CORRELATION BETWEEN DEGENERATIVE DISEASES OF THE LUMBAR SPINE AND TYPES OF LUMBAR LORDOSIS

CORRELAÇÃO ENTRE DOENÇAS DEGENERATIVAS DA COLUNA LOMBAR E OS TIPOS DE LORDOSE LOMBAR

CORRELACIÓN ENTRE LAS ENFERMEDADES DEGENERATIVAS DE LA COLUMNA LUMBAR Y LOS TIPOS DE LORDOSIS LUMBAR

ABSTRACT

Objective:

Evaluate the relationship between the incidence of different types of degenerative diseases of the spine and lumbopelvic biomechanics, according to the types of lordosis of Roussouly's classification.

Methods:

Retrospective study of medical records and results of imaging exams of patients seen at a private hospital in São Paulo. The sagittal alignment of these patients was evaluated by classifying them according to Roussouly into 4 types, based on panoramic radiographs of the spine. These results were correlated with the patient's degenerative diagnosis (Herniated disc, Canal stenosis, Spondylolisthesis, degenerative discopathy and Facet arthrosis). Statistical tests were performed comparing the types of curvature and diagnoses identified.

Results:

418 patients were evaluated, 51.4% male and 49.6% female. The vast majority of patients, about 54%, had a diagnosis of herniated lumbar disc. There was a statistically significant difference that showed a predilection for surgical treatment in cases classified as Type I and Type II in the Roussouly classification. There was no statistically significant difference that correlated the types of lumbar lordosis with the diagnosis presented by the patients.

Conclusion:

There is no statistically significant difference that correlates the type of lumbar lordosis according to Roussouly with lumbar degenerative diseases. In contrast, patients classified as Type 1 and Type 2 by Roussouly underwent a greater number of surgical treatments compared to patients type 3 and 4, with statistical relevance. Level of evidence 2; Retrospective prognostic study.

Keywords:
Low Back Pain; Chronic Disease; Spondylolisthesis; Intervertebral Disc; Spondylolysis

RESUMO

Objetivos:

Avaliar a relação da incidência dos diferentes tipos de doenças degenerativas da coluna com a biomecânica lombopélvica, de acordo com os tipos de lordose segundo Roussouly.

Métodos:

Estudo retrospectivo de prontuários médicos e de resultados de exames de imagens de pacientes atendidos em um hospital privado de São Paulo. Foi avaliado o alinhamento sagital desses pacientes classificando-os de acordo com Roussouly em 4 tipos, com base nas radiografias panorâmicas da coluna vertebral. Esses resultados foram correlacionados com o diagnóstico degenerativo do paciente (Hérnia de disco, Estenose do canal, Espondilolistese, Discopatia degenerativa e Artrose facetaria). Testes estatísticos foram realizados comparando os tipos de curvatura e diagnósticos identificados.

Resultados:

Foram avaliados 418 pacientes, sendo que 51,4% do sexo masculino e 49,6% do sexo feminino. A grande maioria dos pacientes, cerca de 54%, apresentavam como diagnóstico hérnia de disco lombar. Houve uma diferença estatisticamente significativa que evidenciou uma predileção do tratamento cirúrgico nos casos classificados como Tipo I e Tipo II na classificação de Roussouly. Não houve diferença estatisticamente significativa que correlacionasse os tipos de lordose lombar com o diagnóstico apresentado pelos pacientes.

Conclusões:

Não houve diferença estatisticamente significativa que correlacione o tipo de lordose lombar de acordo com Roussouly com as doenças degenerativas lombares. Em contrapartida, os pacientes classificados como Tipo 1 e Tipo 2 de Roussouly foram submetidos em maior número ao tratamento cirúrgico em comparação com os pacientes tipo 3 e 4, com relevância estatística. Nível de evidência 2; Estudo prognóstico retrospectivo.

Descritores:
Dor Lombar; Doenças Degenerativas; Espondilolistese; Disco Intervertebral; Espondilólise

RESUMEN

Objetivos:

Evaluar la relación entre la incidencia de diferentes tipos de enfermedades degenerativas de la columna y la biomecánica lumbopélvica, según los tipos de lordosis de Roussouly.

Métodos:

Estudio retrospectivo de registros médicos y resultados de exámenes de imágenes de pacientes atendidos en un hospital privado en São Paulo. La alineación sagital de estos pacientes se evaluó clasificándolos según Roussouly en 4 tipos, en base a radiografías panorámicas de la columna vertebral. Estos resultados se correlacionaron con el diagnóstico degenerativo del paciente (disco herniado, estenosis del canal, espondilolistesis, discopatía degenerativa y artrosis facetaria). Se realizaron pruebas estadísticas comparando los tipos de curvatura y los diagnósticos identificados.

Resultados:

Se evaluaron 418 pacientes, 51.4% hombres y 49.6% mujeres. La gran mayoría de los pacientes, alrededor del 54%, tenían un diagnóstico de hernia de disco lumbar. Hubo una diferencia estadísticamente significativa que mostró una predilección por el tratamiento quirúrgico en los casos clasificados como Tipo I y Tipo II en la clasificación de Roussouly. No hubo diferencias estadísticamente significativas que correlacionaran los tipos de lordosis lumbar con el diagnóstico presentado por los pacientes.

