SEARCH

SEARCH BY CITATION

Abstract

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

Objective

To describe a group of patients who were treated with tumor necrosis factor α (TNFα) antagonists and who developed coccidioidomycosis, and to test the hypothesis that patients with inflammatory arthritis receiving TNFα antagonist therapy are at higher risk for developing symptomatic coccidioidomycosis.

Methods

Cases of coccidioidomycosis were identified and reviewed from among patients receiving TNFα antagonist therapy from May 1998 through February 2003 in 5 practices within the areas endemic for coccidioidomycosis (Arizona, California, and Nevada). In addition, the relative risk of developing symptomatic coccidioidomycosis was calculated in patients with inflammatory arthritis who were receiving treatment with infliximab, in comparison with patients with inflammatory arthritis who were not receiving infliximab, from January 2000 to February 2003 in a single medical center.

Results

Thirteen cases of documented coccidioidomycosis were associated with TNFα antagonist therapy. Twelve cases were associated with the use of infliximab and 1 case with etanercept. Among the cohort of patients from a single medical center, 7 of the 247 patients receiving infliximab and 4 of the 738 patients receiving other therapies developed symptomatic coccidioidomycosis (relative risk 5.23, 95% confidence interval 1.54–17.71; P < 0.01).

Conclusion

Patients with inflammatory arthritis who are undergoing treatment with infliximab appear to be at higher risk for developing symptomatic coccidioidomycosis as compared with those not receiving infliximab.

Biologic agents are becoming increasingly popular in the treatment of inflammatory disease. The tumor necrosis factor α (TNFα) antagonists are currently the most commonly used, with more than 600,000 patients being treated worldwide (Centocor, Inc.: unpublished data). While infliximab and etanercept are the most widely used of these agents, adalimumab has also been recently approved. Infliximab is a chimeric mouse/human monoclonal IgG1 antibody directed at TNFα. Etanercept consists of 2 copies of a recombinant human TNF receptor p75 monomer attached to the Fc domain of human IgG1. Adalimumab is a fully human monoclonal antibody that was released into the market late in 2002 and has indications for use in rheumatoid arthritis.

The Food and Drug Administration announced that the use of these biologic agents is associated with an increased risk of serious infections. There have been reports of tuberculosis associated with all of the TNFα antagonists (1, 2). Most of these cases appear to have been reactivations and not acute infections. In addition, there was a report describing cases of histoplasmosis associated with the use of infliximab or etanercept (3). This latter report was unable to conclude whether these were either reactivations or acute cases.

The soil-dwelling dimorphic fungus, Coccidioides, causes coccidioidomycosis. This type of infection is endemic to the San Joaquin Valley of California, southern Arizona, southern New Mexico, western Texas, and areas of Mexico as well as Central and South America. In most cases, infection results after inhalation of coccidioidal arthroconidia. The annual risk of infection in Tucson, Arizona, a known endemic region, has been estimated at 3% (4). However, 60% of the cases are completely asymptomatic and are only discovered because of a positive response on the coccidioidal skin test. Among patients with depressed cellular immunity, symptomatic illness has been estimated to be much higher, at ∼7% per year (5, 6).

The present report describes a number of cases of coccidioidomycosis in patients receiving TNFα antagonist treatment. The compilation of these data was prompted by the observed increased incidence of symptomatic coccidioidomycosis in our university-based outpatient practice. Furthermore, in this report, a cohort of patients with inflammatory arthritis from a single medical center was retrospectively reviewed to determine the risk of developing coccidioidomycosis during treatment with infliximab as compared with the risk with other therapies.

PATIENTS AND METHODS

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

Case series.

Documented cases of coccidioidomycosis were retrospectively collected from 5 participating practices within the regions endemic for coccidioidomycosis, comprising Arizona (Tucson and Phoenix), California (Bakersfield), and Nevada (Henderson), from May 1998 to February 2003. Clinicians from these 5 practices identified patients who developed serologically and/or pathologically proven coccidioidal infection while receiving TNFα antagonist therapy. Cases were first collected in Tucson, Arizona from the university-based outpatient practice. However, through networking with colleagues from other endemic regions, more cases were discovered. Therefore, the patients were from both university-based outpatient practices and private practices in regions endemic for coccidioidomycosis.

