eTable 1. Severity of Systemic Symptoms based on the BILAG2004 (British Isles Lupus Assessment Group’s activity index)
eTable 2. Data From Telephone Interviews for Patients Not Seen for >1 Year
Wieczorek IT, Propert KJ, Okawa J, Werth VP. Systemic Symptoms in the Progression of Cutaneous to Systemic Lupus Erythematosus. JAMA Dermatol. 2014;150(3):291-296. doi:10.1001/jamadermatol.2013.9026
Patients with cutaneous lupus erythematosus (CLE) who develop systemic lupus erythematosus (SLE) may have few and mild systemic symptoms.
To characterize the types and severity of systemic symptoms in a longitudinal cohort of patients with CLE.
Design, Setting, and Participants
Prospective, longitudinal cohort study of 77 patients with CLE who presented between January 2007 and April 2011 at a university autoimmune skin disease clinic.
Main Outcomes and Measures
Systemic symptoms and severity were determined from data recorded at each study visit and from medical records.
Of 77 patients with CLE, 13 (17%) went on to meet criteria for SLE, with a mean (SD) time from CLE diagnosis to SLE of 8.03 (6.20) years. Of the 13 patients, 1 (8%) solely met the mucocutaneous American College of Rheumatology (ACR) criteria of malar rash, discoid rash, photosensitivity, and oral ulcers, and 3 (23%) met the mucocutaneous ACR criteria plus positive antinuclear and other antibody titers. After a mean (SD) follow-up time of 2.81 (1.34) years, only 5 of the 13 patients with CLE (38%) who progressed to meet SLE criteria developed moderate to severe additional systemic disease.
Conclusions and Relevance
Patients with CLE who developed SLE during our study did so mostly by meeting the mucocutaneous ACR criteria, and the majority developed none to mild additional systemic disease during the study period. Thus, our study suggests that a small percentage of patients with CLE eventually develop SLE and that even if they do, most patients will have mild systemic disease.
Lupus erythematosus (LE) presents as a continuum, with some patients only demonstrating skin lesions, as in cutaneous LE (CLE), and others life-threatening manifestations of systemic LE (SLE).1 For the purpose of classification, the diagnosis of SLE is based on meeting 4 of 11 criteria established by the American College of Rheumatology (ACR) in 1982.2 The criteria are fulfilled when 4 of 11 are present at the same time or serially during a period of observation.
Studies have shown that 75% of patients with newly diagnosed CLE do not already have an SLE diagnosis.3 A subset of these patients go on to develop SLE at a later time, and there have been mixed reports on the percentage who experience a transition from CLE to SLE. Durosaro et al4 reported a 5% and 23% cumulative incidence of SLE in patients with newly diagnosed CLE at 5 and 25 years, respectively.4 Another population-based cohort study by Grönhagen et al5 in Sweden showed that the probability of developing SLE in the first 1 and 3 years after CLE diagnosis was 12.1% and 20.0%, respectively.5
In addition, because the 1982 ACR criteria include malar rash, discoid rash, photosensitivity, and oral ulcers, which are all cutaneous findings, patients with CLE who meet criteria for SLE may have disease limited to the skin.3 Note that we use the term rash in this article because malar and discoid rash are terms set by the ACR; these are more general terms than dermatologists use to describe such skin eruptions. Prior studies did not report the specific systemic symptoms experienced by patients with CLE that converted to SLE.4,5 There is also minimal data on the severity of systemic symptoms developed by patients who experience a transition from CLE to SLE. One study that compared inpatients with both SLE and subacute CLE (SCLE) with inpatients with SLE only found an increased prevalence of central nervous system disease, renal disease, arthritis, anemia, and pleuritis in the SLE-only group.6 Another study that matched outpatients with SCLE (63% with SLE) with those with SLE (but not SCLE) found an increased frequency of serositis and hematologic symptoms in the SLE-only group.7
Therefore, we propose to categorize which ACR criteria patients develop and identify the types and severity of systemic symptoms in patients who develop SLE after a CLE diagnosis.
To accomplish the study goals, we used our ongoing database of patients with LE and skin manifestations.8 We identified 202 patients with lupus-specific skin disease enrolled between January 5, 2007, and April 25, 2011. We excluded 67 patients who already had a diagnosis of SLE at their first visit; 44 patients who had only 1 study visit; 8 patients who were followed for less than 6 months; and 6 patients whose disease overlapped with dermatomyositis, systemic sclerosis, or limited cutaneous sclerosis. The protocol for the study was approved by the institutional review board of the University of Pennsylvania School of Medicine and was in accordance with the Declaration of Helsinki in its current form. All patients consented by means of signed institutional review board–approved informed consent and Health Insurance Portability and Accountability Act forms.
For the 77 eligible patients, we used the lupus database to download 3 documents, which are completed at every study visit. The first, a general form called the DRSA (which is not an acronym), contains information about date and type of CLE diagnosis, other skin manifestations, other diagnoses, the 11 ACR criteria, photosensitivity testing, and laboratory data including presence of antibodies, leukopenia, and proteinuria. An antinuclear antibody (ANA) titer of 1:80 was considered positive. The diagnosis of CLE was made by V.P.W., based on current or historical clinical and pathologic findings. The second document is the SLE Disease Activity Index (SLEDAI), which evaluates systemic disease.9 The third document is the CLE Disease Activity and Severity Index (CLASI), which scores the activity and damage of CLE on the skin and also reports on the presence of oral ulcers.10,11
From the DRSA, SLEDAI, and CLASI, it was possible to track which of the 11 ACR criteria patients had at the first visit or developed since the previous visit. If 2 sources of data were incongruent (arthritis was checked off in the SLEDAI but not the DRSA), the patient’s electronic medical chart was reviewed. If the arthritis, renal, hematologic, pleuritis, or neurologic boxes were checked in the DRSA, a medical chart review was conducted to better categorize the symptoms within each criteria. Each of these categories was evaluated for whether the symptom actually fulfilled ACR criteria as well as the severity of the systemic involvement.
For 5 patients (6%), data were also obtained from records of visits at the Hospital of the University of Pennsylvania with other dermatologists, rheumatologists, or primary care physicians. If a patient had seen more than 1 health care clinician, all notes were used as information to determine which SLE symptoms patients were having. We also contacted by telephone 16 patients (21%) who had not been seen at the Hospital of the University of Pennsylvania for more than 1 year.
A patient was considered to have experienced a transition to SLE if they had a CLE diagnosis at the first visit and went on to accrue 4 of the 11 ACR criteria during the follow-up time; we describe these patients as having “SLEc.” A history of a specific criterion still counted toward meeting 4 of the 11 criteria.
We determined the severity of all systemic symptoms present at the first visit or developed during the study period. Symptoms were divided into musculoskeletal, hematologic, renal, cardiopulmonary, and neurologic categories. We used a modified version of the British Isles Lupus Assessment Group’s activity index (BILAG2004), to classify the severity of systemic involvement.12 The BILAG2004 was developed and validated based on the likely treatment warranted by a host of symptoms, with the more severe grades representing clinical features that would lead to the prescription of medium to large doses of corticosteroids or immunosuppressive drugs and lower grades representing symptoms that would necessitate less potent therapies.12- 14 eTable 1 in the Supplement shows the severity breakdown that we used. Mild overall severity was classified as having none to mild involvement of all organ systems, while moderate to severe overall severity was defined as having at least 1 organ system classified as moderate to severe.
We used the Fisher exact test for binary outcomes and the Mann-Whitney rank sum test for continuous measurements. Because of the large number of statistical tests, we set the significance level at .005 to partially control for multiple comparisons.
Of the 77 patients eligible for this study, 64 received a diagnosis of CLE and remained with CLE only during the mean (SD) study period of 2.09 (1.33) years. Thirteen patients with CLE at baseline (17%) went on to meet criteria for SLE during the study period, with a mean (SD) follow-up time of 2.81 (1.34) years.
Table 1 contains baseline characteristics of the 64 patients who had CLE only during the study and the 13 patients who developed SLEc (see Table 1 for specific numbers). There was no significant difference in the time from CLE diagnosis to the first visit, total follow-up time, and number of women and white patients between the 2 groups.
Although a greater percentage of CLE-only patients had localized CLE (33% vs 8%) and a greater percentage of patients with SLEc had generalized CLE (31% vs 11%) compared with the other group, these differences were not statistically significant. There was also no significant difference in the distribution of chronic CLE and SCLE between the 2 groups.
