This article is more than 5 years old. Information may no longer be current.
Opinions vary on Scheuermanns kyphosis treatment
Orthopedic spine surgeons offer their tips for diagnosing and treating this troubling condition.
Click Here to Manage Email Alerts
In 1921, Danish radiologist Holger Werfel Scheuermann described the characteristic rigid kyphotic deformity that differed from the more flexible postural round back. Today, the etiology remains unknown.
In 1964, Sorensen defined the radiographic findings of Scheuermann’s kyphosis — anterior wedging greater than 5° on three or more consecutive vertebra. The kyphosis is characterized by a sharper, fixed curve that includes vertebral wedging, Schmorl’s nodes and apophyseal irregularities. Although the kyphosis was originally attributed to the thoracic spine and without scoliosis, surgeons now know that it can occur in the thoracolumbar region, and about 20% of patients have concomitant scoliosis of about 20°.
In addition to pain at the deformity level, there are reports of tight hamstrings, low back pain and spondylolysis associated with the disorder. This may be related to the hyperlordosis of the lumbar spine seen in these patients.
While the guidelines for the surgical treatment of idiopathic scoliosis are straightforward, they are not applicable to kyphosis. Pain is a factor heavily weighed in this decision-making. There is also a higher risk of neurological complications associated with the surgical treatment of kyphosis than with scoliosis. In congenital kyphosis, the danger has been associated with anterior release of the segmental vessels, whereas in noncongenital kyphosis, there are questions surrounding vascular vs. mechanical reasons.
Below we have convened a panel of leading surgeons who have reported on the condition to update our readers based on our queries.
Alvin H. Crawford, MD
Moderator
| ||||||||
Alvin H. Crawford MD
Keith H. Bridwell, MD
Charles E. Johnston, MD
A. Atiq Durrani, MD 
Jose Herrera-Soto, MDAlvin H. Crawford, MD: Is there a level of genetic or clinical research for the etiology of Scheuermann’s kyphosis you can offer to other doctors and patients?
Keith H. Bridwell, MD: There is no definitive multicenter study regarding the etiology of Scheuermann’s kyphosis. This is why many of us refer to it as idiopathic kyphosis.
A. Atiq Durrani, MD: This is a very common question asked by many parents. The etiology of Scheuermann disease remains unclear. Either genetic or mechanical factors, or a combination of the two, have been postulated to explain Scheuermann disease.
As early as 1978, Halal et al described five families in which the disease seemed to follow an autosomal dominant pattern of inheritance. Recently, Damborg reported an overall prevalence of 2.8% with heritability of 74% in more than 46,000 twins from the Danish Registry. Since the pair-wise and proband-wise concordance and the odds ratio were two to three times higher in monozygotic than in dizygotic twins with high heritability, the researchers concluded that there is a major genetic contribution to the etiology of Scheuermann disease.
Axenovich et al reported on segregation analysis of Scheuermann disease in 90 families from Siberia. According to their model, Scheuermann disease should never occur in the absence of the mutant allele. All male carriers of the mutant allele develop the disease, but only half of the female carriers manifest it. They reported a high frequency of idiopathic scoliosis in the families with Scheuermann disease (0.08 vs. 0.01-0.02 in the general population), with a succession of idiopathic scoliosis and Scheuermann disease in consecutive generations, indicating genetic unity of Scheuermann disease and idiopathic scoliosis. This echoes our clinical observation.
Crawford: What should you as a doctor tell the parents of a 16-year-old child with a 65° curve?
Bridwell: Assuming that this 16-year-old child is skeletally mature, I would tell the parents that this is not a serious spine problem; It is a mild to moderate clinical deformity. We don’t really know what the likelihood is that a 65° kyphosis will gradually progress into adulthood. Maybe it will and maybe it won’t. It may slightly increase the likelihood of having thoracic back pain, but it is not likely to cause substantial or severe back pain. It should not interfere with function. The likelihood that a 65° kyphosis will result in any kind of neurologic deficit is extremely low.
Durrani: This is at least 20° above the normal age-matched kyphosis. Briggs et al have reported that increases in thoracic kyphosis are associated with significantly higher multisegmental spinal loads and trunk muscle forces when the patient is in an upright stance. These factors are likely to accelerate degenerative processes in spinal motion segments.
