LUMBAR LORDOSIS VARIATION ACCORDING THE TYPE OF POSITIONER USED IN LUMBAR ARTHRODESIS

ABSTRACT Objective: Evaluate the influence of the most used surgical positioners for lumbar lordosis (LL) in asymptomatic individuals. Methods: Cross-sectional study based on demographic data and radiographic parameters of asymptomatic individuals. For this study, 16 volunteers, 15 males, and one female were selected, and the average age was 24.6 years. They were submitted to lateral radiographs of the lumbar spine in orthostasis in use of the following positioners: gel cushion, gel cushion with hip extension, four-point Relton-Hall and Wilson-type positioner. Results: The mean LL in the orthostatic position was 58.76º, whereas in the gel cushion positioner it was 52.51; on the gel cushion with hip extension of 58.23º, Relton-Hall/4points 37.63º and, finally, on the Wilson-type positioner of 40.87º. An average reduction of 5.42º of the LL was observed when positioning on the gel cushion in relation to the orthostasis. In the linear regression analysis, the data presented statistically significant results (p<0.05), demonstrating that the L4-S1 segment influences 60% in LL. Conclusion: The positioner with gel cushion and hip extension reproduces an LL similar to physiological values. Relton-Hall and Wilson-type positioners with hip flexion promote hypolordotic positioning compared to basal lordosis in orthostasis. Hip extension alone generated a 5.96º increase in the subject’s lordosis. The L4-S1 segment has a 60% influence on the LL when the individuals are in the positioners. Level of evidence III; Controlled cross-sectional study.


INTRODUCTION
Procedures for spinal arthrodesis are becoming more frequent.There was a 137% increase in the annual number of spinal fusion surgeries between 1998 and 2008 in the US, from 174,223 to 413,171 procedures in 2008. 1 An optimal sagittal balance is conducive to better arthrodesis results, such as a higher fusion rate and less adjacent-level degeneration. 2][5] Measures for optimization of the LL involve knowledge of the individual's previous sagittal parameters, proper surgical technique, and also correct intraoperative positioning.Surgical positioning is critical to maintain LL and prevent iatrogenic flatback. 6Lenke and colleagues observed that patients with deformities in the sagittal plane and reduced LL (average of 25.9°) showed an average gain of 17.2° with surgical placement alone. 7everal studies have demonstrated the influence of intraoperative positioning on LL, especially the gain of LL in the prone position with hip extension, showing it to be a valid strategy for optimizing LL. [7][8][9][10] These studies evaluated patients positioned mainly on Jackson and Allen tables.However, these surgical tables and positioners are not available in the operational routine of most hospitals.Furthermore, to the authors' knowledge, no work has been published to evaluate the influence of surgical positioners available in our environment on LL.
Thus, this study aimed to evaluate the influence of the most commonly used surgical positioners in our environment on lordosis of the lumbar spine in asymptomatic individuals.

METHODS
After approval by the Research Ethics Committee (CAAE: 55519622.0.0000.0023)and signing the Informed Consent Form (ICF), 16 volunteers aged 18 years or older, asymptomatic, and with no previous history of spinal pathology or surgery were included.Initially, the demographic data of each individual (gender, age, weight, and height) were collected, and lateral radiographs of the lumbar spine were taken in orthostasis and the following positioners: gel cushion, gel cushion with hip extension, Relton-Hall (4 points), and Wilson-type positioner.(Figures 1 to 4) The same researcher adjusted the volunteers in the positioners in a standardized way.Then, for a period of 5 minutes, waited for accommodation and relaxation of the paravertebral musculature between successive positions.Then, all radiographs were taken according to technique and in digital format, encompassing the upper L1 plateau and the femoral heads in the same image to allow a correct measurement of the spinopelvic parameters.
The parameters measured in each radiograph were: lumbar lordosis (angle between the upper plateau of L1 and S1), pelvic incidence (PI), pelvic tilt (PT), and sacral slope (SS) (Figure 5), using Surgimap® software (Nemaris Inc., New York, NY, USA).The variables obtained from lateral radiographs of the lumbar spine in orthostasis were compared with those of the radiographs taken with each surgical positioner.Continuous variables were expressed as mean and standard deviation for this evaluation and compared between groups using paired Student's t-test.The results and p-values (considering values less than 0.05 as statistical significance) are listed in Tables 1 and 2. A simple linear regression analysis was also performed to evaluate the effect of lumbar lordosis L4-S1 (dependent variable) on LL (independent variable), that is, the influence of L4-S1 tracking on LL composition and change (Table 3).The analyses were performed in the statistical program IBM® SPSS Statistics (version 22.0; SPSS, Chicago, IL, USA).

