Long-Term Outcomes of Surgical and Nonsurgical Management of ...

SPINE Volume 30, Number 8, pp 927–935 ©2005, Lippincott Williams & Wilkins, Inc.

Long-Term Outcomes of Surgical and Nonsurgical Management of Sciatica Secondary to a Lumbar Disc Herniation: 10 Year Results from the Maine Lumbar Spine Study Steven J. Atlas, MD, MPH,* Robert B. Keller, MD,† Yen A. Wu, MPH,* Richard A. Deyo, MD, MPH,‡ and Daniel E. Singer, MD*

Study Design. A prospective cohort study. Objective. To assess 10-year outcomes of patients with sciatica resulting from a lumbar disc herniation treated surgically or nonsurgically. Summary of Background Data. There is little information comparing long-term outcomes of surgical and conservative therapy of lumbar disc herniation in contemporary clinical practice. Prior studies suggest that these outcomes are similar. Methods. Patients recruited from the practices of orthopedic surgeons, neurosurgeons, and occupational medicine physicians throughout Maine had baseline interviews with follow-up questionnaires mailed at regular intervals over 10 years. Clinical data were obtained at baseline from a physician questionnaire. Primary analyses were based on initial treatment received, either surgical or nonsurgical. Secondary analyses examined actual treatments received by 10 years. Outcomes included patient-reported symptoms of leg and back pain, functional status, satisfaction, and work and disability compensation status. Results. Of 507 eligible consenting patients initially enrolled, 10-year outcomes were available for 400 of 477 (84%) surviving patients; 217 of 255 (85%) treated surgically, and 183 of 222 (82%) treated nonsurgically. Patients undergoing surgery had worse baseline symptoms and functional status than those initially treated nonsurgically. By 10 years, 25% of surgical patients had undergone at least one additional lumbar spine operation, and 25% of nonsurgical patients had at least one lumbar spine operation. At 10-year follow-up, 69% of patients initially treated surgically reported improvement in their predominant symptom (back or leg pain) versus 61% of those initially treated nonsurgically (P ⫽ 0.2). A larger proportion of surgical patients reported that their low back and

From the *General Medicine Division and the Clinical Epidemiology Unit, Medical Services, Massachusetts General Hospital, Harvard Medical School, Boston, MA; †the Center for the Evaluative Clinical Sciences, Dartmouth Medical School, Hanover, NH; and ‡the Center for Cost and Outcomes Research and the Departments of Medicine and Health Services, University of Washington, Seattle, WA. Acknowledgment date: March 8, 2004. First revision date: May 14, 2004. Acceptance date: May 27, 2004. Supported by grants from the Agency for Healthcare Research and Quality (HS-06344, HS-08194, and HS-09804). The manuscript submitted does not contain information about medical device(s)/drug(s). Federal funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. Address correspondence and reprint requests to Steven J. Atlas, MD, General Medicine Division, Massachusetts General Hospital, 50 Staniford Street, Boston, MA 02114. E-mail [email protected]

leg pain were much better or completely gone (56% vs. 40%, P ⫽ 0.006) and were more satisfied with their current status (71% vs. 56%, P ⫽ 0.002). Treatment group differences persisted after adjustment for other determinants of outcome in multivariate models. Change in the modified Roland back-specific functional status scale favored surgical treatment, and the relative benefit persisted over the follow-up period. Despite these differences, work and disability status at 10 years were comparable among those treated surgically or nonsurgically. Conclusions. Surgically treated patients with a herniated lumbar disc had more complete relief of leg pain and improved function and satisfaction compared with nonsurgically treated patients over 10 years. Nevertheless, improvement in the patient’s predominant symptom and work and disability outcomes were similar regardless of treatment received. For patients in whom elective discectomy is a treatment option, an individualized treatment plan requires patients and their physicians to integrate clinical findings with patient preferences based on their symptoms and goals. Key words: sciatica, herniated lumbar disc, prospective cohort study, outcome research, lumbar disc surgery, natural history. Spine 2005;30:927–935

Sciatica resulting from a lumbar intervertebral disc herniation is the most common cause of radicular leg pain in adult working populations.1 Such patients have a favorable natural history associated with resorption of extruded disc material, but surgical treatment is frequently considered and performed in those with persistent or severe symptoms.2 Indeed, discectomy is the most frequent lumbar spine operation in the United States. The likelihood that a patient with an intervertebral disc herniation will undergo surgery varies widely across geographic regions.3–7 This may reflect different opinions among healthcare providers about the relative benefits of surgical and nonsurgical treatment.4 The primary evidence for the relative benefit of surgical over nonsurgical treatment comes from one randomized clinical trial performed over 30 years ago. That study was conducted among 126 inpatients at a single referral hospital in Norway.8 It reported that surgery was superior at 1 year and nonsignificantly better at 4 years compared with nonsurgical treatment. However, after 10 years, treatment outcomes were similar. No comparable data are available from patients treated in the United States, but a large multicenter randomized trial is currently ongoing.9 For individuals 927

