|
You mean a study like this one?
Minocycline in Rheumatoid Arthritis: A 48-Week, Double-Blind, Placebo-Controlled Trial
Barbara C. Tilley; Graciela S. Alarcon; Stephen P. Heyse; David E. Trentham; Rosemarie Neuner; David A. Kaplan; Daniel O. Clegg; James C. C. Leisen; Lenore Buckley; Sheldon M. Cooper; Howard Duncan; Stanley R. Pillemer; Marilyn Tuttleman; and Sarah E. Fowler
15 January 1995 | Volume 122 Issue 2 | Pages 81-89
Objective: To assess the safety and efficacy of minocycline in the treatment of rheumatoid arthritis.
Design: A double-blind, randomized, multicenter, 48-week trial of oral minocycline (200 mg/d) or placebo.
Setting: 6 clinical centers in the United States.
Patients: 219 adults with active rheumatoid arthritis who had previous limited treatment with disease-modifying drugs.
Measurements: As the primary outcomes, 60 diarthrodial joints were examined for tenderness, and 58 joints were examined for swelling (hips excluded). Grip strength, evaluator's global assessment, morning stiffness, Modified Health Assessment Questionnaire, patient's global assessment, hematocrit, erythrocyte sedimentation rate, platelet count, and IgM rheumatoid factor levels were also assessed; radiographs of both hands and wrists were taken.
Results: 109 and 110 patients were randomly assigned to receive minocycline and placebo, respectively. At entry, demographic, clinical, and laboratory measurements were similar in both groups. Most patients had mild to moderate disease activity and some evidence of destructive disease. At the week 48 visit, 79% of the minocycline group and 78% of the placebo group continued to receive the study medication. At 48 weeks, more patients in the minocycline group than in the placebo group showed improvement in joint swelling (54% and 39%) and joint tenderness (56% and 41%) (P < 0.023 for both comparisons). The minocycline group also showed greater improvement in hematocrit, erythrocyte sedimentation rate, platelet count, and IgM rheumatoid factor levels (all P values < 0.001), and more patients receiving minocycline had laboratory values within normal ranges at 48 weeks. For the remaining outcomes, P values ranged from 0.04 to 0.76, all greater than the critical value of 0.005 (Bonferroni adjustment for multiple comparisons). The frequency of reported side effects was similar in both groups, and no serious toxicity occurred.
Conclusions: Minocycline was safe and effective for patients with mild to moderate rheumatoid arthritis. Its mechanisms of action remain to be determined.
For current author affiliations, see end of text.
*For members of the MIRA Trial Group, see the Appendix. For current author affiliations, see end of text.
Deceased.
Rheumatoid arthritis is a chronic, sometimes incapacitating systemic disease of unknown cause that affects at least two million Americans [1, 2]. Various pharmacologic agents are used to manage rheumatoid arthritis, but none is completely effective. On the basis of the theory that persistent Mycoplasma infection may cause rheumatoid arthritis, some physicians have prescribed and reported benefit from lengthy courses of tetracyclines [3, 4]. A small 1-year clinical trial comparing tetracycline, 250 mg/d, with placebo could not show significant benefit in patients with rheumatoid arthritis [5]. Because of these contradictory outcomes, the treatment of rheumatoid arthritis with antibiotics has remained controversial.
Tetracyclines have several nonantibiotic effects that may be beneficial for patients with rheumatoid arthritis. Minocycline, a semi-synthetic tetracycline that is rapidly absorbed and has a prolonged half-life, inhibits collagenase activity of gingival fibroblasts from diabetic rats, even in a germ-free environment [6, 7]. Oral minocycline at a dose of 100 mg twice a day for 10 days reduced collagenase activity in the synovial tissue of patients with rheumatoid arthritis [8]. Moreover, the addition of minocycline to synovial tissue cultures from patients with rheumatoid arthritis also resulted in the inhibition of this enzyme.
Other properties of tetracycline analogs include inhibition of protein synthesis by rapidly dividing cells [9], perturbation of leukocyte functions [10, 11], interference with lymphocyte proliferation [12-14], and anti-inflammatory effects [15] that are probably related to its antioxidant activity [16, 17]. Minocycline can also suppress the induction of experimental arthritis in rats [18].
Recently, favorable results from uncontrolled clinical studies of minocycline in rheumatoid arthritis were reported [19, 20]. In addition, Kloppenburg and colleagues [21, 22] observed improvement when minocycline was added to baseline nonsteroidal anti-inflammatory drugs (NSAIDs) and disease-modifying antirheumatic drugs or corticosteroid therapy in a 26-week randomized, placebo-controlled trial. We conducted a 48-week, randomized, double-blind, multicenter trial to determine the efficacy and safety of minocycline in treating rheumatoid arthritis when it is added to background NSAIDs or low-dose prednisone therapy in patients not receiving concomitant disease-modifying antirheumatic drugs.
Methods
Top
Methods
Results
Discussion
Author & Article Info
References
Protocol Development
The protocol for the Minocycline in Rheumatoid Arthritis (MIRA) Trial was developed by the investigators from the clinical centers, the coordinating center, and the National Institute of Arthritis and Musculoskeletal and Skin Diseases on the basis of protocols for clinical trials of therapies for rheumatoid arthritis that were previously done by the Cooperative Systematic Studies of Rheumatic Diseases (CSSRD) program [23, 24]. Inclusion criteria for those CSSRD trials were two of the following: nine or more tender joints capable of response, an erythrocyte sedimentation rate of 28 mm/h or greater, and morning stiffness lasting 45 minutes or longer. To simplify recruitment of patients, we modified these criteria to require the presence of nine or more tender joints capable of response with no minimum specified for either the erythrocyte sedimentation rate or for morning stiffness. The study protocol was reviewed and approved by institutional review boards at each participating clinical center.
Patients
To be eligible for the minocycline trial, patients had to meet the 1987 American Rheumatism Association (now the American College of Rheumatology) classification criteria for rheumatoid arthritis [25]. Patients were required to have active disease, defined as nine or more tender joints and six or more swollen joints capable of assessable improvement. Patients had to be 18 years of age or older and to have experienced onset of rheumatoid arthritis at 16 years of age or older. Patients receiving NSAIDs or low-dose corticosteroids (prednisone equivalent of a daily dose of 10 mg or less) were required to be receiving stable concurrent medication for 4 weeks before entering the study; these medications were to be kept stable throughout the study. One change from equivalent doses of one NSAID to another was permitted during the trial. Phenylbutazone, oxyphenbutazone, or parenteral steroids (including intra-articular corticosteroids) and disease-modifying antirheumatic drugs were not allowed for 4 weeks before the study; the use of these drugs during the trial was a protocol violation. Women were required to be postmenopausal, surgically sterile, or practicing a reliable method of contraception and to have a negative pregnancy test result.
