Comparison testing of current (PPD-S1) and proposed (PPD-S2) reference tuberculin standards

Since 1951, the tuberculin PPD-S1 has been used to standardize commercial PPD reagents and perform special tuberculin surveys. PPD-S1 is now in short supply and a new standard (PPD-S2) has been manufactured. To determine if PPD-S2 is equivalent and can replace PPD-S1, we conducted a double-blind clinical trial. Between May 14 and October 28, 1997, 69 subjects with a history of culture-proven tuberculosis (TB patients) and 1,189 subjects with a very low risk for TB infection were enrolled, received four skin tests (with PPD-S1, PPD-S2, and one each of the commercially available PPDs), and had reactions measured by two trained observers. Among the TB patients, we found statistically indistinguishable immunogenicity (mean reaction size +/- standard deviation): 15.6 +/- 6.6 mm for PPD-S1 and 14.8 +/- 5.6 mm for PPD-S2. Among low-risk subjects, the tests had equally high specificities (PPD-S1, 98.7% and PPD-S2, 98. 5%), using a 10-mm cutoff. The number of discordant (negative versus positive) interpretations for PPD-S2, assuming that low-risk subjects who had a >/= 10 mm reaction to PPD-S1 were truly infected, was low (0.5%) and indistinguishable from the rate of discordant interpretations of the same test when read by two different observers (0.8%). The study results indicate that PPD-S2 is qualified to be used as the new U.S. reference standard for PPD tuberculin.

Detection of coccidioidal antibodies by 33-kDa spherule antigen, Coccidioides EIA, and standard serologic tests in sera from patients evaluated for coccidioidomycosis

During a 9-month study of patients being evaluated for coccidioidomycosis, 1 or more serum samples were obtained from 138 patients with an illness suggestive of recent infection. In this group, standard immunodiffusion tests of unconcentrated sera were positive for 25; 49 additional patients had at least 1 reactive test result by newer enzyme-linked serologic tests. At least 11 of these 49 patients had coccidioidomycosis as determined by culture or subsequent standard serologic tests. Patients with coccidioidomycosis identified only by newer tests had fewer or milder clinical abnormalities than did patients in whom the disease was detected by standard tests. For 31 other patients with illness of a chronic or undetermined duration, newer tests detected only 10 more than the 18 identified by standard tests, suggesting that later in the course of illness, standard testing gains in sensitivity for coccidioidal infection.

New serologic tests for early detection of coccidioidomycosis

Arizona college students suspected of having recently acquired coccidioidomycosis were tested for anticoccidioidal antibodies and circulating fungal antigens using conventional antibody detection methods and new ELISA procedures. Of 233 patients with compatible symptoms, 26 had anticoccidioidal antibodies detected by conventional tests. ELISA detected antibodies in sera from 20 of these patients and also from another 25 patients. Patients with antibodies detected by either conventional or ELISA procedures were significantly more likely to have abnormal chest radiographs, elevated erythrocyte sedimentation rates, or absent upper respiratory symptoms than were other patients. Circulating antigen was found in sera from 35 patients, 33 of whom had no detectable anticoccidioidal antibodies at that time. Detectable antigen was noted frequently in sera obtained within the first month after the onset of symptoms and was infrequently detected later when more patients exhibited antibodies. These results indicate the feasibility of developing ELISA procedures using spherule-derived antigens for earlier detection of coccidioidal infections.

Symptoms and routine laboratory abnormalities associated with coccidioidomycosis

To assess the relationships of various symptoms and other early findings to the diagnosis of primary coccidioidomycosis, we devised a 40-question survey that was completed by 556 college students seeking medical care for illness possibly due to Coccidioides immitis. The results of routine laboratory studies on these patients were also compiled. Of 269 who had coccidioidal antibody determinations and other diagnostic tests, coccidioidomycosis was diagnosed in 36 (13%). By logistic regression procedures, an elevated erythrocyte sedimentation rate, male gender, "red lumps on shins," recent arrival to an endemic area, acuteness of symptoms, and decreased total peripheral blood lymphocyte counts were independent factors positively associated with infection (P less than .05). Relative risk analysis indicated that 60% of patients with four or more of these factors were found to have coccidioidomycosis. Other significantly but not independently associated factors were an increased total leukocyte count, chest pain with breathing, fever, an absence of hoarseness, and an abnormal chest roentgenogram.

Coccidioidomycosis at a university health service

At a single university health service within an endemic area, 172 cases of coccidioidomycosis were retrospectively identified by fungal cultures, serologic studies, or intercurrent skin test conversions. The average annual incidence of symptomatic infection was 0.43% for susceptible students. Peak case rates occurred in November and June (11 cases/10,000 clinic visits) and were 1.5 to 6 times greater than in other months. Although mononucleosis was diagnosed approximately twice as frequently as coccidioidomycosis, coccidioidal infections required 74% more visits per patient and over 3 times longer clinical supervision than did mononucleosis. Although only 2 patients disseminated, diagnosis and management of patients with coccidioidomycosis constituted more than 2% of our clinic's visits, for an estimated annual cost of more than $34,000. Because our patient population is young and otherwise healthy, our estimates of the impact of primary coccidioidal infection may underestimate that on civilian public health as a whole.