AUTHOR(S): Pappas, Conrad T. E., M.D., Ph.D.; Harrington, Timothy, M.D.; Sonntag, Volker K. H., M.D. Division of Neurological Surgery, Barrow Neurological Institute, Phoenix, Arizona Neurosurgery 30; 862-866, 1992 ABSTRACT: This article reports the outcomes of 654 consecutive patients treated during a 4.5-year period. Patients had a micro-discectomy, a laminectomy plus microdiscectomy, or a decompressive laminectomy with a microdiscectomy. The causes of ruptured discs were lifting (31.4%), falls (10.2%), and sports (10.0%). Almost all patients had complained of leg pain (99%), and 79% had radicular pain in a dermatomal distribution. Thirtythree percent of the patients had been involved in industrial accidents, and 6% had legal claims pending during the surgical period. Almost 11% of the patients had complications, and there was one death caused by abdominal arterial bleeding. Patients were also rated according to the Prolo FunctionalEconomic Outcome Rating Scale to improve the ability to compare series in the future. Almost 80% of the patients had good outcomes as defined by scores on this scale of 8 (16.2%), 9 (33.2%), and 10 (26.9%). Several conclusions can be drawn from the results of this series: 1) most patients had good outcomes; 2) patients with nonindustrial injuries had better outcomes than did patients with industrial injuries; 3) professionals with legal concerns and laborers with industrial insurance had good outcomes; and 4) the Functional-Economic Outcome Rating Scale appears to be a useful tool for comparing different procedures more objectively and for comparing the outcomes across series. KEY WORDS: FunctionalEconomic Outcome Rating Scale; Herniated lumbar disc; Laminectomy; Microdiscectomy INTRODUCTION The surgical outcome for lumbar discectomy has improved as surgical techniques and patient selection have been refined. The use of the operating microscope has been widely accepted by neurosurgeons for more than 10 years (7,8,24,27,29,32). At our institution, lumbar microdiscectomy has been the method of choice of all teaching neurosurgeons for at least that long. This technique offers a small incision, excellent magnification, gentle handling of the nerve root, and good exposure. We report the outcome of a large series of patients treated for herniated lumbar discs that can be easily compared with series reported in the future.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/862/2751797 by Technische Universitaet Muenchen user on 09 July 2018
PATIENTS AND METHODS Patient population Data from the hospital and office charts of 654 consecutive patients treated during a 4.5-year period were reviewed retrospectively. Sixty-one percent of the population were men (average age, 42 yr), and 39% were women (average age, 43 yr; range, 15-83 yr). The criterion for inclusion was the presence of a herniated lumbar disc that had been treated surgically. Patients with primarily bony spinal stenosis or lateral recess stenosis with resultant nerve root entrapment were not included. Operative technique Measuring outcome from surgery has always presented a problem, especially when new methods are used. Many articles have discussed the technique of microdiscectomy, its advantages and disadvantages, and surgical outcome. It is difficult to analyze the surgical outcome of patients undergoing microdiscectomies or to compare the results with other therapeutic methods. Many authors use terms such as excellent, good, satisfactory, and poor without defining specific characteristics for each category, or each category is defined differently by different authors. Developing an objective method of assessing outcome that would not depend on the procedure used and that would yield comparable results whether rated by the operating surgeon, the patient, or a neutral observer is therefore a worthwhile objective. The Functional-Economic Outcome Rating Scale for back surgery has been proposed by Prolo et al. (25) as an objective method by which to evaluate patients undergoing microdiscectomy. We report the outcome of a large series of patients treated for herniated lumbar discs based on this scale. Eight neurosurgeons from the Barrow Neurological Institute performed the operations. Four surgeons performed 75% of the operations, and two of these four surgeons performed 50% of the total number of operations in this series. Three operative procedures were used. The first, microdiscectomy, was considered to be a small incision, with removal or opening of the yellow ligament; minimal bone removal; and the use of the operating microscope to remove the disc material. In each case, the disc space itself was also entered and all available disc material was removed. Typically, only the bulk of the intra-annular material within reach of rongeurs and curettes was removed; however, some surgeons made a greater attempt to remove disc material and end plates fully by using ring curettes. A search for midline and foraminal fragments was routinely performed, and a medial facetectomy and/or foraminotomy was done if bony encroachment was present. A microdiscectomy was performed on 54% of the patients. The second technique, laminectomy plus microdiscectomy, was defined as any operation that described a large opening in or complete removal of the lamina. In these instances, an operating
