Abstract
Introduction
Choosing between the different indices of activity evaluation in RA is often difficult considering the very heterogeneous clinical expression of the disease. The objective of our study was to evaluate the level of similarity between SDAI, CDAI, DAS28- ESR and DAS28- CRP indices in the evaluation of RA activity.
Patients and methods
In this transversal study, a total of 100 patients with RA responding to the ACR 87 criteria were followed up for a period of 20 months. The correlations between the four indices were studied through the Pearson’s correlation coefficient ( r ). The similarity between these tools was evaluated through Kendall’s (K) “tau” similarity coefficient.
Results
The 87 female and 13 male patients (sex ratio: 6.7F/1M) were of a mean age of 52.9 ± 11.6 years (17–77) and have been living with RA for a mean of 8.3 ± 9 years (2 months–41 years). The DAS28- ESR mean score was 5.53 ± 1.46 [1.25–8.05]. The DAS28- CRP mean score was 5.01 ± 1.44 [1.68–7.81]. The CDAI mean score was 30.72 ± 15.04 [2–62] and that of SDAI was 28.86 ± 15.86 [2.3–71.3]. A positive, statistically significant correlation was noted between the four indices of RA activity. The level of similarity between the different indices was good (K variation between 0.758 and 0.943). DAS28- ESR allowed classifying the patients in the same category of disease activity than DAS28- CRP in a proportion of 85%. This proportion was 88% when comparing DAS28- ESR to CDAI and SDAI, respectively. Regarding DAS28- CRP and CDAI, these two indices classified the patients in the same category in a proportion of 80%, compared to 87% regarding DAS28- CRP and SDAI. Finally, CDAI and SDAI classified the patients in the same category in a proportion of 92% with an excellent level of similarity.
Conclusion
Different evaluation indices of RA activity are currently available. DAS28 is the most used. CDAI and especially SDAI have a good level of similarity with DAS28. Their advantage is to be simple and quick, and seem therefore well adapted to the follow-up of outpatients.
Résumé
Introduction
Le choix entre les indices d’évaluation de l’activité de la PR est souvent difficile vu l’expression clinique très hétérogène de la maladie. L’objectif de notre étude était d’évaluer le niveau de concordance des indices SDAI, CDAI, DAS28-VS et DAS28 -CRP dans l’évaluation de l’activité de la PR.
Patients et méthodes
Il s’agit d’une étude transversale portant sur 100 patients atteints de PR répondant aux critères de l’ACR 87, suivis sur une période de 20 mois. Les corrélations entre les quatre indices ont été étudiées par le coefficient de corrélation ( r ) de Pearson. La concordance entre ces outils a été évaluée par le coefficient de concordance « tau » de Kendall (K).
Résultats
Il s’agit de 87 femmes et 13 hommes (sexe ratio : 6,7F/1H), âgés en moyenne de 52,9 ± 11,6 ans (17–77). La PR évoluait en moyenne depuis 8,3 ± 9 ans (deux mois–41 ans). Le score moyen du DAS28-VS était de 5,53 ± 1,46 [1,25–8,05]. Le score moyen du DAS28-CRP était de 5,01 ± 1,44 [1,68–7,81]. Le score moyen du CDAI était de 30,72 ± 15,04 [2–62] et celui du SDAI était de 28,86 ± 15,86 [2,3–71,3]. Une corrélation positive statistiquement significative a été notée entre les quatre indices d’activité de la PR. Le niveau de concordance entre les différents indices était bon (K varie entre 0,758 et 0,943). Le DAS28-VS permettait de classer les patients dans la même rubrique d’activité de la maladie que le DAS28-CRP dans une proportion de 85 %. Cette proportion était de 88 % en comparant le DAS 28 VS respectivement au CDAI et au SDAI. Quant au DAS28-CRP et au CDAI, ces deux indices classaient les patients dans la même rubrique dans 80 % des cas alors que le DAS28-CRP et le SDAI le permettaient dans 87 % des cas. Enfin le CDAI et le SDAI classaient les patients dans la même rubrique dans 92 % des cas avec un excellent niveau de concordance.
Conclusion
Différents scores d’évaluation de l’activité de la PR sont actuellement disponibles. Le DAS28 reste l’indice le plus utilisé. Le CDAI et surtout le SDAI ont un bon niveau de concordance avec le DAS28. Ils ont l’avantage d’être simples et rapides et semblent ainsi bien adaptés au suivi des patients en ambulatoire.
