Abstract
Study Design
Retrospective Cohort.
Objective
We investigate whether duration of symptoms a patient experiences prior to lumbar microdiscectomy affects pain, lifestyle, and return to work metrics after surgery.
Methods
A retrospective review of patients with a diagnosis of lumbar radiculopathy undergoing microdiscectomy was conducted using a statewide registry. Patients were grouped based on self-reported duration of symptoms prior to surgical intervention (Group 1: symptoms less than 3 months; Group 2: symptoms between 3 months and 1 year; and Group 3: symptoms greater than 1 year). Radicular pain scores, PROMIS PF Physical Function measure (PROMIS PF), EQ-5D scores, and return to work rates at 90 days, 1 year, and 2 years after surgery were compared using univariate and multivariate analysis.
Results
There were 2408 patients who underwent microdiscectomy for lumbar disc herniation for radiculopathy with 532, 910, and 955 in Groups 1, 2, and 3, respectively. Postoperative leg pain was lower for Group 1 at 90 days, 1 year, and 2 years compared to Groups 2 and 3 (P < .05). Postoperative PROMIS PF and EQ-5D scores were higher for Group 1 at 90 days, 1 year, and 2 years compared to Groups 2 and 3 (P < .05).
Conclusion
Patients with prolonged symptoms prior to surgical intervention experience smaller improvements in postoperative leg pain, PROMIS PF, and EQ-5D than those who undergo surgery earlier. Patients undergoing surgery within 3 months of symptom onset have the highest rates of return to work at 1 year after surgery.
Keywords: lumbar, degenerative disc disease, degenerative, disc, discectomy, decompression
Introduction
Background
Intervertebral disc herniations represent one of the most common spinal pathologies requiring surgery. Lumbar disc herniations can cause severe debilitating pain which places a great burden on patients and the health care system alike. 1 The radiculopathy associated with lumbar disc herniation results in reduced patient quality-of-life across multiple metrics including the Patient-Reported Outcomes Measurement Information System PROMIS PF, and Euroqol (EQ-5D).2,3
Initial management of intervertebral disc herniation involves conservative measures with physical therapy, anti-inflammatory medications, and epidural spinal injections. 4 Most patients experience relief of symptoms within 6 weeks; however, some have progression of symptoms. 5 In these cases, surgical treatment has been shown to provide effective relief of symptoms from lumbar disc herniation.6,7 Despite the high success ranging from 78-95% for microdiscectomies, some authors argue that poor outcomes may occur due to the duration of pre-operative symptoms (i.e., motor deficits, sensory changes, persistent low back pain, or persistent radiculopathy).6,8–10
Earlier surgical intervention has been shown to result in faster pain relief and better functional and clinical outcomes relative to conservative management, while delay in surgery may result in significantly worse outcomes.2,10,11 Multiple studies have evaluated the effects of the duration of symptoms on outcomes following microdiscectomy, suggesting that increased symptom duration caused by lumbar disc herniation may result in poorer outcomes following treatment.12,13 However, other studies have suggested that there is limited or inconclusive evidence to suggest a relationship between the duration of symptoms and post-operative success. 14 Movassaghi et al 15 (n = 210) assessed patient-reported outcomes and reported that the duration of symptoms prior to surgery did not notably influence postoperative clinical outcomes, the frequency of subsequent operations, or the level of patient satisfaction. The polarized differences in the literature have emphasized the indeed for large multi-institutional studies assessing whether there truly is a relationship between symptom chronicity and postoperative outcomes.
Large multi-center research registries are increasing in popularity. The Michigan Spine Surgery Improvement Collaborative (MSSIC) is a collaborative quality initiative with the goal of improving spine surgery outcomes throughout the state of Michigan.16,17 MSSIC maintains a comprehensive data registry from across 29 hospital sites that includes demographic and pre-surgical clinical data, as well as individual patient surgical procedure information including complications and/or adverse events. MSSIC quality initiatives have led to a 15.1% reduction in readmissions from 2017 to 2018, allowing the organization to expand its agenda on patient-reported outcomes pre- and post-spine surgery. 17
The purpose of this study was to use the MSSIC database to examine the effect of the duration of symptoms on outcomes following lumbar microdiscectomy. We hypothesized that patients with a longer duration of pre-operative symptoms have less favorable clinical outcomes compared to those with a shorter duration of symptoms. This information may help spine surgeons and their patients to understand the optimal time for surgical intervention of lumbar disc herniations.
Patients/Methods
A retrospective review was conducted by examining prospectively compiled data from the MSSIC registry. The MSSIC is a quality improvement initiative that involves a diverse range of orthopedic surgeons and neurosurgeons from private, academic, rural, and urban contexts. The registry includes data from elective spine surgeries for degenerative diseases, excluding non-degenerative and/or complex pathologies.
This study protocol was approved by our institutional review board. A retrospective cohort study was performed using the MSSIC registry between June 1, 2016 to December 4, 2020. Informed consent was waived since data collection was part of routine care in a de-identified fashion. Patients who underwent a standard primary single-level lumbar microdiscectomy for radiculopathy were identified based on International Classification of Diseases 10 th revision code (ICD-10). Patients who were under 18 years of age at the time of surgery and patients who had undergone fusion were excluded. 2408 patients were identified who met inclusion criteria. Patients were excluded if they had any previous lumbar surgery, were pregnant, or had more than 1 level operated on.
