SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Objective

Patients with rheumatoid arthritis (RA) have been shown to have an increased susceptibility to the development of infections. The exact causes of this increased risk are unknown, but may relate to immunologic disturbances associated with the disease or to the immunosuppressive effects of agents used in its treatment. This study was undertaken to identify predictors of serious infections among patients with RA. Identification of such factors is the necessary first step in reducing the excess risk of infection in RA.

Methods

Members of a population-based incidence cohort of Rochester, Minnesota residents ages ≥18 years, who had been diagnosed with RA between 1955 and 1994, were followed up longitudinally through their complete medical records until January 1, 2000. We examined potential risk factors for the development of all objectively confirmed (by microbiology or radiology) infections and for infections requiring hospitalization. Potential risk factors included RA severity measures (rheumatoid factor positivity, elevated erythrocyte sedimentation rate, extraarticular manifestations of RA, and functional status), comorbidities (diabetes mellitus, alcoholism, and chronic lung disease), and other risk factors for infection (presence of leukopenia, smoking). Predictors were identified using multivariate time-dependent Cox proportional hazards modeling.

Results

The 609 RA patients in the cohort had a total followup time of 7,729.7 person-years (mean 12.7 years per patient). A total of 389 patients (64%) had at least 1 infection with objective confirmation, and 290 (48%) had at least 1 infection requiring hospitalization. Increasing age, presence of extraarticular manifestations of RA, leukopenia, and comorbidities (chronic lung disease, alcoholism, organic brain disease, and diabetes mellitus), as well as use of corticosteroids, were strong predictors of infection (P < 0.004) in both univariate and multivariate analyses. Notably, use of disease-modifying antirheumatic drugs was not associated with increased risk of infection in multivariate analyses, after adjustment for demographic characteristics, comorbidities, and disease-related variables.

Conclusion

We identified a number of strong predictors of infections in a population-based cohort of patients with RA. These results can be used to prospectively identify high-risk patients, who may benefit from closer followup and implementation of preventive strategies.

Concern about an increased risk of infection among patients with rheumatoid arthritis (RA) has become heightened with recent reports describing severe and opportunistic infections in patients treated with new biologic agents (1–4). Although an increase in the risk of septic arthritis in patients with RA has been reported (5), little is known about the rates of various other infections among these patients. In a recent study reported elsewhere in this issue of Arthritis & Rheumatism (6), we found a higher rate of virtually all types of infection in an incidence cohort of patients with RA compared with an age- and sex-matched group of subjects without RA. The objective of this companion study was to identify factors that are predictive of the observed increased risk of infection in RA patients. The identification of determinants of the risk of infection in RA is the first step toward reducing the excess disease burden associated with infection in these patients.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Ascertainment of RA patients.

An incidence cohort of patients with RA was identified using the data resources of the Rochester Epidemiology Project. This diagnostic indexing and medical records linkage system at the Mayo Clinic affords access to medical records from all sources of care for community residents. Population-based epidemiologic research in Rochester, Minnesota is possible because of its relative geographic isolation from other urban centers and the fact that nearly all medical care is delivered to local residents by a small number of providers. These providers include the Mayo Clinic, Olmsted Medical Group, and a few private practitioners. Each provider uses a comprehensive medical record system whereby all data collected on an individual are assembled in a single record. Medical records are available for all residents dating back to 1910. Medical diagnoses and other key information are routinely abstracted in a summary record (“master sheet”) and entered into computerized indices. The medical records linkage system composed of these indices facilitates identification of all cases of a given condition. Thus, this system ensures virtually complete ascertainment of all clinically diagnosed cases in a geographically defined community. The potential utility of this data retrieval system for population-based studies has been described in detail previously (7, 8).

To identify individuals with RA for this study, all potential cases of RA were identified by searching the computerized diagnostic index of the Rochester Epidemiology Project for any diagnosis of arthritis (excluding degenerative arthritis or osteoarthritis) made between January 1, 1955, and December 31, 1994, among Rochester residents ≥18 years of age. The complete medical records were reviewed by a team that included 3 trained nurse abstractors and 1 physician (MFD), using a pretested data collection form. The diagnosis was confirmed or rejected based on the American College of Rheumatology (formerly, the American Rheumatism Association) 1987 diagnostic criteria for RA (9).

Patients with RA were followed up through their entire (inpatient and outpatient) community medical record, until death, migration from the county, or the date of study end (January 1, 2000).

Data collection.

