Coccidioidomycosis (valley fever) is an endemic fungal infection of the American Southwest, an area with a large population of patients with rheumatic diseases. There are currently no guidelines for management of patients who develop coccidioidomycosis while under treatment with biologic response modifiers (BRMs) or disease-modifying antirheumatic drugs (DMARDs). We conducted a retrospective study of how both concurrent diseases were managed and the patient outcomes at 2 centers in Tucson, Arizona.
A retrospective chart review identified patients who developed coccidioidomycosis during treatment with DMARDs or BRMs. Patients were seen at least once in a university-affiliated or Veterans Affairs outpatient rheumatology clinic in Tucson, Arizona, between 2007 and 2009.
Forty-four patients were identified. Rheumatologic treatment included a BRM alone (n = 11), a DMARD alone (n = 8), or combination therapy (n = 25). Manifestations of coccidioidomycosis included pulmonary infection (n = 29), disseminated disease (n = 9), and asymptomatic positive coccidioidal serologies (n = 6). After the diagnosis of coccidioidomycosis, 26 patients had BRMs and DMARDs stopped, 8 patients had BRMs stopped but DMARD therapy continued, and 10 patients had no change in their immunosuppressive therapy. Forty-one patients had antifungal therapy initiated for 1 month or longer. Followup data were available for 38 patients. BRM and/or DMARD therapy was continued or resumed in 33 patients, only 16 of whom continued concurrent antifungal therapy. None of the patients have had subsequent dissemination or complications of coccidioidomycosis.
Re-treating rheumatic disease patients with a BRM and/or a DMARD after coccidioidomycosis appears to be safe in some patients. We propose a management strategy based on coccidioidomycosis disease activity.
Coccidioides species are dimorphic fungi endemic to California, Arizona, New Mexico, Texas, northern Mexico, and parts of Central and South America. Infection results from inhalation of arthroconidia of Coccidioides species after they are released from the soil into the air. Approximately 150,000 new infections occur annually in the US (1). In normal hosts, most infections are either asymptomatic or produce a self-limited pulmonary illness difficult to distinguish from other forms of community-acquired pneumonia. Small studies have found that as much as 29% of the cases of community-acquired pneumonia in the endemic region may be due to coccidioidomycosis (2, 3), and a similar risk would be expected in travelers to endemic regions who develop pneumonia shortly after their return. Up to 5% of patients develop complicated pulmonary disease, and approximately 1% develop disseminated disease (4). The risk of dissemination is much higher in immunocompromised patients. The most common sites of dissemination are skin, joints, bones, and meninges. Reactivation of previously controlled disease has also been reported in immunosuppressed patients (5).
Treatment of rheumatologic diseases has been greatly improved by the use of biologic response modifiers (BRMs), defined as agents derived from biologic production systems that target and modify key components of the immune system. These include tumor necrosis factor (TNF) antagonists and soluble inhibitors of T cell activation, such as abatacept, among others. BRMs are often given in conjunction with traditional disease-modifying antirheumatic drugs (DMARDs), including methotrexate, azathioprine, and leflunomide. Treatment with BRMs and DMARDs, however, is associated with an increased risk of infection (6, 7). The risk of coccidioidomycosis in patients treated with TNF inhibitors has been examined in 2 previously published studies (8, 9). Both found a significant (1–2%) occurrence of coccidioidomycosis, with a rate of disseminated disease that appeared to be higher than the general population. The dilemma facing rheumatologists is whether a BRM and/or a DMARD after infection with coccidioidomycosis can be continued or must be withdrawn because of the risk of worsening or recurring infection.
Significance & Innovations
This is the largest series to date of patients who developed coccidioidomycosis (valley fever), an endemic fungal infection of the American Southwest, during treatment with biologic response modifiers (BRMs) and/or disease-modifying antirheumatic drugs (DMARDs).
A significant number of patients in our study were able to resume or continue BRM and/or DMARD treatment without experiencing complications from coccidioidomycosis.
This problem will become increasingly important as the population of patients with rheumatic diseases in the endemic area increases and new immunosuppressive drugs become available.
PATIENTS AND METHODS
Human subjects and data collection.
The project was approved by the University of Arizona Human Subjects Protection Program and by the Southern Arizona Veterans Affairs Health Care System (SAVAHCS) Research and Development Committee and the Subcommittee on Research Safety.
