Portal de Periódicos da CAPES

Reflections of an Orthopaedic Surgeon on Patient Care and Research Into the Condition of Scoliosis

2011; Lippincott Williams & Wilkins; Volume: 31; Linguagem: Inglês

10.1097/bpo.0b013e3181f73beb

1539-2570

Jean Dubousset,

Scoliosis diagnosis and treatment

This paper discusses the lessons I have learned in a lifetime of practicing medicine and my thoughts on research into the condition of scoliosis. As a 3-dimensional (3D) spinal deformity, adolescent idiopathic scoliosis (AIS) destroys the harmony, mobility, and function of the spinal column that begins at the head (cephalic vertebrae) and ends at the pelvis (pelvic vertebrae) and is responsible for the balance of the human body (Fig. 1). Despite decades of investigation the etiology is still unknown. Nevertheless, I believe AIS is genetically programmed with neurocentral and/or neurohormonal dysfunction producing the deformity. In 2 earlier studies, my collaborators and I found abnormal melatonin signaling that we postulate is responsible for impaired transmission of postural control signals between the left and right brain resulting in scoliosis in chickens.1Another study showed inherited anomalies of the vestibular system at the level of the semicircular ducts in patients with scoliosis.2FIGURE 1: The chain of balance.I also believe there are 2 major kinds of idiopathic scoliosis, "ascending" or developmental scoliosis and "descending" or degenerative scoliosis. Ascending scoliosis occurs during a child's growth period. It is linked to the individual's genetic code and simultaneously affects development of the vertebrae, lungs, thorax, and the central nervous system. Understanding hormonal and radiographic growth maturation assists in planning the proper treatment at the proper time for ascending scoliosis. Descending scoliosis develops in either a scoliotic or a normal spine in late adulthood and is precipitated by an age-related loss of disc and vertebral height. For this paper, I will concentrate on the ascending developmental scoliosis, keeping in mind that the ultimate goal of treatment is to also reduce the adult sequelae of descending degenerative scoliosis. Just as the spinal column provides for the harmonious function of the trunk, the orthopaedic surgeon must provide harmonious care to the patient with scoliosis. In this paper, I will describe my recommendations for giving such care to patients with scoliosis and their families, and I will offer my thoughts on the 3D deformity called scoliosis. INITIAL PATIENT ENCOUNTER The first contact with the patient (regardless of age) and the family can be summarized in a few words: listening, kindness, sincerity, and clarity. The key is mutual confidence and transparency. I recommend clarity in the diagnosis, explanations, and proposals for treatment. Avoid prematurely asking the patient or family to choose among the various possibilities. However, do not be elusive when asked questions or hesitate to give your advice. As a surgeon and physician, you must be convinced of the treatment course you recommend. Ambivalence in the surgeon's mind provokes doubt in the mind of the patient and family, potentially leading to compromised results. As I believe in the techniques of Stagnara and Salanova in France and Nachemson in Sweden, I always first consider casting or bracing for young patients with progressive deformities. The advantages of not doing early surgery are usually intuitive to the child and the parents, but all parties must also understand that some patients still require surgery despite having a rigorous nonoperative program. In addition, if surgical treatment is proposed, the complications should be openly discussed along with techniques to prevent them. If complications do occur, they are much better understood. A second lesson is to carry out a complete history and clinical examination before rushing for radiographs. Patients and families appreciate the personal attention displayed by a physician, who first learns about the individual before resorting to technological tools. Determine why the patient came to the clinic, assess for respiratory issues, pain, and neurological complaints. Try to understand the psychological impact of the deformity for both the patient and the parents. Perform a complete physical examination including a careful neurological assessment. Determine the cosmetic deformity, maturation signs, spinal motion, stiffness, and balance. Always remember scoliosis may exist with neurological conditions such as the Chiari malformation, spinal cord syrinx, or tethering. Magnetic resonance imaging should be performed if the patient has asymmetry of the abdominal reflexes, a left thoracic curve, significant pain, or rapid curve