928 Spine • Volume 30 • Number 8 • 2005

treated in contemporary clinical practice, the only prospective comparison of surgical and nonsurgical treatment is from an observational study.10 The Maine Lumbar Spine Study has previously reported results after 1 and 5 years11,12 that were similar to those from the Weber study.8 The goal of the current study was to assess the relative benefits of surgical and nonsurgical treatment over a 10-year follow-up period using a broad range of validated patient-reported outcome measures. Materials and Methods Details about the Maine Lumbar Spine Study design and methods, and 1- and 5-year outcomes for patients with sciatica due to an intervertebral disc herniation have been previously published.10 –12 The study prospectively followed patients treated by orthopedic surgeons, neurosurgeons, and occupational medicine physicians in community-based practices throughout the state of Maine. Treatment, either surgical or nonsurgical care, was determined in a routine clinical manner by the patient and the physician. Almost all patients undergoing surgery had an open discectomy to remove an extruded disc fragment or herniation.11 Percutaneous discectomies were uncommon (4.3%), and no patient underwent fusion at the initial operation. For nonsurgically treated patients, back exercises, physical therapy, bedrest, spinal manipulation, narcotic analgesics, and epidural steroids were most frequently used, with most patients receiving several treatments.11

Study Population. The diagnosis of sciatica resulting from an intervertebral disc herniation was based on the physician’s assessment. Specific radiographic findings were not required for study entry. To restrict the study to patients for whom surgery would be elective and acceptably safe, patients were excluded if they had prior lumbar spine surgery, cauda equina syndrome, developmental spine deformities, vertebral fractures, spine infection or tumor, inflammatory spondylopathy, pregnancy, or severe comorbid conditions. A total of 507 patients were enrolled from 1990 to 1992, with enrollment stratified to obtain roughly equal numbers of surgical and nonsurgical patients.10 Patients who initially chose nonsurgical treatment but underwent surgery before the first follow-up evaluation at 3 months were included in primary analyses as having been surgically treated (n ⫽ 38, 14%).11,12

Study Protocol. For eligible patients, written informed consent was obtained at study entry. Baseline interviews were conducted in person by trained research assistants. Follow-up was obtained by mailed questionnaires at 3, 6, and 12 months, and then yearly through 10 years. Physicians completed a detailed baseline questionnaire including history, physical and neurologic findings, diagnostic procedure results, and planned treatment. Baseline imaging studies (CT, MRI, or myelogram) were ordered as directed by the treating physicians, and about half were available for independent review by a neuroradiologist.11 All study activities were approved by the Institutional Review Boards at the University of Washington, Seattle, ME Medical Center, Portland, ME, and Massachusetts General Hospital, Boston, MA. Baseline Variables and Findings. Baseline findings, including demographic information, employment and disability status, comorbid conditions, past spine history, physical exami-

Table 1. Patient Characteristics and Features of Back Disorder at Baseline Evaluation Characteristic* Age (yr) 关mean (SD)兴 Gender, male Education, college graduate Smoking status, never or quit ⬎ 6 mo ago Comorbid illnesses, yes‡ Employed in past 4 wk Receiving or applying for Workers’ Compensation Past episodes of back pain, none Positive straight leg raise test Abnormal examination findings 关mean (SD)兴 § Radiographic image reviewed (n) Moderate or severe findings# Quebec classification, category 4 or 6** Length of current episode, ⬍ 6 mo Unilateral leg pain, yes SF-36 score (0–100), mean (SD) †† Physical function Bodily pain Role-emotional Mental health General health

Surgical (n ⫽ 217)

Nonsurgical (n ⫽ 183)

42.2 (11.0) 63.6 34.1 59.3

41.8 (11.7) 58.5 26.8 63.9

26.7 59.9 29.2

21.9 50.3 48.6

0.26 0.05 ⬍0.001

20.7

23.0

0.59

72.8 1.1 (0.9)

47.8 1.4 (1.0)

115

P† 0.77 0.29 0.11 0.34

⬍0.001 ⬍0.001

75

82.6 49.1

61.3 30.2

0.001 ⬍0.001

70.5

70.9

0.94

85.3

71.1

⬍0.001

31.4 (25.5) 19.2 (15.6) 48.2 (43.3) 61.4 (18.6) 79.1 (17.3)

46.7 (26.5) 35.5 (23.6) 57.4 (43.7) 66.6 (20.1) 72.8 (19.1)

⬍0.001 ⬍0.001 0.04 0.009 ⬍0.001

*The data are expressed as the percentage, except as noted. Denominators differ slightly among variables because not all patients answered each question on the survey. †P values compare surgical and nonsurgical treatment groups using Fisher’s exact test or t test. ‡Any self-reported chronic pulmonary disease, heart disease, stroke, cancer, or diabetes. §The mean no. of positive physical examination findings for a patient, including unilateral strength, sensation, or reflex abnormality (range, 0 –3 findings). ¶Any computerized tomography, magnetic resonance imaging, or myelogram available for independent review. #Global rating from normal to severe by study neuroradiologist blinded to treatment group and clinical information. **The Quebec classification system defines 11 categories including sciatica with distal extremity radiation and neurologic findings (category 4), or the above with radiographic findings of nerve root compression (category 6). ††Higher scores indicate better function.

nation and imaging findings, symptoms, and functional status, have been reported previously.11,12 Patients initially receiving surgical or nonsurgical treatment were similarly young, healthy, predominantly male, had not graduated from college, had been employed in the past month, and had symptoms for less than 6 months (Table 1). Patients treated surgically were less likely than patients treated nonsurgically to be receiving Workers’ Compensation but were more likely to have abnormal physical findings, moderate or severe findings on imaging studies, severe back and leg pain, and greater back-related and overall disability (Tables 1 and 2). Despite worse back-related symptoms and findings, surgically treated patients had better baseline SF-36 general health perceptions. Patients not completing 10-year follow-up (n ⫽ 107) had similar baseline characteristics as those not completing follow-up at 1 and 5 years.11,12 Nonresponders tended to be younger and were significantly more likely to be male, smokers, not married or working, and less educated (data not shown). Among other baseline characteristics, nonresponders also had significantly worse low back pain severity and general health perceptions.

