Monitoring of airborne fungus and yeast species in a hospital unit

Abstracts

OBJECTIVE: To monitor and characterize airborne filamentous fungi and yeasts from abiotic and biotic sources within a hospital unit. METHODS: Collections were carried out on a monthly basis, at two different time periods, from the adult and pediatric intensive care units and surgical center of a hospital in Araraquara, Southeastern Brazil. Collection of airborne fungi was carried out using a simple-stage Andersen sample. The presence of yeasts was investigated in samples taken from the hands and oropharynx of staff members as well as from the surface of beds and doorknobs inside the critical areas. RESULTS: Thirty-two genera of airborne fungi and were recovered from the surgical center and 31 from the intensive care units. Genera most frequently isolated were Cladophialophora spp., Fusarium spp., Penicillium spp., Chrysosporium spp. e Aspergillus spp. During the study period, a new unit was built in the hospital, which coincided with an increase in Cladophialophora spp., Aspergillus spp., and Fusarium spp. colony counts. Yeasts were found in 39.4% of samples obtained from healthcare staff (16.7% from interdigital spaces, 12.1% from nailbeds, and 10.6% from oropharynx) and in 44% of furniture samples, with a predominance of the Candida genus ((C. albicans, C. guilliermondii, C. parapsilosis e C. lusitaniae), followed by Trichosporon spp. CONCLUSIONS: We found a relatively high number of airborne fungi (potentially pathological) in special areas and expressive levels of yeasts in both biotic and abiotic samples. Microbiological and environmental monitoring should be conducted, especially in special areas which include immunocompromised patients, who are more susceptible to the exposure to environmental and staff-derived pathogens.

Fungi; Yeasts; Air conditioning; Cross infection; Aerosols; Infection control

OBJETIVO: Monitorar e caracterizar fungos anemófilos e leveduras de fontes bióticas e abióticas de uma unidade hospitalar. MÉTODOS: As coletas foram realizadas mensalmente e em dois períodos, do centro cirúrgico e unidades de terapia intensiva adulto e neonatal em hospital de Araraquara, Estado de São Paulo. Para coleta de fungos anemófilos foi utilizado amostrador tipo Andersen de simples estágio. A pesquisa de leveduras foi feita das mãos e de orofaringe de profissionais de saúde, bem como de superfícies de leitos e de maçanetas das áreas críticas. RESULTADOS: Foram recuperados do centro cirúrgico 32 gêneros de fungos anemófilos e 31 das unidades de terapia intensiva. Os gêneros mais freqüentemente isolados foram Cladophialophora spp., Fusarium spp., Penicillium spp., Chrysosporium spp. e Aspergillus spp. Durante o período de estudo, houve reforma e implantação de uma unidade dentro do hospital, que coincidiu com o aumento na contagem de colônias de Cladophialophora spp., Aspergillus spp. e Fusarium spp. Leveduras foram encontradas em 39,4% dos profissionais de saúde (16,7% das amostras dos espaços interdigitais, 12,1% do leito subungueal e 10,6% da orofaringe) e, em 44% das amostras do mobiliário, com predomínio do gênero Candida (C. albicans, C. guilliermondii, C. parapsilosis e C. lusitaniae) seguido por Trichosporon spp. CONCLUSÕES: Observou-se número relativamente elevado de fungos anemófilos (potencialmente patogênicos) em áreas especiais e níveis expressivos de leveduras em fontes bióticas e abióticas. O monitoramento microbiológico ambiental deve ser realizado, principalmente em salas especiais com pacientes imunocomprometidos, sujeitos à exposição de patógenos do meio ambiente, assim como, advindos de profissionais de saúde.

Fungos; Leveduras; Ar condicionado; Infecção hospitalar; Aerossóis; Controle de infecções

ORIGINAL ARTICLES

Monitoring of airborne fungus and yeast species in a hospital unit

José Nelson Martins-Diniz; Rosangela Aparecida Moraes da Silva; Elaine Toscano Miranda; Maria José Soares Mendes-Giannini

Departamento de Análises Clínicas. Faculdade de Ciências Farmacêuticas. Universidade Estadual Paulista. Araraquara, SP, Brasil

^{Correspondence}

ABSTRACT

OBJECTIVE: To monitor and characterize airborne filamentous fungi and yeasts from abiotic and biotic sources within a hospital unit.

METHODS: Collections were carried out on a monthly basis, at two different time periods, from the adult and pediatric intensive care units and surgical center of a hospital in Araraquara, Southeastern Brazil. Collection of airborne fungi was carried out using a simple-stage Andersen sample. The presence of yeasts was investigated in samples taken from the hands and oropharynx of staff members as well as from the surface of beds and doorknobs inside the critical areas.

