SASP is a useful DMARD in RA and is probably useful in early ankylosing spondylitis, psoriatic arthropathy and reactive arthritis. It shares many of the characteristics of other DMARDs such as gold and penicillamine. It produces improvement in clinical and laboratory parameters of disease activity, with a slow onset of action--8-12 weeks elapse before beneficial effect is noted. SASP probably slows radiological progression of RA but definitive proof is difficult to obtain. Although side-effects are common, these are often managed by a reduction in dose, and serious adverse events requiring cessation of therapy are uncommon. Serious side-effects do occur however, and regular monitoring with full blood counts is recommended. The mechanism of action of SASP is unknown and this remains one of the principal areas of research interest.

Since systemic lupus erythematosus most frequently affects women of childbearing years, the management of patients during pregnancy is an important and common problem facing the clinician. This review concerns the effects of pregnancy on the course of maternal disease and fetal well-being. On the maternal side are the problems of renal disease which may exacerbate and be difficult to differentiate from pre-eclampsia especially when occurring in the third trimester. An active urinary sediment, falling C3 and CH50 and elevated complement split products of the alternative pathway and terminal attack complex may serve as useful parameters of lupus activity. In general, maternal disease is not an imposing threat and prospective studies suggest that the exacerbation rate is not significantly greater in the pregnant lupus patient than in the non-pregnant patient. On the fetal side are the problems of placental insufficiency and in utero attack on developing organs. Maternal antibodies such as those reactive with negatively charged phospholipids are associated with second trimester miscarriages and suggested, but not firmly established, thrombosis of placental vessels. The placental transfer of maternal antibodies against components of the rapidly expanding group of SSA/Ro-SSB/La ribonucleoproteins is strongly implicated in the transient and permanent manifestations of neonatal lupus. Using various techniques for defining the specificity of the antibody response most associated with heart block, the data suggest that mothers whose sera contain antibodies which recognize antigens of SSA/Ro-SSB/La on SDS-immunoblot are at greatest risk. In the absence of antibodies to SSB/La, mothers whose sera contain antibodies reactive only to bovine SSA/Ro by ELISA do not appear to be at high risk. A rational approach to in utero treatment of autoantibody mediated fetal myocarditis includes plasmapheresis and the use of dexamethasone. Finally, the safety of the commonly used medications for the treatment of lupus such as the nonsteroidal anti-inflammatory agents, glucocorticoids and anti-malarials during gestation and breast feeding, is addressed.

NLE is manifested most typically as transient subacute cutaneous lupus lesions or isolated complete congenital heart block. Babies with NLE have maternal anti-Ro/SSA, anti-La/SSB, or anti-U1RNP autoantibodies. It is presumed, but not proven, that transmission of these autoantibodies through the placenta to the baby has resulted in disease. However, other factors such as inflammatory cells or complement activation may be necessary for disease to be expressed. About half of babies reported with NLE have had heart disease and about half have had skin disease. There have been a few reports of liver disease and a few of thrombocytopenia. Any combination of these findings is possible in a given infant. Possibly, other haematologic abnormalities, pneumonitis or neurological disease could occur, but the evidence that these other abnormalities are part of NLE is scant. Mortality in NLE has occurred in babies with severe cardiac disease. It is estimated that 10% or more of babies with cardiac NLE die in infancy. Of the remainder, perhaps half will require permanent pacemaker implantation. Thus, there is substantial morbidity and mortality with cardiac NLE. The skin disease, by contrast, is not serious and typically leaves little or no residua. Individuals who have had NLE may develop connective tissue disease in adulthood. Whether this is a common or an unusual occurrence is not yet known, since a large cohort of individuals with NLE has not yet been followed into adulthood. Mothers of babies with NLE are often initially asymptomatic. With time, they frequently develop connective tissue disease symptoms. In our experience, these have been largely symptoms of Sjögren's syndrome and have generally not been debilitating. Most babies of mothers with anti-Ro/SSA, anti-La/SSB, or anti-U1RNP autoantibodies do not develop NLE. There is no way to determine prospectively which fetus or infant will be affected and which of those affected will have life-threatening disease. Systemic therapies should be reserved for those infants who have life-threatening manifestations of NLE. It is not yet known whether treatment of the mother during gestation will be beneficial or harmful to fetuses with severe NLE cardiac disease.

