A widely under-recognized form of arthritis, ankylosing spondylitis (AS) is a chronic, progressive disease targeting the spine.
Commonly striking in young adulthood, before age 35, AS causes inflammation, pain, and stiffness in the spinal joints--the vertebrae--and the sacroiliac joint, where the spinal column meets the pelvis. In advanced cases, this disease can result in deforming, crippling spinal fusion and organ damage. According to expert estimates, AS afflicts at least half a million people in the United States.
While there is no cure for AS, numerous studies have affirmed the effectiveness of non-steroidal anti-inflammatory drugs (NSAIDs) for improving physical function, as well as providing rapid relief from back pain and stiffness. Yet, physicians generally recommend taking NSAIDs on a short-term basis, only as needed for severe AS symptoms, due to the risk of gastrointestinal complications associated with long-term use. Inspired by the improved gastoprotective safety profile of COX-2-selective NSAIDs, an international team of rheumatologists set out to explore the impact of ongoing NSAID treatment, over a 2-year span, on the course of AS. Featured in the June 2005 issue of Arthritis & Rheumatism, their study offers hope for decreasing the progression of AS, without increasing the risk of peptic ulcers or other adverse events.
The clinical trial began with 215 outpatients, drawn from the records of 76 hospitals and private practices in France, with a history of AS. Using a computer-generated randomization list, the patients were divided into two treatment groups. The first group, comprised of 111 patients, was prescribed twice-daily treatment with a NSAID, regardless of symptoms. The second group, comprised of 104 patients, was also prescribed a NSAID, but instructed to take it only when they suffered pronounced pain or stiffness. Both groups started treatment with celecoxib at a dosage of 100 milligrams. All patients were allowed to increase their dosage to 200 milligrams, if necessary, or switch to another NSAID, as long as they maintained their assigned treatment strategy--either continuous or on-demand. Compliance was assessed by pill count.
At baseline, after 1 month, at 7 follow-up visits conducted at 3-month intervals, and at the final visit, patients in both groups were thoroughly evaluated, through questionnaires and laboratory tests, for clinical signs and symptoms of AS, as well as for any adverse events. X-rays of the spine were taken at the study's onset and 24-month culmination.
Overall, differences in hallmarks of disease activity--including spinal pain, night pain, morning stiffness, fatigue, and restricted mobility--were not statistically significant between the two treatment groups. There were also no significant differences in the gastrointestinal, respiratory, cardiovascular, and other health problems experienced by the groups, with the exception of symptoms of depression, which occurred more frequently in the continuous-treatment group. Only a single adverse event--a case of severe abdominal pain requiring hospital admission in the on-demand group--was deemed directly related to NSAID use, by the treating physician. There were, however, statistically significant differences in the X-ray evidence of AS progression between the continuous-treatment and the on-demand groups.
Complete sets of radiographs were available for 76 of the patients in the continuous-treatment group and for 74 of the patients in the on-demand group. Scored by a single observer blinded to treatment strategy, using the modified Stoke Ankylosing Spondylitis Spine Score (SASS), the X-rays showed both more pronounced disease progression and in a greater proportion of patients among patients who had taken NSAIDs only as needed for pain management. In fact, twice as many patients in the on-demand group were scored as having moderate to high levels of spinal joint damage at the 2-year mark than patients in the continuous-treatment group. The between-group difference in radiographic progression remained stable and significant after adjusting for baseline values of joint damage and disease-activity variables.
As noted by Professors Maxime Dougados and Desiree van der Heijde, the study's leading authors and participating rheumatologists, the findings have implications for both the treatment of AS and the application of NSAIDs. On the strength of the data, inflammation and progression of joint damage may be two separate processes in AS, which may be different from the situation in rheumatoid arthritis. And NSAIDs, generally considered symptom modifiers, may have unexplored disease-controlling properties.
"We conclude that a strategy of continuous use of NSAIDs decreases the radiographic progression in patients with AS without substantially increasing toxicity," Prof. Dougados states. "While awaiting confirmation of these results, we carefully recommend that if patients need treatment with NSAIDs to reduce the signs and symptoms of AS, they should take NSAIDs continuously instead of as needed based on symptoms."