Inappropriate patient selection continues to be one of the most common causes of failed laminectomy syndrome in the surgical treatment of lumbar disc herniation.7,8,17,21,23,27,29,31 The failure rate reported in the literature varies from 1% to 48% and averages between 10% and 20% in most series.1,2,5,6,9-13,16,18-20,24-26,30 An initial ill-advised surgery is often followed by further operations that fail to improve the patient's condition. Patients who continue to report severe pain and who remain disabled for months or years after back surgery pose tremendous socioeconomic as well as psychological costs.
In 1985, Herron and Turner reported their use of the revised objective rating system for selecting patients for lumbar laminectomy and discectomy from among 106 patients treated for disc herniation.14 This revised system had been modified from that developed by Spengler and Freeman28 and Spengler et al.30 The revised objective rating score was found to be highly predictive of surgical outcome. The authors have continued to use the objective rating system and in the current study have reported their results with 275 patients with at least 1 year of followup.
MATERIALS AND METHODS
Between 1979 and 1992, lumbar laminectomy and discectomy for disc herniation were performed for 291 patients. All patients literate in English completed a Minnesota Multiphasic Personality Inventory before surgery; 195 completed the Minnesota Multiphasic Personality Inventory, and 47 completed the Minnesota Multiphasic Personality Inventory-2 which replaced the Minnesota Multiphasic Personality Inventory in 1991. All patients (with the exception of 1 blind patient) completed a pain drawing.22 Followup of at least 1 year was done for 275 patients (173 males, 102 females). Two patients had died of causes unrelated to surgery, and 14 patients did not return for a followup evaluation 1 year or more after surgery.
The diagnosis of disc herniation was made on the basis of medical history, physical examination, and an appropriate imaging study. Three percent of patients presented with bowel or bladder dysfunction and cauda equina syndrome and thus underwent immediate surgery. Otherwise, the patients were first treated with conservative therapy, to which they did not respond. Conservative treatment consisted of a minimum of back school, analgesics and/or antiinflammatory medications, and rest or modified activities. Before evaluation by the surgeon, most patients had also been treated with physical therapy and spinal manipulation therapy.
The average age of the patient at time of surgery was 43 years (range, 15-83 years). Patient characteristics are listed in Table 1. The average duration of preoperative leg pain symptoms was 7 months (range, 1-60 months). Subjective complaints are listed in Table 2. Examination findings are listed in Table 3.
Before surgery. 248 patients underwent a myelogram or magnetic resonance imaging (MRI) alone, 27 patients underwent a computed tomography (CT) scan alone, and 69 patients underwent a CT scan combined with an MRI or myelogram. The myelogram or MRI defect was graded by the surgeon as severe in 25% of the patients, moderate with lateralization in 67%, and moderate ventral without lateralization or minimal lateral defect in 8%. The CT scan defect was graded as severe in 17% of the patients, moderate with lateralization in 77%, and moderate without lateralization or minimal lateralization in 6%.
A unilateral laminectomy and discectomy were performed in 71% of patients and a bilateral procedure in 29%. Nine percent of patients were noted to have significant associated spinal stenosis. Five percent of patients underwent a 2-level decompressive laminectomy for associated spinal stenosis. Ten patients underwent 2-level discectomy. Discectomy was unilateral in 77% of patients and bilateral in 23%. Laminectomy level was L2-3 in 4 patients, L3-4 in 21 patients, L4-5 in 140 patients, and L5-S1 in 144 patients. A 1-level laminectomy was performed in 241 patients and a 2-level laminectomy in 34 patients. Eight patients underwent associated fusion: 1, L4 to sacrum fusion; 1, L5-S1 fusion; and 6, L4-5 fusion. Disc disease was degenerative in 2 patients (1%), protruded in 135 patients (49%), extruded in 101 patients (37%), and sequestrated in 37 patients (13%).
After surgery, patients were maintained on modified bed rest with bathroom privileges and a twice-daily walk to tolerance for 2 weeks. They then attended a postoperative back school program for spinal education and conditioning exercises. After this, they were encouraged to begin lap swimming, aquakinetics, or a Nautilus gym program. Gradual return to normal activities was allowed by 6 to 8 weeks with appropriate ergonomic and recreational restrictions.
The revised objective rating system, summarized in Table 4, was used to provide a numeric score for each patient before surgery. A maximum score of 25 points in each of 4 categories (neurologic signs, root tension signs, imaging study, and psychosocial environment) could be awarded, summing to a possible 100 points. In the current series, the objective rating score averaged 58 points (range, 15-100 points).
At the followup examination, the patients were interviewed and examined by the surgeon. Patient outcome was graded with the system used previously, which includes categories for back and leg pain relief, return to work, restriction of activities, and analgesic use (Table 5).13 The study was prospective, and patients were seen for followup at 3 weeks, 6 weeks, 3 months, 6 months, 1 year, and yearly thereafter. Followup ranged from 1 to 12.8 years, with an average of 4 years. All analyses of followup data were based on the longest followup assessment available. Data were analyzed by means of descriptive statistics, Chi-squared analyses, and logistic regression analyses to determine the probabilities of good, fair, and poor outcomes given specific patient characteristics. All significance levels were determined using alpha = 0.05.
On the basis of the surgical outcome grading scale, 226 (82%) patients had good outcomes; 26 (10%), fair outcomes; and 23 (8%), poor outcomes at long term followup (Table 6). The objective rating score was highly associated with outcomes at followup (Chisquared = 93.396, p < 0.001). Table 7 shows the distribution of the objective rating scores in the good, fair, and poor outcome categories. The average preoperative objective rating score was 34 (range, 15-60, standard deviation = 14.2) among the poor outcome patients, 47 (range, 25-75, standard deviation = 16.4) among the fair outcome patients, and 62 (range, 25-100, standard deviation = 16.8) among the good outcome patients.
Further analysis was performed to determine the association of the individual components of the rating method with outcome. Preoperative sensory deficit, motor weakness and reflex asymmetry were not associated with outcome, nor was anatomic sensory deficit and motor weakness.
Excluding the 20 patients who underwent L2-3 or L3-4 discectomy (who would be expected to have normal straight-leg raising), unilateral straight-leg raising restriction because of leg pain was significantly associated with the surgical result. Again, excluding the 20 patients with L2-3 or L3-4 discectomy, unilateral straight-leg raising of <60 ° was present before surgery in 141 patients, of whom 130 subsequently had good outcomes; 6, fair; and 5, poor. There were 65 good, 12 fair, and 13 poor outcomes among the 90 patients with straight-leg raising restricted to 60 ° to 75 °. For patients with unilateral straight-leg raising of <60 °, the probability of a poor outcome was 3.6%; of a fair outcome, 4.3%; and of a good outcome, 92.1%. For patients with unilateral straight leg raising of 60 ° to 75 °, the corresponding probabilities were 14.5%, 13.3%, and 72.2% for poor, fair, and good outcomes, respectively.
Positive results of a Lasegue test were present in 136 patients, of whom subsequent outcomes were good for 123, fair for 7, and poor for 6. The Lasegue test was significantly associated with outcome (Chi-squared = 12.54, p = 0.002). For patients with negative Lasegue test results, the probability of a poor outcome was 12.2%; of a fair outcome, 13.7%; and of a good outcome, 74.1%. For patients with a positive Lasegue test result, the corresponding probabilities were 4.4%, 5.2%, and 90.4%.
A cross straight-leg raising response, the Cram or bowstring test, and a list or flexed-knee stance were not significantly associated with outcome. For patients with a cross straight-leg raising response, 46 had good, 3 had fair, and 2 had poor results. The Cram or bowstring test results were positive for 230 patients, with 193 good, 18 fair, and 19 poor results. The test results were negative for 45 patients, with 33 good, 8 fair, and 4 poor results. A list or flexed-knee stance was seen in 96 patients, with 82 good, 7 fair and 7 poor results.
The myelogram or MRI defect was not associated with surgical outcome, probably because of the small number of patients who did not have a severe or moderate defect with lateralization. Of the 226 patients with good results who underwent MRI or myelogram before surgery, 211 (93%) had either a severe or moderate defect with lateralization.