progression. In children less than 10 years of age, treatment of an existing neurological abnormality has led to the resolution of the scoliosis in many. Lastly, an examination for scoliosis should be carried out on any accompanying related family members. NATURAL HISTORY AND PROGNOSTIC LESSONS During childhood, AIS is generally pain free. When a young patient presents with pain and stiffness in the spinal column the physician must consider other etiologies, including tumors. However, in cases with significant 3D imbalance the patient may have muscular pain in the back. Neck pain is also possible in cases of significant thoracic lordosis. Cosmesis is a major reason for surgery for both female and male patients. The emotional impact of the deformity is substantial in children, parents (often with significant feelings of guilt), and surprisingly even in adults. Many patients with AIS function well, even with a significant curve. I have seen patients postpone surgical treatment until after the age of 45 to 50 years to be "free" during the adult active life to participate in sports, jobs, and active family life. It is worth remembering that a spinal fusion done in adolescence leads to improved cosmesis but also results in a stiff spine, which can limit some activities of life that require flexion and motion of the trunk. However, when a curve shows trunk imbalance and progression of approximately 50 degrees, it is best to surgically correct and fuse the spine. Untreated, the curve will invariably progress and can lead to terrible deformities early in adulthood. Correcting severe spinal deformities in adulthood is associated with high rates of complications. Cardiorespiratory impairment is seen mainly in 2 conditions: (1) very severe thoracic curves with a Cobb angle above 85 to 90 degrees and (2) marked thoracic lordosis in which the thoracic vertebral bodies protrude into the anterior chest, described by the "spinal penetration index" (Fig. 2).3 This condition can produce an impingement of the airway (main bronchi or even trachea) with intermittent atelectasis. Thoracic lordosis is also associated with horizontal ribs, leading to decreased expansion of the thoracic cage.FIGURE 2: Spinal penetration index.Even with a severe deformity we have never seen any spontaneous spinal cord compromise in developmental idiopathic scoliosis. The spinal canal remains intact despite severe deformity, and when spinal cord injury occurs it is always linked to treatment. In contrast, nerve root compression is seen in adult deformity cases, especially with degenerative changes in the aging lumbar spine. This is related to the narrowing of the canal owing to facet degeneration or to rotary dislocation secondary to the degenerative changes in the disc and facet joint. Thus, pain is a major reason for surgical treatment of the aging scoliotic spine. MY BREAKTHROUGH: A BETTER UNDERSTANDING OF THE 3D DEFORMITY IN SCOLIOSIS From 1972 to 1975, 3 clinical observations led me to better understand the difference between the normal spine and the scoliotic one. The normal spine is straight in the coronal plane, but it has a succession of balanced curvatures from head to pelvis in the sagittal plane. These include cervical lordosis, thoracic kyphosis, and lumbar lordosis all linked by junctional zones. The amount of lumbar lordosis is determined by the anatomy of each unique pelvis and is related to the sagittal alignment by the positioning of the sacroiliac joint, described as the "incidence angle" by Mme Duval Beaupere (the angle between the line from the center of the femoral head to the center of the S1 plateau and the line perpendicular to the S1 plateau). This determines the position of the pelvis, its version or tilt, and the subsequent sacral slope (inclination of S1 plateau on the horizontal). The classical definition of scoliosis was "lateral deviation with rotation." I came to understand that radiographic images were only a glimpse of what was really happening: spines must be studied as stacks of horizontal slices. The first case that showed this new concept was in a patient with paralytic pelvic obliquity. The pelvic obliquity was in the coronal, sagittal, and horizontal planes, not just what we were measuring on the anterior-posterior radiograph. The second observation came some months later with a case of congenital deformity with "rotatory dislocation." Again, the 3D nature of the deformity was evident when the entire length of the twisted segment was studied (Fig. 3). The third observation was the discovery of the "crankshaft phenomenon," whereby I understood how the continued anterior growth of the twisted vertebrae could occur despite a posterior spinal fusion. In addition, there was marvelous anatomic