CRISTINA DRENKARD, ANTONIO R. VILLA, CARLOS GARCRA-PADILLA, MARQA
ESTHER PIREZ-V;ZQUES AND DONATO ALARC[N-SEGOVIA
It has been little recognized that systemic lupus erythematosus (SLE), even when severe, may go into remission whether spontaneously or under the influence of treatment and require no further treatment for some time or even indefinitely (6, 7, 12, 13, 24, 30, 31). Further, the frequency, recurrence, and duration of remissions have not been determined, nor have the factors that may influence their occurrence or the characteristics of patients in whom remissions are more likely to occur.
Inasmuch as SLE has had a reputation in the past as a uniformly fatal condition, a notion that is still commonly held, and since there is no actual cure for this disease, achieving remission is the current goal of management. It seemed therefore important to study the occurrence of remissions in our cohort of patients with SLE to determine if we can counteract that pessimistic notion and learn how often that therapeutic goal is achieved.
We found that nearly one-fourth of our patients with SLE in the cohort had had treatment-free remission for at least 1 year, that more than half of those whose disease lasted more than 18 years had achieved remission, and that three-fourths of those with over 30-years disease duration had had remission. Survival of patients who underwent remission was significantly longer than of those who did not. There were 41 patients with severe renal disease who achieved remission.
Patients
We have previously described our lupus cohort with 667 patients with SLE entered into a database (1). Briefly, patients entered into this cohort were seen at the Lupus Clinic of our institute between April 1986 and December 1989. All patients included fulfilled 4 or more criteria of the American College of Rheumatology (29). Prospective follow-up consisted of clinical and laboratory evaluation at 3-month intervals and any additional visits deemed necessary for best management of patients. Prospective follow-up was completed December 31, 1990, which represents 2,039 person/years. The average follow-up within the cohort was 3.1 " 1.5 years (median, 3.7 yr). In our cohort we identified all patients who had undergone remission, defined as at least 1 year during which lack of clinical disease activity permitted withdrawal of all treatment for lupus proper. We did not consider as being in remission any patient whose disease was so mild that it had not required treatment, nor any patient who had become asymptomatic but was still taking corticosteroids, immunosuppressors, antimalarials, nonsteroidal anti-inflammatory drugs, or any other form of treatment intended to suppress general lupus activity or that of a particular clinical manifestation. The period of remission was considered from the time the patient stopped all medications: to be considered in remission, a patient had to be without any such medication for at least 1 continous year. Changes in laboratory parameters without concurrent clinical evidence of disease activity, thus not requiring treatment, did not invalidate the criteria for remission. Clinical manifestations representing disease activity, whether they were treated or not, that invalidated including a patient as in remission are presented in the Appendix.
A new database was created to include ad hoc demographic, clinical and therapeutic variables. Main variables and their definitions are presented in the Appendix. Initial manifestations and therapeutic coefficients were considered only when they occurred before the first remission, even in patients who achieved remission within the first 2 years of disease. Data on these variables were obtained from the patient record by one of us (CGP) using a precoded questionnaire with the definitions included in the Appendix.
Methods
The study included the report of 3 selected illustrative patients in remission and a description of all patients who fulfilled criteria for remission at any time during follow-up under our care. In addition, we compared survival between patients who achieved remission and those who had never entered in remission up to cutoff time.
Statistical analysis: We estimated the probability of going into the first remission across time in the whole cohort by the Kaplan-Meier method (16). Risk time was defined as that from the first manifestation attributable to SLE to the date of onset of the first remission or to cutoff time (uncensored observations). We used the same method to calculate survival curves taking death as dependent variable and remission as prognostic variable, and we measured the time at risk to death of each patient. Time zero corresponded to the first clinical manifestation of SLE. Differences between curves were estimated by generalized Wilcoxon test. Multivariate analysis of mortality risks was performed by means of Cox proportional hazard regression by stepwise method (4). Mortality risks were expressed as hazard functions.
SPSS/PC (27) was used to perform Kaplan-Meier analysis. Cox proportional hazard regressions were performed using EGRET (8) package.