1
English version
1.1
Introduction
Evaluating the activity of rheumatoid arthritis (RA) is fundamental for defining its progression properties at one time point as well as for the patients’ follow-up and the evaluation of therapeutic response. Several tools are currently proposed for the evaluation of RA activity. Some of them are in use in clinical trials, like the ACR 20, 50 and 70 criteria and those of the FDA . Others are more adapted to daily practice, like the Disease Activity Score (DAS28) , the Simplified Disease Activity Index (SDAI) and the Clinical Disease Activity Index (CDAI) ( Appendices 1–4 ). The choice between these indices is actually not supported by objective criteria, but is rather depending on the physicians’ affinities. The choice is made even more difficult given the very heterogeneous expression of the patients’ clinical expression of RA.
The objective of our study was to evaluate the level of similarity between the SDAI, CDAI, DAS28 score indices calculated with the sedimentation rate (DAS28- ESR ) and DAS28 calculated with the C-reactive protein (DAS28- CRP ) in the evaluation of RA activity.
1.2
Patients and methods
In this transversal study, a total of 100 patients with RA fulfilling the ACR 87 criteria were followed-up in the rheumatology department of the Institute Mohamed Kassab (Manouba, Tunisia) over a period of 20 months (January 2008–February 2010). A form was established to record the different parameters necessary for the calculation of the disease activity indices. The different recorded data were analyzed with SPSS, release 13.0, for Windows. The correlations between the four indices were studied through Pearson’s correlation coefficient ( r ). The similarity between these tools was evaluated through the Kendall (K) similarity coefficient “tau”. The significance cut-off value ( P ) was fixed to 0.05.
1.3
Results
The 87 female and 13 male patients (sex ratio: 6,7F/1M) were of a mean age of 52.9 ± 11.6 years (17–77) and have been living with RA for a mean of 8.3 ± 9 years (2 months–41 years). The rheumatoid factor (RF) and the anti-cyclic citrullinated peptide (anti-CCP) antibodies were positive in respectively 72% and 51.1% of patients. RA was erosive in 80% of patients. The number of painful joints (NPJ) was of a mean of 12.6 ± 9.35 (0–28) and that of swollen joints (NSJ) of a mean of 4.58 ± 4.65 (0–24). The mean joint pain evaluation by the patient on the visual analogue scale (VAS-patient) was of 62.25 mm ± 24.73 (0–100), and mean global disease evaluation by the physician (VAS-physician) was of 54.6 mm ± 22.03 (10–95). Mean ESR and CRP were of 39.45 mm ± 25.73 (1–130) and 18.6 mg/L ± 25.48 (3–179), respectively.
The mean calculation time of DAS28- ESR , DAS28- CRP , SDAI and CDAI was of 1 minute, regardless of the clinical and biological data collection time. The results of these four activity scores are presented in Table 1 .
| Activity scores | Mean ± DS | Activity level | |||
|---|---|---|---|---|---|
| Remission | Low | Moderate | High | ||
| DAS28-ESR | 5.53 ± 1.46 [1.25–8.05] | < 2.6 n = 5 | ≤ 3.2 n = 2 | > 3.2 et ≤ 5.1 n = 30 | > 5.1 n = 63 |
| DAS28-CRP | 5.01 ± 1.44 [1.68–7.81] | < 2.6 n = 8 | ≤ 3.2 n = 4 | > 3.2 et ≤ 5.1 n = 31 | > 5.1 n = 57 |
| CDAI | 30.72 ± 15.04 [2–62] | ≤ 2.8 n = 2 | ≤ 10 n = 10 | > 10 et ≤ 22 n = 21 | > 22 n = 67 |
| SDAI | 28.86 ± 15.86 [2.3–71.3] | ≤ 3.3 n = 3 | ≤ 11 n = 9 | > 11 et ≤ 26 n = 28 | > 26 n = 60 |
A positive, statistically significant correlation was noted between the four activity indices of RA ( Table 2 ). The similarity level between the different indices was good (K between 0.758 and 0. 943). SDAI presented the best level of similarity with the other activity indices. However, DAS28- ESR and DAS28- CRP presented a better level of similarity between themselves than with SDAI ( Table 2 ).
