Portal de Periódicos da CAPES

Debate: Degenerative Scoliosis

2006; Lippincott Williams & Wilkins; Volume: 31; Issue: Suppl Linguagem: Inglês

10.1097/01.brs.0000234732.43489.0f

1528-1159

Behrooz A. Akbarnia, James W. Ogilvie, Kim Hammerberg,

Spine and Intervertebral Disc Pathology

The operative treatment of adult lumbar degenerative scoliosis is a formidable challenge to the spinal deformity surgeon. Recent reports indicate that the major surgical complication rate of adult lumbar scoliosis fused to the sacrum ranges from 56% to 75%, and the unplanned reoperation rate ranges from 18% to 58%. In addition, complication rates are further increased by advanced age and medical comorbidities. The improvement in pain and physical activity achieved by fusion has not been as extensively published as other orthopedic procedures such as total hip arthroplasty. The decision to operate on an elderly patient with significant comorbidities must weigh the high complication and reoperation rates against anticipated improvement. If surgery is contemplated, one must consider complete preoperative medical evaluation. Postoperative care also must involve an appropriate interdisciplinary team and include established care patterns for at-risk patients. The surgical procedure itself may need modification to achieve lesser goals than would be attempted in a more robust patient. An informed consent discussion should include the patient and her support group. Risks and complications must be fully understood, and the patient will ultimately choose between those risks and her quality of life. The surgeon should offer options and probabilities, but the patient must bear the consequences of that choice. Case Presentation: Behrooz A. Akbarnia, MD A 65-year-old woman presents with a progressive lumbar curve. The patient has significant radiculopathy and neurogenic claudication. The patient has failed standard nonoperative treatments and states, "I cannot live like this." Imaging studies show rotatory subluxation at L3–L4, fixed tilt at L4–L5, marked spinal stenosis at both of those segments, and foraminal stenosis at L5–S1. At a minimum, surgery will require multiple-level decompression and fusion and instrumentation from the distal thoracic spine to the sacrum. She is an insulin-dependent diabetic, well controlled. She quit smoking 5 years ago, is 5′4″ in height and weighs 200 lb. The decision to proceed with surgery may result in too high a likelihood of complications, whereas not proceeding with surgery decreases the chance of improving her quality of life (Figures 1, 2). Thus, the debate becomes whether to operate or not to operate.Figure 1: Preoperative anteroposterior (A) and lateral (B) radiographs showing significant loss of coronal and sagittal balance as well as degenerative lumbar scoliosis, rotatory subluxation, and fixed tilt at L4–L5.Figure 2: Preoperative CT/myelogram confirming significant degenerative stenosis at L4–L5, L5–S1.The following paragraphs represent each debaters argument for their side of the debate. To Operate: James W. Ogilvie, MD This case illustrates a frequent dilemma for the adult spine surgeon: a potentially high-risk patient with a difficult, but treatable lesion. Assuming that all reasonable nonsurgical options have been exhausted, surgery must be viewed as the balance between quality of life and the possibility of surgical complications including death. Nonintervention also has its risks of decreased mobility with its negative implications for general health and depression from debilitating pain or neurologic deficits. A decision-making process must include risk stratification at several levels. The first decision is to consider the diabetic patient in general. The average longevity for a diabetic female is 70.7 years, which is considerably shorter than for nondiabetics, but still represents 6 additional years of this patient's life. It is noteworthy that patients with diabetes at the age of 85 are less active (32%) than their nondiabetic counterparts (42%). This highlights the fact that there are 85-year-old patients with diabetes that can be evaluated for their activity level. By implication, the 65-year-old patient in this debate may have several decades of life ahead. It has been documented that, although patients with diabetes have a higher complication rate, they generally do well in surgery.1 Even with diabetes, spine surgery is reasonable in geriatric patients.2 Patients with diabetes have a greater in-hospital morbidity and mortality in a group of elderly hip fracture patients, but they do regain mobility in a similar fashion to nondiabetics. This study was over the period 1987 to 1996, and few details are given of the medical parameters mentioned in the following discussion. Can a lesser procedure be done? However, it is not uncommon to encounter patients who have had a "conservative surgery," which was composed of decompression that destabilized a motion segment. The result is a patient with postoperative spondylolisthesis, worsening symptoms, prolonged convalescence, and increasing disuse atrophy of bone and muscle that has crossed the threshold into the "no surgery" category. Would a transforaminal lumbar interbody fusion (TLIF) allow decompression, deformity correction, and interbody fusion without an