Case 1
A 31-year-old woman was admitted at our institution with a diagnosis of SLE. Two years before, she had begun with weight loss, fever, myalgia and arthralgia. Some months later she developed arthritis and alopecia. Antinuclear antibodies and positive lupus erythematosus cells were documented at that time. She was admitted in September 1969. On physical examination we found butterfly rush; alopecia; cervical, axillar and supraclavicular lymph nodes; Raynaud phenomena; subungual cyanosis; erythema over the dorsum of hands; livedo reticulairs; arthritis; and splenomegaly. Laboratory data on admission showed anemia, lymphopenia, positive lupus erythematosus cells, antinuclear antibodies of 1/16 (complement fixation) and immunofluorescence with peripheral and homogeneous pattern, anti-dsDNA antibodies of 1/160 by a hemagglutination method, C3 of 27.9 mg/dL (normal > 80), C4 of 7.6 mg/dL (normal > 18.5). There were no alterations in urinalysis, creatinine, or blood urea nitrogen values. She was found to have gall stones.
During hospitalization the patient developed pericarditis and ascites. Skin, muscular, and renal biopsies showed acute and chronic vasculitis. Renal biopsy showed also a focal proliferative nephritis. Prednisone, 60 mg/d, was started with improvement and the patient was discharged 40 days later with prednisone, 30 mg/d. A cholecystectomy was done. The clinical course was good and prednisone was stopped on June 1971. She was asymptomatic for 6 years and 3 months, but in September 1977 she flared with asthenia, fever, vascular purpura, and active urinary sediment with proteinuria, hematuria and casts. She had anemia, positive antinuclear antibodies, and hypocomplementemia. She was restarted on 30 mg/d of prednisone, but vasculitis extended, with ensuing arthritis, alopecia, and oral ulcers. Prednisone was therefore increased to 60 mg/d with hydroxychloroquine and aspirin. One year later prednisone was withdrawn again. This second remission lasted 3 years and 3 months when the patient flared again with general, cutaneous and articular manifestations in March 1982. Two weeks later she developed seizures, hypocomplementemia, and anti-dsDNA uptake of 98.5 % (normal < 36 %). Computed axial tomography showed only mild cortical atrophy. The patient was treated with prednisone, 45 mg/d, and valproic acid. She had no more seizures, and levels of C3 and DNA uptake became normal 3 months later allowing decrease of prednisone to its complete withdrawal. She improved, and in February 1985, 2 years and 8 months later, valproic acid was also stopped. Since then, until her last visit in September 1995, she has remained in remission, without evidence of SLE activity or any treatment.
Case 2
A 12-year-old girl, with family history of SLE in 2 paternal aunts, started with polyarthralgia in December 1984. Two months later, after sun exposure, she developed malar erythema with skin ulceration on the nose, asthenia, fever and dyspnea. She was hospitalized at another institution where she was febrile, with hypertension, butterfly rash and palm arythema. She had leukopenia of 2,600 cells, sedimentation rate of 56 mm/hour, telescoped urinary sediment, a blood urea of 64 mg/dL, a serum creatinine of 1 mg/dL, positive rheumatoid factor, antinuclear and anti-DNA antibodies. SLE with renal involvement was diagnosed and she was given prednisone, 90 mg/d, and antihypertensive drugs. One week later a renal biopsy showed diffuse proliferative lupus nephritis with an activity index of 8 and a chronicity index of 4. Days later the renal function worsened with blood urea increasing to 89 mg/dL and serum creatinine to 1.3 mg/dL. Cyclophosphamide was started with good response. Three months later she was admitted at our institution. She complained of headache and was found to have periungual erythema and arthritis in elbows and left knee. She was taking prednisone, 45 mg/d, and cyclophosphamide, 100 mg/d. Laboratory data at admission showed antinuclear antibodies 1:100 with a homogeneous pattern, hypocomplementemia, hemoglobinuria, and normal serum creatinine and blood urea. Cyclophosphamide was changed to azathioprine, 100 mg/d, and the prednisone dose was tapered gradually. During follow-up her disease remained clinically inactive, allowing us to withdraw prednisone 2 years later and azathioprine in October 1989. Since then the patient has remained asymptomatic without treatment, but occasionally she has had lymphopenia, increased anti-dsDNA antibodies and hypocomplementemia. Her renal function is normal.
