SPINE Volume 39, Number 2, pp 172-176 ©2014, Lippincott Williams & Wilkins
SURGERY
Lumbar Clinical Adjacent Segment Pathology Predilection for Proximal Levels Paul C. Celestre, MD,* Scott R. Montgomery, MD,t Asher I. Kupperman, MD,+ Bayan Aghdasi, BS,t Hirokazu Inoue, MD,t and Jeffrey C. Wang, MD+
Study Design. Retrospective case series. Objective. To evaluate reoperations for lumbar adjacent segment pathology (ASP) during a 10-year period. Summary of Background Data. ASP after lumbar arthrodesis is an important clinical problem. There remains controversy, however, on the distribution of the most commonly affected levels. Methods. Thirty-one patients undergoing revision operation for ASP in the lumbar spine were included in this study. Patients' charts were evaluated for demographic data including age at index and revision operations, time to revision operation, and index and revision levels fused. Resuits. L4-L5 was the most commonly instrumented level in both single-level (n = 12), and multilevel (n = 13) index fusions. The mean length of time from the index operation to revision surgery was 81 months (range, 11-570 mo). Kaplan-Meier analysis predicted a disease-free survival rate of 32.3% at 5 years and of 12.9% at 10 years after the index operation. L3-L4 was the most commonly affected level by ASP with 75% (16/20) requiring reoperation. L2-L3 was the next most commonly affected level at 52% (14/27). The L5S1 disk was relatively protected from ASP, with only 4/1 7 (24%) disks at risk developing ASP. A subgroup analysis of patients undergoing revision after a single-level L4-L5 arthrodesis revealed ASP at L3-L4 in 83% (10/12) of patients, compared with only 3/12 (25%) at L5-S1 [P < 0.05). Of all cases of ASP, the proximal segments were involved 90% of the time. Conciusion. ASP most commonly affects proximal levels in the lumbar spine. In this cohort of patients undergoing revision fusion for ASP, 90% of affected levels were rostral to the index level(s).
From tbe 'Norton Leatherman Spine Center, Louisville, KY; +UCLA Department of Ortbopaedic Surgery, Los Angeles, CA; and iUSC Spine Center, Los Angeles, CA. Acknowledgment date: May 23, 2013. First revision date: August 30, 2013. Second revision date: October 10, 2013. Acceptance date: October 14, 2013. Tbe device(s)/drug(s) is/are FDA-approved or approved by corresponding national agency for this indication. No funds were received in support of this work. Relevant financial activities outside tbe submitted work: board membership, expert testimony patents, royalties, stocks, travel/accommodations/meeting expenses. Address correspondence and reprint requests to Jeffrey C. Wang, MD, USC Spine Center, 1520 San F^blo St, Suite 2000, Los Angeles, CA 90033; E-mail: [email protected] DOI: 10.1097/BRS.0000000000000094 172
www.spinejournal.com
In patients undergoing L4-L5 single-level arthrodesis, L3-L4 is at high risk, whereas L5-S1 is somewhat protected. Surgeons should pay particular attention to proximal levels when planning a lumbar arthrodesis, however, motion segments distal to fusion may not be as protected as previously thought. Key words: adjacent segment pathology, adjacent segment disease, adjacent segment degeneration, lumbar arthrodesis, lumbar fusion, spine, spondylosis.
