In this cross-sectional study, 197 patients were recruited at the time of enrolment on waiting lists for total knee replacement to measure pain, stiffness, function and HRQoL and to identify demographic, clinical, socioeconomic and psychosocial characteristics associated with these outcomes. We found that subjects reported important pain, stiffness and loss of function. HRQoL was also significantly impaired in these subjects, compared to the Canadian norms, for all the domains and components measured. These findings likely reflect the surgical indication of the subjects' condition. Interestingly however, mental aspects of HRQoL were also impaired. This finding may have implication regarding prehabilitation interventions for these patients, since it suggests that it might be beneficial to include mental health interventions to better help these patients. It is coherent with previous evidence that show that good mental health is a protective factor of functional decline in subjects suffering from knee osteoarthritis. [50
We believe that one very interesting finding of our study is that contralateral knee pain is associated with worse pain, stiffness, function and HRQoL related to the knee scheduled for replacement surgery. Although no studies have formally identified this kind of association, from a clinical point of view, it seems logical that patients suffering from both knees have a worse condition. Only one study by Merle-Vincent (2007) has looked at that specific factor but the authors did not find a significant association. [18
] Clearly, further research is needed to evaluate the effects of bilateral knee pain on patient's status and outcomes while waiting, as well as after knee replacement surgery. However, this phenomenon may have important clinical implication, as conservative treatment in patients waiting for knee replacement could realistically target both knees to maximize patients' status.
Psychological distress was low in this cohort of patients waiting for knee replacement surgery. Nonetheless, it was significantly associated with worse pain, stiffness, function and HRQoL. Other studies have outlined the important role of psychological distress on the health status of patients suffering from knee pain or undergoing knee replacement surgery. [14
] High BMI was also significantly associated with worse pain, function and HRQoL. Other studies have found that BMI is a risk factor for the incidence and progression of knee osteoarthritis [51
] and that it is also related to post-operative outcomes. [15
] In terms of treatment, weight-loss therapy and exercise have been found to be beneficial for this population and could be an important component of a prehabilitation program. [52
Low social support was significantly associated with worse mental HRQoL in our study. This finding is compatible with the results published by Ethgen et al. (2004), who found a significant association between social support and mental and physical aspects of HRQoL in subjects suffering from knee osteoarthritis. These authors recommended that physical health interventions should also add a social support component to improve health outcomes in these patients. [53
] In our study, subjects married or living in common-law had a better HRQoL compared to single, separated, divorced or widowed subjects. Further adjustment of this regression model with social support did not change the strength of the association between marital status and the role-physical component of HRQoL. Therefore, we believe that this association is more likely to be related to the help of the spouse on coping skills than to an effect of social support. [54
Contrary to what other studies found, sociodemographic factors were not related to pain, stiffness, function or HRQoL in our study. This may results from the fact that in this cohort of Canadian patients, access to surgery is equitable as it is not diminished in subjects of lower socioeconomic backgrounds nor that workers are fast tracked to see the surgeon. [14
Although the association between longer duration of symptoms and knee stiffness found in our study was small, it is unclear why subjects with longer duration would show less stiffness. Maybe it results from a response shift and possible adaptation to the chronic condition of knee osteoarthritis or arthritis. [56
] However this would potentially reflect also in the other sections of the WOMAC, an effect we did not observe.
Several factors were significantly associated with increased pain, stiffness, loss of function and loss of HRQoL in the study subjects. One of the strengths of our study is that many of these independent factors have a consistent effect across the scales of the WOMAC or the SF-36, which further supports the validity of our results. Other strengths include a high participation proportion (81.7%), thorough and relevant independent variables selection and no indication of selection bias (there were no significant differences between participants and eligible non participants on age and gender (data not shown). The regression models were adjusted for age and gender and further adjustments with other potential confounding factors only marginally changed the strength of the associations and were therefore not kept in the final models.
This study used a cross-sectional design and therefore caution is warranted when interpreting our results. Hypothesis related to the causality of the different independent variables on patients' health status need to be validated prospectively. Nonetheless, we believe our results provide valuable information regarding the patients' condition right at their enrollment on pre-surgery waiting list for knee replacement.
Another limitation was that the main outcome measures were self-reported and we did not include performance-based measures. The WOMAC and the SF-36 have been found to be valid instruments; still, it has been reported that performance-based measures provide distinct impressions of pain and function that complement self-reported measures. [57
] Therefore, the associations or strength of associations between patients' characteristics and performance-based measures could be different from the findings of our study. Although we found a significant association between low social support and worse mental HRQoL, the social support measure only reflected the size of the social network and not the two other components of social support. Therefore associations or strength of associations with the full validated social support measurement tool could be different. It is important to point out that this study focused on patients scheduled for primary unilateral knee replacement and excluded patients undergoing a revision or with a previous contralateral knee replacement or with a hip replacement, therefore results may differ for these patients.