Case 3
In October 1971 a 28-year-old woman was admitted to our hospital with a 1-month history of fever (39.5 EC), weight loss, asthenia and myalgia. On admission she was found to have somnolence; alopecia; livedo reticularis; erythematous lesions in her legs; splenomegaly; and cervical, supraclavicular and axillar lymph nodes. Laboratory data showed anemia, leukopenia, hypergammaglobulinemia, high sedimentation rate, hypocomplementemia and a false-positive VDRL test. Urinalysis revealed hemoglobin, erythrocytes, protein and casts. Infection was excluded. Biopsy of an axillary node showed nonspecific adenitis. While hospitalized she developed erythematous plaques on her face, petechiae on the palate and vascular purpura with hematic bullae on palms and soles. She had evidence of neurologic involvement with somnolence, indifference, tremor in her extremities and hyperreflexia. Anti-dsDNA antibodies were found in fluid from the bullae but not in serum where antinuclear antibodies were also negative. A diagnosis of SLE with cutaneous, neurologic, and renal involvement was made, and prednisone, 150 mg/d, was begun with prompt disappearance of fever, anemia, leukopenia and skin and neurologie manifestations. She was discharged with decreasing prednisone dose. In February 1972 the patient was started on hydroxychloroquine, remaining without clinical activity. Nine months later she was off prednisone and hydroxychloroquine was stopped because of her first pregnancy. She did not require treatment for lupus during pregnancy, which was uneventful. She remained without SLE activity and without treatment 2 years, after which she developed a butterfly rash treated with hydroxychloroquine. Treatment was stopped in May 1976, and 4 months later the patient became pregnant again. During that pregnancy, she remained without evidence of clinical activity and no treatment. Three months after successful delivery, she presented with cutaneous vasculities, arthralgia, and alopecia, which were controlled with low doses of prednisone and hydroxychloroquine. She continued to be asymptomatic and was able to stop treatment again in April 1979 although she had positive antinuclear antibodies (homogeneous pattern) and low C3 levels. She remained in remission until June 1980, when she again had skin rash and arthritis with leukopenia and low C4 levels. She was given prednisone, 10 mg/d, with fast improvement. The fourth remission was achieved in March 1981 and lasted 6 years. In February 1987 she started with fever, headache, somnolence, hyperreflexia, Babinski sign and myoclonus. A leukocyte count was 3,000 with 1,170 lymphocytes and 72,000 platelets. Antinuclear antibodies were positive, an anti-dsDNA uptake was 50 % and C3 level was 41,6 mg/dL. Prednisone, 45 mg/d, was started and 4 months later, while it was being decreased, she developed fever and skin rash and subsequently seizures and arthritis. She was hospitalized and no abnormalities in the electroencephalogram were found. However, a computed tomography (CT) scan and magnetic resonance imaging (MRI) revealed white substance increase in periventricular areas, consistent with progressive multifocal leukoencephalopathy. The patient had only apathy but no memory, judgment, or calculation alterations. She remained without lupus activity or neurologic deficit, and in May 1988 prednisone was stopped again. Since then she has been asymptomatic and takes only a baby aspirin daily. She has refused a new MRI scan.
In our cohort of 667 patients with SLE we identified 156 who had undergone remission at least once during follow-up. This represents a cumulative incidence of 0.234 for the whole period. Sum of risk time to first remission was 5,546 person/years for the 667 patients with SLE, giving an incidence density of remission was 0.028 new cases/person/year. Of the 156 patients who underwent remission, 150 were female. The mean follow-up of this group was 11.6 " 6.0 standard deviation (SD) years. The mean age " SD at SLE onset in this group of patients was 28.9 " 11.2 years, and the mean length of time from SLE onset to the first remission was 7.5 " 5.6 years (range, 0.5-31.7 yr). Total number of visits in this group during their first remission was 2,298 with a mean of 14.7 visits and a rate of 3.2 visits per patient/year. Twelve of the 156 patients did not complete follow-up but contributed with risk time up to their last visit. Complete information regarding initial manifestations and remission was obtained in these patients.
Table 1 shows the frequency of the main initial manifestations in 156 patients who achieved remission. There were at least 41 patients who achieved remission after having renal disease during the initial period (cases 1, 2 and 3). Another 19 patients had central nervous system involvement (case 3), 15 had thrombocytopenia, and 8 had hemolytic anemia as initial manifestations.
Table 1. Frequency of main initial manifestations in 156 patients with SLE who achieved remission
Variable
No. of %
Patients
Arthralgia 127 81.4
Mucocutaneous manifestations 120 76.9
Arthritis 111 71.2
Lymphopenia 50/86 58.1
Butterfly rash 73 46.8
Fever 73 46.8
Asthenia 69 44.2
Renal involvement 41/97 42.3
Mucosal ulcerations 51 32.7
Raynaud phenomenon 47 30.1
Alopecia 47 30.1
Leukopenia 26/90 28.9
Pleuritis 44 28.2
Palmar erythema 28 17.9
Thrombocytopenia 15/91 16.5
Central nervous disease 19 12.2
Hemolytic anemia 8/87 9.2
Discoid lupus 13 8.3
Periungual erythema 13 8.3
Carditis 13 8.3
Myositis 12 7.7
Seizures 10 6.4
Pericarditis 9 5.8
Psychosis 3 1.9
Organic brain syndrome 3 1.9
Transverse myelitis 3 1.9
Classical migraine 3 1.9
Abbreviations: SLE = systemic lupus erythematosus.