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Neurosurgery 1992-98 June 1992, Volume 30, Number 6 862 Outcome Analysis in 654 Surgically Treated Lumbar Disc Herniations Experimental and Clinical Study
Outcome analysis The records for each patient were analyzed with respect to the cause of injury, clinical findings, radiological examination, operative procedure and related complications, and the incidence of subsequent operations. The type of injury and the type of insurance coverage and whether litigation was involved were also recorded. The outcome was measured by the FunctionalEconomic Outcome Rating Scale of Prolo et al. (Table 1) (25), which consists of two subscales: the economic and the functional. Outcome scores were assigned by the reviewing physician on the basis of chart review. The subscale scores were summed to obtain an overall score. Total scores of 5 or less were considered poor outcomes; scores of 6 to 7 were considered to be moderate outcomes; and scores between 8 and 10 were considered to be good outcomes. The functional scale, as reported by Prolo et al. (25) gives an economic grade expressing the patients' capacity for gainful employment and a function grade expressing the effects of pain on daily activity. RESULTS Cause of injury The highest incidence of ruptured disc in this series was caused by lifting (31.4%), followed by falls (10.2%), sports or exercise injuries (10.0%), and automobile accidents (6.1%). Only 1.8% of the patients associated the occurrence of the disc rupture with coughing or sneezing. No definite cause could be determined in 40.5% of the cases.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/862/2751797 by Technische Universitaet Muenchen user on 09 July 2018
Clinical findings Almost all patients (99%) complained of leg pain, and 79% had radicular pain in a dermatomal distribution. Seventy-five percent of the patients had a sensory or reflex loss compatible with the location of the disc lesion operated upon. Motor weakness was present in 27% of the patients. Almost all patients had undergone conservative care. Ninety-three percent had symptoms for at least 3 months before surgery, and 88% had at least 2 weeks of bed rest, either as an inpatient or an outpatient. More than one third of the group had been given systemic or epidural corticosteroids before surgery. Diagnostic testing No patient underwent an operation in this series without an imaging study in addition to plain lumbosacral spine x-rays. A myelogram was performed on 77% of the patients, and 92% underwent computed tomography (CT). Seventy percent of the patients had a combination of myelography and CT or myelography and magnetic resonance imaging (MRI) (Signa 1.5-Tesla system; General Electric, Milwaukee, WI). In the final year of the study, MRI was the only diagnostic study performed in 50% of the patients who underwent operations. Complications of surgery Seventy-one patients (10.8%) had complications. There were 45 patients with wound infections above the fascia, and 2 patients with wound infections requiring a subsequent operation below the fascia. Three patients with discitis were diagnosed (one positive culture, two negative cultures), and a cerebrospinal fluid leak was identified in 6 patients. Three additional patients developed a postoperative pseudomeningocele and required further surgery for this complication. In 3 patients, the tissues outside the laminar space were injured. Two arterial injuries and one small bowel injury were identified. One of the arterial injuries resulted in death from complications of the abdominal surgery undertaken for repair. Six patients had pulmonary complications, but cardiac ischemia, arrhythmias, and deep vein thrombosis were also present. Reoperations Sixty-one patients had a second lumbar disc operation. Thirty underwent the operation at the same level as previous surgery, and 15 underwent surgery at a new level. In 16 patients, no data about the previous level of operation were available. Of the 15 who underwent surgery at a new level, the original level of surgery was reexplored at the same operation in nine. Nineteen patients had previously undergone surgery at the L4-L5 level, 18 patients at the L5-S1 level, and 1 patient at the L3-L4 level where the information was available. There were four recurrences at the L3-L4 level in patients where this level had not previously been operated on. Additional operations, decompression, laminectomy, or spinal stabilization were needed in 3.6% of this series of patients.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