Charts were reviewed by one of the investigators (LB). Patients with either a new diagnosis of coccidioidomycosis or a reactivation of a previous infection were analyzed. The diagnosis of coccidioidomycosis was confirmed by serology and/or pathology. Baseline characteristics were noted, comprising age, sex, race, rheumatologic diagnosis, comorbid medical conditions, any potentially immunosuppressive medications, date of diagnosis of coccidioidomycosis or recurrence of infection, location of infection, evidence of symptoms, mode of diagno-sis, serologic evidence of infection, and outcome of infection.

Retrospective cohort study.

The single-center cohort study of coccidioidomycosis in patients receiving infliximab was conducted by retrospectively reviewing the medical records at the University Medical Center in Tucson, Arizona to identify all patients ages ≥18 years with the diagnosis of inflammatory arthritis who were treated at this medical center from January 2000 to February 2003. The diagnoses were rheumatoid arthritis, juvenile rheumatoid arthritis, reactive arthritis, and psoriatic arthritis. Patients with a new diagnosis of symptomatic coccidioidomycosis or reactivation of a previous infection were identified by a chart review of all patients with inflammatory arthritis, using International Classification of Diseases, Ninth Revision codes for coccidioidomycosis, and this was verified by their treating rheumatologists. All of these patients were from a university-based outpatient practice.

The cumulative incidence of symptomatic coccidioidomycosis was obtained both in patients receiving TNFα antagonist therapy and in those receiving other therapies, as a group. Adalimumab was not available at the time of this study. There were no cases of coccidioidomycosis observed in patients receiving etanercept therapy in the single site where the retrospective cohort analysis was performed. Therefore, the incidence of disease among patients receiving infliximab was compared with that among patients receiving other therapies, including etanercept.

Statistical analysis.

A relative risk (RR) was calculated and 95% confidence intervals (95% CIs) estimated using Epi-Info 2002 (Centers for Disease Control and Prevention, Atlanta, GA). A chi-square statistic and its corresponding P value were also calculated using Stata, version 6.0 (Stata Corporation, College Station, TX). Univariate and multivariate logistic regression modeling were used to test for the association between infliximab use and symptomatic coccidioidomycosis, adjusting for age, methotrexate use, and prednisone use. SPSS, version 11.5 (SPSS, Chicago, IL) was used for the logistic regression analyses.

RESULTS

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

Characteristics of the case series.

Thirteen patients with active coccidioidomycosis associated with the use of TNFα antagonist therapy (Table 1) were identified. One of these cases was associated with etanercept and 12 were associated with infliximab. None of the patients had significant comorbid conditions that have been associated with an increased risk of disseminated coccidioidomycosis, such as diabetes, human immunodeficiency virus (HIV) syndrome, or pregnancy. Eleven of the 12 patients who were receiving infliximab were also receiving methotrexate. Two of these patients were also taking additional immunosuppressive medication.

Table 1. Case series of coccidioidomycosis associated with tumor necrosis factor α antagonist therapy*
Patient/age/sexDate of entryRaceRheumatologic diagnosisHistory of coccidioidomycosis (time since prior infection)Medications and dosagesInterval between infliximab treatment and infection, weeksSerology results, pretreatment/posttreatmentMethod of diagnosis of infectionCourse of infection
  • *

    RA = rheumatoid arthritis; q8 weeks = every 8 weeks; JRA = juvenile rheumatoid arthritis; bid = twice daily; PsA = psoriatic arthritis.

  • Neg. = negative for anticoccidioidal antibodies; pos. = positive for both IgM and IgG anticoccidioidal antibodies. ND = serology not done.

  • Interval in parentheses indicates the number of weeks between a dosage change and infection.

