Research Article: Patient-reported outcome after rheumatoid arthritis-related surgery in the lower extremities

Date Published: April 24, 2012

Publisher: Informa Healthcare

Author(s): Anna Clara Benoni, Ann Bremander, Anna Nilsdotter.


Although decreasing with the development of effective pharmacological regimes, joint surgery has improved the function and quality of life of patients with rheumatoid arthritis (RA). Few studies have assessed patient-reported outcomes after RA surgery to the lower extremities. Here we report patient-relevant outcome after RA-related surgery based on the first data from the Swedish National Register of Rheuma Surgery (RAKIR).

258 RA patients (212 women) who had joint surgery performed at the Department of Orthopaedics, Spenshult Hospital between September 2007 and June 2009 were included. Mean age at surgery was 64 (20–86) years. The patients completed the SF-36 and HAQ questionnaires preoperatively and 6 months postoperatively, and 165 patients completed them after 12 months.

Improvement was seen as early as at 6 months. At 12 months, 165 patients (141 women)—including hip (n = 15), knee (n = 27), foot (n = 102), and ankle (n = 21) patients—reported statistically significant improvements from preoperatively to 12 months postoperatively in HAQ (mean change: –0.11) and SF-36 subscales physical function (11), role physical (12), bodily pain (13), social functioning (6.4), and role emotional (9.4). Hip and knee patients reported the greatest improvements.

Orthopedic RA-related surgery of the lower extremities has a strong effect on pain and physical function. Improvement is evident as early as 6 months postoperatively and remains after 12 months.

Partial Text

At the 12-month follow-up, responses were available from 165 patients (141 women) with a mean age at surgery of 64 (27–84) years (Table 2). The distribution of surgery was as follows: foot (n = 102), ankle (n = 21), knee (n = 27), and hip (n = 15). 93 patients did not complete the 12-month follow-up questionnaires. There were no statistically significant differences between responders and non-responders regarding age, preoperative HAQ score, or preoperative SF-36 score. Friedman’s test showed a difference in mean ranks for HAQ and for SF-36 subscales PF, RP, BP, and SF (Table 3). Patients reported statistically significant improvements in scores from before surgery to 12 months postoperatively in HAQ and in SF-36 subscales PF, RP, BP, SF, and RE. Statistically significant improvements were evident at 6 months (compared to preoperatively) in the same subscales, with the exception of RE. VT improved at 6 months but not at 12 months. When the subgroups were analyzed separately, foot patients reported significant improvement in SF-36 PF, RP, and BP scores at 6 and 12 months relative to preoperative levels (Table 4). Ankle patients had significant improvements in BP and SF compared to preoperatively (Table 5). Ankle patients who underwent total joint replacement (n = 10) had significant improvement in pain, and the improvement in PF was close to significant (p = 0.05) while ankle patients who underwent arthrodesis surgery (n = 11) only had significant improvement in SF. Knee patients (Table 6) had significant improvements in HAQ and in SF-36 PF, RP, BP, VT, SF, and RE at 12 months. Hip patients had significant improvements in HAQ and in SF-36 PF, BP, VT, SF, and RE at 12 months (Table 7).

Lower limb function has been shown to deteriorate more than upper limb function over a 10-year period in RA patients (Ringen et al. 2008). However, few studies have assessed patient-relevant outcome after surgery to the lower extremities. Osnes-Ringen et al. (2009) compared surgery in the upper and lower extremities and replacement surgery with non-replacement surgery in 255 patients with inflammatory arthropathies (two-thirds of whom had RA). As might be expected, the largest improvement at 12 months—as measured by the standard response mean (SRM)—was for pain reported from the specific joint that had been operated on. SF-36 subscales PF and BP also showed high SRMs, while SRMs for GH and HAQ were lower. In a study by March et al. (2008) assessing costs and outcomes of total joint replacement surgery for 42 Australian RA patients (31 patients undergoing total knee arthroplasty (TKA) and 11 patients undergoing total hip arthroplasty (THA)), the subjects were followed for 12 months postoperatively. The results were similar to ours, with statistically significant improvements in PF and BP for both knee and hip patients, while neither group improved regarding GH. Hip patients also showed better results in subscales related to other aspects of psychology (VT and MH). Similar results have been found for osteoarthritis patients (Bachmeier et al. 2001) and in RA drug trials (Mease et al. 2008, Wells et al. 2008), where aspects of patient psychology were less affected by pharmacological treatment. It has been proposed that the effect of successful RA treatment may first affect relief of pain, mobility, physical function, and physical role activities—and emotional well-being only later (Kosinski et al. 2002). In a follow-up study of osteoarthritis patients undergoing THA (n = 108) and TKA (n = 86) (Bachmeier et al. 2001), the SF-36 at 12 months was less responsive than the WOMAC, which is a disease-specific questionnaire for patients with osteoarthritis of the lower limb (Bellamy et al. 1988). The relative improvement was most pronounced for physical functioning and physical role functioning. Similarly to our data, the GH subscale after TKA remained unaltered.According to the authors this might be explained by the fact that—even after a successful operation—pain and restrictions in daily life remain, making patients rate their general health as insufficient. Moreover, the chronic course of RA can make evaluation of a specific surgical procedure and its effect on the patient more challenging to interpret (Nelissen 2003), which should be kept in mind when comparing RA surgery and surgery for osteoarthritis. In joint replacement studies including both osteoarthritis patients and RA patients (Espehaug et al. 1998, Bullens et al. 2001), RA patients were more satisfied than osteoarthritis patients after surgery. It was suggested that the moderate correlation between satisfaction and postoperative function could be partly explained by the different expectations of the patients preoperatively (Bullens et al. 2001). As for osteoarthritis patients undergoing knee arthroplasty, outcome of pain and function is best after 12 months (Nilsdotter et al. 2009) but major improvement is evident as early as 6 months postoperatively for RA patients. Improvement in pain has high priority in RA patients (Heiberg et al. 2005) and pain remains the most important indication for RA surgery. In our analysis, for every patient group there was a statistically significant improvement in the SF-36 subscale BP postoperatively compared to preoperatively.