A comparison and review of three sets of classification criteria for systemic lupus erythematosus for distinguishing systemic lupus erythematosus from pure mucocutaneous manifestations in the lupus disease spectrum

Abstract

Although the original purpose of the systemic lupus erythematosus (SLE) classification criteria was to distinguish SLE from other mimic diseases, and to facilitate sample selection in scientific research, they have become widely used as diagnostic criteria in clinical situations. It is not known yet if regarding classification criteria as diagnostic criteria, what problems might be encountered? This is the first study comparing the three sets of classification criteria for SLE, the 1997 American College of Rheumatology (ACR’97), 2012 Systemic Lupus International Collaborating Clinics (SLICC’12) and 2019 European League Against Rheumatism/American College of Rheumatology (EULAR/ACR’19), for their ability to distinguish patients with SLE from patients with pure mucocutaneous manifestations (isolated cutaneous lupus erythematosus without internal disease, i-CLE) in the lupus disease spectrum. 1,865 patients with SLE and 232 patients with i-CLE were recruited from a multicenter study. We found that, due to low specificity, none of the three criteria are adept at distinguishing patients with SLE from patients with i-CLE. SLICC’12 performed best among the original three criteria, but if a positive ANA was removed as an entry criterion, EULAR/ACR’19 would performed better. A review of previous studies that compared the three sets of criteria was presented in this work.

Diagnostic accuracy of the European League against rheumatism/American College of Rheumatology-2019 versus the Systemic Lupus International Collaborating Clinics-2012 versus the ACR-1997 classification criteria in adult systemic lupus erythematosus: A systematic review and meta-analysis

“…A total of nine original studies were included in our systematic review, including five retrospective cohort studies (including six pairs of study cohorts) (6,7,9,10,14) and four case-control studies (11)(12)(13)25); of them, three studies were conducted under the circumstances that the clinical diagnosis was prior to classification criteria scoring, causing a high risk of bias (10,12,14). The scoring of EULAR/ACR-2019 in derivative cohort by Aringer et al was obtained by rheumatologists without knowing its specific classification threshold, thus its bias risk is rated as "high" (6).…”

“…In a Chinese study by Teng et al, there were 9 patients only satisfying EULAR/ACR-2019, of which 4 were attributed to fever item; one with oral ulcers, leucopenia, hemolytic anemia, thrombocytopenia and antinuclear antibody can just be classified by SLICC-2012; on the contrary, another one with acute cutaneous lupus, photosensitivity, anti-dsDNA, ANA met ACR-1997 and EULAR/ACR-2019 but not SLICC-2012; notably, a total of 33 patients satisfying both EULAR/ACR-2019 and SLICC-2012 lost SLE classification using ACR-1997 (11). Jin et al pointed out 132 SLE patients with missed diagnosis by ACR-1997, with a high prevalence of renal disorder, non-scarring alopecia and hypocompletemia; no photosensitivity in SLICC-2012 was accountable for 8 SLE patients who lost SLE classification by SLICC-2012; there were 29 SLE patients not classified by EULAR/ACR-2019, of whom the proportion was higher in negative ANA, mucocutaneous lesions, joint involvement, renal disorder and hematological diseases (25).…”

Validation of the 2019 EULAR/ACR classification criteria for systemic lupus erythematosus in ANA-positive Chinese patients

“…Lee et al 17 showed similar specificity of all three SLE criteria in their Korean patients. Jin et al 21 studied 2097 Chinese patients with SLE and isolated cutaneous lupus. The sensitivity of the EULAR/ACR criteria for SLE was 96.8%, which was similar to the SLICC criteria (98.3%).…”

Performances of different classification criteria for systemic lupus erythematosus in a single-center cohort from Turkey

“…Previous studies that analyzed the performances of these three classification criteria in different populations have yielded various results. 1,2,8,11,14,[16][17][18][19][20][21][22][23][24][25][26] The sensitivities and specificities varied considerably in these studies. Sensitivities as low as 56% 21 and as high as 100% 14,21,25 and specificities as low as 56.5% 20 and as high as 100% 21,23 have been reported.…”

“…1,2,8,11,14,[16][17][18][19][20][21][22][23][24][25][26] The sensitivities and specificities varied considerably in these studies. Sensitivities as low as 56% 21 and as high as 100% 14,21,25 and specificities as low as 56.5% 20 and as high as 100% 21,23 have been reported. One possible reason for this variation is the different patient subgroups and different populations in which the studies were conducted.…”