Conclusión:

No hubo diferencias estadísticamente significativas que correlacionen el tipo de lordosis lumbar según Roussouly con las enfermedades degenerativas lumbares. En contraste, los pacientes clasificados como Tipo 1 y Tipo 2 por Roussouly se sometieron a un mayor número de tratamientos quirúrgicos en comparación con los pacientes tipo 3 y 4, con relevancia estadística. Nivel de evidencia 2; Estudo de pronostico retrospectivo.

Descriptores:
Dolor de la Región Lumbar; Enfermedad Crónica; Espondilolistesis; Disco Intervertebral; Espondilólisis

INTRODUCTION

Lumbar arthrodesis is a widely-used surgical technique for treatment of various spinal pathologies, including degenerative diseases, traumas, and deformities.^1,2 The initial objective of the procedure is to obtain fusion between vertebral segments to promote the reestablishment of stability and alignment lost due to pathologies that affect the spine.^3,4

Although spinal arthrodesis is often effective in relieving pain and providing some degree of functional recovery, the procedure is not without potential issues. Spinal mobility is an integral component of the activities of daily life and the stiffness associated with arthrodesis can lead to limitations on individual functional capacity.^5,6 Thus, the lumbar stiffness disability index questionnaire was developed in order to better understand the limitations on activities of daily life resulting from stiffness secondary to lumbar spine arthrodesis.⁷

The objective of the present study was to evaluate the impact of stiffness associated with lumbar arthrodesis on functional capacity and the quality of life in order to gain a better understanding of the functional limitations that arthrodesis at different levels of the lumbar spine can cause.

METHODS

Type of study and population

This is a retrospective study evaluating 40 patients who underwent spinal arthrodesis surgery, including the lumbar segment. The study was approved by the institutional review board (CAAE: 82012017.6.0000.5463) and all patients signed the informed consent form. Patients who underwent spine surgery with arthrodesis, the extent of which included at least one lumbar segment (from L1-L2 to L5-S1), for the treatment of degenerative diseases, traumas, or deformities of the spine, with minimum postoperative follow-up of 24 months were included. Patients whose arthrodesis extended only as far as T12-L1 were not included, since we consider said segment to be the transition between the thoracic and lumbar spines and without the same biomechanics as the lumbar spine.

Patients who had undergone surgical procedures to treat oncologic conditions were excluded because systemic compromise from the disease can interfere with the functional capacity assessment. Patients with other associated orthopedic diseases such as sacroiliitis, coxarthrosis, gonarthrosis, and pseudarthrosis in bones of the lower limb were also excluded. Another exclusion factor was the occurrence of complications associated with the spine surgery, such as loosening of the implants, pseudarthrosis, or adjacent level disease, which were limiting the postoperative functional assessment.

Data collection

All the patients who met the inclusion and exclusion criteria were invited to participate in the study. Those who accepted moved on to the data collection phase. Demographic data, including sex, age, age at the time of surgery, and duration of follow-up, were considered, in addition to information about the surgical procedure, especially the extent of the arthrodesis. The version of the LSDI questionnaire translated and adapted for Brazilian Portuguese⁸ was applied to evaluate limitations on the activities of daily life due to stiffness secondary to lumbar spine arthrodesis. The higher the LSDI score, the greater the functional limitation indicated by the patient.

To quantify clinical postoperative lumbar stiffness/mobility, the modified-modified Schöber test (MMST)⁹ was administered to the patients. With the patient in orthostatism, the evaluator locates and demarcates the posterior superior iliac spine, also making a corresponding mark in the midline of the spine (caudal mark). Then, a point is drawn 15 centimeters above this caudal mark. Finally, the patient flexes the trunk with the knees in extension and the new distance between the points is calculated. The MMST value is indicated by the difference obtained between the two measurements. The lower the MMST value, the greater the lumbar stiffness.

Statistical analysis

The statistical analysis was conducted using SPSS v.20 software (IBM Corp., Armonk, NY, USA). The normality of the distribution of the samples was analyzed using the Shapiro-Wilk test. The linear correlation between the LSDI score and the number of arthrodesed levels, as well as between the LSDI score and the MMST measurement, were analyzed using the Spearman rank correlation test.

The patients were then divided into two groups: Group 1, arthrodesis extending to the sacrum; Group 2, arthrodesis not extending to the sacrum. The Mann-Whitney test was performed to compare the LSDI score values between the two groups.

A value of p < 0.05 was considered for the level of significance.

RESULTS

Sample

Eighteen patients (45%) were male and 22 (55%) were female. The age of the patients ranged from 18 to 79, with a mean of 57.7 years of age (standard deviation [SD]: 16.2). The minimum postoperative follow-up time was 2 years, and the maximum was 19 years (mean 7.5 years, SD: 4.2). The mean body mass index (BMI) of the sample was 28 (SD: 4.6).

Correlation between lumbar stiffness and functional limitation

The mean MMST value was 3.75 cm (SD: 1.5 cm), ranging from 0 to 7 cm. (Table 1) The mean LSDI score was 41.7 (SD: 20.6), ranging from 0 to 75. (Table 1, Figure 1) There was a moderate negative (r= −0.320) but statistically significant (p= 0.04) correlation between the MMST value and the LSDI score. (Figure 2)

Figure 1
Graph illustrating the distribution of Lumbar Stiffness Disability Index questionnaire scores among the patients of the sample.

Figure 2
Comparison of the Lumbar Stiffness Disability Index scores between patients with and without extension of the arthrodesis to the sacrum.

