SPINE Volume 39, Number 12, pp E719-E727 ©2014, Lippincott Williams & Wilkins
SURGERY
Utilization of Spinal Cord Stimulation in Patients With Failed Back Surgery Syndrome Shivanand P. Lad, MD, PhD,* Ranjith Babu, MS,* Jacob H. Bagley, BS,* Jonathan Choi, MD,* Carlos A. Bagley, MD,* Billy K. Huh, MD, PhD,* Beatrice Ugiliweneza, PhD, MSPH,† Chirag G. Patil, MD,‡ and Maxwell Boakye, MD†
Study Design. Retrospective analysis of a population-based insurance claims data set. Objective. To evaluate the use of spinal cord stimulation (SCS) and lumbar reoperation for the treatment of failed back surgery syndrome (FBSS), and examine their associated complications and health care costs. Summary of Background Data. FBSS is a major source of chronic neuropathic pain and affects up to 40% of patients who undergo lumbosacral spine surgery for back pain. Thus far, few economic analyses have been performed comparing the various treatments for FBSS, with these studies involving small sample sizes. In addition, the nationwide practices in the use of SCS for FBSS are unknown. Methods. The MarketScan data set was used to analyze patients with FBSS who underwent SCS or spinal reoperation between 2000 and 2009. Propensity score methods were used to match patients who underwent SCS with those who underwent lumbar reoperation to examine health care resource utilization. Postoperative complications were analyzed with multivariate logistic regression. Health care use was analyzed using negative binomial and general linear models. Results. The study cohort included 16,455 patients with FBSS, with 395 undergoing SCS implantation (2.4%). Complication rates at 90 days were significantly lower for SCS than spinal reoperation (P < 0.0001). Also in the matched cohort, hospital stay (P < 0.0001) and associated charges (P = 0.016) were lower for patients with SCS. However outpatient, emergency room, and medication charges were similar between the 2 groups. Overall cost totaling $82,586 at
From the *Division of Neurosurgery, Department of Surgery, Duke University Medical Center, Durham, NC; †Department of Neurosurgery, University of Louisville, Louisville, KY; and ‡Department of Neurosurgery, Cedars-Sinai Medical Center, Los Angeles, CA. Acknowledgment date: July 4, 2012. Revision date: February 2, 2014. Acceptance date: February 25, 2014. The 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 the submitted work: grants/grants pending. Address correspondence and reprint requests to Shivanand P. Lad, MD, PhD, Division of Neurosurgery, Department of Surgery, Duke University Medical Center, Box 3807, Durham, NC 27710; E-mail: [email protected] DOI: 10.1097/BRS.0000000000000320 Spine
2 years was slightly higher in the lumbar reoperation group than in the SCS group with total cost of $80,669 (P = 0.88). Conclusion. Although previous studies have demonstrated superior efficacy for the treatment of FBSS, SCS remains underused. Despite no significant decreases in overall health care cost with SCS implantation, because it is associated with decreased complications and improved outcomes, this technology warrants closer consideration for the management of chronic pain in patients with FBSS. Key words: complications, failed back surgery syndrome, health care cost, lumbar reoperation, spinal cord stimulation. Level of Evidence: 4 Spine 2014;39:E719–E727
F
ailed back surgery syndrome (FBSS) is a major source of chronic neuropathic pain, estimated to affect 5% to 40% of all patients who undergo lumbosacral spine surgery for back pain.1–4 FBSS often results in significant longterm disability, and its treatment results in high health care costs. Because of the rising rates of spine surgery, FBSS is a significant public health problem, serving as a major contributor to the estimated $19.8 billion dollars of indirect costs of back pain.5 Despite its frequency, there are no clear management guidelines due to its complexity and diverse etiology.6 Because there is no “gold standard” for treatment, FBSS is typically first treated with conventional medical management (CMM), which includes analgesics and antidepressants, as well as physical and psychosocial therapy. Other nonsurgical interventions may also involve epidural injections, percutaneous epidural adhesiolysis, transcutaneous electrical nerve stimulation, and medial branch blocks. If these therapies fail, patients may undergo repeat operation or spinal cord stimulation (SCS). However, repeat surgery is associated with increased patient morbidity, health care costs, and poor success rates, which range from 22% to 40%.1,4,7 As a result, there has been an increased interest in the use of SCS due to its reversibility, lower morbidity, and increased efficacy. Several randomized controlled trials have demonstrated superior pain relief after treatment with SCS than both CMM and reoperation.8–10 Because of rising health care costs, it is increasingly important to identify appropriate means for disease management www.spinejournal.com
E719
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE120939_LR E719
24/04/14 3:49 AM
SURGERY and treatment while providing effective care. Few economic analyses have been performed comparing the various treatments for FBSS, and these studies typically involved small sample sizes.11–15 In addition, despite favorable data for the use of SCS in FBSS, nationwide practices in the use of this technology are unknown. In this study, we examined a nationally selected cohort of patients diagnosed with FBSS to evaluate the use of SCS and lumbar reoperation, the associated procedural complications, and long-term health care costs.