Ten-Year Outcomes of Sciatica • Atlas et al 929

Table 2. Change in Symptoms and Functional Status From Baseline to 10-Yr Follow-up According to Initial Treatment Surgical Treatment* Variable 关mean score (SD)兴† Low back pain in the past week Frequency score Bothersome score Leg pain in the past week Frequency score Bothersome score Sciatica Index Frequency Bothersome Modified Roland scale SF-36 general health perceptions**

Nonsurgical Treatment*

Baseline‡

10 Yr‡

Change§

Baseline‡

10 Yr‡

Change§

P¶

4.4 (2.0) 4.2 (2.0)

1.9 (1.8) 1.9 (2.0)

⫺2.5 (2.3) ⫺2.3 (2.5)

3.7 (2.1) 3.5 (2.0)

2.3 (1.8) 2.3 (1.8)

⫺1.4 (2.4) ⫺1.2 (2.4)

⬍0.001 0.007

5.1 (1.4) 5.3 (1.4)

1.5 (1.8) 1.5 (1.9)

⫺3.6 (2.1) ⫺3.8 (2.2)

3.5 (2.1) 3.6 (2.2)

1.8 (1.7) 1.7 (1.8)

⫺1.7 (2.3) ⫺1.9 (2.4)

⬍0.001 0.002

17.8 (5.2) 17.6 (5.1) 17.7 (4.0) 2.2 (1.1)

6.1 (6.6) 5.7 (6.7) 6.0 (7.0) 2.4 (1.0)

⫺11.8 (7.6) ⫺11.9 (7.4) ⫺11.7 (7.2) 0.26 (1.2)

12.2 (6.6) 11.8 (6.4) 13.3 (5.9) 2.4 (1.0)

6.7 (6.0) 6.1 (6.0) 7.6 (7.0) 2.6 (0.9)

⫺5.5 (7.9) ⫺5.8 (7.4) ⫺5.8 (7.6) 0.25 (1.1)

0.01 0.004 ⬍0.001 0.20

*The no. of patients reporting each variable ranged from 188 to 217 for surgical patients and 152 to 183 for nonsurgical patients. †Low back and leg pain scores range from 0 to 6. Sciatica frequency and bothersome indices are the sum of 4 questions (each response scored from 0 to 6): leg pain, leg or foot weakness, leg numbness, and pain in the back or leg while sitting. Sciatica Index scores range from 0 to 24, modified Roland Scale scores range from 0 to 23, and SF-36 general health perceptions range from 1 to 5. ‡Higher mean scores at baseline and at 10 yr indicate worse symptoms or function for all variables. §Change is calculated as score at 10 yr minus score at baseline. Negative values indicate improvement for all variables. ¶P values compare the change between surgical and nonsurgical treatment groups using multiple linear regression models that control for baseline score. **Assessed using the SF-36 question: “In general, would you say your health is” “excellent” (1) to “poor” (5).

Outcome Measures. On each follow-up questionnaire, patients were asked to describe improvement in low back and leg pain relative to baseline. Responses ranged from “much worse” to “completely gone” on a 7-point scale. The primary symptom outcome was improvement in the patient’s predominant symptom, either back or leg pain, as indicated at baseline. The outcome was categorized as improved if the response was “better,” “much better,” or “completely gone,” the same if the response was “about the same” or “a little better,” and worse if the response was “a little worse” or “much worse.” To assess for a more definitive positive treatment outcome, symptom responses were also classified as “much better” or “completely gone” compared with other responses. Reports of symptoms in the past week were assessed at baseline and follow-up, including the frequency (from 0 “not at all” to 6 “always”) and bothersomeness (from 0 “not bothersome” to 6 “extremely bothersome”) of low back pain, leg pain, leg or foot weakness, leg numbness, and pain in the back or leg while sitting. Sciatica frequency and bothersome indexes, each with scores ranging from 0 to 24, were created by summing the four leg-related questions.13 Back-specific functional status was measured using the modified Roland disability scale.13 General health perceptions were assessed with a question from the Medical Outcomes Study Short Form 36-item questionnaire, “In general, would you say your health is” “excellent” (1) to “poor” (5).14 For each variable, higher scores indicate more severe symptoms or dysfunction. Patients were considered to be satisfied with their current state if they replied that they were “delighted,” “pleased,” or “mostly satisfied” on a 7-point scale. Satisfaction with treatment decisions was assessed at follow-up visits by determining if patients would still choose their initial treatment. Disability and work status were assessed at follow-up and are reported according to whether or not the patient was receiving Workers’ Compensation at baseline. Finally, first operations for patients initially treated nonsurgically and reoperations for those treated surgically were assessed for all patients from physician office records, state hospital discharge data, and patient responses to follow-up surveys and telephone contact.

Analysis. The rating of current findings at follow-up used categorical responses, so distributions were directly compared between treatment groups using ␹2 tests or Fisher’s exact tests. Although baseline clinical features of the treatment groups differed, there was considerable overlap. To adjust for baseline differences between the two treatment groups, logistic regression models were used to estimate the marginal effect of surgical compared with nonsurgical treatment for the predominant symptom and satisfaction with the current state outcomes at 10 years. Changes in symptoms and functional status were assessed by subtracting results at 10-year follow-up from those at baseline, and linear regression models were used to examine the effect of treatment group after controlling for baseline score. In all analyses, the effect of the patient’s initial treatment decision, either surgical or nonsurgical care, was assessed. Reoperation rates among patients initially undergoing surgery and surgical crossover rates among those initially receiving nonsurgical treatment were assessed using survival analysis methods (in this case, time-to-an-event analysis) over 10 years. Secondary analyses of treatment effect considered the actual treatment received at 10-year follow-up. To examine the pattern of change over time, repeatedmeasures analysis was performed using data from all followups between 2 and 10 years (n ⫽ 468, 92%). Mixed-effects models and logistic regression models with Generalized Estimating Equations were used to model the correlation structure of the repeated-measures within each patient.15 The treatmentby-time interaction was used to test how treatment effect differed over time in these models. All analyses were performed using a commercial software package (Statistical Analysis System, SAS Institute, Cary, NC). Results Thirty of 507 eligible patients enrolled in the study died during follow-up; a 93.4% 10-year survival rate (using survival analysis). Among those alive after 10 years, outcomes were available for 400 of 477 (84%) patients: 217