Before randomization, eligible patients for the minocycline trial could have received therapeutic trials of hydroxychloroquine and, at most, one other disease-modifying antirheumatic drug (oral or parenteral gold salts, D-penicillamine, azathioprine, methotrexate, or sulfasalazine) unless therapy was discontinued for toxicity. Patients could have tried any number of disease-modifying antirheumatic drugs if discontinued for toxicity, provided that no drug was administered for more than 3 months or, for parenteral gold salts, was not given in a cumulative dose of more than 500 mg. These exclusion criteria were designed to avoid recruiting patients with a history of repeated failures of therapeutic trials of disease-modifying antirheumatic drugs but did allow entry of those who had discontinued therapy with these drugs because of toxicity before prespecified dosages or durations were reached.
Patients were not eligible at screening if they had known hypersensitivity to tetracycline; required long-term tetracycline therapy for another disorder; required ongoing therapy with antacids containing aluminum, calcium, or magnesium; or had received anticoagulant therapy. We excluded patients with chronic illnesses that would limit successful participation in the trial and patients classified as American Rheumatism Association-Steinbrocker functional class IV [26].
Three patients who met eligibility criteria at their screening visit were found to have too few affected joints at their baseline (randomization) visit and were inadvertently randomized. We included these patients in the analysis. All three patients were in the placebo group and were not considered responders to treatment. Removing these patients did not change the results of the trial.
Study Design
The MIRA trial was a six-center, double-blind, randomized, placebo-controlled trial comparing minocycline and placebo in treating rheumatoid arthritis. Patients were randomly assigned within the clinical center (that is, the clinical center was a stratifying factor), and randomization was blocked using randomly chosen block sizes between 4 and 6 to minimize the possibility that staff would guess the next treatment assignment. For 48 weeks, patients received two 50-mg capsules of minocycline hydrochloride or a visually similar placebo taken with water twice daily on an empty stomach. To minimize interference with the absorption of minocycline, patients were instructed not to ingest food or iron within 2 hours of receiving study medication. Investigators were allowed to adjust the dosage (without breaking the treatment code) if patients experienced side effects.
Measurements
The following measurements were obtained at the baseline and randomization visits and every 12 weeks thereafter.
Evaluator Assessments
Sixty diarthrodial joints were examined for the presence or absence of tenderness, and 58 were examined for swelling (hips excluded). Joint counts for tenderness and for swelling were the sum of the number of affected joints [27]. These joints were also evaluated for the severity of both tenderness and swelling on a 4-point scale. For tenderness, 0 = none, 1 = mild (positive response on questioning), 2 = moderate (spontaneous response elicited on examination), and 3 = severe (withdrawal by patient on examination); for swelling, 0 = none, 1 = mild (detectable synovial thickening without loss of bony contours), 2 = moderate (loss of distinctness of bony contours), and 3 = severe (bulging synovial proliferation with cystic characteristics). The sum of these determinations for all joints constituted a collective scores for either swelling or tenderness [23]. To foster agreement, evaluators attended a training session at the coordinating center before the trial. During the trial, all joint assessments for each patient were done by the same evaluator if possible. (During the trial, 94% of patients receiving minocycline and 92% of those receiving placebo had their joints evaluated by the same examiner at every visit.)
Grip strength was measured bilaterally using a mercury strain sphygmomanometer [28]. An evaluator assessed disease activity on the day of the examination using a scale of 1 to 5: absent, mild, moderate, severe, or very severe activity [29]. For functional class, patients were assessed according to the American Rheumatism Association-Steinbrocker classification [26] (at baseline and 48 weeks only).
Patient Assessments
Patients were asked about the duration of morning stiffness they experienced on the day before each study visit [23]. Function was assessed on eight activities of daily living with the Modified Health Assessment Questionnaire (MHAQ) [30]. Patients used a scale of 0 to 3: activity without any difficulty, with some difficulty, with much difficulty, and unable to do; thus, the higher the score, the greater the incapacitation. Patients were also asked to give an overall assessment of their disease activity for the preceding day on a scale of 1 to 5: absent, mild, moderate, severe, and very severe [29].
Ancillary Assessments
Laboratory determinations included urinalysis, complete blood count with differential, and a chemistry profile at the screening visit and every 4 weeks through week 12 and then every 6 weeks through week 48 of their treatment. Erythrocyte sedimentation rate (Westergren sedimentation rate) was measured at baseline and every 12 weeks. These tests were done by laboratories at each of the clinical centers.
Blood was drawn at baseline and every 12 weeks thereafter for centrally done analyses and serum banking. Serum samples were shipped frozen to the coordinating center. Immunoglobulin M rheumatoid factor was assayed by SmithKline Beecham Clinical Laboratories (Farmington Hills, Michigan) using nephelometry [31]. Seropositivity was defined as an IgM rheumatoid factor level of 30 U/dL or greater. For the few samples that were too lipemic to be tested reliably by nephelometry, IgM rheumatoid factor was measured by the latex fixation agglutination method [32]. The National Center for Infectious Diseases of the Centers for Disease Control and Prevention assayed serum samples for the presence of antibodies to Borrelia burgdorferi using a whole-cell sonicate and flagella-based enzyme-linked immunosorbent assay [33-35].
A posteroanterior radiograph of both hands and wrists was taken at baseline. Erosions were identified using the average score of two independent readers.
Adverse Drug Reactions
The patients were screened for adverse drug reactions at every visit. Patients were to stop receiving study medication for significant leukopenia, thrombocytopenia, anemia, hepatotoxicity, renal toxicity, or other serious adverse drug reactions. Data on adverse events were reviewed monthly at the coordinating center and every 6 months by the independent Data Safety Monitoring Review Committee.
Compliance
Patients were queried at each visit about all medications taken, including background medications. Trial medication capsules were counted at each visit to monitor compliance. Blood samples for minocycline assays were drawn at baseline and every 12 weeks. Pharmaco, Inc. (Richmond, Virginia) analyzed serum samples for minocycline using high-performance liquid chromatography with ultraviolet detection. This method provides a minimum quantifiable level for minocycline of 0.02 µg/mL [36].
