MODIFICATION OF THE SPINOPELVIC PARAMETERS WITH SURGICAL POSITIONING AND THE ROLE OF THE HIP

Post-operative lumbar pain is related to alterations in the sagittal and spinopelvic parameters. A lumbar fusion that fails to maintain, or that worsens the physiological lordosis alters the sagittal balance. Objective: To analyze spinopelvic variation in different surgical positions. Methods: A prospective, analytical and comparative study of spinograms, lumbosacral radiographs in the surgical position over a 4-pole quadratus lumborum, and lumbosacral radiographs with quadratus lumborum and support in the knees. A sample of 129 patients, of both sexes, aged between 18 and 60 years, and presenting with lower back pain. Lumbar Lordosis (LL), Pelvic Tilt (PT), Pelvic Incidence (PI) and Sacral Slope (SS) were measured. Results: PI was the most stable parameter. With quadratus lumborum, a slight increase in PT, a decrease in SS and a significant reduction in LL were found. With quadratus lumborum and support in the knees, a decrease in PT and a slight increase in SS were found, while the LL value remained similar to that of the spinogram. Conclusion: The intraoperative position with hip flexion of between 40o and 45o over quadratus lumborum reduced LL to 10.52o in men and 16.21o in women, increased PT, and decreased SS. The intraoperative position with hip flexion of between 0o and 10o showed the same values as the reference spinogram. Level of Evidence II; Prospective comparative study.


INTRODUCTION
The spinal cord is the main axis for maintaining the correct standing position in humans, with Sagittal Balance (SB) being one of the most important factors. Lumbar degeneration results in a loss of lumbar lordosis, with the subsequent activation of compensatory mechanisms in order to maintain the capacity to walk, but which create greater muscular tension and stress on the static subsystems, causing them to become exhausted. Clinical consequences are lumbar pain and loss of quality of life. [1][2][3][4][5] Postoperative lumbar pain is also closely related to alterations in sagittal parameters and spinopelvic angulation. [6][7][8] In recent decades, the increase in vertebral instrumentations has led to a clear increase in the rate of instrumented lumbar arthrodesis. Current technology has brought several benefits for surgeons, although it has also led to the appearance of a greater incidence of flat black syndrome.
The term "flat back syndrome", also known as lumbar kyphosis, was initially described by Doherty (1973), based on a case of forward tilt of the core due to loss of normal lumbar lordosis. This can be caused by many factors, but the most common origin is iatrogenic, based on lumbar attachments in hypolordosis. 9 Incorporating the spinopelvic parameters in the surgical planning is of great importance, and is considered essential for a good postoperative result. [1][2][3][4][5][6][7][8][9][10] It is well established that the spinal cord adapts itself to the pelvic parameters, hence the importance of a spinopelvic analysis. 11 A lumbar fusion that fails to maintain, or that worsens physiological lumbar lordosis results in accelerated wear on the adjacent musculature and loss of the sagittal balance. 12 Many works have shown the importance of the study of spinal sagittal balance and spinopelvic angulation, not only in healthy individuals, but also in spinal cord surgery. 6-7-13-14 The aim of this work is to analyze spinopelvic variation in different surgical positions.

METHODS
Following approval by the Ethics Committee of Sanatorio Allende, we conducted a prospective, analytical and comparative study, with two independent observers, of spinograms in the standing position, lumbosacral radiographies simulating surgical position in ventral decubitus over a 4-pole quadratus lumborum (30 centimeters high) with hip flexion of between 40º to 45º, and lumbosacral radiography with quadratus lumborum with extra knee support (40 centimeters high) and hip flexion of between 0º and 10º. The sample consisted of 129 patients who presented with lower back pain between April 2017 and February 2018, at the Orthopedic and Trauma Service of our institution. All the patients who participated in the study signed an Informed Consent Form. The inclusion criteria were: patients who had previously presented with lower back pain in their first appointment, patients of both sexes, aged between 18 to 60 years. The exclusion criteria were: patients who had previously undergone spinal cord surgery and had chronic lumbar symptomatology or the existence of a structural pathology of the spinal cord. There was no conflict of interest in relation to this study.