MATERIALS AND METHODS Data We used the MarketScan commercial Claims and Encounters, Medicare Supplemental and Medicare database records from 2000 to 2009. The MarketScan data contains claims records from employers, health plans, government, and public organization. The databases contain person-specific clinical use and cost as well as enrollment information with more than 158 million unique patients since 1996.16 Within each database of the MarketScan, data are grouped into different tables representing the type of claim. For this project, we used the inpatient tables, the outpatient tables, the prescription drugs tables and the enrollment tables. Each patient in MarketScan is assigned a unique encrypted enrollee ID that can be used to link the different tables. This study has been approved by institutional review board.
Patient Selection Data Preprocessing To form the comparative groups, we queried both the inpatient and the outpatient tables. In the MarketScan, the inpatient tables’ rows group all the diagnoses and all the services provided in a stay. In the outpatient tables, each service is listed in separate rows. Thus, we have many services occurring on the same day but listed in different rows. First, we
Utilization of SCS in FBSS • Lad et al
collapsed all the services occurring on the same day in one row for each patient. Then, we merged the inpatient and the outpatient tables by enrollee ID and by date of service and we created an indicator for the source of each row data. Indexing the First Day of Follow-up In the resulting data set, we extracted cases in which a lumbar surgery or an implantation or replacement of spinal neurostimulator lead was performed for FBSS or postlaminectomy pain syndrome. For this query, we used the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis and procedure codes and the Current Procedural Terminology, Fourth Edition (CPT-4) procedure codes. For patients in our lumbar surgery cohort, we queried all inpatient stays for patients receiving a laminectomy or hemilaminectomy with or without simultaneous facetectomy or foraminotomy (ICD-9-CM: 03.09; CPT-4: 63005, 63012, 63030, 63042, 63044, 63047, 63056, 63035, 63048, and 63057). In addition, we included patients who received a lumbar fusion (ICD-9-CM: 81.06, 81.07, and 81.08; CPT-4: 22558, 22630, and 22612). For implantation or replacement of spinal neurostimulator lead, we used ICD-9-CM code 03.93 and CPT-4 codes 63650 and 63655 with the insertion or replacement of single/dual/rechargeable/nonrechargeable neurostimulator generator (ICD-9-CM: 86.94–86.98; CPT-4: 63685). The diagnosis of FBSS was recognized by the ICD9-CM code 225.1 and postlaminectomy pain syndrome by ICD-9-CM code 722.83. Among the cases retained by the above algorithm, we excluded those in which the patient was younger than 18 years. For each patient, the first occurring record satisfying all the conditions listed in the algorithm was indexed to be the first day of follow-up. Figure 1 illustrates the patient selection algorithm. Extraction of Postoperative Data Using the enrollee ID, we extracted all the 2000–2009 inpatient, outpatient, prescription drugs, and enrollment records
Figure 1. Flowchart illustrating the patient selection algorithm. FBSS indicates failed back surgery syndrome; SCS, spinal cord stimulation.
E720
www.spinejournal.com
May 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE120939_LR E720
24/04/14 3:49 AM
SURGERY for the retained patients. For the analysis, we only considered the index first day of follow-up record and subsequent records. Calculation of Postoperative Continuous Enrollment Follow-up Time The postoperative continuous enrollment follow-up time was calculated as the difference between the end enrollment date and the index date. For patients without the end enrollment date, this follow-up time was calculated as the difference of December 31, 2009 (last day of our data) and the index date. We formed 2 cohorts for our analysis, the cohort of all patients and the cohorts of patients with at least 2-year postoperative data.