| DAS28-ESR | DAS28-CRP | CDAI | SDAI | |
|---|---|---|---|---|
| DAS28-ESR | r : 0.932 ** K: 0.781 ** | r : 0.901 ** K:0.758 ** | r : 0.897 ** K: 0.760 ** | |
| DAS28-CRP | r : 0.932 ** K: 0.781 ** | r : 0.944 ** K: 0.844 ** | r : 0.957 ** K: 0.885 ** | |
| CDAI | r : 0.901 ** K: 0.758 ** | r : 0.944 ** K: 0.844 ** | r : 0.988 ** K: 0.943 ** | |
| SDAI | r : 0.897 ** K: 0.760 ** | r : 0.957 ** K: 0.885 ** | r : 0.988 ** K: 0.943 ** |
The analysis of the indices similarity according to their definition of activity level (remission, low, moderate or high activity) revealed a very good similarity level between the different indices (K between 0.724 and 0.849) ( Tables 3a and b ). Likewise, SDAI presented the best level of similarity with the other RA activity indices. DAS28- ESR allowed classifying the patients in the same category of disease activity than DAS28- CRP in a proportion of 85%. This proportion was of 88% when comparing DAS28- ESR respectively to CDAI and SDAI. As regards DAS28- CRP and CDAI, these two indices entered the patients in the same category for up to 80%, whereas DAS28- CRP and SDAI allowed it in 87% of patients. Finally, CDAI and SDAI entered the patients in the same category for up to 92% with an excellent level of similarity.
| Scores | DAS28-CRP | CDAI | SDAI | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Activity level | Remission | Low | Moderate | High | Remission | Low | Moderate | High | Remission | Low | Moderate | High | |
| DAS28-ESR | |||||||||||||
| Remission | 5 | 0 | 0 | 0 | 2 | 3 | 0 | 0 | 3 | 2 | 0 | 0 | |
| Low | 2 | 0 | 0 | 0 | 0 | 2 | 0 | 0 | 0 | 2 | 0 | 0 | |
| Moderate | 1 | 4 | 24 | 1 | 0 | 5 | 21 | 4 | 0 | 5 | 24 | 1 | |
| High | 0 | 0 | 7 | 56 | 0 | 0 | 0 | 63 | 0 | 0 | 4 | 59 | |
| DAS28-CRP | Remission | 2 | 6 | 0 | 0 | 3 | 5 | 0 | 0 | ||||
| Low | 0 | 2 | 2 | 0 | 0 | 2 | 2 | 0 | |||||
| Moderate | 0 | 2 | 19 | 10 | 0 | 2 | 25 | 4 | |||||
| High | 0 | 0 | 0 | 57 | 0 | 0 | 1 | 56 | |||||
| CDAI | Remission | 2 | 0 | 0 | 0 | ||||||||
| Low | 1 | 9 | 0 | 0 | |||||||||
| Moderate | 0 | 0 | 21 | 0 | |||||||||
| High | 0 | 0 | 7 | 60 | |||||||||
| DAS28-ESR | DAS28-CRP | CDAI | SDAI | |
|---|---|---|---|---|
| DAS28-ESR | X 2 : 153.1 ** K: 0.724 ** | X 2 : 151 ** K: 0.766 ** | X 2 : 161.9 ** K: 0.776 ** | |
| DAS28-CRP | X 2 : 153.1 ** K: 0.724 ** | X 2 : 152.9 ** K: 0.635 ** | X 2 : 131.3 ** K: 0.753 ** | |
| CDAI | X 2 : 151 ** K: 0.766 ** | X 2 : 131.3 ** K: 0.635 ** | X 2 : 227.1 ** K: 0.849 ** | |
| SDAI | X 2 : 161.9 ** K: 0.776 ** | X 2 : 152.9 ** K: 0.753 ** | X 2 : 227.1 ** K: 0.849 ** |
1.4
Discussion
The therapeutic management of RA has seen an important evolution over the last years. Multiple molecules are currently available, allowing good therapeutic responses, and even remissions. As a consequence, an objective evaluation of RA activity is essential. Several RA activity indices have therefore been worked out and made available to us, making it not always easy to choose among this vast range . The majority of them have indeed been validated by different teams and present comparable metrologic properties . The acceptability is good for the different indices since they globally use the same parameters, with the use of a special calculator for DAS28.