Study personnel collected the data according to a prespecified and pretested protocol. Reliability testing was carried out at the outset of the abstraction process: a sample of medical records was reviewed by all abstractors to ensure good interobserver and intraobserver agreement. Regular meetings were held throughout the abstraction period to identify and correct problems in data collection, interpretation of definitions, and application of study criteria. Before commencing data analysis, we performed an extensive series of checks for data consistency, proper sequences of dates, and an evaluation of missing or incomplete data. If necessary, medical records were reviewed again, and questions were resolved by consensus of the investigative team.

Data on all episodes of infection requiring medical care occurring after the incidence date were collected. Information on minor upper respiratory tract infections was excluded. The operational definitions for each infection type were as follows: bacteremia/septicemia, isolation of a pathogenic microorganism from 1 or more blood cultures, with fever (>38.0°C); septic arthritis, positive microbiologic culture from joint aspirate fluid in the presence of suggestive clinical features; urinary tract infection, including pyelonephritis and urosepsis, isolation of >100,000 colony-forming units/ml of urine in the presence of suggestive clinical features; pneumonia, presence of new infiltrates, consolidation, or effusion seen by chest radiography and suggestive clinical features; osteomyelitis, clinical suspicion with confirmation by definite radiologic findings or positive bone culture. Lower respiratory tract infections, skin and soft tissue infections, and acute gastroenteritis could be included on the basis of a physician's diagnosis and relevant clinical findings alone, but microbiologic culture results were recorded if available. Skin and soft tissue infections included cellulitis, abscesses, wound infections, herpes zoster, and diabetic foot infections. Intra-abdominal infections could be included on the basis of clinical findings alone, and comprised acute cholecystitis, ascending cholangitis, suppurative appendicitis, and peritonitis. The category “other infections” included episodes of otitis media and sinusitis that required hospitalization, eye infections, male and female genital tract infections, culture-proven tuberculosis, endocarditis, and infectious hepatitis.

For each episode of infection, we collected information on accompanying fever, leukocytosis, and findings of relevant investigations, including microbiologic cultures and radiologic findings. We also recorded whether the infection required hospitalization, and length of hospital stay. In the case of urinary tract infections (other than those classified as urosepsis/acute pyelonephritis), we recorded total number of culture-positive infections.

Information on RA severity measures (rheumatoid factor [RF] positivity, elevated erythrocyte sedimentation rate [ESR], extraarticular manifestations of RA, and functional status at baseline), comorbidities (diabetes mellitus, alcoholism, and chronic lung disease), and other potential risk factors for infection (presence of leukopenia, smoking status) was ascertained through review of medical records. Extraarticular manifestations of RA included Felty's syndrome, amyloidosis, rheumatoid vasculitis, and rheumatoid lung disease. Information on use of corticosteroids and disease-modifying antirheumatic drugs (DMARDs) was collected, including duration of therapy, and, in the case of corticosteroids, cumulative doses.

Data analysis.

Baseline characteristics of the study population and frequency of medication use were summarized using descriptive statistics. For the portion of the study being reported herein, 2 outcomes were studied: all infections with objective confirmation (defined as those with positive results of microbiologic cultures and/or radiologic imaging), and all infections that required hospitalization.

For each potential predictor variable, the univariate hazard ratio for developing an infection and the 95% confidence interval (95% CI) was calculated using Andersen-Gill proportional hazards models (10). These models took into account all episodes of infection that occurred during the followup period. A stepwise selection process was used to create an optimal multivariate model of potential predictors of infection. All 2-way interaction terms were examined. P values less than 0.05 were considered significant.

The following variables were included in the analyses as time-varying covariates: alcoholism, diabetes mellitus, chronic lung disease, extraarticular manifestations of RA, rheumatoid nodules, leukopenia, ESR, and use of medications, including corticosteroids and DMARDs. The other covariates (smoking, age, sex, functional capacity, body mass index, and RF positivity) were recorded at baseline only. Corticosteroid therapy was analyzed in terms of cumulative dose and duration of therapy, in addition to ever/never use.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The incidence cohort comprised 609 patients, of whom 73.1% were female. The mean age at RA incidence was 58.0 years. The total followup time was 7,729.7 person-years, and the mean followup was 12.7 years. The general characteristics of these patients, and the distribution of potential risk factors for infection, are shown in Table 1.

Table 1. Distribution of characteristics among the RA incidence cohort (1955–1994; n = 609)*
CharacteristicNo. (%)
  • *

    RA = rheumatoid arthritis; RF = rheumatoid factor; ESR = erythrocyte sedimentation rate.