Systemic alterations of the maternal inflammatory and immune system occur during pregnancy. These changes alone are unlikely to be responsible for the acceptance of the fetal semiallograft. Numerous local events at the maternal-fetal interface appear to be more important. The alterations of the maternal inflammatory and immune systems are subtle enough for no significant increase of infections or malignancy to be apparent. However, 75% of women with rheumatoid arthritis are clinically improved during pregnancy. The effects of pregnancy on polymorphonuclear cells are not likely to be responsible because cell function actually appears enhanced in vivo, despite the fact that pregnancy serum is suppressive in vitro. There is no clear evidence for reduction of monocyte/macrophage function during pregnancy, either in vivo or in vitro. It is unlikely that modulation of B cell phenotype or function is responsible because no suppression is noted, either in vivo or in vitro. Selected products of B cells, immune complexes, appear to be reduced during pregnancy. In patients, the reduction in the concentration of complexes may be due to adsorption by the placenta. The importance of this reduction as a causative factor in the improvement of women with rheumatoid arthritis during pregnancy remains to be determined. Natural killer cell cytotoxicity is decreased during pregnancy. This may in part be due to the release of progesterone induced blocking factor. It is also possible that circulating factors, capable of inhibiting IL-2 release or IL-2 function in vivo, might be responsible. Natural killer cytotoxicity can be normalized by incubation with IL-2. It is unclear how the reduction of natural killer cell activity might systematically affect inflammation or immunity in vivo during pregnancy. In vivo delayed type hypersensitivity appears somewhat reduced during pregnancy. This observation appears consistent with the improvement of rheumatoid synovitis, which is also thought to be T cell mediated. T cell function, measured in vitro, generally appears normal. However, most recent studies have employed mitogens, such as PHA, which is not physiological. Subtle defects involving antigen processing or antigen presentation might be missed in this system. These observations suggest that circulating factors might be important in modulating the cell mediated immune system, in vivo, during pregnancy. While anti-HLA-DR antibodies eluted from the human placenta may be effective therapy in patients with rheumatoid arthritis, their occurrence is too infrequent to account for the improvement seen in afflicted patients.(ABSTRACT TRUNCATED AT 400 WORDS)

The natural inclination of patients with rheumatic diseases wishing to become pregnant or to breast feed will be to take as few medications as possible. The guidelines outlined above can be used to balance the risk of drug effect on the fetus or neonate with the risk of inducing a flare in disease activity by stopping the drug. Although there are situations where no information on drug behaviour during pregnancy or lactation exists, some guidelines can be developed from a knowledge of the drug's inherent metabolism. In the majority of the rheumatic diseases, disease activity can be reduced to a minimum using the smallest possible dose of drugs known to be safe in pregnancy and lactation, thus providing minimum risk to mother, fetus and neonate.

Patients with rheumatic diseases who become pregnant are justifiably categorized as having high-risk pregnancies. Utilizing a multidisciplinary approach, including perinatologists, rheumatologists and anaesthetists, successful pregnancies have become the rule rather than the exception. However, women with rheumatic disease are particularly prone to develop serious obstetric problems which often result in early hospitalization and delivery. Although vigilant obstetric care improves perinatal outcome, prematurity will continue to be a major problem complicating pregnancies in women with rheumatic disease.

Sex hormones have physiological and pathological (autoimmune conditions) effects on the immune system. Studies in experimental animal models of human autoimmune diseases have clearly shown that sex hormones regulate the expression, severity and course of autoimmune diseases. Sex hormones affect the function of T, B and NK cells, and macrophages. Precisely how sex hormones affect lymphocytes is a highly complex question. Sex hormones can modulate the immune system, perhaps directly (e.g. thymic reticular tissue), or indirectly via host and many oestrogen target tissues, including the central nervous system hypothalamic-pituitary axis (the neuroendocrine tissues). The effects of sex hormones on the immune system (immunosuppression or immunopotentiation) may vary, even with the same hormone. For example, oestrogen can increase IgA levels in the uterus, but decrease IgA levels in the vagina or have no effect in lacrimal tissues (Sullivan, 1989). Therefore the effects of sex hormones on the immune system cannot be generalized but must be evaluated independently. Some of the reasons for variability in results have been reviewed in detail elsewhere (Steinberg et al, 1979; Ansar Ahmed et al, 1985b). These include, dose of hormones, age and sex-hormonal status of animals, route and time of administration, the immunocompetence of the host, stress, the metabolism of hormones (e.g. metabolism of testosterone to oestrogen) resulting in alteration of biological activity, and differential response to various antigens. The initial encounter of sex hormones with the type of target cells, the variability of secondary messengers and gene activation events are other important considerations. The effects of sex hormones on the immune system to modulate immune responses are unequivocal. The burgeoning advances in cellular immunology, endocrinology and molecular biology, should provide a better understanding of: (1) the interactions of hormones with the immune system; (2) how hormones activate specific genes; and (3) how hormones influence intracellular communication. In a clinical situation, it is hoped that androgenic compounds which lack virilizing effects, but possessing the desired immunomodulatory effects, will eventually be synthesized. These hormone analogues, in combination with specific (non-toxic) oestrogen antagonists, may offer new therapeutic avenues.