Arizona         
 1/64/FMarch 2001WhiteRAYes (2.5 years)Infliximab 3 mg/kg × 2, methotrexate 7.5 mg, itraconazole2Neg./pos.SerologyLeft upper lobe pulmonary cavity, pneumonia, resolution
 2/71/FOctober 2001WhiteRAYes (2 years)Infliximab 4 mg/kg q8 weeks, methotrexate 10 mg44ND/IgM+SerologyPneumonia, resolution
 3/70/FMay 1998WhiteRANoInfliximab 10 mg/kg q8 weeks, methotrexate 15 mg40ND/NDPathologyDissemination, death
 4/51/FSeptember 2000WhiteJRANoInfliximab 3 mg/kg q8 weeks, prednisone 8 mg daily, cyclosporine 100 mg bid, methotrexate 25 mg9ND/pos.SerologyDissemination, resolution
 5/72/FJune 2002HispanicRANoInfliximab 3 mg/kg q8 weeks, methotrexate 20 mg1Neg./pos.Serology + pathologyPneumonia, resolution
 6/66/MJuly 2002WhiteRANoInfliximab 10 mg/kg q8 weeks, methotrexate 12.5 mg4Neg./pos.SerologyPneumonia, resolution
 7/76/FSeptember 2002WhiteRANoInfliximab 3 mg/kg q8 weeks, methotrexate 15 mg13ND/IgM+SerologyRight lower lobe pneumonia, resolution
 8/68/FSeptember 2002WhiteRANoInfliximab 5 mg/kg q8 weeks, azathioprine 100 mg daily, methotrexate 7.5 mg36 (9)Neg./IgG+SerologyRight upper lobe pneumonia, resolution
 9/54/FDecember 2002WhiteRANoInfliximab 5 mg/kg q6 weeks, hydroxychloroquine 200 mg bid, methotrexate 12.5 mg44 (1.5)Neg./pos.SerologyRight middle lobe pneumonia, resolution
 10/72/MNovember 2002WhiteRANoInfliximab 5 mg/kg q8 weeks, methotrexate 20 mg48Neg./pos.Serology + pathologyDissemination, resolution
Nevada         
 11/65/FJanuary 2002HispanicPsANoInfliximab 5 mg/kg q8 weeks, methotrexate 22.5 mg6ND/neg.PathologyDissemination, death
California         
 12/58/MAugust 2002WhiteRANoInfliximab 3 mg/kg q8 weeks, azathioprine 100 mg daily12ND/pos.SerologyPneumonia, resolution
 13/52/FJanuary 2003WhitePsANoEtanercept 25 mg biweekly, methotrexate 20 mg96ND/pos.SerologyLeft upper lobe pneumonia, resolution

On presentation, all 13 patients in the case series had pneumonia. The chest radiograph from a represen- tative case is shown in Figure 1 (case 5). Four patients had documented dissemination of infection (including in the blood, spleen, peritoneum, and central nervous system). Eleven patients had positive serologic findings of IgM and/or IgG anticoccidioidal antibodies. Four patients had pathologic evidence (i.e., biopsy or culture) of coccidioidomycosis. One patient had negative coccidioidal serologic findings but had histologic evidence of active coccidioidomycosis.

thumbnail image

Figure 1. Chest radiograph revealing a representative case of coccidioidomycosis in a 72-year-old female patient with rheumatoid arthritis who presented with pneumonia.

Download figure to PowerPoint

Five patients were hospitalized. All but 1 patient was treated with antifungal medication. Two patients died. One death was due to staphylococcal sepsis secondary to a line placed to administer antifungal therapy. The other death was in a patient presenting with a large pulmonary infiltrate and negative sputum and blood cultures; she was originally treated with antibiotics, and the day following her death, Coccidioides was isolated from her lung biopsy specimen. Among the remaining 11 cases in which the individuals survived, the infection was recognized and treatment with antifungal therapy was begun.

Illustrative cases.

Patient 1 (reactivation).

Patient 1 is a 64-year-old woman with a 24-year history of rheumatoid arthritis who was previously treated with hydroxychloroquine, prednisone, methotrexate, and sulfasalazine. In September of 1998, while taking methotrexate and prednisone, she was diagnosed as having coccidioidomycosis. Treatment with fluconazole, and later itraconazole, was started and continued. Although the serologic findings of coccidioidomycosis were originally positive, they became negative after initiation of antifungal treatment. The patient had a stable pulmonary cavity in the left upper lobe. In May of 1999, leflunomide was added to her regimen, but was discontinued secondary to lack of response. In November of 1999, she was placed on etanercept. She continued to receive itraconazole, and after 17 months of exhibiting a good response to etanercept, the patient's arthritis worsened.

In March of 2001, the patient's regimen was changed to infliximab and methotrexate. No serologic tests were done prior to placing the patient on infliximab. After her first infusion of infliximab, she developed a productive cough. A computed tomography scan of her thorax revealed an increase in the size of the pulmonary cavitary lesion in association with a positive anticoccidioidal IgM serologic response. The dose of itraconazole was increased and the infliximab was discontinued. Her symptoms subsequently resolved and the cavity stabilized.

Patient 8 (probable acute disease).