Statistical Analysis
We used two prespecified primary outcome measures: improvement in joint swelling and improvement in joint tenderness. Definitions of the primary outcomes were taken from protocols for the CSSRD rheumatoid arthritis treatment trials [23, 24, 37]. Patients were considered to show improvement in joint swelling (or joint tenderness) if they had a 50% or greater improvement in the number of swollen (tender) joints at 48 weeks compared with baseline. Improvement in an individual joint was defined as a change in swelling (tenderness) from severe, moderate, or mild to no swelling (tenderness). Worsening was defined as involvement of a previously uninvolved joint or a change in swelling (tenderness) to severe in a joint previously rated as moderate or mild or none. For example, if a patient with 10 moderately tender joints capable of responding at baseline had 8 joints improve and 3 worsen during the trial (a net of 5 joints or 50% improvement), the patient would be classified as showing an improvement in tenderness. If 4 joints improved in the same patient and 2 worsened (20% improvement), this patient would not meet the definition of improvement.
The study was designed to detect an absolute difference between the minocycline and placebo groups of 30% using a critical level of 0.05 (two-sided test). In calculating the sample size, we assumed that the joints of 20% of the patients receiving placebo would improve. We assumed a dropout rate of 24% and considered this rate in calculating the sample size [38]. A sample size of 200 (100 per group) would provide greater than 90% power to detect such a difference. We did not plan an adjustment for multiple comparisons for the two primary outcome measures because these were prespecified.
One interim analysis of the primary outcome data was done using a Lan-DeMets statistical approach [39]. Because of this interim analysis, the final P value for the primary outcomes should be interpreted against a critical level of 0.0498 instead of the planned 0.05. The clinical center investigators and all patients remained blinded for the duration of the study and were unaware of the results of the interim analysis.
The primary outcomes of the treatment and placebo groups were compared using a Mantel-Haenszel [40] test with the clinical center included because it was a stratifying factor for randomization. (We used a critical value of 0.05 as a criterion for identifying differences between treatment groups at baseline that should be included in the model to test for treatment differences at follow-up. No baseline variables met the criterion.) In the analysis, we ignored the blocking. A statistical analysis that ignores blocking is less powerful, making detection of treatment differences more difficult [38]. We used an intention-to-treat analysis [38]; that is, all patients were analyzed in the group to which they were assigned. All patients were followed to the completion of the trial, even if they had discontinued therapy with the study medication and received other treatments. For patients who did not have a visit at 48 weeks, we imputed a final visit value using the worst of the previous visits (including the randomization visit) as a proxy for the 48-week visit.
We examined 10 other outcomes and used the Mantel-Haenszel approach for binary variables, with the clinical center as the stratifying variable. For continuous or ordinal outcomes, we used analysis of covariance on the ranks, adjusting for clinical center (five dummy variables) and testing for a treatment effect. To adjust for multiple comparisons, we interpreted P values against a critical level of 0.005 (Bonferroni correction for 10 tests) [41]. Our study was not designed to have high power to detect differences in these 10 other outcomes.
Baseline comparisons between the minocycline and placebo groups were made using t-tests for continuous variables, chi-square tests for nominal variables, and the Wilcoxon rank-sum [42] or the robust rank-order statistical test [43] for ordinal variables. We used an exact test for 2-by-K tables to compare reasons for withdrawal from study medication between the treatment groups.
Results
Top
Methods
Results
Discussion
Author & Article Info
References
Patients
Of the 219 patients entered in the study, 109 were randomly assigned to receive minocycline and 110 were assigned to receive placebo. The demographic and baseline clinical features for patients in the two treatment groups are shown in Tables 1 and 2. No patient had positive serologic results for B. burgdorferi.
View this table:
[in this window]
[in a new window]
Table 1. Demographic Characteristics at Baseline
View this table:
[in this window]
[in a new window]
Table 2. Patient Clinical Characteristics at Baseline*
Almost one half of the patients had received disease-modifying antirheumatic drug therapy before entering the study, including oral and parenteral gold salts (23%), hydroxychloroquine (22%), oral and parenteral methotrexate (12%), sulfasalazine (6%), D-penicillamine (2%), and azathioprine (1%); one third were receiving low doses of oral corticosteroids.
Drug Compliance and Withdrawal
At the 48-week visit, 79% of the minocycline group and 78% of the placebo group continued to receive the study medication. Of the 219 patients, 94% had a visit at 48 weeks (7 patients in the minocycline group and 7 patients in the placebo group missed this visit). Of the 47 patients who had previously discontinued therapy with the study drug, 36 had a visit at 48 weeks. For those receiving medication, the mean percentage of medication taken by capsule count was 84% ±26% for the minocycline group and 85% ±25% for the placebo group. Minocycline assays confirmed the capsule count data.
The primary reasons for discontinuation of therapy with the study medication were inefficacy in the placebo group and toxicity in the minocycline group (Table 3). We observed an association between treatment group and the primary reason for stopping medication (P = 0.002).
View this table:
[in this window]
[in a new window]
Table 3. Physician-reported Primary Reason for Discontinuing Therapy with Study Drug
Concurrent Antirheumatic Drugs
At baseline, both groups were similar in the concurrent use of antirheumatic drugs. Twenty-nine percent of the minocycline group and 32% of the placebo group were receiving oral corticosteroids. As shown in Table 4, there were more therapeutic interventions in the placebo group than in the minocycline group during the trial; some of these interventions were clearly protocol violations (for example, 17% of patients in the placebo group and 6% in the minocycline group received an intra-articular corticosteroid injection, and 14% of patients in the placebo group and 7% in the minocycline group had more than one NSAID substitution). Other interventions occurred after therapy with the study drug had been discontinued.
View this table:
[in this window]
[in a new window]
Table 4. Medication Changes during the Trial
Primary Outcomes
Table 5 shows the intention-to-treat data for the primary outcomes. For joint swelling, 54% of patients receiving minocycline showed improvement compared with 39% of patients receiving placebo (P = 0.023). We observed similar differences (56% and 41% of patients, respectively) for joint tenderness (P = 0.021).
View this table:
[in this window]
[in a new window]
Table 5. Intention-To-Treat Analysis for Changes in Primary Outcomes at 48 Weeks
Other Outcomes
At the week 48 evaluation, the median joint counts for tenderness and swelling decreased from the baseline counts for patients in both the minocycline and placebo groups. The median decreases for tenderness were 14 for the minocycline group and 8.5 for the placebo group; for swelling, the decreases were 9 and 6, respectively (P < 0.001). No patients met American Rheumatism Association criteria for remission [44]. Figure 1 shows the temporal change in the proportion of patients with meaningful improvement by intention-to-treat analysis. For both groups, improvement began by week 12. Improvement in joint swelling and in joint tenderness continued through week 48 for patients receiving minocycline but reached a plateau at week 24 for patients receiving placebo.
