in a systematic review of 1,126 patients treated with synthetic bone graft materials. © 2018 R. Grant Steen. Hosting by Science Repository. All rights reserved.


Introduction
About two-thirds of adults have lower back pain (LBP) at some point during their lifetime [1]. LBP may be even more prevalent in a military population, due to acute trauma in combat and stress injuries in training [2]. LBP caused more than 457,000 Americans to seek spinal fusion in 2011, with ~75% of fusion procedures for spondylolisthesis, spondylosis, or other painful disk disorders [3,4].
Lumbar fusion surgery is done to achieve solid fusion, restore spinal stability, and enable return to normal activities. Spinal fusion procedures use various biologic materials to promote new bone growth, with iliac crest bone graft (ICBG) considered the "gold standard" of treatment, because of its osteoconductivity, osteoinductivity, osteogenicity, and angiogenic potential [5,6]. Though ICBG is widely used, problems such as donor site morbidity, limited supply of graft material, potential increase in surgical complication rates, and the increase in surgical times argue against use of ICBG [5,7]. Thus, there has been an effort to develop other bone grafting options that include allograft bone (e.g., demineralized bone matrices, allograft chips), synthetic graft materials (e.g. hydroxyapatite or HA), β-tricalcium phosphate (β-TCP), bioactive glass, collagen-based matrix, or recombinant growth factors (e.g., rhBMP-2). Synthetic graft materials provide the osteoconductive element that ICBG provides but may lack the osteoinductive and osteogenic components of ICBG [6]. To circumvent these limitations, some synthetic grafts incorporate bioactive glass to provide an osteostimulatory signal and many surgeons combine synthetics with bone marrow aspirate (BMA), to provide an osteogenic component. Synthetic graft materials are attractive because of the low risk of disease transmission, minimal immunogenicity, and superior intra-operative handling capabilities [6]. Additionally, synthetics avoid the potential adverse events reported in some patients following use of rhBMP-2, including retrograde ejaculation, ectopic bone formation, and dysphagia [8,9].
Here we report an inception cohort study of 8 patients treated in a military clinic who underwent lumbar fusion procedures to alleviate symptoms of degenerative disk disease and other spinal disorders. All patients were treated with SIGNAFUSE® Bioactive Bone Graft (Bioventus, Durham, NC), a combination synthetic graft composed of hydroxyapatite with β-tricalcium phosphate (HA/β-TCP) and bioactive glass in a resorbable polymer carrier, and all patients were evaluated for fusion by computed tomographic (CT) imaging. We also present a systematic review of publications reporting synthetic bone graft use in lumbar fusion surgery, as a point of comparison for our observed fusion results with SIGNAFUSE.

Patients and Methods
A planned retrospective chart review was combined with prospective computed tomographic (CT) imaging at follow-up, to evaluate the primary endpoint of radiographic fusion rate in patients treated with SIGNAFUSE at William Beaumont Army Medical Center (WBAMC) in El Paso, TX. The study received full Institutional Review Board (IRB) approval in October 2016. Potentially eligible patients were contacted after surgery by research staff to assess their interest in participating in the study. Patients were eligible for inclusion in this study if they: Patients interested in the study were invited for an on-site visit with a follow-up CT scan at WBAMC. A total of 19 patients were consented for study participation; 8 patients met inclusion criteria and completed a follow-up CT scan. Fusion status of the patient was assessed using CT scans and the patients were considered fused if there was evidence of bilateral continuous trabecular bone across the fusion bed.
A systematic review was also performed to compare the observed healing outcomes to healing outcomes from a range of different spinal fusion techniques that used many different commercially available synthetic bone graft substitutes. Sources were identified by several PubMed searches, augmented with sources drawn from a recent systematic review [10]. We sought to identify lumbar fusion procedures that used a range of commercially available synthetic bone grafts. Specific inclusion criteria were that studies: • Were identified using the search phrase "lumbar spinal fusion" or "lumbar spine fusion"; • Had been published within the last 20 years; • Were randomized controlled trials (RCTs), prospective cohorts, or retrospective cohorts; • Reported sufficient data to calculate fusion rate following fusion surgery; • Used synthetic bone graft materials, bone grafts augmented with autograft, or bone marrow aspirate during the procedure;  Patient #3 was a 34-year old male with a long history of lower back pain, beginning after an injury while on active-duty military service. His pain was midline in the lower back, with bilateral lower leg numbness and pain on the posterior aspect of both legs, with subjective muscle weakness. The patient was diagnosed with L5/S1 degenerative disc disease with herniation. The patient attempted conservative management, including physical therapy, pain medications, traction, and bracing, without relief, so he underwent an ALIF at L5/S1 without complication.
Three months later, X-rays demonstrated L5/S1 incorporation of the graft into the endplates with bony fusion. Patient #12 was a 25-year old male with bilateral leg pain and lower back pain of 2 years duration. Onset of pain was insidious, with no overt injury. Dorsal leg pain projected to the toes, with numbness and tingling, and patient had positional relief when leaning forward, but most other activities increased pain. Patient was diagnosed with L5/S1 spondylolisthesis with radiculopathy. The patient underwent an ALIF at L5/S1 without complication and CT scan demonstrated L5/S1 fusion of 40% of the disc at 20 months. Exclusion criteria (for the whole study or for a particular study arm) were that: • Patient follow-up was less than 12 months; • Surgery involved the cervical spine; • Bone graft materials were augmented with rhBMP; • Fusion was reported by level, not by patient; • Radiographic assessment of fusion was not reported; • Publication was as an abstract, conference proceeding, or editorial.