Evaluation of the individual components of the psychosocial portion of the rating system was also performed. Patients with compensation or litigation issues were significantly more likely to have poor outcomes (Chi-squared = 55.65, p = 0.00). As shown in Table 6, the 89 patients with compensation or litigation issues had 52 (58%) good, 16 (18%) fair, and 21 (24%) poor results. Of the 186 patients without compensation or litigation issues pending surgery, 174 (94%) had good; 10 (5%), fair; and 2 (1%), poor results. If the patient was involved in compensation or litigation, his/her probability of a poor outcome was 21.4%; of a fair outcome, 21.4%; and of a good outcome, 57.2%. Patients with no compensation or litigation issues had corresponding probabilities of 2.4%, 4.0%, and 93.6%.
Patients who had been employed before surgery had better outcomes than those who were unemployed (Chi-squared = 19.32, p = 0.0001). Of the 161 patients employed at the time of surgery, 144 (89%) had good results; 13 (8%), fair results; and 4 (3%), poor results versus 82 (72%), good results; 13 (11%), fair results; and 19 (17%) poor results for unemployed patients. For patients employed at the time of surgery, the probability of a poor outcome was 4.5%; of a fair outcome, 14.8%; and of a good outcome, 89.7%. For patients who were unemployed at the time of surgery, the corresponding probabilities were 14.1%, 14.8%, and 71.1%. The Minnesota Multiphasic Personality Inventory Hypochondriasis (F = 16.29, p < 0.01) and Hysteria (F = 14.25, p < 0.01) subscales were also significantly related to outcome.
Age, gender, subjective weakness or numbness, reports of bowel or bladder dysfunction, and gait disturbance were not significantly related to outcome. The probability of a good outcome decreased with increasing numbers of previous surgeries, although only 7 patients had undergone >1 prior surgery.
Excluding the 10 patients who had a 2-level discectomy, patients with L2-3 or L3-4 discectomy had poorer outcomes than did those with L4-5 or L5-S1 discectomy. At the L2-3 and L3-4 levels, there were 10 (63%) good, 5 (31%) fair, and 1 (6%) poor surgical results. Those patients treated with L4-5 discectomy had 97 good (84%), 11 fair (9%), and 8 poor (7%) results. Those undergoing L5-S1 discectomy had 95 (85%) good, 8 (7%) fair, and 9 (8%) poor surgical results. Patients with surgery at the L4-5 level did not differ significantly in outcomes from those with surgery at the L5-S1 level.
The revised objective rating score assigned before surgery was again found to be highly associated with patient outcomes at long-term followup after laminectomy and discectomy for lumbar disc herniation. The use of this system eliminates reliance on much of the subjective interpretation of physical findings, imaging results, and psychosocial factors. The system weights these parameters by assigning varying values to lesser or more severe findings. One of the strengths of the system is that the sum of the individual parts of the score is more highly predictive of outcome than is any single variable.
The interpretation of the imaging study continues to be of paramount importance in this system. Too frequently, the radiology report of a small central disc herniation or protrusion is used to justify surgery, when in fact the patient has an annular bulge consistent with degenerative disc disease. Minimal ventral defects should be ignored. Moderate ventral defects without lateralization or minimal lateral defects are graded low on this system. Moderate defects with lateralization and severe defects are significant predictors of surgical success (93% good outcome).
Whether the neurologic signs portion of the objective rating score is necessary remains to be determined. These items are part of the routine examination of patients and traditionally have been useful in the diagnosis of disc herniation. However, objective weakness or sensory deficit analyzed alone or together were not predictive of patient outcomes in the current series. Similarly, reflex asymmetry, after controlling for level of discectomy (ankle reflex for L5-S1 laminectomies and knee reflex for L2-3 or L3-4 laminectomies), was not associated with outcome. With the current capabilities of MRI, it would appear that the neurologic findings, if present, merely confirm the level diagnosed by imaging and are not predictive of the outcome in and of themselves. The incidence of positive imaging findings in asymptomatic patients is well known,3,15,32 and it is important that the subjective reports and neurologic findings be consistent with the defect imaged with respect to side and level.
