Coccidioidomycosis is caused by the dimorphic fungi Coccidioides immitis and Coccidioides posadasii, which are endemic in the southwestern United States, with most cases (97%), reported from Arizona and California.1–3 The organism resides in arid soil and spreads as airborne spores when soil is disrupted. The incidence of coccidioidomycosis in California has increased over the past decade with the largest number of cases reported in 2017.4 Clinical manifestations range from a self-limited influenza-like illness to pneumonia and, rarely, dissemination beyond the lungs.5 Differences in disease severity may be related to variations in immunologic responses to the infection among individuals.6,7 The risk of dissemination is higher in infants, Filipinos, Hispanics, African Americans, diabetics, immunocompromised hosts, and pregnant women. Published literature on coccidioidomycosis in the pediatric population is limited, particularly regarding disseminated diseases.5,8–10 The objectives of this study are to review the clinical course, diagnostic studies, treatment and outcomes of children with extrapulmonary coccidioidomycosis at a tertiary children’s hospital in central California.
MATERIALS AND METHODS
Study Design and Patients
The study was conducted at Valley Children’s Hospital; a 358-bed tertiary center serving 12 counties in central California, a coccidioidomycosis-endemic area. We conducted a retrospective case review of patients followed at our facility with a diagnosis of coccidioidomycosis from January 1, 2007 to December 31, 2016. Cases were identified through electronic medical records using ICD-9 and ICD-10 codes. Inclusion criteria were (1) having a clinical illness compatible with coccidioidomycosis; (2) having positive coccidioidal precipitin or complement fixation (CF) serology, cultures being positive for Coccidioides immitis/posadasii from any source, or biopsy materials demonstrating coccidioidal spherules; (3) having clinical or radiographic evidence of disease extending beyond the lungs [ie, to the mediastinum, pericardium, lymph nodes, skin, bones, joints, and/or central nervous system (CNS)]; and (4) being ≤21 years of age at diagnosis. Of note, 25 patients including 7 with mediastinitis and 18 with disseminated disease have been previously reported.5,11 All coccidioidal serologies were performed at the University of California, Davis Coccidioidomycosis Serology Laboratory.12 Fungal culture identification was performed by the Fungus Testing Laboratory, Department of Pathology, University of Texas Health Science Center in San Antonio. The Valley Children’s Hospital Institutional Review Board approved the study.
Demographic, clinical, laboratory, radiographic, treatment and outcome data were collected by medical chart review. Clinical outcomes were defined as: resolved, if therapy was completed with no return of signs or symptoms; stable, remaining on therapy without complete resolution of sign or symptoms or on life-long therapy for CNS infection; relapsed, therapy was completed followed by a return of signs or symptoms and an increase in coccidioidal CF titers, with or without radiographic evidence, resulting in re-initiation of treatment; progressive, on therapy with worsening signs or symptoms; and fatal, death resulted from coccidioidomycosis. Mediastinitis was defined as the presence of mediastinal lymph node enlargement with radiographic evidence of necrosis and/or purulence. Patients were followed for 1 to 10 years depending on disease severity.
Demographic, clinical, laboratory, radiographic characteristics and outcome data were compared between groups using the Mann-Whitney U test for continuous variables and χ2 or Fisher exact test for categorical variables. Descriptive results were expressed as medians and interquartile range (IQR) for continuous variables or as frequency and percentage for categorical variables.
Binary logistic regression and linear regression models were used to explore predictors for adverse clinical outcomes. More specifically, binary logistic regression models were used to predict the odds of having unresolved disease, >1 organ involvement, and >1 drug therapy. Linear regression models were used to estimate the length of hospitalization, inlet duration (interval between onset of symptoms and diagnosis in days), illness duration (interval between diagnosis and resolution of illness in days) and number of drug days [defined as the number of drug(s) × days]. Since these continuous variables were not normally distributed, they were log-transformed. For all models, we implemented the step-wise model selection technique to select variables that significantly predict the clinical outcomes mentioned above.13
Statistical analyses were performed using SAS version 9.4 (Cary, NC). Statistical significance was defined as P < 0.05.
