Evidence-Based Advances in Spinal Care: Where Do We Stand Today?

Alexander R. Vaccaro, MD, PhD, MBA; Charles G. Fisher, MD


Spine. 2021;46(4):E274-E276. 

Welcome to the age of robotics. Historically many spine surgeons have understandably proceeded with trepidation into new technologies due to suboptimal long-term outcomes, less than advertised performance, implant malfunctions, attendant liability, and inflated cost considerations. Will the introduction of robotics in spine surgery be different?

The operating room is a physically and psychologically demanding arena and technology to minimize these stresses is welcome. Robotic technology may alleviate some of this anxiety due to minimized hospital personnel radiation exposure, relatively lessened time consumption related to the learning curve, and improved surgical precision.[1] Reproducible surgical accuracy has been demonstrated by multiple retrospective and prospective studies about pedicle screw placement, but its effect on short and long-term outcomes has not been elucidated.[1–4] The study by Han et al, although an RCT, did not have a navigation without radiation control group, which has become the standard in many centers. The study, however, demonstrated an interesting additional consideration relevant to today's concern for junctional degeneration and that is the avoidance of collateral damage to the junctional facet joint due to precise implant placement with robotic guidance.[1] This may be the most important added benefit of this technology as the other findings such as complications, return to operating room, computed tomography (CT)-based accuracy of screw placement, although better in the robotic group, were not clinically impressive in light of the added expense of this technology.[1] Robotics may ultimately be beneficial in several ways: when preoperative magnetic resonance imaging instead of CT can be used to merge the robotic navigational system to lessen radiation exposure, and in revision cases and deformity due to improved accuracy of screw placement and less junctional facet violation.[1–5] We need to analyze further the cost of this technology offset by its benefits, and changes in workflow, which can be challenging to make informed decisions as to the value of robotics in specific cases. This is an exciting time for this technology and hopefully these issues can be better understood over the next several years as we strive to improve the safety of spine surgery and minimize the collateral radiation risk to the operating room staff.

It is generally recognized that surgery in the setting of significant axial back pain often portends an unsatisfactory outcome in patients who undergo surgery for degenerative conditions of the lumbar spine.[6,7] Surgeons are taught during training that low back pain surgery will at best improve low back symptoms in approximately 50% to 70% of patients.[8] The only well-performed studies that demonstrated the effectiveness of surgery for low back pain were those utilized in company sponsored IDE trials assessing the efficacy of lumbar arthroplasty devices.[9–11] There has been a resurgence of discussion on the effect of lumbar decompressive surgery on the subjective symptoms of low back pain in the setting of spinal stenosis.[12,13] It is now often quoted that up to 50% of one's back pain may improve after such surgery even though we counsel patients that surgery is only designed to improve lower extremity symptoms. The study by Hermansen et al[12] followed a cohort of patients in a healthcare system (Norwegian Registry for Spine Surgery) with the ability to collect long-term follow-up in the majority of patients who underwent surgery. The authors demonstrated that surgery for spinal stenosis in patients with more back pain than leg pain complaints have less satisfactory outcomes compared to those with more significant lower extremity symptoms. It would have been interesting to cull out the degree of low back improvement in subsets of patients with uniform lower extremity complaints so one could advise a patient with significant lower extremity symptoms what the potential for back pain improvement may be. This article supports a long-term belief among spine surgeons that if a patient has primarily low back pain with ambulation, it may be difficult to differentiate pain due to degenerative disk disease versus neurogenic low back pain and until that diagnosis can be better understood, surgery should not be uniformly recommended in this patient population.[12,14]

The age-old question "should one universally perform a fusion in the setting of instability related to degenerative spondylolisthesis" is debated fiercely in the literature. The consensus is that if there is obvious instability and a decompression is contemplated, then instability may worsen and a fusion may be warranted to prevent worsening back pain and leg discomfort.[15–17] Recent high-level literature has provided conflicting evidence as to the value of an adjunctive fusion in this setting.[16,18] The study by Chan et al[19] used the Quality Outcomes Database to get a better understanding if fusion is beneficial in terms of back pain relief in the treatment of degenerative spondylolisthesis. We all agree that fusion is a well-accepted treatment option for isthmic spondylolisthesis, iatrogenic spondylolisthesis, or spondylolisthesis due to trauma or infection. This study, however, examined the benefit of fusion for low back pain rather than improvement of leg discomfort and is rarely ever chosen as a primary treatment option in this degenerative condition; therefore, this study adds more confusion to the literature as to the benefits of fusion in this degenerative disorder.[19] The authors found that a fusion improved certain outcome measures but was not significantly better in others.[19] As a result, this study does not provide any further information on the value of fusion in this setting. Degenerative spondylolistheses patients need to be better categorized from an imaging and clinical perspective to determine which patients require a fusion and if so what type of fusion. We need to then study this so we can optimize resource utilization and minimize the morbidity associated with a fusion when such a procedure is not actually beneficial.[15]