anterior approach? By using the TLIF approach, a circumferential fusion can be obtained with minimal retraction of nerve roots and dura while allowing for wide decompression of the subarticular recesses and foramens.3 In addition to noting the lesser cost, studies have shown that those undergoing TLIF have fewer complications than those undergoing anteroposterior surgery in the first year following surgery. This procedure, which can be done with a minimally invasive surgical approach, requires less surgical exposure than standard lumbar fusion techniques without compromising the objectives of decompression, stabilization, and deformity correction.4 This allows earlier postoperative mobilization of the patient. In short, would a base hit suffice instead of a home run? The use of pedicle instrumentation is necessary for the TLIF procedure and usually allows brace-free recovery. The presence of dual-energy x-ray absorptiometry scan documented osteoporosis, >2 standard deviation below age-matched controls, may preclude their use. Internal fixation is an important adjunct to any proposed surgery in this patient, and the inability to use it may obviate surgery. Obesity, more than 20% above the ideal body mass index, generally increases the incidence of surgical complications. Morbid obesity, greater than 100 lb above the ideal weight, presents the combination of increased wound infection, greater stress on spine instrumentation with higher failure rates, difficulty in postoperation mobilization, and greater risk of deep vein thrombosis and pulmonary embolism and is usually a contraindication to multilevel spine surgery. If this patient is to have surgery, there is a presumption that convalescence will be longer than the younger, better-conditioned spine patient, and strong consideration should be given to postoperative anticoagulation. Would the use of bone morphogenic protein (BMP) shorten the operating time and morbidity? The use of osteoinductive agents can improve fusion rates and decrease donor site morbidity. Although the comorbidities of diabetes, obesity, and multilevel fusion have not been specifically addressed in studies of BMP and its cost-effectiveness, inductive logic suggests that in may be of value in this clinical situation. There are two particular categories of concern in diabetic patients. The first addresses ischemic vascular episodes during surgery and in the immediate postoperative period. The metrics of cardiovascular risk stratification are well known. If there are two or more of the following features [1) history of ischemic heart disease, 2) a complex surgical procedure, 3) insulin-dependent diabetes mellitus, 4) creatinine greater than 2 mg/dL, 5) a history of CVA, or 6) a history of congestive heart failure], a cardiac stress or isotope test is indicated.5 This patient has, at a minimum, criteria 2 and 3 and therefore requires a thorough cardiology evaluation. The results may preclude surgery. In Japanese patients with diabetes, there is no association between mortality from ischemic heart disease and glycosylation, but there is a correlation with diabetic retinopathy and/or massive proteinuria and ischemic heart disease.6 Those features are not known in the above patient but are critical parameters in patient selection. The surgeon must have the support of dedicated internists or endocrinologists, intensivists, anesthesiologists, and cardiologists for perioperative care.7 A perioperative β-blocker protocol has shown to result in fewer cardiac complications and is indicated in this setting.8 Any member of the care team, surgeon, cardiologist, anesthesiologist, or internist may advise against a surgical procedure based on their own individual judgment. Those decisions must be parametric and based on the current standard of care. The second concern is wound infection in the immediate postoperative period. Serum glucose levels are closely related to wound infections in patients with diabetes. Tight glycemic control by continuous insulin infusion is both cost-effective and has been shown to be effective in reducing wound infection rates in cardiac surgery patients.9 This should be considered in any insulin-dependent diabetic who is undergoing surgery. Particularly in the obese diabetic, superficial wound healing is problematic. Retraction of wound edges can result in epidermal necrosis, and retracting fat can cause triglycerides and free fatty acids to be released into the wound. These do not produce an environment conducive to healing. Vacuum-assisted wound closure has been demonstrated to be useful in closing diabetic foot wounds. This may be an appropriate adjunct for the diabetic, overweight spine patient. The ultimate informed consent is a decision between the surgeon and the patient along with her family support group. Each patient has their own perspective on life itself versus the quality of life, and there are few absolute guidelines. Surgeons may not choose to assume the legal, emotional, and physical challenges of treating high-risk patients such as this one. Although that is a personal decision, the words of the American essayist Donald Shedd come to mind, "All boats are safe in (the) harbor, but that is not why we build ships." Would I do surgery? Possibly. Would I do further preoperative