Patients were separated into three groups based on self-reported duration of symptoms prior to surgical intervention. Prior to surgery, Group 1 had symptoms for <3 months; Group 2 had symptoms >3 months but <1 year; and Group 3 had symptoms for 1 year or greater (Figure 1). Baseline patient characteristics include age, gender, race, body mass index (BMI), education, diabetes, deep vein thrombosis (DVT), coronary artery disease (CAD), depression, anxiety, osteoporosis, ambulation status, private insurance, previous spine surgery, hypertension, current smoking status, current work status, and ASA scores were collected. Pre- and post-operative 0-10 radicular pain rating scores, PROMIS PF Physical Function four-item Short Form (PROMIS PF), EQ-5D scores, as well as self-reported return to work rates at 90 days, 1 year, and 2 years after surgery were collected on all patients and compared among groups. For PROMIS PF, EQ-5D, and 0-10 pain ratings, patients were classified as “yes” or “no” in terms of whether they had achieved at least a “Minimum Clinically Important Difference” (MCID) degree of improvement from baseline to follow-up point.
Figure 1.
Patient distribution based on symptom duration.
Ethical Approval
This study was reviewed by Beaumont Hospital IRB committee and was determined to be exempt from institutional review board approval (IRB number # 2019-379).
Statistical Analysis
Univariate comparisons were performed for demographic, medical history, and complication variables between patient symptom duration groups of <3 months, >3 months but <1 year, and 1 year or more. T-tests were used for normally distributed continuous variables, while Wilcoxon Rank Sum tests were used for continuous skewed variables, and Chi-square tests were performed for categorical variables. Poisson generalized estimating equation (GEE) models were used for multivariate analysis to determine patient and operative characteristics associated with PROMIS PF minimum clinically important difference (MCID), EQ-5D MCID, Leg pain MCID, and complications at 90 days, 1 year, and 2 years. For continuous EQ-5D at 90 days, 1 year, and 2 years, normal GEE models were used for multivariate analysis. Variables adjusted for during multivariate analysis include age, gender, race, education, BMI, diabetes, DVT, CAD, depression, anxiety, osteoporosis, ASA >2, independent ambulation pre-op, private insurance, previous spine surgery, hypertension, current smoking status, and appropriate baseline score. All analyses are performed using SAS 9.4 (SAS Institute Inc., Cary, NC, USA) with statistical significance set at P < .05.
Results
Demographics
Of the 2408 patients who underwent microdiscectomy for lumbar disc herniation, 543 reported symptom duration less than 3 months (Group 1), 910 reported symptoms for longer than 3 months but less than 1 year (Group 2), and 955 reported symptoms for longer than 1 year prior to surgical intervention (Group 3) (Table 1). Gender, BMI, most comorbidities (diabetes, history of DVT, CAD, osteoporosis, and hypertension), and ASA scores were similar between groups. Patient groups displayed significant baseline differences when comparing education (P < .05), pre-operative diagnosis of depression and anxiety (P < .05), pre-operative independent ambulation (P < .05), possession of private insurance (P < .05), smoking (P < .05), and return to work plans (P < .05).
Table 1.
Descriptive Statistics Comparing Symptom Duration Groups for Demographics and Comorbidities.
| Demographic/Comorbidity | <3 months (N = 543) | >3 months but <1 year (N = 910) | 1 year or > (N = 955) | P-Value | |