Another exciting topic that has greater interest within the spinal community is the complication of junctional breakdown following correction of adult spinal deformities.[20] Yang et al[21] evaluated 736 patients who underwent surgical treatment for adult spinal deformity. The authors observed that junctional breakdown was more common in patients with a larger preoperative deformity or those with overcorrection of their deformity.[21] The authors recommended that an individual approach to deformity correction should be taken and that striving for complete curve correction in the sagittal plane may not be in the best interest of the patient.[21] This may not be due to a miscalculation of optimal deformity realignment but the fact that we may not understand where patients like to exist three dimensionally in space due to a myriad of factors related to spinal balance. We have noted anecdotally that under-correction tends to lead to less junctional breakdown than perfect calculated correction or over-correction. This may be a reflection of our lack of understanding of the confounding variables related to balance such as age, curve type, and bone health. These findings have been reported in the literature over the last several decades especially in the pediatric neuromuscular scoliosis literature.[22] These observations are assuring, as most surgeons rely on deformity correction through passive positioning on the operating table or spinal osteotomies after judging a curves magnitude in the absence of multiple spinal pelvis parameter measurements.[23] More research has to be done to elucidate the risk factors related to curve progression and junctional breakdown. Over time and with large data registries, this information should become clearer and will help guide a surgeon's selection on appropriate prophylactic methods to minimize junctional breakdown such as vertebroplasty, tethering, or preservation of the interspinous ligaments.[24]

Understanding complications related to surgical intervention in the setting of cervical deformity warrants as intense scrutiny as the plethora of papers that have been published related to the complications associated with the surgical management of adult spinal deformity.[25–28] To understand cervical deformity, a surgeon must understand global spinal alignment and its effect on cervical alignment. This is a limitation of the study by Passias et al,[29] which only considered cervical measurement parameters. The authors found that cervical kyphosis or an increase in cervical SVA increased the potential for complications following surgical intervention with >64% of patients having a complication and almost one-quarter of such complications being related to a neurologic cause.[29] Without an understanding of global spinal alignment, it is unclear whether these complications are primarily related to correction of an intrinsic cervical deformity or due to the fact that an unrecognized underlying thoracolumbar deformity predisposed a procedure to fail due to disturbance of underlying compensatory cervical deformity. We unfortunately do not have this information, but what we do know is that if one decides to operate on a cervical deformity, there is an inherently high risk for perioperative complications.[29] However, this risk may be mitigated to a certain extent with a thorough understanding of global spinal alignment and an appreciation for the compensatory nature, if present, of the cervical deformity.[30]

Finally, the study by Crawford et al[27] evaluating the indications for reoperations following an index surgery in the setting of adult symptomatic lumbar scoliosis found that these procedures were frequently performed as a consequence of complications known to occur after adult spinal deformity surgery such as rod fracture, pseudarthrosis and proximal junctional breakdown. This finding once again reflects our lack of understanding of the optimal spinal alignment one must obtain following surgical intervention for this type of deformity. The incidences of pseudarthrosis and rod fracture are a function of the amount of stresses on a fusion mass.[31] If someone is optimally aligned, the stresses are thought to be relatively lower. Can one overcome the complications of rod failure in pseudarthrosis with more metal or better biologics? These strategies may delay the onset of such complications or result in a different type of adverse event such as junctional breakdown. Again, these results serve to emphasize the fact that with aging and normal changes in body habitus over time, in addition to other unknown confounders, we truly do not know what the optimal degree or method of spinal correction should be for patients with spinal deformities. As with the findings of Yang et al, it is foolish to assume that we truly understand spinal alignment and the myriad of factors that affect it; thus, acknowledging our current deficiencies should prompt a less aggressive approach to surgical correction until we better understand the natural history of spinal deformities, especially in the sagittal plane.