SAVAHCS patients were chosen by an administrative search (programmed extraction of patient files within the SAVAHCS electronic medical records system) conducted by SAVAHCS Decision Support Service personnel. The search identified living or deceased patients with one or more coccidioidomycosis diagnostic codes, who had taken one of the following medications between January 1, 2007 and December 31, 2009: abatacept, adalimumab, etanercept, infliximab (all BRMs), methotrexate, leflunomide, or azathioprine (all DMARDs). The university-based rheumatology clinic patients were chosen by reviewing the charts of all patients seen at least once between January 1, 2007 and December 31, 2009 who developed coccidioidomycosis while being treated with one of the above medications. The charts reviewed were selected by a programmed extraction of patient names from the clinic's electronic medical records system by using the same criteria mentioned above.
We reviewed each chart and identified the mode of diagnosis, clinical manifestations, antifungal therapy and duration, and the management of BRMs and DMARDs after the diagnosis of coccidioidomycosis was made. Rationale for management decisions was noted wherever available. In both clinics, it is standard practice to annually screen patients on BRMs for coccidioidomycosis using a serologic test (10).
Coccidioidomycosis was defined as at least 1 of the following: any positive Coccidioides serologic test (enzyme immunoassay, immunodiffusion, or complement fixation); a positive fungal culture; detection of spherules in a tissue or fluid specimen; or a clinical diagnosis of coccidioidomycosis by a highly experienced clinician (NMA or JNG) specializing in coccidioidomycosis. Coccidioidomycosis was defined as asymptomatic if the only criterion met was a positive serologic test, with no compatible symptoms. Pulmonary coccidioidomycosis was defined as a positive test plus a new radiographic finding with or without lower respiratory symptoms, a rash compatible with erythema nodosum, fever, or night sweats. Disseminated coccidioidomycosis was defined as a positive culture or histologic finding from a nonpulmonary site.
We reviewed a total of 485 charts and identified 44 patients who developed coccidioidomycosis during treatment with BRMs and/or DMARDs (Table 1). The majority had a diagnosis of rheumatoid arthritis (n = 33), were white (n = 33), and were age >50 years (n = 38) (Table 1). Twenty-nine patients had positive Coccidioides serologic test results (5 patients with titers of 1:16 or greater, as detailed below), 4 patients had negative serologic test results but a positive sputum culture and/or cytology finding, 9 patients had a tissue diagnosis of disseminated disease with or without a positive serologic test result, and only 2 patients were diagnosed on the basis of clinical presentation alone.
Table 1. Summary of characteristics of 44 patients who developed coccidioidomycosis during treatment with biologic response modifiers and/or disease-modifying antirheumatic drugs*
No. of patients
IBD = irritable bowel disease.
IBD with spondylitis
Systemic lupus erythematosus
At the time of diagnosis of coccidioidomycosis, 11 patients were being treated with a BRM alone, 8 with a DMARD alone, and 25 with a combination of a BRM and a DMARD. BRMs being taken were infliximab (n = 21), adalimumab (n = 8), etanercept (n = 6), and abatacept (n = 1). DMARDs being taken were methotrexate (n = 26), azathioprine (n = 5), and leflunomide (n = 2). The most common combination was infliximab and methotrexate (n = 11) (Table 2).
Six (14%) patients had asymptomatic coccidioidomycosis discovered on routine annual serologic testing. One of these patients had an antibody titer of 1:16, which subsequently declined to negative. Twenty-nine (66%) patients had pulmonary coccidioidomycosis. Presentation of the illness was a lung infiltrate in 3 patients, lung nodule in 11 patients, infiltrate evolving into a nodule or mass in 7 patients, cavity in 2 patients, pleural effusion in 4 patients, and lower respiratory symptoms with positive serology test results in 2 patients. One patient, who was taking a DMARD alone, required resection of a coccidioidal lung cavity due to persistent scarring, bronchiectasis, and hemoptysis. The DMARD was subsequently stopped due to remission of rheumatic disease, and antifungal therapy was stopped 4 months postoperatively. One patient with ankylosing spondylitis had progressive interstitial lung disease and pulmonary hypertension 3 years after diagnosis of pulmonary coccidioidomycosis. These findings were believed by the treating clinician to be not clearly related to coccidioidomycosis. No other patient required hospital admission or surgery for pulmonary disease. One patient with pulmonary disease had a high titer antibody (1:16) that declined over time. Nine (20%) patients had disseminated disease, with involvement of the skin in 4 patients, joints in 2 patients (1 knee and 1 ankle), larynx in 1 patient, extrathoracic lymph node in 1 patient, and meninges in 1 patient. Both patients with articular coccidioidomycosis required surgery for diagnosis and/or treatment. Three patients with disseminated disease had antibody titers greater than 1:16. Two declined to 1:2 or less in followup, and 1 had no subsequent titer available in the record.