confirmation by Rene Perdriolle in a meticulous study of an AIS specimen, and multiple radiographs in his 1979 book (Maloine 1979). All these observations led me to a new understanding of the scoliotic curve as a torsion of the spine with lateral translation, intervertebral lordosis, and axial (or horizontal) rotation along a partial helicoid pattern that is defined by the apical and junctional zones between 2 successive curvatures.FIGURE 3: Rotatory dislocation or Siphon of the spine.At the same time I realized that efficient spinal balance led to a spatial cone of economy. This followed the demonstration of my teacher, Pol le Coeur, who showed how an articulated skeleton was able to stand with proper alignment of the skull, spine, and pelvis when connected only by 1 string on each side between the heels and femoral condyles and with a rubber strap in front of each hip. This led to the concept of a cone of economy (Fig. 4) in which minimum muscular energy is required to maintain balance between the heavy cephalic vertebrae (the head) and the polygon of support (both feet). The predominant role of the pelvis (pelvic vertebrae) is to adapt this posture to conserve muscular energy.FIGURE 4: The cone of economy.Anatomic specimens showed the 3D nature of the scoliotic deformity and pushed us to use computer reconstructions as early as 1977 to 1978. Three-dimensional reconstructions of normal and scoliotic spines were accomplished using reference points on successive orthogonal standing radiographs and a rudimentary model for each vertebra. Looking at the model from the top down was so impressive, so close to the anatomic specimens, and so different from classic radiographs that 3D modeling became, for us, crucial. Torsion is the real deformity of AIS. This has been clearly shown and precisely measured with today's sophisticated tools. The scoliotic spine is defined by a succession of segments with torsion and countertorsion. Scoliosis is a partial twist in one direction followed by another segment twisted in the opposite direction. This occurs to keep the orientation of the head (or vision) horizontal and always front-facing. This is why we must consider the segments with their torsion and countertorsion united by an intermediate neutral vertebra, which is a vertebra with no torsion. Torsion can be "structural" in some segments, meaning not reducible by any side-to-side traction maneuver, or "compensatory" and thus completely reducible by side bending. When we analyze the reducibility of a scoliotic curvature we should not rely on the Cobb angle but rather on the amount of detorsion obtained. If the detorsion seen on the axial rotation is complete it is a compensatory curve that we must protect and leave outside of any surgical fusion. The 3D measurements of a structural curve show a clear difference between the apex, where the lateral deviation from the midline is the most significant, and the amount of axial rotation. In a standing position, the apical vertebra (or sometimes the apical disc) is the most horizontal. But if we measure the intervertebral axial rotation with the adjacent vertebrae, immediately above or below the apical vertebra, we have very little difference, meaning that it is the most rigid part of the structural curve. Conversely, the least axial rotation is seen in the vertebra at both ends of a structural curve as it transitions to another structural or compensatory curve above or below. These transitional zones are also where the vertebrae are more tilted on the horizontal in the frontal plane. Finally, it is the zone where intervertebral rotation between adjacent vertebrae shows axial rotations in opposite directions. The junctional zones are the most mobile part of the structural curve and respond best when we apply the corrective forces through casting, bracing, or surgical instrumentation. patients often look kyphotic when viewed from the side. When this kyphotic appearance is analyzed in 3D, 2 categories of deformity emerge (Fig. 5). In the first group, the sagittal apex coincides exactly with the coronal apex of the curve; it is a "false kyphosis." The appearance of a kyphosis is given by a very severe axial rotation of the apical vertebrae. In reality, the apex is lordotic. The anterior length of the spine is longer than the posterior part. These are "hyper-rotatory kyphoscolioses." Shawet al showed this hyper-rotatory kyphoscoliosis. For the second category of lateral deformity overlapping the anteroposterior (AP) and lateral views shows the sagittal apex matching the junction of 2 successive curves in the AP. This is a real kyphosis, well shown in the past during instrumentation of both curves with Harrington instrumentation. Today this is shown by the