RESULTS: Thirty-two genera of airborne fungi and were recovered from the surgical center and 31 from the intensive care units. Genera most frequently isolated were Cladophialophora spp., Fusarium spp., Penicillium spp., Chrysosporium spp. and Aspergillus spp. During the study period, a new unit was built in the hospital, which coincided with an increase in Cladophialophora spp., Aspergillus spp., and Fusarium spp. colony counts. Yeasts were found in 39.4% of samples obtained from healthcare staff (16.7% from interdigital spaces, 12.1% from nailbeds, and 10.6% from oropharynx) and in 44% of furniture samples, with a predominance of the Candida genus (C. albicans, C. guilliermondii, C. parapsilosis e C. lusitaniae), followed by Trichosporon spp.

CONCLUSIONS: We found a relatively high number of airborne fungi (potentially pathological) in special areas and expressive levels of yeasts in both biotic and abiotic samples. Microbiological and environmental monitoring should be conducted, especially in special areas which include immunocompromised patients, who are more susceptible to the exposure to environmental and staff-derived pathogens.

Keywords: Fungi. Yeasts. Air conditioning. Cross infection, prevention and control. Aerosols. Infection control.

INTRODUCTION

The importance of bioaerosols has been emphasized in the last decades due to their relationship to human health, leading to the appearance of diseases that range from allergies to disseminated infections in susceptible patients.⁵ Determining the composition and concentration of airborne microorganisms in internal and/or external areas of critical hospital wards is considered as an extremely necessary measure.¹ Hospitals are environments that require great care in terms of the environmental monitoring of critical areas. The aim of such a measure is to identify possible sources of contamination and dissemination, as well as the etiological agents potentially involved. On the other hand, in climate-controlled environments, the accumulation of humidity and organic matter inside air-conditioning trays may render them powerful sources of dispersion of bioaerosols.

Fungal infections of hospital origin are gaining in importance in recent years due to their progressive increase and to the high rates of morbidity and mortality with which they are associated.^4,19 Many of these infections are endogenous in nature, but others can be acquired by exogenous routes, through the hands of healthcare workers, contaminated infusion products and biomaterials, and abiotic environmental sources.^19,25 Monitoring bioaerosols and the microbiota of adjacent areas and hands of healthcare professionals may provide epidemiological information regarding the microorganisms involved in nosocomial infections.^8,25

Thus, in the present study, we carried out the environmental monitoring of hospital areas, isolating and quantifying both fungal bioaerosols and yeasts from abiotic sources and healthcare workers.

METHODS

The present study was conducted in critical areas of a hospital in the city of Araraquara, Southeastern Brazil, with monthly collections between October 2001 and August 2002. Critical areas included adult and pediatric intensive care units (ICUs) and surgical center, all of which are climate-controlled by air-conditioning and/or ventilation, but which lack HEPA (high-efficiency particulate air) filters. Collection of air samples from the critical areas was carried out at two different times, one during the morning, immediately following cleaning, and another during late afternoon, at the end of the regular shift.

Sampling totaled 196 air samples from critical areas of the hospital (surgical center and adult and neonatal ICUs) collected using appropriate equipment, following the recommendations of Andersen.¹⁶ Samples from the access routes to the critical areas and from a spot outside the building were collected on the same days as the sampling of internal areas.

Following the results of a pilot-test, airborne fungi were isolated by seeding the contents of 80 liters of air on Sabouraud agar + chloramphenicol culture medium. The identification of airborne fungi was based on the association of macroscopic aspects and microscopic characteristics of primary cultures. Identifications were later confirmed by sporulation characteristics following microculture on Lactrimel medium.^10,11

Simultaneously, the presence of yeasts was investigated on the hands, nails (interdigital spaces and nailbeds), and oral mucosa of 66 professionals directly in contact with ICU and surgical-center patients. Ninety-one swab samples were collected from doorknobs, beds, and telephones located inside the ICUs and surgical center.

Yeasts from biotic and abiotic sources were isolated on Sabouraud agar + chloramphenicol culture medium. Yeasts were identified previously on CHROMagar Candida medium. These results were associated to filamentation, germ-tube formation after two hours, assimilation and fermentation of carbon and nitrogen sources for confirmation. Identification was based on Kurtzman & Fell⁹ and Mendes-Giannini & Melhem.¹²

Kruskal-Wallis, Mann-Whitney, Anova-Manova, and t-tests were performed in order to analyze variables associated to collection time, location, and month of isolation, as well as the prevalence among the isolated genera. Statistical analyses were performed using Biostat 3.0 software.

RESULTS

Potentially pathogenic and toxigenic fungi were isolated from both environment and staff (Table). Thirty-one and 32 genera were isolated from surgical center and ICUs, respectively. Quantitative analysis of colony counts at the two collection times was relatively similar in the ICUs, but not in the surgical center (p<0.05). Cladophialophora was the predominant genus during morning (59.2%) and afternoon (74.4%), with significant differences (p<0.05) between collection times and during certain months.

