Understanding Spine Biologics: A Systematic Review of Demineralized Bone Matrix in Spinal Fusion From 2014-2024

Abstract

Study Design

Systematic Review.

Objectives

Demineralized bone matrix (DBM) is an allograft material commonly used as an adjunct in spinal fusion. This review summarizes literature from the past decade describing DBM use in spinal fusion and provides an updated analysis of fusion outcomes based on product type and characteristics.

Methods

A systematic search was conducted using MEDLINE and CENTRAL. Risk of bias was assessed with the Cochrane Back and Neck criteria for randomized studies and the Methodological Index for Non-Randomized Studies criteria for nonrandomized studies. Extracted data included study design, sample size, procedure type, DBM product, fusion outcomes, complications, and follow-up duration. Additional product characteristics, such as DBM form (fiber-based or particulate-based) and sterility method (aseptic or terminal irradiation), were identified from tissue bank data.

Results

Twenty-one studies met inclusion criteria: 3 randomized (level I evidence), 4 level II, ten level III, and 4 level IV. Seven studies examined cervical procedures, and 14 addressed lumbar procedures. Posterior lumbar fusion (n = 6) and anterior cervical discectomy and fusion (n = 6) were most common. Among 18 comparative studies, 17 found no significant difference in fusion outcomes between DBM and comparison groups.

Conclusions

DBM used as an autograft extender demonstrated fusion rates comparable to alternative graft materials. However, the predominance of nonrandomized, retrospective designs and limited controlled data result in low overall evidence quality. Further high-quality prospective studies are warranted to better define the role of DBM in spinal fusion.

Keywords

spinal fusion bone matrix allografts autografts

Introduction

Over the past several decades, the incidence of spine fusion surgeries has increased significantly, driven by advancements in surgical techniques and an aging population.¹ Despite advancing techniques, pseudarthrosis continues to be a commonly associated complication, contributing to nearly one-fourth of spine fusion revisions.² Accordingly, graft augmentation has long been an extensively studied principle.^3,4 Autologous iliac crest bone graft (ICBG) is considered the “gold-standard” graft material for spine fusion due to high fusion rates without the risk of rejection or disease transmission.⁵ However, the limited availability and associated donor site morbidity limit the use of ICBGs clinically, prompting greater exploration and development of alternative solutions.^6,7 This has translated to an expansive market of commercially available bone graft products, which spine surgeons in the current era must understand to select the most appropriate products.⁸

Allopathic bone grafts, derived from human cadavers, are the most used alternative to autologous grafts such as ICBG.⁹ Demineralized bone matrix (DBM) is an allograft material that has become an increasingly popular adjunct to fusion procedures, comprising approximately half of all allografts used in the US.⁹ DBM is created by the acid treatment of cadaveric bone which removes most of the mineral content and retains the organic matrix, composed primarily of Type I collagen and growth factors including bone morphogenic proteins (BMP). The 0.5 M hydrochloric acid treatment is used by most tissue banks to mimic the natural bone regeneration process. Pre-clinical studies have demonstrated that a 2% residual calcium value leads to the greatest bone formation with less than 8% residual calcium required to meet the definition of a DBM per ASTM International Standard (F2529).¹⁰

Many different DBM products are currently available, with variances in further processing and sterilization techniques resulting in differing properties.¹¹ One such variance is the processing of DBM either as fiber-based or as particulate-based. Martin et al studied the improved fusion outcomes of fiber-based DBM over particulate-DBM in a preclinical, posterolateral intertransverse process spine arthrodesis model.¹² The posterolateral spine can be a more challenging healing environment than interbody spine procedures and thus was used to provide a better understanding of the differences between fiber-based DBM and particulate-based DBM. It was demonstrated that the devitalized (nonosteoinductive) fiber-based DBM outperformed the devitalized particulate-based DBM as the fiber-based DBM also provided better osteoconductive properties. A comparison of DBM products as fiber-based vs particulate-based in thus warranted.

As the commercial DBM industry grew, tissue regulations transitioned from voluntary standards with the establishment of the AATB (American Association of Tissue Banks) in 1976 to the first published “the standards of tissue banking” in 1984 and finally to the FDA’s good tissue practices (GTP) enforcement in 2005 of all tissue establishments.¹³ While an FDA tissue establishment is now required of all tissue banks, most musculoskeletal tissues are also distributed by tissue banks with AATB-accreditation. A sterility assurance level (SAL) 10⁻³ (1 in 1000) was established by the FDA as adequate for implantable medical biological devices. However, the Association for the Advancement of Medical Instrumentation (AAMI) and the AATB use SAL 10⁻⁶ as their standard. The FDA audits tissue banks and their methods of tissue processing, including sterilization, that include different biologic detergents, antibiotics, alcohol, or hydrogen peroxide with or without terminal irradiation as a final step.

Thus, another variance in processing of DBM is the sterilization technique. Early investigations of DBMs by Dr Marshall Urist in the 1960s-1970s led to an understanding that radiation affects the bone induction of DBM.^14,15 Dr Everard Munting confirmed these findings in the 1980s by noting a 60% reduction in implanted bone content in a preclinical rat model due to sterilizing with gamma irradiation.¹⁶ This has been studied further by others, including in 2005 by Dr Arthur Gertzman who noted a 53% reduction in the osteoinductivity score due to the terminal use of cold gamma irradiation.¹⁷ The negative side-effects of irradiation have been observed when employed as terminal irradiation of the DBM whereby bone proteins are more readily exposed rather than as initial steps to lower the bioburden from a mineralized allograft pre-demineralization. Of note, Musculoskeletal Tissue Foundation (MTF) may pretreat some allograft tissue donors with low-dose gamma irradiation but does not use terminal irradiation. While the first commercial DBM, launched in 1991, (Grafton DBM Gel by Osteotech) does not use irradiation and follows aseptic sterility processing techniques, many tissue banks currently use low-dose gamma irradiation, high-dose gamma irradiation, or electronic beam (e-beam) irradiation such as AlloSource, Bone Bank Allografts, Community Tissue Services (CTS), LifeNet, RTI Biologics, Inc, and Tissue Banks International.¹³ The possible mechanism by which irradiation reduces the osteoinductivity of DBMs has been investigated. A study in 2016 by Dr Uri Antebi observed reductions in the total protein of DBM samples as well as specifically BMP-2 and BMP-7 due to both gamma irradiation and e-beam irradiation of samples that were either lyophilized or frozen.¹⁸ In 2024, Dr Ganesh Shankar presented at the Cervical Spine Research Society (CSRS) Meeting on the negative impacts of both e-beam and gamma irradiation on the biological properties of DBMs. Structural changes to the DBM fibers were observed under the scanning electron microscope (SEM) with increased enzyme degradation and increased protein loss over time. BMP-2 elution at 24 hr was very rapid for all irradiated samples. The aseptic processing preserved a three-dimensional surface while both the gamma irradiated and the e-beam irradiated surfaces were smoothed over. The biological changes due to irradiation observed by Drs. Antebi and Shankar provide a possible mechanism for the reduction in osteoinductive potential observed previously by Drs. Urist, Munting, and Gertzman.

More recently, Dr Yong-jie Zhao investigated the variability in BMP-2 content in DBM when processing cortical bone from the femur, tibia, humerus, or ulna.¹⁹ Dr Zhao determined that BMP-2 concentration was greatest from the femur followed by the tibia, humerus, and ulna in descending order. Using a guanidine-HCL extraction (GuHCL), the enzyme-linked immunosorbent assay (ELISA) measured the femur as 5.6 times more than the tibia, 24.7 times more than the humerus, and 34.5 times more than the ulna. While it is unknown which tissue banks may or may not focus on specific cortical shafts in the production of their DBM, these differences in BMP-2 concentration may be an additional variance between different DBM formulations. Dr Hyun Bae in 2006 originally observed an intervariability and intravariability on BMPs in commercially available DBM products.²⁰ The DBM formulations studied varied in terms of fiber-based vs particulate-based, aseptic vs terminally irradiated, the percentage of DBM vs handling carriers, and likely the harvesting site of the DBM from the donors. Thus, many variables exist when reviewing DBM formulations and their clinical outcomes.

Despite the widespread use of DBM, few systematic reviews have been done assessing its use in spinal fusion. Furthermore, existing studies have analyzed DBM use only for specific spine procedures or in combination with non-spine surgeries and all were published more than 5 years ago.^21-23 Here, we aim to systematically review the available clinical literature describing the use of DBM products in spinal fusion within the past 10 years, present an updated analysis of their impact on fusion outcomes, and assess results by known variances such as fiber-based vs particulate-based and aseptic processing vs terminally irradiated processing. By synthesizing the latest evidence, we seek to provide insights into the current state of DBM use in spine fusion and identify areas for future research.

Methods

We conducted this systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Cochrane Handbook for Systematic Reviews of Interventions guidelines.^24,25 A literature search of MEDLINE (via PubMed) and Cochrane Central Register of Controlled Trials (CENTRAL) databases was conducted in July 2024. Search terms included “demineralized bone matrix”, “DBM”, “spine”, “fusion” and “arthrodesis” as described in previous studies.²⁶ The reference lists of selected papers captured in the search were also analyzed to identify any studies that may have been missed.