evaluation? Yes. Would I offer surgery? That depends on the medical evaluation and discussion with the patient. Not to Operate: K. W. Hammerberg, MD The patient is 65 years old with a progressive lumbar scoliosis. The severity of her symptoms and limitation of function are not given. She does have multiple comorbidities including diabetes, obesity, and a recent smoking history. This case represents a typical scenario of an older patient with extensive degenerative changes superimposed on what probably once was a modest idiopathic adolescent curve. In clinical practice, the rotatory subluxations and the segmental loss of lordosis seem to be the most significant components of the deformity in terms of pain generation and symptoms. In my experience, these represent a global perturbation of spinal balance which cannot be satisfactorily addressed by local procedures. I would agree that at a minimum the patient would require multilevel decompression and fusion with instrumentation from the distal thoracic spine to the sacrum. The need for anterior spinal reconstruction is another debate. The decision to operate on this patient must be shared by the patient, the patient's family, and the surgeon. The decision is based on a balance between the severity of the patient's symptoms, the natural history of the disease process, and the outcomes of the intended intervention. The generally accepted indications for surgery are pain refractory to nonoperative management, pulmonary or neurologic compromise secondary to the deformity, and possibly the magnitude of the deformity. Similar to this patient, the chief complaints of the majority of older patients who present for evaluation are pain and limitation of physical function. In other words, they complain of diminished quality of life. Therefore, surgical intervention is an elective procedure and not an absolute. In addition, it is frequently difficult to distinguish whether the diminished quality of life is due to deformity-related problems or to the general aging process and associated comorbidities. The natural history of lumbar curves in the elderly is probably one of continued progression. Schwab et al recently reported a 68% prevalence rate of scoliosis in a group of elderly patients with no previous history of spinal deformity.10 Pain does seem to be associated with lumbar curves. However, there appears to be a discrepancy between the association of pain with increasing age, and no correlation between pain and increasing curve magnitude. A surgical recommendation based solely on curve magnitude does not appear well supported. The outcomes of a surgical procedure include the anticipated complication rate and the expected levels of effectiveness. Level I studies, randomized clinical trials, provide the highest level of evidence for decision-making. In analyzing the outcomes for adult lumbar scoliosis surgery, almost all of the studies are relatively low, Level IV, case series reports providing the treating surgeon only limited evidence for decision-making. Recent case series of lumbar scoliosis fused to the sacrum demonstrate a high complication rate and only modest improvement in clinical parameters. Emani et al reported 34% perioperative and 37% late complication rates with an 18% unplanned reoperation rate.11 Edwards et al compared long fusions to L5 with fusions to S1.12 The sacral cohort was more problematic with 33% medical morbidity, 42% pseudarthrosis, and 58% unplanned reoperation rates. Other studies analyzing adult lumbar scoliosis surgery in general note that complications increase with increasing age and the presence of multiple comorbidities.13–15 Our patient must be regarded as a high-risk candidate for surgical complications. Complications such as pseudarthrosis and infection have been shown to have a negative impact on the outcome of the surgical intervention. An interesting retrospective, comparative study was reported by Dickson et al in 199516; they compared self-reported pain and functional status between a group of operated patients and a group who were offered surgery but declined. The average age of the patients was 56 years; fusion levels were not specified. The study had a number of significant limitations but concluded that the operated patients had better outcomes. Severe postoperative pain was reported by 35% of the operated patients compared to 70% of the nonoperated. This comparison would seem to favor surgery, but the preoperative pain status was not indicated. As well, the high rate of severe residual pain does not reflect favorably on the effectiveness of surgery. The operated patients reported perceived improvement in physical, functional, and self-care activities. However, there was no statistically significant difference in the scores for actual physical, functional, or self-care performance between the 2 groups. The operated group did have a 46% complication rate and 17% unplanned reoperation compared to none for the nonoperative group. Another interesting study was reported by Albert et al in 1995.17 The outcomes of a prospective series of adult deformity surgeries were evaluated with preoperative and postoperative self-reported health assessment and function. The patients demonstrated statistically significant improvement in 4 of 9 health score domains on