Patient’s perspective of sustained remission in rheumatoid arthritis

Clinical remission has become a widely accepted treatment goal in rheumatoid arthritis (RA) patients with early disease. Reaching remission, however, is only the first step in a targeted treatment approach, and sustained remission (SR) is the desirable subsequent step. Previous studies have demonstrated that the progression of joint damage decreases with increasing duration of remission [1]; nonetheless, the ultimate target to drive treatment decisions in RA patients should be better long-term patient-reported outcomes [2].

In the last few decades, international efforts had been committed to defining remission in RA. Various definitions are available, with different levels of stringency [3‐5]. Ultimately, in RA patients, remission may be operationalized as either a complete absence of disease activity or a level of disease activity that is so low that it is not troublesome to the patient and portends a good prognosis [6]. This definition is particularly relevant given that during the course of RA, affected individuals face considerable physical, psychological and social changes that have profound negative effects on their health-related QoL (HRQoL) [7]. However, when remission is achieved, RA patients also achieved population norms in terms of HRQoL, unlike in the case of patients with other rheumatic diseases, such as patients with ANCA-associated vasculitides in remission, who exhibit substantially reduced patient-reported outcomes (PRO) [8].

Current RA management guidelines recommend incorporating patient-reported measures of functioning and quality of life into clinical trials as they are as effective as traditional physician- or laboratory-reported outcomes in reflecting long-term morbidity and mortality, easier to administer and less expensive than physician-observed health status measures [9‐11]. Moreover, clinicians and patients have different perspectives, and patients prioritize clinical outcomes that are not routinely measured [10, 12]. PROs include outcomes that are limited to a symptom (such as pain or fatigue) or those that evaluate complex constructs, such as patient’s function, disability or the overall disease status. One relevant aspect of the impact of SR on PROs is that it may differ for each particular outcome. In the present study, we sought to identify SR patients from an inception and ongoing cohort of RA patients with recent onset disease at cohort enrollment and examine the impact of SR in an ample spectrum of PROs.

The specific objectives were as follows:

Remission (sustained) has become the best clinical outcome in RA patients and may be approached as a state of no or minimal inflammatory disease activity associated with a halt of progression of joint damage [1], better PROs, working productivity and lower costs [24]. Today, it is a feasible target and the first recommendation from the “Treat-to-Target” approach [25]; nonetheless, patient’s perception of SR does not necessarily correspond to the physician construct of SR.

In the present study, we examined the impact of SR on PROs in a real clinical setting of an ongoing early arthritis clinic where completed, periodic and standardized assessments were performed by a dedicated rheumatologist. Two definitions of SR were used with a different degree of strictness [26] that allowed us to compare how each construct affected patient’s health related quality of life. The PRO examination included those adopted through consensus by the ACR, EULAR and Outcomes measures in Rheumatology (OMERACT) as well as fatigue, which was more recently added based on input from patient research partners [27, 28]. In addition, acute reactant-phase determinations and physician-derived measures were also included for additional (instead of a competitive) interpretation of the data.

We first found that SR was achieved in a substantial proportion of patients (i.e., 43% and 54%, depending on the definition used), and up to 35% met both definitions. In early disease, up to 60% of patients achieve clinical remission according to DAS28 [2, 29, 30]. We found a high percentage of B-SR and performed a 28-joint count, although it is recommended to include feet and ankle assessments when applying these criteria [4]. An extended joint count may favor the eventual detection of additional joints as either tender or swollen. Similar to previous publications, the highest percentage of SR patients was achieved with the DAS28 cut-off [8, 26, 29]. We additionally found that follow-up to first SR and time in SR favor DAS28-SR that was easier to achieve and maintain. DAS28 is considered to have lesser stringency than the Boolean definition of remission [29‐31]. The more stringent DAS28 cutoff value (2.4 instead of 2.6), allowed the presence of up to 12 swollen joints while the ESR was restricted to practically normal values; in addition, it is possible to have up to 5 swollen joints and still being in remission using the DAS28 criteria if the other variables are nearly normal [29]. Finally, DAS28 was also less stringent with respect to elevation of the global assessments of activity and pain as compared to an other validated index, the Simplified Disease Activity Index (SDAI) [32].

