6.1 Introduction

As discussed in previous chapters, human bipedalism is an exclusive, vertical, stable, and ergonomic posture. This upright position adoption has resulted in the pelvis being more vertical and, in the same way, in the appearance of characteristic spinal curves. ¹ These spinal curves are unique to Homo sapiens and resulted in a better energy-sparing morphology for the transition to the above-described upright posture. The interest in spinal shapes and segmentation started since ancient Greece with Hippocrates of Kos (460–370 BC). Hippocrates, regarded as the father of medicine, was the first to describe the normal spinal curvatures with their anatomical limits and orientation. Hippocrates also realized, by observation of cadavers on the battlefield, that the spine was held together by means of intervertebral disks, ligaments, and muscles. ² Therefore, since Hippocrates, then Galen, the spinal curvatures are classically segmented in sacral kyphosis, lumbar lordosis (LL), thoracic kyphosis (TK), and cervical kyphosis (Fig. 6‑1).

Despite this ideal segmentation in anatomical well-established limits, recently, some authors tend to use a functional segmentation dictated by the intervertebral orientation in the sagittal plane (Fig. 6‑2). This chapter provides an overview of the spinal curve segmentation leading to LL classification. A modern knowledge of the principles of sagittal spine segmentation is pivotal to understand spinal pathology and to implement a surgical strategy. This understanding is intended to provide optimum outcomes when treating spinal disorders. Hippocrates had recognized, a long time ago, the importance of the spine as stated in his famous manuscript On Joints: “One should first get a knowledge on the structure of the spine; this is also a requisite for many diseases.” ³

6.2 Classical Spinal Segmentation: Kyphosis and Lordosis

The term “lordosis” comes from the Greek word lordos (λόρδος) meaning “bent forward,” while “kyphosis” originates from Greek kyphos (κυφός) meaning “hump.” It seems that “lordos” was used for normal curvature, whereas “kyphos” was used for pathological curves. In his book On Articulations and Mochlikon, Hippocrates described the classical anatomical teaching on the spinal curvatures with consecutive lordosis and kyphosis going from the cervical spine to the sacrum. He described three categories of vertebrae: the cervical vertebrae located above the clavicle and going from C2 to the “great vertebra” (C7 or T1), the thoracic vertebrae, and the lumbar vertebra. The normal alignment of the lumbar vertebrae was termed “ithiscolios,” that means that the spine is straight in the coronal but curved in the sagittal plane. ²

This succession of curvature is of primary importance for an economical bipedal gait of H. sapiens. In fact, dorsal kyphosis is the only sagittal curvature present at birth (Fig. 6‑3). ⁴ The development of CL is caused by the lifting of the head and horizontal gaze allowing economical load transmission from the skull to the pelvis. In addition, the human cervical spine is unique by being perpendicular to the cranial base and the foramen magnum being more anteriorly (or less posteriorly) placed as compared to other mammals. ⁵ On the other hand, the lumbar curve acquires its lordosis with the acquisition of standing and walking (Fig. 6‑4). It is characterized, in humans, by being long and mobile allowing a fixed position of the upper spine and the cranium (fixed horizontal gaze) when the pelvis rotates during walking. As mentioned above, sagittal spinal curves increase resistance to vertical loads by directing deformations into preordered directions, which can be quickly controlled by the fast intervention of muscle contraction. ⁴ Compared to the quadrupeds in which the sacrum is positioned forward, the femoral heads, the vertical erected human has the sacrum backward, the femoral heads with a retroverted pelvis (Fig. 6‑5). This position combined with the LL allows the balance of the body weight above the pelvis generating an economical bipedal gait. ⁶

The SRS-Schwab classification (and subsequent North American publications) define LL as the angle between the upper endplate of S1 and the upper endplate of L1. ⁷ TK was defined as the angle between the lower endplate of T12 and the upper endplate of T1 (Fig. 6‑6). Because T1 vertebra is frequently overshadowed by the humeral head and rib cage on the lateral films, and the reliability of measuring T1-T5 kyphosis is low, many authors recommend the measurements of the T4-T12 angle. ⁸ This segmentation of the thoracic and lumbar spine is anatomical and divides the spine into 12 thoracic and 5 lumbar vertebrae. The main problem with the anatomical segmentation is that a fixed limit (T12 or L1) does not take into account the kyphotic and lordotic curvatures with respective lengths and magnitudes. Therefore, there is the importance of the application of the functional segmentation to the spine (Fig. 6‑6).