Seventy-eight cases were identified, including 63 (81%) with disseminated disease and 15 (19%) with mediastinitis. Comparison of baseline demographics, clinical characteristics, diagnostic measures, and outcomes are summarized in Table 1. Almost half of the cases were diagnosed between 2011 and 2012 (Figure, Supplemental Digital Content 1A, http://links.lww.com/INF/D627). Overall, the median age at diagnosis was 9.7 years (IQR, 4.5–14.8 years). The numbers of patients within each age group were 16 (20%) in 0–2 years of age, 25 (32%) in 3–11 years of age and 37 (47%) in 12–21 years of age. The majority of patients was males (55%), Hispanic (65%), without comorbid conditions (85%) and 97% resided in the San Joaquin Valley in central California (Figure, Supplemental Digital Content 1B, http://links.lww.com/INF/D627).
Common presenting symptoms were fever (60%), cough (44%), fatigue (30%) and headache (23%). Erythema nodosum was seen only in 3 (4%) patients. The median duration of symptoms before diagnosis was 34 days (IQR, 18–75 days). Organ involvement included bones and joints (n = 26, 33%), mediastinum (n = 15, 19%), CNS (n = 15, 19%), cervical lymph nodes (n = 12, 15%), larynx (n = 5, 6%) and skin (n = 4, 5%). Bones involved included the skull (n = 7), vertebrae (n = 5), ribs (n = 4), pelvis (n = 3), phalanges (n = 3), tarsal bones (n = 3), mastoid (n = 2), calcaneus (n = 2), mandible (n = 1), clavicle (n = 1), radius/ulna (n = 1) and tibia/fibula (n = 1). Two-thirds of patients with underlying comorbid conditions presented with bone and joint disease.
Bone disease, joint disease and meningitis were seen more often in older patients (mean age, 10.1, 13.0 and 11.6 years, respectively) while laryngeal and mediastinal disease were seen more often in younger patients (mean age, 5.5 and 6.5 years, respectively) (Fig. 1). Concomitant pulmonary involvement occurred in 53 (68%) cases.
Coccidioides precipitin antibodies were positive in 61 (78%) cases and CF antibody titers in 70 (90%). Median initial Coccidioides CF titers upon diagnosis were 1:32 (IQR, 1:16–1:128) and median maximum titers were 1:64 (IQR, 1:32–1:256). Overall, 61 (78%) had maximum CF titers ≥ 1:32 with the highest titer of 1:16,384. Forty-seven (60%) patients had positive cultures for Coccidioides immitis/posadasii from clinical specimens.
Treatment and Outcomes
The median treatment duration was 633 days (IQR, 385–861 days). Fluconazole was the initial therapy in 47 (60%) patients while 28 (36%) patients required intravenous liposomal amphotericin B as initial treatment. The remainder was started on voriconazole (n = 1, 1%), itraconazole (n = 1, 1%) and combined voriconazole and caspofungin (n = 1, 1%). Of 47 who were initially started on fluconazole therapy, 20 (42 %) failed this treatment. Fifty-nine (76%) required treatment with 2 or more antifungal drugs. Eight (10%) failed to respond to liposomal amphotericin B and required salvage combination therapy with voriconazole and caspofungin for progressive disease [mediastinitis (n = 4), vertebral osteomyelitis (n = 2), cutaneous disease (n = 1) and laryngeal disease (n = 1)]. One patient with coccidioidal meningitis received intra-thecal amphotericin for 16 months due to disease progression while on azole therapy and was subsequently maintained on long-term twice-weekly IV liposomal amphotericin B in combination with oral posaconazole as salvage therapy. Five patients (6%) received interferon γ at a dose of 50 µg/m2 subcutaneously 3 times per week as an adjunctive therapy due to worsening disease while on antifungal treatment and 6 (7%) received adjunctive corticosteroids. Among 5 patients who received interferon γ, 3 achieved disease stabilization but 2 continued with disease progression and 1 died.