|---|---|---|---|---|---|
| Age: Mean ± SD (N) | 53.7 ± 14.0 (N = 543) | 51.4 ± 15.0 (N = 910) | 50.5 ± 15.1 (N = 955) | .0002 | |
| Age: Median (min-max) | 55.0 (18.0-91.0) | 51.0 (19.0-90.0) | 51.0 (18.0-91.0) | ||
| Male | 314 (57.8%) | 476 (52.3%) | 499 (52.3%) | .074 | |
| Race | |||||
| Black | 19 (3.5%) | 42 (4.6%) | 70 (7.3%) | .0090 | |
| Other | 52 (9.6%) | 96 (10.6%) | 108 (11.3%) | ||
| White | 471 (86.9%) | 770 (84.8%) | 776 (81.3%) | ||
| Education | |||||
| College | 192 (35.4%) | 282 (31.0%) | 237 (24.8%) | .0001 | |
| High school | 226 (41.6%) | 433 (47.6%) | 506 (53.0%) | ||
| Less than high school | 17 (3.1%) | 34 (3.7%) | 46 (4.8%) | ||
| Some college | 108 (19.9%) | 161 (17.7%) | 166 (17.4%) | ||
| BMI | |||||
| Normal | 122 (22.5%) | 196 (21.7%) | 180 (18.9%) | .083 | |
| Obese | 246 (45.4%) | 380 (42.0%) | 454 (47.7%) | ||
| Overweight | 174 (32.1%) | 329 (36.4%) | 317 (33.3%) | ||
| Diabetes | 86 (15.8%) | 124 (13.6%) | 148 (15.5%) | .40 | |
| DVT | 21 (3.9%) | 29 (3.2%) | 32 (3.4%) | .78 | |
| CAD | 36 (6.6%) | 56 (6.2%) | 72 (7.5%) | .49 | |
| Depression | 115 (21.2%) | 203 (22.3%) | 310 (32.5%) | <.0001 | |
| Anxiety | 120 (22.1%) | 183 (20.1%) | 255 (26.7%) | .0027 | |
| Osteoporosis | 33 (6.1%) | 57 (6.3%) | 74 (7.7%) | .33 | |
| ASA >2 | 174 (32.0%) | 254 (27.9%) | 312 (32.7%) | .063 | |
| Independent ambulation pre-op | 452 (83.2%) | 812 (89.2%) | 841 (88.1%) | .0029 | |
| Private insurance | 366 (67.4%) | 636 (69.9%) | 579 (60.6%) | <.0001 | |
| Previous spine surgery | 145 (27.1%) | 209 (23.1%) | 238 (25.1%) | .23 | |
| Hypertension | 210 (38.7%) | 313 (34.4%) | 310 (32.5%) | .052 | |
| Current smoker | 91 (16.8%) | 199 (21.9%) | 231 (24.2%) | .0035 | |
| Plan to return to work | |||||
| No | 129 (26.3%) | 207 (24.9%) | 225 (26.1%) | <.0001 | |
| Yes | 330 (67.2%) | 558 (67.1%) | 515 (59.8%) | ||
| Unknown | 32 (6.5%) | 67 (8.1%) | 121 (14.1%) | ||
| Employed baseline | |||||
| No | 284 (54.4%) | 444 (50.3%) | 485 (52.6%) | .31 | |
| Yes | 238 (45.6%) | 439 (49.7%) | 437 (47.4%) | ||
Comparison of Metrics Pre-Operatively
Baseline leg pain was similar across the groups (7.07 vs 6.86 vs 6.90, P = .086) (Table 2). Baseline PROMIS PF scores were lower for patients experiencing symptoms for less than 3 months (33.5 vs 35.4 vs 35.2, P < .05), and baseline EQ5D scores were similar across the groups (.53 vs .54 vs .53, P = .34). All groups demonstrated clinically significant improvements in leg pain, PROMIS PF, and EQ-5D scores post-operatively when compared to baseline (P < .05).
Table 2.
Descriptive Statistics Comparing Symptom Duration Groups for Baseline Values.
| <3 months (N = 543) | >3 months but <1 year (N = 910) | 1 year or > (N = 955) | P-Value | |
|---|---|---|---|---|
| Baseline: Leg pain (mean ± SD) | 7.07 ± 2.66 | 6.86 ± 2.59 | 6.90 ± 2.55 | .086 |
| Baseline: PROMIS score (mean ± SD) | 33.5 ± 6.2 | 35.4 ± 5.9 | 35.2 ± 6.0 | <.0001 |
| Baseline: EQ-5D score (mean ± SD) | .53 ± .18 | .54 ± .22 | .53 ± .23 | .34 |
Analysis of Leg Pain Scores & Lifestyle Metrics
Postoperative leg pain was lower for Group 1 at 90 days, 1 year, and 2 years when compared to Group 2 and Group 3 (P < .05 at 90 days, 1 year, and 2 years) (Table 3) (Figure 2). PROMIS PF and EQ-5D scores were higher for Group 1 at 90 days, 1 year, and 2 years when compared to Group 2 and Group 3 (P < .05 for PROMIS PF at 90 days, 1 year, and 2 years) (P < .05 for EQ5D at 90 days, 1 year, and 2 years) (Figure 3) (Figure 4).
Table 3.
Univariate Analysis for Continuous Leg Pain, PROMIS Scores, EQ-5D Scores, and Return to Work Rate Post-Op.
| <3 months (N = 543) | >3 months but <1 year (N = 910) | 1 year or > (N = 955) | P-Value | |
|---|---|---|---|---|
| 90 day: Leg pain (mean ± SD (N)) | 2.04 ± 2.65 (N = 434) | 2.17 ± 2.75 (N = 718) | 2.69 ± 3.03 (N = 756) | .0006 |
| 90 day: Leg pain (median (min-max)) | 1.0 (.0-10.0) | 1.0 (.0-10.0) | 2.0 (.0-10.0) | |
| 1 year: Leg pain (mean ± SD (N)) | 1.69 ± 2.4 (N = 321) | 2.11 ± 2.77 (N = 541) | 2.88 ± 3.11 (N = 543) | <.0001 |
| 1 year: Leg pain (median (min-max)) | .0 (.0-10.0) | 1.0 (.0-10.0) | 2.0 (.0-10.0) | |
| 2 year: Leg pain (mean ± SD (N)) | 1.89 ± 2.75 (N = 212) | 1.96 ± 2.73 (N = 334) | 3.0 ± 3.09 (N = 339) | <.0001 |
| 2 year: Leg pain (median (min-max)) | .0 (.0-10.0) | 1.0 (.0-10.0) | 2.0 (.0-10.0) | |
| 90 day: PROMIS score (mean ± SD (N)) | 45.7 ± 8.0 (N = 434) | 44.8 ± 8.3 (N = 715) | 43.6 ± 8.7 (N = 747) | .0003 |
| 1 year: PROMIS score (mean ± SD (N)) | 47.0 ± 7.7 (N = 315) | 46.2 ± 8.4 (N = 536) | 44.6 ± 8.9 (N = 541) | <.0001 |
| 2 year: PROMIS score (mean ± SD (N)) | 48.0 ± 8.7 (N = 212) | 48.4 ± 8.7 (N = 327) | 45.3 ± 9.5 (N = 347) | <.0001 |
| 90 day: EQ-5D score (mean ± SD (N)) | .82 ± .18 (N = 363) | .80 ± .20 (N = 598) | .76 ± .23 (N = 634) | <.0001 |
| 1 year: EQ-5D score (mean ± SD (N)) | .83 ± .17 (N = 310) | .82 ± .20 (N = 516) | .77 ± .22 (N = 508) | <.0001 |
| 2 year: EQ-5D score (mean ± SD (N)) | .83 ± .19 (N = 213) | .84 ± .18 (N = 338) | .77 ± .22 (N = 355) | <.0001 |
| Return to work at 90 days | ||||
| No | 47 (19.3%) | 103 (24.5%) | 102 (26.4%) | .12 |
| Yes | 197 (80.7%) | 318 (75.5%) | 284 (73.6%) | |
| Return to work at 1 year | ||||
| No | 23 (11.2%) | 51 (15.6%) | 60 (20.5%) | .021 |
| Yes | 182 (88.8%) | 275 (84.4%) | 233 (79.5%) | |
| Return to work at 2 years | ||||
| No | 19 (14.2%) | 29 (13.7%) | 31 (15.0%) | .93 |
| Yes | 115 (85.8%) | 183 (86.3%) | 176 (85.0%) | |
Figure 2.
Mean leg pain scores.
Figure 3.
Mean PROMIS PF scores.
Figure 4.
Mean EQ-5D scores.
Analysis of Return to Work
Patients who had symptoms less than 3 months prior to surgery were more likely to return to work (P < .05) at 1 year after surgery (return to work rate = 88.8% vs 84.4% vs 79.5% for Group 1, Group 2, and Group 3, respectively) (Figure 5). When comparing Group 1 to both Group 2 and Group 3, minimal clinically improved differences (MCID) were achieved by a higher proportion of patients after surgery for PROMIS PF at all time points (P < .05), EQ-5D at 1 year and 2 years (P < .05), and leg pain at 90 days and 1 year (P < .05).
Figure 5.
Return to work rates.
Multivariate Analysis of Metrics
Table 4 presents the multivariate results of the normal GEE models for continuous leg pain, PROMIS PF, and EQ-5D at 90 days, 1 year, and 2 years. After adjusting for covariates, Group 3 had a lower average post-operative PROMIS PF score compared to Group 1 at 90 days (1.68 points lower, P = .004), 1 year (2.01 points lower, P < .05), and 2 years (2.23 points lower, P < .05). In addition, the post-operative PROMIS PF score at 90 days was lower in Group 2 compared to Group 1 (1.37 points lower, P = .008). The average EQ-5D at 1 year was also lower for Group 3 compared to Group 1 (.04 points lower, P < .05). When looking at leg pain, post-operative scores were on average .49 points higher at 90 days (P < .05), .98 points higher at 1 year (P < .05), and .93 points higher at 2 years (P < .05) for Group 3 compared to Group 1. Leg pain scores were on average .36 points higher for Group 2 compared to Group 1 (P < .05).
Table 4.
Multivariate Analysis for Continuous Leg Pain, PROMIS Scores, EQ-5D Scores, and Return to Work Rate Post-Op.
| Outcome | Symptom Duration (Ref: Less than 3 months) | B a | 95% LCL | 95% UCL | P-Value |
|---|---|---|---|---|---|
| Leg pain at 90 days | 1 year or more | .49 | .11 | .87 | .011 |
| Greater than 3 months but less than 1 year | .11 | −.21 | .42 | .503 | |
| Leg pain at 1 year | 1 year or more | .98 | .54 | 1.42 | <.001 |
| Greater than 3 months but less than 1 year | .36 | .02 | .69 | .037 | |
| Leg pain at 2 years | 1 year or more | .93 | .55 | 1.32 | <.001 |
| Greater than 3 months but less than 1 year | .11 | −.26 | .48 | .562 | |
| PROMIS at 90 days | 1 year or more | −1.68 | −2.82 | −.53 | .004 |
| Greater than 3 months but less than 1 year | −1.37 | −2.38 | −.36 | .008 | |
| PROMIS at 1 year | 1 year or more | −2.01 | −3.08 | −.93 | <.001 |
| Greater than 3 months but less than 1 year | −.97 | −2.04 | .10 | .076 | |
| PROMIS at 2 years | 1 year or more | −2.23 | −3.15 | −1.31 | <.001 |
| Greater than 3 months but less than 1 year | −.10 | −1.11 | .90 | .841 | |
| EQ-5D at 90 days | 1 year or more | −.03 | −.06 | .00 | .065 |
| Greater than 3 months but less than 1 year | −.01 | −.03 | .01 | .358 | |
| EQ-5D at 1 year | 1 year or more | −.04 | −.07 | −.01 | .013 |
| Greater than 3 months but less than 1 year | −.01 | −.03 | .01 | .470 | |
| EQ-5D at 2 years | 1 year or more | −.03 | −.06 | .01 | .147 |
| Greater than 3 months but less than 1 year | .03 | −.01 | .06 | .109 | |
| Return to work at 90 days b | 1 year or more | .91 | .86 | .97 | .005 |
| Greater than 3 months but less than 1 year | .94 | .90 | .99 | .017 | |
| Return to work at 1 year b | 1 year or more | .89 | .84 | .94 | <.001 |
| Greater than 3 months but less than 1 year | .94 | .89 | .99 | .012 | |
| Return to work at 2 years 2 | 1 year or more | .93 | .87 | 1.00 | .059 |
| Greater than 3 months but less than 1 year | .95 | .87 | 1.02 | .173 |
aVariables adjusted for: age, gender, race, education, BMI, diabetes, DVT, CAD, depression, anxiety, osteoporosis, ASA >2, independent ambulation pre-op, private insurance, previous spine surgery, hypertension, current smoking status, and baseline scores (as appropriate).