Levei of Evidence: 4 Spine 2014;39:172-176
W
ith lumbar fusions being commonly performed as part of the treatment for a variety of spine conditions, the development and management of adjacent segment pathology (ASP) has become an important consideration for surgeons.''^ As the use of surgical arthrodesis has increased, there has been a focus on the effect of fusion on the health of adjacent intervertebral levels, or ASP.^ Lumbar clinical adjacent segment pathology (CASP) is defined as symptomatic spondylosis occurring within 2 motion segments of a prior arthrodesis. Symptoms of CASP include radiculopathy, neurogenic claudication, back pain, or any combination thereof. This is in distinction to radiographical adjacent segment pathology, defined as radiographical evidence of spondylosis occurring within 2 motion segments of a prior arthrodesis. Put more simply, CASP is a clinical problem whereas radiographical adjacent segment pathology is a radiographical finding. Although many think that ASP is the result of increased motion and stresses at disks adjacent to the level of fusion, there are also data that support degeneration of adjacent disks as the result of a natural degenerative process."*"^ Yet, multiple biomechanical and cadaveric studies have shown increased motion and increased intradiscal pressure at disks adjacent to the level of lumbar fusion,*"" and several clinical studies have demonstrated development of ASP after arthrodesis.'''"'* In their landmark article, Ghiselli et aï^'' defined the natural history of lumbar radiographical adjacent segment pathology and CASP after intertransverse fusion in 215 patients for degenerative conditions of the spine during a 10-year period. The authors found that CASP requiring reoperation developed at a relatively constant rate of 3.9% per year, with January 2014
Spine
SURGERY
16.5% and 36.1% of patients undergoing reoperation at 5 and 10 years, respectively. Furthermore, GASP of the L5-S1 motion segment was less frequent compared with L4-L5. In a separate article, Ghiselli et aF examined the fate of the L5-S1 disk in 32 patients undergoing L4-L5 single-level posterolateral fusion for degenerative spinal conditions. GASP requiring reoperation was found in only 1/32 (3%) patients at an average follow-up of 7.3 years. On the basis of the results of this study, the authors concluded that GASP requiring reoperation was rare at the L5-S1 disk after isolated L4-L5 posterolateral fusion, and that fusion for L5-S1 GASP was unnecessary in their case series. Similarly, in a study of 1069 patients undergoing instrumented lumbar or lumbosacral fusion with a minimum 1-year follow-up. Lee etaF found a 2.6% incidence of GASP, with only 1/21 (5%) patients undergoing reoperation for distal GASP. On the basis of the results of these studies, the L5-S1 disk was thought to be at low risk for GASP after posterolateral arthrodesis, and routine inclusion of the L5-S1 motion segment to avoid a "floating fusion" of L4-L5 was thought to be unnecessary. However, a paucity of data exists on this topic despite the great number of lumbar fusions performed each year in the United States. The purpose of this investigation was to evaluate patients undergoing revision surgery for lumbar GASP and to further elucidate the frequency of development of GASP, the relative location of affected levels, and the surgical procedure chosen for revision surgery.
MATERIALS AND METHODS The case log of the senior author was searched during a 10-year period (2002-2011) and all patients undergoing reoperation for lumbar GASP were identified. Institutional review board approval was granted to perform this retrospective review. All revision operations were performed at the same institution. To be included in the study, patients must have had a prior single- or multilevel posterior lumbar fusion, and have symptomatic spondylosis within 2 levels of the fused segment, both refractory to conservative management and requiring reoperation. Patients with any combination of radiculopathy, stenosis, or mechanical symptoms were included. Minimum follow-up was set at 1 year. Patients were excluded from the study if they had fusion into the cervical or thoracic spine, or multiple prior surgical procedures with the exception of a single-level laminotomy or microdiscectomy at the index fusion level. Patients with a history of pseudarthrosis or infection were also excluded from the study. Patients' charts were evaluated for demographic data including age at index and revision operations, time to revision operation, and index and revision levels fused. Analysis of variance was used to compare the diseased levels. A Kaplan-Meier survivorship analysis was performed to determine the time to reoperation for ASP. A separate subgroup analysis was performed on the 12 patients undergoing single-level L4-L5 fusion. Student t test was used to evaluate proximal versus distal ASP. Spine
Lumbar Clinical Adjacent Segment Pathology • Celestre et al