Table 1
Total sample modified-modified Schöber test and Lumbar Stiffness Disability Index questionnaire score values.

Characteristics of the sample (n = 418)		n	%
Sex
	Female	203	48.60%
	Male	215	51.40%
Roussouly Classification
	Type 1	47	11.20%
	Type 2	159	38.00%
	Type 3	168	40.20%
	Type 4	44	10.50%
Diagnosis
	Facet arthrosis	23	5.50%
	Degenerative discopathy	95	22.70%
	Spondylolisthesis	31	7.40%
	Canal stenosis	41	9.80%
	Disc herniation	219	52.40%
	Normal	9	2.20%
Type of treatment performed
	Surgical	209	50.00%
	Conservative	209	50.00%

• IQR = Interquartile Range

Arthrodesis levels and functional limitation

Regarding the levels of the lumbar spine arthrodesis, only one level was involved in most patients (18, 45%) (Table 2). Eleven (27.5%) patients had arthrodesis in two levels, five (12.5%) in three levels, five (12.5%) in four levels, and only one patient (2.5%) with all five lumbar levels involved in the arthrodesis. There was no correlation between the number of levels involved in the lumbar arthrodesis and the LSDI score (p= 0.160).

Table 2
Lumbar Stiffness Disability Index questionnaire scores by number of arthrodesed lumbar levels.

Factors	Roussouly Classification				p value
	Type 1	Type 2	Type 3	Type 4
Type of treatment (n=418)					0.008a
Surgical	30 (63.8%)	90 (56.6%)	71 (42.3%)	18 (40.9%)
Conservative	17 (36.2%)	69 (43.4%)	97 (57.7%)	26 (59.1%)
Sex (n=418)					0.632a
Female	19 (40.4%)	77 (48.4%)	86 (51.2%)	21 (47.7%)
Male	28 (59.6%)	82 (51.6%)	82 (48.8%)	23 (52.3%)
Diagnosis (n=409)					0.246 b
Facet arthrosis	2 (4.3%)	11 (7%)	7 (4.3%)	3 (7.3%)
Degenerative discopathy/disc herniation	35 (74.5%)	120 (75.9%)	131 (80.4%)	28 (68.3%)
Spondylolisthesis	3 (6.4%)	9 (5.7%)	11 (6.7%)	8 (19.5%)
Canal stenosis	7 (14.9%)	18 (11.4%)	14 (8.6%)	2 (4.9%)

• ^a
  Chi-squared test;
• ^b
  Likelihood ratio test.

Influence of the extension of the arthrodesis to the sacrum on functional limitation

In the sample, 22 (55%) of the patients had arthrodesis extending to the sacrum. (Table 3) The mean LSDI score of the patients with extension of arthrodesis to the sacrum was statistically higher than that of patients with arthrodesis that did not extend to the sacrum (p= 0.002), (Table 3, Figure 3) indicating greater functional limitation in those with extension to the sacrum.

Figure 3
Comparison of the Lumbar Stiffness Disability Index scores between patients with and without extension of the arthrodesis to the sacrum.

Table 3
Lumbar Stiffness Disability Index questionnaire scores for patients with and without extension of the arthrodesis to the sacrum.

Diagnosis	Roussouly Type 1		Roussouly Type 2		Roussouly Type 3		Roussouly Type 4
	No	Yes	No	Yes	No	Yes	No	Yes
Facet arthrosis
Other diagnosis	380 (87.8%)	53 (12.2%)	276 (63.7%)	157 (36.3%)	254 (58.7%)	179 (41.3%)	389 (89.8%)	44 (10.2%)
Yes	35 (85.4%)	6 (14.6%)	26 (63.4%)	15 (36.6%)	29 (70.7%)	12 (29.3%)	33 (80.5%)	8 (19.5%)
P value	0.657		0.967		0.132		0.067
Adjusted p value	0.846		0.988		0.66		0.66
Degenerative discopathy
Other diagnosis	330 (87.1%)	49 (12.9%)	240 (63.3%)	139 (36.7%)	233 (61.5%)	146 (38.5%)	334 (88.1%)	45 (11.9%)
Yes	85 (89.5%)	10 (10.5%)	62 (65.3%)	33 (34.7%)	50 (52.6%)	45 (47.4%)	88 (92.6%)	7 (7.4%)
P value	0.526		0.725		0.116		0.209
Adjusted p value	0.846		0.846		0.66		0.789
Spondylolisthesis
Other diagnosis	379 (87.7%)	53 (12.3%)	272 (63%)	160 (37%)	257 (59.5%)	175 (40.5%)	388 (89.8%)	44 (10.2%)
Yes	36 (85.7%)	6 (14.3%)	30 (71.4%)	12 (28.6%)	26 (61.9%)	16 (38.1%)	34 (81%)	8 (19%)
P value	0.705		0.276		0.761		0.079
Adjusted p value	0.846		0.789		0.846		0.66
Canal stenosis
Other diagnosis	369 (88.1%)	50 (11.9%)	268 (64%)	151 (36%)	247 (58.9%)	172 (41.1%)	373 (89%)	46 (11%)
Yes	46 (83.6%)	9 (16.4%)	34 (61.8%)	21 (38.2%)	36 (65.5%)	19 (34.5%)	49 (89.1%)	6 (10.9%)
P value	0.349		0.756		0.355		0.988
Adjusted p value	0.789		0.846		0.789		0.988
Disc herniation
Other diagnosis	202 (86.7%)	31 (13.3%)	152 (65.2%)	81 (34.8%)	141 (60.5%)	92 (39.5%)	204 (87.6%)	29 (12.4%)
Yes	213 (88.4%)	28 (11.6%)	150 (62.2%)	91 (37.8%)	142 (58.9%)	99 (41.1%)	218 (90.5%)	23 (9.5%)
P value	0.578		0.498		0.724		0.312
Adjusted p value	0.846		0.846		0.846		0.789