Our inclusion criteria were as follows: (1) clinical research studies involving the use of DBM in spinal fusion for at least ten patients, (2) studies explicitly stating the graft products used, (3) studies that report the rate of radiological fusion, and (4) studies published in English from January 2014 through June 2024. Studies that did not report original clinical data, including systematic reviews and meta-analyses, animal studies, and studies of DBM use in non-spinal surgeries were excluded. Two authors (M.B. and C.C.) independently reviewed the title, abstract, and, if necessary, the full text of studies to determine eligibility for inclusion. Any disagreement regarding the inclusion of studies was resolved by the senior author (F.V.). Two authors (M.B. and C.C.) independently extracted data including study design, sample size, type of procedure, DBM product used, DBM machining method, sterility methods, fusion outcomes, complications, and follow-up time from each study. There were no discrepancies between the authors.

Risk of bias was independently assessed by 2 authors (M.B. and C.C) and any disagreements were resolved by discussion, with involvement of the senior author (F.V.) as needed. We assessed randomized studies using the criteria recommended by the Cochrane Back and Neck Group, as described previously in other studies.^23,27 Non-randomized studies were assessed using the Methodological Index for Non-Randomized Studies (MINORS), described in a previous systematic review of anterior cervical spine procedures.¹³ Briefly, MINORS contains 12 items rated as follows: 0 when the article does not report the criterion, 1 if the criterion is reported but not sufficiently fulfilled, or 2 if the criterion is fulfilled.²⁸ A study’s total score is the sum of the criteria ratings, with higher scores indicating low risk of bias. Four of the 12 items only apply to comparative studies, so the highest possible score is 24 for comparative studies and 16 for non-comparative studies. A level of evidence, as described by Wright et al, was assigned for each included manuscript by 2 agreeing authors (M.B. and C.C.).²⁹

Results

Our database search generated 321 search results, of which 176 were duplicates and subsequently removed and 9 additional studies were identified by searching reference lists of relevant articles. Each study was screened by title and abstract, after which 116 were removed including 23 animal studies, 1 cadaver study, 34 review articles, and 58 unrelated or non-spine studies. The remaining 38 studies were reviewed in full to determine eligibility and 17 additional studies were excluded including 14 that did not report the DBM product used and 3 that did not report the rate of fusion or pseudoarthrosis. In total, 21 studies met the criteria for inclusion in our review (Figure 1).

Figure 1.

Flow diagram outlining our study search and inclusion process

Three randomized studies were included in our review, which were assessed using the Cochrane Back and Neck (CBN) group recommended criteria to assess the risk of bias (Table 1). There were no disagreements between the 2 assessors on any of the criteria items for these studies. There was an unclear risk of bias regarding both patient blinding and care provider blinding in the studies performed by Xie et al and Yi et al.^30,31 Only Yi et al reported allocation concealment, and, in the study performed by Hyun et al, there was a significant difference in both weight and body mass index between treatment groups.³² All 3 studies had a low risk of bias regarding the rest of the criteria items. We did not use a cutoff for classifying studies as higher or lower risk of bias, as recommended by the CBN group.²⁷ The remaining 18 studies included in our review were non-randomized and assessed using the MINORS checklist (Table 2). Fourteen of 18 were comparative studies, and a total of 200 risk of bias items were assessed with no disagreements between the assessors.

Table 1.

Risk of Bias for Randomized Studies Assessed With the 13-Item Criteria Recommended by the Cochrane Back and Neck Group

	Was the method of randomization adequate?	Was the treatment allocation concealed?	Was the patient blinded to the intervention?	Was the care provider blinded to the intervention?	Was the outcome assessor blinded to the intervention?	Was the drop-out rate described and acceptable?	Were all randomized participants analyzed in the group to which they were allocated?	Are reports of the study free of suggestion of selective outcome reporting?	Were the groups similar at baseline regarding the most important prognostic indicators?	Were the cointerventions avoided or similar?	Was the compliance acceptable in all groups?	Was the timing of the outcome assessment similar in all groups?	Are other sources of potential bias unlikely?
Xie et al, 2015¹⁹	L	U	U	U	L	L	L	L	L	L	L	L	L
Yi et al, 2015²⁰	L	L	U	U	L	L	L	L	L	L	L	L	L
Hyun et al, 2021²¹	L	U	L	L	L	L	L	L	H	L	L	L	L

H = high risk (no), L = low risk (yes), U = unclear (unsure).

Table 2.

Risk of Bias for Non-randomized Studies Assessed With the MINORS Checklist

	A clearly stated aim	Inclusion of consecutive patients	Prospective collection of data	Endpoints appropriate to the aim of the study	Unbiased assessment of the study endpoint	Follow-up period appropriate to the aim of the study	Loss to follow up less than 5%	An adequate control group	Contemporary groups	Baseline equivalence of groups	Adequate statistical analyses	Total
Baumann et al, 2015²²	2	2	1	2	0	2	1	2	1	2	2	17
Chung et al, 2016²³	2	2	2	2	2	2	2	-	-	-	-	14
Fu et al, 2016²⁴	2	2	1	2	0	2	2	2	1	2	2	18
Kim et al, 2016²⁵	2	2	1	2	0	2	1	2	2	2	2	18
Nam & Yi, 2016²⁶	2	2	2	2	2	2	2	2	2	2	2	22
Ryu et al, 2016²⁷	2	2	1	2	0	2	2	-	-	-	-	11
Gatam et al, 2017	2	2	2	2	0	2	1	2	2	2	2	19
Kim et al, 2017²⁸	2	0	1	2	2	2	0	2	1	2	2	16
Han et al, 2018²⁹	2	2	1	2	2	2	0	-	-	-	-	11
Ricart et al, 2018³⁰	2	2	1	2	2	2	0	1	2	2	2	18
Lee et al, 2019³¹	2	0	1	2	0	2	0	2	2	2	2	15
Chang et al, 2021³²	2	2	2	2	2	2	1	1	2	2	2	20
Ko et al, 2022³³	2	0	1	2	2	2	0	2	2	2	2	17
Roh et al, 2022³⁴	2	0	1	2	0	2	0	2	1	2	2	14
Sathe et al, 2022³⁵	2	2	2	2	0	2	0	1	1	1	2	15
Lee et al, 2023³⁶	2	2	2	2	2	2	0	-	-	-	-	12
Luo et al, 2023³⁷	2	2	1	2	2	2	2	1	2	2	2	20
Lee et al, 2024³⁸	2	2	1	2	2	2	0	1	2	2	2	18

0: Not reported; 1: Reported but inadequate; 2: Reported and adequate.

Seven included studies involved the use of DBM products in cervical spine procedures (Table 3). In a randomized clinical trial, Xie et al¹⁹ compared fusion outcomes between patients who underwent anterior cervical discectomy and fusion (ACDF) procedures with calcium sulphate plus DBM or autogenous iliac cancellous bone filled polyetheretherketone (PEEK) cages. Both treatment groups had a 100% fusion rate at 24-month follow-up, and the autogenous bone graft group experienced significantly greater intraoperative blood loss. In another ACDF clinical trial, Yi et al compared fusion outcomes between patients in which PEEK cages filled with hydroxyapatite (HA) and β-tricalcium phosphate (BTP) vs HA and DBM were used.³¹ In this trial, there was no significant difference in fusion rate at 12-month follow-up. Four non-randomized studies also investigated DBM in ACDF procedures, including 1 prospective case series using PEEK cages filled with a mixture of DBM fibers and cancellous bone chips, a retrospective case series using PEEK cages filled with DBM and autologous local bone, and 2 retrospective comparative studies. In the prospective case series, Chung et al reported a fusion rate of 100% at 6 months.³⁴ In the retrospective case series, Han et al reported a 100% fusion rate at 28 months.³⁹ In a retrospective study, Kim et al compared PEEK cages filled with DBM vs ICBG and reported 100% fusion rate in both groups at follow-up (25 months in the DBM group and 29 months in the ICBG group).³⁸ In the last retrospective ACDF study, Lee et al compared iliac autograft to cortical ring allograft plus DBM and reported greater blood loss in the autograft group with no statistically significant difference in fusion rates between the 2 treatment groups at 24 months.⁴⁰ The remaining cervical spine study was a retrospective case series by Ryu et al³⁷ reporting a 100% fusion rate at 12-month follow-up in patients who underwent atlantoaxial fusion using titanium mesh cages filled with DBM.

Table 3.