Numbers for each manifestation may be larger, as some patients may have had manifestations before being seen at our institution. Treatment given to the patients to control manifestations of disease activity before they achieved remission is shown in Table 2. Most of these patients received prednisone and chloroquine, 30 % of them took azathioprine, and only 14 patients took cyclophosphamide. Cumulative time under treatment with prednisone at low doses represented only 49 % of the time under our follow-up before entering in remission, as shown by its mean coefficient. Other coefficients are lower.
Table 2. Treatment in 156 patients with SLE before the first remission
Drug No. of
% Coefficient*
Patients Mean
Chloroquine 88 56 0.30
Prednisone, < 15 mg/day 135 87 0.49
Prednisone, 15-30 mg/day 94 60 0.09
Prednisone, > 30 mg/day 72 46 0.06
Azathioprine 46 30 0.15
Oral cyclophosphamide 12 8 0.02
Azathioprine plus oral 1 0.6 0.00
cyclophosphamide
Cyclophosphamide bolus 1 0.6 0.00
----------------
* See appendix for description.
Sixty-two of the 156 patients underwent remission within 2 years of the first clinical manifestation attributable to SLE (initial period) as illustrated by case 3. The remaining 94 patients began their first remission after having had from 1 to 28 episodes of disease reactivation following the initial period (Figure 1). The mean duration of the first remission was 4.6 " 3.6 years (range, 1-17.3 yr), but the mean duration of remission of the 81 patients remaining in remission up to cutoff time was 5.9 " 4.2 years (range, 1-17.3 yr). Of the 75 patients who flared after remission, there were 37 who did not remit again and 38 who did. Twenty-six of these 38 patients remained in remission while the other 12 patients had subsequent flares and remissions as shown in Figure 2 (cases 1 and 3). Seventy-three percent of patients (n = 114) were still in remission at cutoff time, while the remaining 42 required treatment.
The total time each patient was in remission up to cutoff time is shown in Figure 3. Only 35 patients had remission of less then 2 years. Half of the patients were at least 5 years in remission, and 24 were in remission 10 years or longer (cases 1 and 3).Two patients had been in remission 17.3 years up to cutoff time and continue to be in remission at the time of this writing (21 years).
Mean total time in remission was 5.8 years (" 4.0). This represented 40 % " 24 % of their total disease duration. However, this amount of time in remission means that these patients were in remission half of their follow-up time (50 % " 26 %).
Gladman and coworkers (13) described patients with SLE who had low C3 determinations and/or high DNA uptake despite having quiescent disease. However, most of our patients with clinical remission also had negative serologic tests most of the time while in it. Of 134 patients tested when in remission, 70 had no positive anti-dsDNA uptake (Figure 4A). Of the 86 patients whose C3 levels were tested while in remission, 50 had normal levels at all times (Figure 4B). Lymphopenia has also been proposed as indicative of disease activity (23). Patients with SLE in remission had lymphopenia a median of 18 % of the times they were tested (Figure 5).
Figure 6 shows the Kaplan-Meier curve made to estimate the probability across time of entering remission in the whole cohort. There was a continuous increase in the likelihood of having a first remission from the beginning of disease up to 20 years of disease, at which time it reached 50 %. The probability of remission rose to 70 % after 30 years of disease duration.
Additional analysis was made to compare the survival of the patients who underwent remission during follow-up to those patients who did not (Figure 7). Two of the 156 patients who achieved remission and 48 of the 511 who did not died during follow-up. The difference between the 2 survival curves was highly significant (p = 0.000). Cox multivariate analyses controlling for other manifestations previously reported as associated with increased mortality (5) showed that patients who achieved remission had a 12.5-fold (1/0.08) smaller chance of dying than those who did not, irrespective of the presence of renal disease and thrombocytopenia (Table 3).
Our findings indicate that treatment-free remission can be achieved in nearly one-fourth of patients with SLE and that about one-half of these remained in their first remission for a mean of nearly 6 years. It should be noted that with the strict definition of remission we used, this length of time is of both quiescent disease and absent treatment after having had active disease that required it. This may imply a much longer time devoid of symptoms while medications were gradually withdrawn. However, treatment-free remission accounted for half the time of total follow-up of the 156 patients. This represented a mean of 5.8 years in remission for the entire group. Applying indices of disease activity to this study was not deemed necessary; they would have been uniformly negative except for the laboratory manifestations included in some of them, which only represent disease activity when accompanied by clinical manifestations (for example, low C3, positive anti-DNA, lymphopenia). The list of manifestations we used to exclude patients from the remission group is more comprehensive than those of most indices of disease activity. In addition, the retrospective nature of part of our study prevented the use of these indices since they have been planned mostly for prospective studies (14, 20).