DISCUSSION

The present study demonstrates the impact of stiffness secondary to lumbar spine arthrodesis on the limitation of daily activities in 40 patients with a minimum postoperative follow-up of two years. The treatment of various pathological conditions of the spine through solid intervertebral arthrodesis is widely supported in the literature,^3,4,10,11 despite the knowledge that the loss of segmental mobility associated with arthrodesis can impair, at least to some degree, the functional capacity for different activities.^12,13 However, a relationship is not yet fully established between the number of levels included in the arthrodesis and either the degree of rigidity or the degree of limitation of functional capacity, especially when the inclusion or not of the sacrum in the arthrodesed levels are compared.

Extension of the arthrodesis to the sacrum has always been viewed as a challenge, both because of considerable levels of fusion failure when compared to arthrodesis without extension to the sacrum and due to a fear of limitation of functional capacity resulting from stiffness in the region of the lumbosacral transition.^14–16 In the present study, the patients with lumbar arthrodesis without extension to the sacrum had significantly better LSDI scores than the patients with the sacrum included in the arthrodesis, i.e., the addition of the sacrum was associated with greater functional limitation related to stiffness.

In a literature review article, Bridwell et al. observed that the extension of the arthrodesis to the sacrum, in addition to increasing the risk of pseudarthrosis, compromised mobility in the lumbosacral junction, which can change the mechanics of gait due to rigidity of the sacroiliac joints.¹⁵ On the other hand, Edwards e t al. observed that the extension of arthrodesis to the sacrum in long fusions did not alter the functional outcome as evaluated by the Scoliosis Research Society-24 questionnaire, as compared to patients with arthrodesis extending to L5.¹⁷

The LSDI questionnaire was developed to assess the limitation of daily activities specifically related to lumbar spinal stiffness after arthrodesis surgery, in order to facilitate understanding of the impact of arthrodesis.⁷ It is an easy to apply and easy to understand tool that has proven to be valid for quantifying functional capacity limitations in these patients. Recently, the LSDI questionnaire was translated into Portuguese and adapted for use in the Brazilian population.⁸

In the present study, it was observed that less lumbar mobility, identified by lower MMST values, was indicative of worse functional capacity in the patients, as represented by higher LSDI scores. This is the first study to show a significant, albeit moderate, correlation between LSDI scores and clinical stiffness identified by the MMST value. Other studies have demonstrated a correlation between the LSDI score and lumbar stiffness evaluated by the range of motion in dynamic lateral radiographs (flexion and extension) of the lumbar spine.⁷ However, this examination cannot be considered the gold standard for evaluating lumbar mobility because of technical limitations, such as being dependent on the way it is executed (operator-dependent),¹⁸ in addition to exposure to radiation and the costs involved in performing it.

In terms of the number of lumbar levels included in the arthrodesis, the data from this series showed that the functional capacity as measured by the LSDI had no relationship with the number of levels, i.e., worsening of functional capacity was not proportional to the number of arthrodesed levels. This finding is in line with other published studies. Gotfryd et al. observed no difference between the quality of life indicators of patients who underwent single-level arthrodesis and those with arthrodesis in two or more levels.¹⁹ Hart et al. observed that patients submitted to pan-lumbar arthrodesis, involving all lumbar levels, did not present a worsening of functional capacity related to lumbar stiffness during a minimum of two years of follow-up.²⁰

Limitations can be identified in the present study. As it is a retrospective analysis, the preoperative functional capacity data could not be evaluated. Thus, the statement that lumbar arthrodesis causes functional limitations cannot be made. Although the study included 40 patients, considered a robust number compared to other published series,⁷ this number would limit more complex analyses considering specific subgroups.

CONCLUSIONS

Functional capacity related to lumbar stiffness, measured by the LSDI score in patients who underwent spinal arthrodesis, was shown to be causally related to clinical lumbar stiffness, measured by the modified-modified Schöber test. Inclusion of the sacrum in the arthrodesis was associated with greater impairment of functional capacity related to lumbar stiffness than arthrodesis without extension to the sacrum.

• Study conducted at the Hospital Israelita Albert Einstein, São Paulo, SP, Brazil.
• Reviewed by: Carlos A. Tello

• ERRATUM

  When one reads the article CORRELATION BETWEEN DEGENERATIVE DISEASES OF THE LUMBAR SPINE AND TYPES OF LUMBAR LORDOSIS, DOI: http://dx.doi.org/10.1590/S1808-185120212001235427, published in Revista Coluna/Columna V.20 N.1/2021, from the introduction to the conclusion, the figures and references on pages 26-29 should be read from pages 30-33 of the article CORRELATION BETWEEN DEGENERATIVE DISEASES OF THE LUMBAR SPINE AND TYPES OF LUMBAR LORDOSIS, DOI: http://dx.doi.org/10.1590/S1808 185120212001235427, in the errata of Revista Coluna/Columna V.20 N.4/2021.
  INTRODUCTION

  Low back pain is a significant social and economic problem that leads to the loss of billions of dollars a year worldwide.¹,² The etiology is multifactorial, but degenerative changes in the lumbar spine are closely associated with this problem³ and are frequent causes of a reduction in the quality of life in the active population and especially among the elderly.⁴ The most common degenerative lumbar spine conditions involve the degeneration of the intervertebral disc, facet joints, capsule, and vertebral ligaments, which leads to diseases such as disc herniation, spondylolisthesis and canal stenosis.⁵ Although degenerative conditions are part of the natural progression of aging, it is suspected that in the spine these are related to the load that the vertebrae bear over time. Load distribution in the lumbar region would be directly linked to the anatomy and design of the physiological curves of the spine (lordosis and kyphosis), as well as to the positioning of the pelvis in relation to the vertebral axis.⁶

  In 2005, Roussouly et al.,⁷ created a classification that addresses the normal variation in the sagittal alignment of the human lumbar spine and pelvis in the orthostatic position in order to quantify and classify common variations in the sagittal alignment of the spine, the sacrum, and the pelvis.

  While developing their classification, Roussouly et al.,⁷ observed that the types of lordosis could be related to some of the most common degenerative lumbar spine diseases, suggesting that patients with symptomatic disc herniation fit into types 1 and 2 while stenoses were most often seen in cases classified as type 4. Patients classified as type 3 rarely had significant complaints. However, there was no evidence or statistical analysis of this observation. Given this gap in the literature, the objective of this study was to evaluate the relationship between the incidence of the different types of degenerative spine disease and lumbopelvic biomechanics, according to the types of lordosis as classified by Roussouly and their correlation with the treatment performed in these patients.

  METHODS

  This study was approved by the Institutional Review Board. A retrospective search was conducted of the medical records of patients treated at a private hospital in the city of São Paulo, during the period from 2012 to 2017, who were diagnosed with degenerative lumbar spine disease and had previously received a surgical indication for this reason, but who did not necessarily undergo surgery. The diagnostic and treatment information of these patients was reviewed and the imaging examinations (radiographs and magnetic resonance imaging of the lumbosacral spine) were analyzed to confirm the lumbar lordosis diagnosis and classification. Extraction of patient imaging examinations was performed from the PACS Platform (Carestream Health, Rochester, New York, USA) at the hospital. The radiographs were imported to Surgimap^® software (version 2.2.15.1) (Nemaris Inc.™, Audubon, Pennsylvania, USA) for verification of the angles and classification of the lumbar curvature. They were assigned to one of Roussouly's four lordosis curve types according to the radiographic analysis of the lumbar spine. These steps will be described in detail later. Magnetic resonance images of the lumbar spine, together with the medical history on record, were used to define the patient's diagnosis of degeneration. In the presence of two concomitant diseases observed in the magnetic resonance images, the diagnosis of greater clinical severity, which in these patients was the cause of seeking treatment was considered. As such, diagnoses of degenerative discopathy, lumbar disc herniation, spinal canal stenosis, degenerative spondylolisthesis, and facet arthropathy were considered. These data were cross-referenced to correlate the pattern of the curve with the type of lumbar degeneration.

  Patients between the ages of 18 and 75 with a diagnosis of degenerative lumbar spine disease who had radiographic and magnetic resonance examinations and complete medical records were included. Patients with prior spine surgery, pediatric spinal deformity, a history of infection or active infection, oncologic diseases or spinal fracture were excluded from the study. In order to divide the groups by the types of lordosis according to the classification of Roussouly et al.,⁷ four types of lordosis were defined below and shown in Figure 1. In type 1 the inflection point (the point where there is a change in the orientation of the vertebral bodies) is L3/L4, sacral inclination is less than 35°, the pelvic incidence is small, and long kyphotic and short lordotic curves are present in an 80:20 ratio of the length of the thoracolumbar spine. In type 2, which has more vertebral bodies, the inflection point is above level L1/L2, sacral inclination is less than 35°, pelvic incidence is small, short kyphotic and long lordotic curves are present. They are in a proportion of 60:40 of the total length of the thoracolumbar spine. In type 3, the inflection point is in T12/L4, the sacral inclination is between 35° and 45°, pelvic incidence is high, and the kyphotic and lordotic curves are almost equal in a ratio of 50:50 of the total length of the thoracolumbar spine, and the spine is balanced. In type 4, the inflection point is in T9/T10, the sacral inclination is greater than 45°, pelvic incidence is high, and the lordotic curve is longer than the kyphotic curve in an inverse ratio of 20:80 of the total length of the thoracolumbar spine.

  

  Figure 1
  Types of lumbar lordosis, according to Roussouly.
  

  Magnetic resonances of the lumbar spine were used to define the patients’ diagnoses. Patients were classified as having disc herniation/degenerative discopathy, spondylolisthesis, spinal canal stenosis, or facet arthrosis. The disc degeneration diagnosis was considered in patients with any degree of degenerative disc changes in the magnetic resonance, without other major changes, complaining of axial pain, especially with trunk flexion. Disc herniation was considered in patients who presented this condition in the magnetic resonance examination, with lumbosciatalgia, paresthesia and/or the loss of strength in the lower limbs. Spinal canal stenosis was considered when viewed in the examination and presenting with neurological claudication. Degenerative spondylolisthesis was considered in patients with vertebral slippage of any degree in the examination, with possible symptoms of axial or root pain. Facet arthropathy was considered in those patients with joint changes without any other findings in the magnetic resonance and with complaints of axial pain.