bAmong patients planning to return to work.
Univariate & Multivariate Analysis of MCID
From the univariate analysis (Table 5), it was observed that patients who experienced symptoms for less than 3 months (Group 1) were more likely to show clinically significant improvements, reaching the minimum clinically important difference (MCID) in leg pain, PROMIS PF, and EQ-5D scores across all post-operative time points (90 days, 1 year, and 2 years), compared to the other groups (Group 2 and Group 3). The differences in MCID achievement were statistically significant for both leg pain and PROMIS PF scores at all time points and for EQ-5D scores at 1 year and 2 years.
Table 5.
Univariate Analysis for MCID Leg Pain Scores, PROMIS Scores, and EQ-5D Scores Post-Op.
| <3 months (N = 543) | >3 months but <1 year (N = 910) | 1 year or > (N = 955) | P-Value | |
|---|---|---|---|---|
| MCID: 90 Day Leg Pain | ||||
| No | 58 (14.2%) | 114 (16.9%) | 154 (21.5%) | .0053 |
| Yes | 351 (85.8%) | 560 (83.1%) | 562 (78.5%) | |
| MCID: Leg pain at 1 year | ||||
| No | 28 (9.3%) | 81 (15.9%) | 128 (24.9%) | <.0001 |
| Yes | 272 (90.7%) | 427 (84.1%) | 386 (75.1%) | |
| MCID: Leg pain at 2 years | ||||
| No | 29 (14.4%) | 52 (16.5%) | 70 (22.2%) | .051 |
| Yes | 172 (85.6%) | 263 (83.5%) | 245 (77.8%) | |
| MCID: PROMIS at 90 Days | ||||
| No | 79 (18.9%) | 189 (28.2%) | 225 (31.6%) | <.0001 |
| Yes | 338 (81.1%) | 482 (71.8%) | 488 (68.4%) | |
| MCID: PROMIS at 1 Year | ||||
| No | 45 (14.9%) | 109 (21.3%) | 156 (30.4%) | <.0001 |
| Yes | 258 (85.1%) | 402 (78.7%) | 357 (69.6%) | |
| MCID: PROMIS at 2 Years | ||||
| No | 30 (14.6%) | 49 (15.6%) | 106 (31.5%) | <.0001 |
| Yes | 175 (85.4%) | 265 (84.4%) | 231 (68.5%) | |
| EQ-5D MCID at 90 days | ||||
| No | 99 (31.9%) | 178 (34.6%) | 204 (38.1%) | .18 |
| Yes | 211 (68.1%) | 337 (65.4%) | 332 (61.9%) | |
| EQ-5D MCID at 1 Year | ||||
| No | 72 (29.6%) | 120 (29.0%) | 159 (39.7%) | .0022 |
| Yes | 171 (70.4%) | 294 (71.0%) | 242 (60.3%) | |
| EQ-5D MCID at 2 years | ||||
| No | 47 (28.3%) | 63 (24.5%) | 112 (41.9%) | <.0001 |
| Yes | 119 (71.7%) | 194 (75.5%) | 155 (58.1%) | |
From the multivariate analysis (Table 6), patients who had symptoms for 1 year or more (Group 3) had a less likely chance to meet the minimum clinically important difference (MCID) for leg pain scores at 1 year (Relative Risk (RR): .86, 95% Confidence Interval (CI): [.80, .92], P < .05), PROMIS PF scores at 90 days (RR: .94, 95% CI: [.88, 1.00], P = .047), 1 year (RR: .88, 95% CI: [.82, .95], P = .002), and 2 years (RR: .87, 95% CI: [.80, .95], P = .003), as well as EQ-5D scores at 2 years (RR: .88, 95% CI: [.79, .98], P = .021).
Table 6.
