Spine Surgery Outcomes More Reliable at 12 Months Than 3 Months

Nancy A. Melville

May 21, 2015

Washington, DC — Three-month outcomes after spine surgery likely do not reliably predict longer-term outcomes at 12 months on the patient level, underscoring the need for prospective spine registries to span at least 12 months in identifying true efficacy, authors of a new study conclude.

The findings are particularly important as clinical registries emerge as the most valid means of measuring real-world care in an era of value-based reform, they say.

"Registry science is here to stay in all fields of medicine, and I think we have to be at the forefront," said lead author Matthew Joseph McGirt, MD, from Carolina NeuroSurgery and Spine Associates, Charlotte, North Carolina.

"It's clear from these findings that we need to stay connected to patients early on and measure their response after spine surgery for at least 1 year. During that time, many patients will likely take a step back, while equally as many will continue to improve."

The results were presented here at the American Association of Neurological Surgeons (AANS) 83rd Annual Meeting.

With the role of registries in mind, Dr McGirt and his colleagues asked the important question of exactly how long spinal surgery patients need to be tracked in order to define success.

Their prospective, longitudinal study included 593 patients undergoing lumbar spine surgery for degenerative disease at a single center by eight surgeons over a 2-year period.

Patient-reported outcomes provided at baseline and at 3 and 12 months after surgery showed an overall correlation between 3-month and 12-month outcomes in terms of improvement in general health (EQ-5D: P < .0001) and disability (Oswestry Disability Index; P < .0001).

Measures on the individual-patient level in terms of clinically significant improvement, however, showed wide discrepancies.

Postoperative disability figures were contradictory, with 11.5% of patients who achieved significant improvement at 3 months falling below the improvement threshold by 12 months, while 10.5% of those who did not reach clinically significant improvement at 3 months surpassing the improvement threshold at 12 months.

The specificity of 3-month improvement in predicting 12-month improvement in disability was therefore only 62.6%, while sensitivity was 86.8%.

In terms of postoperative general health (EQ-5D), 8.5% of patients lost significant improvement from 3 to 12 months, while 4.0% who did not reach the threshold at 3 months gained it by 12 months after surgery.

In terms of quality-adjusted life-years, achievement of a clinically significant of EQ-5D improvement at 3 months accurately predicted 12-month improvement with only 87.7% specificity and 87.2% sensitivity.

Dr McGirt noted that a separate outcomes database from the National Neurosurgery Quality and Outcomes Database on 3073 patients also showed wide variability and movement in continued health improvement between 3 and 12 months after spinal surgery.

"We consider this to be more external validation of what we saw in our study — it's the exact same thing with much greater statistical power," Dr McGirt said.

"There is definitely not a steady state achieved at 3 months, and as many people in fact take a small step backward as do continuing forward."

Dr McGirt said the need to monitor such important aspects of patient outcomes as accurately as possible in the real world has never been more pressing.

"Consumers have vocally and consistently agreed that our surgical care product has great variability in quality and often unproven value, and because of this, we are now entering the largest transformation in US healthcare history," he said.

"Without quality numerators on the safety and effectiveness of care measurement in the [real world] where we deliver it, we are going to be left with cost-based reform."

Commenting on the study, discussant Peter D. Angevine, MD, from the Department of Neurological Surgery at Columbia University College of Physicians & Surgeons in New York, said the findings offer valuable information on outcomes.

"The study provides useful individual analysis of 3- and 12-month outcomes after surgery. With more detailed outcomes data, we will be better able to characterize the magnitude of durability of the effect of surgical interventions for spinal disorders."

Such data are particularly important in developing meaningful registries, he added.

"With clinically relevant data for registries, we will be able to construct probability models to refine the singular inductive inference of individual outcome. This is the main promise of registries but one that still remains to be realized."

Justin Smith, MD, PhD, an associate professor at the University of Virginia's Department of Neurosurgery in Charlottesville, put things further into perspective.

"Prospective longitudinal outcomes registries are critical for assessment of the effectiveness of spine treatments," he told Medscape Medical News.

"Dr McGirt and colleagues assessed an important question with regard to registry design — whether the longer-term outcomes for lumbar surgery can be sufficiently predicted by 3-month outcomes assessment."

"Their findings demonstrate that the added efforts and costs of following patients to 1 year, and potentially further, to assess outcomes following lumbar spine surgery are warranted, since the recovery process and gained benefits are not always achieved in the short term."

"Beyond the implications for registry design, this study from Dr McGirt and colleagues may prove useful for preoperative counseling of patients undergoing lumbar spine surgery with regard to outcomes expectations and the time horizon for achieving these."

Dr McGirt, Dr Angevine, and Dr Smith have disclosed no relevant financial relationships.

American Association of Neurological Surgeons (AANS) 83rd Annual Meeting. Abstract 601. Presented May 4, 2015.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: