Clinical features of systemic lupus erythematosus in Sri Lankan patients : results from a lupus clinic

Objectives To find the common clinical features, pattern of visceral involvement, treatment received and outcome in patients diagnosed as having systemic lupus erythe­ matosus (SLE) on American Rheumatological Association (ARA) criteria. Setting Clinic for patients referred or admitted to the University Medical Unit, National Hospital of Sri Lanka, Colombo, with diagnosed or suspected SLE. Design and methods A prospective descriptive study. Clinical features of patients collected at time of registra­ tion in the clinic were maintained in a database. Patients were followed up prospectively and changes recorded. Data were analysed after 3 years of follow up. Results Of the 111 patients registered during this period, 96 (86%) were clinically diagnosed as having SLE. Of these, 77 patients (80%) satisfied ARA criteria for diagnosis of SLE. 72 were females (93%). The mean age of patients who satisfied the ARA criteria was 32 years (range 11 to 58), and the mean duration of disease 7 years (range 1 to 15). The commonest presentation was with mucocutaneous features (98%) and alopecia in 87%. Systemic features were found in 92% of patients. 67 (87%) of patients had visceral involvement with 60 (78%) having it at time of diagnosis. 53 (69%) had renal, 42 (54%) haematological, 33 (42%) neu­ rological, 12 (16%) cardiac and 8 patients pulmonary in­ volvement. Five patients died during the 3-year follow up and 2 developed chronic renal failure. Three patients un­ derwent successful pregnancy after diagnosis of SLE. Conclusions Our study confirmed the wide variability of clinical features seen in SLE. Alopecia and visceral involve­ ment were common in Sri Lankan patients.


Introduction
SLE is a multi-system autoimmune disease where the incidence, prevalence and the clinical features are known to vary with factors such as race, gender, ethnicity, age and country of birth (1,2).The disease occurs world-wide, and is most commonly found among women of child-bear ing age (2).
Publications from different centres have shown marked variation in the clinical features of SLE among dif ferent races (2), and several reports have noted a differ ence in the clinical features of SLE between Asians and other subjects (3,4,5,6.).
There are no published data on clinical features of patients with SLE in Sri Lanka.We have analysed the clini cal features of SLE diagnosed on ARA criteria in patients attending our lupus clinic.

Methods
We did a descriptive analysis of data in SLE patients prospectively in the University Lupus Research clinic which was started in 1996 with the objective of providing better care and follow up for patients with SLE.Detailed data on clinical features, investigations, complications and treatment of these patients were entered in a database.The ARA classification criteria were used for the diagno sis of SLE.Those not satisfying the diagnostic criteria were further investigated for other autoimmune diseases and followed up.

Results
The results are summarised in Tables 1 to 7. The demo graphic data of patients are shown in Table 1.Of the 77 patients satisfying the ARA criteria for diagnosis, 72 (93%) were female.There were 80% Sinhalese, 13% Moor and 5% Tamils.There were patients from all 9 provinces of Sri Lanka: 44% were from Western province, 19% from Southern and 13% from North-Western.The commonest presentation was with mucocutane ous features (98%) which included alopecia, malar rash, discoid rash, photosensivity, oral ulcers, vasculitis and Raynaud's phenomenon (Table 2).A majority (92%) also had systemic features such as fever, loss of appetite and weight loss, and 85% had musculoskeletal features such as arthralgia or arthritis: 67 patients (87%) had visceral in volvement, and 49 (80%) had visceral involvement at diag nosis.
There were 53 patients (69%) with renal involvement.The renal histology was available in the majority, and was classified according to the WHO classification of lupus nephritis on renal biopsy appearance (Table 3).Table 3 gives the details of clinical manifestations.The specific findings at investigations are given in Table 4.A majority of patients (>90%) needed treatment with immunosuppressive drugs (Table 5).

'Senior Registrar in Medicine and
Five patients died during follow up (Table 6) giving an incidence of death of 7% for a mean duration of illness of 7 years.Three patients became pregnant, and where pregnancy progressed beyond the first trimester, deliv ered live healthy babies.Two patients had a complicated puerperium, but recovered fully.

Discussion
The diverse clinical features in SLE may result from . the variable influence of genetic, immunological, hormonal and environmental factors (7), with epidemiological stud ies from different parts of the world giving varying preva lence rates for common features of SLE (7,8).
In the absence of a single screening test with a high specificity and sensitivity, the diagnosis of SLE is made on the basis of the ARA (now the American College of Rheu matology) criteria for the diagnosis of SLE (9).The pres ence of any 4 of these 11 criteria either concurrently or sequentially at any time of the disease enables the diagno sis of SLE with 98% specificity and 97% sensitivity (9).
Although the ARA criteria for lupus are used for clini cal diagnosis, they were developed for classifying patients in clinical trials and epidemiological studies (10).Some char acteristic visceral involvements such as chorea, peripheral neuropathy, and restrictive lung disease, which were seen in our patients, are not part of the ARA criteria.Conse quently, some of our patients did not fulfill the ARA diag nostic criteria.The late development of some clinical fea tures included in the ARA criteria, several months or years after initial illness, is another reason why all SLE patients do not fulfill ARA criteria initially.However, we used the ARA criteria to include patients for our study, as in other studies which have analysed the clinical features of SLE (11,12,13,14).Some patients who were referred to our clinic did not have SLE, and 26 had a different autoimmune disease.

Papers
The SLE features seen in our study compared with other series (11,12,13,14) are given in Table 7.This compari son shows a higher prevalence of alopecia (88%) in our patients, and the lowest prevalence for malar rash (66%).There were no striking differences noted in other clinical manifestations.The female preponderance shown in all previous studies is confirmed in our study.
The visceral involvement seen in 87% patients is in keeping with the observation of high prevalence of inter nal organ disease in Asians (6).The majority of our pa tients with renal involvement had focal proliferative or dif fuse proliferative nephritis (WHO classes III and IV) which are considered as severe lupus nephritis.Neurological in volvement was also noted in a number of patients and chorea was more prevalent in this study population com pared to others (17,18).
Atherosclerosis is emerging as a significant cause of death and illness in patients with long standing SLE (19).The mortality rate from coronary, artery disease in patients with SLE is estimated to be 9 times that predicted on popu lation based rates (20).The reasons for accelerated athero sclerosis in SLE include the high prevalence of risk factors such as hypertension, hyperlipidaemia and obesity, and treatments such as corticosteroids (9).
The treatment of patients in our clinic depended on the severity of the disease and visceral involvement.Patients with severe disease and with life-threatening major organ involvement received induction therapy with methylprednisolone, dexamethasone and cyclophospha mide as recommended in accepted treatment protocols (21,22).
Recent studies have documented substantial improve ment in the survival of patients with SLE (23,24,25) with 5year survival rates of over 90% and 10-year survival rates of over 80% (24,25,26).Thus our mortality figure of 6% during 3-year follow up for a mean duration of illness of 7 years, is comparable to other series.The leading causes of death in patients with lupus are infectious complications and clinical manifestations directly related to lupus itself (24,25).In many patients infections develop in the setting of active lupus under aggressive treatment: thus it is often difficult to identify a single cause of death (27).The suc cessful outcome of pregnancies noted in our patients is also in keeping with results from recent studies (28).
This study of clinical features of SLE in Sri Lankan patients confirms the wide variability of clinical features in the disease and its ethnic variations.We observed alope cia and visceral involvement to be more common in Sri Lankan patients.SLE remains a serious disease with sig nificant morbidity and mortality.

Table 2 .
Systemic, musculoskeletal and mucocutaneous features in patients with SLE Lecturer, Department of Pharmacology, Registrar in Medicine, ^Senior Lecturer, Department of Medicine, ^Professor of Medicine, Faculty of Medicine, University of Colombo.(Revised version accepted 16 September 2000).

Table 3 .
Visceral involvement in patients with SLE

Table 4 .
Specific findings of investigations in 77 SLE patients