microscope was used for disc removal beneath the nerve root. In almost all of these procedures, a medial facetectomy and/or foraminotomy was performed. This procedure was performed on 32% of the patients. The third technique, decompressive laminectomy plus microdiscectomy, was defined as a bilateral laminectomy with removal of the spinous process and medial portion of the facets. Fourteen percent of the patients underwent this procedure, in which the microscope was used in the majority of cases. These patients usually had large central disc extrusions with elements of either congenital or acquired spinal stenosis. Two surgeons routinely placed free fat grafts over the dura at the end of the procedure; the others did not cover the dura with any material. Gelfoam or Avitine was used for hemostasis, but there was no consistent policy on the full removal of these materials at the end of the operation. Gelfoam is an absorbable material made from a gelatin solution (Upjohn Co., Kalamazoo, MI). Gelfoam has no hemostatic action of its own, unlike Avitine (Avicon, Inc., Ft. Worth, TX), which is a microfibrillar collagen believed to act by reacting with platelets (2). The L5-S1 disc was operated on in 40% of the patients, the L4-L5 in 49% of the patients, the L3-L4 in 9%, and the L2-L3 in 2%.
Hospitalization and follow-up The mean length of stay in the hospital was 6 days (standard deviation, 4 days; range, 1-30 d; Table 2). The very young and the older patients had longer hospital stays, although the reason was not determined. The hospital stay of 8 patients was extended because of medical complications. The mean length of surgical follow-up was 9.7 months. The last patient in this study underwent surgery 1 year before the series was finally reviewed. Insurance coverage Of the 654 patients, 33% were patients involved in industrial accidents and 6% had legal claims pending during the surgical period. No legal or industrial claims were present during the surgical and follow-up periods of 61% of the patients. Outcome as measured by the Functional-Economic Outcome Rating Scale The mean economic outcome score of all patients, as determined by the physician reviewer, was 4.5. The mean functional outcome score was 4.08 (total, 8.32) (Table 3). Outcome scores based on type of insurance coverage are shown in Table 4. Combining the functional and economic scores, patients in the spontaneous and "other" categories had better total scores than did either the industrial accident patients, who had the lowest scores, or patients involved in lawsuits, who had intermediate scores. When broken into economic and functional categories, the industrial injury patients again had the lowest economic outcome scores. There was no statistical difference between the scores of patients in the spontaneous/other and legal groups. Interestingly, when the functional categories were examined, the industrial accident patients were the only group that slightly increased their functional score over their economic score. In fact, in that category, they did not differ significantly from patients with legal suits and were only slightly different than patients in the spontaneous or other categories. We also examined whether outcome scores differed by physicians because there was some variability in the type of operations performed. When the four physicians who operated most were evaluated in this fashion, there was no difference in the outcomes.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/862/2751797 by Technische Universitaet Muenchen user on 09 July 2018
DISCUSSION Strict criteria for surgical intervention should result in a satisfactory outcome for most patients treated for lumbar disc herniations. Indications for urgent surgery in this series were motor weakness or bowel or bladder dysfunction. Most patients had surgery for persistent pain in dermatomal distributions, sensory change, or reflex loss appropriate to the lesion on medical imaging (4,7,8,14,29,30,33). The location of the herniations was comparable with those from other series (7,8,14,29,30,33). As MRI improves, it is becoming an increasingly useful diagnostic tool. It has begun to replace myelography and CT as both the initial screening technique and definitive presurgical study at the Barrow Neurological Institute. This confidence appears justified: Forristall et al. (9) reported that surgical findings supported the MRI diagnosis with a 90% rate of accuracy compared with only 77% for CT. The sensitivity and specificity of MRI are also greater than CT. Complications of lumbar disc surgery appear to be few in all published series (1,4,7,8,11,17,19,20,26,27,29,32). Perforation of the iliac vessels is the most serious complication reported and probably arises from tooaggressive removal of interspace disc material. In our series, all