IMMEDIATE RECIPROCAL CHANGES AT ADJACENT LEVEL FOLLOWING SINGLE-LEVEL ALIF

Objective: To assess the segmental sagittal parameters and the adjacent mobile segment after interbody fusion by anterior approach in single-level L5-S1. Methods: Retrospective study. Inclusion: Interbody fusion by anterior approach (10°/15° angle), due to DDD, low-grade spondylolisthesis and/or stenosis without prior lumbar fusion. Thirty-five cases were included (25 women; mean age 47±15 years). Analysis of lumbar radiographs were performed preoperatively and 3 months after surgery. The following parameters were evaluated: lumbar lordosis; segmental lordosis L4-L5 and L5-S1; sacral and L5 slope. Two independent evaluators made the measurements. Student’s t test was used. Results: There was correction of lordosis at the L5-S1 level, from an average of 19° prior to surgery to 28° three months after surgery (p<0.001). It was observed reduction of the L4-L5 angle from 17° to 14° (p <0.001). Moreover, there was a small, but statistically significant, reduction of L5 inclination from 17 ° to 13 ° (p=0.007) and increase in the sacral slope from 37° to 40° (p=0.002). The change of total lumbar lordosis was not significant: 52° versus 53° (p=0.461). Before surgical correction, lordosis of the L5-S1 level contributed, on average, 37% of lumbar lordosis, and after, 54% (p<0.001). The L4-L5 level accounted for 34% and decreased to 24% (p<0.001). Conclusion: In the group studied, it was observed a decrease in lordosis of the adjacent mobile level following the treatment of the degenerate level by anterior interbody fusion and concomitant increase in lordosis.


INTRODUCTION
Diseases of the spine with degenerative intervertebral disc components tend to reduce the curvature in lordosis.Compensatory mechanisms are created, to maintain regional and global sagittal alignment of the spine.However, these alterations usually involve pain, loss of function, and acceleration of the degenerative cascade. 1 The compensation mechanisms involve modification of the sagittal curves of the spine and retroversion of the pelvis, in addition to flexion of the knee.Among the sagittal modifications of the spine is an increase in lordosis in the mobile levels adjacent to a vertebral segment with a loss of lordosis, which may be iatrogenic, or the result of the degenerative process, and may be associated with disease at the adjacent level. 2,3iven that the spontaneous correction of this adjacent compensation mechanism is not well described, the objective of this study is to assess the short-term changes to the regional sagittal parameters and the adjacent mobile segment, following interbody fusion and restoration of segmental lordosis for isolated discopathy of L5-S1.

METHODS
A retrospective observational study approved by the Institutional Review Board (CAAE 48112515.2.0000.5511).The inclusion criteria were: indication of surgical treatment at a single level in L5-S1 as a result of degenerative disc disease (DDD defined in this work as Pfirrmann 4 grade IV or V and the presence of Modic changes 5 ), low grade spondylolisthesis and/or stenosis in the presence of instability; indication of lumbar interbody arthrodesis by the anterior approach only (without supplemental pedicle screws); anterior lumbar interbody fusion (ALIF) using an autoblocking titanium cage as an interbody spacer, with a 10-15°angle of lordosis (FUSIMAX Cage on Cage/Sistema SAGA, MDT Implantes/LAS Brasil, São Paulo); the existence of preoperative x-rays and x-rays taken after 3 months of postoperative follow-up.The exclusion criteria were: no presentation of an extruded herniated disc at the level to be treated; no presentation of arthrodesis/arthroplasty prior to the ALIF in any lumbar level.This retrospective analysis assessed 90 cases that underwent interbody fusion at level L5-S1 at the same center and, in accordance with the inclusion/exclusion criteria, 35 cases were selected for inclusion.In this group of 25 women and 10 men, the average was 47 years of age.24 cases had been diagnosed with DDD with severe degenerative alterations and 11 with low grade spondylolisthesis.
The radiological parameters were measured in neutral lateral orthostatic radiographs (Figure 1) taken preoperatively and after 3 months of postoperative follow-up.The following parameters were evaluated: lumbar lordosis (from the plateau above L1 to the plateau above S1); lordosis of the L5-S1 and L4-L5 disc space (the angle between the cranial and caudal vertebral plateaus in relation to the intervertebral disc of reference); segmental lordosis of L5-S1 and L4-L5 (the angle between the vertebral plateaus above the vertebrae adjacent to the vertebral disc of reference); L5 inclination and sacral inclination (angle between the upper plateau of the vertebra and a horizontal line parallel to the ground.Two independent evaluators took the measurements.The averages of the values were used, but if they differed by more than 5°, the measurements were retaken.
The PSAW Statistics 18 program was used for statistical analysis.To test the differences between the two groups the Student's t-test was used and the Pearson coefficient was used to correlate the two groups of measurements.Results with p-values less than 0.05 were considered to be statistically significant.