View larger version (52K):
[in this window]
[in a new window]
Figure 1. Top. Proportion of patients with 50% or greater improvement in joint swelling, by visit. Bottom. Proportion of patients with 50% or greater improvement in joint tenderness, by visit.
Table 6 shows the data for some other outcome variables at randomization and at 48 weeks. The Table also heart rate alterans shows differences between the two assessments and the corresponding summary statistics for these differences. With two exceptions, both groups on average showed some improvement from baseline to 48 weeks. The exceptions were in the placebo group, in which hematocrit decreased and IgM rheumatoid factor levels increased. We detected treatment group differences for erythrocyte sedimentation rate, hematocrit, platelet count, and IgM rheumatoid factor, with the minocycline group showing greater improvement for each measurement (all P values < 0.001).
View this table:
[in this window]
[in a new window]
Table 6. Intention-to-Treat Analysis for Changes in Other Outcomes at 48 Weeks*
Fifty-four percent of patients in the minocycline group and 39% in the placebo group had erythrocyte sedimentation rates within the normal range at 48 weeks. Eighty-seven percent and 77% of patients, respectively, had hematocrits within the normal range, and 98% and 96% had platelet counts within the normal range. Of the 60 patients in the minocycline group who tested positive for rheumatoid factor at baseline, 13 tested negative at 48 weeks compared with only 1 of 63 patients in the placebo group. However, 3 of the 48 patients receiving minocycline who had negative test results became positive, and 2 of 47 patients in the placebo group became positive.
In comparisons of the treatment groups for the remaining outcomes—grip strength, patient and physician global assessment, MHAQ, and hours of morning stiffness—P values ranged from 0.04 to 0.76, all greater than the critical value of 0.005 (Bonferroni adjustment for multiple comparisons).
Adverse Drug Reactions
One death occurred—a suicide by a patient receiving placebo who discontinued therapy with study medication early in the trial. Table 7 shows the reported adverse effects that lasted 1 day or more. Dizziness was expected to be an adverse effect of minocycline [19, 20, 45], yet we observed only slight differences between the two treatment groups. Thirty-nine percent of the minocycline group and 34% of the placebo group reported dizziness lasting more than 1 day at least once during the trial. Six percent of the minocycline group but no patients in the placebo group stopped medication because of possible toxicity. Elevated liver function test results were reported for two patients in the minocycline group and three patients in the placebo group. Two patients receiving minocycline had total bilirubin levels greater than 25.65 µmol/L (1.5 mg/dL) during the trial. The values in one of these patients subsequently returned to normal. The other patient, a suspected alcohol abuser, had persistently elevated liver enzymes, stopped receiving medication, and moved away from the clinical center areas. One patient receiving minocycline and one patient receiving placebo had elevated creatinine levels during the study (range, 132.6 to 203.3 µmol/L [1.5 to 2.3 mg/dL]). Two patients receiving minocycline developed raised blue spots on their tongues, and another reported "graying" teeth. Fewer patients in the minocycline group than in the placebo group had urinary tract infections (5% compared with 12%; P = 0.05) and upper respiratory tract infections (11% compared with 16%; P = 0.25). More patients in the minocycline group had vaginitis than did those in the placebo group (9% compared with 3%; P = 0.04).
View this table:
[in this window]
[in a new window]
Table 7. Patient-reported Symptoms Lasting Longer Than 1 Day
Discussion
Top
Methods
Results
Discussion
Author & Article Info
References
The MIRA trial showed that minocycline is both effective and safe for treating patients with mild to moderately active rheumatoid arthritis. Benefit became evident after 12 weeks of therapy, and the proportion of patients treated with minocycline showing improvement continued to increase through week 48 of the study. Along with clinical improvement in joint swelling and tenderness, objective laboratory features of active inflammation such as hematocrit, erythrocyte sedimentation rate, platelet count, and IgM rheumatoid factor level showed favorable changes. No serious toxicity occurred. This experience complements reports of other studies of minocycline for treating rheumatoid arthritis [19, 20, 22]. These data contrast with the negative results of a small trial of rheumatoid arthritis in which 250 mg of tetracycline daily was used [5]. However, minocycline is more potent and more readily absorbed and has a better tissue distribution than tetracycline hydrochloride [45].
The proportion of patients in the placebo group who improved was twice that anticipated in the study design. Several factors may have contributed to this increased response and may have minimized the differences in outcomes between groups at 48 weeks in an intention-to-treat analysis. Clinic coordinators maintained frequent contact with patients and reinforced study compliance. We provided NSAIDs to all patients as an incentive to maintain optimal and stable background therapy for the duration of the trial. In addition, patients receiving placebo had more protocol-violating interventions during the trial, such as injections of intra-articular corticosteroids and increases in prednisone or NSAID therapy. Finally, more patients receiving placebo than those treated with minocycline who discontinued therapy with the study medication began receiving a disease-modifying antirheumatic drug before their week 48 study visit. However, all available study patients were evaluated at 48 weeks, and their data were analyzed according to their original treatment group assignments.
In our study, adverse drug reactions clearly attributable to minocycline were both infrequent and mild. Patient-reported problems previously believed to be specifically related to the administration of tetracycline and its derivatives, such as vestibular dysfunction [19], occurred with similar frequency in the two treatment groups. Only 3 of the 109 patients receiving minocycline reported dizziness as a reason for stopping medication. This profile contrasts with the experience of others [20, 22], who reported treatment-limiting gastrointestinal and vestibular symptoms in their patients with rheumatoid arthritis who received minocycline. Consistent with the clear dose response for vestibular symptoms shown in patients receiving preventive minocycline therapy for meningitis [46], a possible reason for the low frequency of adverse effects in our minocycline trial is that investigators were allowed to adjust the dosage of the study medication.
Patients in the minocycline trial can be compared with patients in similar clinical trials of therapies for rheumatoid arthritis [22-24, 47-52]. Disease duration was shorter than in the study of minocycline by Kloppenburg and colleagues [22] but longer than in several studies of other antirheumatic agents [24, 48-51]. Our patients had a lower erythrocyte sedimentation rate at baseline [22, 47, 52], and fewer patients were positive for IgM rheumatoid factor when compared with patients in other studies [48, 49, 51]. Sixty-eight percent of our patients had erosive disease compared with 80% in a CSSRD trial [52].