rotatory dislocation phenomenon seen in some dystrophic pathologies and in double major AIS curves. The change in the direction of the axial rotation is associated with a reverse torsion above and below and suspected on the opposite widening side of the disc space surrounding that vertebra on an AP view radiograph (Fig. 6). The other sign is observed on the convex side-bending film, where the end vertebra of a curve remains twisted in the same direction as the other end of that curve. Imbalance is automatically observed if we leave this vertebra free of instrument or fusion. This becomes clear when a 3D reconstruction is carried out and we can measure the vector of maximum slope of each vertebra. Sometimes, this instability lies, not at the vertebral level itself but at a disc space level, often at L3/L4. Ending the instrumentation at such an unstable level always results in imbalance.FIGURE 5: Kyphoscoliosis.FIGURE 6: Postoperative balance assessment.HOW UNDERSTANDING TORSION CHANGED MY CLINICAL PRACTICE This basic concept of scoliosis as a 3D torsional deformity helped me understand why some early onset idiopathic scoliosis spontaneously regressed with or without cast or brace treatment and other cases progressed despite increasingly aggressive intervention. The difference was in the shape and amount of apical torsion seen on the 3D view looking from the top (Fig. 7). This concept also helped me understand the progression of a structural curve during growth, with torsion and countertorsion developing to compensate or balance the body posture. It was also helpful to understand that in cast or brace correction, it is the horizontal plane that must be corrected, trying to "untwist" the apex rather than trying to translate it. On the radiograph in the cast or brace, success is not predicted by the Cobb angle correction but rather by the degree of correction of the apical axial rotation.FIGURE 7: Schematic 3-dimensional reconstruction of the spine (1978).This also explains why, the nucleus pulposus is ejected toward the convexity in developmental thoracic scoliosis. Together with the subsequent remodeling of the bodies a limitation to any instrumented attempt at reduction occurs, and until this "locking nut" is removed, a harmonious reduction is difficult to achieve. A "straight" spine in the coronal plane is often also "straight" in the lateral plane and thus is not in the desired degree of kyphosis. Some surgeons feel that it is necessary to perform an anterior release before posterior instrumentation. My personal experience has led to the conclusion that we can expect a 40 –to 45 degrees improvement in the Cobb angle from anterior release. However, there is always a loss of respiratory function even when entering the chest with a scope. The surgeon considers the risks and advantages. Finally, some surgeons, when faced with very severe rigid apical deformities, advocate a posterior vertebral column resection. In my opinion this technique necessitates a very experienced team. LESSONS OF HARMONY AND BALANCE, OR WHY THE MAXIMUM CORRECTION IS NOT ALWAYS OPTIMAL When a segment of the spine is fused with or without instrumentation, the result is a rigid spinal segment between mobile spinal segments. The residual deformity included in this piece of bone has relatively little effect on the 3D positioning of the mobile segments above and below the fusion mass, if the fusion mass is properly aligned within the gravity line (Fig. 8). The most important part of the spine after a fusion is the unfused, remaining mobile section, which will be required to maintain trunk balance. This is why my friend, Alain Dimeglio, used to say, "Maximum reduction is not always the optimum." With modern and powerful techniques one can obtain a straight spine in the coronal plane—but frequently a straight spine in the lateral view does not reflect the harmony required by a normal spine. Ideal 3D balance must be the goal.FIGURE 8: Quality of the location of the fusion mass.THE LESSON OF THE THORACIC CAGE The rib hump, a deformity of the thoracic cage associated with torsion, is often a major cosmetic concern for the patient. Modern "derotation" devices attempt to correct this rib deformity with some success, but this can flatten the entire thoracic cage. As this is insufficient for severe rib humps, various thoracoplasty methods have been developed. Unfortunately, the cosmetic improvement comes with some impairment of pulmonary function and occasionally persistent chest pain. However, there is a second hump, the intrathoracic vertebral hump.3 It is created by penetration of the vertebral bodies at the apex of the curve into the thorax. It is defined by the spinal penetration index and can cause airway compression with