Patient 8 is a 68-year-old woman with a long history of rheumatoid arthritis and hypertension and no prior occurrence of coccidioidomycosis. For 9 months, she had been taking infliximab at 3 mg/kg every 8 weeks, azathioprine at 100 mg per day, and methotrexate at 7.5 mg once per week. She subsequently developed worsening of her joint pain and swelling. Her infliximab dose was increased to 5 mg/kg. Within 9 weeks of this increase, she developed a pulmonary infiltrate in the right upper lobe. Coccidioidal serologic findings were positive for IgG anticoccidioidal antibodies, whereas prior to infliximab therapy, the serologic results had been negative. The patient was treated with fluconazole and her pneumonia resolved.

Patient 9 (probable acute disease).

Patient 9 is a 54-year-old woman with a 22-year history of rheumatoid arthritis who, for 11 months, had been treated with infliximab at 5 mg/kg every 8 weeks, methotrexate at 12.5 mg once per week, and hydroxychloroquine at 200 mg twice per day. After her arthritis worsened, the interval for treatment with infliximab was decreased to every 6 weeks. Two weeks later, she developed pneumonia in the right middle lobe, and the coccidioidal serologic findings were positive for IgM and IgG anticoccidioidal antibodies. Prior to the initiation of infliximab, the chest radiograph and coccidioidal serologic results were negative. The patient was treated with fluconazole and her symptoms resolved.

Incidence of coccidioidomycosis in a cohort of infliximab-treated patients.

A total of 985 patients with inflammatory arthritis (ages ≥18 years) were evaluated from January 2000 to February 2003 at University Medical Center in Tucson, Arizona. Rheumatologic diagnoses for this cohort included 845 patients with rheumatoid arthritis, 50 with juvenile rheumatoid arthritis, 70 with psoriatic arthritis, and 20 with reactive arthritis. Eleven of the 985 patients developed symptomatic coccidioidomycosis during this period, yielding a cumulative incidence of 1%. Among these patients, 7 of the 247 patients receiving infliximab therapy developed symptomatic coccidioidomycosis (group A), while 4 of the 738 patients treated with medications other than infliximab developed symptomatic coccidioidomycosis (group B) (Table 2) (RR 5.23, 95% CI 1.54–17.71; P < 0.01). There was one other case in which a patient with reactive arthritis who was taking hydroxychloroquine was found to have a nodule in her lung on a routine preoperative chest radiograph. The patient had no documented symptoms, but a biopsy was performed to rule out the possibility of malignancy. The biopsy results were consistent with a coccidioidal nodule, indicating previous infection. The patient was asymptomatic and was not treated with antifungal medications. Therefore, this patient was excluded from the above analysis.

Table 2. Comparison of inflammatory arthritis cohorts with and without infection
 CoccidioidomycosisTotalIncidence of infection
Yes*No
  • *

    Group A = patients with coccidioidomycosis receiving infliximab therapy; group B = patients with coccidioidomycosis treated with medications other than infliximab.

Taking infliximab, no.    
 Yes7 (group A)2402470.028
 No4 (group B)7347380.005
Total119749850.011
Relative risk5.23   
95% confidence interval1.54–17.71   
χ28.80   
P<0.01   

Using logistic regression modeling, we found that the use of infliximab was associated with the presence of symptomatic coccidioidomycosis even after adjusting for methotrexate use, prednisone use, and age. Table 3 shows the results of separate univariate and multivariate models that tested the relationship between these variables and symptomatic coccidioidomycosis.

Table 3. Logistic regression models for the risk of coccidioidomycosis
ModelVariableCoefficientSEPOR95% CI for OR
LowerUpper
  1. * OR = odds ratio; 95% CI = 95% confidence interval; MTX = methotrexate; Pred. = prednisone.

Model 1Infliximab1.680.630.0085.351.5518.44
Model 2Age0.030.020.2131.030.981.07
Model 3MTX1.490.780.0584.440.9520.64
Model 4Pred.0.210.630.7391.230.364.25
Model 5Infliximab1.430.650.0274.171.1714.80
 MTX1.140.800.1573.120.6515.08
Model 6Infliximab1.780.640.0055.921.7020.66
 Age0.030.020.1371.030.991.08
Model 7Infliximab1.550.650.0174.721.3116.95
 MTX1.130.800.1603.090.6414.92
 Age0.040.020.1361.040.991.08

A total of 974 patients had no evidence of symptomatic coccidioidomycosis (group C). There was no statistical difference in the mean age between all 3 groups. The sex, race, and comorbidities were also similar among the patients in each of the 3 groups (Table 4). The mean length of rheumatic disease was similar between group A and group B, with durations of 15.9 years and 19.8 years, respectively. Among the patients in group A, all were taking infliximab in combination with methotrexate. In groups B and C, 50% of the patients were taking methotrexate. The mean dose of methotrexate was similar between all 3 groups (15.36 mg, 18.75 mg, and 14.12 mg in group A, group B, and group C, respectively). Fourteen percent of patients in group A were taking prednisone, compared with 75% of patients in group B, and the mean doses were 8 mg and 10 mg, respectively. Of the patients in group C, 32% were taking prednisone.