Table 3. Patient Reported Improvement in Symptoms and Satisfaction and Disability Status at 10-Year Follow-up 10-Year Outcomes (%)

Surgical

Nonsurgical

P*

Low back pain vs. baseline† Improved Same Worse Leg pain compared to baseline† Improved Same Worse Predominant symptom compared to baseline†‡ Improved Completely gone Much better Better Same Worse Satisfied with current state, yes§ Still choose the same initial treatment, yes¶ If receiving Workers’ Compensation at entry, current status Receiving any disability compensation Working If employed at entry, current status Receiving any disability compensation Working

(n ⫽ 212) 68.9 20.3 10.9 (n ⫽ 208) 69.2 19.2 11.5 (n ⫽ 207)

(n ⫽ 179) 58.7 27.9 13.4 (n ⫽ 171) 64.3 23.4 12.3 (n ⫽ 175)

0.11

0.19

69.1 32.9 22.7 13.5 18.4 12.6 70.5 86.6

61.1 16.6 23.4 21.1 25.7 13.1 55.5 75.7

0.002 0.006

(n ⫽ 63)

(n ⫽ 88)

19.1

17.1

0.83

81.0 (n ⫽ 130)

75.0 (n ⫽ 92)

0.43

5.4

5.5

1.00

81.5

82.6

0.86

0.56

*P values assessed using Fisher’s exact or ␹2 tests for categorical variables. †Symptom severity was reported to be improved if the response was “better” to “completely gone,” the same if the response was “about the same” or “a little better,” and worse if the response was “a little worse” or “much worse.” ‡The predominant symptom, either back or leg pain, as rated by the patient at baseline. §N ⫽ 217 for surgical and N ⫽ 182 for nonsurgical cohorts. ¶N ⫽ 217 for surgical and N ⫽ 181 for nonsurgical cohorts.

of 255 (85%) initially treated surgically and 183 of 222 (82%) initially treated nonsurgically. Patients’ Global Evaluation at 10 Years Patients were asked to contrast the current status of their low back and leg pain symptoms at 10 years with that at baseline. A slightly higher percentage of patients treated surgically reported improved low back pain compared with patients treated nonsurgically (69% vs. 59%, P ⫽ 0.11), but leg pain improvement was similar in both treatment groups (Table 3). The predominant pain symptom, either low back or leg pain based on the patient’s report at entry, was improved in 69% of surgical patients compared with 61% of nonsurgical patients (P ⫽ 0.19). However, the predominant symptom was reported to be “much better” or “completely gone” in 56% of surgical patients compared with 40% of nonsurgical patients (P ⫽ 0.006). Although improvement in the primary symptom measures was not significantly different between treatment groups, patient satisfaction at 10-year follow-up was significantly better among those treated surgically (Table

3). Satisfaction with the patient’s current state was reported by 70.5% of surgically treated patients versus 55.5% of nonsurgically treated patients (P ⫽ 0.002). Eighty-seven percent of patients treated surgically would still definitely or probably choose their initial treatment compared with 76% of those treated nonsurgically (P ⫽ 0.006). Work and disability outcomes were similar regardless of initial treatment received (Table 3). Among patients receiving Workers’ Compensation at study entry, 19% of surgically treated and 17% of nonsurgically treated patients were receiving some form of disability compensation at 10 years (P ⫽ 0.83). Patients who were working at study entry were very unlikely to be receiving disability compensation at 10 years (5%), regardless of treatment. The percentage of patients working at 10-year follow-up was also similar regardless of initial work, disability, or treatment status. Change in Symptoms and Functional Status Over 10 Years Patients rated the frequency and bothersomeness of low back pain and leg symptoms in the past week at baseline and 10 year follow-up (Table 2). For both low back pain and leg pain, surgically treated patients rated the frequency and severity of symptoms as worse at baseline evaluation and better at 10-year follow-up compared with patients treated nonsurgically. The frequency and bothersomeness of sciatica symptoms were also worse at baseline and better at follow-up for surgical patients. For each symptom question, the change in symptoms at 10 years was significantly greater among surgically treated patients than among nonsurgically treated patients (all P ⬍ 0.01). Similarly, surgically treated patients reported significantly greater improvement in back-specific functional status using the modified Roland scale (11.7 point improvement vs. 5.8 point improvement, P ⬍ 0.001). Although back-related symptoms and function improved over 10 years, general health perceptions worsened slightly to a comparable degree in both treatment groups. Independent Predictors of Symptom Improvement and Satisfaction at 10 Years Initial treatment was not randomly assigned. At baseline, surgical patients on average had more severe symptoms and findings than nonsurgical patients. Variables in all models included treatment group, age, and sex as well as independent baseline predictors of outcome. Logistic regression models examining the independent effect of treatment on the patients’ predominant symptom and satisfaction with the current state at 10 years confirmed the unadjusted findings in Table 3 (Table 4). Specifically, when “improved” was defined as a patient response of “better,” “much better,” or “completely gone,” surgical treatment was not a statistically significant independent predictor of the predominant symptom outcome (odds ratio [OR], 1.4; 95% confidence interval [CI], 0.9 –2.3). However, if symptom improvement was restricted to re-