Multivariate Analysis for MCID Leg Pain Scores, PROMIS Scores, and EQ-5D Scores Post-op.
| Outcome | Symptom Duration (Ref: Less than 3 months) | Relative Risk | RR LCL | RR UCL | P-Value |
|---|---|---|---|---|---|
| Leg pain MCID 90 days | 1 year or more | .95 | .89 | 1.01 | .096 |
| Greater than 3 months but less than 1 year | .99 | .94 | 1.04 | .576 | |
| Leg pain MCID 1 year | 1 year or more | .86 | .80 | .92 | <.001 |
| Greater than 3 months but less than 1 year | .94 | .89 | 1.00 | .058 | |
| Leg pain MCID 2 years | 1 year or more | .95 | .89 | 1.01 | .110 |
| Greater than 3 months but less than 1 year | .98 | .91 | 1.04 | .488 | |
| PROMIS MCID 90 days | 1 year or more | .94 | .88 | 1.00 | .047 |
| Greater than 3 months but less than 1 year | .94 | .89 | 1.00 | .043 | |
| PROMIS MCID 1 year | 1 year or more | .88 | .82 | .95 | .002 |
| Greater than 3 months but less than 1 year | .97 | .90 | 1.04 | .346 | |
| PROMIS MCID 2 years | 1 year or more | .87 | .80 | .95 | .003 |
| Greater than 3 months but less than 1 year | 1.04 | .95 | 1.13 | .424 | |
| EQ-5D MCID 90 days | 1 year or more | .98 | .88 | 1.10 | .771 |
| Greater than 3 months but less than 1 year | 1.02 | .94 | 1.12 | .558 | |
| EQ-5D MCID 1 year | 1 year or more | .93 | .83 | 1.06 | .279 |
| Greater than 3 months but less than 1 year | 1.08 | .98 | 1.18 | .102 | |
| EQ-5D MCID 2 years | 1 year or more | .88 | .79 | .98 | .021 |
| Greater than 3 months but less than 1 year | 1.09 | .99 | 1.20 | .082 |
However, other comparisons did not show statistically significant results. Specifically, there was no significant difference in achieving MCID for leg pain scores at 90 days (RR: .95, 95% CI: [.89, 1.01], P = .096), EQ-5D scores at 90 days (RR: .98, 95% CI: [.88, 1.10], P = .771) and 1 year (RR: .93, 95% CI: [.83, 1.06], P = .279) for Group 3.
For patients who had symptoms for more than 3 months but less than 1 year (Group 2), none of the comparisons reached statistical significance, including PROMIS PF scores at 1 year (RR: .97, 95% CI: [.90, 1.04], P = .346) and 2 years (RR: 1.04, 95% CI: [.95, 1.13], P = .424), EQ-5D scores at 90 days (RR: 1.02, 95% CI: [.94, 1.12], P = .558), 1 year (RR: 1.08, 95% CI: [.98, 1.18], P = .102), and 2 years (RR: 1.09, 95% CI: [.99, 1.20], P = .082).
Discussion
While conservative management is often the first line of treatment for symptomatic lumbar disc herniation, surgery is ultimately required in about 10% of cases. 18 Despite the high success with standard open lumbar discectomy, less than ideal outcomes may occur in patients with prolonged duration of pre-operative symptoms. Prior Spine Patient Outcomes Research Trial (SPORT) and observational studies have associated worse post-operative pain and quality-of-life outcomes with longer duration of pre-operative symptoms. 19 In their SPORT study (n = 1192), Rihn et al 19 reported that patients who had symptoms for more than six months reported lower increases on Short Form-36 and higher scores in the Oswestry Disability Index score when compared to patients who had symptoms for less than six months following both operative and conservative measures. 19 Kerr et al also conducted a SPORT study (n = 1244) which found that patients exhibiting symptoms less than or equal to six months had significantly better outcomes with both surgical and nonoperative treatment cohort. 20 However, the authors note that patients who had symptoms longer than six months benefitted relatively more from surgery than other patient groups. Ng and Sell 18 reported in their prospective cohort study (n = 113) that patients who had sciatica for more than 12 months had an overall less favorable outcome. Peul et al 21 noted in their multi-center prospective study (n = 283) that patients who had early surgery reported faster relief of leg pain and faster rate of perceived recovery compared to those who underwent conservative management and eventual surgery. The purpose of this study was to examine whether the duration of symptoms a patient experiences prior to lumbar microdiscectomy affects their outcomes after surgery.
The principal findings of our study showed that even with similar baseline scores, patients who undergo surgery following a reported symptom duration of longer than 3 months had significantly worse outcomes (increased leg pain and lower PROMIS PF and EQ-5D scores) at 90 days, 1 year, and 2 years post-operative compared to those patients with symptom duration of less than 3 months. Patients who had symptoms for over a year before undergoing surgery (Group 3) were observed to have lower average post-operative PROMIS PF scores at various time points (90 days, 1 year, and 2 years) compared to patients with symptoms that persisted for less than 3 months (Group 1). Furthermore, patients from Group 3 experienced higher average post-operative leg pain scores at these same time intervals relative to Group 1, as evidenced by our multivariate analysis. Univariate analysis revealed that a greater proportion of patients with symptom duration for less than 3 months achieved MCID as well as attained higher return to work rates at 1-year follow-up compared to other groups. Specifically, 88.8% of subjects in group 1 were able to return to work within a year following their surgery, in contrast to the 84.4% in Group 2 and 79.5% in Group 3 (P < .05). This information can help guide physicians when counseling patients on a timeline of treatment or possible treatment options when treating their lumbar disc herniation. While there is no optimal duration of conservative treatment before the consideration of surgery, our findings suggest that early intervention offers the best outcomes after surgery.