the SF-36, although the actual improvement in the scores was marginal, only 10 to 15 points. In discussing the Albert et al article,17 Katz pointed out the modest improvement documented by SF-36 scores for adult deformity surgery in contrast to total hip replacement18: the improvement in pain for scoliosis surgery 36% versus 79% for total hip replacement and improvement in physical function for scoliosis surgery 16% versus 111% for total hip replacement. He raised the question as to whether adult scoliosis surgery is as minimally effective as it appears or whether the SF-36 is insensitive to its benefits. Would a disease-specific outcome measure such as the SRS instrument be a more valid tool? The SF-36 is sensitive to pain and physical activity, but not to self-image and deformity. Returning to our patient, how important are self-image and deformity in a 65-year-old, obese woman? Certainly, relief of pain and improvement in physical activity are important considerations, but only modest gains can be anticipated at a significant risk and expense. Surgery should be discussed with this patient and family but discouraged, unless her quality of life is severely diminished. A modest improvement in pain and function might be worth the risks in an otherwise healthy, older adult incapacitated by deformity-related pain. However, the benefit of surgical intervention in someone with only mild to moderate symptoms is debatable. Equally debatable is the benefit of surgery in the severely limited patient with multiple comorbidities, in whom the surgical risks would be the greatest and the outcomes most unpredictable. Even though a patient may say "I can't go on like this," the surgeon is not given carte blanche but has an obligation to counsel the patient toward an informed decision. In the compromised patient with specific complaints that can be anatomically identified, such as a focal stenosis or nerve root entrapment, a limited intervention can be of value. However, decompression without stabilization should be used judiciously. In healthier, more active patients, this strategy often fails, resulting in an even more complex reconstruction. The use of bone morphogenic protein and other biologic enhancers may reduce the pseudarthrosis rate but will not reduce the patient's age or moderate other significant comorbidities. Lumbar scoliosis surgery in adults has high complication and reoperation rates. Extension of the fusion to the sacrum further increases these risks. Complications have been shown to increase with age and additional medical comorbidities, resulting negatively on the clinical outcome. The effectiveness of surgery to improve pain and physical activity in similar patients has been demonstrated to be modest with a high financial cost. The need for large prospective outcome studies employing validated self-assessment instruments to justify the allocation of healthcare resources to the surgical treatment of adult spinal deformity is advocated.19 Debate Summary As the general life expectancy increases in our society, age becomes less of an issue for surgical decision-making. We are faced more and more with the dilemma of determining indications for surgery in geriatric patients with spinal deformities who desire to remain active later into life. When comorbidities such as diabetes, cardiac problems, and history of smoking exist, the decision to operate or not to operate becomes even more important for the treating surgeon because of the significant risks associated with both options. Considering the natural history of progressive spine deformity in adults, the question becomes how we can balance the risks of the treatment method we choose with expected outcomes. Ogilvie and Hammerberg debate this important issue in order to search for an answer and to define guidelines for surgical intervention in progressive adult scoliosis, especially in those cases when fusion to the sacrum and/or pelvis is necessary. There is general agreement that the rate of complications for this surgery is high, particularly when comorbidities exist. The improvement in pain and level of physical activity following fusion to the sacrum in a 65-year-old patient has not been extensively compared with other orthopedic procedures such as total hip replacement. However, in making the decision, one must recognize the patient's limitations and significance of her progressive disability. I agree with Ogilvie that surgery must be viewed as the balance between quality of life and surgical complications. As the treating surgeon, we should evaluate each risk factor individually against the expected outcomes. Ogilvie uses the example of diabetes to illustrate this point. He feels that spine surgery is a reasonable alternative in geriatric patients with diabetes in comparison to some of the other common comorbidities and the affected patient population generally benefits from surgery. The other issue that the treating surgeon should address is whether or not lesser surgery such as decompression or limited fusion could provide reasonable and lasting relief for the patient. Or do these patients require more extensive surgery including anterior exposure, decompression, and posterior instrumentation? What we should fully evaluate is the patient's dominating symptoms. Her complaints