In recent years, it has become apparent that systemic lupus erythematosus (SLE) is not a very rare disease. Epidemiologie studies have suggested that it is a common disorder, with a prevalence of about six per 100,00O1 in white women; the prevalence is considerably higher in non-whites. The overall incidence and prevalence in younger aged individuals has not been determined.
In most series from investigators who have studied both children and adults, the percentage of affected children has varied depending on the definition of "child", In our own series of 182 patients first seen between 1968 and 1977, 32 (18%) had the onset of the first symptom of SLE before the age of 1 6.2 Thus, the first symptoms of the disease may be seen by pediatricians in 18 percent of all SLE patients. Thirty-five of our 182 SLE patients developed their first symptom between the ages of 16 and 20; thus, if pediatricians continue to be the primary physicians for patients under the age of 2 1 , they will be the first physicians to observe the onset of the first symptom of the disease in 32 percent of all SLE patients. It is clear that pediatricians should become very familiar with the manifestations of SLE in children and adolescents to facilitate diagnosis as early as possible in the disease, and institute appropriate therapy and observation.
The female to male ratio in the teenage group of about nine to one is similar to that in adults; among young children, however, the ratio falls to three to one. Thus, the pediatrician should always consider the possibility of SLE in young boys.
There are no controlled studies on the mortality and morbidity of SLE patients treated early in the course of disease compared with those treated after symptoms have been present for longer periods of time. However, it is my impression that patients who have not been diagnosed early in the course of the disease frequently suffer severe cases.
Pediatricians should become familiar with the early signs of SLE. The disease may become a full-blown, multisystem disease within a few weeks after the onset of the first symptoms; conversely, the first symptom may be present for months or years before the multisystem disease becomes gradually clinically apparent. In the latter situation it is most important for the pediatrician to consider the possibility of SLE and to obtain the appropriate laboratory tests in order to make a diagnosis.
In 42 children under the age of 1 5 studied in New York, 48 percent manifested the first symptom of SLE as joint pain.2 Very often, the children did not actually complain of pain, but the parents observed them limping or not using a hand. In most instances, more than one joint was involved, although, on occasion, the disease may be present as a monoarthritis. We did see two patients with pain in one ankle and one knee as their respective presenting complaints. In both of these young boys, fever and nephritis occurred within a few weeks of the onset of joint pain. In some patients, there was no objective evidence of arthritis - only joint pain. Polyarthritis involving most of the joints of the body was most frequent.
Rash was the second most frequent initial manifestation or presenting complaint. The rash was a butterfly blush in about one-half of the patients; in others it was generalized, erythematous and blotchy. The rash occurred most frequently on the extremities and chest, but was also present in some patients on the trunk. In some patients, the rash was pruritic. In about one-third of the patients, the rash occurred in sun-exposed areas after sun exposure at the time of the onset of the disease. Vasculitic lesions presenting as a tender erythematous lesion with a central small necrotic area occurred in some patients, especially on the hands or feet. In a small number of patients, urticarial lesions - especially on the extremities - were present at the time of onset. These urticarial lesions revealed leukocytoclastic angiitis on histologie examination.
Other dermatologie abnormalities may be present in a child who comes to the office with the major complaint of rash, and these must be sought by the pediatrician. Thus, the pediatrician should be alert to the possibility of SLE and pay particular attention to other manifestations of the disease that may support its diagnosis. Mucosal ulcers occur in about 25 percent of children and are even more common among those with rashes. The ulcers occur most often on the hard palate, but may be seen on the soft palate and in the anterior nasal septum. They are nearly always asymptomatic, although some children with nasal septal ulcerations may describe a pink discharge when they blow their noses. The presence of mucosal ulcers in a child with a rash is very suggestive of SLE and should lead to a thorough work-up for the disease.
The presenting complaint in a few children is an acute neurological symptom such as sudden blindness, severe ataxia or convulsion. Other initial complaints include generalized weakness, chest pain, and abdominal pain.
It is of interest that half of our children had been initially diagnosed as having diseases other than SLE. In the group of patients seen between 1956 and (968, rheumatic fever was the most common diagnosis.2 More recently, juvenile rheumatoid arthritis has been the most common initial diagnosis, primarily because of the presenting feature of joint pain in the majority of SLE patients. Discoid lupus has been seen in some children as the initial diagnosis and manifestation. Enlarged lymph nodes occur more frequently among children than among adults with active disease. In some of our patients, the initial diagnosis was lymphoma (Hodgkin's disease, reticuloendotheliosis).