Variables Independent Variables We considered the sex, insurance (commercial, Medicaid, Medicare), Charlson index, and surgery type (SCS, lumbar reoperation) at the time of the index procedure to be the independent variables for our analyses. Dependent Variables The dependent variables were: (1) reoperation, the occurrence of either SCS revision or removal, fusion, revision fusion, or laminectomy during postoperative follow-up period; (2) complications, the presence of any of renal, cardiac, neurological, deep vein thrombosis/pulmonary embolism (DVT/PE), pulmonary, infection, or wound complications in any of the postoperative claims; (3) health care resources use, hospital days, outpatient services, and prescription medications occurring after the index operation; and (4) health care resources cost, the total cost associated with the health care resources.
Propensity Score Matching We used propensity score methods to match patients who underwent SCS with patients who underwent lumbar reoperation. The propensity score was calculated using a multivariate logistic regression with surgery type as the outcome and age, sex, year of surgery, Charlson index, postoperative follow-up time and insurance as predictors. We used the SAS macro with greedy algorithm.
Statistical Analysis Data were summarized using means and standard deviations for continuous variables, count and percentages for categorical variables. Group characteristics and demographics were compared with Mann Whitney U test for continuous variables and χ2 test or Fisher exact test for categorical variables. Dichotomous outcomes (reoperation, complications) were analyzed with multivariate logistic regression, count outcomes (number of postoperative outpatient services, and medications) were analyzed with negative binomial models and continuous variables (hospital days, health care resources cost) were analyzed with general linear models. All tests were Spine
Utilization of SCS in FBSS • Lad et al
2-sided and a P < 0.05 was considered statistically significant. We used SAS 9.3 (Cary, NC) for all statistical analyses.
RESULTS Patient Cohort Characteristics A total of 16,455 patients with FBSS met our inclusion criteria, with a total of 395 patients undergoing SCS implantation (2.4%) and the remaining 16,060 undergoing spinal reoperation (97.6%) (Table 1). Of these patients, 6497 (6386, lumbar surgery; and 111, SCS) had at least 2 years of postoperative continuous enrollment. In the propensity score–matched cohort, the characteristics of the 111 patients who underwent SCS were matched to those of the 111 patients who underwent lumbar surgery. In the entire cohort, the mean age of patients undergoing lumbar reoperation or SCS for FBSS was similar (54 ± 13 vs. 54 ± 12 yr, P = 0.28) (Table 1). However, the proportion of females who underwent SCS (63.8%) was significantly higher than those who underwent lumbar surgery (55.3%) (P = 0.0008). Patients who underwent SCS also had more comorbidities, with greater Charlson indices than those in the lumbar surgery group (SCS: 22.3% with a Charlson score greater than zero vs. lumbar surgery: 16.2%, P = 0.0003). Those with Commercial and Medicare insurance were more likely to undergo lumbar reoperation, with significantly more Medicaid patients undergoing SCS implantation (19.0% vs. 7.5%).
Complications The incidence of postoperative complications during the index hospitalization was significantly higher in those who underwent lumbar reoperation (11.7%) than patients who underwent SCS (5.1%) (P < 0.0001) (Table 2). Analysis of complication type only revealed a significant difference in the rate of postoperative wound complications, with those who underwent lumbar reoperation having a significantly higher incidence (5.8% vs. 1.3%, P = 0.0001). Even at 90-day follow-up, those in the lumbar reoperation group experienced complications at more than 2 times the rate as those in the SCS group (14.4% vs. 6.5%, P < 0.0001). Evaluation of mortality during the index hospitalization did not reveal a significant difference between those who underwent lumbar reoperation or SCS, although all instances of mortality occurred in the lumbar surgery group (0.12% vs. 0.0%, P = 0.46).