  Two observers conducted the analysis of the radiographic and magnetic resonance images of the patients included in the study to define the diagnosis and classify the type of lordosis. A reliability analysis was conducted between the observers resulting in concordance greater than 90%, which was considered acceptable.

  General data, such as age, sex, and treatment received were collected from the medical records and analyzed.

  For the statistical data analysis, the quantitative variables were described as mean, standard deviation, minimum and maximum values and the qualitative variables as absolute and relative frequencies.

  Comparisons between the Roussouly classifications⁷ by sex, diagnosis, and type of treatment were verified via the chi-squared test⁸ and multinomial logistic regression.⁹

  The analyses were conducted using the Statistical Package for the Social Sciences – SPSS, v26.0¹⁰ software (IBM – Armonk – New York – USA) and the level of significance considered was 5%.

  RESULTS

  The sample consisted of 418 patients with radiographs and magnetic resonance images of the lumbar spine, 203 of whom were women and 215 of whom were men, the equivalent of 48.6% and 51.4%, respectively.

  As regards the Roussouly classification⁷, 47 (11.2%) patients were classified as type 1 lordosis, 159 (38%) as type 2, 168 (40.2%) as type 3, and 44 (10.5%) as type 4.

  The type of treatment performed was proportional, with 50% (209) of the sample undergoing surgical treatment and 50% (209) conservative treatment. Diagnoses of the type of degenerative lumbar spine disease were distributed as follows: 23 (5.5%) patients with facet arthrosis, 92 (22.7%) with degenerative discopathy, 31 (7.4%) with spondylolisthesis, 41 (9.8%) patients with spinal canal stenosis, 219 (52.4%) disc herniation, and 9 (2.2%) patients with no changes in the imaging examinations. Patient characteristics are shown in Table 1.

  Table 1
  Characterization of the sample.
  

  Characteristics of the sample (n = 418)		n	%
  Sex
  	Female	203	48.60%
  	Male	215	51.40%
  Roussouly Classification
  	Type 1	47	11.20%
  	Type 2	159	38.00%
  	Type 3	168	40.20%
  	Type 4	44	10.50%
  Diagnosis
  	Facet arthrosis	23	5.50%
  	Degenerative discopathy	95	22.70%
  	Spondylolisthesis	31	7.40%
  	Canal stenosis	41	9.80%
  	Disc herniation	219	52.40%
  	Normal	9	2.20%
  Type of treatment performed
  	Surgical	209	50.00%
  	Conservative	209	50.00%

  • IQR = Interquartile Range

  One of the study objectives was to check possible associations between the type of lordosis, according to the Roussouly classification,⁷ and the sex of the patients, the diagnosis, and the type of treatment performed.

  We found no evidence of significant association with sex (p value = 0.632). As for the type of treatment performed, patients with type 1 and type 2 lordosis had a higher predominance of surgical treatment (63.8% vs. 36.2%) and type 3 and type 4 lordosis had a higher predominance of conservative treatment (59.1% vs. 40.9%), as observed in Figure 2, representing a significant difference (p value = 0.008). For the purpose of comparison, we grouped diagnoses of degenerative discopathy and disc herniation together, since they are both considered intervertebral disc diseases, and we disregarded the nine cases of patients with normal examinations. Additionally, due to the low contingency table frequencies, we opted for the likelihood ratio test obtained through the multinomial regression model. However, we found no evidence of significant associations (p value = 0.246). The results are shown in Table 2.

  

  Figure 2
  Types of Treatment x Roussouly Classification.
  

  Table 2
  Comparisons by Roussouly classification
  

  Factors	Roussouly Classification				p value
  	Type 1	Type 2	Type 3	Type 4
  Type of treatment (n=418)					0.008a
  Surgical	30 (63.8%)	90 (56.6%)	71 (42.3%)	18 (40.9%)
  Conservative	17 (36.2%)	69 (43.4%)	97 (57.7%)	26 (59.1%)
  Sex (n=418)					0.632a
  Female	19 (40.4%)	77 (48.4%)	86 (51.2%)	21 (47.7%)
  Male	28 (59.6%)	82 (51.6%)	82 (48.8%)	23 (52.3%)
  Diagnosis (n=409)					0.246 b
  Facet arthrosis	2 (4.3%)	11 (7%)	7 (4.3%)	3 (7.3%)
  Degenerative discopathy/disc herniation	35 (74.5%)	120 (75.9%)	131 (80.4%)	28 (68.3%)
  Spondylolisthesis	3 (6.4%)	9 (5.7%)	11 (6.7%)	8 (19.5%)
  Canal stenosis	7 (14.9%)	18 (11.4%)	14 (8.6%)	2 (4.9%)

  • ^a
    Chi-squared test;
  • ^b
    Likelihood ratio test.

  We also compared each of the diagnoses with the Roussouly classifications⁷ individually. The comparisons were verified using the chi-squared test and we used the Benjamini-Hochberg correction to control type 1 errors, but none of the comparisons were significant (p value > 0.05). The results are presented in Table 3.