By the time the diagnosis of SLE was made in our group of children, fever was present in 90 percent, joint complaints in 88 percent and rashes or alopecia in 74 percent. Thus, multisystem disease was evident. It is interesting to note that neurological complaints such as vertigo, ataxia, headache, tremors, seizures and paraplegia were present in 25 percent of the patients at the time of diagnosis. Organic psychosis was present in 10 percent at the time of diagnosis and prior to steroid therapy.2 Hepatosplenomegaly occurs more often in children than in adults. Renal disease (abnormal urinalysis) was present in 69 percent of the children at the time of diagnosis. The hematologie system was involved in 60 percent of patients; this will be discussed below under laboratory abnormalities. Gastrointestinal abnormalities in 19 percent of patients included abdominal pain with and without signs of peritoneal irritation. Crampy abdominal pain, while much less common among adults with SLE, is not an infrequent complaint among children. Pleurisy, pleural effusion and lung infiltrates occurred in 25 percent of the children.
Thus, there are manifestations of SLE that occur more frequently in children than in adults, and should be looked for with particular attention by the pediatrician. They include lymphadenopathy, hepatomegaly, splenomegaly and abdominal pain. He;matologie abnormalities also occur more often in children than in adults with SLE.
When a child comes to the pediatrician's office with one of the manifestations of SLE, a complete diagnostic work-up for SLE should be contemplated. If the child has fever and clinical evidence of multisystem disease when first seen, the child should be hospitalized for a complete work-up and for observation during initiation of therapy; if the child is afebrile but has a rash or joint pain, the work-up may be initiated in the office. It is very important to obtain the results of the laboratory tests as soon as possible since significant abnormalities may be present in a child who, save for mild symptoms, appears quite well.
To confirm a clinical diagnosis of SLE or to make a diagnosis in a child without signs and symptoms of multisystem disease, a large number of laboratory tests usually need to be obtained. In either case, specific tests for SLE should be ordered.
The best test to eliminate the diagnosis of SLE is an antinuclear antibody (ANA) test. This should be ordered from a laboratory that reports both the pattern of nuclear fluorescence produced by the patient's undiluted serum and the highest titer of the positive reaction, A peripheral pattern is nearîy always diagnostic of SLE. The speckled and diffuse patterns are not specific, and may be seen in juvenile rheumatoid arthritis and other childhood diseases. In a child with untreated SLE, the titer should be above 1 :32 and is usually considerably higher. A negative ANA using patient's undiluted serum rules out the diagnosis. In a child with mild clinical disease, the pattern is usually not peripheral. Thus, the ANA test, while positive, is not diagnostic. It is therefore necessary to obtain additional laboratory confirmation of the diagnosis. Antibodies to double-stranded (native) DNA are specific for SLE; therefore, this test should be ordered when SLE is first considered as a diagnosis.3'4 The sensitivity of this test is much lower than that of ANA and many SLE patients do not have antì-DNA antibodies at the time of onset of disease, unless they are acutely ill. Thus, a negative test for anti-DNA antibodies does not rule out the diagnosis of SLE. Antibodies to the Sm antigen (a nuclear antigen different from DNA or nucleoprotein) appears to be quite specific for SLE patients, even in those with mild disease. Unfortunately, since this antibody occurs in only about 20 percent of the SLE patients, a negative test again does not rule out the diagnosis of SLE. An LE preparation should also be ordered, although this test is positive in only about 70 percent of patients and may be negative, particularly in patients with mild disease. In addition, since the test usually depends on the number of polymorphonuclear leukocytes present in the patient's peripheral blood, and since leukopenia is common in children with SLE, the test may be negative. Again, a negative LE cell preparation does not rule out the diagnosis of SLE.
If the child has a positive ANA in a diffuse or speckled pattern and a significant titer, the pediatrician should continue to pursue the laboratory work-up of the patient.
A complete blood count is essential for the work-up in children. Anemia, thrombocytopenia and leukopenia are more common among children than among adults with the disease. In a child with joint pain, the presence of one of these abnormalities is strongly suggestive of SLE and should lead to more intensive investigation. The anemia most commonly seen is not hemolytic, but the patient should be investigated for hemolysis. A positive Coombs' test occurs in SLE patients who are not hemolyzing, and is another indication of the presence of SLE. A prothrombin time and partial thromboplastin time should be ordered; if the PTT is prolonged the patient should be investigated for the presence of a lupus anticoagulant.
