Impact of selective cyclooxygenase-2 inhibitors on anti-ulcer medication and non-steroidal anti-inflammatory drug use in patients with rheumatic disease.

BACKGROUND: Anti-inflammatory medications are the most common treatment for rheumatic disease in Australia. Recent years have seen large increases in the use of selective cyclooxygenase-2 (COX-2) inhibitors. Predictors of use, costs and benefits of the new medications have not been evaluated. AIMS: To determine trends in selective COX-2 inhibitor, non-steroidal anti-inflammatory drug (NSAID) and anti-ulcer medication (AUM) prescription following the introduction of selective COX-2 inhibitors; to determine predictors of selective COX-2 inhibitor, NSAID and AUM prescribing and to perform a limited evaluation of the costs and benefits associated with the introduction of selective COX-2 inhibitors. METHODS: Groups of consecutive patients attending a hospital rheumatology clinic, private rooms of consulting rheumatologists and a dermatology outpatient clinic were surveyed by investigator-administered questionnaire on three separate occasions. Information was sought about AUM, NSAID and selective COX-2 use and about factors likely to influence selective COX-2 prescribing. Sampling was carried out at 3, 10 and 16 months after the release of COX-2 selective inhibitors in Australia. The final survey was 3 months after Pharmaceutical Benefits Scheme (PBS) listing of celecoxib in Australia. Costs of treatment were calculated from survey findings of frequency of drug use as well as published drug prices and hospitalisation costs. RESULTS: Four-hundred and fifty-eight patients were surveyed. From the 3 months post-release to the 3 months post-PBS listing, a period of 13 months, COX-2 use in rheumatology patients increased from 18 to 57%. De novo prescription of selective COX-2 inhibitors increased from 42 to 61%. During the same period there was a fall in both NSAID (43-20%) and AUM use (41-27%). Neither selective COX-2 inhibitor nor NSAID prescription was related to risk factors for gastro-intestinal (GI) complications, but AUM use was found to correlate strongly to histories of gastroscopy, GI ulceration or GI bleed. The calculated increase in the cost of treatment was $1 033 002/10 000 patients per year. The net cost per serious GI event prevented was $71 736, compared with the normal cost of treatment of $2004. CONCLUSIONS: Among rheumatology patients, selective COX-2 inhibitors have largely replaced NSAIDs and have resulted in a reduction in AUM consumption, but prescribing patterns for selective COX-2 inhibitors have not been related clearly to risk factors for GI complications. The introduction of selective COX-2 inhibitors has been associated with a significant increase in expenditure.

Increased binding of D-penicillamine to monocytes in rheumatoid arthritis.

Long-term therapy of D-penicillamine (D-Pen) for rheumatoid arthritis (RA) is associated with a fall in rheumatoid factor, but many patients develop autoantibodies. In vitro binding of D-Pen to human peripheral blood monocytes was examined in 37 patients with RA and 75 healthy subjects. Mononuclear cells were reacted with D-Pen coupled to a fluorescein isothiocyanate-bovine serum albumin (BSA) conjugate in the presence of sodium azide and BSA, and analyzed by flow cytometry. Patients showed significantly higher D-Pen binding to monocytes than did healthy subjects. The proportion of monocytes binding D-Pen increased with age in the patients but not in healthy subjects. None of 6 patients who had D-Pen-induced autoimmune side effects was associated with increased D-Pen binding though patients with therapeutic responses showed high D-Pen binding. These results suggest that D-Pen binding to monocytes may be important in mediating therapy and inducing autoimmune side effects.

Genetic markers in rheumatoid arthritis relationship to toxicity from D-penicillamine.

In a 3-centre study involving 144 patients with rheumatoid arthritis (RA), a relationship between side effects from D-penicillamine and HLA antigens, allotypic markers of the IgG heavy chain (Gm) and allotypes of complement components Bf, C4A and C4B was sought. There was a significant association between proteinuria induced by D-penicillamine and the antigens DR3 and B8. However, the presence of DR2 seemed to protect against the development of proteinuria. Thrombocytopenia from D-penicillamine was significantly associated with HLA-A1 and DR4; 15 of 23 patients who possessed both antigens developed thrombocytopenia (p less than 0.001 uncorrected, approximate relative risk (RR) = 5.5). A null complement allele located at the C4B locus (C4BQO) was also associated with thrombocytopenia from D-penicillamine (p less than 0.005, RR = 17.3). Our study confirms the findings from other series which indicate that there is a genetic predisposition for the development of proteinuria from D-penicillamine in RA and suggests that this may also be the case in D-penicillamine induced thrombocytopenia.

D-penicillamine induced suppression of B cell function: in vivo effect of D-penicillamine.

We assessed the immunoglobulin secretory capacity of circulating B lymphocytes in 9 patients with classical rheumatoid arthritis (RA) before and after treatment with D-penicillamine. Peripheral blood lymphocytes (PBL) from patients with RA spontaneously synthesized more IgG and IgA than normals. The secretory rate of rheumatoid PBL could not be induced by the polyclonal activator, pokeweed mitogen (PWM). The presence of D-penicillamine in cultures significantly suppressed PWM stimulated immunoglobulin synthesis of control PBL but did not inhibit synthesis of mitogen stimulated RA PBL. After D-penicillamine therapy for 3 months immunoglobulin synthesis by PBL from patients with RA was reduced with or without PWM. The T mu:T gamma ratio was also decreased after therapy. These results support the hypothesis that D-penicillamine selectively impairs helper T cells in vivo, preventing the T dependent expansion and activation of B cells characteristic of RA.

Regional migratory osteoporosis.

The case history of a patient with regional migratory osteoporosis and associated electromyographic abnormalities is reported. The changes seen in this patient suggest that radiculopathy secondary to traumatic ischaemia may be the pathogenic basis of this disorder.

Lateral subluxation of the atlas in rheumatoid arthritis: a case report and post-mortem study.

The atlantoaxial joints of a patient with lateral subluxation of the atlas were studied post mortem by dissection. There was severe erosion of the left lateral atlantoaxial joint and an asymmetrical erosion of the cartilages of the median atlantoaxial joint which prevented reduction of the lateral subluxation. The atlantoaxial ligaments were remarkably intact. It is emphasised that osseous factors rather than ligaments confer lateral stability to the atlantoaxial joints and that destruction of these leads to lateral subluxation.

Gold colitis.

A case of severe colitis, which developed in a patient with rheumatoid arthritis after administration of sodium aurothiomalate, is reported. Early recognition of this rare, but potentially fatal, complication is essential. Rectal biopsy should be performed as part of the early assessment.

Penicillamine-induced immunosuppression: in vitro studies of inhibition of immunoglobulin synthesis.

An in vitro system of T-dependent polyclonally activated B lymphocytes has been used to identify the cellular basis of immunosuppression induced by penicillamine. Suppression of immunoglobulin synthesis was dose dependent, and required the drug to be present at the initial stages of the culture. Coculture studies identified T helper cells as the primary target. Study of T cell membrane receptors identified a susceptible subpopulation characterized by its inability to form tight sheep red blood cell rosettes, a delay in the generation of T mu receptors and reduced "help" for polyclonally activated B cells. B lymphocytes were not primarily suppressed by penicillamine.

Immune complexes in acute pancreatitis.

Immune complexes were detected in the sera of ten of 22 patients with acute pancreatitis using a Clq deviation assay. Five of these were positive using a second technique. There was no correlation between immune complexes and clinical or aetiological features of the pancreatitis. Two patients with immune complexes developed a benign and transient pancreatic polyarthritis. Immune complexes may provide one common path in the sequence of pathogenic events that lead to pancreatitis.