Table 4. Independent Baseline Predictors of Patient Symptoms and Satisfaction at 10 Yr Predominant Symptom† Improved ⫽ At Least Better Baseline Features* Initial treatment, surgical Unadjusted Adjusted model Education, college graduate Married or living together Receiving Workers’ Compensation Comorbid illnesses, yes § Abnormal examination findings ¶ Location of pain, unilateral leg Low Back Pain Score (0–12), per 2-point increment Sciatica Bothersome Index (0–24), per 4-point increment SF-36 General Health (0–100), per 10point increment SF-36 Mental Health (0–100), per 10point increment

Improved ⫽ At Least Much Better

Satisfied With Current State‡

OR (95% CI)

P

OR (95% CI)

P

OR (95% CI)

P

1.4 (0.9–2.2) 1.4 (0.9–2.3) NS NS 0.6 (0.4–0.9) NS NS NS 0.8 (0.7–0.9)

0.10 0.17

1.9 (1.2–2.8) 2.1 (1.2–3.7) 2.3 (1.4–3.9) 2.0 (1.1–3.7) 0.4 (0.2–0.6) NS 1.3 (1.0–1.7) 2.6 (1.4–4.9) NS

0.003 0.01 0.002 0.02 ⬍0.001

1.9 (1.3–2.9) 2.2 (1.4–3.6) NS NS 0.5 (0.3–0.8) 0.5 (0.3–0.9) NS NS 0.8 (0.7–0.9)

0.002 0.001

0.7 (0.6–0.8)

⬍0.001

NS

1.2 (1.1–1.4)

0.005

NS

0.02

0.002

NS 1.2 (1.1–1.4)

0.004

NS

NS

0.01 0.003

1.1 (1.0–1.3)

0.002 0.02 0.002

0.04

*Stepwise multiple logistic regression models were developed for the three outcomes reported. The table presents the final model for each outcome that adjusts for initial treatment, age, gender, and other independent baseline predictors of outcome (c-statistic ⫽ 0.69, predominant symptom improved; 0.78, predominant symptom much better; 0.74, satisfaction). NS ⫽ not (statistically) significant. †The predominant symptom, either back or leg pain, as indicated by the patient at baseline symptom was categorized as improved if the response was “better,” “much better,” or “completely gone.” To assess for a more definitive positive treatment outcome, symptom responses were also classified as “much better” or “completely gone.” ‡Patient considered to be satisfied with their current state if they replied that they were “delighted,” “pleased,” or “mostly satisfied” on a 7-point scale. §Any self-reported chronic pulmonary disease, heart disease, stroke, cancer, or diabetes. ¶ The no. of positive physical examination findings for a patient, including unilateral strength, sensation, or reflex abnormality (range, 0 –3 findings). **Sum of low back pain frequency and bothersomeness questions in the past week at baseline (score range, 0 –12; higher scores indicate worse low back pain).

sponses of “much better” or “completely gone,” surgical treatment was significantly associated with a favorable outcome (OR, 2.1; 95% CI, 1.2–3.7). Patients treated surgically were also more likely to be satisfied with their current state (OR, 2.2; 95% CI, 1.4 –3.6). Depending on the outcome variable, several other baseline factors were independently associated with outcome (Table 4). In all models, receiving Workers’ Compensation at baseline was associated with worse outcomes (OR, 0.4 – 0.6). Greater severity of baseline symptoms was also associated with worse outcomes, although the particular symptom measure (low back pain or sciatica bothersomeness) varied depending on the outcome selected (OR, 0.7– 0.8). On the other hand, better SF-36 general health status at study entry was associated with improved symptoms, while SF-36 better mental health status was associated with higher satisfaction at 10 years (OR, 1.2 per 10-point change). Time Course of Outcomes Mean scores at baseline and at each follow-up over 10 years for the frequency of sciatica symptoms and backspecific functional status are shown in Figure 1. The percentage of patients reporting satisfaction with their current state at each follow-up is shown in Figure 2. As previously reported,12 most of the improvement in outcome due to surgery was seen shortly after patients’ entry into the study. From years 2 through 10, there was a small but significant improvement in the frequency of sciatica symptoms and satisfaction over time in both

Figure 1. Time course of symptom and functional status outcomes. Assessed at initial evaluation and at 3, 6, and 12 months, and then yearly follow-up for nonsurgical (-F-) and surgical (-E-) treatment. Mean ⫾ 2 SE. A, Sciatica Frequency Index. B, Modified Roland score.