  • Percent is based on the number of patients with available information; status of some patients not known.

  • By Steinbrocker criteria (28).

Female445 (73.1)
Alive at last followup326 (53.5)
Ever smoked317 (55.2)
Functional class 
 I54 (8.9)
 II427 (70.5)
 III117 (19.3)
 IV8 (1.3)
Positive RF341 (64.0)
Alcoholism42 (6.9)
Diabetes mellitus63 (10.3)
Chronic lung disease158 (25.9)
RA complications54 (8.9)
Organic brain disease82 (13.5)
Rheumatoid nodules144 (23.6)
Received chemotherapy for cancer17 (2.8)
Leukopenia102 (16.7)
Abnormal ESR at or before diagnosis317 (52.1)

Details of medication use for RA are shown in Table 2. Almost half of all patients (47.9%) received corticosteroids at some stage, and the median cumulative length of therapy was 798 days (2.2 years). Hydroxychloroquine (HCQ) had been prescribed for more than one-third of the patients (36.6%), and intramuscular gold and methotrexate (MTX) were also frequently used (22.5% and 21.8% of patients, respectively). The median length of therapy was longest for MTX (1,497 days [4.1 years]). Only 3 (0.5%) of the patients had been treated with a biologic therapy (etanercept) prior to the end of followup.

Table 2. Medication use by patients in the rheumatoid arthritis incidence cohort
Medication*No. (%) ever receivedDays received, median (interquartile range)
  • *

    IM = intramuscular; TNFR = tumor necrosis factor receptor; IV = intravenous.

  • Of patients who ever received the medication.

IM gold137 (22.5)456 (133–1,366)
Methotrexate133 (21.8)1,497 (585–2,338)
Oral gold64 (10.5)363 (159–1,344)
Sulfasalazine44 (7.2)509 (131–1,039)
Hydroxychloroquine223 (36.6)644 (237–1,871)
Azathioprine30 (4.9)676 (186–1,480)
D-penicillamine50 (8.2)298 (165–1,010)
Leflunomide19 (3.1)234 (103–297)
Cyclophosphamide7 (1.1)153 (87–743)
Cyclosporine3 (0.5)300 (290–394)
Etanercept (TNFR:Fc)3 (0.5)107 (30–239)
Corticosteroids (IV or IM)292 (47.9)798 (153–2,238)

A total of 389 patients (64%) had at least 1 infection with objective confirmation, and 290 (48%) had at least 1 infection requiring hospitalization. There were a total of 1,481 infections with objective confirmation observed during followup, yielding a rate of 19.64/100 person-years. Seven hundred forty infections requiring hospitalization occurred, with a rate of 9.57/100 person-years (6).

With regard to all infections for which there was objective confirmation, univariate analyses revealed several statistically significant (P ≤ 0.016) predictors of increased infection risk, including age, male sex, smoking, various comorbidities (including alcoholism, leukopenia, organic brain disease, diabetes mellitus, and chronic lung disease), and disease-related factors (including extraarticular RA, positive RF, rheumatoid nodules, poor functional capacity, and high ESR) (Table 3). Of the medications included in our analyses, use of corticosteroids was a strong and statistically significant predictor of infection (P < 0.001). Use of cyclophosphamide (CYC), cyclosporine, and etanercept was difficult to assess, due to the small number of patients who took these medications (7, 3, and 3 patients, respectively) and the short duration of use for these agents during the study period (median number of days 153, 300, and 107, respectively) (Table 2). However, the available data suggest that there was an increased risk of infection with use of cyclosporine or CYC (Tables 3 and 4).

Table 3. Univariate predictors of objectively confirmed infections in the RA incidence cohort*
PredictorHazard ratio95% CIP
  • *

    RA = rheumatoid arthritis; 95% CI = 95% confidence interval; ESR = erythrocyte sedimentation rate.