The majority of patients required hospitalization (84%) with a median duration of 30 days (IQR, 51–129 days). At the time of study conclusion, disease in 48 (62%) patients was considered resolved, 17 (22%) were stable, 2 (2.6%) were progressive and 7 (9%) had relapsed with a median interval to relapse of 4.9 months (IQR, 2.9–24 months). Five of these relapsed patients had osteomyelitis and/or arthritis and 2 had cutaneous disease. There were 2 fatal cases, 1 with meningitis with disease progression due to medication noncompliance and 1 with delayed diagnosis of mediastinal and joint disease (5 months from the onset of illness). Two cases were lost to follow-up near the end of therapy.
Predictors of Severe Illness
Patients who were 10 years or older at the time of diagnosis were more likely to require more days of drug therapy (611 vs. 349 days, P = 0.02), had higher ESRs (69.5 vs. 53.0 mm/h, P = 0.04), and were more likely to have maximum Coccidioides CF titers ≥ 1:32 (89% vs. 70%, P = 0.05) (Table 2). There was also a trend towards having >1 organ involvement (47% vs. 25%, P = 0.06) and experiencing relapsed/progressive/fatal disease (21% vs. 5%, P = 0.06) among those who were 10 years or older compared with those <10 years (Fig. 2).
Non-Hispanic patients were much older (13.8 vs. 5.6 years, P < 0.01), were more likely to have maximum Coccidioides CF titers ≥1:32 (89% vs. 72%, P = 0.04), to require >1 drug therapy (85% vs. 70%, P = 0.04) and were more likely to experience relapsed/progressive/fatal disease (22% vs.8%, P = 0.02) when compared with Hispanic patients (Table 2). African American and Asian/Pacific Islander patients were more likely to develop relapsed/progressive/fatal disease when compared with Caucasian and Hispanic patients (Table, Supplemental Digital Content, http://links.lww.com/INF/D628).
Coccidioides CF Antibody Titers
When analyzing Coccidioides CF antibody titers, those with >1 organ involvement had higher median maximum Coccidioides CF titers than those with 1 organ involvement (1:256 vs. 1:64, P < 0.01). Those who required therapy with >1 drug also had higher median maximum Coccidioides CF titers than those requiring therapy with only 1 drug (1:256 vs. 1:32, P < 0.01).
In multivariable analyses, age remained an independent risk factor for relapsed/progressive/fatal disease and more days of drug therapy. With each year of increase in age, there is a 35% increase in the odds (95% CI: 1.09–1.66, P < 0.01) of having relapsed/progressive/fatal disease, and a 32% increase in the number of days of drug therapy (95% CI: 5.2%–64.9%, P = 0.05). Those having >1 organ involvement had 29 times higher odds of longer duration of illness (95% CI: 2.5%–63.5%, P = 0.03) compared with those with 1 organ involvement. Those with initial Coccidioides CF titer ≥ 1:32 also had 31 times higher odds of having longer duration of illness compared with those with titers <1:32 (95% CI: 5.2–64.9, P = 0.02).
Most patients with coccidioidomycosis have asymptomatic or mild, self-limited pulmonary infections.14–16 We demonstrate that pediatric coccidioidomycosis can be severe and can disseminate beyond the lungs. The majority of cases were diagnosed between 2011 and 2012 when there was overall increase in incidence of coccidioidomycosis.1,3,17 In this study, we found a significant delay in diagnosis in our patients. The long interval duration prior to diagnosis is of concern since delayed diagnosis may contribute to increased disease severity and dissemination.
Similar to the studies by Dimitrova et al8 and Sondermeyer et al,1 the majority of our patients were previously healthy and without comorbid conditions.16 Recent data suggest that defects in the interleukin 12/interferon gamma and STAT3 axes may predispose to disseminated coccidioidomycosis.7 Genetic screening was not performed on our patients to detect these defects.
