Trends Analysis of rhBMP Utilization in Single-Level Posterior Lumbar Interbody Fusion in the United States.

STUDY
DESIGN: Retrospective study.
OBJECTIVES: Recombinant human bone morphogenetic protein-2 (rhBMP-2) has been widely used in spinal fusion surgery, but there is little information on rhBMP-2 utilization in single-level posterior lumbar interbody fusion (PLIF). The purpose of our study was to evaluate the trends and demographics of rhBMP-2 utilization in single-level PLIF.
METHODS: Patients who underwent single-level PLIF from 2005 to 2011 were identified by searching ICD-9 diagnosis and procedure codes in the PearlDiver Patient Records Database, a national database of orthopedic insurance records. The year of procedure, age, gender, and region of the United States were recorded for each patient. Results were reported for each variable as the incidence of procedures identified per 100 000 patients searched in the database.
RESULTS: A total of 2735 patients had single-level PLIF. The average rate of single-level PLIF with rhBMP-2 maintained at a relatively stable level (28% to 31%) from 2005 to 2009, but decreased in 2010 (9.9%) and 2011 (11.8%). The overall incidence of single-level PLIF without rhBMP-2 (0.68 cases per 100 000 patients) was statistically higher (P < .01) compared to single-level PLIF with rhBMP-2 (0.21 cases per 100 000 patients). The average rate of single-level PLIF with rhBMP-2 utilization was the highest in West (30.1%), followed by Midwest (26.9%), South (20.5%), and Northeast (17.8%). The highest incidence of single-level PLIF with rhBMP-2 was observed in the age group <65 years (0.3 per 100 000 patients).
CONCLUSIONS: To our knowledge, this is the first study to report on the demographics associated with rhBMP-2 use in single-level PLIF. There was a 3-fold increase in the rate of PLIF without rhBMP-2 compared to PLIF with rhBMP-2, with both procedures being mainly done in patients less than 65 years of age.

Introduction

Posterior lumbar interbody fusion (PLIF) was initially introduced by Cloward, and it is used for various spinal conditions. Successful fusion rates (>90%) have been reported for degenerative disc disease and spondylolisthesis.[1-4] An important objective of the PLIF procedure is to achieve solid arthrodesis of the spinal segment and to restore segmental lordosis at the involved level.[5,6] Recombinant human bone morphogenetic protein-2 (rhBMP-2) belongs to a family of growth factors that promote bone formation and remodeling.[7] In 2002, the Food and Drug Administration (FDA) approved the use of rhBMP-2 for anterior lumbar spine fusion.[8] Potential benefits of rhBMP-2 are an increase in fusion rates and a reduction in morbidity from the bone graft harvest. However, use of rhBMP-2 might be associated with complications such as heterotopic ossification, radiculitis, prevertebral soft-tissue swelling, and bone resorption.[9-12]

The main objective of this study was to define the epidemiology of rhBMP-2 utilization in single-level PLIF using a national private insurance database for the period between 2005 and 2011.

Materials and Methods

Patients undergoing single-level PLIF with and without rhBMP-2 were identified using the PearlDiver Patient Records Database (Pearl-Diver Technologies, Fort Wayne, IN). Medicare database between 2005 and 2011 was used for the purpose of this study. Patients were identified using Current Procedural Terminology (CPT) code 22 630 (“Arthrodesis, posterior interbody technique, including laminectomy and/or discectomy to prepare interspace [other than for decompression], single interspace; lumbar”) and combination or exclusion of International Classification of Diseases, Ninth Edition (ICD-9) code 84.52 (“Insertion of recombinant bone morphogenetic protein”), as listed in the Table 1. Patients with ICD-9 codes 81.30 to 81.39 for “correction of pseudarthrosis of spine, refusion of spine” and CPT code 22 585 for “each additional interspace” were excluded. The search was defined to exclude the same code being counted more than once for the same patient. Patients were stratified by gender, geographic region (Midwest, Northeast, South, and West), and age group (less than 65, 65-69, 70-74, 75-79, 80-84, and more than 84 years). In this study, PLIF included transforaminal lumbar interbody fusion and minimally invasive transforaminal lumbar interbody fusion. Our study was deemed exempt from institutional review board review as all patient information was de-identified.

Table 1.

ICD-9 Diagnosis Codes and CPT Procedure Codes Searched.

Code	Diagnosis/Procedure
CPT code 22 630	Arthrodesis, posterior interbody technique, including laminectomy and/or discectomy to prepare interspace (other than for decompression), single interspace; lumbar
ICD-9 code 84.52	Insertion of recombinant bone morphogenetic protein (rhBMP)

Abbreviations: CPT, Current Procedural Terminology; ICD-9, International Classification of Diseases, Ninth Edition.

Results were reported as incidence, number of single-level PLIF cases with/without rhBMP-2 identified per every 100 000 patients searched in a particular year, gender, age group, or region. This was done to account for differences in the number of patients in the database for a given variable. We used χ2 analysis to determine the statistical significance, and linear regression was performed to test the significance of trends over time. The level of significance was P < .05.

Results

A total of 2735 patients undergoing single-level PLIF were identified. There were 1711 women and 1024 men. There were 2094 patients undergoing single-level PLIF without rhBMP-2 and 641 patients undergoing single-level PLIF with rhBMP-2. The overall incidence of single-level PLIF without rhBMP-2 (0.68 cases per 100 000 patients) was 3 times higher than the incidence of single-level PLIF with rhBMP-2 (0.21cases per 100 000 patients; P < .01; Figure 1). Looking at the annual increase, incidence of PLIF without rhBMP-2 had a steady increase from 2005 to 2011 (Figure 2). Interestingly, single-level PLIF with rhBMP-2 maintained at a relatively stable level (0.22 to 0.28 cases per 100 000) from 2005 to 2009, but sharply decreased in 2010 and 2011 (Figure 2).

