TEXT

Top Abstract Text References

Aspergillus species are ubiquitous fungi (13). Invasive aspergillosis (IA), which is often fatal, is the second most common opportunistic mycosis, after candidiasis, affecting humans (13). Of the more than 200 Aspergillus species known, Aspergillus fumigatus and Aspergillus flavus are frequently noted to cause infections in immunocompromised patients. In certain high-risk populations, such as bone marrow transplant recipients (3, 12), the attributable mortality rate approaches 85 to 95% (3, 10). Outbreaks of Aspergillus infection have been reported in association with hospital construction (1, 6, 7, 14), prompting experts to recommend strategies for minimizing dust production in areas where high-risk patients may be exposed (1, 11, 12, 14).

Although the consequences of construction in patient care areas are well described (1, 6, 7, 11, 14), the aftermath of construction that occurs in non-patient care areas is rarely discussed. We investigated a cluster of Aspergillus niger cultures that were obtained during construction in our clinical microbiology laboratory and here suggest ways to minimize specimen contamination during periods of construction.

During the period of July to September 1996, A. niger was isolated from 17 clinical specimens ("clinical specimens" refers to all tissue, blood, sputum, cerebrospinal fluid (CSF), and urine specimens presented for routine bacterial culture) (Table 1) submitted for 15 inpatients at our hospital (6.4/10,000 patient days), compared to A. niger isolation from 2 inpatients in the preceding 12 months (0.17/10,000 patient days) (rate ratio [RR], 7.50 [RR is the ratio of two instantaneous incidence rates, i.e., the incidence rate of cultures positive for A. niger during the pseudoepidemic period compared to the incidence rate of cultures positive for A. niger during the pre-pseudoepidemic period.]; 95% confidence interval [CI₉₅], 1.74 to 67.59 [CI₉₅ indicates that there is a 95% probability that the true value lies between the confidence limits. A CI₉₅ that does not include 1.0 is similar to finding that the association is statistically significant.]; P = 0.002). During the epidemic period, the overall rate of isolation of A. niger from clinical specimens was 7.9 per 100,000 clinical specimens, compared to 0.73 per 100,000 specimens in the preceding 9 months (RR, 10.82; CI₉₅, 1.28 to 887.92; P = 0.005) (Fig. 1). Initially, we heard of this potential outbreak as A. niger was being isolated from patients in the intensive care unit for whom the pretest probability of infection was high; these included three neutropenic patients, two bone marrow transplant patients, and three solid organ transplant patients. We discovered, however, that A. niger was also being isolated from patients with no recognizable risk factors for IA, including one patient whose urine specimen was collected during a routine obstetrical evaluation. We initiated an investigation in order to determine the etiology of this increased rate of isolation of A. niger from clinical specimens.

View larger version (24K): [in this window] [in a new window]   FIG. 1.   Isolation of A. niger from inpatient clinical specimens.

Using standardized definitions of nosocomial pneumonia, invasive pulmonary aspergillosis, disseminated extrapulmonary aspergillosis, and local extrapulmonary aspergillosis (4, 13), we reviewed all cases to determine if the patients were infected with A. niger and whether the infections were nosocomial. No inpatients met our criteria for a definite case of nosocomial IA, either pulmonary or extrapulmonary. One inpatient was initially classified as possibly infected because he had a positive culture and a compatible clinical syndrome. We determined that 14 inpatients were unlikely to be infected because they (i) had no clinical manifestation of disease, (ii) went untreated without adverse sequelae, or (iii) had a dedicated fungal culture that was negative.

We discovered that nine patients had specimens that were split at the point of collection, with one part going directly to the mycology laboratory for fungal culture while the remainder was sent to the main microbiology laboratory for bacterial culture. When specimens were split as such, A. niger was isolated from the portion processed in the main microbiology laboratory but not from the portion plated explicitly for fungi. We concluded that either patient colonization or contamination occurring during collection and transport of specimens was an unlikely explanation for the increased number of positive cultures being reported.

We conducted a series of experiments in the microbiology laboratory to investigate possible sources of contamination with A. niger. We noted that many of the positive cultures (12 of 26, 46%) were from specimens plated on buffered charcoal yeast extract (BCYE) medium, the medium that is routinely used in our laboratory for isolation of Legionella spp. Unlike standard blood plates that are held for 48 to 72 h, BCYE plates are held for 7 days, allowing additional time for growth of A. niger. We suspected that growth was detected on BCYE agar because these plates were held longer than the standard blood plates, but we also considered that A. niger may grow preferentially on BCYE medium due to nutrients in the BCYE medium (5). After performing sterility checks on our BCYE and blood agar plates, we used both in a subsequent series of experiments. (In retrospect, we acknowledge that conventional mold agars would have been more appropriate and less expensive in looking for environmental sources of A. niger.)

Experiment 1. BCYE settle plates were left open for 48 h throughout the microbiology laboratory, including the specimen evaluation benches. Similarly, BCYE plates were left open for 24 h in the incubators, including the one used for respiratory specimens. Although specimens are initially processed in the central microbiology laboratory biological safety cabinet (BSC), we wanted to analyze all areas within the laboratory where plates are screened, assessed, or incubated. These settle plates showed no growth of mold after one week of incubation.

Experiment 2. Sterile water, instead of sputum, was plated onto six BCYE agar plates by the standard laboratory protocol for plating of Legionella cultures. These plates were incubated and checked daily to simulate what is normally done with respiratory cultures that are submitted for Legionella detection. These showed no growth after 1 week of incubation.

Experiment 3. To evaluate the integrity of the central microbiology laboratory BSC, six settle plates were placed in the back of the BSC while the cultures in experiment 2 were being done. Three BCYE plates and three blood plates were left open for 5 min to approximate normal processing time. The plates were incubated and checked daily for growth. On the 4th day, a black mold, which was later identified as A. niger, was present on one of three BCYE plates.

Experiment 4. Next, we inspected the central microbiology laboratory BSC, which is the standard receiving point for all specimens entering the bacteriology section of the microbiology laboratory. We noted a fine black dust on the interior ceiling of the BSC, on the vortex machine that was inside the BSC, and on the BSC floor beneath the vortex machine. Laboratory personnel denied using a Bunsen burner or other soot-producing device inside the BSC that could explain the black dust. Using a sterile swab, we took specimens from all three locations, plated them directly on Sabouraud-dextrose agar slants, and incubated them at 34 to 37°C in the mycology laboratory's incubator. The mycology laboratory is physically separated from the main microbiology laboratory. After 1 week, A. niger grew from two of three specimens.

Experiment 5. While inspecting the area around the BSC, we found that the medium plates stored on the laboratory bench immediately adjacent to the BSC were dusty. BCYE plates were not kept in the area adjacent to the BSC; therefore, we did not suspect that the medium was contaminated prior to specimen processing. The ceiling tiles above the medium plates were dusty as well, and a culture of one of the ceiling tiles grew A. niger.