Demographic variables   
 Age, /10-year increment1.371.24–1.51<0.001
 Male sex1.401.06–1.850.016
 Smoking1.421.10–1.840.008
 Body mass index0.990.97–1.020.68
Comorbidities   
 Alcoholism1.911.23–2.990.004
 Leukopenia2.181.60–2.97<0.001
 Organic brain disease2.871.98–4.16<0.001
 Diabetes mellitus2.141.63–2.80<0.001
 Chronic lung disease2.612.03–3.35<0.001
Disease-related variables   
 Extraarticular RA3.092.17–4.41<0.001
 Rheumatoid factor1.781.37–2.33<0.001
 Rheumatoid nodules2.031.56–2.66<0.001
 Functional capacity1.821.46–2.27<0.001
 ESR1.671.29–2.16<0.001
Medications   
 Cancer chemotherapy4.462.40–8.27<0.001
 Methotrexate0.960.64–1.450.85
 Azathioprine1.160.69–1.950.57
 Hydroxychloroquine1.170.90–1.530.23
 Sulfasalazine1.390.69–2.790.35
 Intramuscular gold1.290.94–1.780.12
 Oral gold1.100.70–1.710.69
 D-penicillamine1.540.98–2.420.061
 Leflunomide1.680.49–5.780.41
 Cyclophosphamide2.120.77–5.830.15
 Cyclosporine2.311.17–4.570.016
 Etanercept0.360.06–2.140.26
 Corticosteroids1.921.51–2.45<0.001
Table 4. Univariate predictors of infections requiring hospitalization in the RA incidence cohort*
PredictorHazard ratio95% CIP
  • *

    See Table 3 for definitions.

Demographic variables   
 Age, /10-year increment1.491.33–1.67<0.001
 Male sex1.280.96–1.720.099
 Smoking1.290.98–1.700.071
 Body mass index1.000.97–1.030.83
Comorbidities   
 Alcoholism2.001.27–3.160.003
 Leukopenia2.171.58–2.98<0.001
 Organic brain disease2.942.08–4.16<0.001
 Diabetes mellitus2.451.84–3.27<0.001
 Chronic lung disease2.832.15–3.72<0.001
Disease-related variables   
 Extraarticular RA3.222.17–4.77<0.001
 Rheumatoid factor1.651.24–2.20<0.001
 Rheumatoid nodules1.761.32–2.33<0.001
 Functional capacity1.871.49–2.35<0.001
 ESR1.631.25–2.13<0.001
Medications   
 Chemotherapy5.022.44–10.3<0.001
 Methotrexate0.910.57–1.450.69
 Azathioprine1.240.70–2.200.45
 Hydroxychloroquine1.080.81–1.430.61
 Sulfasalazine1.200.53–2.720.67
 Intramuscular gold1.120.79–1.600.51
 Oral gold1.060.70–1.620.77
 D-penicillamine1.350.81–2.250.25
 Leflunomide2.220.88–5.640.093
 Cyclophosphamide6.143.12–11.8<0.001
 Cyclosporine1.991.25–3.160.004
 Etanercept0.360.06–2.100.26
 Corticosteroids1.901.47–2.47<0.001

The factors that were independently associated with an increase in the risk of objectively confirmed infections in multivariate models are shown in Table 5. Factors associated with the highest infection risk included increasing age (hazard ratio [HR] 1.30/10-year age increment [95% CI 1.16–1.46]) and comorbidities, including chronic lung disease (HR 1.84 [95% CI 1.40–2.41]), leukopenia (HR 1.91 [95% CI 1.40–2.61]), organic brain disease (HR 1.88 [95% CI 1.22–2.89), diabetes mellitus (HR 1.60 [95% CI 1.12–2.30]), and alcoholism (HR 1.67 [95% CI 1.16–2.41]). A number of markers of disease severity were also associated with infection risk, including extraarticular manifestations of RA (HR 2.07 [95% CI 1.41–3.06]), positive RF (HR 1.36 [95% CI 1.06–1.75]), rheumatoid nodules (HR 1.41 [95% CI 1.04–1.90]), elevated ESR (HR 1.33 [95% CI 1.05–1.68]), and reduced functional capacity (HR 1.27 [95% CI 1.04–1.56]). Of the medications analyzed, only corticosteroid use was associated with increased infection risk (HR 1.56 [95% CI 1.22–2.01]). We found that accounting for cumulative dose or duration of corticosteroid therapy provided little additional information beyond that obtained simply by accounting for a history of corticosteroid use (data not shown).

Table 5. Predictors of objectively confirmed infections in the RA incidence cohort, identified in the multivariate model*
PredictorHazard ratio95% CIP
  • *

    See Table 3 for definitions.