Radiological Studies
This work was conducted in the Diagnostic Imaging Service, which was given an explanatory protocol that included the inclusion and exclusion criteria and the rules to follow in order to simulate intraoperative positions. The imaging studies used for the analysis were: spinogram in the standing position (side projection) (Figure 1), lumbosacral radiography (side) in the surgical position with quadratus lumborum (hip flexion) ( Figure 2) and lumbosacral radiography (side) in the surgical position with additional support in the knees (extended hip) ( Figure 3).
The variables analyzed were as follows: Lumbar Lordosis (LL) Physiological lumbar lordosis is an important parameter that must be considered in spinal cord surgeries, as it maintains normal sagittal balance and is related to Pelvic Incidence.
This ratio can be used to determine the ideal LL in a given patient using the following formula LL=PI +/-10º. 15

Pelvic Parameters
In the side projection, the three parameters that determine pelvic configuration were measured. First, Pelvic Incidence (PI), which   determines the width of the pelvis using an anatomically fixed angle, which is specific to each individual and does not vary after adolescence. Second, Pelvic Tilt (PT), which shows the rotation of the pelvis around the femoral heads: in retroversion, PT increases, while in anteversion, PT decreases, and lastly, Sacral Slope (SS), which is a compensatory angle of PT and is characterized by the position of the base of the sacrum (S1).
Mathematically, the pelvic parameters are expressed by the expression PI= PT+SS. These four parameters were measured in each patient, with the aid of the software program SurgiMap. Figure 4 shows how they were measured and how the observations of each indicator were obtained. This measurement software can be applied not only in research but also in clinical fields. Its high reliability was recently demonstrated; consequently, it can be incorporated as a tool for sophisticated sagittal alignment. 16

RESULTS
The patient's average age was comparable for both sexes (p=0.949), with average ages (standard deviations) of 31.7 (9.4) years for men and 31.5 (9.9) for women. This was not associated with the lumbopelvic parameters evaluated (PT, p=0.356; IP, p=0.401; LL, p=0.694; SS, p=0.791), in any of the rehearsed positions. The estimated rates of correlation were below 0.18 in all cases, except for the position that included the surgical quadratus, in which a slight tendency towards lordosis (LL, r=0.24, p=0.091) was observed. However, noticeable differences were found between sexes in the measured parameters in the side spinogram. Table 1 presents the statistical summaries for each characteristic under the reference position (spinogram) and by sex, reflecting values that are significantly greater for lumbar lordosis and sacral slope in women, apart from the smaller value obtained for pelvic tilt when being compared to the other parameters at a general level (both sexes).
To complement this, Figure 5 illustrates the empirical distribution for each variable used in the reference evaluation (spinogram), compared to the two innovated positions (with quadratus lumborum and additional support in the knees), and shows the lack of adhesion among the three series and in each parameter. This lack of adhesion, as mentioned previously, was closely associated with sex (p-global value 0.0107), as can be seen in Tables 2 and 3, with distinctive features in the female group.
Pelvic incidence, a parameter considered an invariant quality of each individual, was measured and also mathematically calculated by adding PT and SS. In the adopted positions (spinogram, with quadratus lumborum and quadratus lumborum plus support in the   obtained from the three positions depended largely on the patient's sex, and not on the patient's age. This is why the following description is only related to that biological factor. This adhesion was not proportional, but has a particular behavior for each sex. In order to obtain reference values and predictions by sex, several regression models were performed. Intercepts and linear coefficients of those models were estimated to describe the relationship between pelvic parameters, considering the different positions, and using the spinogram as a baseline. Table 4 shows the estimated coefficients for each sex. It can be observed that for each parameter, the y-intercepts (expressed as subscript 1 and 2 for the relation with quadratus lumborum and quadratus lumborum plus support in the knees, respectively) interpreted as discharges in absolute terms in relation to the reference value (spinogram), are significantly different between sexes. For example, the model between PT (quadratus lumborum) and PT (spinogram) is estimated as indicating that for each unit of increase in PT (spinogram), the PT to be obtained in an assisted position with quadratus lumborum does not increase in the same proportion, but increases by 0.70 compared to the former, with a discharge of 2.80. This means that the lack of adherence between both measurements could not be described as a small degree of change, but maintains a model of linear regression with positive and significant slope. These models obtained for each parameter were all significant and depended on sex (except for PI), with more extreme (and significant) estimates in lordosis and sacral slope. The exhaustive analysis of these estimates would indicate the existence of alterations to the parameters in the first and second positions, incorporating quadratus lumborum and support, and the reference values could only be recovered via the spinogram through those functional ratios. Figures 4-8 illustrate, for some of the parameters, the satisfying performance of these predictive models.
In practical terms, for both sexes, the ratio PT (quadratus lumborum):PT (spinogram) was 1.38, this ratio being arrived at by averaging 0.91 for men and 1.85 for women (opposite ratio). When the position with support in the knees is considered, the ratio PT(C+R):PT was 0.32 for men and 0.48 for women.
For pelvic incidence (quadratus lumborum), this ratio was estimated at 0.95, and with support in the knees, 0.98, this being the most stable parameter of the studies carried out in this work. For lordosis, these ratios were also significantly different from 1: under the quadratus lumborum position: 0.83 and 0.71 for men and women, respectively, and with the addition of support in the knees: 1.06 and 0.95 for men and women, respectively. This behavior was similar   Table 4. Values and estimates of the coefficients of the linear regression model (y-intercept and slope) for each pelvic parameter and for both sexes, considering a spinogram as reference (independent variable) and quadratus lumborum (1) and quadratus lumborum with support in the knee (2) as a variable answer (to be predicted or dependent).    for the evaluation of sacral slope, with 0.98 and 0.89 (quadratus lumborum), 1.12 and 1.07 (quadratus lumborum plus support) for men and women, respectively.