In addition to the complete blood count and platelet count, a urinalysis should be performed on any child suspected of having SLE. If it is abnormal in a child with joint pain and a positive ANA, the diagnosis of SLE should be strongly suspected, the child admitted to hospital, and a complete work-up performed including renal tests. Procrastination or delay of any sort in such a patient may lead to rapid exacerbation of the disease before adequate therapy can be instituted. The work-up for lupus nephritis in children should include consultation with a nephrologist and renal biopsy.
If the urinalysis is normal at the time the child is first seen, the pediatrician should continue to obtain urinalyses at weekly intervals while awaiting the results of the other laboratory studies. In patients with mild symptoms at the time of first observation, the disease may progress rapidly over the next few weeks or months, and the onset of renal disease will be overlooked unless the patients are tested for urinary abnormalities at very frequent intervals.
The diagnosis of SLE can be supported by other laboratory tests. A positive VDRL and a negative FTA is strongly suggestive of SLE. The diagnosis of SLE can be made in a child with a sun-sensitive rash and ieukopenia, a false positive VDRL and a positive ANA in significant titer. Therefore, the VDRL should be ordered as part of the work-up for SLE.
The complement system is involved in the pathogenesis of immune complex disease seen in SLE patients. A low serum euro3 and euro4 are nearly always present in patients with active lupus nephritis. In addition, many patients who have clinical evidence of active disease (arthritis, rashes, pleurisy, etc.) have low serum Cj and Ca levels.3'5 The presence of low complement levels is very suggestive of SLE; on the other hand, however, not every patient with untreated active disease will be found to have low complement levels. It is very important for the pediatrician to remember that patients with certain genetic complement deficiences may develop a lupus-like syndrome or clear-cut SLE. Thus, a total hemolytic complement assay should be obtained in order to determine an absence of one of the complement proteins in a patient with manifestations of SLE. Some of the complement-deficient patients have been found to have very mild disease; it is especially important to rule out genetic complement deficiency in children with a mild SLE-like syndrome (arthralgias, some rashes, low titer ANA, negative anti-DNA antibodies). If the total hemolytic complement level is zero, a complete study of the levels of the individual complement proteins should be conducted. Additionally, it is important to test the parents and siblings of the symptomatic child. The most common genetic abnormalities associated with SLE are absent C2 and C4. In these patients, the Cj level is usually normal or elevated. If the child has a deficiency of a complement protein such as C*, it is important to know this at the onset of treatment. The serum level of d in such a patient will never return to normal and, therefore, cannot be used to monitor the efficacy of the treatment.
The erythrocyte sedimentation rate should be checked using the Westergren method.* This is elevated in nearly all children with SLE. This test may be a good indication of future disease activity, if an initially elevated ES R falls to normal as the treatment is instituted and the patient responds.
Once the diagnosis of SLE is established, it is important to have a baseline for the above laboratory tests prior to institution of therapy. These tests will be used to help estimate disease activity during management of the patient, which may last for 20 to 30 years.
A physician, experienced with either children or adults afflicted with SLE, should manage patients with the disease. It is important that the appropriate laboratory tests, such as Cj, C4 and anti-native DNA antibody test, be readily available.5 The mainstay of therapy is corticosteroids, which are used to suppress clinical and laboratory evidence of active disease. In patients with major dermatologie manifestations (suchas rash.alopecia, mucosal ulcers, etc.) antimalarial drugs should also be employed. (Side effects of the antimalarials are minimal, with the exception of the major problem of oculotoxicity. All patients should be seen by an ophthalmologist prior to institution of antimalarials, seen at six month intervals while on the drug, and at yearly intervals if the drug is discontinued.) The initial prednisone therapy should always be given in a four-times-daily dosage. After initial response, the dose should be gradually tapered; it can eventually be changed to a twice-daily dosage. The laboratory tests, especially the Cs and anti-DNA antibodies, should be monitored, with initial improvement of the abnormalities used as an indication for reduction of the prednisone dose. If both of these parameters become abnormal as the dose is lowered, it should be increased again to the level at which these tests were normal.