Figure 2. Time course of satisfaction outcome. Assessed at 3, 6, and 12 months, and then yearly follow-up for nonsurgical (-F-) and surgical (-E-) treatment. Percent satisfied with current state ⫾ 2 SE.

treatment groups and to a similar extent. There was no change in functional status between 2 and 10 years for patients initially treated surgically, although there was a small amount of improvement for nonsurgical patients (P ⫽ 0.2 for the interaction between time and treatment group). Lumbar Spine Surgery After Initial Treatment Among patients initially undergoing surgical treatment, the 10-year reoperation rate was 25% (n ⫽ 64; median time to reoperation, 24 months). Among patients initially receiving nonsurgical treatment, the crossover rate to surgery between 3 months and 10 years was 25% (n ⫽ 53; median time to crossover, 24 months). Baseline characteristics and findings were similar for surgical patients undergoing a reoperation or not, and for nonsurgical patients having subsequent surgery or not (data not shown). 10-Year Outcomes According to Actual Treatment Received Secondary analyses of treatment effect considered the actual treatment received at 10-year follow-up for the

138 patients treated completely nonsurgically, and for the 262 patients treated surgically, including 51 patients initially in the nonsurgical group who crossed over to surgical treatment. There were no significant outcome differences according to actual treatment received at 10year follow-up for symptom improvement, satisfaction, work and disability status, and change in symptom frequency or bother. However, surgically treated patients had greater improvement in Roland functional status (10.6 vs. 6.0 point change for surgical and nonsurgical groups, respectively, P ⫽ 0.03). Ten-year outcomes were also compared according to the patients’ initial and final treatment status: surgery with or without a second spine operation (reoperation) and initial nonsurgical treatment with or without crossover to surgical treatment after 3 months (Table 5). Patients who remained in their initial treatment group throughout the study, especially those treated surgically, had better outcomes than those undergoing subsequent surgery. Surgical patients having a reoperation or nonsurgical patients having subsequent surgery had outcomes that were inferior to those initially receiving surgical treatment. For example, 51% and 40% of those having a reoperation or crossing over to subsequent surgery were satisfied with the current state compared with 76% and 61% of surgical or nonsurgical treatment continuing without subsequent surgery. There was also a trend for patients undergoing subsequent surgical procedures to have worse disability and work outcomes than those continuing with their initial treatment, whether surgical or nonsurgical. Outcomes of Those Not Completing 10-Year Follow-up A total of 107 patients did not complete 10-year followup: 56 surgically and 51 nonsurgically treated. Eightysix of these 107 patients (80%) returned at least 1 prior follow-up survey (most recent follow-up: range, 3–108 months; median, 60 months). Outcomes at the most recent follow-up reported by these patients were worse

Table 5. 10-Yr Outcomes for Patients Based Upon Initial and Final Treatment Status Initial Surgical Treatment

Initial Nonsurgical Treatment

No Reoperation

Reoperation

Crossover to Surgery

10-yr Outcomes

(n ⫽ 166)

(n ⫽ 51)

P*

(n ⫽ 45)

P*

(n ⫽ 138)

P*

Predominant symptom, % improved† Satisfied with current state, % yes Sciatica Frequency Index, mean change (SD)‡ Roland scale, mean change (SD)‡ SF-36 general health perceptions, mean change (SD)§ If receiving Workers’ Compensation at entry, current status Receiving any disability compensation Working

71.8 76.5 ⫺12.5 (7.1) ⫺12.6 (6.9) 0.16 (1.2)

60.8 51.0 ⫺9.3 (8.7) ⫺8.4 (7.1) 0.60 (1.2)

0.14 ⬍0.001 0.008 0.002 0.02

48.8 40.0 ⫺4.5 (7.4) ⫺5.4 (7.6) 0.22 (1.1)

0.006 ⬍0.001 ⬍0.001 ⬍0.001 0.08

64.9 60.6 ⫺5.9 (8.0) ⫺6.0 (7.6) 0.26 (1.1)

0.21 0.003 0.02 ⬍0.001 0.19

(n ⫽ 48)

(n ⫽ 15)

12.5 85.4

40.0 66.7

(n ⫽ 22) 0.02 0.12

27.3 68.2

No Subsequent Surgery

(n ⫽ 66) 0.14 0.10

13.6 77.3

0.86 0.28

*Using analysis of variance with Dunnett’s adjustment for multiple comparisons and using surgical with no reoperation group as reference. †The predominant symptom, either back or leg pain, as rated by the patient at baseline. Symptom severity was reported to be improved if the response was “better” to “completely gone.” ‡Change is calculated as score at 10 yr minus the score at baseline. Negative values indicate improvement for all variables. §Assessed using the SF-36 question: “In general, would you say your health is” “excellent” (1) to “poor” (5).


than those completing 10-year follow-up, but they were similarly worse for both surgical and nonsurgical patients. If one includes outcomes from the last available follow-up survey (3–120 months) for the surgical (n ⫽ 266, 97.4%) or nonsurgical (n ⫽ 220, 94.0%) cohort, the comparative results and conclusions are unchanged. For example, the predominant symptom was improved in 66.4% of surgical and 58.2% of nonsurgical patients (P ⫽ 0.06). In multiple regression models examining symptom improvement and satisfaction with the current state that included variables for treatment group, time of the last follow-up, and whether the patient died or not, only death was significantly associated with the outcome (patients lost to follow-up because of death had better outcomes). Discussion Ten-year follow-up in this study was available for 84% of surviving patients with sciatica due to a herniated lumbar disc initially treated either surgically or nonsurgically. Although patients treated surgically had more severe symptoms and worse functional status at baseline than those treated nonsurgically, surgical patients reported better functional status and were more satisfied with their current state after 10 years. However, a similar percentage of patients in both groups reported improved symptoms. Disability and work status were also similar regardless of initial treatment. Although the Weber study has been criticized for not meeting current standards for randomized trials,16 it is the only randomized comparative trial of surgical versus nonsurgical treatment for patients with sciatica due to a lumbar intervertebral disc herniation.8 Our results after 5 years were similar to Weber’s 4-year results.12 At 10 years, Weber reported good results in 63.6% of patients initially randomized to surgery and 56% of those initially receiving conservative treatment, a difference that was not statistically significant. The narrowing of outcomes between 4 and 10 years represented a small improvement for conservatively treated patients and worse outcomes in surgically treated patients. In our study, favorable 10-year outcomes were somewhat higher in both treatment groups (69% and 61%, respectively, among surgical and nonsurgical patients), but the absolute difference in good results was nearly identical in the two studies (see Figure 3 in accompanying manuscript).17 Like Weber’s findings, these overall results did not achieve statistical significance in unadjusted or adjusted analyses, even with our larger sample size. Few other prospective studies have reported longterm outcomes of discectomy, chemonucleolysis, or conservative care for patients with an interververtebral disc herniation. At 10 years, Gogan reported 77% of patients initially randomized to chymopapain and 38% of patients receiving placebo injections were at least moderately improved.18 Other nonrandomized studies comparing long-term outcomes of discectomy to other nonsurgical treatments have found modest differences favoring surgery.19,20