Positive root tension signs were significantly associated with outcome. Limitation of straight-leg raising to <60 °, and a positive Lasegue test result were associated with good patient outcomes. A positive result on the femoral nerve stretch test was also related to outcome in patients with L2-3 or L3-4 disc herniation. Straight-leg raising restriction should be unilateral and should be restricted by leg pain, not back pain, to be considered positive. A bilateral straight-leg raising restriction should be viewed skeptically unless accompanied by a large imaging defect or anatomic neurologic findings.
Psychosocial factors were highly associated with outcome. The presence of compensation or litigation issues was a strong risk factor for poor outcome. Among the patients with these issues, 58% had good; 18%, fair; and 24%, poor results. Among the patients without such issues, 94% had good; 5%, fair; and 1%, poor results. Similarly, unemployment at the time of surgery was a strong risk factor for poor outcome, and elevated scores on the Hypochondriasis and Hysteria subscales were associated with poor outcome as well. Further research is needed for determination of how best to treat patients with workers' compensation and litigation issues. This might involve the use of a different system of criteria for surgery or a higher score on the objective rating scale than would be required for patients without such issues. A complex set of issues involving multiple parties (multiple physicians, claims managers, employers, and attorneys) characterizes patients in the workers' compensation system. Better outcomes may not be achieved until major changes are made in the management of patients with work-related injuries from the onset of injury.
In the absence of a cauda equina syndrome or progressive motor weakness, patients with lumbar disc herniation deserve a trial of appropriate conservative care. At least 75% of patients so treated will recover and thereby avoid surgery. For patients who do not respond to conservative care and for those who wish to proceed with laminectomy and discectomy, the results of this study suggest that an objective rating system score of at least 30 or 40 points is needed before surgery can be undertaken. Patients scoring <30 points should be told that the probability of a successful outcome is inadequate to justify surgery. A score of 40 points or more indicates a reasonable probability of success of laminectomy and discectomy for lumbar disc herniation.
The current study did have limitations. First, 16 of the 291 patients were unavailable for followup. Possibly, these 16 patients were somehow different in a way that might have affected the relationships between the objective rating scores and followup results. However, the followup rate of 95% in this study makes it highly likely that the findings are valid for this patient population. Second, the outcome measure was a rating made by the surgeon on the basis of the patient's reports. Such a rating is influenced by biases on the part of the surgeon and the patient. There is a need for a replication study in which valid and reliable measures of pain and functioning are determined by an independent evaluator before surgery and at followup and in which analyses are conducted to examine the ability of the objective rating scale to predict pain and functioning (including work) at followup.
The authors thank Gloria Carrasco, CMT, and Capi Glines, CMT, for manuscript preparation.
1. Abramovitz JN, Leff SR: Lumbar disc surgery: Results of the prospective lumbar discectomy study of the Joint Section on Disorders of the Spine and Peripheral Nerves of the American Association of Neurological Surgeons and the Congress of Neurological Surgeons. Neurosurgery 29:301-308, 1991.
2. Balderston RA, Gilyard GG, Jones AM, et al: The treatment of lumbar disc herniation: Simple fragment excision versus disc space curettage. J Spinal Disord 4:22-25, 1991.
3. Boden SD, Davis DO, Dina TS, et al: Abnormal magnetic resonance scans of the lumbar spine in asymptomatic patients. A prospective investigation. J Bone Joint Surg 72A:403-408, 1990.
4. Cram RH: A sign of sciatic nerve root pressure. J Bone Joint Surg 35B:192-195, 1953.
5. Dvorak J, Gauchat MH, Valach L: The outcome of surgery for lumbar disc herniation. I. A 4-17 year's follow-up with emphasis on somatic aspects. Spine 13:1418-1422, 1988.
6. Ebersold MJ, Quast LM, Bianco Jr AJ: Results of lumbar discectomy in the pediatric patient. J Neurosurg 67:643-647, 1987.
7. Fager CA, Froidberg SR: Analysis of failures and poor results of lumbar spine surgery. Spine 5:87-94, 1980.
8. Finnegan WJ, Fenlin JM, Marvel MP, et al: Results of surgical intervention in the symptomatic multiply-operative back patient. J Bone Joint Surg 61A:1077-1082, 1979.