resulting temporary or permanent areas of atelectasis. ANOTHER LESSON: DO NOT ABANDON THE ANTERIOR APPROACH The major advantage of surgical correction of the spinal deformities through the anterior route is the protection of the posterior spinal muscle layer, a perfectly coordinated network of muscles from the nape of the neck to the pelvis. A posterior approach, stripping the muscles from the bone, destroys this beautiful, harmonious mechanism. The objections to the anterior approach are the lower fusion rate than with posterior routes, pulmonary function consequences, and the possibility of ending the fusion too short. Nevertheless, an anterior approach must be carried out (with or without a posterior procedure) if the patient is young enough that a crankshaft deformity might occur. THE VALUE OF DELAYING SURGERY IN EARLY ONSET IDIOPATHIC SCOLIOSIS Early onset idiopathic scoliosis, formerly known as "infantile" or" juvenile" scoliosis, is a special diagnosis. Some cases spontaneously regress or regress completely with cast or brace treatment. For those who do not, there is a strong tendency to perform early surgical treatment. The lesson I learned from this difficult situation is that one gets caught in a trap of repeated surgeries and complications. I believe we must delay surgery by performing repeated and well done casting under anesthesia with intervals of bracing in between the casts. We must delay surgery as long as we can. These casts and braces must provide protection to the thoracic cage, understanding the pressure areas and providing wide windows for pulmonary function (eg, the Garchois Brace). We often add respiratory positive pressure ventilation therapy 3 or 4 times a day. DEVELOPMENT OF NEW SYSTEMS, NEW INFORMATION, AND THE PROSPECT OF MORE UNDERSTANDING The EOS biplanar upright x-ray imaging system was developed from the joint work of Nobel laureate, Georges Charpak and his team, the biomechanical engineers of ENSAM Paris, (W Skalli and F Lavaste), LIO Montreal (J de Guise), the clinicians of the St Vincent de Paul Hospital of Paris, including radiologist Dr G Khalifa, and myself.4 The system allows simultaneous AP and lateral digital radiograph imaging of the entire body in a standing (functional) position using a 15-second ultra-low-dose radiation scan. The x-ray is always perpendicular to the target, resulting in no distortion. The radiation dose is 1/10th that of radiographs and 1/800th of 3D reconstructions using computed tomographic imaging. Another advantage is that EOS is done in the standing position with the effect of gravity. Biomechanical engineering software provides a 3D surface reconstruction as reliable as that achieved by computed tomography (Fig. 9A to D).FIGURE 9: A, EOS imaging device. B, Standing 3-dimensional reconstruction. C, Three-dimensional surface reconstruction. D, Top view of a scoliotic spine.The system is useful for all orthopaedic surgery, including the upper limbs, but it is particularly interesting for the pathology of the spinal column. With just 1 radiation exposure it is possible to assess balance, volumetric representation of the spine, the thoracic cage with the torsion of each vertebra, the 3D measurement of the pelvis, and the 3D positioning of each segment of the skeleton, relative to one another, and finally the classic measurements such as the Cobb angle. Preoperative and postoperative comparisons are easy and reproducible as are measurements of the effect of bracing on the spine and thoracic cage. EOS measurements allow for new information such as the "vertebral vector" developed by T Iles, the role of the pelvic vertebrae to compensate for any change in the posture secondary to spinal surgery, and measurement of decompensation in the aging spine where pelvic retroversion is easily measured with EOS. EOS also allows for various correction strategies to be evaluated after entering data regarding rigidity and motion of the segments involved in the surgery. It is possible to link this system to external markers to measure the 3D motion using noninvasive methods and so approach the real dynamic balance of the patient. With this system we enter a new era of real measurement of spinal deformities and will eventually replace the so-called "gold standard" Cobb angle, which measures only the collapse of the spine. Will this new technology give us, in the near future, a real anatomic and functional 3D spinal score? Analyzing this and a parallel score, which would include questionnaires and other subjective data, would allow us to reach a better assessment of the disorder and the outcomes of our treatments. My real dream is that the etiology of idiopathic scoliosis will be discovered perhaps even leading to a medical treatment for AIS.