Although the relative benefit of surgical compared with nonsurgical treatment was small for improvement in the patient’s predominant pain symptom, other important outcomes continued to favor surgery after 10 years, including complete pain relief, back-specific functional status, and satisfaction. Few other comparable studies have included the broad range of validated outcome measures used in the Maine Lumbar Spine Study. Nevertheless, these differences favoring surgical treatment did not translate into better disability and work outcomes over 10 years. Few patients, irrespective of initial treatment or Workers’ Compensation status, were receiving any disability compensation and approximately 80% were working at 10 years. These results suggest that specific medical treatments may have less impact on disability and work status than factors such as workplace accommodations, job characteristics (tasks, autonomy, satisfaction, etc.), and other income and local economic factors.21–24 Among those initially treated surgically, one in four had at least 1 additional spinal operation over 10 years. These reoperation rates are generally higher than those reported in prior studies.19,25–27 For patients initially treated nonsurgically, another 1 in 4 had subsequent spine surgery between 3 months and 10 years. Few studies have examined long-term rates of subsequent surgery in patients initially managed conservatively.18,19 Although we lack detailed information about the indications and findings that led to a subsequent operation, outcomes associated with these procedures were generally poor. Prior studies have reported conflicting outcomes from patients undergoing subsequent surgery.28 –31 Future research is needed to confirm these generally poor findings and to perform interventional trials on patients considering these procedures. Strengths of our study include prospective, long-term follow-up in a large percentage of patients from contemporary comparison groups treated in community-based clinical practice using validated outcome measures, all making it likely that the results are broadly generalizable. The study’s major limitation is its observational, nonrandomized design.11,12 One cannot be certain that the differences in outcomes between treatment groups were exclusively due to surgery, rather than unmeasured confounders. Important differences at baseline in demographic, past history, and current symptoms, backrelated disability, and objective findings existed among those patients treated surgically or nonsurgically. However, the effect of surgical treatment on the predominant symptom and satisfaction were similar after adjusting for other predictors of outcome in regression models. Also, the overall results are similar to those of Weber’s randomized trial, although both of our treatment groups had somewhat better outcomes. Finally, although our follow-up rate is high, these 10-year results likely overstate the benefit of both surgery and conservative care to a similar extent based on dropouts in this study appearing to have worse outcomes than those remaining in the study.


Although our results support the relative benefit of surgery for properly selected patients, this does not mean that nonsurgical treatment should no longer be a recommended initial treatment. Conservative treatment for at least 4 weeks is recommended for patients with an intervertebral disc herniation without significant neurologic examination findings.32 After 6 weeks, our results should not be interpreted as requiring surgery in those without improvement since many of the conservatively treated patients in this study had good outcomes. For those with very bothersome symptoms, surgical treatment can hasten recovery and may result in better long-term outcomes. However, pain outcomes narrow over time, a sizeable number undergo additional surgery, and work and disability outcomes are not improved with surgical treatment. Important questions remain for patients and physicians about optimal conservative treatment beyond 6 weeks. Patients with persisting symptoms who are reluctant to undergo surgery may continue conservative care knowing that their symptoms will likely improve, but the magnitude of the improvement may not be as great as for surgical treatment. It is unclear whether there is a time window beyond which the relative benefit of surgery diminishes and thus when conservative treatment should be deemed to have failed. These results demonstrate how complex it is to interpret outcomes of studies that evaluate the relative impact of alternative treatments in patients with a lumbar disc herniation. Patients need to better understand that their specific symptoms and concerns may determine how relative treatment benefits and risks should be weighed. Differences in decision-making depend on whether the most important outcome is symptoms, function, satisfaction, disability, or work. Efforts are needed to integrate unbiased information that combines the patient’s perspective with his or her clinical findings into the clinical decision-making process.9,33 Additional work is needed to assess whether helping patients and their physicians translate the complexity of these findings into individualized treatment plan results in improved outcomes of care.34,35 Key Points ● Outcomes of 400 patients with sciatica resulting from a lumbar disc herniation treated surgically or nonsurgically were followed over a 10-year period. ● Because treatment was determined in a routine clinical manner by the patient and the physician, those surgically treated had more severe symptoms and worse functional status at baseline. ● Patients initially treated surgically had more complete relief of leg pain and improved function and satisfaction compared with patients initially treated nonsurgically over 10 years. ● However, improvement in the patient’s predominant symptom and work and disability outcomes were similar regardless of treatment received.

Acknowledgments The authors thank YuChiao Chang, PhD, for assistance with statistical analyses and Valerie Soucie for assistance with managing long-term patient follow-up.