Only few studies have compared directly two or several of these indices. In accordance with data from literature, we note the good level of similarity between DAS28, CDAI and SDAI with K > 0.70 and a proportion of about 80% of the subjects classified in the same category of RA activity ( Table 4 ) . The only problem was the definition of remission with the different indices where the level of similarity was less obvious (K between 0.48 and 0.58) . Indeed, DAS28 seems to be more permissive in its definition of remission . Our study confirms these data. So, among the 12 patients with a low disease activity, DAS28- CRP classified 66% of them in remission, vs 41% with DAS28- ESR , 25% with SDAI, and 16% with CDAI ( Table 3a ). Actually, SDAI and CDAI take into account, like DAS28, the number of painful and swollen joints and global VAS-patient, with in addition the global VAS-physician. This latest parameter seems of importance in the objective evaluation of RA activity and allows getting back the patients considered as in remission by DAS28.
| Study | Activity scores | Results |
|---|---|---|
| Aletaha et al. | CDAI/DAS28 | K = 0.70 |
| Aletaha et al. | SDAI/DAS28 | K = 0.70 |
| Ranganath et al. | SDAI/DAS28 CDAI/DAS28 SDAI/DAS CDAI/DAS | K = 0.80 K = 0.74 K = 0.80 K = 0.74 |
| Khanna et al. | SDAI/DAS28 CDAI/DAS28 CDAI/SDAI | Remission: K = 0.48 Low activity: K = 0.68 Remission: K = 0.52 Low activity: K = 0.67 Remission: K = 0.97 |
| Mierau et al. | SDAI/DAS28 CDAI/DAS28 | Remission: K = 0.63 Remission: K = 0.58 |
| Shaver et al. | CDAI/DAS28 | Remission DAS28 = 28.5 % Remission CDAI = 6.5% |
| Our study | SDAI/DAS28-ESR SDAI/DAS28-CRP CDAI/DAS28-ESR CDAI/DAS28-CRP | K = 0.76 K = 0.88 K = 0.75 K = 0.84 |
Concerning DAS28- ESR and DAS28- CRP , a good correlation was noted between these two scores. However, the cut-off values defining the activity levels, and especially remission, do not appear to be completely superimposable and are probably higher for DAS28- CRP . CRP indeed gets back to normal quicker than ESR. On the other hand, ESR can be accelerated outside RA attacks, especially in the frequently associated gammaglobulinopathy or Sjögren syndrome .
Furthermore, a meta-analysis conducted on the metrologic properties of these indices has shown that DAS, DAS28, SDAI and CDAI have a good converging validity with good to moderate correlations with HAQ and with the structural effects . It has also demonstrated that the discriminative abilities of these indices were satisfying for the ACR criteria-related therapeutic change or remission . The sensitivity to change was evaluated for SDAI and CDAI in two studies . These two indices detected a difference between the ACR responders and non-responders with a good level of sensitivity. Only one study has directly compared the intra-observer reproducibility of DAS28, SDAI and CDAI. It has found intra-class correlation coefficients between 0.85 and 0.89, highlighting a good reproducibility .
Otherwise, SDAI and CDAI allow a simple calculation of RA activity, without mathematical formulations, whereas DAS and DAS28 calculation is obtained through complex equations and often need the use of special calculators .
These four indices present however common limits. Indeed, they don’t take into account some affections like persisting synovitis in the foot and the ankle, frequently observed in RA . On the other hand, these indices can be inappropriately increased in patients with concomitant fibromyalgia, often having 28 painful joints .
1.5
Conclusion
Different evaluation scores of RA activity are currently available. DAS28 is the most frequently used. CDAI, and especially SDAI, have a good level of similarity with DAS28. Their advantage is to be simple and quick, and they appear therefore well adapted to the follow-up of outpatients. The levels of low and high activity seem to be comparable between the different indices. However, the definition of remission is much stricter with SDAI and CDAI which appear therefore better recommended in clinical trials.
Disclosure of interest
The authors declare that they have no conflicts of interest concerning this article.
2
Version française
2.1
Introduction
L’évaluation de l’activité de la polyarthrite rhumatoïde (PR) est fondamentale aussi bien pour définir son évolutivité à un moment donné que pour le suivi des patients et l’évaluation de la réponse thérapeutique. Plusieurs outils sont actuellement proposés pour évaluer l’activité de la PR. Certains sont utilisés dans les essais cliniques comme les critères ACR 20, 50 et 70 et ceux de la FDA . D’autres sont plus adaptés à la pratique courante, tels que le Disease Activity Score (DAS28) , le Score Disease Activity Index (SDAI) et le Clinical Disease Activity Index (CDAI) ( Annexes 1–4 ). En réalité, le choix entre ces indices ne repose pas sur des critères objectifs mais dépend plutôt des affinités des praticiens. Il est d’autant plus difficile que les patients sont très hétérogènes quant à l’expression clinique de leur PR.