RESULTS
Of the sample of 16 volunteers, 15 were male and one female, with a mean age of 24.6 years (minimum 19, maximum 39, and standard deviation 5.6), mean height 176 cm (minimum 168, maximum 188, and standard deviation 6.2), and mean body mass 77.4 kg (minimum 49, maximum 102, and standard deviation 14.5).
The mean LL in the orthostatic position ("orthostatism") was 58.76°, in the gel-cushion positioner ("gel") 52.51°, in the gel-cushion positioner with hip extension ("extension") 58.23°, in the Relton-Hall positioner/four-point positioner ("4-point") 37.63° and finally in the Wilson-type positioner ("Wilson") 40.87°.(Table 1) When performing the t-test, we observed a mean reduction of 5.42° of LL when positioning on the gel cushion to orthostasis, a reduction of 19.74° with the 4-point positioning, and 17.73° with the Wilson-type positioner, all values with statistical significance (p<0.05).When comparing the LL in orthostasis with the "extension" group, a mean reduction of only 0.5° in lordosis was observed without statistical significance (P=0.754).When comparing the "gel" group with the "extension" group, there was an increase of 5.96°, also with statistical significance (P<0.05).Hip extension alone favored a significant increase in LL. (Table 2) When specifically evaluating the variations in lordosis of the L4-S1 follow-up, we observed a reduction of 3.90° in the "gel" group (p=0.067),0.58° in the "extension" (p=0.694),12.58° in the 4-point (P=0.000) and 17.16° (p=0.000) in the Wilson positioner relative to the standing position.When analyzing the group positioned on the gel cushion to the group with hip extension, there was a 3.32° increase in lordosis L4-S1; again hip extension alone showed statistical significance (p<0.05).(Table 3) In the linear regression analysis, it was observed with statistical significance that the L4-S1 segment has a 60% influence on the LL when the subjects are in the positioners, even though the L4-S1 segment has only two movable discs (Table 4).For example, hip extension alone promotes a 5.42° increase in LL ("gel" vs. "gel with the extension"), and 3.32° (61%) of this increase occurred in the L4-S1 segment.

DISCUSSION
The mean LL of the 16 volunteers was 58.76° in the standing position, with a significant reduction in the prone position in all positioners studied, except for the gel cushion with hip extension, which reproduced lordosis in the standing position.The linear regression analysis showed the importance of the L4-S1 follow-up in the composition and alteration of the LL, with an influence of up to 60%.Different studies have also evaluated the influence of intraoperative positioning on lumbar lordosis. 7,9,11an and colleagues analyzed the variation in intraoperative positioning in the LL of 10 volunteers.They found a reduction of up to 57% in the LL of subjects positioned with hip flexion.In contrast, the LL of the standing position was reproduced by positioning the volunteers with pads supporting the thorax and pelvis.In addition, they also reported that the LL change was accompanied by a change in the angles between the vertebral bodies and that most of this change was located in the L4-L5 and L5-S1 segments. 11These   data agree with our study, contributing to the understanding of hip extension as a tool to optimize the LL and the need for more attention to the optimization of lordosis of the L4-S1 segment, since this influences the LL in up to 60%.Just as hip extension showed a gain in LL, hip flexion worsens these values.In the study conducted by Stephens and colleagues, 9 the average LL in 10 asymptomatic volunteers was 51.7°.By positioning them on the Andrews table with 90° of hip flexion, the LL reduced significantly to 17°.When reduced the flexion to 60°, the average LL increased to 27.3º.However, the L1-L4 follow-up accounted for 80% of the change in lumbar lordosis with minimal changes in the L4-L5 and L5-S1 discs, a finding that diverges from our linear regression analysis.
Lenke and colleagues 7 evaluated the effect of intraoperative positioning (prone position with hips extended on a Jackson table) on LL of adults with spinal deformity.They concluded that the patients with the greatest lordosis gain with intraoperative positioning were hypolordotic (LL mean 25.9°) and had worse sagittal balance.While the group with a preoperative mean LL of 54.2° and better sagittal parameters showed no significant change with intraoperative positioning.This study differs from our population by evaluating patients with spinal deformities already with surgical indication, while the present study evaluated asymptomatic patients.The authors believe that the optimization of LL with intraoperative positioning will be even greater in patients with complex deformities who already present with reduced LL and sagittal imbalance.This is precisely the group of patients that require the greatest corrections.
As already described, there was a statistically significant reduction in the LL of individuals in the "4-point" and "Wilson" groups, 19.74° and 17.73°, respectively.The difference in reduction between these positioners was not statistically significant, i.e., both negatively influence the lordosis of the individual similarly.The authors believe that this reduction is mainly the result of greater hip flexion.In spinal arthrodesis surgeries, these positioners may provide a fixation in hypolordosis relative to the patient's baseline, favoring iatrogenic flatback, leading to pain, disc degeneration, and future changes in sagittal balance. 4,6he main limitations of this study are the small sample size of the volunteers, the fact that it was carried out in only one center, which reduces the generability of the data, and finally, for ethical reasons, it did not evaluate the effect of anesthesia (muscle relaxation) on lordosis gain.However, it is believed that there will be an even greater increase in lumbar lordosis when the patient is anesthetized in the prone position and when hip extension is used as a tool in positioning hypolordotic patients with spinal deformities.Furthermore, to the authors' knowledge it was the only paper to date that evaluated the cushions available in our environment.

CONCLUSION
Positioning with gel pads and hip extension reproduces an LL similar to physiological values.Relton-Hall and Wilson-type hip flexors promote hypolordotic positioning compared to basal lordosis in orthostasis.Hip extension alone was able to generate a 5.96° increase in lordosis for the individual.The L4-S1 segment has a 60% influence on the LL when the subjects are in the positioners.Hip extension is a valid strategy when seeking to optimize LL gain intraoperatively.

Figure 1 .
Figure 1.Positioning in gel cushions with support for face, thorax and pelvis.

Figure 2 .
Figure 2. Positioning on gel cushion with hip extension.

Table 1 .
Mean values of the lumbar lordosis in the different positioners.

Table 2 .
Comparison of lumbar lordosis (LL) between different positions.
MV: Mean L1-S1 variation between positions; p: paired T-test (p<0.05).Positive values (MV LL) indicate reduced LL of the second group compared to the first.

Table 3 .
Comparison of lordosis of L4-S1 between different positions.

Table 4 .
Results of the simple linear regression analysis.