neurosurgeons removed more than the fragmented disc. The literature supports the excision of as much disc material as possible to help lower the recurrence rate. Williams (31) and Balderston et al. (3) advocate fragment removal and minimal evacuation of disc material without curettage in order to avoid injuring healthy disc material. This disc material, however, may become a source of recurrent disc herniation. Williams reports a 9% recurrence rate. After more radical excision of disc material with curretage, Yasargil's and Caspar's recurrence rate, as stated in Maroon (24), was only 4%. Disc removal from the interspace may therefore be important, but the depth of the instrument into the interspace must be restrained in order to prevent penetration of the anterior annulus. Eleven percent of the patients required a second operation. Our recurrence rate at the same level was 3%, which compares favorably with those of other reports (3-15%) (6-8,10,29). In early recurrences (
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/862/2751797 by Technische Universitaet Muenchen user on 09 July 2018
PATIENTS AND METHODS Patient population Data from the hospital and office charts of 654 consecutive patients treated during a 4.5-year period were reviewed retrospectively. Sixty-one percent of the population were men (average age, 42 yr), and 39% were women (average age, 43 yr; range, 15-83 yr). The criterion for inclusion was the presence of a herniated lumbar disc that had been treated surgically. Patients with primarily bony spinal stenosis or lateral recess stenosis with resultant nerve root entrapment were not included. Operative technique Measuring outcome from surgery has always presented a problem, especially when new methods are used. Many articles have discussed the technique of microdiscectomy, its advantages and disadvantages, and surgical outcome. It is difficult to analyze the surgical outcome of patients undergoing microdiscectomies or to compare the results with other therapeutic methods. Many authors use terms such as excellent, good, satisfactory, and poor without defining specific characteristics for each category, or each category is defined differently by different authors. Developing an objective method of assessing outcome that would not depend on the procedure used and that would yield comparable results whether rated by the operating surgeon, the patient, or a neutral observer is therefore a worthwhile objective. The Functional-Economic Outcome Rating Scale for back surgery has been proposed by Prolo et al. (25) as an objective method by which to evaluate patients undergoing microdiscectomy. We report the outcome of a large series of patients treated for herniated lumbar discs based on this scale. Eight neurosurgeons from the Barrow Neurological Institute performed the operations. Four surgeons performed 75% of the operations, and two of these four surgeons performed 50% of the total number of operations in this series. Three operative procedures were used. The first, microdiscectomy, was considered to be a small incision, with removal or opening of the yellow ligament; minimal bone removal; and the use of the operating microscope to remove the disc material. In each case, the disc space itself was also entered and all available disc material was removed. Typically, only the bulk of the intra-annular material within reach of rongeurs and curettes was removed; however, some surgeons made a greater attempt to remove disc material and end plates fully by using ring curettes. A search for midline and foraminal fragments was routinely performed, and a medial facetectomy and/or foraminotomy was done if bony encroachment was present. A microdiscectomy was performed on 54% of the patients. The second technique, laminectomy plus microdiscectomy, was defined as any operation that described a large opening in or complete removal of the lamina. In these instances, an operating
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Neurosurgery 1992-98 June 1992, Volume 30, Number 6 862 Outcome Analysis in 654 Surgically Treated Lumbar Disc Herniations Experimental and Clinical Study
Outcome analysis The records for each patient were analyzed with respect to the cause of injury, clinical findings, radiological examination, operative procedure and related complications, and the incidence of subsequent operations. The type of injury and the type of insurance coverage and whether litigation was involved were also recorded. The outcome was measured by the FunctionalEconomic Outcome Rating Scale of Prolo et al. (Table 1) (25), which consists of two subscales: the economic and the functional. Outcome scores were assigned by the reviewing physician on the basis of chart review. The subscale scores were summed to obtain an overall score. Total scores of 5 or less were considered poor outcomes; scores of 6 to 7 were considered to be moderate outcomes; and scores between 8 and 10 were considered to be good outcomes. The functional scale, as reported by Prolo et al. (25) gives an economic grade expressing the patients' capacity for gainful employment and a function grade expressing the effects of pain on daily activity. RESULTS Cause of injury The highest incidence of ruptured disc in this series was caused by lifting (31.4%), followed by falls (10.2%), sports or exercise injuries (10.0%), and automobile accidents (6.1%). Only 1.8% of the patients associated the occurrence of the disc rupture with coughing or sneezing. No definite cause could be determined in 40.5% of the cases.