DISCUSSION
Spinopelvic sagittal balance, and in particular, lumbar lordosis, must be maintained through correct intraoperative positioning; otherwise, changes will occur that will result in a reduction in lumbar lordosis and a positive balance at the postoperative sagittal level. This positive balance has been reported in the literature as    the beginning of lumbar pain. 17 When planning a lumbar fusion, positions should be used that maintain this spinopelvic physiological balance. Therefore, two surgical positions were tested in this work, taking the spinogram in the standing position as reference. Significant differences were found when the surgical position with 4-pole quadratus lumborum was simulated (hip flexion between 40º to 45º), with a slight increase in PT, a decrease in SS and a significant reduction in LL. In the simulation of the surgical position with quadratus lumborum and additional support in the knees (hip flexion 0º to 10º), less accentuated differences were found, with a reduction in PT and a slight increase in SS, while the value of LL remained similar to that of the spinogram. This last position was the one that was most similar to reference values. This work contributes valuable additional information for future practice, as it obtains and estimates models that can be used for predictive purposes to analyze changes that would be obtained in different clinical situations in relation to the reference values obtained from the spinogram.
Many authors have carried out research on normal physiological lordosis. Anderson and Cols showed average lumbar lordosis from L1 of 59.8. Other authors such as Tan and Cols, obtained an average of 55.6º and a range of 38º to 70º.
Our results demonstrate an average lordosis of 58.77º in men and 65.06º in women for the spinogram in the standing position. 18,19 LL was the most variable parameter found in women, but did not vary significantly with age. Some publications state that the degenerative process reduces this variability. PI was the most stable parameter found in different clinical trials, with average values of 50.88º and 51.41º for men and women, respectively, which is in agreement with the literature. 20,21 There have been some studies on the behavior of lumbar lordosis in instrumented surgeries. Works such as those of Guanciale's et al. and Stephens et al. show a variation in lumbar lordosis between the Jackson and the Andrews positions (hip flexion between 60º and 90º), which explains the importance of hips at the moment of surgical positioning. 22,23 Different works compare the variation in lordosis using four different surgical supports, but none of the analyses examines the change in pelvic parameters as separate from lumbar lordosis, in simulated surgical positions in relation to the reference spinogram.
Our predictive model of adjustable parameters with surgical decubitus could be of great use for preoperative planning especially in deformity pathologies, in which the priority is to restore the sagittal balance.
A weakness of our work is its relatively small sample and the age of the patients, who were mainly young. This is because at older ages, the degenerative process may influence the degree of variation in the studied parameters.

CONCLUSION
The intraoperative position with hip flexion between 40º and 45º over four-pole quadratus lumborum used in spinal cord surgeries not only reduces lumbar lordosis to 10.52º in men and 16.21º in women, but also modifies pelvic parameters, with a slight increase in PT and a decrease in SS, compared to the spinogram in the standing position. But the intraoperative position with hip flexion between 0º and 10º shows the same values as the reference spinogram.
We believe that a precise knowledge of these changes is very important when planning the surgical strategy.

CONTRIBUTION OF THE AUTHORS:
Each author made significant individual contributions to this manuscript. CM (0000-0001-6444-9921)* and DO (0000-0003-0988-305X)* were the main contributors to the writing, design and discussion of the results of the manuscript. These two authors, together with AG (0000-0002-9480-2906)* and RC (0000-0001-9176-074X)*, contributed to the acquisition, analysis and interpretation of the data. ON (0000-0001-7461-3879)* and GJ (0000-0002-8887-8947)* contributed to the review and approval of the final version of the work. All authors carried out the bibliographic research and contributed to the intellectual concept of the study. *ORCID (Open Researcher and Contributor ID).