About one-third of our patient had >1 site of involvement with the skeletal system being the most common site followed by CNS; this finding is similar to previous reports.8,10 Not previously reported, CNS and skeletal disease were noted more frequently in older children whereas laryngeal and mediastinal diseases were noted more frequently in younger children. This highlights the need for careful and prompt evaluation for extrapulmonary disease tailored specifically to each age group.
About two-thirds of our patients had Coccidioides CF titers of 1:32 or higher at the time of diagnosis. CF titers of 1:32 or higher are associated with disseminated disease and warrant detailed evaluation based on clinical presentation.8 In our cohort, patients with >1 organ involvement who required >1 drug therapy were more likely to have maximum CF titers of 1:32 or higher (Table 3), consistent with previous studies.8,18 Of note, 10% of our patients had negative CF titers at the time of diagnosis, indicating that negative coccidioidal CF titers cannot completely exclude disseminated disease. Among those with negative CF titers, 5 were tested about 2 weeks and the remainder 1–2 months after disease onset.
Infectious Disease Society of America guidelines recommend the use of fluconazole as the initial treatment of coccidioidomycosis, which was used in 62% of our patients.6 However, due to disease severity, some patients required liposomal amphotericin B therapy; also consistent with current guideline. Combination antifungal therapy is not routinely used in patients with coccidioidomycosis due to limited clinical trial data as well as theoretical concerns about possible drug-to-drug antagonism. However, due to poor response to antifungal monotherapy and continued disease progression, some patients received salvage combination therapy with voriconazole and caspofungin, according to the treating physician’s discretion.19 A good response was noted in all of these patients with the exception of one with advanced CNS disease. Five patients received adjunctive therapy of interferon γ due to disease progression while on antifungal therapy. The use of interferon γ has been previously reported in refractory disseminated coccidioidomycosis but requires further study and must be considered investigational at this time.20–22
Extrapulmonary coccidioidomycosis requires prolonged treatment and follow-up due to the risk of disease relapse. Overall, clinical outcome in our cohort was good with the majority of patients experiencing either complete resolution or stable disease during the study period.1,5,10
When compared with those with disseminated coccidioidomycosis, patients with mediastinal disease were younger, were diagnosed more rapidly, had higher maximum Coccidioides CF titers and required longer hospitalizations, as previously noted in a case series.5 In that series, patients with mediastinitis were younger (median age 3 years) and required prolonged hospitalization (median duration 130 days) when compared with non-mediastinitis patients.5 We also found that older patients and those with higher CF titers were at higher risk for >1 organ involvement, relapsed or progressive disease and more often required >1 drug treatment7,23
There are several limitations to this study. This is a retrospective study, which relied on the accuracy and completion of medical records. This study is representative of only 1 hospital in central California and its findings may not necessarily be generalizable to others. Immunologic investigations were not conducted in most cases and these may have identified underlying immune defects that could explain the risk for dissemination.
In summary, while coccidioidomycosis is usually a self-limited illness, it can cause serious and life-threatening disease in children resulting in prolonged hospitalizations, multi-drug therapy and death. Non-Hispanic patients, patients >10 years of age and those with higher Coccidioides CF titers are at increased risk for extrapulmonary disease, suggesting a need for earlier recognition and intervention. Healthcare providers should consider testing for coccidioidomycosis in children who reside in or have visited endemic regions and present with symptoms compatible with coccidioidomycosis to avoid delays in diagnosis and proper treatment. Prospective studies are needed to identify predictors for adverse outcomes in pediatric disseminated coccidioidomycosis.
1. Sondermeyer GL, Lee LA, Gilliss D, et al. Epidemiology of pediatric coccidioidomycosis
in California, 2000-2012. Pediatr Infect Dis J. 2016;35:166–171.
2. Stockamp NW, Thompson GR 3rd. Coccidioidomycosis
. Infect Dis Clin North Am. 2016;30:229–246.
3. Cooksey GS, Nguyen A, Knutson K, et al. Notes from the field: increase in coccidioidomycosis
- California, 2016. MMWR Morb Mortal Wkly Rep. 2017;66:833–834.