Table 4. Comparison of demographic and clinical characteristics of cohort patients*
 Coccidioidomycosis (n = 11)No coccidioidomycosis, group C (n = 974)
Group A (n = 7)Group B (n = 4)
  • *

    Except where indicated otherwise, values are the number (%) of patients. RA = rheumatoid arthritis; JRA = juvenile rheumatoid arthritis; PsA = psoriatic arthritis; ReA = reactive arthritis; MTX = methotrexate.

  • Group A = patients receiving infliximab therapy; group B = patients treated with medications other than infliximab.

Mean age (range), years64.8 (51–72)64.0 (56–69)57.8 (18–89)
Female5 (71)3 (75)746 (77)
Race
 White6 (86)3 (75)707 (73)
 Hispanic1 (14)1 (25)180 (18)
RA6 (86)3 (75)836 (86)
JRA1 (14)049 (5)
PsA01 (25)69 (7)
ReA0020 (2)
Infliximab7 (100)0240 (25)
Etanercept0076 (8)
MTX7 (100)2 (50)490 (50)
 Mean MTX dose, mg15.418.814.1
Prednisone1 (14)3 (75)308 (32)
History of coccidioidomycosis1 (14)1 (25)6 (0.006)
Dissemination of infection2 (29)1 (25)0

DISCUSSION

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

Coccidioidomycosis is a common infection in the southwestern United States that has been increasing in incidence over the last decade (7, 8). Cellular immunity plays a critical role in the host defense against this organism, as evidenced by the increased risk of severe coccidioidomycosis among those with depressed cellular immune function (9, 10). In this case series, we have described 13 cases of symptomatic coccidioidomycosis in association with inflammatory arthritis and TNFα antagonist therapy.

All but 1 of the 13 patients in the case series had normal chest radiographs prior to TNFα antagonist treatment. Coccidioidal serologic results were also negative in the 6 patients whose samples were tested serologically prior to TNFα antagonist therapy. Two of the patients had a history of coccidioidomycosis. On the basis of this observation, it is believed that 2 of the 13 cases were consistent with a reactivation of a prior coccidioidal infection. Although serologic testing for anticoccidioidal IgM and IgG is useful for detecting acute, active infection, titers of both types of antibodies become negative within months after infection in those patients whose infection is controlled (11, 12).

Tests for delayed-type hypersensitivity, which is an indication of prior infection and can persist for prolonged periods, are not currently available for general use in the United States and might not have been useful in this group of patients with impaired immunity. Despite these limitations, cases 3–11 (Table 1) appear to represent new infections with coccidioidomycosis. These findings are consistent with those of a prospective study of coccidioidomycosis among HIV-infected persons living in a coccidioidomycosis-endemic area (13). The 41-month study found that symptomatic coccidioidomycosis developed in nearly 25% of an HIV-infected cohort who did not have a previous delayed-type hypersensitivity response to coccidioidomycosis.

TNFα may play a unique role in controlling infections associated with granulomatous inflammation, such as tuberculosis, histoplasmosis, and coccidioidomycosis. When TNF-knockout mice are infected with Mycobacterium tuberculosis, they have altered granuloma formation and reduced macrophage apoptosis (14). Investigations in mouse models and humans suggest that TNFα plays a role in the control of coccidioidal infection. TNFα is produced by human peripheral blood monocytes as well as murine peritoneal macrophages in response to incubation with coccidioidal spherules (15, 16). Human monocytes incubated with recombinant TNFα possessed greater fungicidal capabilities (17). Increased levels of TNFα have been found in the spleens of genetically resistant mice compared with genetically susceptible mice after coccidioidal infection (18).