Subsequent remission occurs in about half of those patients who relapse and, at cutoff time,
73 % of patients were still clinically quiescent without treatment. The probability of undergoing remission increased with disease duration, reaching 70 % at 30 years. Nineteen patients had central nervous disease as initial SLE manifestations and afterward achieved remission. Although several authors (9, 19, 28) consider neuropsychiatric SLE as having a bad prognosis, there are others (11, 25, 26) who disagree. It is known that neuropsychiatric lupus, other than the vasoocclusive, is often reversible (11, 17, 26), and that some neurologic syndromes, such as psychosis or convulsions, do not yield a poor prognosis (9). Examples of this are cases 1 and 3 who suffered severe central nervous system involvement but later remitted for long periods without neurologic sequelae.
Forty-one of 97 patients who could be evaluated at the initial period had renal disease. This tells us that although remission may be difficult to achieve in patients with renal disease, it does occur. This notion can be showed by the 3 patients reported herein who, in spite of having had severe renal involvement, were able to achieve remission. We should also consider that physicians may, in general, be more hesitant to stop treatment in patients with known lupus nephritis.
Cox multivariate analyses showed that remission was associated with better survival independently of the effect of other disease manifestations known to cause increased mortality (5).
Although there was the notion that some patients with SLE can achieve treatment-free remission, our findings indicate that the potential to do this has been heretofore underestimated. Actually, there was a surprising paucity of information on remission in SLE. Tumulty (31), Dubois (7) and Ropes (24) called attention to the fact that some patients with SLE may undergo spontaneous remission following their first symptoms, which were usually mild and consisted mainly of arthritis, fever, pleuropericarditis, or anemia (7). They seldom found it to occur in patients with major organ involvement. However, 17 of Dubois' 132 patients with SLE who required prednisone administration achieved remission for a year or longer after such treatment (7). In 1964 Zeldis (33) reported 4 patients with SLE in whom all medication could be stopped for at least 2 years. He considered these patients possibly cured. We are hesitant to consider cured even those patients whose remission has exceeded 10 years. On the one hand we have seen relapses after as many as 14 years of remission which, due to continued observation, have been promptly and efficaciously treated and, on the other hand, immunoregulatory studies done with peripheral blood mononuclear cells from patients with SLE in remission have shown improvement, but usually not disappearance, of the abnormalities in cell populations (21, 22), functional studies (3, 18), and cytokine production (2). Gladman and coworkers (13) found 14 patients who became clinically quiescent, albeit having persistently positive antinuclear antibody tests, high levels of DNA binding in 50% of the tests, and low complement in 59 % of the determinations. Some of their patients could be taken off all medication and the minority of them subsequently flared (32). Herein we confirmed this notion but found that such patients constitute a minority and that most others have increased DNA uptake and/or decreased C3 levels only a small percentage of the time they are in remission.
In some series remission of individual clinical manifestations was seen but seldom was there remission of all disease activity allowing discontinuation of treatment (12). In a retrospective study (15) of 148 patients with SLE no complete remission occurred in any patient followed for more than 10 years. This would be quite different from our findings and could not be attributed to milder disease in our patients since among those who achieved remission there was a rather large contingent of patients with renal, neuropsychiatric, or hematologic disease. That patients with severe SLE may undergo remission was recently brought up by Euler and coworkers (10) who reported 8 patients with severe SLE who went into remission within 6 months of synchronization of plasmapheresis with subsequent pulse cyclophosphamide and remained treatment-free for a mean observation period of 5.6 years. Our findings here indicate that such specific form of treatment is not a requirement for the remission of patients with severe SLE. Remission in patients with renal disease did not seem to be related to any particular treatment program.
Table 3. Cox multivariate model of surivial in 667 patients with SLE*
Variable Hazard
95 %
Funcition Confidence p Value
Intervals
Remission 0.08 0.02-0.33 <0.001
Renal involvement 3.3 1.8-6.2 <0.001
Thrombocytopenia 2.5 1.4-4.4 0.002
* Model=s p value = 0.003
In conclusion, patients with SLE, including those with severe organ involvement, may undergo remission permitting prolonged withdrawal of all treatment. The proportion of our patients who achieve remission increases as the disease lasts longer reaching 70 % at 30 years. Our findings support the notion that, in general, SLE is a more benign disease than previously considered
and that a significant number of patients may become, perhaps indefinitely, symptom-free and in need of no more medication.