Demographic variables   
 Age, /10-year increment1.301.16–1.46<0.001
 Male sex1.180.90–1.550.22
Comorbidities   
 Chronic lung disease1.841.40–2.41<0.001
 Leukopenia1.911.40–2.61<0.001
 Organic brain disease1.881.22–2.890.004
 Diabetes1.601.12–2.300.011
 Alcoholism1.671.16–2.410.006
Disease-related variables   
 Extraarticular RA2.071.41–3.06<0.001
 Rheumatoid factor1.361.06–1.750.015
 Rheumatoid nodules1.411.04–1.900.025
 ESR1.331.05–1.680.019
 Functional capacity1.271.04–1.560.018
Medications   
 Corticosteroids1.561.22–2.01<0.001

A similar pattern of predictor variables was seen for infections requiring hospitalization, i.e., extraarticular manifestations of RA, chronic lung disease, and leukopenia were most strongly associated with the development of serious infection (Table 6). Sex was not associated with risk of infection in these analyses. There was a statistically significant interaction between age and diabetes mellitus, such that the excess risk of infection associated with diabetes mellitus was greater for younger compared with older RA patients. This suggests that for young RA patients, the presence of diabetes mellitus is a more important predictor of infection risk, while age is the strongest predictor of infection risk among older RA patients, with no difference between those with and those without diabetes mellitus.

Table 6. Predictors of infections requiring hospitalization in the RA incidence cohort, identified in the multivariate model*
PredictorHazard ratio95% CIP
  • *

    See Table 3 for definitions.

  • The excess risk of infection associated with diabetes mellitus was relatively high for younger people and fell with increasing age, as age dominated the excess risk.

Demographic variables   
 Male sex1.080.81–1.430.62
Comorbidities   
 Chronic lung disease2.051.53–2.74<0.001
 Leukopenia1.921.37–2.67<0.001
 Organic brain disease1.711.17–2.500.006
 Alcoholism1.851.25–2.740.002
Disease-related variables   
 Extraarticular RA2.381.58–3.60<0.001
 Rheumatoid factor1.270.98–1.660.072
 Functional capacity1.351.08–1.680.008
Medications   
 Corticosteroids1.561.20–2.040.001
Interaction term   
 Age, diabetes<0.001

Extraarticular manifestations were a consistently strong predictor of infection in all of our analyses. The cohort included 12 patients with Felty's syndrome, 17 with Sjögren's syndrome, 17 with rheumatoid lung disease, 9 with rheumatoid vasculitis, and only 2 with amyloidosis. Each of these extraarticular manifestations of RA was associated (univariately) with an increased risk of objectively confirmed infections, with the following hazard ratios: Felty's syndrome 2.31 (95% CI 1.33–3.99), Sjögren's syndrome 2.12 (95% CI 1.09–4.13), rheumatoid lung disease 2.85 (95% CI 1.56–5.18), and vasculitis 4.65 (95% CI 2.07–10.40). Extraarticular manifestations were similarly associated with increased risk of infections requiring hospitalization (Felty's syndrome 2.19 [95% CI 1.15–4.18], Sjögren's syndrome 2.33 [95% CI 0.99–5.49], rheumatoid lung disease 2.68 [95% CI 1.54–4.68], and vasculitis 6.19 [95% CI 3.20–12.00]).

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

This is the first study to examine predictors of infection in a population-based cohort of patients with RA. We identified several factors that increase the risk of development of infection in RA patients. These include measures of RA severity, advanced age, male sex, and several comorbidities. Of the disease-modifying therapies we examined, only corticosteroids consistently increased infection risk.

Measures reflecting RA severity (RF positivity, rheumatoid nodules, extraarticular manifestations of RA, increased ESR, and poor functional status) were strong predictors of infection risk in this study. This suggests that infection risk in RA is related to disease severity. It has been postulated that RA is associated with an immunologic deficit that predisposes people with this condition to develop infections. Evidence for this theory comes from laboratory data indicating that patients with RA have a contraction of their T cell receptor repertoire, the diversity of which is necessary for recognition of a wide variety of antigens (11). It has also been shown that T cell dynamics are fundamentally altered in RA, such that the ability of these patients to react to infections is compromised (12).

Alternatively, the link between RA severity and infection may simply result from immobility due to joint damage, the presence of skin defects, and multiple joint surgeries in patients with severe disease. These would all predispose to infections of the skin and joints. However, the results of our companion study comparing this cohort of RA patients with a matched control group without RA revealed statistically significant increases in overall infection risk in the RA patients for virtually all infection types studied (6).

Coexisting medical conditions were common in the RA cohort, and some were strongly associated with occurrence of infections. Leukopenia affected 102 patients in the cohort (16.7%) at some stage during followup and, not surprisingly, was associated with an increased risk of infection. In a small proportion of patients, leukopenia was secondary to Felty's syndrome, and in the other patients it may have been related to bone marrow suppression secondary to either rheumatoid disease or DMARD therapy. Diabetes mellitus has been shown to increase the risk of certain types of infection (13). Chronic lung disease such as chronic obstructive airway disease, as well as alcoholism and dementia, have all been shown to increase the risk of infections of the respiratory tract (14–16).