RESULTS Seventy-five patients were identified as undergoing reoperation for lumbar GASP. However, after application of strict exclusion criteria, 44 of these patients were excluded from the study, leaving a cohort of 31 patients. Demographic data including gender distribution, age at index operation, time to revision operation, and age at revision operation are presented in Table 1. Figure 1 is a graphical representation of the distribution of the index fusion levels. There were 14 single-level index operations and 17 multilevel index operations. L4-L5 was the most commonly fused level, both in single and multilevel procedures, with a total of 25 patients. This was followed by L5-S1, which was included in the index fusion in 14 patients, and L3-L4, which was fused in 8 patients during the index operation. Seventeen cases were instrumented posterior spinal fusions, 2 were noninstrumented posterior spinal fusions, 1 was an isolated anterior lumbar interbody fusion (AUF), 4 were posterior lumbar interbody fusions (PLIFs), 4 were transforaminal lumbar interbody fusions, 2 were ALIFs with posterior instrumented fusion, and 1 was an ALIF with noninstrumented posterior fusion. The total number of disks at risk (i.e., within 2 levels of the index fusion) at each level, stratified as affected or spared by ASP, is shown in Figure 2. The percentage of levels affected by GASP and thus requiring revision surgery is shown in Table 2. L3-L4 was the most commonly affected level, and of 20 disks at risk, 16 (75%) developed GASP. L2-L3 was the next most commonly affected level, with 14/27 levels (52%) undergoing reoperation for GASP. The L5-S1 disk was noticeably less affected by adjacent segment disease, with only 4/17 (24%) disks at risk undergoing revision operation. Overall, 90% of reoperation for GASP was performed at levels proximal to the index fusion, compared with only 10% at distal levels (P
SURGERY
Lumbar Clinical Adjacent Segment Pathology Predilection for Proximal Levels Paul C. Celestre, MD,* Scott R. Montgomery, MD,t Asher I. Kupperman, MD,+ Bayan Aghdasi, BS,t Hirokazu Inoue, MD,t and Jeffrey C. Wang, MD+
Study Design. Retrospective case series. Objective. To evaluate reoperations for lumbar adjacent segment pathology (ASP) during a 10-year period. Summary of Background Data. ASP after lumbar arthrodesis is an important clinical problem. There remains controversy, however, on the distribution of the most commonly affected levels. Methods. Thirty-one patients undergoing revision operation for ASP in the lumbar spine were included in this study. Patients' charts were evaluated for demographic data including age at index and revision operations, time to revision operation, and index and revision levels fused. Resuits. L4-L5 was the most commonly instrumented level in both single-level (n = 12), and multilevel (n = 13) index fusions. The mean length of time from the index operation to revision surgery was 81 months (range, 11-570 mo). Kaplan-Meier analysis predicted a disease-free survival rate of 32.3% at 5 years and of 12.9% at 10 years after the index operation. L3-L4 was the most commonly affected level by ASP with 75% (16/20) requiring reoperation. L2-L3 was the next most commonly affected level at 52% (14/27). The L5S1 disk was relatively protected from ASP, with only 4/1 7 (24%) disks at risk developing ASP. A subgroup analysis of patients undergoing revision after a single-level L4-L5 arthrodesis revealed ASP at L3-L4 in 83% (10/12) of patients, compared with only 3/12 (25%) at L5-S1 [P < 0.05). Of all cases of ASP, the proximal segments were involved 90% of the time. Conciusion. ASP most commonly affects proximal levels in the lumbar spine. In this cohort of patients undergoing revision fusion for ASP, 90% of affected levels were rostral to the index level(s).
From tbe 'Norton Leatherman Spine Center, Louisville, KY; +UCLA Department of Ortbopaedic Surgery, Los Angeles, CA; and iUSC Spine Center, Los Angeles, CA. Acknowledgment date: May 23, 2013. First revision date: August 30, 2013. Second revision date: October 10, 2013. Acceptance date: October 14, 2013. Tbe device(s)/drug(s) is/are FDA-approved or approved by corresponding national agency for this indication. No funds were received in support of this work. Relevant financial activities outside tbe submitted work: board membership, expert testimony patents, royalties, stocks, travel/accommodations/meeting expenses. Address correspondence and reprint requests to Jeffrey C. Wang, MD, USC Spine Center, 1520 San F^blo St, Suite 2000, Los Angeles, CA 90033; E-mail: [email protected] DOI: 10.1097/BRS.0000000000000094 172
www.spinejournal.com
In patients undergoing L4-L5 single-level arthrodesis, L3-L4 is at high risk, whereas L5-S1 is somewhat protected. Surgeons should pay particular attention to proximal levels when planning a lumbar arthrodesis, however, motion segments distal to fusion may not be as protected as previously thought. Key words: adjacent segment pathology, adjacent segment disease, adjacent segment degeneration, lumbar arthrodesis, lumbar fusion, spine, spondylosis.
Levei of Evidence: 4 Spine 2014;39:172-176
W
ith lumbar fusions being commonly performed as part of the treatment for a variety of spine conditions, the development and management of adjacent segment pathology (ASP) has become an important consideration for surgeons.''^ As the use of surgical arthrodesis has increased, there has been a focus on the effect of fusion on the health of adjacent intervertebral levels, or ASP.^ Lumbar clinical adjacent segment pathology (CASP) is defined as symptomatic spondylosis occurring within 2 motion segments of a prior arthrodesis. Symptoms of CASP include radiculopathy, neurogenic claudication, back pain, or any combination thereof. This is in distinction to radiographical adjacent segment pathology, defined as radiographical evidence of spondylosis occurring within 2 motion segments of a prior arthrodesis. Put more simply, CASP is a clinical problem whereas radiographical adjacent segment pathology is a radiographical finding. Although many think that ASP is the result of increased motion and stresses at disks adjacent to the level of fusion, there are also data that support degeneration of adjacent disks as the result of a natural degenerative process."*"^ Yet, multiple biomechanical and cadaveric studies have shown increased motion and increased intradiscal pressure at disks adjacent to the level of lumbar fusion,*"" and several clinical studies have demonstrated development of ASP after arthrodesis.'''"'* In their landmark article, Ghiselli et aï^'' defined the natural history of lumbar radiographical adjacent segment pathology and CASP after intertransverse fusion in 215 patients for degenerative conditions of the spine during a 10-year period. The authors found that CASP requiring reoperation developed at a relatively constant rate of 3.9% per year, with January 2014