9. Frymoyer JW, Hanley E, Howe J, et al: Disc excision and spine fusion in the management of lumbar disc excision. A minimum ten-year follow-up. Spine 3:1-6, 1978.
10. Garfin SR, Glover M, Booth RE, et al: Laminectomy: A review of the Pennsylvania Hospital experience. J Spinal Disord 1:116-133, 1988.
11. Gurdjian ES, Osrowski AZ, Hardy WG, et al: Results of operative treatment of protruded and ruptured discs with 82% follow-up of 3-13 years. J Neurosurg 18:783-789, 1961.
12. Hakelius A: Prognosis in sciatica: A clinical follow-up of surgical and nonsurgical treatment. Acta Orthop Scand 129(Suppl):1-76, 1970.
13. Herron LD, Pheasant HC: Bilateral laminectomy and discectomy for segmental lumbar disc disease: Decompression with stability. Spine 8:86-97, 1983.
14. Herron LD, Turner J: Patient selection for lumbar laminectomy and discectomy with a revised objective rating system. Clin Orthop 199:145-152, 1985.
15. Hitselberger W, Whitten R: Abnormal myelograms in asymptomatic patients. J Neurosurg 28:204-206, 1968.
16. Hoffman RM, Wheeler KJ, Deyo RA: Surgery for herniated lumbar discs: A literature synthesis. J Gen Intern Med 8:487-497, 1993.
17. Holmes HE, Rothman RH: The Pennsylvania Plan: An Algorithm for the Management of Lumbar Disc Disease. AAOS Instructional Course Lectures. Vol 28. St. Louis, CV Mosby, 193-200, 1979.
18. Lewis PJ, Weir, BKA, Broad, RW, Grace MG: Long-term prospective study of lumbosacral discectomy. J Neurosurg 67:49-53, 1987.
19. Naylor A: The late results of laminectomy for lumbar disc prolapse. A review after ten to twenty-five years. J Bone Joint Surg 56B:17-29, 1974.
20. Pappas CTE, Harrington T, Sontag VKW: Outcome analysis in 654 surgically treated lumbar disc herniations. Neurosurgery 30:862-866, 1992.
21. Pheasant HC, Dyck P: Failed lumbar disc surgery: Cause, assessment, treatment. Clin Orthop 164: 93-109, 1982.
22. Ransford AO, Cairns D, Mooney V: The pain drawing as an aid to the psychological evaluation of patients with low back pain. Spine 1:127-134, 1976.
23. Rask M: Knee flexion test and sciatica. Clin Orthop 134:221-223, 1978.
24. Rish BL: A critique of the surgical management of lumbar disc disease in a private neurosurgical practice. Spine 5:500-504, 1984.
25. Salenius P, Laurent LE: Results of operative treatment of lumbar disc herniation. A survey of 886 patients. Acta Orthop Scand 48:630-634, 1977.
26. Spangfort EV: The lumbar disc herniation. A computer-aided analysis of 2,504 operations. Acta Orthop Scand 142(Suppl): 1-95, 1972.
27. Spengler DM: Lumbar discectomy: Results with limited disc excision and selective foraminotomy. Spine 7:604-607, 1982.
28. Spengler DM, Freeman CW: Patient selection for lumbar discectomy, an objective approach. Spine 4:129-134, 1979.
29. Spengler DM, Freeman W, Westbrook R, Miller JW: Low back pain following multiple lumbar spine procedures. Failure of initial selection? Spine 5:356-360, 1980.
30. Spengler DM, Quellette EA, Battie M, Zeh J: Elective discectomy for herniation of a lumbar disc. Additional experience with an objective method. J Bone Joint Surg 72A:230-237, 1990.
31. Waddell G, Kummel EG, Lotto WN, et al: Failed lumbar disc surgery and repeat surgery following industrial injuries. J Bone Joint Surg 61A:201-207, 1979.
32. Wiesel SW, Tsourmas N, Feffer HL, et al: A study of computer-assisted tomography: I. The incidence of positive CAT scans in an asymptomatic group of patients. Spine 9:549-551, 1984.
ORIGINAL ARTICLES© Lippincott-Raven Publishers.