Details and Pertinent Results of Studies Reporting the Use of DBM Products in Cervical Spine Procedures

Authors & year	Study design; level of evidence	Procedure(s)	Treatment groups	Results	Fusion assessment	Determination of successful fusion
Xie et al, 2015¹⁹	RCT; I	ACDF	PEEK cages w/CS/DBM (Osteoset®): n = 35; PEEK cages w/autogenous iliac cancellous bone (AIB): n = 33	Fusion rate at 12 mos.: 94.3% CS/DBM, 100% AIB. Fusion rate at 24 mos.: 100% both groups. Higher total complication rate (P < 0.05) & blood loss (P < 0.01) in AIB group	Radiographs at hospital discharge, 6 weeks and 3, 6, 12, & 24 mos. after surgery; fine cut CT 12 mos. after surgery	Fusion was defined as the presence of continuous trabecular bone bridges in at least one of the following locations: Anterior, within, or posterior to the PEEK cage. Absence of such bridges or the presence of an anterior-posterior discontinuation was classified as non-fusion.
Yi et al, 2015²⁰	RCT; I	ACDF	PEEK cages w/HA/BTP: n = 42; PEEK cages w/HA/DBM (Bonion^TM): n = 43	Fusion rate at 12 mos.: 87% HA/DBM, 72% HA/BTP (P = 0.16)	Dynamic lateral radiographs taken in flexion and extension at 3, 6, and 12 mos. after surgery, CT at 6 and 12 mos. after surgery	Successful fusion defined as ≤ 2 mm difference in the interspinous distances on dynamic lateral radiographs taken in flexion and extension and the definite presence of a continuous fusion mass either inside or outside of the cage on CT
Chung et al, 2016²³	Prospective case series; IV	ACDF	PEEK cages w/DBM (CGDBM100^TM): n = 20	Fusion rate at 6 mos.: 100%	Plain radiographs and CT at 1, 3, and 6 mos. after surgery	Successful fusion defined as evidence of a bridging bone between the cage and the adjacent endplate of the vertebral body on at least 4 of 6 surfaces (anterior, posterior, superior, inferior, and both lateral sides)
Ryu et al, 2016²⁷	Retrospective case series; IV	Posterior cervical fusion (atlantoaxial)	Titanium mesh cage w/DBM (Synthes DBX®): n = 14	Fusion rate at 12 mos.: 100%	Flexion-extension lateral radiographs at 1, 3, 6, and 12 mos. after surgery	Successful fusion if the following criteria met: (1) osseous trabecular formation observed to bridge the C1-C2 lamina through or around the graft, (2) absence of abnormal motion of C1-C2 on flexion-extension lateral radiographs, (3) absence of hardware failure, and (4) absence of a significantly different atlantodental interval on radiographic images at 1 mo. and 12 mos. postoperatively
Kim et al, 2017²⁸	Retrospective cohort study; III	ACDF	PEEK cages w/DBM (Grafton^TM): n = 37; PEEK cages w/ICBG: n = 33	100% fusion rate at 29 mos. (ICBG) and 25 mos. (DBM)	Simple lateral and flexion-extension lateral radiographs, CT images immediately after surgery, at 8 weeks, 6 mos., and 12 mos. after surgery, and then annually	Fusion was defined as <2 mm motion between the spinous processes on flexion–extension lateral radiographs, absence of radiolucent gap between the graft and end plate junctions, and the presence of continuous bridging trabeculae at the graft and end plate junctions on simple lateral radiographs or CT images
Han et al, 2018²⁹	Retrospective case series; IV	ACDF	PEEK cages w/DBM (Grafton^TM) + ALB: n = 98	Fusion rate at 28 mos.: 100%	Anteroposterior and lateral radiographs at 6 weeks and 3, 6, 9, and 12 mos. after surgery then annually, flexion-extension radiographs at 6 mos. after surgery and on subsequent visits	Successful fusion if the following criteria met: (1) <2 mm change in the posterior interspinous process distance at treated level on lateral flexion-extension radiographs, (2) bridging bone between the end plates, (3) no sign of implant failure of the anterior plate system and (4) <50% of radiolucency covering the outer surface of the implant
Lee et al, 2019³¹	Retrospective cohort study; III	ACDF	Cortical ring allograft w/DBM (Orthoblast® II): n = 24; tricortical iliac autograft: n = 17	Fusion rate at 24 mos.: 94% DBM, 96 % iliac autograft (P = 0.66). Greater blood loss in autograft group (P = 0.013)	Flexion and extension lateral radiographs at 12 mos. after surgery	Successful fusion if < 1 mm of interspinous motion on dynamic radiographs magnified to 150%

RCT = randomized controlled trial, ACDF = anterior cervical discectomy and fusion, PEEK = Polyetheretherketone, CS = calcium sulphate, HA = hydroxyapatite, BTP = β-tricalcium phosphate, ALB = autologous local bone.

The remaining 14 studies assessed DBM use in lumbar spine procedures (Table 4). One study was a randomized controlled trial while the rest were non-randomized studies, including 8 that were retrospective and 5 that were prospective. In the randomized trial, Hyun et al compared DBM gel with and without recombinant human bone morphogenetic protein-2 (rhBMP-2) for transforaminal lumbar interbody fusion (TLIF) procedures, reporting no significant difference in fusion rates at 11 months.

Table 4.

Details and Pertinent Results of Studies Reporting the Use of DBM Products in Lumbar Spine Procedures

Authors & year	Study design; level of evidence	Procedure(s)	Treatment groups	Results	Fusion assessment	Determination of successful fusion
Baumann et al, 2015²²	Retrospective cohort study; III	Instrumented PLF	DBM (Synthes®): 16 patients; ICBG: n = 46	Fusion rate at 18 mos.: 94% DBM, 100% ICBG (P = 0.26)	Subjective evaluation of pain, physical examination, and conventional radiographs in 2 planes at 6 weeks and 3, 12, and 18 mos. after surgery	Spinal segment considered fused when there was no pain in the clinical examination and when radiographic signs of consolidation were apparent
Fu et al, 2016²⁴	Retrospective cohort study; III	Instrumented PLF ≧ 3 levels	DBM (Allomatrix^TM) + ALB + HA/BTCP: n = 26; ICBG + ALB + HA/BTCP: n = 21	Fusion rate at 12 mos.: 80.8% DBM, 85.7% ICBG (P = 0.72). More blood loss in ICBG group (P = 0.02)	Anteroposterior and flexion-extension dynamic plain radiographs at 3, 6, 12 and 18 mos. after surgery; The fusion success evaluation was based on findings on the 12-month plain radiographs	A successful fusion was determined as continuously qualitative intertransverse bony bridging at the target level on the follow-up radiographs
Kim et al, 2016²⁵	Retrospective cohort study; III	ALIF, PLIF, TLIF	HA-DBM (Bonfuse^TM): n = 65; ALB: n = 65	Fusion rate at 12 mos.: 52% HA-DBM, 62% ALB (P = 0.21)	CT at 12 mos. after surgery; The fusion grade was classified from Grade I to IV; Grade I = cortical union of the graft material and central trabecular continuity, Grade II = cortical union of the structural graft material with partial trabecular incorporation, Grade III = superior or inferior cortical non-union of the central graft material with partial trabecular discontinuity centrally, Grade IV = both superior and inferior cortical non-union with a complete lack of central trabecular continuity	Grades I and II considered successful fusion
Nam & Yi, 2016²⁶	Prospective cohort study; II	Instrumented PLF	HA-DBM (Bonfuse^TM): n = 38; HA: n = 41	Fusion rate at 12 mos.: 73% DBM, 58% HA (P = 0.15)	Anteroposterior, lateral and flexion-extension radiographs at 3, 6, and 12 mos. after surgery	Successful fusion if radiographs demonstrated a bilateral continuity in the fusion mass between the cephalad and caudad transverse processes, <2° of angular motion and <2 mm of sagittal motion at the level of the PLF
Gatam et al, 2017³⁹	Prospective cohort study; II	TLIF	ICBG: n = 18; HA + DBM (Bongener^TM): n = 17	Fusion rate at 12 mos.: 77.8% ICBG, 76.5% HA/DBM (P = 0.99)	Radiographs and CT images at 3, 6, and 12 mos. after surgery; Girasole et al criteria (based on spacer margin, bone within cage, and bone bridge between endplates)	A score of 3 or 4 considered successful fusion
Ricart et al, 2018³⁰	Retrospective cohort study; III	Instrumented PLF	DBM (Grafton^TM) + ALB: n = 21; BTP + ALB: n = 20	Fusion rate at 12 mos.: 90% DBM, 70% BTP (P = 0.09)	12 mo. CT; each spine level was given a fusion grade from 1-5 on both sides based on 3 radiographic findings: (1) bridging trabecular bone on at least 2 sequential images, (2) cortication of the peripheral edges of the fusion mass, and (3) the presence of an identifiable cleft on sequential images. “Grade 1″ represented definitely not fused, with radiographic finding number 3 on imaging. “Grade 2” represented probably not fused, with the absence of radiographic finding 1 or 2 and the presence of radiographic cleft on a single image but not sequential imaging. “Grade 3” was considered indeterminate, with radiographic finding 1, or 2 and 3 on a single image. “Grade 4” represented probably fused, with radiographic finding 1 and partial radiographic finding 2, with the absence of radiographic finding 3. “Grade 5” represented definitely fused, and was considered so if the patient had both radiographic finding 1 and 2, with the definite absence of radiographic finding 3.	A spine motion segment was considered fused if at least the right or the left was graded >4. A patient was considered “fused” if at least 1 side in each motion segment had a score >4.
Chang et al, 2021³²	Prospective cohort study; II	Instrumented PLF	DBM (Grafton^TM) + ALB: n = 19; DBM (Synthes DBX®) + ALB: n = 19	Fusion rate at 12 mos.: 59.5% Grafton^TM DBM, 51.9% DBX® DBM (P = 0.425). Fusion rate at 24 mos.: 70.4% Grafton^TM DBM, 66.7% DBX® DBM (P = 0.574)	Anteroposterior and flexion/extension lateral radiographs and 1-mm slice coronal CT	Fusion success at the surgical level if these criteria were met: (1) < 3° of angulation on flexion/extension lateral radiographs, (2) < 2 mm of translation on flexion/extension lateral radiographs, and (3) bridging bone connecting the transverse process on consecutive 1-mm coronal images from CT.
Hyun et al, 2021²¹	RCT; I	TLIF	DBM gel (Rafugen^TM) + rhBMP-2: n = 40; DBM gel (Rafugen^TM): n = 36	Fusion rate at 11 mos.: 82.86% DBM + rhBMP-2, 78.13% DBM (P = 0.6247)	Flexion and extension radiographs and CT scan at 12, 24, and 48 weeks post-treatment	Fusion on CT was defined as evidence of trabecular bone bridging at an intracompartmental site. Fusion on radiographs was defined as <5° of angular motion on flexion and extension radiographs or the absence of radiolucency lines >2 mm thick covering >50% of the superior or inferior surface of the grafts
Ko et al, 2022³³	Retrospective cohort study; III	Instrumented PLIF	ALB + DBM (Surfuse^TM): n = 20; ALB (cancellous only) + DBM (Surfuse^TM): n = 20	Fusion rate at 12 mos.: 100% both groups	Simple radiography at 3, 6, 12 mos. postoperatively; Brantigan-Steffee classification	Grades 4 and 5 considered successful fusion
Roh et al, 2022³⁴	Retrospective cohort study; III	LLIF	DBM (Grafton^TM): n = 34; DBM (Grafton^TM) + rhBMP-2: n = 20	Fusion rate at 12 mos.: DBM 97.4%, DBM + rhBMP-2100% (P = 0.552)	CT at 6 and 12 mos. postoperatively; modified Bridwell interbody fusion grading system (grade 1, fused with remodeling and trabeculae present; grade 2, not fully remodeled and incorporated, but no lucency present; grade 3, potential lucency present at the top and bottom of the graft; grade 4, fusion absent with collapse/resorption of the graft).	Grades 1 and 2 considered successful fusion
Sathe et al, 2022³⁵	Prospective cohort study; II	ALIF, LLIF	ABM/P-15: n = 46; rhBMP-2: n = 44; DBM(Grafton^TM): n = 50	ABM/P-15: 97.9% fusion rate at 16 mos.; rhBMP-2: 93.2% fusion rate at 18 mos.; DBM 98% fusion rate at 26 mos.	CT at 3 and 12 mos. after surgery and flexion-extension lateral radiographs at 3, 6, 9, and 12 mos. postoperatively	Successful fusion when osseous continuity observed on CT images and mobility of <4° seen on radiographs
Lee at al., 2023³⁶	Prospective case series; IV	OLIF	DBM (Grafton^TM): n = 79	Fusion rate at 12 mos.: 32.3%; 18 mos.: 58.3%; 24 mos.: 72.9%	Dynamic views of lateral radiographs and CT scans at 12, 18, and 24 mos. postoperatively; modified Brantigan-Steffee-Fraser (mBSF) scale (grades I-III; based on the presence of bridging bone and movement on dynamic radiographs)	Grade III (the presence of trabecular bridging bones in more than half of the fusion area on sagittal or coronal CT without movement) considered successful fusion
Luo et al, 2023³⁷	Retrospective cohort study; III	Instrumented PLF	ALB + DBM (Grafton^TM): n = 67; ALB + BCP: n = 81	Fusion rate at 24 mos.: DBM 89.6%, BCP 98.8%	Anteroposterior and flexion/extension radiographs, CT at 2 years postoperatively	Successful fusion if the following criteria met: (1) Bridging trabecular bone from the lateral-most aspect of the transverse processes to the medial-most aspect of the facet joints on at least 2 sequential CT images, (2) Cortication of the peripheral edges of the fusion mass, (3) Absence of an identifiable cleft on sequential CT images, and (4) Absence of motion on flexion–extension radiographs
Lee at al., 2024³⁸	Retrospective cohort study; III	PLIF	Exfuse^TM DBM: n = 98; Bongener^TM DBM: n = 72; Bonfuse^TM DBM: n = 66	Fusion rate at 12 mos.: 88.3% exfuse^TM, 94.3% Bongener^TM, 87.7% Bonfuse^TM (P = 0.214)	Flexion/extension lateral radiographs at 3, 6, and 12 mos. postoperatively and 1 mm slice coronal and sagittal CT at 6 and 12 mos. postoperatively; Bridwell grading system	Grades 1 and 2 considered successful fusion