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/862/2751797 by Technische Universitaet Muenchen user on 09 July 2018
Clinical findings Almost all patients (99%) complained of leg pain, and 79% had radicular pain in a dermatomal distribution. Seventy-five percent of the patients had a sensory or reflex loss compatible with the location of the disc lesion operated upon. Motor weakness was present in 27% of the patients. Almost all patients had undergone conservative care. Ninety-three percent had symptoms for at least 3 months before surgery, and 88% had at least 2 weeks of bed rest, either as an inpatient or an outpatient. More than one third of the group had been given systemic or epidural corticosteroids before surgery. Diagnostic testing No patient underwent an operation in this series without an imaging study in addition to plain lumbosacral spine x-rays. A myelogram was performed on 77% of the patients, and 92% underwent computed tomography (CT). Seventy percent of the patients had a combination of myelography and CT or myelography and magnetic resonance imaging (MRI) (Signa 1.5-Tesla system; General Electric, Milwaukee, WI). In the final year of the study, MRI was the only diagnostic study performed in 50% of the patients who underwent operations. Complications of surgery Seventy-one patients (10.8%) had complications. There were 45 patients with wound infections above the fascia, and 2 patients with wound infections requiring a subsequent operation below the fascia. Three patients with discitis were diagnosed (one positive culture, two negative cultures), and a cerebrospinal fluid leak was identified in 6 patients. Three additional patients developed a postoperative pseudomeningocele and required further surgery for this complication. In 3 patients, the tissues outside the laminar space were injured. Two arterial injuries and one small bowel injury were identified. One of the arterial injuries resulted in death from complications of the abdominal surgery undertaken for repair. Six patients had pulmonary complications, but cardiac ischemia, arrhythmias, and deep vein thrombosis were also present. Reoperations Sixty-one patients had a second lumbar disc operation. Thirty underwent the operation at the same level as previous surgery, and 15 underwent surgery at a new level. In 16 patients, no data about the previous level of operation were available. Of the 15 who underwent surgery at a new level, the original level of surgery was reexplored at the same operation in nine. Nineteen patients had previously undergone surgery at the L4-L5 level, 18 patients at the L5-S1 level, and 1 patient at the L3-L4 level where the information was available. There were four recurrences at the L3-L4 level in patients where this level had not previously been operated on. Additional operations, decompression, laminectomy, or spinal stabilization were needed in 3.6% of this series of patients.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
microscope was used for disc removal beneath the nerve root. In almost all of these procedures, a medial facetectomy and/or foraminotomy was performed. This procedure was performed on 32% of the patients. The third technique, decompressive laminectomy plus microdiscectomy, was defined as a bilateral laminectomy with removal of the spinous process and medial portion of the facets. Fourteen percent of the patients underwent this procedure, in which the microscope was used in the majority of cases. These patients usually had large central disc extrusions with elements of either congenital or acquired spinal stenosis. Two surgeons routinely placed free fat grafts over the dura at the end of the procedure; the others did not cover the dura with any material. Gelfoam or Avitine was used for hemostasis, but there was no consistent policy on the full removal of these materials at the end of the operation. Gelfoam is an absorbable material made from a gelatin solution (Upjohn Co., Kalamazoo, MI). Gelfoam has no hemostatic action of its own, unlike Avitine (Avicon, Inc., Ft. Worth, TX), which is a microfibrillar collagen believed to act by reacting with platelets (2). The L5-S1 disc was operated on in 40% of the patients, the L4-L5 in 49% of the patients, the L3-L4 in 9%, and the L2-L3 in 2%.