We have successfully managed many children with SLE with extremely low doses of prednisone. The dose is lowered very gradually to 10 mgm per day and then is reduced by one milligram increments every two to three months. Most frequently, this is accomplished without a major flare of the disease. Since relapses usually involve the same abnormalities seen at the time of diagnosis, these should be established. For example, in some SLE patients the first indication of an impending flare-up may be a rising ESR; in other patients, the hemoglobin may fall to the level present at diagnosis prior to the onset of clinical symptoms of rash, arthritis or other manifestations of the disease. These flare-ups may occur months to years after the patient has initially responded to therapy. Since each patient is different, it is important to know which laboratory tests reflect impending disease activity in that patient. In most patients, the anti-DNA antibody level is an extremely sensitive indicator of active disease, as may be the Cj level.5 Both of these abnormalities (high antiDNA antibody levels and low serum C3 levels) are usually present in severely ill patients. If these tests are normalized by treatment and then both of them return to the initial abnormal levels as the patient is followed under therapy, the disease should be considered active and the therapy altered to return these parameters to within the normal range. Persisting high anti-DNA antibodies levels in association with persisting low complement levels nearly always lead to onset or exacerbation of renal disease.
Immunosuppressive agents should be used only by those with ongoing protocols, since their use is still controversial and side effects are significant.6
Any local infection should be treated immediately and blood cultures obtained.
The survival of children with SLE has improved markedly over the past 10 to 15 years. In our patients studied in New York before 1968, the five-year survival in children (under 16 years of age) with renal disease was 42 percent compared with our recent finding of 89 percent survival in children with renal disease. The data on survival of children without renal disease was 72 percent in the older study and 100 percent in our recent group of patients.7 Both of these studies were based on duration of disease since the time of diagnosis.
This improved survival of our patients followed during different periods of time may be attributed to a variety of factors. In the New York patients, no attempt was made to normalize laboratory tests and no tests for Ci and anti-DNA antibodies were available. Clearly the availability of better antibiotics may have made a major difference, since fewer deaths from infection occurred in the later series. It is important to point out that immunosuppressive drugs were not used in either group of patients.
It is interesting to note that, in our recent series, fewer patients with renal disease were dead after five years than in our previous groups of patients. Even in our children with biopsy-proven diffuse lupus glomerulonephritis, survival occurred for longer than five years. Thus, we feel that the improved survival may be related to monitoring complement and anti-DNA antibody levels, which may become abnormal before the urinalysis.
Children with SLE are now reaching adulthood and the question of use of birth control pills must be considered by the physician. In general, weprefertheuseofaformof contraception other than birth control pills. Pregnancy should be avoided unless the disease is in stable remission for at least six months and the patient requires less than 10 to 15 mgm of prednisone per day. Many of our patients have had successful, uncomplicated pregnancies later in the courses of their diseases. Thus, preadolescent girls with SLE should be reassured that they can live normal lives and have children, although they will require constant supervision by their physicians. If a patient becomes pregnant during a period of active clinical or serologie disease, the disease activity should be brought under control. Abortions should be avoided during active disease.
Finally, children with SLE should be encouraged to be as active as possible in school and at play. This activity may help prevent osteoporosis and certainly has a beneficial effect on their psychologic attitudes towards themselves. The support of family, nurses, and other children with the disease is very important, especially during the initial stages of the disease when corticosteroidinduced disfigurement occurs. The first period of the illness may be terrifying to both parents and child; support from knowledgeable adults and children with the disease may be essential.
SLE in children is now a disease which carries a favorable prognosis for most patients. The management of the disease requires considerable effort on the parts of the physician, the family and the child - all of whom need to cooperate so that successful control can be accomplished.
1. Serdula MK and RhoadsGG. Frequency of SLE in different ethnic groups in Hawaii. Arthritis Rheum 22:328, 1979.
2. Meislin AG and Rothfield N. Systemic lupus erythema tos us in childhood: Analysis of 42 cases, wilh comparative data on 200 adult cases followed concurrently. Pediatrics 42:37, 1968.
3. Lehman TJ, et al. The role of antibodies directed against double-stranded DNA in the manifestations of SLE in childhood. J Pediatrics 96:657, 1980.
4. Kornreich H. Systemic lupus erythema t osus in childhood. In: Rheumatic Disorders In Childhood. Ansell BM (ed). Clinics in Rheumatic Diseases, vol. 2, number 2, August 1976.
5. Garin EH. et al. The significance of serial measurements of serum complement Ci and Cj components and DNA binding capacity in patients with lupus nephritis. Clin Nephrol 12:148, 1979.
6. Cameron JS. Immunosuppressive agents in the treatment of the nephrotic syndrome and glomerulonephritis in childhood. Paediatrician 8:364, 1978.
7. Abeles M. et al. Systemic lupus e ry t he ma tos us in the younger patient: Survival studies. J Rheumatol 7:515, 1980.