“To address the growing focus on using patient-reported outcomes (PRO) and individual-specific outcomes when assessing osteoarthritis (OA), a strategy that defines clinically meaningful improvement or worsening is needed,” explains Jeffrey B. Driban, PhD. “Although well-established methods to identify clinically meaningful improvement exist, such as the Osteoarthritis Research Society International (OARSI) and Outcome Measures in Rheumatology (OMERACT) responder criteria set, there is less agreement on how to define clinically meaningful worsening. One possibility is to apply the inverse of the OARSI-OMERACT responder criteria to assess clinically meaningful worsening. This approach, however, has not been validated in relation to clinically relevant worsening outcomes in people with OA.”
For a paper published in The Journal of Rheumatology, Dr. Driban and colleagues aimed to assess if the inverse OARSI-OMERACT criteria relate to concurrent radiographic knee OA (KOA) progression and decline in walking speed, as well as future knee replacement. “We adapted the OARSI-OMERACT criteria to define worsening,” Dr. Driban says. All knees in the study were required to have an inclusive definition of symptomatic OA—at least doubtful OA [Kellgren-Lawrence (KL) grade ≥ 1] and knee pain ≥ 10/100 [Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain subscale] at the 12-month visit.
One in Five Met Inverse OARSI-OMERACT Criteria
The researchers conducted knee-based analyses of data from the Osteoarthritis Initiative, a multicenter study in the United States. They used generalized linear mixed models to assess the relationship of the inverse OARSI-OMERACT criteria, at 12-month and 36-month visits, with worsening radiographic severity (any increase in KL grade from 12 months to 36 months) and decline in self-selected 20-minute walking speed (from 12 months to 36 months). A Cox model was used to assess time to knee replacement for the 6 years after the 36-month visit.
“The study focused on patients with symptomatic knee OA and was representative of the patient population we would typically see in a clinical setting or clinical trial,” says Dr. Driban. “For example, about 60% of participants were female and many were overweight or obese.”
Among the 1,746 knees analyzed, 19% met the inverse OARSI-OMERACT criteria for clinically meaningful worsening of KOA. “Patients who met the criteria for clinically meaningful worsening were almost twice as likely to experience slowed walking speed (odds ratio [OR], 1.89) and radiographic disease progression (OR, 1.89),” Dr. Driban notes. “Furthermore, a patient whose knee met the inverse OARSI-OMERACT criteria was more likely to receive a knee replacement after the 36-month visit (23%) compared with someone who did not experience clinically meaningful worsening (10%; hazard ratio, 2.54).” Results were similar when only including knees with KL grade 2 or greater.
The inverse OARSI-OMERACT criteria, Dr. Driban adds, accounts for multiple aspects of OA-related symptoms—specifically, self-reported knee pain, knee function, and global impact of OA (Figure).
A Valid Measurement of Clinically Meaningful Worsening of OA
“These findings indicate that the inverse OARSI-OMERACT criteria may be a valid way to identify clinically meaningful worsening of OA,” Dr. Driban says. “The percentage of knees that met the inverse OARSI-OMERACT criteria is comparable with prior reports of self-reported worsening. Therefore, the inverse OARSI-OMERACT responder criteria may be appropriate for use in epidemiological studies, as well as in longitudinal studies for which investigators are interested in defining changes in symptom states.”
Dr. Driban and colleagues would like to see future research monitor the validity of the inverse OARSI-OMERACT criteria as clinical practice changes. “Furthermore, it would be interesting to see how the inverse OARSI-OMERACT criteria perform compared with simply asking a patient if they feel like their treatment was a failure or if they feel worse than they did before,” Dr. Driban adds.