PLF = posterolateral lumbar fusion, ICBG = autologous iliac crest bone graft, ALB = autologous local bone, ALIF = anterior lumbar interbody fusion, PLIF = posterior lumbar interbody fusion, TLIF = transforaminal lumbar interbody fusion, HA = hydroxyapatite, BTP = β-tricalcium phosphate (BTP), RCT = randomized controlled trial, rhBMP-2 = recombinant human bone morphogenetic protein-2, RR = relative risk, LLIF = lateral lumbar interbody fusion, ABM/P-15 = anorganic bone matrix/15-amino acid peptide fragment, ASD = adult spinal deformity, DCF = demineralized cortical fibers, BMA = bone marrow aspirate, OLIF = oblique lateral interbody fusion, BCP = biphasic calcium phosphate.

Six studies investigated DBM use in posterolateral lumbar fusion (PLF) including 2 prospective studies, one of which compared 2 different commercial DBM products (detailed in Table 5) and the other which compared HA with and without DBM.^36,41 There was no significant difference in fusion rate at 12 months in either study. Chang et al also reported no difference in the fusion rate between treatment groups after 24 months of follow-up. The remaining PLF studies compared DBM to BTP (n = 1), biphasic calcium phosphate (BCP; n = 1), and autologous ICBG with (n = 1) and without (n = 1) the use of autologous local bone.^33,35,42,43 In their study comparing DBM with autologous local bone to ICBG with autologous local bone, Fu et al reported a significant increase in blood loss in the group receiving ICBG. Fusion rates were similar across all treatment groups in these studies, except for Luo et al who compared DBM to BCP in one- and two-level PLFs. They found similar fusion rates between treatment groups for one-level surgeries but reported a greater fusion rate for patients receiving BCP in two-level surgeries (98.3% vs 80.0%, P = 0.014).

Table 5.

Characteristics of Commercial DBM Products Used in Included Studies

DBM product	Manufacturer	Description; form	Sterility processing method	Indications	Studies
DBX®	DePuy Synthes, Paoli, PA, USA	DBM (particulate) w/hyaluronic acid carrier; paste, putty, mix or strips	Aseptic	Autograft extender	Baumann et al, 2015,²² Ryu et al, 2016,²⁷ Chang et al, 2021³²
Orthoblast® II	IsoTis Orthobiologics, Inc., Irvine, CA, USA	DBM (particulate) w/reverse-phase medium carrier + cancellous bone; paste or putty	Terminally irradiated	Autograft extender	Lee et al, 2019³¹
Osteoset® 2 DBM	Wright Medical Technology, Inc, Memphis, TN, USA	DBM (particulate) w/calcium sulphate; pellets	Terminally irradiated	Autograft extender	Xie et al, 2015¹⁹
Allomatrix^TM	Wright Medical Technology, Inc, Memphis	DBM (particulate) w/calcium sulphate; putty	Terminally irradiated	Autograft extender	Fu et al, 2016²⁴
CGDBM100^TM	CG Bio Co, Ltd., Seoul, Korea	DBM (fiber) + cancellous bone chips; putty	Terminally irradiated	Autograft substitute or extender	Chung et al, 2016²³
Grafton^TM DBM	Medtronic, Inc, Minneapolis, MN, USA	DBM (fiber) w/glycerol carrier; putty or strips	Aseptic	Autograft extender	Kim et al, 2017,²⁸ Han et al, 2018,²⁹ Ricart et al, 2018,³⁰ Chang et al, 2021,³² Roh et al, 2022,³⁴ Sathe et al, 2022,³⁵ Lee et al, 2023³⁶ Luo et al, 2023³⁷
Bonfuse^TM	CG Bio Co, Ltd., Seoul, Korea	DBM (fiber) + hydroxyapatite; putty	Terminally irradiated	Autograft extender	Kim et al, 2016,²⁵ Nam & Yi, 2016,²⁶ Lee et al, 2024³⁸
Bonion^TM	CG Bio Co, Ltd., Seoul, Korea	DBM (particulate) w/poloxamer hydrogel carrier; paste	Terminally irradiated	Autograft extender	Yi et al, 2015²⁰
Surfuse^TM	Hans Biomed Co., Ltd., Seoul, Korea	DBM (particulate) w/carboxymethylcellulose carrier; putty	Terminally irradiated	Autograft extender	Ko et al, 2022³³
Exfuse^TM	Hans Biomed Co., Ltd., Seoul, Korea	DBM (particulate) w/carboxymethylcellulose carrier + cancellous bone chips; putty	Terminally irradiated	Autograft extender	Lee et al, 2024³⁸
Bongener^TM	CG Bio Co, Ltd., Seoul, Korea	DBM (fiber) + blood controlled thermal responsive polymer(B-TRP); paste	Terminally irradiated	Autograft substitute or extender	Gatam et al, 2017,³⁹ Lee et al, 2024³⁸
Rafugen^TM	Cellumed Co., Ltd., Seoul, Korea	DBM (particulate) + carboxymethylcellulose, starch, glycerol; gel	Terminally irradiated	Bone void filler	Hyun et al, 2021²¹

The remaining 8 studies involved DBM use for lumbar interbody fusion. Each of the comparative studies reported no significant difference in outcomes between treatment groups. In a retrospective study, Kim et al compared a commercially available HA/DBM product to autologous local bone in patients undergoing anterior lumbar interbody fusion (ALIF), posterior lumbar interbody fusion (PLIF) or transforaminal lumbar interbody fusion (TLIF) procedures.⁴⁴ Two other retrospective studies used DBM for PLIFs, including Ko et al who compared autologous local bone with DBM to only cancellous autologous local bone with DBM, and Lee et al who compared 3 different commercial DBM products.^33,38 Roh et al compared DBM with and without rhBMP-2 in patient undergoing lateral lumbar interbody fusion (LLIF) procedures.⁴⁵ In the remaining 3 prospective studies, Sathe et al compared anorganic bone matrix/15-amino acid peptide fragment (ABM/P-15), rhBMP-2, and DBM for ALIF and LLIF procedures, Gatam et al compared HA with DBM to ICBG for TLIFs, and Lee at al. conducted a prospective case series studying DBM use in oblique lateral interbody fusion (OLIF) procedures reporting 72.9% fusion rate at 24 months.^46-48 Additional descriptions of the commercial DBM products used in included studies are provided in Table 5.

Discussion

The Overall Quality of Evidence Among Included Studies of DBM in Spinal Fusion is Limited

The aim of this systematic review was to assess the clinical evidence for the use of DBM products in spinal fusion in the past 10 years. An extensive search of the literature yielded 21 studies that met our inclusion criteria and just 3 of these were randomized, constituting level I evidence. Four studies represented level II evidence, while most studies represented level III (10 studies) and IV (4 studies) evidence. Overall, studies were limited mainly to retrospective studies and many lacked a standard control group or used historical controls. Almost all studies reported no significant difference in fusion rates between treatment groups. However, in more than half of the non-randomized studies, it was unclear if their outcome assessment was unbiased as many did not disclose whether the assessors were blinded to the interventions. In the randomized studies, blinding of the patients and providers was unclear in 2 and only 1 reported concealment of treatment allocation which may represent sources of performance and selection bias, respectively.²⁷

DBM Products in the Studies Varied

Commercial DBM products are available by many different companies and in varying formats. Within the studies that met out inclusion criteria, Grafton^TM DBM was the most studied product (8 studies) followed by Synthes DBX® and Bonfuse^TM which were each used in 3 studies. Products commercially available in the US consisted of Grafton^TM DBM, Synthes DBX®, Allomatrix^TM, Osteoset® II DBM, and Orthoblast® II DBM. Other products available outside of the U.S. consisted of Bonfuse^TM, Bongener^TM, CGDBM100^TM, Bonion^TM, Rafugen^TM, Surfuse^TM, and Exfuse^TM.

For the US commercial products, the DBM products were mostly used in combination with local bone as autograft extenders, but several were also used without local bone. For the Grafton DBM studies, 4 were used with autologous local bone and 4 were used alone. Grafton DBM Putty consists of DBM fibers with a glycerol carrier processed aseptically. One study (Sathe et al) used Grafton Orthoblend which includes allograft bone chips with the other components of Grafton DBM Putty. Synthes DBX was used once with autologous local bone and twice used alone. Synthes DBX consists of particulate-based DBM with a hyaluronan carrier processed aseptically. Osteoset II DBM consists of a mixture of particulate-based DBM and calcium sulfate terminally processed with irradiation and was used alone. Allomatrix is similar to Osteoset II DBM, but in the study it was combined with a biphasic hydroxyapatite/beta-tricalcium phosphate synthetic granules (Foramic: Maxigen Biotech, Inc, Taiwan) as well as autologous local bone. Lastly, Orthoblast II is a particulate-based DBM used alone inside a cortical allograft ring terminally processed with irradiation.

For the products outside the US commercially, the DBM products were often processed as a combination with hydroxyapatite synthetic granules. Bonfuse was used in 3 studies and consists of a combination of hydroxyapatite with DBM fibers terminally processed with irradiation. Bongener consists of DBM fibers terminally processed with irradiation and was combined with hydroxyapatite for the studies. Bonion consists of particulate-based DBM terminally processed with irradiation and was also combined with hydroxyapatite. Additionally, 2 products are a combination of DBM with allograft bone chips. CGDBM100 consists of DBM fibers with bone chips terminally processed with irradiation. Exfuse consists of particulate-based DBM with bone chips terminally processed with irradiation. Surfuse was used in a study (Ko et al) and combined either with autologous local bone or only the cancellous portion of the local bone. Surfuse consists of particulate-based DBM terminally processed with irradiation. Rafugen is the only non-US product used alone and was used in an ACDF study. Rafugen consists of particulate-based DBM terminally processed with irradiation. Fusion outcomes stratified by DBM sterility processing method and graft form are summarized in Table 6, highlighting variability across formulations and follow-up intervals.

Table 6.

Fusion Rates by Sterility Processing Method and DBM Form

Product characteristics			Fusion rate (%)
Product characteristics			6 mos.	12 mos.	18 mos.	24+ mos.
Processing Method	Aseptic	DBX®	—	51.9-100	94	66.70
	Aseptic	Grafton^TM DBM	—	32.3-100	58.30	70.4-100
	Terminally irradiated	Orthoblast® II	—	—	—	94
		Osteoset® 2 DBM	—	94.30	—	100
		Allomatrix^TM	—	80.80	—	—
		CGDBM100^TM	100	—	—	—
		Bonfuse^TM	—	52-87.7	—	—
		Bonion^TM	—	87	—	—
		Surfuse^TM	—	100	—	—
		Exfuse^TM	—	88.30	—	—
		Bongener^TM	—	76.5-94.3	—	—
		Rafugen^TM	—	78.13-82.86 (11 mos.)	—	—
DBM form	Fiber	CGDBM100^TM	100	—	—	—
		Grafton^TM DBM	—	32.3-100	58.30	70.4-100
		Bonfuse^TM	—	52-87.7	—	—
		Bongener^TM	—	76.5-94.3	—	—
	Particulate	DBX®	—	51.9-100	94	66.70
		Orthoblast® II	—	—	—	94
		Osteoset® 2 DBM	—	94.30		100
		Allomatrix^TM	—	80.80	—	—
		Bonion^TM	—	87	—	—
		Surfuse^TM	—	100	—	—
		Exfuse^TM	—	88.30	—	—
		Rafugen^TM	—	78.13-82.86 (11 mos.)	—	—

Understanding the Percentage DBM of the Products

One limitation of comparing the outcomes across these studies is not knowing the percentage composition of demineralized bone content within each product. Han et al characterized a quantitative assay for understanding the relative osteoinductivity of DBM products using both a C2C12 cell in-vitro assay as well as an intramuscular nude rat assay.⁴⁹ DBM particulate without a handling agent was prepared and then a portion rendered inactive. DBM was inactivated by deproteinizing the DBM and thus denaturing the BMP proteins from the collagen structure that remains when bone is demineralized. Within the in-vitro assay, alkaline phosphatase (ALP) bone activity varied linearly as the volume composition of active DBM increased vs inactive DBM. These same volume compositions with various combinations of active DBM and inactive DBM were also implanted into the in-vivo model. A linear correlation was observed when comparing the in-vitro ALP activity to the ALP activity of explants at 28 days from the in-vivo model. Thus, the percentage by volume of the active DBM component is the best predictor of osteoinductive activity.

Often the percentage value by weight may be known but this is not the best characterization of a DBM product for comparison. For example, the handling agents/carriers, such as glycerol, hyaluronic acid, or a reverse phase medium carrier, will vary by weight as well. Thus, a heavier carrier, like glycerol, reduces the relative weight percentage of the active DBM component but the actual DBM volume is often higher than other DBMs with a low-weight carrier. Another complication is that most handling agents/carriers will break down and be removed from the implant site by the body within 48-72 hours of implantation. The overall volume of bone graft will settle into less than the original implant volume and may variably affect the clinical use. Thus, the percentage by volume of the DBM component is more important than the percentage by weight.

The DBM products included within these studies also vary in their composition by other non-osteoinductive components, such as synthetic granules or cancellous chips. These mineralized components provide structural scaffolding that may be clinically useful to maintain the graft volume space, ie, for use in the posterolateral spine, but with no active release of BMPs from these components, the potential osteoinductivity of the overall product is reduced by the volume percentage of these components as well.

Additionally, the volume of DBM used in each study varied and in 1 study was not well controlled between the study groups (39). Luo et al compared the use of a biphasic calcium phosphate (BCP) to DBM for posterolateral fusions.⁴⁷ In 1 group, 5cc of the BCP was used per side combined with local autogenous graft. However, the other group was combined with only 2.5cc of DBM. Also, TLIFs were performed as needed and patients with or without a TLIF were not controlled for between the 2 study groups with a portion of the available autogenous bone graft used in the TLIFs as well.

Understanding Fusion Classifications Used in the Studies

The studies in this review used multiple fusion classification systems, each with distinct criteria that shaped the interpretation of surgical outcomes. Most authors relied on the Bridwell grading system, which defines fusion based on the continuity of trabecular bone across the graft and ranges from Grade I (definite fusion) to Grade IV (definite nonunion).^50-53 This system remains the most widely adopted because it provides a reproducible way to distinguish between solid fusion and pseudarthrosis on plain radiographs or CT. Several studies instead used the Brantigan–Steffee classification, which grades interbody fusion from I to V and incorporates both radiographic and clinical stability, with Grades IV and V generally accepted as successful fusion.⁵⁴ Other investigators used binary definitions of fusion, categorizing patients simply as fused or not fused based on the presence of bridging trabecular bone on radiographs, CT scans, or dynamic studies.^55,56 In more recent work, researchers applied CT-based grading systems, including modifications such as the Tan et al CT fusion grading scale, which separates complete fusion from partial fusion.⁵⁷ Collectively, these approaches reveal significant variability in how surgeons and radiologists define and report fusion. Understanding these differences is essential, as the choice of classification system directly influences reported success rates, complicates comparisons across studies, and highlights the need for standardization in future spinal fusion trials.