MCID is a valuable metric for interpreting the clinical relevance of statistical differences, particularly in patient-reported outcomes. It provides a benchmark for the smallest difference in scores that patients perceive as beneficial, thus bridging the gap between statistical significance and clinical relevance. Despite the observed 2.23 points lower post-operative PROMIS PF score in Group 3 compared to Group 1 at 2 years, our multivariate MCID results give a deeper perspective. For instance, when considering MCID in PROMIS PF scores, we found that Group 3 was significantly less likely to achieve MCID at 90 days (RR = .94, P < .05), 1 year (RR = .88, P < .05), and 2 years (RR = .87, P < .05) compared to Group 1.
Importantly, all groups displayed significant improvements in leg pain and lifestyle metrics compared to baseline, suggesting that patients can be assured that they may have relative improvement post-operation regardless of the duration of their symptoms. Thus, consistent with the findings of Siccoli et al as well as other prior studies, earlier surgical intervention can lead to better postoperative clinical and functional outcomes.2,21–24
Limitations
The present study is not without limitations. Differences in baseline data including the location of lumbar disc herniation and operative time between groups may have altered the results as our study also showed significant differences between patient groups at baseline in relation to factors like education level, diagnoses of depression and anxiety, pre-operative independent ambulation, private insurance status, smoking habits, and return to work plans. One important consideration is that nearly 33% of patients with symptoms greater than 12 months presented with concurrent depression which was significantly higher compared to other patient groups. Physicians may bias against patients with a psychiatric diagnosis and understate the severity of their reported statements. These findings are consistent with epidemiological studies which have revealed that nearly 1 in 5 patients with symptomatic lumbar stenosis report depression.19,22 Additionally, higher preoperative mental health scores have been shown to produce better patient satisfaction and outcome scores, suggesting that depression may skew reported patient data.25,26 Nevertheless, our study suggests that patients with longer duration of symptoms have higher rates of depression and the avoidance of this bias may be unavoidable. The retrospective nature is a limitation in its own, but we believe the large number of patients provides a signal that duration of symptoms may influence outcomes after microdiscectomy.
Conclusion
Microdiscectomy for lumbar disc herniation within 3 months of symptom onset leads to greater improvements in postoperative leg pain, PROMIS PF, and EQ-5D than if surgery is performed after 3 months of symptom onset. Furthermore, patients undergoing surgery within 3 months of symptom onset have the highest rates of return to work at 1 year after surgery. These results suggest that the swift treatment of lumbar disc herniations within 3 months gives patients the best outcomes after surgery. While our study does suggest a potential influence of symptom duration on certain surgical outcomes, this relationship is likely complex and may be influenced by various patient-specific factors. The direct causal relationship between symptom duration and surgical outcomes warrants further exploration.
Footnotes
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: All authors have no relationships related to this study specifically. Although BCBSM and MSSIC work collaboratively, the opinions, beliefs, and viewpoints expressed by the authors do not necessarily reflect the opinions, beliefs, and viewpoints of BCBSM or any of its employees. Support for MSSIC is provided by BCBSM and Blue Care Network as part of the BCBSM Value Partnerships program. Although BCBSM and MSSIC work collaboratively, the opinions, beliefs, and viewpoints expressed by the authors do not necessarily reflect the opinions, beliefs, and viewpoints of BCBSM or any of its employees. Support for MSSIC is provided by BCBSM and Blue Care Network as part of the BCBSM Value Partnerships program.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical & Disclaimer Statements
Ethical Approval
This study was reviewed by Beaumont Hospital IRB committee and was determined to be exempt from institutional review board approval (IRB number # 2019-379).
Disclaimer Statement
Although Blue Cross Blue Shield of Michigan and MSSIC work collaboratively, the opinions, beliefs and viewpoints expressed by the authors do not necessarily reflect the opinions, beliefs, and viewpoints of BCBSM or any of its employees.
Support Statement
Support for MSSIC is provided by Blue Cross and Blue Shield of Michigan and Blue Care Network as part of the BCBSM Value Partnerships program.