Of the 8 patients who were taking a DMARD alone at the time of diagnosis, 2 developed asymptomatic disease (both taking methotrexate), 4 developed pulmonary coccidioidomycosis (2 each taking azathioprine and methotrexate), and 2 developed disseminated coccidioidomycosis to the joints (1 each taking azathioprine and methotrexate) (Table 2).
At the time of coccidioidomycosis diagnosis, 14 of the 25 patients who were taking both BRMs and DMARDs stopped both drugs, while 8 stopped BRMs and continued on DMARDs, and 3 continued on both. Of the 11 patients taking BRMs alone, 8 stopped the drug and 3 continued taking it. Of the 8 patients taking DMARDs alone, 4 stopped the drug and 4 continued taking it. Overall, 34 (77%) of 44 patients stopped 1 or more immunosuppressive drugs, while 10 (23%) patients had no change in therapy (Table 2). Forty-one (93%) patients were started on antifungal therapy, most commonly fluconazole at a dosage of 400 mg daily. Three patients were not treated with antifungal therapy, 2 because they were asymptomatic and 1 for reasons not documented. Duration of antifungal therapy was 0 to 96 months with a median of 12 months.
Followup data were available for 38 patients. As of the last followup visit, 23 patients were receiving BRMs and DMARDs, and 10 were receiving DMARDs alone. Time from diagnosis of coccidioidomycosis to restarting a DMARD was 0 to 48 months, with a median duration of 1 month. Time to restarting a BRM was 0 to 72 months, with a median duration of 10 months. Four patients were not re-treated with a BRM or DMARD due to the remission of their rheumatic disease, and 1 patient was not re-treated due to disseminated coccidioidomycosis. Of the 33 patients who were on BRMs and/or DMARDs at the last followup visit, 17 had stopped antifungal treatment after 1 to 96 months (median 9 months) of therapy. The most common reason for stopping antifungal therapy was a subsequent negative serology test result (n = 9), followed by resolution of clinical illness in patients who were never seropositive (n = 3) and adverse reactions (n = 3). The rationale was not clearly documented in 2 cases. Followup for patients who stopped antifungal therapy or were never treated is 3 to 96 months (median 30 months).
The remaining 16 patients continued on antifungal therapy as of the last followup visit. The 2 most common reasons for doing so were persistent positive serologies (n = 8) and disseminated coccidioidomycosis (n = 6).
None of these 38 patients, including the 17 whose antifungal therapy was discontinued, developed more severe coccidioidomycosis disease during the followup period. However, there was 1 patient with possible coccidioidomycosis reactivation while taking a BRM: a 59-year-old female was diagnosed with asymptomatic coccidioidomycosis in 2005 based on positive serology test results. She was treated with fluconazole at that time and also continued to receive infliximab. Her serologic studies became negative for 3 years and then became positive again, although she remained asymptomatic for a followup period of 6 years. The reappearance of a positive IgG antibody was thought to possibly represent reactivation of her coccidioidomycosis after a period of negative serologic testing. Therefore, the decision was made to continue treatment with fluconazole.
Twelve patients were taking prednisone at various doses when diagnosed with coccidioidomycosis. Seven of them developed pulmonary disease, 4 disseminated disease, and 1 asymptomatic coccidioidomycosis. One patient was restarted on prednisone after initial antifungal treatment without any evidence of active coccidioidomycosis as of the last followup visit. Nineteen patients were not on prednisone at the time of their coccidioidomycosis diagnosis. Data about prednisone use are not available for the remaining 13 patients.