We observed significant differences (as defined in the Methods section) favoring minocycline over placebo in the primary outcome measures, in improvement in joint swelling and tenderness, and in the laboratory measurements. For some outcomes, such as the patient and physician global assessments, we observed differences in favor of minocycline that did not meet our criteria for statistical significance. These results were similar to the observations of Kloppenburg and colleagues [22] and the investigators of the CSSRD sulfasalazine study [50].
We cannot completely explain the failure to find larger differences in the global assessments and MHAQ. We had power of 0.83 and 0.77 to detect effects of the size observed in Weinblatt and colleagues' trial [47] comparing methotrexate with auranofin for the physician and patient global assessments, respectively, using a critical level of 0.005. In Weinblatt and colleagues' trial [47], the methotrexate group showed larger mean differences in the physician and patient global assessments than we observed in our minocycline group. However, the methotrexate differences would not have been considered significant had there been an adjustment for multiple comparisons. In our trial, pain appeared to be well controlled by NSAIDs in both groups, and multitrait analyses [53] suggest that the two global outcomes represent another assessment of the patient's level of pain. By controlling pain, it is also possible that level of activity as measured by the MHAQ would also improve. Thus, the background effects of NSAIDs may have made it difficult to find differences in our 48-week trial. Given the effect of minocycline on disease activity as shown by the differences in laboratory measurements and in the primary outcomes, it is possible that over a longer period of observation, larger differences in physician or patient global assessments could have emerged.
Whether the antirheumatic activity of minocycline is mediated by its antimicrobial, anti-inflammatory, or immunomodulatory properties remains to be determined. Although the efficacy of minocycline became evident at 12 weeks, improvement in patients treated with minocycline continued through the end of the trial; this time lag by itself does not rule out the possibility that the antirheumatic properties of this compound are caused by its antimicrobial activity. In addition, patients receiving minocycline experienced beneficial changes in laboratory measurements such as hematocrit, erythrocyte sedimentation rate, platelet count, and IgM rheumatoid factor level. Although the magnitude of the median differences in Table 6 is small for some laboratory measurements, the change in median values in the minocycline group resulted in more patients in the minocycline group having values within the normal ranges at 48 weeks than patients in the placebo group. The consistency of these laboratory results suggests that other well-documented nonantibiotic properties of these compounds, as shown both in vitro and in experimental animals [6-18], could explain the antirheumatic effect of minocycline.
In conclusion, we showed that minocycline is a safe, effective treatment for patients with mild to moderate rheumatoid arthritis. The observed adverse effects were mild to infrequent. On average, all outcome measures in the minocycline group improved from baseline to 48 weeks. For patients in the placebo group, hematocrit and IgM rheumatoid factor level on average worsened. The proportion of patients in the minocycline group whose joint counts improved by 50% or more continued to increase through week 48 of the trial, whereas the corresponding proportion of patients receiving placebo appeared to reach a plateau by week 24. The long-term effects of minocycline, its relative potency and safety in relation to disease-modifying antirheumatic drugs, and its mechanisms of action remain to be determined.
Appendix
The following are members of the MIRA Trial Group: The Beth Israel Hospital, Boston, Massachusetts: D. Trentham, R. Dynesius-Trentham, K. Sewell, F. Kantrowitz, G. Craven, M. Finell; State University of New York Health Sciences Center, Brooklyn, New York: D. Kaplan, R. Neuner, K. Orloff, C. Mignone, I. Caicedo, I. Kansariwala, S. Kolaninski, E. Ginzler, I. Antoniadis, R. Joks, S. Bakels; The University of Alabama at Birmingham, Birmingham, Alabama: G. Alarcon, W. Blackburn, Jr., S. Baum, I. Mikhail, R. Ayers, N. Abraham, M. Sehn; The University of Utah Medical Center, Salt Lake City, Utah: D. Clegg, T. McAllister, C. Jackson, M. Lundberg, P. Knibbe, J. Williams, G. Cannon, D. Stromquist, J. Ward, J. Huan; Henry Ford Hospital, Detroit, Michigan: J. Leisen, C. Sherlitz, M. Lubetsky, B. Segal, J. Sigler, L. Lasichak; The University of Vermont College of Medicine, Burlington, Vermont: S. Cooper, M. Merchant, L. Buckley, E. Leib, K. Cartularo, R. Budd, M. Davitt, K. Clark; Case Western Reserve University, The Henry Ford Health Sciences Center, Detroit, Michigan: B. Tilley, S. Fowler, H. Duncan, G. Bluhm, S. Fagan, S. Kardasz, M. Rusinova, N. Maddy, H. Boldarini, D. Whitted; National Institute of Arthritis and Musculoskeletal and Skin Diseases, Bethesda, Maryland: L. Shulman, S. Heyse, S. Pillemer, M. Tuttleman, S. Shores, E. Webster-Cissel; Tifton Medical Center, Tifton, Georgia: J. Sharp. Data and Safety Monitoring Review Committee Members included the following: The Medical Center at Princeton, New Jersey: R. Pinals; The New York Hospital-Cornell Medical Center New York, New York: M. Reidenberg; Fred Hutchinson Cancer Research Center, Seattle, Washington: S. Green; Howard University, Washington, D.C.: W. Greaves; National Eye Institute, Bethesda, Maryland: D. Everett; Georgetown University Medical Center: Washington, D.C.: C. Taylor.
Author and Article Information
Top
Methods
Results
Discussion
Author & Article Info
References
From the Henry Ford Health Sciences Center, Detroit, Michigan; the University of Alabama at Birmingham; National Institutes of Health, Bethesda, Maryland; Beth Israel Hospital, Boston, Massachusetts; the State University of New York at Brooklyn; the University of Utah Medical Center, Salt Lake City, Utah; and the University of Vermont, Burlington, Vermont.
For the MIRA Trial Group*
Requests for Reprints: The MIRA Trial Group, c/o Stephen P. Heyse, MD, Office of Disease Prevention, Epidemiology, and Clinical Applications, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Natcher Building, Room 5AS-53, 45 Center Drive, MSC 6500, Bethesda, MD 20892-6500.
Acknowledgments: The authors thank the physicians who referred their patients to the clinical centers, the clinic coordinators, the patients who participated in the study, and Helene Boldarini and Ella Henderson for secretarial assistance.