The Use of DBM as an Autograft Extender Instead of a Graft Substitute

Graft extenders promote bone formation and can be used in conjunction with autologous bone to reduce the amount required for achieving a similar fusion rate, or to enable the fusion of additional levels with a standard amount of autologous graft. DBM graft extenders generally contain a lower percentage of DBM due to the addition of handling carriers, synthetic granules, or mineralized bone chips.¹⁰ Graft extenders may utilize sterility processing methods either aseptically or with terminal irradiation. Some DBM products are also indicated for use as graft substitutes, which can achieve similar fusion rates without the need for any autologous graft.²⁶ These “stand-alone” biologics typically consist of 100% DBM. Compared to stand-alone biologics, DBM graft extenders typically have lower DBM content and therefore less osteoinductivity, or degree to which graft materials induce host mesenchymal stem cells to differentiate into bone-forming osteoblasts. However, there can be a wide range of osteoinductivity between different DBM bone graft extenders. These products all contain various amounts of DBM, while some include carrier compounds and others include additional bone graft materials such as hydroxyapatite.^47,50 Furthermore, previous studies have shown significant variations in BMP concentrations across different lots of the same product.^51,52 These factors highlight the importance of analyzing the efficacy of DBM in a product-specific context. Due to the heterogenicity of product development and processing, we suggest that for future studies the specific DBM products used in a study should be reported.

Several studies in our review investigated DBM used alone and not in combination with allograft bone chips, autologous local bone, or synthetic granules such as hydroxyapatite. Kim et al published a retrospective cohort study (level III evidence) with 100% fusion of Grafton DBM alone in an ACDF procedure.³⁸ Roh et al published a retrospective cohort study (level III evidence) with 97% fusion of Grafton DBM alone in an LLIF procedure.⁴⁵ Lee et al published a case series (level IV evidence) with 73% fusion of Grafton DBM alone in an OLIF procedure.⁴⁷ Baumann et al published a retrospective cohort study (level III evidence) with 94% fusion of Synthes DBX alone in an instrumented PLF procedure.³³ Ryu et al published a case series (level IV evidence) with 100% fusion of Synthes DBX alone in a posterior cervical procedure.³⁷ All of these studies consisted of DBM processed aseptically without the use of terminal irradiation. Hyun et al published an RCT study (level I evidence) with 78% fusion of Rafugen in a TLIF procedure.³² Lee et al published a retrospective cohort study (level III evidence) with 94% fusion of Orthoblast II in an ACDF with a cortical allograft ring.⁴⁰ Both Rafugen and Orthoblast II are terminally processed with irradiation. These studies support the use of DBM used alone, but clinical evidence is more limited compared to that of DBM used in combination with other materials.

The Use of DBM for Cervical Fusion Procedures Was Associated With Similar Fusion Rates and Fewer Complications Compared to the Use of Iliac Autograft

Studies reporting the clinical use of DBM products for cervical spine procedures comprised about one-third of all included studies and 2/3 randomized studies. Three of these, including 1 RCT (level I evidence) and 2 retrospective cohort studies (level III evidence) compared DBM to iliac autograft in ACDF procedures and reported no significant difference in fusion rates between treatment groups. Overall, all studies of DBM in cervical fusion did report high fusion rates (100% in 3 of 5 studies) or noninferiority of the DBM products. Studies in which ICBGs were used showed significantly greater blood loss in ICBG patients. A 2016 review of DBM use in ACDF procedures by Zadegan et al included all published studies to date and found 1 additional randomized study that compared DBM with freeze-dried allograft to autograft and reported non-inferior fusion results.^23,53 There were several features of this study conferring a high-risk of bias, though, and Zadegan et al demonstrated a lack of high-quality evidence for DBM use in ACDF procedures. The findings of our review also indicate this, with the majority of our included studies constituting level III-IV evidence (67%). Furthermore, our review included only 1 study of DBM in posterior cervical fusion, which was a retrospective case series involving a small number of patients in which titanium mesh cages filled with DBM were used in atlantoaxial fusion (Level IV evidence).³⁷

DBM is Non-inferior to Iliac Autograft for Posterior Lumbar Fusion

Other studies in our review investigated DBM products for lumbar spine procedures, most commonly posterior lumbar fusions. Two retrospective cohort studies (level III evidence) compared DBM alone (without other graft products) to iliac autograft for posterior lumbar fusion and found no significant differences in fusion rate between the groups. The remaining studies used DBM products for posterolateral fusions and various lumbar interbody fusion procedures, and most indicate high or noninferior fusion rate in the DBM treatment groups. Among all studies of DBM in lumbar fusion, Luo et al were the only authors who reported a significantly lower fusion rate in the DBM group, concluding that DBM was inferior to BCP for two-level PLFs.

Hydroxyapatite With DBM Results in Similar Fusion Rates Compared to Autograft and Higher Fusion Rates Compared to Hydroxyapatite Alone for Lumbar Fusion Procedures

Hydroxyapatite with DBM was also investigated by several studies in our review. Three lumbar spine studies, including 2 interbody fusion and 1 PLF study, investigated HA with DBM compared to autologous local bone, ICBG, and HA alone. Both studies comparing HA-DBM to autograft reported no difference in fusion rates (level II and III evidence), and the study that compared HA-DBM to HA alone reported higher fusion in the group receiving DBM, though it was not statistically significant (level II evidence).^36,44,48 Overall, these results suggest the noninferiority of HA plus DBM to autograft in lumbar fusions.

Clinical Implications and Suggested Algorithm

Based on the synthesis of available clinical data, a pragmatic, procedure-specific algorithm for DBM use can be proposed. In cervical fusion, particularly ACDF, DBM demonstrates consistently high fusion rates when used as an autograft extender, with several studies reporting noninferior fusion compared to ICBG and reduced donor-site morbidity. Given the relatively favorable biologic environment of the interbody space and the lower graft volume requirements, aseptically processed, fiber-based DBM used as an ICBG extender represents a reasonable first-line adjunct in ACDF, especially in single-level procedures or in patients in whom avoidance of iliac crest harvest is desirable.

A similar approach may be applied to PLF, where DBM is most appropriately used as an autograft extender rather than a stand-alone graft. Across included PLF studies, fusion rates using DBM combined with local autograft were generally comparable to ICBG, particularly in single-level or short-segment constructs. In this setting, aseptic fiber-based DBM may be favored over particulate or terminally irradiated formulations, as preclinical and clinical data suggest superior osteoconductive structure preservation and potentially greater osteoinductive capacity. This is particularly relevant in the posterolateral environment, which is more biologically demanding than interbody fusion beds. However, caution is warranted in long-segment or multi-level PLF, where graft volume requirements are substantially higher and fusion demands are increased. One included study demonstrated inferior fusion rates for DBM compared with biphasic calcium phosphate in two-level PLF, highlighting that DBM performance may be more variable in this context.⁴⁷ Accordingly, in multi-level PLF, DBM should be used judiciously as a supplement to robust local autograft, and surgeons should consider alternative or additional biologic strategies when host or construct-related risk factors for pseudarthrosis are present.

Limitations

This systematic review has several important limitations that should be considered when interpreting the findings. First, the overall quality of evidence remains limited. Most included studies were retrospective and nonrandomized, with relatively small sample sizes and heterogeneous control groups, restricting the ability to draw definitive conclusions regarding the comparative efficacy of DBM across procedures and product formulations. In addition, DBM was frequently used in combination with local autograft, synthetic materials, or other biologics, making it difficult to isolate the independent contribution of DBM to fusion outcomes.

Second, there was substantial methodologic heterogeneity across studies, including variability in surgical indications, number of fused levels, graft volumes, and DBM formulations. Fusion assessment methods were inconsistent, ranging from plain radiographs to CT-based grading systems, with differing definitions of successful fusion. This variability limits cross-study comparability and precluded meaningful quantitative synthesis or meta-analysis.

Third, product-specific factors—including DBM form (fiber vs particulate), sterility processing method, carrier composition, and relative DBM content—were not consistently reported or controlled for, despite known biologic variability among commercial products. Similarly, lot-to-lot variability in osteoinductive potential and BMP content could not be assessed from the available clinical data.

Finally, clinical outcomes beyond radiographic fusion were incompletely reported. Revision surgery, pseudarthrosis-related reoperation, and patient-reported outcomes were inconsistently captured, and none of the included studies incorporated cost or economic analyses. As a result, conclusions regarding the relative value of DBM compared with other graft options remain limited. Taken together, these limitations highlight gaps in the existing literature and underscore the need for more rigorous, standardized investigation.

Future Directions

Advancing the clinical understanding of DBM in spinal fusion will require prospective, procedure-specific randomized controlled trials, particularly in posterior lumbar and multilevel constructs where fusion demands are greatest and outcomes appear more variable. To enhance consistency and interpretability across studies, future trials should incorporate standardized CT-based fusion assessment, such as the Tan et al fusion grading scale, which allows more precise differentiation between complete fusion, partial fusion, and pseudarthrosis and minimizes variability associated with plain radiographic criteria.⁵⁸

Equally important is the integration of biologic characterization alongside clinical outcomes. Prior experimental work has demonstrated substantial variability in BMP concentration, protein preservation, and elution kinetics across DBM formulations and sterility processing methods. Incorporating BMP quantification and elution assays into clinical trial design may help establish biologic–clinical correlations and clarify why certain DBM products perform differently in more challenging fusion environments, including multilevel posterolateral fusion.