Hospitalization and follow-up The mean length of stay in the hospital was 6 days (standard deviation, 4 days; range, 1-30 d; Table 2). The very young and the older patients had longer hospital stays, although the reason was not determined. The hospital stay of 8 patients was extended because of medical complications. The mean length of surgical follow-up was 9.7 months. The last patient in this study underwent surgery 1 year before the series was finally reviewed. Insurance coverage Of the 654 patients, 33% were patients involved in industrial accidents and 6% had legal claims pending during the surgical period. No legal or industrial claims were present during the surgical and follow-up periods of 61% of the patients. Outcome as measured by the Functional-Economic Outcome Rating Scale The mean economic outcome score of all patients, as determined by the physician reviewer, was 4.5. The mean functional outcome score was 4.08 (total, 8.32) (Table 3). Outcome scores based on type of insurance coverage are shown in Table 4. Combining the functional and economic scores, patients in the spontaneous and "other" categories had better total scores than did either the industrial accident patients, who had the lowest scores, or patients involved in lawsuits, who had intermediate scores. When broken into economic and functional categories, the industrial injury patients again had the lowest economic outcome scores. There was no statistical difference between the scores of patients in the spontaneous/other and legal groups. Interestingly, when the functional categories were examined, the industrial accident patients were the only group that slightly increased their functional score over their economic score. In fact, in that category, they did not differ significantly from patients with legal suits and were only slightly different than patients in the spontaneous or other categories. We also examined whether outcome scores differed by physicians because there was some variability in the type of operations performed. When the four physicians who operated most were evaluated in this fashion, there was no difference in the outcomes.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/862/2751797 by Technische Universitaet Muenchen user on 09 July 2018
DISCUSSION Strict criteria for surgical intervention should result in a satisfactory outcome for most patients treated for lumbar disc herniations. Indications for urgent surgery in this series were motor weakness or bowel or bladder dysfunction. Most patients had surgery for persistent pain in dermatomal distributions, sensory change, or reflex loss appropriate to the lesion on medical imaging (4,7,8,14,29,30,33). The location of the herniations was comparable with those from other series (7,8,14,29,30,33). As MRI improves, it is becoming an increasingly useful diagnostic tool. It has begun to replace myelography and CT as both the initial screening technique and definitive presurgical study at the Barrow Neurological Institute. This confidence appears justified: Forristall et al. (9) reported that surgical findings supported the MRI diagnosis with a 90% rate of accuracy compared with only 77% for CT. The sensitivity and specificity of MRI are also greater than CT. Complications of lumbar disc surgery appear to be few in all published series (1,4,7,8,11,17,19,20,26,27,29,32). Perforation of the iliac vessels is the most serious complication reported and probably arises from tooaggressive removal of interspace disc material. In our series, all neurosurgeons removed more than the fragmented disc. The literature supports the excision of as much disc material as possible to help lower the recurrence rate. Williams (31) and Balderston et al. (3) advocate fragment removal and minimal evacuation of disc material without curettage in order to avoid injuring healthy disc material. This disc material, however, may become a source of recurrent disc herniation. Williams reports a 9% recurrence rate. After more radical excision of disc material with curretage, Yasargil's and Caspar's recurrence rate, as stated in Maroon (24), was only 4%. Disc removal from the interspace may therefore be important, but the depth of the instrument into the interspace must be restrained in order to prevent penetration of the anterior annulus. Eleven percent of the patients required a second operation. Our recurrence rate at the same level was 3%, which compares favorably with those of other reports (3-15%) (6-8,10,29). In early recurrences (