Grant Support: By research contracts from the National Institutes of Health, National Institute of Arthritis and Musculoskeletal and Skin Diseases (contracts N01-AR-1-2202, N01-AR-1-2203, N01-AR-1-2205, N01-AR-1-2207, N01-AR-1-2206, N01-AR-2-2210, and N01-AR-2-2205). Study medication and placebo were provided by Lederle Laboratories, a Division of Cyanamid, Inc., Pearl River, New York.
References
Top
Methods
Results
Discussion
Author & Article Info
References
1. Lawrence RC, Hochberg MC, Kelsey JL, McDuffie FC, Medsger TA Jr, Felts WR, et al. Estimates of prevalence of selected arthritis and musculoskeletal diseases in the United States. J Rheumatol. 1989; 16:427-41.
2. Pinals R. Management of chronic inflammatory joint disease. In: Kelley WN, ed. Textbook of Internal Medicine. Philadelphia: J.B. Lippincott; 1989:1094-100.
3. Brown TM, Bush SW, Felts WR. In: Wohl MG, ed. Long-Term Illness: Management of the Chronically Ill Patient. Philadelphia: W.B. Saunders; 1959:93-125.
4. Sanchez I. Tetracycline treatment in rheumatoid arthritis and other rheumatic diseases (As tetraciclinas no tratamento da artrite reumatoide e doenas reumaticas inflamatorias). Brasil Medico. 1968; 82:22-31.
5. Skinner M, Cathcart ES, Mills JA, Pinals RS. Tetracycline in the treatment of rheumatoid arthritis. A double-blind controlled study. Arthritis Rheum. 1971; 14:727-32.
6. Golub LM, Lee HM, Lehrer G, Nemiroff A, McNamara TF, Kaplan R, et al. Minocycline reduces gingival collagenolytic activity during diabetes. Preliminary observations and a proposed new mechanism of action. J Periodontal Res. 1983; 18:516-24.
7. Golub LM, Ramamurthy NS, McNamara TF, Cianco S, Schneir M, Perry H. Tetracyclines inhibit tissue collagenase activity. J Periodont Res. 1984; 19:651-5.
8. Greenwald RA, Golub LM, Lavietes B, Ramamurthy NS, Gruber B, Laskin RS, et al. Tetracyclines inhibit human synovial collagenase in vivo and in vitro. J Rheumatol. 1987; 14:28-32.
9. Riesbeck K, Bredberg A, Forsgren A. Ciprofloxacin does not inhibit mitochondrial functions but other antibiotics do. Antimicrob Agents Chemother. 1990; 34:167-9.
10. Martin RR, Warr GA, Couch RB, Yeager H, Knight V. Effects of tetracycline on leukotaxis. J Infect Dis. 1974; 129:110-6.
11. Forsgren A, Schmeling D, Quie PG. Effect of tetracycline on the phagocytic function of human leukocytes. J Infect Dis. 1974; 130:412-5.
12. Thong YH, Ferrante A. Inhibition of mitogen-induced human lymphocyte proliferative responses by tetracycline analogues. Clin Exp Immunol. 1979; 35:443-6.
13. Ingham E, Turnbull L, Kearney JN. The effects of minocycline and tetracycline on the mitotic response of human peripheral blood lymphocytes. J Antimicrob Chemother. 1991; 27:607-17.
14. MacDonald TT, Carter PB. Requirement for a bacterial flora before mice generate cells capable of mediating the delayed hypersensitivity reaction to sheep red blood cells. J Immunol. 1979; 122:2624-30.
15. Plewig G, Schopf E. Anti-inflammatory effects of antimicrobial agents: an in vivo study. J Invest Dermatol. 1975; 65:532-6.
16. van Barr HM, van de Kerkhof PC, Mier PD, Happle R. Tetracyclines are potent scavengers of the superoxide radical (Letter). Br J Dermatol. 1987; 117:131-2.
17. Wasil M, Halliwell B, Moorhouse CP. Scavenging of hypochlorous acid by tetracycline, rifampicin and some other antibiotics: a possible antioxidant action of rifampicin and tetracycline? Biochem Pharmacol. 1988; 37:775-8.
18. Sewell KL, Furrie E, Trentham DE. The therapeutic effect of minocycline in experimental arthritis: mechanism of action (Abstract). Arthritis Rheum. 1990; 33:S106.
19. Breedveld FC, Dijkmans BA, Mattie H. Minocycline treatment for rheumatoid arthritis: an open dose finding study. J Rheumatol. 1990; 17:43-6.
20. Langevitz P, Bank I, Zemer D, Book M, Pras M. Treatment of resistant rheumatoid arthritis with minocycline: an open study. J Rheumatol. 1992; 19:1502-4.
21. Kloppenburg M, Breedveld FC, Miltenburg AM, Dijkmans BA. Antibiotics as disease modifiers in arthritis. Clin Exp Rheum. 1993; 8:S113-5.
22. Kloppenburg M, Breedveld FC, Terwiel J, Mallee C, Dukmans BA. Minocycline in active rheumatoid arthritis. A double-blind, placebo-controlled trial. Arthritis Rheum. 1994; 37:629-36.
23. Williams HJ, Ward JR, Reading JC, Brooks RH, Clegg DO, Skosey JL, et al. Comparison of auranofin, methotrexate, and the combination of both in the treatment of rheumatoid arthritis. A controlled clinical trial. Arthritis Rheum. 1992; 35:259-69.
24. Williams HJ, Wilkens RF, Samuelson CO Jr, Alarcon GS, Guttadauria M, Yarboro C, et al. Comparison of low-dose oral pulse methotrexate and placebo in the treatment of rheumatoid arthritis. Arthritis Rheum. 1985; 28:721-30.
25. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 1988; 31:315-24.
26. Steinbrocker O, Traeger CH, Batterman RC. Therapeutic criteria in rheumatoid arthritis. JAMA. 1949; 140:659-62.
27. Cooperating Clinics Committee of the American Rheumatism Association. A seven day variability study of 499 patients with peripheral rheumatoid arthritis. Arthritis Rheum. 1965; 8:302-34.
28. Lee P, Baxter A, Dick WC, Webb J. Assessment of grip strength measurement in rheumatoid arthritis. Scand J Rheumatol. 1973; 3:17-23.
29. Felson DT, Anderson JJ, Boers M, Bombardier C, Chernoff M, Fried B, et al. The American College of Rheumatology preliminary core set of disease activity measures for rheumatoid arthritis clinical trials. The Committee on Outcome Measures in Rheumatoid Arthritis Clinical Trials. Arthritis Rheum. 1993; 36:729-40.