How quickly Scheuermann disease becomes symptomatic depends upon two factors: first, the magnitude of the deformity, and second, the location of the apical deformity. Soo et al reported that patients with kyphotic curves exceeding 70° had an inferior functional result in their long-term study. Murray et al reported that patients with a mean kyphosis of 71° had more intense back pain, jobs that tended to have lower requirements for activity, less range of motion of extension of the trunk and less-strong extension of the trunk, and different localization of the pain at a mean follow up of 32 years. Gennari et al studied the association of the apical level of the deformity with back pain and reported that thoracolumbar kyphosis had the highest association with pain.
|
Images: Bridwell KH |
It will be prudent for this patient to start an exercise program to relieve lower extremity contractures and strengthen abdominal musculature, coupled with practiced normal posture in stance and in sitting. Bracing in a skeletally mature patient has very little benefit, if any. If back pain becomes persistent or the deformity progresses, surgical correction may be considered. Family needs to be informed about what may be in store.
Charles E. Johnston, MD: Assuming that the 16-year-old is mature (Risser 4 or 5), the natural history of a 65°, otherwise asymptomatic kyphosis is uncertain regarding progression. Thus, the only rational treatment is observation.
Jose Herrera-Soto, MD: Though there is evidence of kyphosis, and that the chances of curve progression are there, it can be observed for progression. I would also inform them that no treatment besides physical therapy for conditioning and flexibility is warranted. Bracing is not indicated in the mature patient – it has been shown to be ineffective after skeletal maturity. However, observation with annual follow-up radiographs is recommended, and if progression is noted, then a posterior spinal fusion is warranted.
Crawford: What level do you conventionally use to measure kyphosis and what position do you recommend for the arms (ie., at the side at a 90° angle holding onto poles, etc.)?
Bridwell: I usually use levels T5 to T12, but it depends on the personality and character of the kyphosis. The levels of kyphosis that I will typically measure in the thoracic spine include T5 to T12, T3 to T12 (if we can see T3), T10 to T12, T12 to L2 and T10 to L2. I will also measure the segments that are the most kyphotic, which may not always fit to the standard measurement levels. In terms of arm position, I have been using the Horton position — placing the fists in the superclavicular fossa. In this position the arms are probably flexed forward about 30°. The idea is to have the arms flexed forward enough that they are out of the way of the spine so it can be visualized, but not so far flexed in front as to displace the C7 plumb posteriorly. Also, the idea of using the same arm position on all radiographs is to have consistency so the position is identical both preoperatively and postoperatively. Otherwise, arm position will have some impact on the ability to visualize the spine and may have a tendency to displace the C7 plumb/sagittal vertical axis. The arm position is more important for judging global sagittal balance than for judging Scheuermann’s kyphosis, as most Scheuermann’s kyphosis patients have acceptable global sagittal balance. In fact, they tend to be in somewhat negative sagittal balance.
Durrani: T3-T12 is generally good for Scheuermann kyphosis with upper thoracic or midthoracic apical deformity. For thoracolumbar apical Scheuermann kyphosis this will be inaccurate. It is more accurate to use the end kyphotic vertebra as the distal measurement point. There are three commonly used positions for obtaining a lateral 36-inch radiograph; shoulder flexed to 90° with elbows extended, shoulders flexed to 30° and lastly the fist on clavicle position. There is a general agreement that placing the shoulders at 90° flexion results in a negative shift of the sagittal vertical axis. We standardize our lateral radiographs by shoulder flexion at 30 ° to maximize the repeatability of sagittal vertical axis measurement. Whether any of these radiographic positions reproduces the spinal balance of the subject’s functional standing posture remains controversial.
Johnston: With modern PACS system storage of X-rays, measurement of T2 or T3 to T12 is not unreasonable, as the “windowing” ability in the electronic viewing allows visualization of the upper thoracic spine. Positioning for the lateral X-ray is with arms flexed in front of the torso and knuckles in the supraclavicular fossa.
Herrera-Soto: Studies have shown that arm position may alter the sagittal balance. It is recommended to place the hands on the ipsilateral shoulders. This allows the spine to be visualized on both the PA and lateral views. Most of these patients have a negative sagittal balance. This means that the center of the C7 body is posterior to the posterior edge of the sacrum. I measure from T2-T3 (based on my ability to visualize the endplates) to the lowest vertebra that is creating kyphosis. Most of the time is T12, but it can be L1 or L2 depending on the type of Scheuermann’s.