Health Care Resource Use Analysis of unmatched patients revealed those who underwent lumbar reoperation to have a hospital stay twice as long as those who underwent SCS implantation (4 ± 4 vs. 2 ± 1 days, P < 0.0001) (Table 3). However, this increase in stay did not result in significant differences in the cost of the index hospitalization between the groups ($36,371 ± $39,305 [lumbar reoperation] vs. $31,210 ± $27,327 [SCS], P = 0.43). There were no significant differences in hospital days or associated cost at the end of 2 years, with hospital charges totaling www.spinejournal.com
E721
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE120939_LR E721
24/04/14 3:49 AM
SURGERY
Utilization of SCS in FBSS • Lad et al
TABLE 1. Characteristics and Outcomes for All Patients Characteristics at the Time of Procedure
Lumbar Surgery (n = 16,060)
SCS (n = 395)
P
54 (13)
54 (12)
0.2756
777 (721)
540 (461)
SURGERY
Utilization of Spinal Cord Stimulation in Patients With Failed Back Surgery Syndrome Shivanand P. Lad, MD, PhD,* Ranjith Babu, MS,* Jacob H. Bagley, BS,* Jonathan Choi, MD,* Carlos A. Bagley, MD,* Billy K. Huh, MD, PhD,* Beatrice Ugiliweneza, PhD, MSPH,† Chirag G. Patil, MD,‡ and Maxwell Boakye, MD†
Study Design. Retrospective analysis of a population-based insurance claims data set. Objective. To evaluate the use of spinal cord stimulation (SCS) and lumbar reoperation for the treatment of failed back surgery syndrome (FBSS), and examine their associated complications and health care costs. Summary of Background Data. FBSS is a major source of chronic neuropathic pain and affects up to 40% of patients who undergo lumbosacral spine surgery for back pain. Thus far, few economic analyses have been performed comparing the various treatments for FBSS, with these studies involving small sample sizes. In addition, the nationwide practices in the use of SCS for FBSS are unknown. Methods. The MarketScan data set was used to analyze patients with FBSS who underwent SCS or spinal reoperation between 2000 and 2009. Propensity score methods were used to match patients who underwent SCS with those who underwent lumbar reoperation to examine health care resource utilization. Postoperative complications were analyzed with multivariate logistic regression. Health care use was analyzed using negative binomial and general linear models. Results. The study cohort included 16,455 patients with FBSS, with 395 undergoing SCS implantation (2.4%). Complication rates at 90 days were significantly lower for SCS than spinal reoperation (P < 0.0001). Also in the matched cohort, hospital stay (P < 0.0001) and associated charges (P = 0.016) were lower for patients with SCS. However outpatient, emergency room, and medication charges were similar between the 2 groups. Overall cost totaling $82,586 at
From the *Division of Neurosurgery, Department of Surgery, Duke University Medical Center, Durham, NC; †Department of Neurosurgery, University of Louisville, Louisville, KY; and ‡Department of Neurosurgery, Cedars-Sinai Medical Center, Los Angeles, CA. Acknowledgment date: July 4, 2012. Revision date: February 2, 2014. Acceptance date: February 25, 2014. The 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 the submitted work: grants/grants pending. Address correspondence and reprint requests to Shivanand P. Lad, MD, PhD, Division of Neurosurgery, Department of Surgery, Duke University Medical Center, Box 3807, Durham, NC 27710; E-mail: [email protected] DOI: 10.1097/BRS.0000000000000320 Spine
2 years was slightly higher in the lumbar reoperation group than in the SCS group with total cost of $80,669 (P = 0.88). Conclusion. Although previous studies have demonstrated superior efficacy for the treatment of FBSS, SCS remains underused. Despite no significant decreases in overall health care cost with SCS implantation, because it is associated with decreased complications and improved outcomes, this technology warrants closer consideration for the management of chronic pain in patients with FBSS. Key words: complications, failed back surgery syndrome, health care cost, lumbar reoperation, spinal cord stimulation. Level of Evidence: 4 Spine 2014;39:E719–E727
F
ailed back surgery syndrome (FBSS) is a major source of chronic neuropathic pain, estimated to affect 5% to 40% of all patients who undergo lumbosacral spine surgery for back pain.1–4 FBSS often results in significant longterm disability, and its treatment results in high health care costs. Because of the rising rates of spine surgery, FBSS is a significant public health problem, serving as a major contributor to the estimated $19.8 billion dollars of indirect costs of back pain.5 Despite its frequency, there are no clear management guidelines due to its complexity and diverse etiology.6 Because there is no “gold standard” for treatment, FBSS is typically first treated with conventional medical management (CMM), which includes analgesics and antidepressants, as well as physical and psychosocial therapy. Other nonsurgical interventions may also involve epidural injections, percutaneous epidural adhesiolysis, transcutaneous electrical nerve stimulation, and medial branch blocks. If these therapies fail, patients may undergo repeat operation or spinal cord stimulation (SCS). However, repeat surgery is associated with increased patient morbidity, health care costs, and poor success rates, which range from 22% to 40%.1,4,7 As a result, there has been an increased interest in the use of SCS due to its reversibility, lower morbidity, and increased efficacy. Several randomized controlled trials have demonstrated superior pain relief after treatment with SCS than both CMM and reoperation.8–10 Because of rising health care costs, it is increasingly important to identify appropriate means for disease management www.spinejournal.com
E719
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE120939_LR E719
24/04/14 3:49 AM
SURGERY and treatment while providing effective care. Few economic analyses have been performed comparing the various treatments for FBSS, and these studies typically involved small sample sizes.11–15 In addition, despite favorable data for the use of SCS in FBSS, nationwide practices in the use of this technology are unknown. In this study, we examined a nationally selected cohort of patients diagnosed with FBSS to evaluate the use of SCS and lumbar reoperation, the associated procedural complications, and long-term health care costs.
