Nosocomial Infections and Food/Waterborne Infections Symposium Sections
Epidemiology of nosocomial fungal infections: invasive aspergillosis and the environment

https://doi.org/10.1016/S0732-8893(99)00026-7Get rights and content

Abstract

The incidence rates of invasive aspergillosis have increased dramatically during the last two decades, and, despite all diagnostic and therapeutic efforts, outcome is often fatal. Therefore, preventive measures are of major importance in the control of invasive aspergillosis, and require full understanding of the epidemiology of this devastating disease. The environment has been suggested to play a crucial role in the epidemiology of invasive aspergillosis. Aspergillus spores are released in the air and may remain airborne for prolonged periods. As a result, spores are ubiquitously found in air and contaminate anything in contact with air. It has been hypothesized that the inhalation of airborne Aspergillus spores, either directly or through intermediate nasopharyngeal colonization, is a direct cause of pulmonary infection in immunocompromised patients. Recently, water has been suggested as an additional source of “airborne” Aspergillus spp. This review summarizes the current knowledge on the role of the environment in the epidemiology of invasive aspergillosis.

Introduction

With the advance of modern medicine, a shift in the patient population has taken place leaving us with substantially increased numbers of severely immunocompromised and neutropenic patients that are at high risk for acquiring invasive aspergillosis Bodey 1988, Denning 1998. Indeed, autopsy surveys have documented dramatic increases in the incidence rates of invasive aspergillosis during the last two decades: from less than 0.4% of all patients autopsied in the seventies up to 4% in the mid-nineties Yamazaki et al 1997, Groll et al 1996. In a recent population-based laboratory active surveillance the one-year cumulative incidence of invasive aspergillosis was estimated at 12.4 per million persons (Rees et al., 1998).

The clinical spectrum of invasive aspergillosis includes primary cutaneous aspergillosis, (rhino)sinusitis, tracheobronchitis, pulmonary aspergillosis, and disseminated aspergillosis, with cerebral aspergillosis as its most serious manifestation Hagensee et al 1994, Martinez and Ahdab–Barmada 1993, Drakos et al 1993, Kac et al 1995, Kemper et al 1993, Kramer et al 1991, Talbot et al 1991. Despite all diagnostic and therapeutic efforts, the attributable mortality of invasive aspergillosis still ranges from 50–100% in collected case series Denning 1996, Weinberger et al 1992, Pannuti et al 1991, Denning and Stevens 1990. An overall success rate of 34%, varying substantially between different host groups, has been documented with amphotericin B, the currently standard therapeutic agent for the treatment of invasive aspergillosis (Denning, 1996). Under certain circumstances, the new lipid formulations of amphotericin B may be appropriate as well (Hiemenz and Walsh, 1996). The lipid-associated amphotericins are generally regarded as equally effective, but less nephrotoxic than conventional amphotericin B Oppenheim et al 1995, Mills et al 1994, Meunier et al 1991, Stevens 1994, Ellis et al 1996.

The primary route of acquiring Aspergillus infection is probably by the inhalation of fungal spores. Spores of the most commonly involved pathogenic Aspergillus spp., i.e., Aspergillus fumigatus, Aspergillus flavus, and Aspergillus terreus, are relatively small, with sizes ranging from 2–5 micron (Bennett, 1995). When inhaled, spores will become deposited in both the upper and lower respiratory tract (Morrow, 1980). For most individuals deposited airborne spores will be cleared without affecting their health. However, immunocompromised patients are extremely susceptible to local invasion of respiratory tissues by deposited spores, resulting in invasive aspergillosis (Denning, 1998).

Different immunocompromised patient groups vary substantially in their risk of developing invasive aspergillosis (Denning, 1998). The most important documented host risk factors are profound and prolonged neutropenia (Gerson et al., 1984), neutrophil function deficits as in chronic granulomatous disease (Mouy et al., 1989), supraphysiological corticosteroid therapy Gonzalez–Crespo and Gomez–Reino 1995, Gustafson et al 1983, graft-versus-host disease McWhinney et al 1993, Sherertz et al 1987 and/or rejection in transplantation (Gustafson et al., 1983). In addition, several studies that obtained prospective nasopharyngeal cultures in immunocompromised patients have suggested that nasopharyngeal colonization is a major risk factor for the acquisition of invasive aspergillosis Martino et al 1989, Rhame et al 1984, Aisner et al 1979. However, data on nasopharyngeal colonization should be interpreted with caution, as it may be difficult to distinguish between nasopharyngeal colonization prior to invasive aspergillosis and colonization subsequent to invasive aspergillosis.