Crawford: Is there a clear-cut indication for bracing. If so, what is it in your hands? What are the criteria?
Bridwell: I have found it is much harder to brace a Scheuermann’s kyphosis patient than one with idiopathic scoliosis. I believe this is in large measure because the brace has to extend up so high. It is very difficult to get substantially above the kyphosis. Most braces made to address thoracic kyphosis are not able to accomplish such. The TLSO (thoracic lumbar sacral orthosis) braces with a sternal pad fairly high up are most common, which patients fairly accept, but I don’t see accomplishing substantial correction. So I think the only two-brace paradigms that really seem to work are either serial hyperextension casting or the Milwaukee brace. Because compliance is quite poor, I would only consider bracing in a thin, highly motivated patient who is skeletally immature and has a kyphosis in excess of 65°. If the patient is not thin and highly motivated, there is no point in moving forward with this approach.
Durrani: Bracing for Scheuermann kyphosis is a well-established treatment modality. Sachs et al in their classic paper reported an improvement of deformity in 55% of patients, which was maintained for 8 years after consistent Milwaukee Bracing for an average of 3.5 years. Most of the curves in their study more than 70° at the time of bracing eventually required surgical correction. A combination of casting and bracing has been advocated as a viable treatment for Scheuermann’s Kyphosis. I recommend full-time bracing for flexible curves less than 70° in skeletally immature patients. I inform them that full-time compliant bracing is successful in just over 50% of patients. In my practice a decreasing number of patients accept the odds of success vs. length of treatment as a viable option. I do not recommend casting.
Johnston: Bracing is indicated in the unusual circumstance of an immature (< Risser 3) teenager who is motivated to do whatever is necessary to improve their cosmetic or symptomatic deformity of up to 70°-75°. It is unusual because when the teenager is shown the Milwaukee brace, the desire for nonoperative treatment usually dissipates. With nonmotivated individuals, bracing will be ineffective due to noncompliance. In the even rarer instance of a relatively mature patient with a large curve who does not want surgery as the initial treatment, serial casting as described by Ponte can be effective. Few patients will accept this method, but it is definitely a viable alternative to operative management.
Herrera-Soto: Bracing is indicated in the immature spine with a Risser of 0 to 2. Flexible curves can also be braced until the patient has reached skeletal maturity. It is usually most efficacious with curves from 45°-65°. Curves with vertebral wedging of over 10° have been shown to progress despite bracing. The best method of bracing is using the Milwaukee-type brace. However, it is very difficult to convince the patients to use this brace due to appearance issues.
Crawford: What are the indications for surgery? Is there a length of follow-up required during progression before surgery, or do you draw a line in the sand based on the curve angle and skeletal age?
|
|
Bridwell: The tidemark for surgical treatment is not as clear as it is for idiopathic scoliosis. By and large, I will consider surgical treatment if the kyphosis exceeds 75° and the patient is, in some way, shape or form, symptomatic — meaning either having substantial back pain that has failed conservative measures or the deformity is unacceptable to the patient and family. Other factors include the location of the kyphosis and the extent to which it is sharp and angular. If the apex of the kyphosis is at the thoracolumbar junction rather than the mid-thoracic spine, the deformity usually looks worse. The more sharp and angular the kyphosis is may suggest more of a potential consideration for surgical treatment. Probably, the average Scheuermann’s kyphosis that I operate on is a 90° deformity. The type of patient I am most likely to operate on in the teenage population is either a thin female or a nonathletic male. An athletic teenage male or female will occasionally present with a fairly substantial Scheuermann’s kyphosis and I am much less likely to consider surgical treatment than for the other patient groups. If the patient has a substantial kyphosis of 80° or 90° and has demonstrated progression and is skeletally immature, those all represent relative indications for surgical consideration.
Durrani: There are two generally accepted indications for surgical intervention, persistent back pain and respiratory compromise with progressive deformity.
Murray et al reported that patients with kyphosis over 100° and apex of the curve in the first to eighth thoracic segments had restrictive lung disease. My indications for surgery are symptomatic curves over 70° which do not respond to back rehabilitation for 6 months and even asymptomatic curves above 100°.