In examining baseline ANA, anti–double-stranded DNA (anti-dsDNA), anti-Smith, anti-cardiolipin, and lupus anticoagulant titers between the 2 groups, we found a statistically significant difference (P = .002) in the distribution of baseline ANA titer, as patients with SLEc were more likely to have an ANA titer above 1:80 at the first visit compared with patients with CLE only.
At baseline, there were no significant differences in the use of topical or intralesional corticosteroids, oral prednisone, antimalarials or dapsone, immunomodulators like azathioprine, methotrexate, and mycophenolate mofetil, or thalidomide between patients with CLE only and those with SLEc.
Table 1 also contains the mean number of ACR criteria and number of patients with musculoskeletal, renal, and hematologic disease at baseline (no patients had cardiopulmonary or neurological involvement at first visit). There were no significant differences between the patients with CLE only and those with SLEc, although patients with SLEc had a greater number of ACR criteria at baseline (2.38 compared with 1.62 in the CLE-only group; P = .007).
The mean (SD) time from CLE diagnosis to SLE in the 13 patients was 8.03 (6.20) years, with all patients developing SLE more than 6 months after their CLE diagnosis. Nine patients developed 1 additional criterion, and 4 patients developed 2 criteria to go on to be called SLE. The criteria developed were as follows: hematologic involvement (n = 4 [31%]), arthritis (n = 3 [23%]), malar rash (n = 3 [23%]), renal involvement (n=2 [15%]), photosensitivity (n=2 [15%]), oral ulcers (n=2 [15%]), and discoid rash (n=1 [8%]).
Regarding the criteria that these 13 patients met at baseline or subsequently developed, 11 (92.3%) had photosensitivity, 9 (69%) had a positive ANA titer, 8 (62%) had discoid rash, 7 (54%) had oral ulcers, 6 (46%) had arthritis, 4 (31%) had malar rash, 4 (31%) had hematologic involvement, 3 (23%) had positive antibody titers, and 2 (15%) had renal criteria (Table 2). No patient with CLE that converted to SLEc had cardiopulmonary or neurologic involvement. Of the 13 patients, 1 (8%) solely met the ACR criteria because of malar rash, discoid rash, photosensitivity, and oral ulcers, and 3 (23%) met the ACR criteria for malar rash, discoid rash, oral ulcers, and photosensitivity and had positive ANA and other antibody titers (Table 2).
One patient with CLE only (2%) developed new mild musculoskeletal disease, and 3 patients with SLEc (23%) developed new moderate to severe musculoskeletal involvement during the follow-up period; this distribution was significantly different (P = .004) (Table 3). There were no patients with CLE only who developed new renal disease, while there were significantly more patients with SLEc with new renal involvement (P < .001), with 2 (15%) having mild and 2 (15%) moderate to severe disease. Five patients with CLE only (8%) developed new mild hematologic disease, and 4 patients with SLEc (31%) developed new mild hematologic involvement during the follow-up period, but this difference was not statistically significant (P = .04).
There was also a significant difference (P < .001) in the overall severity of new systemic symptoms between the 2 groups, as all 64 patients with CLE only had only mild new systemic involvement, and among the patients with SLEc, 8 (62%) had overall mild and 5 (38%) had overall moderate to severe additional systemic disease.
Of our 77 patients, 16 have not been seen at the University of Pennsylvania for more than 1 year. We called all 16 and were able to conduct telephone interviews with 8 of these patients (eTable 2 in the Supplement).
In this study we followed 77 patients with CLE and observed that 64 patients remained with CLE only and 13 (17%) met SLE criteria during the follow-up period, with a mean time from CLE diagnosis to SLE of 8.03 years. These findings are similar to the study by Durosaro et al,4 which showed that 12% of their patients with CLE had experienced conversion to SLE, with a mean progression time of 8.2 years. However, it is challenging to make a direct comparison with these numbers because in the study by Durosaro et al,4 patients were enrolled at the time of CLE diagnosis and were followed for 40 years, while in our study, patients were enrolled after having their CLE for a mean of 6.62 years and were only followed for a short period.