RESULTS
The results of the radiological measurements are summarized in Table 1.
The average lordosis values of the pathological level, L5-S1, and of the adjacent cranial level, L4-L5, were equal (19° ± 8° versus 17°±6°, respectively; p=0.0880) in the preoperative analysis.Even though the total change in lumbar lordosis after the fusion of L5-S1 was not significant (p=0.461), the results show that there was an increase in segmental lordosis of L5-S1 of 9°, starting at 19° and reaching 28° (p<0.001).We observed a reduction of the average angle of L4-L5, from 17° before fusion to 14° after fusion (p<0.001).Figure 2 shows a case example.
Other values that underwent statistically significant changes were sacral inclination, 37° versus 40° (p=002), and inclination of L5 from 17° to 13° (p=007).The distributions of the sagittal correction in the caudal (change in sacral inclination) and cranial (change in the inclination of vertebra L5) directions were not statistically different (p=0.470).
The correlation between the difference (3 months x preoperative) in the angulation of disc L5-S1 and the difference (3 months x preoperative) in the angulation of disc L4-L5 is shown in Figure 3.The correlation was negative and statistically significant (p<0.001), with a coefficient of determination (R 2 ) of -0.357 and a Pearson correlation coefficient (ρ) equal to -0.597.
To assess the distribution of lumbar lordosis among the distal levels before and after fusion, we calculated the percentage of contribution of the segmental lordosis of L4-L5 and L5-S1 to global lumbar lordosis, shown in Table 2.It was shown that in the preope-  A B

DISCUSSION
Reciprocal change adjacent to a level with degeneration is a recent theme, and we have not yet identified the description of the short term behavior of the adjacent mobile level following restoration of the space and of the disc angulation via interbody arthrodesis.The main objective of this study was to confirm whether changes occur in the sagittal positioning of the mobile level above an anterior approach interbody fusion in the short term.
The key findings of this work show that following an increase in the angulation of L5-S1, several regional changes occur: an increase in the lordosis of L4-L5, a slight reduction in the vertebral inclination of L5 (which is usually inclined downwards), and an increase in sacral inclination.
The normal angle of global lumbar lordosis is generally between 20 and 65°, with an increase in fractional lordosis in the caudal segments. 6,7The average global lumbar lordosis obtained by this study was around 52° and the contribution of the L4 to S1 lordosis to total lumbar lordosis was 70%.According to earlier reports, 8,9 the last two mobile levels (L4-L5 and L5-S1) contribute around 60% of the global lumbar lordosis.The collapse of the segment underneath, due to degenerative disease, therefore results in significant decompression due to the large contribution of these discs to sagittal alignment.The results for total change in lumbar lordosis presented here did not undergo change post-arthrodesis.Through analysis of the distribution of lumbar lordosis among the levels, it was observed that prior to surgery -the period in which level L5-S1 was diagnosed as the principal cause of the clinical symptoms -the contribution of the segmental lordosis of L5S1 was similar to that of L4-L5, and the contribution from disc lordosis was even less.In the analysis of the 3-month period, these contributions increased to practically double those of L4-L5, which decreased in relation to the preoperative period.
The value of lumbosacral lordosis is affected by the type of fusion procedure chosen by the surgeon, and this must be taken into account when planning the surgery.Dimar et al. 10 compared four arthrodesis techniques commonly used in the restoration of lumbar lordosis following single-level instrumented fusion: instrumented posterolateral fusion, transforaminal approach interbody fusion, anteroposterior fusion with posterior instrumentation, and anterior approach arthrodesis with lordotic cage.They discovered that the increase and maintenance of lordosis, and of the anterior and posterior height of the disc space, are better achieved with anterior arthrodesis using lordotic spacers.As with the technique, the form and the angulation of the interbody implants can also alter the sagittal alignment of the lumbar spine, as shown by other authors. 11he degeneration of the segment adjacent to a fusion is a multifactorial process.Besides the contribution of pre-existing changes and of the natural process of disc degeneration, sagittal alignment appears to make a significant contribution.In a case-control study, Djurasovic et al. 2 studied possible predisposing factors for adjacent degeneration.They concluded that spinal fusion in abnormal sagittal alignment with loss of lumbar lordosis predisposes patients to the development of degeneration of the adjacent segment, 2 a fact that can contribute significantly to an increase in the intradiscal pressure of the adjacent mobile segments, 12 which, after fusion, have to work at the maximum limit of their functional movement more frequently. 13In addition to sagittal positioning, another factor that emerged in a systematic review of the literature 3 as a potential protector of the adjacent level is fusion without the use of supplemental percutaneous screws, since these can damage the facet joints of the adjacent segment.
This study has some limitations to be considered.The group studied is relatively small, given that the very specific inclusion and exclusion criteria restricted the selection of a group with more patients.Moreover, not all the radiographs included the head of the femur, which made consideration of the pelvic incidence and pelvic rotation values impossible in those cases.rative period the contribution was statistically the same (34% and 37%, respectively; p=0.167).At the three month point, following placement of the lordotic interbody spacer, the contribution of segment L5-S1 increased (p<0.001) and the contribution of segment L4-L5 decreased (p<0.001).Future studies are needed, to investigate whether restoration of the spinopelvic sagittal parameters of the adjacent levels can provide direct protection against the degenerative cascade.

CONCLUSION
In degenerative diseases of the isolated L5-S1 level, there is loss of segmental lordosis resulting in an increase in the angulation of the upper adjacent level.In the group studied, following anterior interbody fusion using lordotic spacers, it was possible to observe an increase in the lordosis of the level treated and a concomitant decrease in the lordosis of the adjacent mobile level.

Figure 1 .
Figure 1.Case example with baselines for measurements.(A) Preoperative radiograph and (B) radiograph 3 months following surgery.The white dashed lines show the determination of the planes (from top to bottom: upper plateau of L1, upper plateau of L4, lower plateau of L4, upper plateau of L5, horizontal plane, lower plateau of L5, upper plateau of S1, and horizontal plane) for the determination of the angles described in the Methods section.

Table 2 .
Contribution to lumbar lordosis by specific level.