Results
Average patient age was 41.6 years (standard deviation (SD) = ± 10.6) at surgery and 2 patients had prior surgery (Table 1). Diagnoses within the cohort included 6 patients with spondylosis and 2 with spondylolisthesis. In 5 patients, SIGNAFUSE was used in combination with bioglass granules and in 1 patient SIGNAFUSE was used in combination with cancellous allograft bone. Four patients received anterior lumbar interbody fusion (ALIF), 3 received transforaminal lumbar interbody fusion (TLIF), and 1 received oblique lumbar interbody fusion (OLIF) fusion surgery. An average of 1.6 spinal levels were fused (range: 1 to 3 levels). All patients received stabilization hardware via pedicle screw instrumentation or integrated fixation built into the ALIF cages (Table 2). Average duration of surgery was 181 minutes (SD = ± 94 minutes), and operative blood loss was 164 mL (± 260 mL SD). The average hospital stay after surgery was 2.1 days, with the longest stay being 3 days. Seven of 8 patients had successful fusion, as determined by CT, for an overall fusion rate of 87.5% (95% confidence interval: 47.4% to 99.7%) ( Table 2). All 2-and 3-level fusion procedures showed evidence of radiographic fusion by CT scan within 2 years post-surgery, while 3 out of 4 patients with 1-level fusion procedures showed evidence of radiographic fusion. Figures 1 and 2 show representative CT scans of successfully fused patients.
A total of 26 published studies were identified for the systematic review . The process by which these studies were identified is shown (Figure 3). Included studies were predominantly posterolateral fusions (PLF) with instrumentation and used ceramic bone grafts augmented with autograft or bone marrow aspirate. Analysis of these published studies shows that the overall weighted average fusion rate is 84.4% in 1,126 patients (Table 3). Thus, fusion results from SIGNAFUSE-treated patients ( Table 2) are comparable to fusion results in the systematic review (Table 3).  Table 3 was assembled.

Discussion
Our results show that a fusion rate of 87.5% can be attained in a clinical setting with a range of different lumbar fusion approaches combined with SIGNAFUSE ( Table 2). The fusion rates obtained with SIGNAFUSE are thus comparable to fusion rates reported with a range of other synthetic bone grafts ( Table 3). The overall weighted average of fusion in the 26 publications was 84.4% (943 of 1,126 patients fused), while the fusion success of our SIGNAFUSE-treated patients was 87.5% (7 of 8 patients fused). Sub-analysis of the systematic review by graft type (HA, β-TCP, or HA/β-TCP) showed that fusion rates did not differ appreciably, with respective average fusion rates of 80%, 86%, and 88% (Table 3).
This study has several limitations. First, the number of patients in our series is small, and patients were treated with several different surgical techniques, while most published studies were PLF procedures. Second, varying amounts of SIGNAFUSE graft material were used (range: 7.5 -30 g) in our series (Table 2). Third, the systematic review reports a range of different synthetic graft materials which may not all be comparable to one another, but the sample size of the individual graft materials is too small to analyze separately. Fourth, even within the cohort of patients who received SIGNAFUSE, some patients had SIGNAFUSE augmented with bioactive glass, some patients had SIGNAFUSE augmented with cancellous chips, and some patients did not have any supplemental grafting materials. Fifth, the volume and relative ratio of other materials (BMA, autograft, etc.) mixed with graft materials varied, so it is not possible to attribute fusion success to any specific synthetic materials. For example, one study used HA and β-TCP without other supplemental grafting materials, while all other studies in the systematic review used BMA or autograft to augment the synthetic materials [33]. Finally, few controlled trials using synthetic bone graft materials have been conducted overall, so there is a compelling need for additional research.

Conclusions
Lumbar fusion surgery has increased in the recent past and is expected to continue to increase as the population ages [4]. Stable fusion is the goal of surgery, so there is a strong impetus to identify which types of bone grafting materials or bone graft substitutes are most effective in attaining fusion. The use of a synthetic graft such as SIGNAFUSE, alone or in combination with other grafting materials, may provide an advantage in spinal fusion surgery. Table 3: Systematic review of previously published studies utilizing synthetic bone grafts in lumbar fusion approaches HA = hydroxyapatite; β-TCP = beta-tricalcium phosphate; A = autograft; BMA = bone marrow aspirate; β-TCS = beta-tricalcium sulfate; C = collagen; DBM = demineralized bone matrix; Si-CaP = silicate substituted calcium phosphate.