Besides host factors, many investigators have studied the role of the environment in the epidemiology and pathogenesis of invasive aspergillosis. The next section will discuss current knowledge on the relation between human invasive aspergillosis and the presence of Aspergillus spp. in the environment.

The primary ecological niche of A. fumigatus is believed to be decaying vegetation (Mullins et al., 1976). However, air plays a crucial role in the spread of Aspergillus spp. in the environment. Aspergillus spp. efficiently release large amounts of spores in the air from conidiophores that protrude from the mycelial mass. Due to their small size and roughened surface Aspergillus spores may remain airborne for prolonged periods. A settling velocity of 0.03 cm/s (i.e., approximately 1 m/h) has been reported for A. fumigatus spores (Gregory, 1973). As a result, spores are ubiquitously found in air. As spores gradually settle out, anything in contact with air will become contaminated with spores. Disturbance of settled spores will often result in mini-bursts of spores (Rhame, 1991). The process of settling out and becoming airborne again can repeat itself for prolonged periods of time, because Aspergillus spores are viable for months in dry locations. In the presence of water spores will germinate and mycelial growth will occur with subsequent sporulation.

Reported spore concentrations in outdoor air vary from study to study, with mean concentrations ranging from 0.2–15.0/cubic meter Solomon et al 1978, Jones and Cookson 1983, Rose and Hirsch 1979, Rath and Ansorg 1997. Seasonal variation in spore content has been documented in some studies. However, results have been conflicting. Highest spore concentrations have been detected either in summer (Jones and Cookson, 1983), or during fall and winter Mullins et al 1976, Noble and Clayton 1963. Other studies have not been able to detect any seasonal differences in spore counts Hospenthal et al 1998, Arnow et al 1991, nor could a correlation between Aspergillus spore concentrations in air and meteorological data, such as mean temperature, rainfall, air pressure, or humidity be detected Jones and Cookson 1983, Rath and Ansorg 1997. Very high concentrations of airborne A. fumigatus spores, up to 106/cubic meter, have been reported in special circumstances, such as near compost heaps and hay barns (Clark et al., 1983).

Several studies have shown large variability in Aspergillus spp. spore concentration from sample to sample Rose and Hirsch 1979, Rath and Ansorg 1997, Hospenthal et al 1998, Arnow et al 1991. This may, at least in part, be due to sampling error. The abundance of other fungal species in air precludes sampling of large air volumes, unless selective techniques for pathogenic Aspergillus spp. are used. Therefore, single sample measurements of Aspergillus spores are often not precise. Another source for the sample to sample variation in spore concentration may be the occasional release of spores in clouds, that results in significant fluctuations in spore counts (Rose and Hirsch, 1979). In a study comparing paired samples from hospital wards and outside air, no statistically significant difference in A. fumigatus spores was found in 88 paired samples, except for one pair showing a considerable excess of spores in one ward compared to the paired samples from other wards and outside air (Solomon et al., 1978). Another study, that performed continuous air sampling, detected only one burst of Aspergillus spores in the first 182 h of sampling (Rhame et al., 1984). As the frequency and nature of such bursts are insufficiently known, one may argue that sampling of small air volumes over short time periods results in unreliable estimates of the total exposure to Aspergillus spores in individual patients. Future research should therefore concentrate on developing valid measures of fungal exposure.

Several potential sources for Aspergillus spores in hospital air have been described, inadequate filtration of outside air by the air handling system being the most obvious one Sarubbi et al 1982, Rhame et al 1984. Dust with high concentrations of spores accumulates in air ducts and other places that are infrequently cleaned, and produces bursts of airborne Aspergillus spp. when disturbed (Rhame, 1991). Another well documented source for bursts of Aspergillus spores is vacuum cleaning (Anderson et al., 1996), even when HEPA (high-efficiency particulate air)-filtered exhausts are used on the vacuum cleaner (Rhame et al., 1984). Besides airborne spread, Aspergillus spores may enter the hospital with spices that are used in the preparation of food (Schwab et al., 1982), or with other organic material such as the soil of potted ornamental plants, flowers, or fresh fruit (Staib, 1984). Anecdotal reports have linked smoking of tobacco and/or marijuana with exposure to large quantities of Aspergillus spores and the development of invasive aspergillosis Hamadeh et al 1988, Marks et al 1996. Furthermore, growth of Aspergillus on air filters and within fireproofing material that had been used in hospital construction, with subsequent release of spores inside the hospital has been reported Aisner et al 1976, Arnow et al 1991.

