Incidencia y factores relacionados con la degeneración del segmento adyacente en pacientes con patología degenerativa e inestabilidad intervenidos mediante un procedimiento de fusión lumbar

Spinal fusion procedures have increased significantly in the last few years. Both reconstructive surgery for tumor treatment, stabilization of unstable traumatic lesions and deformity correction surgery remain absolute indications for fusion procedures. In degenerative pathology patients, decompression of neural elements remains the leading indication/criteria, however these indications of fusion become controversial if there is no instability factor. Health-related quality-of-life questionnaires (HRQOL) have been proposed as valid indicators of the patient’s own perception of his health state. Visual analogic scale (VAS), the Oswestry Disability Index (ODI) and the Short-Form Health Survey (SF-36) have been proposed as reliable indicators of both lumbar and radicular pain, disability and functional status. Computed tomography (CT) and Magnetic Resonance (MR) imaging allow for an accurate assessment of bone changes and the degree of compression of neural elements. However, full-lengths X-rays standing images retain their value as a basic diagnostic study in assessing both coronal and sagittal balance as well as spinal and pelvic parameters. Medical complications due to secondary disorders and associated with blood loss and length of procedure are not infrequent in elderly patients undergoing surgery. Surgical complications may have a temporary or definitive impact on the outcome. After a spinal fusion procedure has been carried out, the adjacent segment degeneration (ADS) can be a frequent complication. This has been defined as the radiological or clinical changes - radiological or clinical degeneration respectively - where symptoms such as lumbar and radicular pain or disability can reappear following a symptom-free period after the initial procedure, which may require a second surgery. While the question of whether these changes may be caused by the progression of degenerative changes or the fusion procedure remains controversial, they may in fact originate from a range of factors and could ultimately be due to the genetic predisposition to degeneration, result of the natural history of progressive degeneration, the alterations in adjacent segment anatomy at the time of surgery, implantation of spinal instrumentation that may directly alter the adjacent segments, or the condition being solely a biomechanical phenomenon. The clinical signs of the degeneration of the adjacent segment and the incidence of the changes have been described at various degrees of variability due to the retrospective nature of the studies, the chosen criteria and the follow-up periods. OBJECTIVE To determine the incidence of the degeneration of the adjacent segment according to radiological and clinical criteria and the factors associated with its occurrence. MATERIALS AND METHODS An observational longitudinal retrospective study of patients having undergone surgery in the period 2006-2010 with a minimum follow-up of two years and a diagnosis of degenerative pathology and instability was performed. The surgery followed a posterolateral fusion (PL) or circumferential technique via posterior lumbar inter-body fusion (PLIF) with pedicle screws and cages PEEK (Polyetheretherketone) and autologous bone. Clinical variables (back and leg pain and disability) have been analyzed in preoperative and at the end of the follow-up using VAS and ODI and radiological variables total lumbar (TLL) and lumbosacral lordosis (LSL), posterolateral radiological fusión assessment according to Lenke and Bridwell criteria, circumferential fusion with Hackenberg criteria, rotational deformity according to Nash-Moe scale, anterolisthesis or retrolisthesis in all free segments measured in millimeters, radiologic osteoarthritis of the lumbar spine (Weiner radiologic scoring), disc height loss >1mm in all free discs and Pfirrmann magnetic resonance classification of lumbar intervertebral disc degeneration. Radiological criteria for adjacent disc degeneration in all free discs were: rotational deformity ≥ 1 degree, shifting ≥ 3mm (anterolisthesis o retrolisthesis), disc height loss ≥ 1mm (anterior, central and posterior), moderate or severe radiologic disc degeneration according to Weiner scoring and severe disc MR degeneration grade IV and V (Pfirrmann classification). We considered as clinical degeneration criteria, lumbar or radicular pain (>3 points in VAS-back and VAS-legs scores) and moderate or severe disability (according to Oswestry Dissability Index ODI). After being anonymized, data was included in the SPSS (19 v). A descriptive analysis was performed (frequency distribution, central and dispersion tendency measures (CI: 95%) and analysis of variance (ANOVA) to evaluate lumbar and radicular pain and disability changes. Intra and interobserver measures reliability was performed by intraclass correlation coefficient (ICC) for quantitative measures and Kappa index for qualitative measures. Landis and Koch scale established agreement. To determine the patient´s evolution who remain without degeneration we performed a mono-variable analysis by actuarial and Kaplan-Meier methods, the comparison of survival curves by Mantel-Haenszel and multi-variable Cox proportional hazard analysis (Cox regression). Independent variables were transformed into dichotomous variables to increase the statistical power: age (≥ 61 years old and ≤ 60 years old), sex (male/female), surgical method (PL vs. PLIF), pedicle screw device system (top-loading vs. side-connecting lateral), fusion levels (<3 vs. ≥ 3 levels) and diagnosis (canal stenosis vs. others diagnosis). RESULTS Of all patients undergoing surgery in the period 2006-2010, a total of 293 met the criteria for inclusion, with 263 being valid for the study. The observation period ranged between 24-82 months with an average of 45,9 months (DE: 17,8) and a mean of 43. Anterior or posterior shifting (anterolisthesis or retrolisthesis) was observed in 33 patients (12,5%) (CI 95%: 8,4-16,7) at the end of the follow-up period. The rotation of the upper spinal segments not included in the fusion increased at the post-surgery stage in 36 patients (13,6%) (CI 95%: 9,3-18,0). The degree of radiological disc degeneration moderate or severe was observed in 48 discs in 35 patients (13,3%) (CI 95%: 