Spine
SURGERY
16.5% and 36.1% of patients undergoing reoperation at 5 and 10 years, respectively. Furthermore, GASP of the L5-S1 motion segment was less frequent compared with L4-L5. In a separate article, Ghiselli et aF examined the fate of the L5-S1 disk in 32 patients undergoing L4-L5 single-level posterolateral fusion for degenerative spinal conditions. GASP requiring reoperation was found in only 1/32 (3%) patients at an average follow-up of 7.3 years. On the basis of the results of this study, the authors concluded that GASP requiring reoperation was rare at the L5-S1 disk after isolated L4-L5 posterolateral fusion, and that fusion for L5-S1 GASP was unnecessary in their case series. Similarly, in a study of 1069 patients undergoing instrumented lumbar or lumbosacral fusion with a minimum 1-year follow-up. Lee etaF found a 2.6% incidence of GASP, with only 1/21 (5%) patients undergoing reoperation for distal GASP. On the basis of the results of these studies, the L5-S1 disk was thought to be at low risk for GASP after posterolateral arthrodesis, and routine inclusion of the L5-S1 motion segment to avoid a "floating fusion" of L4-L5 was thought to be unnecessary. However, a paucity of data exists on this topic despite the great number of lumbar fusions performed each year in the United States. The purpose of this investigation was to evaluate patients undergoing revision surgery for lumbar GASP and to further elucidate the frequency of development of GASP, the relative location of affected levels, and the surgical procedure chosen for revision surgery.
MATERIALS AND METHODS The case log of the senior author was searched during a 10-year period (2002-2011) and all patients undergoing reoperation for lumbar GASP were identified. Institutional review board approval was granted to perform this retrospective review. All revision operations were performed at the same institution. To be included in the study, patients must have had a prior single- or multilevel posterior lumbar fusion, and have symptomatic spondylosis within 2 levels of the fused segment, both refractory to conservative management and requiring reoperation. Patients with any combination of radiculopathy, stenosis, or mechanical symptoms were included. Minimum follow-up was set at 1 year. Patients were excluded from the study if they had fusion into the cervical or thoracic spine, or multiple prior surgical procedures with the exception of a single-level laminotomy or microdiscectomy at the index fusion level. Patients with a history of pseudarthrosis or infection were also excluded from the study. Patients' charts were evaluated for demographic data including age at index and revision operations, time to revision operation, and index and revision levels fused. Analysis of variance was used to compare the diseased levels. A Kaplan-Meier survivorship analysis was performed to determine the time to reoperation for ASP. A separate subgroup analysis was performed on the 12 patients undergoing single-level L4-L5 fusion. Student t test was used to evaluate proximal versus distal ASP. Spine
Lumbar Clinical Adjacent Segment Pathology • Celestre et al
RESULTS Seventy-five patients were identified as undergoing reoperation for lumbar GASP. However, after application of strict exclusion criteria, 44 of these patients were excluded from the study, leaving a cohort of 31 patients. Demographic data including gender distribution, age at index operation, time to revision operation, and age at revision operation are presented in Table 1. Figure 1 is a graphical representation of the distribution of the index fusion levels. There were 14 single-level index operations and 17 multilevel index operations. L4-L5 was the most commonly fused level, both in single and multilevel procedures, with a total of 25 patients. This was followed by L5-S1, which was included in the index fusion in 14 patients, and L3-L4, which was fused in 8 patients during the index operation. Seventeen cases were instrumented posterior spinal fusions, 2 were noninstrumented posterior spinal fusions, 1 was an isolated anterior lumbar interbody fusion (AUF), 4 were posterior lumbar interbody fusions (PLIFs), 4 were transforaminal lumbar interbody fusions, 2 were ALIFs with posterior instrumented fusion, and 1 was an ALIF with noninstrumented posterior fusion. The total number of disks at risk (i.e., within 2 levels of the index fusion) at each level, stratified as affected or spared by ASP, is shown in Figure 2. The percentage of levels affected by GASP and thus requiring revision surgery is shown in Table 2. L3-L4 was the most commonly affected level, and of 20 disks at risk, 16 (75%) developed GASP. L2-L3 was the next most commonly affected level, with 14/27 levels (52%) undergoing reoperation for GASP. The L5-S1 disk was noticeably less affected by adjacent segment disease, with only 4/17 (24%) disks at risk undergoing revision operation. Overall, 90% of reoperation for GASP was performed at levels proximal to the index fusion, compared with only 10% at distal levels (P