6.3 Functional Segmentation of the Spine

There is no unanimity for the exact angular definition of LL. Galen of Pergamon (130–210 AD) described the anatomy of the spine stating that Nature formed the structure of the spine: “Nature creates nothing without a purpose.” He insisted that vertebral body shapes provide the harmony in spinal motion. ² Nineteen centuries later, the relationship between pelvic and spinal parameters was initially approached by Delmas in 1953 (Fig. 6‑2). He described an array of variation of normal spine ranging from “static” curved back to “dynamic” flat back, with high and low sacral slope (SS), respectively. To better understand the problematic definition of LL, it is necessary to review the different possibilities for measuring this important parameter.

Based on the studies of Delmas, and later, Stagnara et al, ⁹ Dimnet and Berthonnaud ¹⁰ described the “inflection point” in 2005. In fact, the spine represents a dynamic chain where the curves change direction at a specific point: the inflection point (Fig. 6‑7). In this articulating chain, each anatomical segment orientation and shape relates to and influences the adjacent segment in maintaining an upward posture with a minimum of energy expenditure. ¹⁰ The change of the orientation or the shape of one segment will induce a change or orientation of another segment, with the “inflection point” being the fulcrum of this change. Moreover, at the inflection point, the thoracic and lumbar curves are both tangents to a line of its limited vertebrae.

In the functional segmentation of the spine, the limits of LL and TK are based on the individual sagittal shape of the spine. In other words, the upper limit of LL would not be fixed at L1, and LL would be measured from the upper endplate of the sacrum to the upper limit vertebra in a similar way to the coronal Cobb angle of the scoliosis (Fig. 6‑6). ¹¹ The LL spans the vertebrae from the upper sacral endplate to the inflection point where TK begins. Likewise, the TK spans the vertebrae from the inflection point with the LL to the inflection point marking the limit with CL. The conventional method of delineating TK from LL based on anatomical landmarks such as the thoracolumbar junction is overly simplistic. The advantage of this new method is that the segments are defined by the change in spatial relationships between the vertebrae at the inflection point. The present method gives a more accurate definition of each sagittal curve and the number of vertebral bodies in kyphosis and lordosis.

The analysis of the spinal curves comprises other concepts. The angle arc of a curvature of the spine (i.e., LL) may be divided into two arches limited by the apex of the curve, which determines a horizontal line (Fig. 6‑7). Berthonnaud et al ¹⁰ defined two tangent arcs of a circle, delimited by the apex of the curve with each arch being tangential to the vertical axis at the apex of the curve. The radius and length of each arc are independent and are determined by two critical points: the apex of the curve and the upper inflection point (Fig. 6‑7). Thus, the length of the upper arc of LL is localized between the apex of the LL and the inflection point. Likewise, the length of the lower arc of the LL curve is located between the apex of the LL and the sacral endplate. In the same manner, the TK is divided into two arcs: the upper arc of TK and the lower arc of TK. The upper arc of TK is localized between the apex of the TK and the inflection point of the CL. The length of the lower arc of the TK curve is located between the apex of the TK and the inflection point between LL and TK.

Several mathematical characteristics rule these different angles. The lower arch of the LL is geometrically equal to the SS as they are corresponding angles. The upper arch of LL and the lower arch of the TK are also equal for the same reason. This could explain that an increase in LL would be followed by an increase of TK, maintaining the “spinal harmony.” Even more, in a study on a cohort of asymptomatic individuals, Roussouly et al ¹² found the upper arch of LL to be constant and to have a value of ~ 21.5°. As the lower arch of LL and SS are equal, LL could be predicted by the formula LL = SS + lower arc. Then by replacing each arch by its corresponding value, we have the relation LL = SS + 21°. This feature confirms the strong correlation found between SS and LL (R = 0.86) and indicates that the lordosis is directly dependent on SS.