30. Pincus T, Summey JA, Soraci SA Jr, Wallston KA, Hummon NP. Assessment of patient satisfaction activities of daily living using a modified Stanford Health Assessment Questionnaire. Arthritis Rheum. 1983; 26:1346-53.
31. Hicks MJ, Heick M, Finley P, Gall EP, Minnich L, Williams RJ. Rheumatoid factor activity by rate nephelometry correlated with clinical activity in rheumatoid arthritis. Am J Clin Pathol. 1982; 87:342-5.
32. Singer JM, Plotz CM. The latex fixation test: I. Application to the serologic diagnosis of rheumatoid arthritis. Am J Med. 1956; 21:888-92.
33. Hansen K, Hindersson P, Pedersen NS. Measurement of antibodies to the Borrelia burgdorferi flagellum improves serodiagnosis in Lyme disease. J Clin Microbiol. 1988; 26:338-46.
34. Quan TJ, Craven RB, Mayer LW, Gubier DJ, Bailey RE, Carter LG, et al. Sensitivity and specificity of whole-cell sonicate and flagella based ELISA for Borrelia burgdorferi antibody (Abstract). Presented at V International Conference on Lyme Borreliosis. Arlington, Virginia, 30 May-2 June 1992.
35. Quan TJ, Robbins KE, Craven RB, Mayer LW, Johnson BJB, Sviat SL, et al. Precision of whole-cell sonicate and flagella ELISA for Borrelia burgdorferi antibody (Abstract). Presented at V International Conference on Lyme Borreliosis. Arlington, Virginia, 30 May 30-2 June 1992.
36. Sossong JL, Navet DG, Wilkinson JM, II. Method LC63.1, HPLC analysis of minocycline in human serum. Richmond, Virginia: Pharmaco-LSR; October 1992.
37. Ward JR, Williams HJ, Egger MJ, Reading JC, Boyce E, Altz-Smith M, et al. Comparison of auranofin, gold sodium thiomalate, and placebo in the treatment of rheumatoid arthritis. A controlled clinical trial. Arthritis Rheum. 1983; 26:1303-15.
38. Friedman LM, Furberg CD, DeMets DL. Fundamentals of Clinical Trials. 2d ed. Littleton, Massachusetts: PSG Publishing Co.; 1985.
39. Lan KK, DeMets DL. Discrete sequential boundaries for clinical trials. Biometrika. 1983; 70:659-63.
40. Mantel N. Chi-square tests with one degree of freedom. Extensions of the Mantel-Haenszel procedure. Journal of the American Statistical Association. 1963; 58:690-700.
41. Miller RG. Simultaneous Statistical Inference. New York: Springer-Verlag; 1981:67-70.
42. Lehman EL. Nonparametrics: Statistical Methods Based on Ranks. San Francisco: Holden Day; 1975:123-41.
43. Siegel S, Castellan NJ Jr. Nonparametric Statistics for the Behavioral Sciences. 2d ed. New York: McGraw-Hill; 1988:137-44.
44. Pinals RS, Masi AT, Larson RA. Preliminary criteria for clinical remission in rheumatoid arthritis. Arthritis Rheum. 1981; 24:1308-15.
45. Cunha BA, Comer JB, Jonas M. The tetracyclines. Med Clin North Am. 1982; 66:293-302.
46. Altman R, Black K, Goldfield M. Vestibular reactions to minocycline after meningococcal prophylaxis. MMWR Morb Mortal Wkly Rep. 1975; 24:9-11.
47. Weinblatt ME, Kaplan H, Germain BF, Merriman RC, Solomon SD, Wall B, et al. Low-dose methotrexate compared with auranofin in adult rheumatoid arthritis. A thirty-six week, double-blind trial. Arthritis Rheum. 1990; 33:330-8.
48. Hannonen P, Mottonen T, Hakola M, Oka M. Sulfasalazine in early rheumatoid arthritis. A 48-week double-blind, prospective, placebo-controlled study. Arthritis Rheum. 1993; 36:1501-9.
49. Davis MJ, Dawes PT, Fowler PD, Clarke S, Fisher J, Shadforth MF. Should disease-modifying agents be used in mild rheumatoid arthritis? Br J Rheumatol. 1991; 30:451-4.
50. Williams HJ, Ward JR, Dahl SL, Clegg DO, Willkens RF, Oglesby T, et al. A controlled trial comparing sulfasalazine, gold sodium thiomalate, and placebo in rheumatoid arthritis. Arthritis Rheum. 1988; 31:702-13.
51. Clark P, Casas E, Tugwell P, Medina C, Gheno C, Tenorio G, et al. Hydroxychloroquine compared with placebo in rheumatoid arthritis. A randomized controlled trial. Ann Intern Med. 1993; 119:1067-71.
52. Lopez-Mendez A, Daniel WW, Reading JC, Ward JR, Alarcon GS. Radiographic assessment of disease progression in rheumatoid arthritis patients enrolled in the cooperative systematic studies of the rheumatic diseases program randomized clinical trial of methotrexate, auranofin, or a combination of the two. Arthritis Rheum. 1993; 38:1364-9.
53. Hayes RD, Hayashi T, Carson S, Ware JE. MAP: A Multitrait Scaling Analysis Program. Dubuque, Iowa: William C. Brown; 1989.
Related articles in Annals:
Editorials
Minocycline Treatment of Rheumatoid Arthritis
Harold E. Paulus
Annals 1995 122: 147-148. [Full Text]
This article has been cited by other articles:
D. J. J. Park, J. J. Woog, J. S. Pulido, and J. D. Cameron
Minocycline for the Treatment of Ocular and Ocular Adnexal Sarcoidosis
Arch Ophthalmol, May 1, 2007; 125(5): 705 - 709.
[Full Text] [PDF]
S. K. Jo, M. H. Rosner, and M. D. Okusa
Pharmacologic Treatment of Acute Kidney Injury: Why Drugs Haven't Worked and What Is on the Horizon
Clin. J. Am. Soc. Nephrol., March 1, 2007; 2(2): 356 - 365.
[Abstract] [Full Text] [PDF]
A. Gaffo, K. G. Saag, and J. R. Curtis
Treatment of rheumatoid arthritis
Am. J. Health Syst. Pharm., December 15, 2006; 63(24): 2451 - 2465.