  Table 3
  Comparison of diagnoses by Roussouly classification
  

  Diagnosis	Roussouly Type 1		Roussouly Type 2		Roussouly Type 3		Roussouly Type 4
  	No	Yes	No	Yes	No	Yes	No	Yes
  Facet arthrosis
  Other diagnosis	380 (87.8%)	53 (12.2%)	276 (63.7%)	157 (36.3%)	254 (58.7%)	179 (41.3%)	389 (89.8%)	44 (10.2%)
  Yes	35 (85.4%)	6 (14.6%)	26 (63.4%)	15 (36.6%)	29 (70.7%)	12 (29.3%)	33 (80.5%)	8 (19.5%)
  P value	0.657		0.967		0.132		0.067
  Adjusted p value	0.846		0.988		0.66		0.66
  Degenerative discopathy
  Other diagnosis	330 (87.1%)	49 (12.9%)	240 (63.3%)	139 (36.7%)	233 (61.5%)	146 (38.5%)	334 (88.1%)	45 (11.9%)
  Yes	85 (89.5%)	10 (10.5%)	62 (65.3%)	33 (34.7%)	50 (52.6%)	45 (47.4%)	88 (92.6%)	7 (7.4%)
  P value	0.526		0.725		0.116		0.209
  Adjusted p value	0.846		0.846		0.66		0.789
  Spondylolisthesis
  Other diagnosis	379 (87.7%)	53 (12.3%)	272 (63%)	160 (37%)	257 (59.5%)	175 (40.5%)	388 (89.8%)	44 (10.2%)
  Yes	36 (85.7%)	6 (14.3%)	30 (71.4%)	12 (28.6%)	26 (61.9%)	16 (38.1%)	34 (81%)	8 (19%)
  P value	0.705		0.276		0.761		0.079
  Adjusted p value	0.846		0.789		0.846		0.66
  Canal stenosis
  Other diagnosis	369 (88.1%)	50 (11.9%)	268 (64%)	151 (36%)	247 (58.9%)	172 (41.1%)	373 (89%)	46 (11%)
  Yes	46 (83.6%)	9 (16.4%)	34 (61.8%)	21 (38.2%)	36 (65.5%)	19 (34.5%)	49 (89.1%)	6 (10.9%)
  P value	0.349		0.756		0.355		0.988
  Adjusted p value	0.789		0.846		0.789		0.988
  Disc herniation
  Other diagnosis	202 (86.7%)	31 (13.3%)	152 (65.2%)	81 (34.8%)	141 (60.5%)	92 (39.5%)	204 (87.6%)	29 (12.4%)
  Yes	213 (88.4%)	28 (11.6%)	150 (62.2%)	91 (37.8%)	142 (58.9%)	99 (41.1%)	218 (90.5%)	23 (9.5%)
  P value	0.578		0.498		0.724		0.312
  Adjusted p value	0.846		0.846		0.846		0.789

  Finally, we compared the patients’ type of treatment and diagnosis and found evidence of significant association (p value < 0.001). For patients with facet arthrosis and degenerative discopathy, conservative treatment was the most prevalent at 73.9% and 94.7%, respectively. In patients diagnosed with spondylolisthesis, canal stenosis, and disc herniations, surgical treatment was more prevalent, at 64.5%, 58.5%, and 70.3%, respectively. The results are presented in Table 4.

  Table 4
  Type of treatment by diagnosis.
  

  Diagnosis (n = 418)	Type of Treatment		p value
  	Surgical	Conservative
  Facet arthrosis	6 (26.1%)	17 (73.9%)	< 0.001a
  Degenerative discopathy	5 (5.3%)	90 (94.7%)
  Spondylolisthesis	20 (64.5%)	11 (35.5%)
  Canal stenosis	24 (58.5%)	17 (41.5%)
  Disc herniation	154 (70.3%)	65 (29.7%)
  Normal	0 (0%)	9 (100%)

  • ^a
    Chi-squared test
  DISCUSSION

  Around 50-70% of the population will experience low back pain symptoms for various reasons at least once in their life.¹¹

  One of the factors that leads to low back pain is degenerative changes that are more common with the increasing life expectancy of the population. In 2005, Roussouly et al.⁷ proposed a system to classify types of lumbar lordosis and demonstrated that most asymptomatic individuals were classified as type 3, as was observed in symptomatic individuals in our study. They hypothesized that different types of lordosis could be related to certain pathologies, for example, that patients with type 1 and 2 were liable to present disc herniation. In the present study, no statistically significant result was obtained to confirm this hypothesis.

  In 2017, Roussouly¹² conducted a new study of the types of lumbar lordosis, but for patients with degenerative changes, in addition to the 4 already established types, he included type 3 anteverted and type 4 anteverted, which present the same characteristics as the original types, but with pelvic inclination < 5 degrees; types 1, 2, 3, and 4 retroverted, which present the same characteristics as the originals but with pelvic inclination > 25 degrees; and lumbar and overall kyphosis. This new classification was not used in the present study since the main objective was to use the classic Roussouly classification.⁷

  In absolute numbers, we observed a higher incidence of individuals with intervertebral disc disease in all classifications.

  Some studies, like that of Mardare et al.,¹³ demonstrated that there is a relationship between sagittal balance and the different pathologies, as in patients with low sacral inclination and increased pelvic incidence and inclination values who tend towards greater disc degeneration. These patients normally have reduced lumbar lordosis leading to flat back syndrome, which we can assume causes increased pressure on the anterior spine, i.e., on the intervertebral discs, for their entire life, leading to a mechanism of constant overload and early degeneration.