Figure 1.

Average incidence of patients undergoing single-level PLIF with/without rhBMP-2 from 2005 to 2011 (per 100 000 patients).

Figure 2.

Annual incidence of patients undergoing single-level PLIF with/without rhBMP-2 from 2005 to 2011 (per 100 000 patients).

Gender

In 2010, the incidence of patients undergoing single-level PLIF with rhBMP-2 decreased by 61.9% in males and decreased by 64.5% in females (Figure 3). The incidence of female patients undergoing single-level PLIF with/without rhBMP-2 (P < .05) was higher than male patients undergoing the same procedure (Figure 4).

Figure 3.

Annual incidence of patients undergoing single-level PLIF with/without rhBMP-2 by gender from 2005 to 2011 (per 100 000 patients).

Figure 4.

Average incidence of patients undergoing single-level PLIF with/without rhBMP-2 by gender from 2005 to 2011 (per 100 000 patients).

Region

The average rate of single-level PLIF with rhBMP-2 was the highest in West (30.1%), followed by Midwest (26.9%), South (20.5%), and Northeast (17.8%). The incidence of patients undergoing single-level PLIF with rhBMP-2 declined dramatically in 2010 in all regions, especially in the West (by 78.1%) and South (by 57.1%; Figure 5). Overall, the incidence of patients undergoing single-level PLIF with rhBMP-2 was the lowest in Northeast (0.08 cases per 100 000 patients) compared to South (0.25 cases per 100 000 patients, P < .001), Midwest (0.25 cases per 100 000 patients, P < .001), and West (0.22 cases per 100 000 patients, P = .002; Figure 6). As for the incidence of patients undergoing single-level PLIF without rhBMP-2, there were significant differences between Midwest and Northeast, Midwest and South, Northeast and South, and South and West (P < .01; Figure 6).

Figure 5.

Annual incidence of patients undergoing single-level PLIF with/without rhBMP-2 by region from 2005 to 2011 (per 100 000 patients).

Figure 6.

Average incidence of patients undergoing single-level PLIF with/without rhBMP-2 by region from 2005 to 2011 (per 100 000 patients). *P < .05, **P < .01.

Age

The highest incidence of single-level PLIF with rhBMP-2 (0.3 per 100 000 patients) was observed in the age group less than 65 years and was statistically significant compared to the 75 to 79, 80 to 84, more than 84 age groups (P < .05; Figure 7). The age group more than 84 years had the lowest incidence of single-level PLIF with rhBMP-2 (0.02 per 100 000 patients), which was significant compared to the age group 65 to 79 years, P < .001. The highest incidence of single-level PLIF without rhBMP-2 was observed in the age group less than 65 years (1.03 per 100 000 patients, P < .05). The age group more than 84 years had the lowest incidence of single-level PLIF without rhBMP-2 when compared to other age groups (0.07 per 100 000 patients, P < .05).

Figure 7.

Average incidence of patients undergoing single-level PLIF with/without rhBMP-2 by age from 2005 to 2011 (per 100 000 patients).

Discussion

In our study, the rate of rhBMP-2 utilization in single-level PLIF remained stable between 2005 and 2009, but sharply decreased in 2010 and 2011.

The utilization of rhBMP-2 in spine surgery gained momentum due to the advantage of inducing bone formation without the morbidity associated with iliac crest bone graft harvest. Those beneficial characteristics prompted off-label application of rhBMP-2 that led to certain complications.[10,13-15] At the same time, several studies reported nonsignificantly increased risk for cancer or no association after adjusting for comorbidities, demographics, and levels of procedure.[16,17] Reduction in rhBMP-2 use in our study is very similar to the previous report for posterior lumbar fusion[18,19] and could be potentially influenced by the 2008 Food and Drug Administration notification on rhBMP-2 use in cervical fusion.

The average rate of single-level PLIF with rhBMP-2 was the highest in West (30.1%), followed by Midwest (26.9%), South (20.5%), and Northeast (17.8%). Interestingly, the incidence of patients undergoing single-level PLIF with rhBMP-2 declined dramatically in 2010 in all regions, especially in West and South. Similarly, Singh and colleagues observed that for various fusion procedures the utilization rate of rhBMP-2 was the highest in South and the lowest in Northeast.[20]

We found the highest rhBMP-2 incidence in patients less than 65 years of age and the lowest in the age group more than 84 years of age. From 2005 to 2011, the overall incidence of single-level PLIF increased significantly in patients older than 65 years of age. These results may reflect the aging population in the United States; people older than 65 years of age are expected to represent nearly 20% of the population by 2030, compared to 12.4% in 2000.[20] As previously reported, PLIF has been increasingly used for degenerative lumbar disorders in the older patient population (>70 years of age), and the surgical outcomes were very similar to those in younger patients.[21-24]

There are several limitations to our study. This is a retrospective study of a large database searched by ICD-9 and CPT codes and therefore is subject to errors in coding. There are also inherent limitations with gathering information from the PearlDiver database. The patient population included in our study represents a relatively small number of orthopedic patients in the United States. Furthermore, the regional distribution of patients is uneven in the database due to the variations in regional penetration of insurance carriers. This may introduce bias regarding regional trends. Despite these limitations, we believe that the results in our study reflect the real situation of PLIF and rhBMP-2 use in the United States.

To our knowledge, this is the first study to report a trend of rhBMP-2 use in single-level PLIF. The incidence of rhBMP-2 utilization in single-level PLIF decreased in 2010 and 2011. The West region had the highest incidence of rhBMP-2 use.