MATERIALS AND METHODS Data We used the MarketScan commercial Claims and Encounters, Medicare Supplemental and Medicare database records from 2000 to 2009. The MarketScan data contains claims records from employers, health plans, government, and public organization. The databases contain person-specific clinical use and cost as well as enrollment information with more than 158 million unique patients since 1996.16 Within each database of the MarketScan, data are grouped into different tables representing the type of claim. For this project, we used the inpatient tables, the outpatient tables, the prescription drugs tables and the enrollment tables. Each patient in MarketScan is assigned a unique encrypted enrollee ID that can be used to link the different tables. This study has been approved by institutional review board.
Patient Selection Data Preprocessing To form the comparative groups, we queried both the inpatient and the outpatient tables. In the MarketScan, the inpatient tables’ rows group all the diagnoses and all the services provided in a stay. In the outpatient tables, each service is listed in separate rows. Thus, we have many services occurring on the same day but listed in different rows. First, we
Utilization of SCS in FBSS • Lad et al
collapsed all the services occurring on the same day in one row for each patient. Then, we merged the inpatient and the outpatient tables by enrollee ID and by date of service and we created an indicator for the source of each row data. Indexing the First Day of Follow-up In the resulting data set, we extracted cases in which a lumbar surgery or an implantation or replacement of spinal neurostimulator lead was performed for FBSS or postlaminectomy pain syndrome. For this query, we used the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis and procedure codes and the Current Procedural Terminology, Fourth Edition (CPT-4) procedure codes. For patients in our lumbar surgery cohort, we queried all inpatient stays for patients receiving a laminectomy or hemilaminectomy with or without simultaneous facetectomy or foraminotomy (ICD-9-CM: 03.09; CPT-4: 63005, 63012, 63030, 63042, 63044, 63047, 63056, 63035, 63048, and 63057). In addition, we included patients who received a lumbar fusion (ICD-9-CM: 81.06, 81.07, and 81.08; CPT-4: 22558, 22630, and 22612). For implantation or replacement of spinal neurostimulator lead, we used ICD-9-CM code 03.93 and CPT-4 codes 63650 and 63655 with the insertion or replacement of single/dual/rechargeable/nonrechargeable neurostimulator generator (ICD-9-CM: 86.94–86.98; CPT-4: 63685). The diagnosis of FBSS was recognized by the ICD9-CM code 225.1 and postlaminectomy pain syndrome by ICD-9-CM code 722.83. Among the cases retained by the above algorithm, we excluded those in which the patient was younger than 18 years. For each patient, the first occurring record satisfying all the conditions listed in the algorithm was indexed to be the first day of follow-up. Figure 1 illustrates the patient selection algorithm. Extraction of Postoperative Data Using the enrollee ID, we extracted all the 2000–2009 inpatient, outpatient, prescription drugs, and enrollment records
Figure 1. Flowchart illustrating the patient selection algorithm. FBSS indicates failed back surgery syndrome; SCS, spinal cord stimulation.
E720
www.spinejournal.com
May 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE120939_LR E720
24/04/14 3:49 AM
SURGERY for the retained patients. For the analysis, we only considered the index first day of follow-up record and subsequent records. Calculation of Postoperative Continuous Enrollment Follow-up Time The postoperative continuous enrollment follow-up time was calculated as the difference between the end enrollment date and the index date. For patients without the end enrollment date, this follow-up time was calculated as the difference of December 31, 2009 (last day of our data) and the index date. We formed 2 cohorts for our analysis, the cohort of all patients and the cohorts of patients with at least 2-year postoperative data.