ORCID iDs
Rafid Kasir https://orcid.org/0000-0001-5243-3770
Sazid Hasan https://orcid.org/0000-0001-8555-8721
Ilyas Aleem https://orcid.org/0000-0003-4818-8578
Ehab S. Saleh https://orcid.org/0000-0001-7064-3448
References
- 1.Ahmadi SA, Burkert I-P, Steiger H-J, Eicker SO. Multidimensional long-term outcome analysis after single-level lumbar microdiscectomy: a retrospective single-centre study. Eur J Orthop Surg Traumatol. 2018;28:189-196. [DOI] [PubMed] [Google Scholar]
- 2.Alentado VJ, Lubelski D, Steinmetz MP, Benzel EC, Mroz TE. Optimal duration of conservative management prior to surgery for cervical and lumbar radiculopathy: a literature review. Global Spine J. 2014;4:279-286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Michigan Spine Surgery Improvement Collaborative. New York, NY, USA: (MSSIC). 2023. [Google Scholar]
- 4.Al Qaraghli MI, De Jesus O. Lumbar Disc Herniation. Treasure Island, FL: StatPearls; 2023. [PubMed] [Google Scholar]
- 5.Feng F, Xu Q, Yan F, et al. Comparison of 7 surgical interventions for lumbar disc herniation: a Network meta-analysis. Pain Physician. 2017;20:E863-E871. [PubMed] [Google Scholar]
- 6.Basques BA, Haws BE, Khechen B, et al. The effect of preoperative symptom duration on postoperative outcomes after a tubular lumbar microdiscectomy. Clinical Spine Surgery: A Spine Publication. 2019;32:E27-E30. [DOI] [PubMed] [Google Scholar]
- 7.Campbell R, Ju A, King MT, Rutherford C. Perceived benefits and limitations of using patient-reported outcome measures in clinical practice with individual patients: a systematic review of qualitative studies. Qual Life Res. 2022;31:1597-1620. [DOI] [PubMed] [Google Scholar]
- 8.Gugliotta M, da Costa BR, Dabis E, et al. Surgical versus conservative treatment for lumbar disc herniation: a prospective cohort study. BMJ Open. 2016;6:e012938. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hamawandi SA, Sulaiman II, Abdulhameed AM. Effect of duration of symptoms on the clinical and functional outcomes of lumbar microdiscectomy: a randomized controlled trial. Orthop Surg. 2022;14:157-168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Jansson K-A., Németh G, Granath F, Jönsson B, Blomqvist P. Health-related quality of life in patients before and after surgery for a herniated lumbar disc. J Bone Joint Surg. 2005;87-B:959-964. [DOI] [PubMed] [Google Scholar]
- 11.Kerr D, Zhao W, Lurie JD. What are long-term predictors of outcomes for lumbar disc herniation? A randomized and observational study. Clin Orthop Relat Res. 2015;473:1920-1930. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lawless MH, Tong D, Claus CF, et al. The effect of preoperative symptom duration on patient-reported outcomes after anterior cervical discectomy and fusion in nonmyelopathic patients: analyses from the Michigan spine surgery improvement collaborative (MSSIC). Neurosurgery. 2023;92:955-962. [DOI] [PubMed] [Google Scholar]
- 13.Hamilton T, Bartlett S, Deshpande N, et al. Association of prolonged symptom duration with poor outcomes in lumbar spine surgery: a Michigan Spine Surgery Improvement Collaborative study. J Neurosurg Spine. 2023;39(4):1-10. doi: 10.3171/2023.5.SPINE23249 [DOI] [PubMed] [Google Scholar]
- 14.Peul WC, Hout WB, Brand R, Thomeer RTWM, Koes BW. Prolonged conservative care versus early surgery in patients with sciatica caused by lumbar disc herniation: two year results of a randomised controlled trial. BMJ. 2008;336:1355-1358. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Movassaghi K, Basques BA, Louie PK, et al. The duration of symptoms does not impact clinical outcomes following lumbar decompression surgery. Spine 1976 2019;44:305-308. [DOI] [PubMed] [Google Scholar]
- 16.Lutz GK, Butzlaff ME, Atlas SJ, Keller RB, Singer DE, Deyo RA. The relation between expectations and outcomes in surgery for sciatica. J Gen Intern Med. 1999;14:740-744. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Myers SS, Phillips RS, Davis RB, et al. Patient expectations as predictors of outcome in patients with acute low back pain. J Gen Intern Med. 2008;23:148-153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ng LCL, Sell P. Predictive value of the duration of sciatica for lumbar discectomy. J Bone Joint Surg. 2004;86-B:546-549. [PubMed] [Google Scholar]
- 19.Rihn JA, Hilibrand AS, Radcliff K, et al. Duration of symptoms resulting from lumbar disc herniation: effect on treatment outcomes. J Bone Joint Surg. 2011;93:1906-1914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Nygaard ØP, Kloster R, Solberg T. Duration of leg pain as a predictor of outcome after surgery for lumbar disc herniation: a prospective cohort study with 1-year follow up. J Neurosurg Spine. 2000;92:131-134. [DOI] [PubMed] [Google Scholar]
- 21.Peul WC, van Houwelingen HC, van den Hout WB, et al. Surgery versus prolonged conservative treatment for sciatica. N Engl J Med. 2007;356:2245-2256. [DOI] [PubMed] [Google Scholar]
- 22.Kessler RC, Berglund P, Demler O, et al. The epidemiology of major depressive disorder. JAMA. 2003;289:3095. [DOI] [PubMed] [Google Scholar]
- 23.Siccoli A, Wispelaere MPde, Schröder ML, Staartjes VE. Timing of surgery in tubular microdiscectomy for lumbar disc herniation and its effect on functional impairment outcomes. Neurospine. 2020;17:204-212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Yoon WW, Koch J. Herniated discs: when is surgery necessary? EFORT Open Reviews. 2021;6:526-530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rushton A, Zoulas K, Powell A, Staal J. Physical prognostic factors predicting outcome following lumbar discectomy surgery: systematic review and narrative synthesis. BMC Muscoskel Disord. 2018;19:326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Schoeggl A, Maier H, Saringer W, Reddy M, Matula C. Outcome after chronic sciatica as the only reason for lumbar microdiscectomy. J Spinal Disord Tech. 2002;15:415-419. [DOI] [PubMed] [Google Scholar]