The occurrence and characteristics of remissions in patients with systemic lupus erythematosus (SLE) have not been determined. We therefore studied this in a cohort of 667 patients and found that 156 patients had achieved at least 1 period of 1 year or more of treatment-free clinical remission. This represents an incidence density of 0.028 new cases/person/year. Remission occurred within the first 2 years of disease in 62 patients. The mean duration of first remission was 4.6 years (range, 1-21 yr), and 81 patients were still in the initial remission up until cutoff time. Half of the remaining 75 patients who flared after achieving remission have not entered again in remission. Twenty-six of the 38 patients who did remained in remission, and the remaining 12 had subsequent flares and remissions. Treatment-free remission accounted for a mean of 5.8 years, corresponding to half the time of follow-up. Remission was not limited to patients with mild disease: at least 41 patients achieved remission despite renal involvement, 19 had had neuropsychiatric lupus, 15 had had thrombocytopenia, and 8 had had hemolytic anemia. We also found that the longer the time lapse between the initial manifestation and the diagnosis of SLE, the less likely it was for a patient to enter into remission.
There was a continuous increase in likelihood of achieving a first remission from the beginning of disease up to 30 years of disease duration, when it reached 70 %. Patients who achieved remission had increased survival, independently of the effect of other disease manifestations that cause increased mortality.
We conclude that a significant proportion of patients with SLE, including those with severe organ involvement, may become symptom-free and in need of no more medication, perhaps indefinitely. Our findings support the notion that, in general, SLE is a more benign disease than previously considered.
Evidence of disease activity
The following clinical manifestations were considered as evidence of lupus activity if observed by the attending physician and not due to any other cause. Presence of any of them prevented including a patient as in remission.
General manifestations: fever, asthenia, Mucocutaneous manifestations: alopecia, mucosal ulcerations, butterfly rash, new lesions of discoid lupus, cutaneous vasculitis, new palmar or periungual erythema. Reticuloendothelial manifestations: new adenopathies. Articular manifestations: arthralgia or arthritis. Pulmonary manifestations: pleuritis, pneumonitis, pulmonary hemorrhage. Cardiovascular manifestations: pericarditis, carditis. Gastrointestinal manifestations: peritonitis, pancreatitis, mesenteric vasculitis. Neurologic manifestations: new seizures, psychosis, organic brain syndrome, chorea, transverse myelitis, mononeuritis multiplex, cranial neuritis. Muscular manifestations: myositis. Hematologic manifestations: thrombocytopenia, hemolytic anemia. Renal manifestations: hematuria, granular casts, new proteinuria or increase of residual proteinuria, increased serum creatinine > 5 mg/dL.
The main clinical and therapeutic variables and definitions used in the study are described below. Follow-up: from hospital entry to cutoff time. Disease duration: from first symptom attributable to SLE to cutoff time. Initial manifestations: those had during the first 2 years after the first symptom: General manifestations: fever (> 38EC), asthenia (fatigue, loss of energy and strength with functional limitation). Mucocutaneous lesions: definite alopecia (observed by the physician), probable alopecia (reliable history), mucosal ulcerations, butterfly rash, discoid lupus, palm erythema, periungual erythema. Articular involvement: arthralgia, arthritis. Pulmonary involvement: definite pleuritis (pleuritic chest pain with pleural effusion or audible rub), probable pleuritis (only pleuritic chest pain). Cardiovascular involvement: Raynaud phenomenon (2 phases), definite pericarditis (pericardial pain with effusion or audible rub or compatible electro/echocardiogram), probable pericarditis (only pericardial chest pain), definite carditis (cardiac rate > 100/min and arrhythmias or conduction disturbances, excluding other causes), probable carditis (only cardiac rate > 100/min excluding fever, infection and other causes). Central nervous disease (excluding infection, drugs, metabolic alterations and other causes): seizures, psychosis, organic brain syndrome, transverse myelitis, classical migraine. Muscular involvement: definite myositis (proximal weakness with increased creatine phosphokinase, compatible biopsy, or electromyogram), probable myositis (proximal weakness not due to drugs). Hematologic involvement: definite thrombocytopenia (platelet count < 100,000 on 2 or more occasions, not due to drugs), probable thrombocytopenia (reliable history not documented at our institution), definite hemolytic anemia (hemoglobin decrease > 3 g/dL, corrected reticulocyte > 3,5 %, and indirect bilirubin increase > 0.6 mg/dL), probable hemolytic anemia (reliable history not documented by us), leukopenia (< 4,000 white blood cells on 2 or more occasions), lymphopenia (lymphocytes < 1,500 on 2 or more occasions). Renal involvement: hematuria (> 5 red blood cells/highpower field), granular casts, proteinuria (> 0.5 g/24 h), diastolic arterial hypertension.