Variables Independent Variables We considered the sex, insurance (commercial, Medicaid, Medicare), Charlson index, and surgery type (SCS, lumbar reoperation) at the time of the index procedure to be the independent variables for our analyses. Dependent Variables The dependent variables were: (1) reoperation, the occurrence of either SCS revision or removal, fusion, revision fusion, or laminectomy during postoperative follow-up period; (2) complications, the presence of any of renal, cardiac, neurological, deep vein thrombosis/pulmonary embolism (DVT/PE), pulmonary, infection, or wound complications in any of the postoperative claims; (3) health care resources use, hospital days, outpatient services, and prescription medications occurring after the index operation; and (4) health care resources cost, the total cost associated with the health care resources.
Propensity Score Matching We used propensity score methods to match patients who underwent SCS with patients who underwent lumbar reoperation. The propensity score was calculated using a multivariate logistic regression with surgery type as the outcome and age, sex, year of surgery, Charlson index, postoperative follow-up time and insurance as predictors. We used the SAS macro with greedy algorithm.
Statistical Analysis Data were summarized using means and standard deviations for continuous variables, count and percentages for categorical variables. Group characteristics and demographics were compared with Mann Whitney U test for continuous variables and χ2 test or Fisher exact test for categorical variables. Dichotomous outcomes (reoperation, complications) were analyzed with multivariate logistic regression, count outcomes (number of postoperative outpatient services, and medications) were analyzed with negative binomial models and continuous variables (hospital days, health care resources cost) were analyzed with general linear models. All tests were Spine
Utilization of SCS in FBSS • Lad et al
2-sided and a P < 0.05 was considered statistically significant. We used SAS 9.3 (Cary, NC) for all statistical analyses.
RESULTS Patient Cohort Characteristics A total of 16,455 patients with FBSS met our inclusion criteria, with a total of 395 patients undergoing SCS implantation (2.4%) and the remaining 16,060 undergoing spinal reoperation (97.6%) (Table 1). Of these patients, 6497 (6386, lumbar surgery; and 111, SCS) had at least 2 years of postoperative continuous enrollment. In the propensity score–matched cohort, the characteristics of the 111 patients who underwent SCS were matched to those of the 111 patients who underwent lumbar surgery. In the entire cohort, the mean age of patients undergoing lumbar reoperation or SCS for FBSS was similar (54 ± 13 vs. 54 ± 12 yr, P = 0.28) (Table 1). However, the proportion of females who underwent SCS (63.8%) was significantly higher than those who underwent lumbar surgery (55.3%) (P = 0.0008). Patients who underwent SCS also had more comorbidities, with greater Charlson indices than those in the lumbar surgery group (SCS: 22.3% with a Charlson score greater than zero vs. lumbar surgery: 16.2%, P = 0.0003). Those with Commercial and Medicare insurance were more likely to undergo lumbar reoperation, with significantly more Medicaid patients undergoing SCS implantation (19.0% vs. 7.5%).
Complications The incidence of postoperative complications during the index hospitalization was significantly higher in those who underwent lumbar reoperation (11.7%) than patients who underwent SCS (5.1%) (P < 0.0001) (Table 2). Analysis of complication type only revealed a significant difference in the rate of postoperative wound complications, with those who underwent lumbar reoperation having a significantly higher incidence (5.8% vs. 1.3%, P = 0.0001). Even at 90-day follow-up, those in the lumbar reoperation group experienced complications at more than 2 times the rate as those in the SCS group (14.4% vs. 6.5%, P < 0.0001). Evaluation of mortality during the index hospitalization did not reveal a significant difference between those who underwent lumbar reoperation or SCS, although all instances of mortality occurred in the lumbar surgery group (0.12% vs. 0.0%, P = 0.46).
Health Care Resource Use Analysis of unmatched patients revealed those who underwent lumbar reoperation to have a hospital stay twice as long as those who underwent SCS implantation (4 ± 4 vs. 2 ± 1 days, P < 0.0001) (Table 3). However, this increase in stay did not result in significant differences in the cost of the index hospitalization between the groups ($36,371 ± $39,305 [lumbar reoperation] vs. $31,210 ± $27,327 [SCS], P = 0.43). There were no significant differences in hospital days or associated cost at the end of 2 years, with hospital charges totaling www.spinejournal.com
E721
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE120939_LR E721
24/04/14 3:49 AM
SURGERY
Utilization of SCS in FBSS • Lad et al
TABLE 1. Characteristics and Outcomes for All Patients Characteristics at the Time of Procedure
Lumbar Surgery (n = 16,060)
SCS (n = 395)
P
54 (13)
54 (12)
0.2756
777 (721)
540 (461)