When examining risk of infections requiring hospitalization we found an interaction between age and diabetes mellitus, such that the relative impact of diabetes on infection risk becomes lower as age increases. A possible explanation for this is that younger people with more severe type 1 diabetes may die at an earlier age due to diabetic complications, and they would therefore not be represented in the older age groups in this cohort.

Medications were underrepresented in the list of significant risk factors for infection in this RA cohort, with only corticosteroids being associated with increased infection risk in multivariate analyses. Few studies have explored the risk of infection in RA patients treated with corticosteroids. A study comparing 112 corticosteroid-treated RA patients with a matched group of untreated RA patients showed a greatly increased infection risk in the treated group (odds ratio 8.0 [95% CI 1.0–64], P < 0.05) (17). A second study demonstrated that RA patients taking steroids developed significantly more infections than RA patients who were not taking such medications (38% versus 21%; P < 0.05) (18). Similarly, a study examining risk factors for infection among patients with systemic lupus erythematosus showed a 5-fold increase in the frequency of infection among patients treated with prednisone at an average dosage of ≥40 mg/day compared with those not treated with steroids (19).

The dosage of corticosteroids that would be associated with an increased risk of infection is unknown. A meta-analysis examining short-term corticosteroid use in patients with a variety of diseases revealed no increase in infection risk with prednisone at a dosage of <10 mg/day, or a cumulative dose of <700 mg (20). In our study, however, neither average dosage nor cumulative dose of corticosteroids was associated with infection risk when ever/never use was taken into account. This supports the findings from a previous study, in which average prednisone dosage and cumulative dose had much smaller relative risks for infection (1.25 [95% CI 1.0–1.5] for both), compared with ever prednisone use (17).

Surprisingly, in the present study, use of DMARDs, apart from corticosteroids, did not appear to increase infection risk in multivariate analysis, after adjustment for demographic and disease-related variables. Of DMARDs commonly prescribed for the treatment of RA, MTX has been the most studied. At the higher dosages used in cancer chemotherapy, MTX is clearly immunosuppressive. However, the effects of the dosages of MTX used in RA are less clear. In a prospective cohort study of 228 RA patients, the relative risk for infection in patients taking MTX was 1.52 (95% CI 1.04–2.22) (21). A study comparing the toxicity profiles of 7 DMARDs revealed more infections in patients treated with MTX or CYC than in those treated with azathioprine, HCQ, D-penicillamine, or gold (22). Another study comparing MTX with gold showed an increased infection risk in the MTX-treated group (23). In contrast to these reports, and consistent with our study findings, 2 more recent studies did not demonstrate an association between MTX use and risk of infection in patients with RA (24, 25).

Fewer data are available regarding the other DMARDs commonly used in RA. Most studies examining azathioprine, at the dosages used in RA, have not shown an association with infection risk (22, 24, 26). There is little evidence that HCQ, gold, or sulfasalazine therapy predisposes patients with RA to develop infections (27).

Our results need to be interpreted in light of the potential limitations of this study. One such limitation is that only infections that came to medical attention were included in our analyses. For this reason, we did not include infections of the upper respiratory tract or viral infections. We were therefore unable to examine a possible association between MTX therapy and increased risk of herpes zoster infection. However, we believe our data collection process assures virtually complete ascertainment of all clinically important infections that occurred in the cohort. The only available information on patients' functional status in our study was on status at the time of disease onset, and may not reflect the subsequent disease course. Because of the time period covered by this study, biologic agents were infrequently used and the number of patients treated with these agents was too small to allow an accurate examination of associated infection risk. Finally, because some racial and ethnic groups are underrepresented in Rochester, Minnesota, where the population in 1990 was 96% white according to the US Census data, the results of our population-based study are generalizable only to the US white population.

In conclusion, we have identified several strong predictors of infection among patients with RA. These factors include markers of disease severity, comorbidities, leukopenia, increasing age, and corticosteroid therapy. These results advance our understanding of the relationship between infections and RA, and may help to prospectively identify high-risk patients, facilitating extra vigilance and implementation of preventive strategies in these patients.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The authors wish to thank Ms Deborah Fogarty for her secretarial assistance in the preparation of the manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES