|
Alar Transverse Fusion For Spondylolisthesis in Children A Review of 50 Patients
K. El-Abed, JG Kennedy, F. Condon, D. Mulcahy, F. Dowling, E. Fogarty, D. Moore. Orthopaedics Department, Our Lady’s Hospital for Sick Children, Crumlin, Dublin. 12.
Address Correspondence to: K. El-Abed, & JG Kennedy, Department of Orthopaedics, Our Lady’s Hospital for Sick Children, Crumlin, DUBLIN. 12.
Abstract: We reviewed 50 cases of spondylolisthesis treated by surgical stabilisation over a 14 year period between 1983-1997. There were 25 males and 25 females with a mean age of 14 years (range 10 – 19 years). The etiology was dysplastic in 46 patients (92%), isthmic in 3 patients and traumatic in one patient. Twenty patients had Grade I slip, 17 patients had Grade II slip, 1 patient had Grade III slip and 10 patients had Grade IV slip using the Meyerding grading system. There were 2 patients with spondyloptosis. Patients were divided into 2 groups; those with a <50% slip as group A and those with a >50% slip as group B. The indication for surgery was continuous pain in 88%, progressive slip in 10%, worsening neurology in 4% and gait/postural abnormalities in 2%. All patients had alar transverse fusion, 52% had insitu fusion without instrumentation, 28% had insitu fusion with instrumentation and 20% had reduction of slip followed by instrumentation. Results were assessed by relief of pain and return to normal activity; 88% of patients had an excellent/good result and 12% had a fair result with persistent moderate pain. There was a poorer overall outcome in group B patients than group A. This was associated with the severity of the initial symptoms and was not correlated with surgical reduction. The overall complication rate was higher in group B, however this may be related to early difficulties with transpedicular screws. This study supports the continued use of alar transverse fusion for all grades of spondylolisthesis as a safe and effective method of eleviating symptomatology and preventing progression of slip.
Introduction: Spondylolysis with or without spondylolisthesis is a relatively common condition in the paediatric population with an overall incidence between 4-6%. 1 Additionally in children complaining of low back pain it is found to be the underlying cause in up to 33% of presenting cases 2. Many non-operative treatment paradigms have been reported in the literature with satisfactory outcomes3. These measures include restriction of vigorous activities, physical therapy in addition to analgesia. Occasionally patients may need to be immobilised in a cast or thoracolumbar orthosis. Most of these patients have satisfactory relief of pain at long term follow up 4, 5. Surgical intervention is reserved therefore for patients with uncontrolled pain, a progressive slip, development of a neurological deficit and failure of conservative treatment 6. The surgical treatment of spondylolysis and mild spondylolisthesis is well established. The treatment of more severe grades of slippage is often vague and confusing with Ostermann et al 7 recommending fusion insitu to minimise the risk of neurological damage whereas Paussa et al 8 advocates reduction of severe slips to correct sagittal plane deformity. The aim of this study was therefore to assess in-situ and corrective alar transverse fusion in the treatment of all grades of spondylolisthesis in the paediatric population.
Methods: Seventy-five patients who had undergone surgical treatment for spondylolisthesis were identified from the Spinal Unit Database between 1983 and 1997, 50 patients of these were available for a retrospective chart and radiograph review. Minimum time for follow up was 12 months. Radiographs were graded according to the convention of Meyerding 9, the degrees of slip were measured from the pre-operative radiographs 10. A visual analogue scale was used to assess outcome following surgery. This was based on a total possible score of 100. Any patient who was not reviewed in the clinic was contacted by telephone questionnaire, in addition patients were asked to determine whether their daily living activities had changed following the procedure. Logistical regression analysis was used to determine predictors of outcome using the SPSS statistical package 6.1 (Chicago ILL).
Results: There were 25 males and 25 females, the mean age at surgery was 14 years (range 10-19 years). The primary symptom was unremitting low back pain with a mean duration of 13.5 months at the time of surgery in 44 (88%) patients. (Table I) Forty-six patients (92%) in this study had a congenital defect in L5-S1 facet joints. Although dysplastic in type 11, over half of these patients had an associated pars interarticularis defect. Three patients had a pure isthmic type spondylolisthesis. One patient had an acute post-traumatic slip. Using Meyerding’s classification 37 patients (74%) in our study had a slip less than 50%, a further 11 patients had Grade III or IV slip, and 2 patients had spondyloptosis. All patients had alar transverse fusion; 26 patients had insitu fusion alone, 14 patients had insitu fusion with instrumentation and a further 10 patients had reduction of slip and instrumentation. Those that had a slippage of under 50% were considered as group A and those who had a slippage of more than 50% as group B. Twenty-six (70%) of patients in group A had insitu fusion without instrumentation, with 9 patients requiring instrumented fixation. Two patients had reduction prior to instrumentation. Of those patients in group B, 5 patients were fixed insitu all with instrumentation and 8 (62%) patients underwent reduction prior to fixation. Thirty-five of 37 patients in group A had either complete relief of their pain, or non-restricting mild intermittent pain following alar transverse fusion. The VAS rating for this cohort was 89% (range 75-100%). Two patient in this group had pain limiting return to normal activity, VAS 60% (range 50-74%) one had concurrent degenerative disc disease, whilst the other is involved in ongoing litigation following a motor vehicle accident. In group B, however, 4 (31%) of patients had persistent moderate pain preventing return to full activities with a mean VAS of 55% (range 25 – 54%). This correlated well with an inability to return fully to full sporting activities, but allowed unrestricted activities of daily living. The remaining 9 patients in group B all had a satisfactory outcome. Overall, 44 (88%) patients had an excellent or good result and 6 (12%) patient had a fair result. Complications for this cohort included 2CSF leaks post-operatively, 2 patients developed acute pancreatitis post-operatively, and one patient had superficial wound infection. Three patients subsequently required metal removal; one of whom had a symptomatic broken screw and two further patients had local discomfort from the metal work which resolved once the metal was removed (Table II). Logistal regression analysis demonstrated that age, sex and duration of symptoms were not significant as predictors of a poor outcome. When the degree of slip was evaluated as a univariate predictor of a poorer outcome, we found that those patients who had a 50% degree of slip or greated had a statistically significant poorer outcome. No correlation was found between reduction of slip and an unsatisfactory overall outcome. We did not find, however, an association between our overall complication rate and reduction of slip which was statistically significant (p=0.046).
Discussion: In the current literature many conditions may lead to the displacement of the lumbosacral junction. The comprehensive classification of spondylolisthesis proposed by Wiltse et al in 1976 had been generally accepted. We found that dysplastic and isthmic were the two types most common in our study group; 98% of patients. This is in agreement with other authors in the paediatric population 12, 13. Although a history of minor trauma was common (50% in boys and 25% in girls), an episode of trauma often initiated the onset of symptoms, there was seldom a history of severe injury. Just one patient presented with post-traumatic acute slip. The lack of degenerative changes and exclusion of causes producing spondylolisthesis secondary to structural weakness of the bone or supporting tissues suggest that spondylolisthesis in immature patients should be viewed as a distinct clinical entity. The initial presenting symptom was low back pain in 88% of cases limiting activities of daily living and not responding to conservative management. Unlike adults, children seldom have objective signs of nerve-root compression, such as motor weakness, changed reflexes of sensory deficit 14 with only 4% of our patients demonstrating a progressive neurological deficit. We found also that the physical examination may be normal in a child who has mild (Grade I or II) spondylolisthesis. Tight hamstrings were found in 30% of patients, severe enough only in one patient to develop gait/postural abnormalities. The indications of surgery in our group of patients were comparable to other studies and recommendations in the literature 6, 10 with the majority (88%) of patients complaining of unremitting low back pain despite at least 6 months of conservative management. The overall outcome in this study group of those with mild spondylolisthesis; group A was very satisfactory with over 80% return to full activity. This compares favourably with other authors on similar degrees of slip 15, 16, 14. In those with a more severe degree of slippage; group B, over 60% returned to normal daily activities following surgery. This compares to Harris and Weinstein study 6; who had a 24 year follow up on 21 spondylolisthesis cases with >50% slip , of which 57% were asymptomatic following insitu fusion. We found that there was a poorer overall outcome in patients with a >50% (group B) although there were more reductions followed by fusion in this as compared to group A, there was no correlation between reduction of slip and a poorer outcome either as a univariate or multivariate predictor. Of the 4 patients in group B who had a poor outcome, all complained of low back pain and hamstring tightness preventing a normal range of hip movement following surgery, we believe this low back pain is secondary to the altered spinal anatomy predisposing to paraspinal muscle spasm, and the tight extensor group is again secondary to the altered anatomy. Three of these 4 were placed in body jackets or orthoses following fusion and did not begin a physiotherapy programme until 6 weeks after surgery. This study did not support those authors who have found reduction in severe slips to be associated with a poorer outcome 8, 17. We believe this is due to reducing severe grades to a lesser grade rather than attempting anatomical alignment. We did find however, that reduction was associated with greater overall complication rate; this may be related to the use of transpedicular screws and possible dural tear with subsequent CSF leakage. We have previously commented on the aetiology of pancreatitis following alar transverse fusion, although we cannot speculate on how reduction of spondylolisthesis may predispose to this. Two patients of our study group developed acute pancreatitis postoperatively, both had persistent abdominal pain after alar transverse fusion with markedly elevated plasma amylase. Both patients were treated by nasogastric suction, gastrointestinal rest and prophylactic antibiotics with the resolving of symptoms within a one week period. As previously reported in literature 18, 19, a clinical suspicion of pancreatitis should be aroused when persistent abdominal symptoms are encountered in patients after any spinal surgery. This study has demonstrated that alar transverse fusion is an effective and safe procedure for all grades of spondylolisthesis. Although there was a correlation between a >50% slip and a poorer overall outcome, we could find no association between reduction of slip in this group and an unsatisfactory outcome. We recommend that alar transverse fusion continue to be used in all grades of spondylolisthesis to prevent further progression and alleviate symptomatology.
References:
Table 1 Indications for Surgery. * Continuing Pain - 88% * Progressive Slippage - 10% * Gait/Postural Abnormalities - 2% * Neurological Deficits. - 4%
Table 2 Post-operative Complications. * CSF Leak - 2 Cases * Acute Pancreatitis - 2 Cases * Wound Infection - 1 Case * Discomfort Metal Work - 2 Cases * Symptomatic Broken Screw - 1 Case
|