From a health systems perspective, economic evaluation represents a critical gap in the existing literature. Despite DBM frequently being selected to avoid iliac crest harvest or to reduce reliance on recombinant biologics, none of the available studies have formally assessed cost or cost-effectiveness. Future investigations should therefore include economic endpoints, such as graft-related costs, operative time, donor-site morbidity, revision and reoperation rates, and quality-adjusted life years (QALYs), to better define the relative value of DBM compared with ICBG, synthetic grafts, and rhBMP-2 within value-based spine care models.

Conclusions

Most studies of DBM in spinal fusion within the past 10 years have investigated its use as an autograft extender compared to other bone graft materials or other DBM products. Studies commonly reported no significant difference in fusion outcomes between treatment groups, though the nonrandomized, retrospective nature of most studies with few having adequate control groups contributes to a low overall quality of evidence. However, DBM was studied more frequently in ACDF and PLF procedures, and our results suggest its noninferiority to ICBG for these cases. Similarly, HA plus DBM appears to be noninferior to autograft for lumbar fusion procedures. Additional randomized controlled trials are needed to better evaluate the clinical efficacy of DBM products, particularly in lumbar and posterior cervical spine procedures. The biggest challenge is the lack of detailed scientific study design and the heterogenicity in product development. These factors prevent a comprehensive understanding of current technology. Nevertheless, there is no doubt that a better insight into product development and technology applications can serve our future patients better.^20,59-65

Footnotes

ORCID iD

Molly Butler

Ethical Considerations

Ethics approval was not required for this systematic review.

Author Contributions

Conceived and designed the analysis M.B., C.C. and F.L.V.

Data collection: M.B. and B.M.

Data analysis and interpretation: M.B. and B.M.

Preparation of manuscript draft: M.B., B.M., and C.C.

Critical manuscript revision: M.Q., A.F.P. and F.L.V.

Approval of final manuscript draft: M.B., B.M., C.C., M.Q., A.F.P. and F.L.V.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

All datasets generated during and/or analyzed in this study are available from the corresponding author upon reasonable request.*

References

Cowan

Jr Dimick

Wainess

Upchurch

Jr Chandler

La Marca

. Changes in the utilization of spinal fusion in the United States. Neurosurgery. 2006;59(1):15-20. doi:10.1227/01.neu.0000219836.54861.cd; discussion 15-20.

Raizman

O’Brien

Poehling-Monaghan

. Pseudarthrosis of the spine. J Am Acad Orthop Surg. 2009;17(8):494-503. doi:10.5435/00124635-200908000-00003

Macewen

WIV

. Observations concerning transplantation of bone. Illustrated by a case of inter-human osseous transplantation, whereby over two-thirds of the shaft of a humerus was restored. Proc Roy Soc Lond. 1881;32:232-247.

Donati

Zolezzi

Tomba

Viganò

. Bone grafting: historical and conceptual review, starting with an old manuscript by Vittorio Putti. Acta Orthop. 2007;78(1):19-25. doi:10.1080/17453670610013376

Finkemeier

. Bone-grafting and bone-graft substitutes. J Bone Joint Surg Am. 2002;84(3):454-464. doi:10.2106/00004623-200203000-00020

Tuchman

Brodke

Youssef

, et al. Iliac crest bone graft versus local autograft or allograft for lumbar spinal fusion: a systematic review. Glob Spine J. 2016;6(6):592-606. doi:10.1055/s-0035-1570749

Kannan

Dodwad

Hsu

. Biologics in spine arthrodesis. J Spinal Disord Tech. 2015;28(5):163-170. doi:10.1097/bsd.0000000000000281

Chang

Hsu

. Spinal biologics in minimally invasive lumbar surgery. Minim Invasive Surg. 2018;2018:5230350. doi:10.1155/2018/5230350

Roberts

Rosenbaum

. Bone grafts, bone substitutes and orthobiologics: the bridge between basic science and clinical advancements in fracture healing. Organogenesis. 2012;8(4):114-124. doi:10.4161/org.23306

10.

Wang

Eaton

Harrigan

, et al. Demineralized bone matrix and fibers in spinal fusion. Int J Spine Surg. 2023;17(S3):S28-s34. doi:10.14444/8558

11.

Turonis

McPherson

3rd Cuenin

Hokett

Peacock

Sharawy

. The effect of residual calcium in decalcified freeze-dried bone allograft in a critical-sized defect in the Rattus norvegicus calvarium. J Oral Implantol. 2006;32(2):55-62. doi:10.1563/780.1

12.

Martin

GJJ

Boden

Titus

Scarborough

. New formulations of demineralized bone matrix as a more effective graft alternative in experimental posterolateral lumbar spine arthrodesis. Spine. 1999;24(7):637-645.

13.

Vangsness

Jr.

Allografts: graft sterilization and tissue banking safety issues graft sterilization and tissue banking safety issues. Noyes’ Knee Disorders: Surgery, Rehabilitation, Clinical Outcomes 2010;240-244.

14.

Urist

Hernandez

. Excitation transfer in bone. Deleterious effects of cobalt 60 radiation-sterilization of bank bone. Arch Surg. 1974;109(4):586-593. doi:10.1001/archsurg.1974.01360040012004

15.

Buring

Urist

. Effects of ionizing radiation on the bone induction principle in the matrix of bone implants. Clin Orthop Relat Res. 1967;55:225-234.

16.

Munting

Wilmart

Wijne

Hennebert

Delloye

. Effect of sterilization on osteoinduction comparison of five methods in demineralized rat bone. Acta Orthop Scand. 1988;59(1):34-38.

17.

Gertzman

Sunwoo

Raushi

Dunn

. The effect of cold gamma radiation sterilisation on the properties of demineralised bone matrix. In: Kennedy

Philips

Williams

, eds. Sterilisation of Tissues Using Ionising Radiations. Boca Raton, FL: CRC Press; 2005:151-156.

18.

Antebi

Mathor

da Silva

Guimarães

Honda

. Effects of ionizing radiation on proteins in lyophilized or frozen demineralized human bone. Rev Bras Ortop. 2016;51(2):224-230.

19.

Zhao

Yin

Liu

Deng

Cao

Liu

. Variability of BMP-2 content in DBM products derived from different long bone. Cell Tissue Bank. 2024;25:697-703.

20.

Bae

Zhao

Kanim

Wong

Delamarter

Dawson

. Intervariability and intravariability of bone morphogenetic proteins in commercially available demineralized bone matrix products. Spine. 2006;31(12):1299-1306. doi:10.1097/01.brs.0000218581.92992.b7; discussion 1307-8.

21.

Buser

Brodke

Youssef

, et al. Allograft versus demineralized bone matrix in instrumented and noninstrumented lumbar fusion: a systematic review. Glob Spine J. 2018;8(4):396-412. doi:10.1177/2192568217735342

22.

van der Stok

Hartholt

Schoenmakers

DAL

Arts

JJC

. The available evidence on demineralised bone matrix in trauma and orthopaedic surgery: a systematic review. Bone Joint Res. 2017;6(7):423-432. doi:10.1302/2046-3758.67.Bjr-2017-0027.R1

23.

Zadegan

Abedi

Jazayeri

Vaccaro

Rahimi-Movaghar

. Demineralized bone matrix in anterior cervical discectomy and fusion: a systematic review. Eur Spine J. 2017;26(4):958-974. doi:10.1007/s00586-016-4858-9

24.

Higgins

Thomas

Chandler

, et al. Cochrane Handbook for Systematic Reviews of Interventions; 2023. Updated August 2023. https://www.training.cochrane.org/handbook

25.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Br Med J. 2021;372:n71. doi:10.1136/bmj.n71

26.

Aghdasi

Montgomery

Daubs

Wang

. A review of demineralized bone matrices for spinal fusion: the evidence for efficacy. Surgeon. 2013;11(1):39-48. doi:10.1016/j.surge.2012.08.001

27.

Furlan

Malmivaara

Chou

, et al. 2015 updated method guideline for systematic reviews in the Cochrane back and neck group. Spine. 2015;40(21):1660-1673.

28.

Slim

Nini

Forestier

Kwiatkowski

Panis

Chipponi

. Methodological index for non-randomized studies (minors): development and validation of a new instrument. ANZ J Surg. 2003;73(9):712-716. doi:10.1046/j.1445-2197.2003.02748.x

29.

Wright

Swiontkowski

Heckman

. Introducing levels of evidence to the journal. J Bone Joint Surg Am. 2003;85:1-3.

30.

Xie

Yuan

Yang

Zhang

. A prospective randomized comparison of PEEK cage containing calcium sulphate or demineralized bone matrix with autograft in anterior cervical interbody fusion. Int Orthop. 2015;39(6):1129-1136. doi:10.1007/s00264-014-2610-9

31.

Lee

Nam

, et al. A prospective randomized clinical trial comparing bone union rate following anterior cervical discectomy and fusion using a polyetheretherketone cage: hydroxyapatite/b-tricalcium phosphate mixture versus hydroxyapatite/demineralized bone matrix mixture. Asian Spine J. 2015;9(1):30-38. doi:10.4184/asj.2015.9.1.30

32.