Johnston: Surgical indications include a large (>75°) magnitude deformity which has significant associated pain, either in the kyphosis itself, or in the compensatory hyperlordotic cervical or lumbar regions. Cosmetic issues are also important, but in the absence of serious pain, the cosmetic concerns should be evaluated psychologically and an attempt to rectify physical deconditioning is appropriate. Patients with rigid hyperkyphosis are frequently sedentary, overweight, and have a poor body image, and so the surgical option is only offered once the physical and psychological complaints have been addressed nonoperatively and with appropriate PT and psychologic support.
Herrera-Soto: The SRS has proposed several indications for surgical treatment of Scheuermann’s kyphosis. They include progressive curves despite bracing, curves that are larger than 65°, and painful degenerative curves. Most of the thoracolumbar curves are also cosmetically displeasing and warrant surgery, and the most painful. There is excessive kyphosis in a region of the spine (T10-L2) were the sagittal profile should be 0° of kyphosis.
Crawford: What surgical techniques do you apply (endoscopic release, open release), and what are the indications for combined surgery?. In essence, please outline your particular surgical approach.
|
Bridwell: I usually do a supine hyperextension lateral radiograph to assess the flexibility of the curve. I do it both with and without a bolster. Clinical assessment is represented by a prone hyperextension maneuver. In almost all cases surgical treatment consists of placement of bilateral pedicle screws at all levels and performance of Ponte osteotomies through the most kyphotic segments, usually T5 to T12. Correction is achieved by a combination of cantilever and compression. Typical fusion levels are T3 to L2. I usually try not to put much force on the screws at T3, T4, T5 above and L1 and L2 below and apply the forces between T5 and T12. I have gotten away from using iliac bone graft, but will still occasionally use it in a circumstance of kyphosis, especially if the kyphosis correction has not been that dramatic. My preference is to generate a large volume of local bone off the posterior elements with use of Leksell, burr, osteotome and a mallet. I do use bone morphogenetic protein (BMP) and may in some circumstances still use iliac bone, in particular if there is a substantial residual kyphosis that exceeds 55° after the correction. I rarely do anterior releases. I would only consider that in a circumstance where there was absolutely no flexibility whatsoever on the hyperextension films. I used to treat Scheuermann’s kyphosis with anterior releases, bone graft and posterior correction, but with the use of Ponte osteotomies and pedicle screws at all levels I rarely do that now. My preference is to use 5.5-mm smooth stainless steel rods and multiaxial screws at all levels.
Durrani: Once a patient is deemed a candidate for surgical correction of the deformity, standing 36- inch PA and lateral scoliosis radiographs along with a hyperextension cross table lateral over a bolster are obtained. For curvatures that correct to within the physiologic range of kyphosis i.e. less than 50° on the hyperextension film, I recommend posterior instrumented spinal fusion with pedicle screws at each level. I perform Smith Peterson Osteotomies (SPO) at 3-4 apical levels and use spondylolisthesis reduction Pedicle screws at two levels at the proximal and distal ends of the constructs each to aid in reduction of the deformity and prevent screw pull out at these levels. My final maneuver is apical compression to close the osteotomies and obtain final correction. I do not harvest iliac crest for bone graft, but use a combination of allograft and BMPs as graft extenders especially in skeletally mature patients. For patients with deformities that do not correct to within the physiologic range, 50°, without evidence of significant restrictive lung disease, I perform a simultaneous prone video-assisted thoracoscopy (VATS)-assisted anterior spinal release with structural grafting and a posterior instrumented spinal fusion using pedicle screws at each level and multiple apical SPO. For patients with deformities that do not correct to within the physiologic range but have significant restrictive lung disease, I do apical vertebral column resection (VCR ) with posterior instrumented spinal fusion using all pedicle screws construct.
Johnston: Surgical treatment by posterior instrumentation and fusion is the current recommended treatment, with shortening of the posterior column providing the method of correction. As determined in a study from our institution, there no longer appears to be an indication for anterior release and fusion in the treatment of hyperkyphosis, as the posterior method employing Ponte osteotomies at the apical segments combined with compression instrumentation achieves >50% correction with little or no loss of correction. The Ponte osteotomy allows the posterior column shortening, which translates the apex of the kyphosis ventrally without elongation of the anterior column and without significant cantilever forces which can produce junctional kyphosis at either end of the construct.