Due to its clinical relevance, most studies on Aspergillus spp. in hospital air have dealt with A. fumigatus. Less is known about A. flavus and A. terreus. Air sampling studies have found lower spore concentrations for A. flavus and A. terreus than for A. fumigatus in both outside and indoor (hospital) air Solomon et al 1978, Noble and Clayton 1963. In the absence of more data it may well be assumed that the route of entry in the hospital is similar for all Aspergillus spp.

Whether clinical aspergillosis can arise from chronic, low level, endemic airborne Aspergillus spores or whether a large inoculum from a burst of Aspergillus spores is required is unknown, and needs further investigation.

Most cases of (nosocomial) invasive aspergillosis present with pneumonia Young et al 1970, Fridkin and Jarvis 1996, Sherertz et al 1987. Therefore, it has been hypothesized that the inhalation of airborne Aspergillus spores, either directly or through intermediate nasopharyngeal colonization, is a direct cause of pulmonary infection in immunocompromised patients.

Arguments in support of this hypothesis are provided by many studies (Rhame et al., 1984). Reduced incidence rates of invasive aspergillosis have been reported in hospitals after moving from older hospital buildings to new facilities with modernized air handling systems Rosen and Sternberg 1976, Rose 1972. Moreover, the installation of HEPA-filtered air units for the care of immunocompromised patients has been reported to result in both a reduction of Aspergillus spore concentrations in ambient air and a marked decrease in the incidence of invasive aspergillosis (Rhame et al., 1984). However, other coincidental changes in patient care may have contributed to the decrease in cases of aspergillosis as well.

Many investigators have described outbreaks of nosocomial invasive aspergillosis. Some of these case clusters have been associated with construction and/or renovation activities in and around hospitals Buffington et al 1994, Krasinski et al 1985, Sarubbi et al 1982, Streifel et al 1983, Arnow et al 1978, Lentino et al 1982, Weems et al 1987, others with malfunctioning or contamination of hospital ventilation or air filtration systems Mahoney et al 1979, Kyriakides et al 1976, Lentino et al 1982, Ruutu et al 1987. Some anecdotal reports have documented outbreaks of invasive aspergillosis associated with contaminated fire proofing material (Aisner et al., 1976), damp wood, potting soil and carpeting (Gerson et al., 1994). Although the reported outbreaks suggest that the concentration of Aspergillus spores in air plays an important role in the development of invasive aspergillosis, data should be interpreted carefully. With the relatively low frequency of invasive aspergillosis seen at many hospitals, even small changes in the number of cases may appear to be a cluster when in fact it is not. Besides, most studies have documented potential sources of increased spore concentration in air, such as construction activities, contamination of air handling systems retrospectively, that may have resulted in (observer) biased conclusions. Moreover, there are often no data available on the baseline air concentrations of spores to determine whether these case clusters were associated with increased exposure to airborne Aspergillus spp.

Only few studies have evaluated the relation between the airborne concentration of Aspergillus spp. and the risk of invasive aspergillosis directly. One group of investigators conducted a prospective study in which monthly environmental cultures for Aspergillus spp. and surveillance for nosocomial invasive aspergillosis were performed during a 6-year period (Arnow et al., 1991). An increase in the mean concentrations of A. fumigatus and A. flavus spores from <0.2 to >1 spore cubic meter of air was accompanied by a fourfold increase in the incidence of invasive aspergillosis from 0.3–1.2% in immunocompromised patients (p = 0.01). Yet, in a more recent study the occurrence of six cases of invasive aspergillosis could not be linked to changes in the recovery of airborne Aspergillus spp. during a one-year period with weekly air sampling of the ward rooms and corridors (Hospenthal et al., 1998). Several important methodological issues must be taken into account when studying the relation between airborne Aspergillus spp. and invasive aspergillosis. First, as mentioned before, sampling of relatively small volumes of air will hamper the reliable estimation of spore concentrations in air. Both sampling errors and the occasional release of Aspergillus spores in clouds may result in imprecise estimates. Second, as bursts of Aspergillus spores may occur, assessment of general or mean levels of spore concentration in air on a certain ward may not provide valid estimates of the exposure of an individual patient on that ward. Third, the detection of an association between exposure to airborne Aspergillus spp. and invasive aspergillosis may be dependent on the range of spore concentrations observed, i.e., clinically significant variability in exposure to Aspergillus spores is required. Fourth, the observed (lack of) association between airborne Aspergillus spp. and invasive aspergillosis may be confounded or modified by other potential risk factors for aspergillosis, such as pre-admission exposure to Aspergillus spores, the degree of immunosuppression, antibiotic use, and so on.