L’objectif de notre étude était d’évaluer le niveau de concordance des scores des indices SDAI, CDAI, DAS28 calculé avec la vitesse de sédimentation (DAS28 VS ) et DAS28 calculé avec la C-réactive protéine (DAS28 CRP ) dans l’évaluation de l’activité de la PR.
2.2
Patients et méthodes
Il s’agit d’une étude transversale portant sur 100 patients atteints de PR répondant aux critères de l’ACR 87, suivis dans le service de rhumatologie de l’Institut Mohamed Kassab (Manouba, Tunisie) sur une période de 20 mois (janvier 2008–février 2010). Une fiche a été établie pour le recueil des différents paramètres nécessaires pour le calcul des indices d’activité de la maladie. Les différentes données recueillies ont été analysées au moyen du logiciel SPSS version 13,0 pour Windows. Les corrélations entre les quatre indices ont été étudiées par le coefficient de corrélation ( r ) de Pearson. La concordance entre ces outils a été évaluée par le coefficient de concordance « tau » de Kendall (K). Le seuil de signification ( p ) a été fixé à 0,05.
2.3
Résultats
Il s’agit de 87 femmes et 13 hommes (sex-ratio : 6,7F/1H), âgés en moyenne de 52,9 ± 11,6 ans (17–77). La PR évoluait en moyenne depuis 8,3 ± 9 ans (deux mois–41 ans). Le facteur rhumatoïde (FR) et les anticorps anti-peptides cycliques citrullinés (ACPA) étaient positifs dans respectivement 72 % et 51,1 % des cas. La PR était érosive dans 80 % des cas. Le nombre d’articulations douloureuses (NAD) était en moyenne de 12,6 ± 9,35 (0–28) et celui des articulations gonflées (NAG) était en moyenne de 4,58 ± 4,65 (0–24). L’évaluation de la douleur articulaire sur l’échelle visuelle analogique par le patient (Eva-patient) était en moyenne de 62,25 mm ± 24,73 (0–100) et l’évaluation globale de la maladie par le médecin (Eva-médecin) était en moyenne de 54,6 mm ± 22,03 (10–95). La VS et la CRP moyennes étaient respectivement de 39,45 mm ± 25,73 (1–130) et 18,6 mg/L ± 25,48 (3–179).
Le temps moyen de calcul du DAS28 VS , DAS28 CRP , du SDAI et du CDAI était d’une minute, sans tenir compte du temps de recueil des données cliniques et biologiques. Les résultats de ces quatre scores d’activité sont représentés dans le Tableau 1 .
| Indices | Moyennes | Niveau d’activité | |||
|---|---|---|---|---|---|
| Rémission | Faible | Modérée | Forte | ||
| DAS28-VS | 5,53 ± 1,46 [1,25–8,05] | < 2,6 n = 5 | ≤ 3,2 n = 2 | > 3,2 et ≤ 5,1 n = 30 | > 5,1 n = 63 |
| DAS28-CRP | 5,01 ± 1,44 [1,68–7,81] | < 2,6 n = 8 | ≤ 3,2 n = 4 | > 3,2 et ≤ 5,1 n = 31 | > 5,1 n = 57 |
| CDAI | 30,72 ± 15,04 [2–62] | ≤ 2,8 n = 2 | ≤ 10 n = 10 | > 10 et ≤ 22 n = 21 | > 22 n = 67 |
| SDAI | 28,86 ± 15,86 [2,3–71,3] | ≤ 3,3 n = 3 | ≤ 11 n = 9 | > 11 et ≤ 26 n = 28 | > 26 n = 60 |
Une corrélation positive statistiquement significative a été notée entre les quatre indices d’activité de la PR ( Tableau 2 ). Le niveau de concordance entre les différents indices était bon (K varie entre 0,758 et 0,943). Le SDAI présentait le meilleur niveau de concordance avec les autres indices d’activité. Toutefois, le DAS28 VS et le DAS28 CRP présentaient un meilleur niveau de concordance entre eux qu’avec le SDAI ( Tableau 2 ).
Related posts:
Influence of different types of sockets on the range of motion of the hip joint by the transfemoral amputee
Rehabilitation in patients with peripheral arterial disease
Dependent patients discharged home from PRM departments: Relevant indicators
Analyse de livre
Impact of a rehabilitation program on muscular strength and endurance in peripheral arterial occlusive disease patients
Rehabilitation in Women with Breast Cancer
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