Traditionally anterior release was recommended to elongate the anterior column as an improved corrective method. However, such a release requires structural anterior column support to maintain the correction, otherwise the disc spaces will simply “settle” in the postoperative period with loss of correction. If no anterior release/lengthening is attempted, no structural support is required. Additionally, anterior column support requires thoracotomy or thoracoscopy, adding pulmonary morbidity to the procedure. Although rare, anterior elongation has been associated with neurologic deficit, especially if segmental vessels are ligated.
For these reasons the posterior-only procedure is just as effective with less complications. We routinely place pedicle screws at levels immediately above and below the apex of the kyphosis, and use additional screws or transverse process hooks at each level to the end vertebrae, usually T2-L2 for a common hyperkyphosis deformity. In the usual kyphosis patient, the thoracic transverse processes are frequently very robust, and are just as effective as bone anchors as pedicle screws. Remember, sublaminar implants are dangerous in kyphosis correction due to possible canal intrusion during cantilever maneuvers. We have used threaded-rod and solid-rod constructs, the latter being indicated in the larger patients exceeding 80 kg. The threaded rod is used as the corrective device to shorten the posterior column by using nuts to move the bone anchors along the rod, compressing the anchors cephalad to the apex downward, and caudal to the apex upward in a sequential and progressive fashion. When correction is complete – an intraoperative control X-ray should be obtained as overcorrection can occur – the threaded rods are exchanged, one at a time, for 6.4-mm solid rods.
We have not used any platelet derived factors or BMP for a routine case, and use local autogenous bone with allograft for our fusion.
For more information:
- Keith H. Bridwell, MD, can be reached at Washington University School of Medicine, Campus Box 8233, 660 S. Euclid Ave., St. Louis, MO 63110; 314-747-2533; e-mail: Bridwellk@wustl.edu. He is a consultant for DePuy and Medtronic Sofamor Danek. Washington University receives research money from Medtronic Sofamor Danek to study generic outcomes on patients with spinal disorders.
- Alvin H. Crawford, MD, can be reached at Cincinnati Children’s Hospital Medical Center, Division of Pediatric Orthopaedic Surgery, 3333 Burnet Ave., Building C, MLC #2017, Cincinnati, OH 45229; 513-636-4787; e-mail: Alvin.Crawford@cchmc.org.
- A. Atiq Durrani, MD, can also be contacted at Cincinnati Children’s Hospital Medical Center; 513-636-4785; e-mail: aatiq.durrani@cchmc.org.
- Charles E. Johnston, MD, can be reached at Texas Scottish Rite Hospital for Children, 2222 Welborn St., Dallas, TX 75219; 214-559-7558; e-mail: Charles.Johnston@tsrh.org. He receives honoraria and royalties from Medtronic.
- Jose Herrera-Soto, MD, can be reached at Orlando Regional Healthcare, 89 West Copeland St., Orlando, FL 32806; 321-841-3059; e-mail: Jose.herrera@orhs.org.
References:
- Bridwell KH, Lenke LG, Baldus C, Blanke, K. Major Intraoperative Neurologic Deficits in Pediatric and Adult Spinal Deformity Patients: Incidence and Etiology at One Institution. Spine: 1998;23(3)1:324-331.
- Durrani AA, Crawford AH. Choudhury S, et al. Complications of Surgical Management of Scheuermann’s Kyphosis, Proceedings of the Scoliosis Research Society Meeting, September, 1998, New York, NY.
- Horton WC, Brown CW, Bridwell KH, et al. Is there an optimal patient stance for obtaining a lateral 36” radiograph? A critical comparison of three techniques. Spine. 2005;30(4):427-433.
- Johnston CE, Elerson E, Dagher G. Correction of adolescent hyperkyphosis with posterior-only threaded rod compression instrumentation: Is anterior spinal fusion still necessary? Spine. 2006;30(13):1528–1534.
- Soto JH, Parikh SN, Al-Sayyad MJ, et al. Experience with combined video-assisted thoracoscopic surgery (VATS) anterior spinal release and posterior spinal fusion in Scheuermann’s kyphosis. Spine. 2006;30(19):2176–2181.
- Tribus CB. Complication of the surgical management of Scheuermann’s kyphosis secondary to thoracic stenosis. Spine. 2001;26(9):1086–1089.