4. Smith KL. In; Epidemiologcal summary of coccidioidomycosis
in California. 2017.
5. McCarty JM, Demetral LC, Dabrowski L, et al. Pediatric coccidioidomycosis
in central California: a retrospective case series. Clin Infect Dis. 2013;56:1579–1585.
6. Galgiani JN, Ampel NM, Blair JE, et al. 2016 Infectious Diseases Society of America (IDSA) clinical practice guideline for the treatment of coccidioidomycosis
. Clin Infect Dis. 2016;63:e112–e146.
7. Odio CD, Marciano BE, Galgiani JN, et al. Risk factors for disseminated coccidioidomycosis
, United States. Emerg Infect Dis. 2017;23:308–311.
8. Dimitrova D, Ross L. Coccidioidomycosis
: experience from a children’s hospital in an area of endemicity. J Pediatric
Infect Dis Soc. 2016;5:89–92.
9. Lee JM, Graciano AL, Dabrowski L, et al. Coccidioidomycosis
in infants: a retrospective case series. Pediatr Pulmonol. 2016;51:858–862.
10. Kafka JA, Catanzaro A. Disseminated coccidioidomycosis
in children. J Pediatr. 1981;98:355–361.
11. Lee LA, Sondermeyer Cooksey GL, Kim JJ, et al. Pediatric coccidioidomycosis
: case series from a California pediatric
infectious diseases clinic. Pediatr Infect Dis J. 2019;38:115–121.
12. Pappagianis D, Zimmer BL. Serology of coccidioidomycosis
. Clin Microbiol Rev. 1990;3:247–268.
13. Bursac Z, Gauss CH, Williams DK, et al. Purposeful selection of variables in logistic regression. Source Code Biol Med. 2008;3:17.
14. Chiller TM, Galgiani JN, Stevens DA. Coccidioidomycosis
. Infect Dis Clin North Am. 2003;17:41,viii–57, viii.
15. Crum NF, Lederman ER, Stafford CM, et al. Coccidioidomycosis
: a descriptive survey of a reemerging disease. Clinical characteristics and current controversies. Medicine (Baltimore). 2004;83:149–175.
16. Fisher BT, Chiller TM, Prasad PA, et al. Hospitalizations for coccidioidomycosis
at forty-one children’s hospitals in the United States. Pediatr Infect Dis J. 2010;29:243–247.
17. Sondermeyer Cooksey GL, Jain S, Vugia DJ. Epidemiology of coccidioidomycosis
among children in California, 2000-2016. Med Mycol. 2019; 57(supplement_1):S64–S66.
18. McHardy IH, Dinh BN, Waldman S, et al. Coccidioidomycosis
complement fixation titer trends in the age of antifungals. J Clin Microbiol. 2018;56:e0131818.
19. Levy ER, McCarty JM, Shane AL, et al. Treatment of pediatric
with combination voriconazole and caspofungin: a retrospective case series. Clin Infect Dis. 2013;56:1573–1578.
20. Duplessis CA, Tilley D, Bavaro M, et al. Two cases illustrating successful adjunctive interferon-γ immunotherapy in refractory disseminated coccidioidomycosis
. J Infect. 2011;63:223–228.
21. Kuberski TT, Servi RJ, Rubin PJ. Successful treatment of a critically ill patient with disseminated coccidioidomycosis
, using adjunctive interferon-gamma. Clin Infect Dis. 2004;38:910–912.
22. Vinh DC, Masannat F, Dzioba RB, et al. Refractory disseminated coccidioidomycosis
and mycobacteriosis in interferon-gamma receptor 1 deficiency. Clin Infect Dis. 2009;49:e62–e65.
23. Noble JA, Nelson RG, Fufaa GD, et al. Effect of geography on the analysis of coccidioidomycosis
-associated deaths, United States. Emerg Infect Dis. 2016;22:1821–1823.
coccidioidomycosis; extrapulmonary; meningitis; valley fever; pediatric; fungal infection
Supplemental Digital Content
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.