References 1. Frymoyer JW. Back pain and sciatica. N Engl J Med 1988;318:291–300. 2. Andersson GBJ, Brown MD, Dvorak J, et al. Consensus summary on the diagnosis and treatment of lumbar disc herniation. Spine 1996;24(suppl): 75– 8. 3. Cherkin DC, Deyo RA, Loeser JD, et al. An international comparison of back surgery rates. Spine 1994;19:1201– 6. 4. Keller RB, Atlas SJ, Soule DN, et al. Relationship between rates and outcomes of operative treatment for lumbar disc herniation and spinal stenosis. J Bone Joint Surg Am 1999;81:752– 62. 5. Taylor VM, Deyo RA, Cherkin DC, et al. Low back pain hospitalization: recent United States trends and regional variations. Spine 1994;19:1207–13. 6. Volinn E, Mayer J, Diehr P, et al. Small area analysis of surgery for low back pain. Spine 1992;17:575– 81. 7. Weinstein J, Birkmeyer J, eds. The Dartmouth Atlas of Musculoskeletal Health Care. Chicago: American Hospital Association Press, 2000. 8. Weber H. Lumbar disc herniation: a controlled, prospective study with ten years of observation. Spine 1983;8:131– 40. 9. Birkmeyer NJ, Weinstein JN, Tosteson AN, et al. Design of the Spine Patient outcomes Research Trial (SPORT). Spine 2002;27:1361–72. 10. Keller RB, Atlas SJ, Singer DE, et al. The Maine Lumbar Spine Study: I. Background and Concepts. Spine 1996;21:1769 –76. 11. Atlas SJ, Deyo RA, Keller RB, et al. The Maine Lumbar Spine Study: II. 1-year outcomes of surgical and nonsurgical management of sciatica. Spine 1996;21:1777– 86. 12. Atlas SJ, Keller RB, Chang Y, et al. Surgical and nonsurgical management of sciatica secondary to a lumbar disc herniation: five-year outcomes from the Maine Lumbar Spine Study. Spine 2001;26:1179 – 87. 13. Patrick DL, Deyo RA, Atlas SJ, et al. Assessing health-related quality of life in patients with sciatica. Spine 1995;20:1899 –909. 14. Ware JE, Sherbourne CD. The MOS 36-item short-form survey (SF-36): 1. Conceptual framework and item selection. Med Care 1992;30:473– 83. 15. Diggle PJ, Liang K, Zeger SL. Analysis of Longitudinal Data. New York: Oxford University Press, 1994. 16. Bessette L, Liang MH, Lew RA, et al. Classics in spine: surgery literature revisited. Spine 1996;21:259 – 63. 17. Keller RB, Atlas SJ. Is there a continuing role for prospective observational studies in spine research? Spine 2005;30:846 – 848. 18. Gogan WJ, Fraser RD. Chymopapain: a 10-year, double-blind study. Spine 1992;17:388 –94. 19. Nykbist F, Hurme M, Alaranta H, et al. Severe sciatica: a 13-year follow-up of 342 patients. Eur Spine J 1995;4:335– 8. 20. Tregonning GD, Transfeld EE, McCulloch JA, et al. Chymopapain versus conventional surgery for lumbar disc herniation: 10-year results of treatment. J Bone Joint Surg Br 1991;73:81– 6. 21. Dionne CD, Von Korff M, Koespell TD, et al. A comparison of pain, functional limitations, and work status indices as outcome measures in back pain research. Spine 1999;24:2339 – 45. 22. Hadler NM. Back pain in the workplace: what you lift or how you lift matters far less than whether you lift or when. Spine 1997;22:935– 40. 23. Magora A. Investigation of the relation between low back pain and occupation: V. Psychological aspects. Scand J Rehabil Med 1973;5:191– 6. 24. Papageorgiou AC, Macfarlane GJ, Thomas E, et al. Psychosocial factors in the workplace: do they predict new episodes of low back pain? Spine 1997; 22:1137– 42. 25. Keskimaki I, Seitsalo S, Osterman H, et al. Reoperations after lumbar disc surgery: a population-based study of regional and interspecialty variations. Spine 2000;25:1500 – 8. 26. Loupasis GA, Stamos K, Katonis PG, et al. Seven- to 20-year outcome of lumbar discectomy. Spine 1999;24:2313–7. 27. Moore AJ, Chilton JD, Uttley D. Long-term results of microlumbar discectomy. Br J Neurosurg 1994;8:319 –26. 28. Cinotti G, Roysam GS, Eisenstein SM, et al. Ipsilateral recurrent lumbar disc herniation: a prospective, controlled study. J Bone Joint Surg Br 1998;80: 825–32. 29. Cinotti G, Gumina S, Giannicola G, et al. Contralateral recurrent lumbar

Ten-Year Outcomes of Sciatica • Atlas et al 935 disc herniation: results of discectomy compared with those in primary herniation. Spine 1999;24:800 – 6. 30. Suk K-S, Lee H-M, Moon S-H, et al. Recurrent lumbar disc herniation: results of operative management. Spine 2001;26:672– 6. 31. Vik A, Zwart JA, Hulleberg G, et al. Eight year outcome after surgery for lumbar disc herniation: a comparison of reoperated and not reoperated patients. Acta Neurochir (Wien) 2001;143:607–10. 32. Deyo RA, Weinstein JN. Low back pain. N Engl J Med 2001;344:363–70.

33. Deyo RA. A key medical decision maker: the patient. Br Med J 2001;323: 466 –7. 34. Deyo RA, Cherkin DC, Weinstein J, et al. Involving patients in clinical decisions: impact of an interactive video program on use of back surgery. Med Care 2000;38:959 – 69. 35. Goldberg HI, Deyo RA, Taylor VM, et al. Can evidence change the rate of back surgery? A randomized trial of community-based education. Eff Clin Pract 2001;4:95–104.