In this study of 44 patients who developed coccidioidomycosis while taking DMARDs and/or BRMs, we found that many could be managed during their infection and then restarted on these medications, with or without concomitant antifungal therapy. BRMs and/or DMARDs were continued or resumed in 33 of the 38 patients for whom followup was available, and none of these patients had unexpected clinical complications from coccidioidomycosis at a median of 30 months of followup, including 17 for whom antifungal therapy had been discontinued. Almost all of our patients were treated with antifungal therapy for a median of 12 months, a duration in keeping with current treatment guidelines (11). Our results suggest that continued therapy with BRMs and/or DMARDs and discontinuation of antifungal therapy may be safe in some patients with coccidioidomycosis.
The use of corticosteroids may also increase the risk of severe coccidioidomycosis (12), including fungemia (13). We did not find a difference in clinical course or outcome in the 12 patients who were receiving prednisone at the time of their coccidioidomycosis diagnosis, compared with those not taking prednisone. However, of the 12 patients identified, 11 (91%) of them had clinical disease, rather than asymptomatic positive serologies. Although corticosteroid use is felt to be clinically important, this series had limited data regarding prednisone use; therefore, firm conclusions cannot be drawn.
Based on these results, we propose a tentative management strategy for patients who develop coccidioidomycosis during treatment with a BRM and/or DMARD. All patients should be evaluated clinically with a focus on pulmonary, skin, joint, and meningeal symptoms and signs. Coccidioides serology and a chest radiograph should be obtained, with the need for other imaging studies guided by the clinical scenario. All patients should be treated with antifungal therapy according to established guidelines (11). If coccidioidomycosis is asymptomatic and the rheumatic disease remains active, the clinician can consider continuation of BRMs and/or DMARDs with careful followup. For pulmonary or disseminated coccidioidomycosis, it is prudent to discontinue BRMs and DMARDs at least temporarily and evaluate the patient periodically. Pulmonary disease usually resolves over time as demonstrated by symptoms and imaging studies, and in that case resumption of BRM and/or DMARD therapy should be considered. Coccidioides serology may be useful in determining the disease activity, although it should be noted that cell-mediated immunity, while more difficult to measure, is probably more important in protection from coccidioidomycosis (14). For patients with disseminated coccidioidomycosis or pulmonary disease that remains active, caution should be used in resuming BRM and/or DMARD therapy. A few patients with disease limited to the skin resumed such therapy in our study, with good outcomes to date. Our recommendations are outlined in Figure 1.
Limitations of this study include its small size and retrospective nature. In addition, most of the patients in our study had mild disease, so the findings are not applicable to patients with severe coccidioidomycosis, such as miliary lung disease or meningitis. Due to the lack of specific guidelines for the treatment of these patients, the length of antifungal treatment and management of BRMs and DMARDs were decided by individual clinicians on a case-by-case basis. The racial and ethnic makeup of our study population is not representative of all areas of the US, or even all of the endemic areas for coccidioidomycosis. African Americans and Asians both represent a small minority in Arizona, but African Americans and Filipinos are probably at higher risk compared to whites to develop disseminated coccidioidomycosis (11). However, despite these limitations, this is the largest series to date and the first study to focus on the course of action taken during and after patients were diagnosed with coccidioidomycosis.
Coccidioidomycosis in patients taking BRMs and/or DMARDs will continue to be an important challenge. The population of the endemic areas of the Southwestern US is increasing. New immunosuppressive drugs are being developed and marketed that improve outcomes in autoimmune diseases but carry an often unquantified risk of infection. Additional studies are needed with larger groups of patients and longer followup to further clarify the best approach for patients on BRMs and/or DMARDs who develop coccidioidomycosis. Questions still to be answered include the utility and schedule of routine serologic screening, the proper duration of antifungal therapy, and the role of antifungal prophylaxis, if any. Measurement of cell-mediated immunity against Coccidioides in these patients compared with nonimmunosuppressed controls may be informative in the future (15). In the meantime, we are encouraged to find that for many patients, with proper followup, these medications may be used despite a diagnosis of coccidioidomycosis.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Taroumian had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Taroumian, Lisse, Yanes, Vaz, Galgiani, Hoover.
Acquisition of data. Taroumian, Knowles, Lisse, Yanes, Ampel, Vaz, Hoover.
Analysis and interpretation of data. Taroumian, Knowles, Lisse, Yanes, Ampel, Vaz, Galgiani, Hoover.