9,0-17,6) according to the Weiner scoring at the end of the follow-up period. The loss of discal height of the non-fused lumbar discs measured in millimeters of the anterior, medium and posterior height of all free discs was observed in 159 discs in 112 patients (42,6%) (CI 95%: 36,4-48,8). The changes assessed in 73 patients by means of RM (severe degeneration degrees IV and V) were observed in 45 discs in 32 patients (43,8%) (CI 95%: 31,8-55,9). Based on the degeneration criteria, 151 patients showed radiological changes (57,4%) (CI 95%; 51,2-63,6) whereas in 53 patients clinical changes were found (20,2%) (CI 95%: 15,1-25,2). In survival analysis performed using the actuarial method, the number of patients with changes progressively increased at different intervals during follow-up. In the third year 12 patients experienced changes (4.5%), in the fourth year 12 patients (7.1%), in the fifth year 13 patients (11.2%), in the sixth year 9 patients (13.8%) and in the seventh year 7 patients (26.9%). The median survival was 79 months (CI 95%: 76.6-81.4). Older patients (>61 years) did not show a greater probability of degeneration compared to younger patients (<60 years), although the difference was nearly statistically significant (p=0,058). The risk of free disc degeneration did not reveal any statistically significant differences according to gender (p= 0,096). No statistically significant differences were observed as to the risk of degeneration on the post-surgery analysis of lumbosacral lordosis (LSL) (p=0,9). According to the type of fusion, the likelihood of free disc degeneration did not display significant differences in comparing posterolateral to circumferential fusion techniques (p=0,220). Total post-surgery lumbar lordosis (TLL) showed a statistically significant difference in the risk of degeneration (p=0,018), which was higher in patients with hypo-lordosis total post-surgery lumbar lordosis. Fusion levels did reveal statistically significant differences as to free disc survival with a higher probability of degeneration with the fusion of 3 or more segments (p= 0,009). According to the type of pedicular instrumentation, cases where pedicle top-loading screws were used showed a higher probability of free disc degeneration than those where side-connecting screws were used (p < 0,001). As for diagnosis, patients where lumbar canal stenosis was used reveal a significantly higher probability of free disc degeneration than any other diagnosis (p=0,019). The Cox proportional hazard analysis was applied to assess the effect of the different study variables on the probability of degeneration of the adjacent segment. The risk of degeneration of the adjacent segment doubles in patients whose surgery involved three or more levels of fusion (HR=2,076; IC 95%: 1,047-4,116; p=0,037) and triples in patients where top-loading screws were used in their surgery (HR=2,983; IC 95%: 1,548-5,748; p=0,001). DISCUSSION A longitudinal retrospective study was conducted with a patient cohort where a lumbar fusion procedure was used. The causes, magnitude, associated factors and the degeneration of the adjacent segment effect have been closely examined. Due to the varied follow-up periods, a survival analysis has been used to determine the length of time prior to the occurrence of the degeneration of the adjacent segment and its interdependence with other variables. As in previous studies, the incidence of radiological changes has been higher than that of symptomatic changes. The radiological signs in free segments of the degeneration could explain its multifactorial origin. The discal degeneration and loss of discal height are the earliest and more frequent changes which may be due to the age-related degenerative process or could be made more severe by the fusion-induced biomechanical changes, such as an increase in discal pressure, hypermobility and facet joints loads, with the radiological discal degeneration being the most severe condition. Free segment rotation has been linked with changes in tropism facet and the rigidity of the fused segment. Posterior shifting (retrolisthesis) has been associated with the compensating mechanism of lordosis lost in the fused segments. Age and gender did not show any significant differences as to the risk of degeneration although it was higher in patients older than 60 years and in women. Neither were there any significant differences in the occurrence of changes according to fusion type (posterolateral vs. circumferential) and in patients with normal lumbosacral lordosis vs. hyper-lordosis. However, patients diagnosed with canal stenosis, hypo-lordosis lumbar fusion, number of fused levels and type of instrumentation consisting in top-loading pedicle screws did display significant differences in the risk of degeneration. Patients undergoing fusion for a diagnosis of canal stenosis showed a greater risk of degeneration associated with laminectomy and supra and interspinous ligament was removed, whereas hypo-lordosis total lumbar lordosis patients displayed a higher risk of degeneration related to compensating hypermobility of free segments and facet degeneration. According to Cox proportional hazard analysis (Cox regression), in patients with three or more levels of fusion the risk of degeneration increased two-fold (HR= 2,076) and three-fold (HR= 2,983) in patients treated with open screws. The multilevel fusion causes more rigidity and fewer compensating free segments with an increase in the risk of instability. Although the biomechanical changes in free segments are similar, the facet lesion related to the medial point of the pedicle could justify this finding. CONCLUSIONS 1. Radiological changes associated with the adjacent segment degeneration have been studied and examined in non-fused free lumbar discs with a variable incidence. Their frequency is as follows: discal degeneration evaluated by RM (43,8%), discal height loss (42,6%), radiological discal degeneration (13,3%), rotation (13,6%) and shifting (12,5%). 2. In patients treated with a lumbar fusion procedure by posterolateral and circumferential techniques, the incidence of radiological changes related to the degeneration of the adjacent segment represents 57,4% and the symptomatic changes 20,2%. 3. Risk factors associated with the degeneration occurrence were: patients diagnosed with canal stenosis, lumbar hypo-lordosis fusion, number of levels fused and the use of top-load pedicle screws.