  Regarding the type of treatment in these individuals, we confirmed a statistically significant result in which individuals classified as type 1 and type 2 had a propensity for surgical treatment and type 3 and type 4 for conservative treatment. A comparison of types of treatment and diagnoses yielded statistical significance. Most patients with canal stenosis, spondylolisthesis, and disc herniation underwent surgical treatment, while most patients with facet arthrosis and degenerative discopathy received conservative treatment, in agreement with Lindsey T,¹⁴ who in 2020 demonstrated that conservative treatment of facet arthrosis and degenerative discopathy should be the initial treatment for patients with low back pain.

  The retrospective design of the study itself is one of its limitations. There are others, such as the distribution of the patients into groups where there was a much higher number of individuals with disc herniation than those with other diagnoses. Also, patients were included in the study who had previously been indicated for surgery at another institution, creating a much higher possibility of a real surgical outcome.

  Understanding the etiology of lumbar spine degeneration and diseases is of utmost importance in today's world, as these diagnoses contribute to high healthcare costs and a decrease in the productivity of the population. Given the study limitations presented, it was not possible to confirm a relationship between patient diagnosis and lumbar lordosis type. A study with a greater number of individuals is essential such that, if there were a statistically significant difference in the relationship between the diagnosis and the lumbar curvature, specific preventative methods could be established for each type of population to prevent spine diseases.

  CONCLUSIONS

  We can conclude that the patients classified as Roussouly type 1 and type 2 underwent surgical treatment in higher numbers than type 3 and type 4 patients. We did not observe any statistical correlation between the type of lumbar lordosis and the type of diagnosis presented.

  REFERENCES
• ¹
  Katz JN. Lumbar disc disorders and low-back pain: socioeconomic factors and consequences. J Bone Joint Surg Am. 2006;88(Suppl 2):21-4. doi: 10.2106/JBJS.E.01273.
  » https://doi.org/10.2106/JBJS.E.01273
• ²
  Urban JP, Robertos S. Degeneration of the intervertebral disc. Arthritis Res Ther. 2003;5(3):120-30. doi: 10.1186/ar629.
  » https://doi.org/10.1186/ar629
• ³
  Hoy D, Brooks P, Blyth F, Buchbinder R. The Epidemiology of low back pain. Best Pract Res Clin Rheumatol. 2010;24(6):769-81. doi: 10.1016/j.berh.2010.10.002.
  » https://doi.org/10.1016/j.berh.2010.10.002
• ⁴
  Orita S, Inage K, Eguchi Y, Kubota G, Aoki Y, Nakamura J, et al. Lumbar foraminal stenosis, the hidden stenosis including at L5/S1. Eur J Orthop Surg Traumatol. 2016;26(7):685-93. doi: 10.1007/s00590-016-1806-7.
  » https://doi.org/10.1007/s00590-016-1806-7.
• ⁵
  Sąsiadek MJ, Bladowska J. Imaging of Degenerative Spine Disease – the State of the Art. Adv Clin Exp Med. 2012;21(2):133-42.
• ⁶
  Keorochana G, Taghavi CE, Lee KB, Yoo JH, Liao JC, Fei Z, et al. Effect of sagittal alignment on kinematic changes and degree of disc degeneration in the lumbar spine: An analysis using positional MRI. Spine (Phila Pa 1976). 2011;36(11):893-8. doi:10.1097/BRS.0b013e3181f4d212.
  » https://doi.org/10.1097/BRS.0b013e3181f4d212
• ⁷
  Roussouly P, Gollogly S, Berthonnaud E, Dimnett J. Classification of the normal variation in the sagittal alignment of the human lumbar spine and pelvis in the standing position. Spine (Phila Pa 1976). 2005;30(3):346-53. doi: 10.1097/01.brs.0000152379.54463.65.
  » https://doi.org/10.1097/01.brs.0000152379.54463.65
• ⁸
  Altman DG. Practical statistics for medical research. London: Chapman and Hall/CRC press; 1991.
• ⁹
  Hosmer DW, Lemeshow S. Applied Logistic Regression. Nova Jersey: Wiley; 2004.
• ¹⁰
  IBM Corp. Released 2019. IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp; 2019.
• ¹¹
  Hoy D, Bain C, Williams G, March L, Brooks P, Blyth F, et al. A systematic review of the global prevalence of low back pain. Arthritis Rheum. 2012;64(6):2028-37. doi: 10.1002/art.34347.
  » https://doi.org/10.1002/art.34347
• ¹²
  Sebaaly A, Grobost P, Mallam L, Roussouly P. Description of the sagittal aligment of the degenerative human spine. Eur Spine J. 2018;27(2):489-96. doi: 10.1007/s00586-017-5404-0.
  » https://doi.org/10.1007/s00586-017-5404-0
• ¹³
  Mardare M, Oprea M, Popa I, Zazgyva A, Niculescu M, Poenaru DV. Sagittal balance parameters correlate with spinal conformattional type and MRI changes in lumbar degenerative dis disease: results of a retrospective study. Eur J Orthop Surg Traumatol. 2016;26(7):735-43. doi: 10.1007/s00590-016-1842-3.
  » https://doi.org/10.1007/s00590-016-1842-3
• ¹⁴
  Lindsey T, Dydyk AM. Spinal Osteoarthritis. StatPearls. Treasure Island (FL): StatPearls Publishing; 2020.