The onset of pregnancy appears to result in the amelioration of rheumatoid disease activity. This is most likely to be related to the change in hormonal, and thus immunological, status of the mother. Spontaneous abortion appears to be increased in mothers with rheumatoid disease. Whether this is due to disease status or to drug therapy is not entirely clear for all cases, but there is a suggestion that it could be the disease state. If at all possible, drugs should be avoided during pregnancy but simple anti-inflammatory drugs in low dosage probably do not produce any major problems. Immunosuppressant drugs should be avoided at all times except when the mother's health is at serious risk. The continued use of disease remitting agents throughout pregnancy is probably not necessary and there is still sufficient question as to whether these drugs could be potentially toxic to the infant. Primum non nocere.

Despite the considerable interest in antiphospholipid antibodies during the last 10 years, the clinical disorders with which they are associated are still not well understood (Love and Santoro, 1990). The concept of the antiphospholipid syndrome (APS) and a means by which this can be defined (Harris, 1987) may aid us in better classifying these patients, evaluating their clinical features, and in conducting prospective studies. Standard methods for measurement of anticardiolipin antibodies (Harris et al, 1987b; Harris and the Kingston Anticardiolipin Standardization Group, 1990) and more attention to measurement of the lupus anticoagulant (Triplett and Brandt, 1988) will ultimately play an important role in improving the diagnosis of these patients. The initiation of the Kingston AntiPhospholipid Study (KAPS) group in January 1988 paves the way for future multicentre treatment trials of these patients (Harris, 1988; Triplett, 1989; Recker and Leff, 1989). For the practising physician managing a pregnant woman with the APS, there are few clear guidelines for treatment. An appropriate approach would be to assemble a multidisciplinary team for pregnancy management, exclude other causes of pregnancy loss and select the drug regimen which will cause the least harm and has some reported efficacy. Most importantly, the patient should be followed carefully throughout her pregnancy.

Much strain is placed on the shoulder joint in all sports. Sports injuries may be caused by a single, violent incident. Fractures of the clavicle, acromioclavicular and sternoclavicular traumas, isolated fractures of the greater tuberosity and shoulder dislocations are frequent clinical pictures which today can be managed by well established procedures. Other sports injuries, however, involve over-use of the shoulder which, through repetitive harmful movements, can bring about microtraumatic lesions of the rotator cuff, glenoid labrum, acromioclavicular joint and certain peripheral nerves around the suprascapularis. In all cases, a programmed, comparative, clinical examination and complementary, sequential examinations will indicate suitable curative and preventive treatment.

Rotator cuff problems present with shoulder pain on repetitive overhead activity. Chronic irritation may develop into impingement tendonitis, with weakness of abduction and external rotation and night pain. Conservative management with rest, anti-inflammatory medicine and physiotherapy resolves the majority of symptoms. If these persist, surgical decompression affords good relief of pain.

Glenohumeral instability is an important cause of shoulder pain and disability in an active population. An awareness of the prevalence of recurrent instability, either in the form of dislocation or subluxation, is particularly useful in the assessment of the young athlete presenting with shoulder pain. Young adults presenting with rotator cuff tendinitis may have an underlying instability as the primary cause of their problem. A careful clinical examination should determine whether the instability is voluntary or involuntary, of traumatic or atraumatic onset, and the primary direction of the instability, as these factors have important implications with regard to treatment. Anterior glenohumeral instability is most common and the incidence of recurrent instability following on from an initial dislocation is high in the young active patient. An intensive rehabilitation programme is indicated for all initial dislocations or subluxations but surgery may become necessary after failure of conservative treatment. Care must be taken to determine accurately those patients with voluntary or multi-directional instability and a longer trial of conservative treatment is indicated here, as results of operative treatment in those cases are less favourable. Conservative treatment should be directed at strengthening the dynamic stabilizers of the shoulder joint, notably the rotator cuff muscles. Additional X-ray views are needed to demonstrate all the radiological changes associated with recurrent instability and further evaluation with examination under anaesthesia and arthroscopy is beneficial in the assessment of these patients. Arthroscopic surgery also has a role in the treatment of patients with symptomatic labral pathology and is now being used to perform stabilization procedures in selected cases. Many operative procedures have been described for stabilization of the shoulder and these should be directed at correcting the pathology present. Restoration of the patient's flexibility and strength postoperatively is essential, especially in the athlete in order to allow a full return to sporting activity.