With regard to tuberculosis, the median interval from initiation of infliximab therapy to development of active infection was 12 weeks according to one review; it was believed that most of these cases were reactivations of latent tuberculosis (19). There were 2 approximate time points (7 weeks and 40 weeks) when patients presented clinically with coccidioidomycosis in our study. The earlier time point is similar to the time point indicated in the study of tuberculosis. In 2 patients, the latter time point corresponds to a dosage increase or a decrease in the interval between treatments as a result of increased arthritic symptoms. It is possible that the increase in infliximab blood levels that occurred induced sufficient immune suppression to result in reactivation of latent coccidioidomycosis or increased the susceptibility to primary infection. It is also possible that flares in rheumatic symptoms that resulted in the use of infliximab were due to “desert rheumatism,” the articular form of acute coccidioidomycosis (20). However, at the time that infliximab was initiated, most patients had chest radiographs with no documented abnormalities and no skin manifestations that are associated with this syndrome of acute coccidioidomycosis (i.e., erythema multiforme or erythema nodosum). Therefore, this theory is less likely.

Of note, adalimumab was not available for general public use at the time that the data were collected, and for that reason, only infliximab and etanercept are discussed. Only 1 case of active coccidioidomycosis was associated with etanercept use in this case series and none in the cohort analysis. There are several possible explanations for this difference.

Until recently, etanercept had very limited availability, especially over the time period that these cases were collected. In the cohort analysis, because of the limited availability of etanercept, we reviewed 247 patients who were receiving infliximab and only 76 patients who were receiving etanercept.

Another possible explanation for the difference in risk of infection is the difference in pharmacologic properties between these 2 agents. Infliximab has approximately double the half-life of etanercept and a stronger binding avidity. The ability of infliximab to target soluble and cell-bound TNF and, through complement consumption, cause cell lysis (in vitro) may have a more profound immunosuppressive effect than that of the receptor inhibition of etanercept (1).

A final explanation for the difference could be related to the concomitant use of methotrexate with infliximab, resulting in further immune suppression. Eleven of the 2 patients in the case series who developed coccidioidomycosis while receiving infliximab were also taking methotrexate. Within the cohort analysis, 70% of the 247 patients receiving infliximab were also taking methotrexate compared with 39% of the 76 patients receiving etanercept. However, since the logistic regression modeling still found that infliximab was associated with a statistically significant increase in symptomatic coccidioidomycosis, even when adjusting for methotrexate use, this is less likely to be an issue. All of the baseline demographics, including age, were similar between the infliximab- and etanercept-treated patients. Therefore, there is no difference in the patient populations that would explain the increased risk.

Our cohort study was a retrospective analysis that, as a result, had many inherent limitations. These include the potential effect of uncontrolled confounding variables such as rheumatic disease activity. Even though we did not see any significant differences when considering age, prednisone use, and methotrexate use, we were unable to analyze the effect of the severity of rheumatic disease due to a lack of reliable measurement data from all patients. Patient and/or physician assessment scores and joint counts were not available for all patients. We also collected data from multiple participating physicians and relied on medical record reviews to compile our data. Despite these potential weaknesses, we found that use of infliximab represented a significant risk for the development of symptomatic coccidioidomycosis.

There were also potential weaknesses in our case study. We did not collect data from all of the rheumatologists practicing in coccidioidomycosis-endemic regions, nor was it a random sampling of rheumatologists. We elected to analyze patients with inflammatory arthritis, rather than just strictly those with rheumatoid arthritis, to gain a larger sample size. This makes for a heterogeneous sample, but we believe that since these are the patients who would be placed on TNFα antagonists, it makes our data more applicable to everyday practice. We acknowledge that there are many limitations to this form of data collection. However, we believe that the observations are similar to what has been seen in other granulomatous diseases. We also believe that it is important for health care professionals and patients to be aware of this increased risk. Our report provides the first large-scale case series on coccidioidomycosis and infliximab and finds statistically significant conclusions regarding the increased relative risk of coccidioidomycosis in association with infliximab use.

In the southwestern United States, coccidioidomycosis is a potentially fatal disease, especially in those with an impaired immune system. In this study, there was a significant increase in the risk of developing symptomatic coccidioidomycosis in patients with inflammatory arthritis who were taking infliximab. Time and further research will determine whether use of other TNFα antagonists will result in this same risk. The current regional recommendations for patients treated with any of the TNFα antagonists include a screening chest radiograph, tuberculin skin test, and coccidioidal serologic tests for IgM and IgG. However, data from this study suggest that cases of coccidioidomycosis are more likely to be acute and not detected by these screening measures. Therefore, it may be that these recommendations will just serve as baseline markers. In the future, we hope to further explore whether these are truly acute cases by focusing our investigations on immune evaluations prior to and during TNFα antagonist therapy. We continue to recommend these medications for our patients with inflammatory arthritis, but we have a heightened awareness of the symptoms of acute coccidioidomycosis.

REFERENCES

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