Many studies that have attempted to associate cases of invasive aspergillosis with the environmental recovery of Aspergillus spp. have used only speciation in their comparisons. Recently, however, several studies have applied molecular techniques to compare environmental and clinical isolates. Analysis of moderately repeated DNA sequences of A. fumigatus isolated from patients and the environment revealed that A. fumigatus isolates from two of six patients were identical to hospital environment strains (Girardin et al., 1994). Molecular typing of clinical and environmental isolates during an outbreak of invasive A. flavus infections yielded comparable results (Buffington et al., 1994). That is, restriction fragment length polymorphism analysis revealed similar patterns in isolates from one patient and the environmental source, whereas this pattern differed from those found in other cases of invasive aspergillosis. However, other researchers have failed to detect genetic relatedness between patient and environmental isolates Rath and Ansorg 1997, Leenders et al 1996. Sampling problems may again have accounted for these conflicting results. The genetic diversity among clinical and environmental isolates of A. fumigatus is extremely high, as was recently published (Debeaupuis et al., 1997). Even within groups of isolates with the same geographical origin no clustering of isolates was observed. The results of this study indicate that extensive air sampling, as well as typing of all recovered isolates may be necessary to link cases of invasive aspergillosis with environmental recovery of Aspergillus spp.

Studies on the relation between airborne Aspergillus spp. and invasive aspergillosis have primarily focused on in-hospital exposure to Aspergillus spores. However, because little is known on the incubation period of invasive aspergillosis, it may well be that patients are infected outside the hospital, either before or after discharge, and that invasive aspergillosis is, at least partly, a community acquired disease. In that case preventive measures should not be aimed at the hospital environment only.

Recently, hospital water was suggested as a possible route of transmission in invasive aspergillosis (Anaissie, 1998). Opportunistic fungal pathogens have been recovered from sinks and shower heads in several hospitals in the United States. A. terreus and A. niger were cultured from the showerheads as well as Fusarium species. Sampling before and after showering revealed a significant increase of spore counts in the air. It can be hypothesized that spores present in the showerhead are released during showering. Patients may inhale the spore containing aerosols and become infected. Alternatively, spores may be present on the walls and floors of the bath room being disturbed by air flow changes that occur during showering. The clinical significance of this finding as yet remains unclear. A. fumigatus, that is by far the most frequent causing pathogen of invasive aspergillosis has not been recovered from the water in any of the hospitals in the US (Anaissie, 1998). Furthermore, during a recent survey in our hematology ward we were not able to culture Aspergillus spp. from hospital water either (personal communication). Nevertheless, hospital water may be a source of Aspergillus spp. and the role of water in the transmission of invasive aspergillosis needs further investigation.

Section snippets

Conclusions

With the continuing increase in the number of severely immunocompromised patients, hospitals are faced with the growing problem of invasive aspergillosis and other opportunistic infections. Because treatment of aspergillosis is difficult and outcome is often fatal, preventive measures are of major importance in the control of invasive aspergillosis. Therefore, full understanding of the epidemiology of invasive aspergillosis is crucial in the development of effective preventive strategies.

Acknowledgements

We would like to thank Ms. Gerry Hermkens for her excellent support in retrieving hard-to-get references.

References (69)

  • P.M. Arnow et al.

    Pulmonary aspergillosis during hospital renovation

    Am Rev Respir Dis

    (1978)
  • P.M. Arnow et al.

    Endemic and epidemic aspergillosis associated with in-hospital replication of Aspergillus organisms

    J Infect Dis

    (1991)
  • Bennett JE (1995) Aspergillus species. In: Mandell, Douglas and Bennett’s principles and practice of infectious...
  • J. Buffington et al.