Comparison of the CLE-only and SLEc groups at first visit yielded several notable points. First, baseline ANA titer was the only characteristic that was significantly different in distribution; those with CLE only were more likely to have a negative ANA titer at baseline (and less likely to have ANA titers from 1:80-1:1280) than those with SLEc. Although not significant, a greater percentage of women comprised those with SLEc than CLE only. While our data does not show that women with CLE have a higher risk of developing SLE, Grönhagen et al3 reported that the risk of being diagnosed as having SLE after a CLE diagnosis is twice as high for women as for men.
In addition, more patients with SLEc had generalized discoid lupus erythematosus compared with patients with CLE only, although this was not statistically significant. Other studies have shown that generalized discoid lupus erythematosus and SCLE have a higher association with SLE than other forms of CLE.11 Our sample size was likely too small to show a significant difference. Furthermore, 2 patients with SLEc (15%) had moderate to severe systemic disease at baseline, compared with none in the CLE-only group. Although this difference was not statistically significant (P = .03), again, it may be that we do not have enough power to show that having baseline moderate to severe systemic symptoms are a predictor for meeting SLE criteria in the future.
Most importantly, our patients with CLE who met ACR criteria for SLE did so mainly by mucocutaneous criteria. Almost 25% did not have any other criteria aside from photosensitivity, malar rash, discoid rash, oral ulcers, and positive ANA and other antibody titers. In addition, although a significantly greater number of patients with SLEc developed new musculoskeletal, renal, and overall moderate to severe systemic symptoms compared with their CLE-only counterparts, when we followed these patients with SLEc, only 38% were considered to have developed new moderate to severe systemic disease.
There are several limitations to our study. First, our observation period was short, and patients could develop systemic symptoms in the future that mirror those of an SLE-only population. Second, 21% of our patients were lost to follow-up, and the return on telephone interviews conducted was 50%. However, the interview data with 6 patients with CLE only suggests that patients who were lost to follow-up did not experience more activity or progression of their disease. A third limitation is that many patients were not enrolled at the time of diagnosis of CLE or SLE, so that in these cases the date of diagnosis was based on self-report. Lastly, we could not assess the severity of symptoms that patients experienced prior to enrollment.
This study reveals that patients with CLE who developed SLE during our study did so mostly by meeting the mucocutaneous ACR criteria, and the majority developed none to mild additional systemic disease during the study period. Clinicians can use these data to counsel patients with newly diagnosed CLE that few will go on to develop SLE, and that even if they do, they are likely to have mild systemic symptoms. These data are also important in conducting clinical studies where patients with CLE are automatically excluded for having a diagnosis of SLE; we show it is possible that these patients have mild, if any, systemic symptoms. Our study has also shown that the presence of moderate to severe systemic symptoms at baseline could be predictive of going on to meet SLE criteria, but future investigations should better identify patient characteristics that predict development of SLE in the hope of adjusting monitoring based on the risk of progression to SLE.
Corresponding Author: Victoria P. Werth, MD, Department of Dermatology, Hospital of the University of Pennsylvania, 3600 Spruce St, 2 Rhoads Pavilion, Philadelphia, PA 19104 (email@example.com).
Accepted for Publication: October 9, 2013.
Published Online: January 29, 2014. doi:10.1001/jamadermatol.2013.9026.
Author Contributions: Drs Wieczorek and Werth had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Wieczorek, Werth.
Acquisition of data: Wieczorek, Okawa, Werth.
Analysis and interpretation of data: Wieczorek, Propert, Werth.
Drafting of the manuscript: Wieczorek.
Critical revision of the manuscript for important intellectual content: Wieczorek, Propert, Okawa, Werth.
Statistical Analysis: Wieczorek, Propert.
Obtained funding: Werth.
Administrative, technical, and material support: Okawa, Werth.
Study supervision: Werth.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported in part by the Department of Veterans Affairs (Veterans Health Administration, Office of Research and Development, Biomedical Laboratory Research and Development) and by the National Institutes of Health (grant NIH K24-AR 02207) (Dr Werth); and The Lupus Foundation of America, Philadelphia Tri-State Chapter (Dr Wieczorek).
Role of the Sponsor: The funding agencies had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: Aileen Chang, MD, Elizabeth Ghazi, MD, and Saroj Verma, MD, collected data and provided feedback.
Additional Information: The copyright for the CLASI (CLE Disease Activity and Severity Index), a data source used in this study, is owned by the University of Pennsylvania.