[Abstract] [Full Text] [PDF]
C.J. Tredwin, C. Scully, and J.-V. Bagan-Sebastian
Drug-induced Disorders of Teeth
J. Dent. Res., July 1, 2005; 84(7): 596 - 602.
[Abstract] [Full Text] [PDF]
A. S. Ramirez, A. Rosas, J. A. Hernandez-Beriain, J. C. Orengo, P. Saavedra, C. de la Fe, A. Fernandez, and J. B. Poveda
Relationship between rheumatoid arthritis and Mycoplasma pneumoniae: a case control study
Rheumatology, July 1, 2005; 44(7): 912 - 914.
[Abstract] [Full Text] [PDF]
S. A Voils, M. E Evans, M. T Lane, R. H Schosser, and R. P Rapp
Use of Macrolides and Tetracyclines for Chronic Inflammatory Diseases
Ann. Pharmacother., January 1, 2005; 39(1): 86 - 94.
[Abstract] [Full Text] [PDF]
S M Carty, N Snowden, and A J Silman
Should infection still be considered as the most likely triggering factor for rheumatoid arthritis?
Ann Rheum Dis, November 1, 2004; 63(suppl_2): ii46 - ii49.
[Full Text] [PDF]
E Suresh, I M Morris, and P C Mattingly
Use of minocycline in rheumatoid arthritis: a district general hospital experience
Ann Rheum Dis, October 1, 2004; 63(10): 1354 - 1355.
[Full Text] [PDF]
J. R. O'Dell
Therapeutic Strategies for Rheumatoid Arthritis
N. Engl. J. Med., June 17, 2004; 350(25): 2591 - 2602.
[Full Text] [PDF]
L. B. Krupp, L. G. Hyman, R. Grimson, P. K. Coyle, P. Melville, S. Ahnn, R. Dattwyler, and B. Chandler
Study and treatment of post Lyme disease (STOP-LD): A randomized double masked clinical trial
Neurology, June 24, 2003; 60(12): 1923 - 1930.
[Abstract] [Full Text] [PDF]
D. L. Scott and V. Strand
The effects of disease-modifying anti-rheumatic drugs on the Health Assessment Questionnaire score. Lessons from the leflunomide clinical trials database
Rheumatology, August 1, 2002; 41(8): 899 - 909.
[Abstract] [Full Text] [PDF]
M. S. Klempner, L. T. Hu, J. Evans, C. H. Schmid, G. M. Johnson, R. P. Trevino, D. Norton, L. Levy, D. Wall, J. McCall, et al.
Two Controlled Trials of Antibiotic Treatment in Patients with Persistent Symptoms and a History of Lyme Disease
N. Engl. J. Med., July 12, 2001; 345(2): 85 - 92.
[Abstract] [Full Text] [PDF]
I. I. Kuzin, J. E. Snyder, G. D. Ugine, D. Wu, S. Lee, T. Bushnell Jr, R. A. Insel, F. M. Young, and A. Bottaro
Tetracyclines inhibit activated B cell function
Int. Immunol., July 1, 2001; 13(7): 921 - 931.
[Abstract] [Full Text] [PDF]
D Taylor-Robinson and A Keat
How can a causal role for small bacteria in chronic inflammatory arthritides be established or refuted?
Ann Rheum Dis, March 1, 2001; 60(3): 177 - 185.
[Full Text] [PDF]
T. M. Lawson, N. Amos, D. Bulgen, and B. D. Williams
Minocycline-induced lupus: clinical features and response to rechallenge
Rheumatology, March 1, 2001; 40(3): 329 - 335.
[Abstract] [Full Text] [PDF]
H. Bachelez, P. Senet, J. Cadranel, A. Kaoukhov, and L. Dubertret
The Use of Tetracyclines for the Treatment of Sarcoidosis
Arch Dermatol, January 1, 2001; 137(1): 69 - 73.
[Abstract] [Full Text] [PDF]
M.-T. Labro
Interference of Antibacterial Agents with Phagocyte Functions: Immunomodulation or ""Immuno-Fairy Tales""?
Clin. Microbiol. Rev., October 1, 2000; 13(4): 615 - 650.
[Abstract] [Full Text] [PDF]
D. M. Ozog, D. S. Gogstetter, G. Scott, and A. A. Gaspari
Minocycline-Induced Hyperpigmentation in Patients With Pemphigus and Pemphigoid
Arch Dermatol, September 1, 2000; 136(9): 1133 - 1138.
[Abstract] [Full Text] [PDF]
J. L. Anderson, J. B. Muhlestein, J. Carlquist, A. Allen, S. Trehan, C. Nielson, S. Hall, J. Brady, M. Egger, B. Horne, et al.
Randomized Secondary Prevention Trial of Azithromycin in Patients With Coronary Artery Disease and Serological Evidence for Chlamydia pneumoniae Infection : The Azithromycin in Coronary Artery Disease: Elimination of Myocardial Infection with Chlamydia (ACADEMIC) Study
Circulation, March 30, 1999; 99(12): 1540 - 1547.
[Abstract] [Full Text] [PDF]
S. M. Campbell and R. Wernick
Update in Rheumatology
Ann Intern Med, January 19, 1999; 130(2): 135 - 142.
[Full Text] [PDF]
A. R. Amin, M. G. Attur, G. D. Thakker, P. D. Patel, P. R. Vyas, R. N. Patel, I. R. Patel, and S. B. Abramson
A novel mechanism of action of tetracyclines: Effects on nitric oxide synthases
PNAS, November 26, 1996; 93(24): 14014 - 14019.
[Abstract] [Full Text] [PDF]
C. M. Dollery, J. R. McEwan, and A. M. Henney
Matrix Metalloproteinases and Cardiovascular Disease
Circ. Res., November 1, 1995; 77(5): 863 - 868.
[Full Text]
Minocycline May Ease Rheumatoid Arthritis
Journal Watch Dermatology, March 1, 1995; 1995(301): 13 - 13.
[Full Text]
MINOCYCLINE MAY EASE RHEUMATOID ARTHRITIS
Journal Watch (General), January 31, 1995; 1995(131): 2 - 2.
[Full Text]
Article
Table of Contents
Abstract of this article Free
Figures/Tables List
Related articles in Annals
Articles citing this article
Services
Send comment/rapid response letter
Notify a friend about this article
Alert me when this article is cited
Add to Personal Archive
Download to Citation Manager
ACP Search
Get Permissions
Google Scholar
Search for Related Content
PubMed
Articles in PubMed by Author:
Tilley, B. C.
Fowler, S. E.
Related Articles in PubMed
PubMed Citation
PubMed
|
Topic Options