    Investigation of an epidemic of invasive aspergillosisUtility of molecular typing with the use of random amplified polymorphic DNA probes

    Pediatr Infect Dis J

    (1994)
  • Clark CS, Rylander R, Larsson L (1983) Levels of gram-negative bacteria. Aspergillus fumigatus, dust, and endotoxin at...
  • J. Debeaupuis et al.

    Genetic diversity among clinical and environmental isolates of

    Aspergillus fumigatus. Infect Immun

    (1997)
  • D.W. Denning

    Therapeutic outcome in invasive aspergillosis

    Clin Infect Dis

    (1996)
  • D.W. Denning

    Invasive aspergillosis

    Clin Infect Dis

    (1998)
  • D.W. Denning et al.

    Antifungal and surgical treatment of invasive aspergillosisA review of 2,121 published cases

    Rev Infect Dis

    (1990)
  • P.E. Drakos et al.

    Invasive fungal sinusitis in patients undergoing bone marrow transplantation

    Bone Marrow Transplant

    (1993)
  • Ellis M, Spence D, Meunier F, de Pauw D, Bogaerts M, van der Cam B, Doyen C, Marinus A, Collettee L, Sylvester R, EORTC...
  • S.K. Fridkin et al.

    Epidemiology of nosocomial fungal infections

    Clin Microbiol Rev

    (1996)
  • S.L. Gerson et al.

    Prolonged granulocytopeniaThe major risk factor for invasive pulmonary aspergillosis in patients with acute leukemia

    Ann Intern Med

    (1984)
  • S.L. Gerson et al.

    Aspergillosis due to carpet contamination

    Infect Control Hosp Epidemiol

    (1994)
  • H. Girardin et al.

    Molecular epidemiology of nosocomial invasive aspergillosis

    J Clin Microbiol

    (1994)
  • P.H. Gregory
  • T.L. Gustafson et al.

    Invasive aspergillosis in renal transplant recipientsCorrelation with corticosteroid therapy

    J Infect Dis

    (1983)
  • M.E. Hagensee et al.

    Brain abscess following marrow transplantationExperience at the Fred Hutchinson Cancer Research Center, 1984–1992

    Clin Infect Dis

    (1994)
  • J.W. Hiemenz et al.

    Lipid formulations of amphotericin BRecent progress and future directions

    Clin Infect Dis

    (1996)
  • D.R. Hospenthal et al.

    Concentration of airborne Aspergillus compared to the incidence of invasive aspergillosisLack of correlation

    Med Mycol

    (1998)
  • B.L. Jones et al.

    Natural atmospheric microbial conditions in a typical suburban region

    Appl Environ Microbiol

    (1983)
  • G. Kac et al.

    Aspergillus et aspergillosesÉtude rétrospective dans deux hôpitaux parisiens

    J Mycol Med

    (1995)
  • C.A. Kemper et al.

    Ulcerative and plaque-like tracheobronchitis due to infection with Aspergillus in patients with AIDS

    Clin Infect Dis

    (1993)
  • M.R. Kramer et al.

    Ulcerative tracheobronchitis after lung transplantation. A new form of invasive aspergillosis

    Am Rev Respir Dis

    (1991)
  • Cited by (119)

    • Microbiological indoor air quality of hospital buildings with different ventilation systems, cleaning frequencies and occupancy levels

      2022, Atmospheric Pollution Research
      Citation Excerpt :

      Similar results have been reported in previous study conducted in hospital building of Nigeria (Ekhaise and Ogboghodo 2011) and India (Sudharsanam et al., 2012). Mycotoxins produced by Aspergillus spp. are reported to cause nosocomial infections, carcinogenicity, aspergillosis, irritation to skin, allergic problem and mortality in severe condition (VandenBergh et al., 1999; Zain 2011, Kim et al., 2018; Wu et al., 2019). Aspergillus flavus and Aspergillus fumigatus may cause invasive aspergillosis and their long term exposure is life threating particularly for immune compromised patients (Pasqualotto 2009; Tang 2009).

    • A polymicrobial fungal outbreak in a regional burn center after Hurricane Sandy

      2018, American Journal of Infection Control
      Citation Excerpt :

      Weather may play a significant role in fungal outbreaks. Several studies have documented seasonal variation of fungal spores, with higher levels in the fall; however, these results are inconsistent, and indoor samples do not always correlate with outdoor samples.14-25 The seasonal variation in the prevalence of Aspergillus spores inside a hospital has been associated with rainfall and with internal relative humidity and temperature.26

    View all citing articles on Scopus
    View full text