such as pain play a major role in making the proper treatment decision. Is it leg pain or back pain that is more disabling? What are the functional limitations? We should not forget why this patient has come to us in the first place. If treatment of one of her main symptoms, such as leg pain, requires extensive decompression and, in presence of significant deformity, would lead to instability, then fusion and instrumentation is necessary. If her decompression is limited and does not affect the structural integrity of the spine, then a limited surgery may be all she needs. As Hammerberg points out, the decision for surgery should be a shared one. The surgeon as well as the patient and the family should have an understanding of the limitations of surgical treatment, the natural history of the deformity, and the outcomes of intended intervention. It appears that there are few absolute guidelines for the proper selection of a candidate for surgical intervention. Each patient is different and should be considered individually. The results of nonoperative care including physical therapy, injections, and occasionally orthotic treatment should be well documented and outcomes measured objectively. When those methods fail and surgery is being considered, the procedure, risks, and expected outcomes should be discussed with the patient and her family. The surgeon is responsible to advise the patient as to what he or she feels is the most appropriate surgical or nonsurgical treatment. The patient's input is also very important to the discussion. The patient needs to be forthright in reporting his or her disability and quality-of-life issues. Surgery in this age group should be considered only if progressive physical limitations, pulmonary problems, or neurologic compromise exist. It also should be determined if pain and limitation in physical activity are due to the deformity or the aging process. There is no clear evidence that clinical symptoms correlate with curve magnitude; therefore, basing the decision to proceed with surgery solely on the degree of the curve is not well supported. When surgery is being considered as an option the decision process should include risk stratification. Further workup should be directed toward detection of comorbidities and should include medical and cardiovascular testing. Risks such as high blood glucose, the need for β-blockers, or other abnormal conditions should be addressed and corrected before surgery. In addition, potential health risks should be identified by the patient's medical team before surgery. Once all of the above evaluations have been completed, the surgeon should have an informed consent discussion with the patient and his or her support group such as her adult children to review the patient's limitations, previous treatments, and the patient's desire to have surgery. The expected outcomes as well as the risks and complications of the proposed procedure should be carefully outlined. While the patient is responsible for making the ultimate decision regarding surgery, the surgeon has a duty to explain his or her opinion in a fair and unbiased manner and help the patient to make a well-informed decision. The proposed procedure could be modified to obtain lesser goals if it would result in taking fewer risks in a high-risk patient. In this patient, however, because of the need for a multilevel decompression, fusion would be necessary (Figure 3). This patient underwent a same-day anterior and posterior surgery. The anterior portion consisted of multilevel discectomies and fusion from T12 to L5 using structural graft at L3–L4 and L4–L5. The posterior procedure involved extensive decompression from L3 to S1, posterior spinal fusion and instrumentation from T10 through the sacrum with pelvic fixation using a unilateral iliac screw, and posterior lumbar interbody fusion of the L5–S1. Iliac autogenous bone graft was used for posterior fusion. Correction was achieved by a combination of rotation and translation maneuvers.Figure 3: Postoperative anteroposterior (A) and lateral (B) radiographs following decompression, anterior and posterior fusion, and instrumentation to the pelvis. Note the improvement in the curve magnitude and coronal and sagittal balance. (Courtesy of Oheneba Boachie-Adjei, MD, New York.)When performing this type of surgery, the surgeon should be knowledgeable and experienced in treating adult deformity patients. In addition, an adequate support team such as experience anesthesiologist, intensivist, and internist and sufficient technical support in the operating room is important. Finally, one should be prepared for a high rate of complications and should identify each complication early. Both debaters emphasize adherence to details in the preoperative workup, involvement of the patient and her support group in discussions, and balancing the risks with the expected outcomes. Unfortunately, the currently available information is based on low Level IV evidence. Future prospective clinical studies using validated outcome instruments are necessary to provide much needed information to assist patients and surgeons alike in the process of deciding whether or not to pursue surgery. Acknowledgment The authors thank Sarah Canale, BS, for her help with preparation of the manuscript.