Epidemiology of Spinal Surgery: Rates and Trends
Description
Recent technical changes in spine surgery, including new spinal implants and a shift towards ambulatory surgery may affect surgical rates and re-operation rates for lumbar spine surgery. This project uses national and state databases to examine whether surgical rates continue to rise, and if the introduction of interbody fusion cages resulted in acceleration of surgery rates. We will also examine surgical rates among the elderly, examine surgery rates for spinal stenosis, and examine whether or not re-operation rates are increasing over time, and finally to determine if certain surgical procedures are associated with unusually high rates of re-operations.
Principal Investigator
- Richard A Deyo, MD, MPH
Co-Investigators
- Darryl T. Gray
- Sohail Mirza
- William Kreuter
- Brook Martin
- Bryan Comstock
Research Questions
1. To examine the extent to which previous increases in the use of spinal fusion surgery may have continued or accelerated. In particular, we will determine if there was a sharp increase in rates following the introduction of interbody fusion cages, new implants which made spinal fusion surgery technically easier.
Spinal fusion surgery was the most rapidly increasing type of lumbar spine surgery during the 1980s. It has been suggested that advances in technology, including pedicle screw and plate systems, may have contributed to this rise, along with improvements in preoperative care, expanded training of spine surgeons, and reimbursement incentives. Spinal fusion rates appear to vary among geographic areas even more dramatically than rates of other types of back surgery. International rates of spinal fusion vary more than rates of other types of back surgeries, and Keller reported lumbar fusion rates varying ten-fold between communities located within 100 miles of each other in Northern New England. Rates of lumbar discectomy were strikingly less variable.
2. To determine if previously identified rapid increases in spine surgical rates among elderly patients have continued since the 1980s. This group may represent a high priority for further studies of treatment efficacy and effectiveness. If possible, we will also examine vertebroplasty – a new procedure for osteoporotic compression fractures – which is used primarily in the elderly.
In a study of Medicare beneficiaries aged 65 and older, we found that rates of spinal stenosis surgery increased 8-fold between 1979 and 1992. Furthermore, surgical rates for spinal stenosis varied almost five-fold among U.S. states. Perhaps not surprisingly, mortality and operative complications increased with age and comorbidity. This rising surgical rate was dramatically greater than among younger populations or for other diagnoses. Dartmouth investigators have shown that the overall rate of lumbar surgery among Medicare beneficiaries continued to rise by 57% from 1988 to 1997, though rates for specific diagnoses were not itemized. If this trend has continued, the older population represents a high priority for further studies of treatment efficacy and long-term outcome. These issues will become increasingly important with aging of the population
3. To determine if reoperation rates for lumbar spine surgery are increasing over time, as suggested by our previous work. We will also examine whether reoperation rates are highest in geographic areas where primary operation rates are the highest.
Our previous analysis of Medicare claims revealed a disconcerting trend. The probability of reoperation among patients having surgery in 1989 was slightly greater than the probability of reoperation for a cohort having surgery in 1985, during 3 years of follow-up. Because reoperation is generally regarded as an unfavorable outcome, such a finding may have important implications for patient selection or technical quality of care. To determine if this is a real and significant trend, it will be necessary to examine cohorts separated by longer time intervals, and to examine reoperation rates over longer periods. If reoperation rates are indeed increasing, it may suggest that advances in surgical technique are creating more problems (e.g., the need to remove hardware) or that increasing technical ease of surgery is resulting in less careful patient selection, and worse overall outcomes. Further work is also necessary to determine if any trend observed in the Medicare population is also true among younger adults.
4. To determine reoperation rates associated with different types of lumbar spine surgery. If there are diagnoses or procedures which have unusually high reoperation rates, it may suggest potential problems with patient selection or technical quality of care.
Our previous work described higher rates of reoperation among patients receiving certain types of surgery and for certain diagnoses. For example, among patients having surgery for degenerative disc disease, those who had spinal fusion surgery had nearly twice the rate of reoperations compared to those who did not have fusion surgery, even after adjusting for age, gender, prior surgery, comorbidity, and coverage by Workers’ Compensation. In contrast, patients undergoing spinal fusion in association with discectomy for a herniated disc had no differences in reoperation rates.
5. To determine the extent to which ambulatory disc surgery has begun to replace inpatient disc surgery. This trend began in the mid-1990s, but its extent has not been documented.
In the mid-1990s, ambulatory disc surgery became an increasingly common procedure. This included not only percutaneous techniques, but also increasing use of open spine surgery on an ambulatory basis(19). Furthermore, arthroscopic techniques have recently been validated and are more widely used. The trend towards ambulatory surgery seems to be greater in the United States than in other countries, perhaps because of managed care. The shift from inpatient to ambulatory surgery has been greatest for simple discectomy, although percutaneous techniques are being developed for other types of back surgery, as well. The magnitude of this shift has not been quantified, and the current proportion of lumbar spine operations performed on an ambulatory basis is unknown. This question has methodologic importance, because health services researchers studying surgical patterns and making international comparisons may have erroneous information if they are using only inpatient data. This shift also has implications for resource allocation, and largely unknown implications for long-term outcomes of care.
Research Highlights
1. U.S. NATIONAL TRENDS IN LUMBAR FUSION SURGERY FOR DEGENERATIVE CONDITIONS, 1990 – 2000
Objective: In 2001, approximately 122,000 lumbar fusions were performed in the U.S. On a population basis, this represented a 220% increase from 1990 in fusions per 100,000. The slope of the rising rate increased in 1996, when fusion cages were approved. From 1996 to 2001, the number of lumbar fusions increased over 113%, compared with 13-15% for hip replacement and knee arthroplasty. Rates of lumbar fusion rose most rapidly among patients aged 60 and above. In this group, rates increased 230% over the decade, compared with 180% among adults aged 40-59, and 120% among adults under age 40. The proportion of all lumbar operations involving a fusion increased for all diagnoses.
- Lumbar Spine Fusion Volumes: 122,316 lumbar fusions for degenerative conditions in 2001. This represents an increase in fusions from 1990–2001 of 220%
- Primary Diagnosis associated with lumbar fusion: Rate of fusion surgery increased fastest among oldest patients: 230% for age greater than 60, 180% for age 40-59, 120% for age less than 40. For patients with primary diagnosis of degenerative change, instability, or stenosis, the percent of operations involving a fusion increased from 25% in 1988-89 to 51% in 2000 – 2001.
- Lumbar fusion volume by primary diagnosis: Proportion of operations involving fusion by diagnosis in 2001: Degenerative Changes: 70% Possible instability: 93% Spinal Stenosis: 26%.
2. VOLUMES, RATES, AND CHARGES FOR AMBULATORY AND INPATIENT LUMBAR SPINE SURGERY
Objective: To track trends nation-wide and state-specific trends from 1994-2000 in: 1. Population-based distributions of inpatient and outpatient procedures 2. Overall rates of inpatient and outpatient procedures (excluding cancer, infection, trauma cases) 3. Distributions of inpatient vs. outpatient procedure types Estimate nationwide rates of inpatient and outpatient lumbar spine surgery.
- % of Lumbar Spine Surgery Cases Done on Outpatient Basis in 1997-2000: SID and SASD (4 states combined) Combined inpatient and outpatient rates rose from 1994 – 2000.
- Rates of All Inpatient Lumbar Spinal Surgical Procedures