Relapse: new episode of disease activity after the initial period (after second year of disease course). It comprises the new manifestations of disease activity occuring after disease control by drug treatment, or after a remission period: general, mucocutaneous, reticuloendothelial, articular, pulmonary, cardiovascular, intestinal, neurologic, muscular, hematologic, or renal involvement.
Treatment was evaluated by coefficients of time for each drug administered before the first remission. The coefficients created are listed below.
CLQ coeff: chloroquine coefficient, PDN low coeff: prednisone at low doses (< 15 mg/d) coefficient, PDN intern coeff: prednisone at intermediate doses (15-30 mg/d), PDN high coeff: prednisone at high doses (> 30 mg/d), AZA coeff: azathioprine coefficient, CPM po coeff: oral cyclophosphamide coefficient, AZA + CPM coeff: azathioprine plus oral cyclophosphamide coefficient, CPM bolus coeff: cpclophosphamide bolus. The formula applied to create each drug coefficient follows.
n
(* Drug)
¡=š
Drug coefficient =
Follow-up until onset of 1st remission
Where:
Drug coefficient = Coefficient for each studied drug (CLQ,
PDN low, PDN intern, PDN high, AZA,
CPM po, AZA + CPM, CPM bolus)
n
3 (* Drug) = Sum of periods (*) for each drug indicated
¡=š
(drug) from ¡=š to n
Each drug coefficient has as numerator the cumulative time under treatment with the respective drug and as denominator the time from entry to our clinic to the first remission. The value of the coefficient may range from 0 to 1, where 0 would represent a patient who never received the drug and 1 would represent a patient who received it throughout the time defined by the denominator. For example, take the case of a patient who received chloroquine in 2 different periods since admission, the first period spanning 2 years and the second, 6 months. After 4 years follow-up she achieved remission. Thus the numerator would be 2 + 0.5 = 2.5 yr and the denominator = 4 yr. The value of the CLQ coefficients is 0.625 (2.5/4).
1. Alarc\n-Segovia D, PJrez-V<zques ME, Villa AR, Drenkard C, Cabiedes J. Preliminary classification criteria for the antiphospholipid syndrome within systemic lupus erythematosus. Semin Arthritis Rheum 21: 275-86, 1992.
2. Alcocer-Varela J, Alarc\n-Segovia D. Decreased production of and response to interleukin-2 by cultured lymphocytes from patients with systemic lupus erythematosus. J Clin Invest 69: 1388-92, 1982.
3. Alvarado C, Alcocer-Varela J, Llorente L, Richaud-Patin I, Cerbon M, Alarc\n-Segovia D. Effect of CD28 antibody on T cells from patients with systemic lupus erythematosus. J Autoimmunity 7: 763-73, 1994.
4. Cox DR. Regression models and life tables. J R Stat Soc 34: 187-203, 1972.
5. Drenkard C, Villa AR, Alarc\n-Segovia D, PJrez-V<zquez ME. Influence of the antiphospholipid syndrome in the survival of patients with systemic lupus erythematosus. J Rheumatol 21: 1067-72, 1994.
6. Dubois EL. The effect of the L.E. Cell Test on the clinical picture of systemic lupus erythematosus. Ann Intern Med 38: 1265-94, 1953.
7. Dubois EL. Systemic lupus erythematosus: Recent advances in its diagnosis and treatment. Ann Intern Med 45: 163-84, 1956.
8. EGRET v 0.25.l. Statistics and Epidemiology Research Corporation. Seattle, Wa.
9. Estes D, Christian CL. The natural history of systemic lupus erythematosus by prospective analysis. Medicine (Baltimore) 50: 85-95, 1971.
10. Euler HH, Schroeder JO, Harten P, Zeuner RA, Gutschmidt HJ. Treatment-free remission in severe systemic lupus erythematosus following synchronization of plasmapheresis with subsequent pulse cyclophosphamide. Arthritis Rheum 37: 1784-94, 1994.
11. Feinglass EJ, Arnett FC, Dorsch CA, Zizic TM, Stevens MB. Neuropsychiatric manifestations of systemic lupus erythematosus: Diagnosis, clinical spectrum and relationship to other features of the disease. Medicine (Baltimore) 55: 323-39, 1976.
12. Fries JF, Holman HR. In: Smith LH Jr, ed. Systemic lupus erythematosus: A clinical analysis. Vol. VL Philadelphia: WB Saunders, pp. 127-33, 1975.
13. Gladman DD, Urowitz MB, Keystone EC. Serologically active clinically quiescent systemic lupus erythematosus. A discordance between clinical and serologic features. Am J Med 66: 210-15, 1979.
14. Guzm<n J, Cardiel MH, Arce-Salinas A, S<nchez-Guerrero J, Alarc\n-Segovia D. Measurement of disease activity in systemic lupus erythematosus. Prospective validation of 3 clinical indices. J Rheumatol 19: 1551-58, 1992.
15. Halberg P, Alsbjorn B, Trolle Balslov J, Gerstoft J, Lorenzen I, Ullman S, Wiik A. Systemic lupus erythematosus: Follow-up study of 148 patients. I: Classification, clinical and laboratory findings, course and outcome. Clin Rheumatol 6: 13-21, 1987.
16. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457-81, 1958.
17. Kovacs JAJ, Urowitz MB, Gladman DF. Dilemmas in neuropsychiatric lupus. Rheum Dis Clin North Am 19: 795-814, 1994.
18. Laff\n A, Alcocer-Varela J, Alarc\n-Segovia D. Differences in the kinetics of the autologous mixed lymphocyte reaction between the various connective tissue diseases. Rheumatol Int 3: 117-28, 1983.
19. Lee P, Urowitz MB, Bookman AAM, Koehler BE, Smythe HA, Gordon DA, Ogryzlo MA. Systemic lupus erythematosus. A review of 110 cases with reference to nephritis, the nervous system, infections, aseptic necrosis and prognosis. Q J Med 46: 1-32, 1977.
20. Liang MG, Socher SA, Larson MG, Schur PH. Reliability and validity of six system for the clinical assessment of disease activity in systemic lupus erythematosus. Arthritis Rheum 32: 1107-18, 1989.
21. L\pez-Karpovich X, Cardiel M, C<rdenas R, Piedras J, Alarc\n-Segovia D Circulating colony-forming units of granulocytes and monocytes/macrophages in systemic lupus erythematosus. Clin Exp Immunol 77: 43-46, 1989.
22. Melendro EI, Saldate C, Rivero SJ, Alarc\n-Segovia D. T-Cell subpopulations in the peripheral blood of patients with connective tissue diseases as determined by flow cytometry using monoclonal antibodies. Clin Immunol Immunopathol 27: 340-47, 1983.
23. Rivero SJ, DRaz-Jouanen E, Alarc\n-Segovia D. Lymphopenia in systemic lupus erythematosus. Clinical, diagnostic and prognostic significance. Arthritis Rheum 21: 295-305, 1978.
24. Ropes MW. Observations on the natural course of disseminated lupus erythematosus. Medicine (Baltimore) 43: 387-91, 1964.
25. Sergent JS, Lockshin MD, Klempner MS, Lipsky BA. Central nervous system disease in systemic lupus erythematosus. Therapy and prognosis. Am J Med 58: 644-54, 1975.
26. Sibley JT, Olszynski WP, Decoteau WE, Sundaram MB. The incidence and prognosis of central nervous system disease in systemic lupus erythematosus. J Rheumatol 19: 47-52, 1992.
27. SPSS/PC + v 5.0 Chicago III.
28. Swaak AJG, Nossent JC, Bronsveld W, Van Rooyen A, Nieuwenhuys EJ, Theuns L, Smeenk RJT. Systemic lupus erythematosus. I. Outcome and survival: Dutch experience with 110 patients studied prospectively. Ann Rheum Dis 48: 447-54, 1989.
29. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF, Schaller JG, Talal N, Winchester RJ. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 25: 1271-77, 1982.
30. Tozman ECS, Urowitz MB, Gladman DD. Prolonged complete remission in previously severe SLE. Ann Rheum Dis 41: 39-40, 1982.
31. Tumulty PA. The clinical course of systemic lupus erythematosus. JAMA 156: 947-53, 1954.
32. Walz LeBlanc BAE, Urowitz MB, Gladman DD. Serologically active, clinically quiescent systemic lupus erythematosus. Longterm follow-up. J Rheumatol 21: 174-75, 1994.
33. Zeldis A. Curaci\n en el lupus eritematoso sistJmico. Rev Med Chil 92: 299-301, 1964.
From MEDICINE Volume 75, No. 2
E. L. E. F.
European Lupus Erythematosus Federation