Hyun

Yoon

Kim

, et al. A prospective, multi-center, double-blind, randomized study to evaluate the efficacy and safety of the synthetic bone graft material DBM gel with rhBMP-2 versus DBM gel used during the TLIF procedure in patients with lumbar disc disease. J Korean Neurosurg Soc. 2021;64(4):562-574. doi:10.3340/jkns.2020.0331

33.

Baumann

Krutsch

Pfeifer

Neumann

Nerlich

Loibl

. Posterolateral fusion in acute traumatic thoracolumbar fractures: a comparison of demineralized bone matrix and autologous bone graft. Acta Chir Orthop Traumatol Cech. 2015;82(2):119-125.

34.

Chung

Hur

Ryu

Kim

Seong

. Surgical outcomes of anterior cervical fusion using deminaralized bone matrix as stand-alone graft material: single arm, pilot study. Korean J Spine. 2016;13(3):114-119. doi:10.14245/kjs.2016.13.3.114

35.

Wang

Hsieh

Chen

. The fusion rate of demineralized bone matrix compared with autogenous iliac bone graft for long multi-segment posterolateral spinal fusion. BMC Muscoskelet Disord. 2016;17:3. doi:10.1186/s12891-015-0861-2

36.

Nam

. Bone union rate following instrumented posterolateral lumbar fusion: Comparison between demineralized bone matrix versus hydroxyapatite. Asian Spine J. 2016;10(6):1149-1156. doi:10.4184/asj.2016.10.6.1149

37.

Ryu

Bak

Kim

Chun

. Evaluation of the efficacy of titanium mesh cages with posterior C1 lateral mass and C2 pedicle screw fixation in patients with atlantoaxial instability. World Neurosurg. 2016;90:103-108. doi:10.1016/j.wneu.2016.02.087

38.

Kim

Lee

Jang

Park

Hur

. Polyetheretherketone cage with demineralized bone matrix can replace iliac crest autografts for anterior cervical discectomy and fusion in subaxial cervical spine injuries. J Korean Neurosurg Soc. 2017;60(2):211-219. doi:10.3340/jkns.2015.0203.014

39.

Han

Lee

Kim

Lee

Moon

Lee

. Outcomes of anterior cervical fusion using polyetheretherketone cage with demineralized bone matrix and plate for management of subaxial cervical spine injuries. Korean J Nutr. 2018;14(2):123-128. doi:10.13004/kjnt.2018.14.2.123

40.

Lee

Jang

Ahn

Choi

Kang

Shin

. Comparison of cortical ring allograft and plate fixation with autologous iliac bone graft for anterior cervical discectomy and fusion. Asian Spine J. 2019;13(2):258-264. doi:10.31616/asj.2018.0174

41.

Chang

Park

Han

. Surgical outcomes of two kinds of demineralized bone matrix putties/local autograft composites in instrumented posterolateral lumbar fusion. BMC Muscoskelet Disord. 2021;22(1):200. doi:10.1186/s12891-021-04073-3

42.

Ricart

Gandhi

Geisinger

Baker

Park

. Clinical and CT analysis of lumbar spine arthrodesis: β-tricalcium phosphate versus demineralized bone matrix. J Am Acad Orthop Surg Glob Res Rev. 2018;2(9):e024. doi:10.5435/JAAOSGlobal-D-18-00024

43.

Lee

Ham

D-W

Kwon

Park

J-H

Yoon

Kim

H-J

. Comparison of fusion rates among various demineralized bone matrices in posterior lumbar interbody fusion. Medicina. 2024;60(2):265. doi:10.3390/medicina60020265

44.

Kim

Lee

Shin

Kim

. Matched comparison of fusion rates between hydroxyapatite demineralized bone matrix and autograft in lumbar interbody fusion. J Korean Neurosurg Soc. 2016;59(4):363-367. doi:10.3340/jkns.2016.59.4.363

45.

Roh

Lee

Cho

Jang

Choi

Shin

. Comparative study of radiological and clinical outcomes in patients undergoing minimally invasive lateral lumbar interbody fusion using demineralized bone matrix alone or with low-dose escherichia coli-derived rhBMP-2. World Neurosurg. 2022;158:e557-e565. doi:10.1016/j.wneu.2021.11.028

46.

Sathe

Lee

Kim

, et al. Comparative analysis of ABM/P-15, bone morphogenic protein and demineralized bone matrix after instrumented lumbar interbody fusion. J Korean Neurosurg Soc. 2022;65(6):825-833. doi:10.3340/jkns.2021.0296

47.

Luo

Chen

Chiu

Niu

. Use of biphasic calcium phosphate versus demineralized bone matrix: retrospective clinical and CT analysis of posterolateral fusion results. Eur Spine J. 2023;32(2):734-742. doi:10.1007/s00586-022-07499-y

48.

Kaiser

Groff

Watters

3rd , et al. Guideline update for the performance of fusion procedures for degenerative disease of the lumbar spine. Part 16: bone graft extenders and substitutes as an adjunct for lumbar fusion. J Neurosurg Spine. 2014;21(1):106-132. doi:10.3171/2014.4.spine14325

49.

Han

Tang

Nimni

. Quantitative and sensitive in vitro assay for osteoinductive activity of demineralized bone matrix. J Orthop Res. 2003;21(4):648-654.

50.

Faulks

Cabbabe

Biddau

, et al. A single straight expandable cage via a hybrid posterior-transforaminal approach with rhBMP-2 or allograft provides high fusion rates with low risk of subsidence. J Spine Surg 2025;11(1):1-14. doi:10.21037/jss-24-82

51.

Kulkarni

Kim

, et al. The insert and revolve technique: a novel approach for inserting cages during unilateral biportal endoscopic assisted fusion surgery for effective spinal alignment restoration. Asian Spine J 2024;18(4):514-521. doi:10.31616/asj.2024.0066

52.

Chien

KT.

Comprehensive analysis of novel uniportal trans-kambin endoscopic lumbar interbody fusion with over one year of CT-based fusion evaluation. North Am Spine Soc J. 2025;22(supplement);2666-5484. doi:10.1016/j.xnsj.2025.100659

53.

Duong

TVH

Tuan

, et al. Effectiveness of biportal endoscopic lumbar interbody fusion using the multi-layer bone grafting technique: a retrospective study from Vietnam. Asian Spine J 2025;19(2):228-241. doi:10.31616/asj.2024.0522

54.

Ali

EMS

Abdeen

Saleh

. Minimally invasive versus mini-open transforaminal lumbar interbody fusion in managing low-grade degenerative spondylolisthesis. Acta Neurochir (Wien) 2024;166(1):365. doi:10.1007/s00701-024-06231-7

55.

Feng

Xiang

, et al. Clinical efficacy and quality of life in elderly patients with lumbar degenerative disease treated with TLIF combined with unilateral pedicle screw fixation: a randomized controlled study. J Orthop Surg Res 2025;20(1):420. doi:10.1186/s13018-025-05821-0Copy

56.

Tan

Bailey

Dvorak

Fisher

. Interobserver and intraobserver reliability of radiographic fusion assessment in anterior cervical discectomy and fusion. Spine. 2007;32(10):E562-E567. doi:10.1097/BRS.0b013e31805471b0

57.

Gupta

Kukkar

Sharif

Main

Albers

El-Amin Iii

. Bone graft substitutes for spine fusion: a brief review. World J Orthoped. 2015;6(6):449-456. doi:10.5312/wjo.v6.i6.449

58.

Tan

Goss

Thorpe

Williams

. CT-based classification of long spinal allograft fusion. Eur Spine J. 2007;16(11):1875-1881. doi:10.1007/s00586-007-0376-0

59.

Jun

Nam

. Comparison of fusion rate and functional outcome between local cancellous bone plus demineralized bone matrix and local bone in 1-Level posterior lumbar interbody fusion. Clin Spine Surg. 2022;35(7):E621-E626. doi:10.1097/bsd.0000000000001330

60.

Lee

Jung

Lee

, et al. Fusion assessment of oblique lumbar interbody fusion using demineralized bone matrix: a 2-Year prospective study. Neurospine. 2023;20(4):1205-1216. doi:10.14245/ns.2347032.516

61.

Gatam

Lumban Tobing

. Comparison of clinical and radiological outcomes of lumbar interbody fusion using a combination of hydroxyapatite and demineralized bone matrix and autografts for lumbar degenerative spondylolisthesis. Asian Spine J. 2017;11(5):706-714. doi:10.4184/asj.2017.11.5.706

62.

Giannoudis

Dinopoulos

Tsiridis

. Bone substitutes: an update. Injury. 2005;36(Suppl 3):S20-S27. doi:10.1016/j.injury.2005.07.029

63.

Bae

Zhao

Zhu

Kanim

Wang

Delamarter

. Variability across ten production lots of a single demineralized bone matrix product. J Bone Joint Surg Am. 2010;92(2):427-435. doi:10.2106/jbjs.H.01400

64.

Simpson

Glover

Stephany

. Comparison between allograft plus demineralized bone matrix versus autograft in anterior cervical fusion. A prospective multicenter study. Spine. 1995;20(20):2211-2216.

65.

Russell

Walsh

Lovric

, et al. In-vivo performance of seven commercially available demineralized bone matrix fiber and putty products in a rat posterolateral fusion model. Front Surg. 2020;7:10. doi:10.3389/fsurg.2020.00010