Second, at SR, patients from our study had PRO proxy to normal values. Linde et al. [8] found that the health-related quality of life scores from Danish RA patients in clinical remission approached that of the general population. By contrast, Radner et al. [24] showed that patients in remission, on average, did not reach normal health-related quality of life values (from HAQ, SF-36 physical component, EURO Qol-5D and Short Form-6D). Important considerations to be highlighted included that their patients had almost 12 years of disease duration with the consequent accrual of irreversible damage and disability [33] and their population was almost 25 years older than ours. Physical function declines with age [34], and increased age reduces HRQoL in aged RA patients [35]. Finally, in our study, SR (instead of time point remission) was defined, and it has been shown that physical function (as per HAQ) continues to improve over time if remission is reached and sustained [36].

Third, the percentage of patients with normal PRO varied from 97% for the HAQ, but decreased to approximately 50% for absence of fatigue and did not differ according to the SR definition used. Fatigue is a symptom that patients consider to be as severe as pain, and they frequently experience fatigue as incontrollable and overwhelming [37]. Among patients who achieved remission (or low disease activity), fatigue is frequent and may be experienced as a problem that affects quality of life [38‐40]. Additionally, a relationship between female sex and high levels of fatigue has been documented in a systematic review [41]. Our cohort was highly represented by females, which may explain the considerable prevalence of substantial fatigue in patients with SR, a number that was independent of the stringency of the SR definition used.

Fourth, we identified a particular temporal pattern of outcome normalization in patients who achieved SR, characterized by consecutive (rather than simultaneous) patient-centered followed by physician-centered, outcome normalization, with only mild differences between both SR definitions. Acute reactant phase normalization was detected very early during follow-up, followed by the simultaneous normalization of the majority of PRO, physician-dependent outcomes (as were swollen joint count and overall disease VAS) and SR construct. Moreover, 7 to 9 months extended between the normalization of acute reactant phases and the time point when the SR criteria are achieved. The results emphasize that the impact of SR on serologic disease activity, PRO and clinical disease activity do not necessarily occur in parallel, as previously highlighted by Altawil et al. [42]. In addition, SR according to DAS28 and the Boolean definition had a similar degree of stringency regarding their gain in most of the outcomes relevant to patients. In terms of fatigue and the physical component of the SF-36, it did behave similarly, but absence of fatigue and normalization of the physical component of the SF-36 were achieved earlier when the Boolean definition was used. The most strict definition translated into (2 specific) earlier improved patient-reported long-term outcomes. We do not have a clear explanation for the behavior of those outcomes. Nonetheless, Thiele et al. compared the functional ability of German RA patients in remission according to the DAS28, the SDAI and the Boolean definition of remission; they showed small differences between the 3 remission groups in swollen joint counts, moderate differences in acute phase reactants but large differences in PRO; patients achieving Boolean remission had a better patient global assessment as well as lower pain and fatigue scores than those in SDAI remission meanwhile for patients with DAS28 remission, disease duration, pain, fatigue and comorbid spine disease were associated with not achieving the other 2 sets of remission criteria [26].

The limitations of the study should be considered when interpreting our results. B and DAS-28 –SR’s definition include a patient VAS that can be high in patients with comorbid musculoskeletal conditions given that patients may not be able to distinguish between discomfort due to comorbid conditions and due to RA. Comorbid conditions were highly prevalent in our cohort [18]. We applied DAS28-SR using DAS28-ESR. An association between the ESR level and age [43] has been described, and elderly patients (3.2% of our cohort had ≥65 years of age) with true SR may have been misclassified as with some disease activity. Also, age (and gender) influence normative data of PROs (as SF-36 and HAQ); nonetheless, older age was underrepresented in our cohort and its impact on SR assessments is likely to be minimal. Fatigue was assessed using the fatigue short form 36 vitality subscale instead of a validated scale or an appropriated questionnaire primarily designed to assess fatigue [44]. Our cohort of patients is highly represented by females as previously described [17, 18] and data reported should not be generalized to RA males. Finally, additional confounding factors highly prevalent in RA patients [45] with confirmed strong associations with PRO, such as depression and anxiety, were not possible to control for given that these factors were not part of the original program in the patient’s follow-up.

In conclusion, awareness of the patient preferences is a salient premise for priorities in health care as it has been demonstrated that patients and health professionals differ in their perceptions of patients’ health status and need for care [46, 47]. Remission status is the most desirable outcome. For adoption by both patients and physicians, remission status should reflect symptom’s resolution and ideally prevent subsequent structural damage. PRO provide unique information that cannot be collected from a physician, especially regarding symptoms, and complete the picture of patients who live with RA in the clinical context of sustained remission.