• lupus erythematosus
• systemic
• autoimmune diseases
• immune complex diseases

Dear Sir,

In their letter,1 Dr Cui and colleagues include two interesting thoughts. Primarily, they argue, illustrated by one case, that gastrointestinal (GI) involvement may be more common in systemic lupus erythematosus (SLE) than usually thought. This thought is to some degree supported by the patient survey performed within the European League Against Rheumatism (EULAR)/ American College of Rheumatology (ACR) classification criteria project, in which we noted that more than 5% of the patients reported GI symptoms at the time of their SLE diagnosis,2 even though this organ system was not included on the questionnaire.

However, no specific SLE GI pattern has been described so far, which would be a prerequisite of including a GI manifestation into SLE classification criteria. A large number of rare organ manifestations possible in SLE could not be included into the classification criteria, since this would be impracticable. Throughout the classification criteria development process, SLE experts opined for a comprehensive system balanced against computational ease in the clinic.3 4 As in most instances, SLE affected multiple organs in the authors’ patient, easily allowing for classification according to the EULAR/ACR 2019 SLE classification criteria.5 6 On the other hand, in unusual situations of rare, isolated organ manifestations, failure to fulfil classification criteria should never prevent making a clinical diagnosis. While both try to correctly define whether a patient has SLE or not, diagnosis and classification are clearly distinct,7 and we once again caution against using the EULAR/ACR 2019 classification criteria for making or even worse refuting a diagnosis of SLE. As result of a very stringent methodological process, lack of sufficient homogeneity of GI manifestations did not qualify as a classification domain, but this does not argue against their clinical occurrence.

The other idea Dr Cui et al express is direct translation of points received in the EULAR/ACR classification system to the probability that the patient has SLE. This has not yet been tried, but the line of thought is correct in that the system in fact provides for a measure of the probability a patient can be classified as having SLE.