We obtained 33,317 CFU/m³ in the ICUs, with a mean 317.1 CFU/m³ for internal environments and 454.6 CFU/m³ for environments outside special areas but inside the building. From the surgical center, we isolated 34,301 CFU/m³, with a mean 332.2 CFU/m³ from internal environments and 482.8 CFU/m³ from external ones. Forty-six samples were collected from locations outside the adult and pediatric ICUs and the surgical center, totaling 10,487 CFU/m³ and 4,401 CFU/m³, respectively. From the environment outside the building, we isolated 5,388 CFU/m³ from eight samples, with a mean 673.5 UFC/m³.

In the surgical center, Cladophialophora spp., Fusarium spp., Penicillium spp., and Aureobasidium spp. were present in 10 of the internal points evaluated. Fusarium spp. were isolated from all rooms during the morning period and from nine of 10 points in the afternoon collection. Aspergillus spp. were found in four of 10 points during the morning period. Fusarium spp. were prevalent (p<0.05) in the month of March during the morning; in May, Aspergillus spp. were also isolated at similar levels to the dematious fungus, whereas, in the remaining months, there was a prevalence of Cladophialophora spp. This genus increased significantly (p<0.05) in the month of March during the afternoon, falling gradually in subsequent months (Figure 1).

In the ICUs, over 60% of isolates from all rooms were of the Cladophialophora genus. There was a correlation between the number of colonies and the number of occupied hospital beds in both collection periods. Fusarium spp. were found in the five rooms, whereas Aspergillus spp. were found in four. In the adult ICU, during the morning period, Cladophialophora spp. remained stable during the months of October, December, and February, began to increase in April, and peaked in July (p<0.05). The same occurred in the afternoon period, but counts were proportionally lower, and differences between the two periods were significant (p<0.05) in October, February, and July (Figure 2).

In the Neonatal ICU (Figura 3), Cladophialophora spp. prevailed in July (p<0.05) and April (p<0.05), in the morning and afternoon periods, respectively, with a subsequent reduction in counts.

During renovation, there was a significant difference (p<0.05) in airborne fungus counts in relation to the sampled months. Higher rates of the predominant genus were found in external environments when compared to internal ones, especially during the afternoon (data not shown).

Another aspect of the survey focused on the healthcare staff of the adult and neonatal ICUs and of the surgical center. Yeasts were isolated from 39.4% of staff members. From nailbeds 12.1% of samples were isolated, 16.7% from interdigital spaces, and 10.6% from the oropharynx.

Of the positive samples obtained from staff members, Candida albicans accounted for 23%, C. parapsilosis for 19%, Candida spp. for 19%, C. guilliermondii for 8%, and Trichosporon spp. for 31%. C. albicans was isolated mainly from the oropharynx and C. parapsilosis and Trichosporon spp, from interdigital spaces and nailbeds.

A total 91 samples was obtained from furniture samples, of which 44% were positive for the presence of yeasts. Candida was the predominant genus (70%), followed by Trichosporon spp. Candida guilliermondii prevailed in the environment, especially on doorknobs (52%), where C. lusitaniae (5,0%), C. parapsilosis (3,0%), Candida spp. (10%), and Trichosporon spp. (30%) were also isolated. Trichosporon spp. were isolated from all the environments investigates, prevailing on doorknobs and in the air inside the ICUs.

DISCUSSION

The incidence of fungi is widely variable, depending on season, temperature, relative humidity, time of the day, speed and direction of winds, presence of human activity, and artificial climate control.^6,11,13

The composition and concentration of airborne organisms in external and internal environments of critical hospital areas have been little investigated, but there are studies that emphasize their importance due to the appearance of these agents in nosocomial infections.¹

In the present study, about 30 different genera were isolated from a hospital environment, with a predominance of Cladophialophora spp., Fusarium spp., Penicillium spp., Chrysosporium spp., and Aspergillus spp. The number of studies addressing this subject in Brazil is limited, and data on the airborne microbiota in hospitals are sparse. Silva et al²² investigated the fungal microbiota of the air and floor of a hospital, and found mainly Cladosporium spp. (65.0%), Aspergillus spp. (37.1%), Mycelia sterilia (26.9%), Fusarium spp. (20.1%), Penicillium spp. (19.8%), Aureobasidium spp. (18.4%), Curvularia spp. (16.2%), and Nigrospora spp. (15.3%). The prevalence of Cladophialophora has been reported in studies of aerosols from several Brazilian cities.^6,13

During the study period, the hospital underwent extensive renovation for the implementation of a new unit, and this coincided with an increase in propagule counts, especially of Cladophialophora spp. In the period preceding renovation, there was a significant increase in Fusarium spp. fungi (p<0.05), which are potentially pathogenic and are currently being reported in cases of invasive disease.²⁰Aspergillus spp. were isolated from the special areas in both periods, and, at similar levels to Cladophialophora spp., from inside the surgical center during the afternoon in the month of May, coinciding with renovation. In this period there was an increase in air contamination outside the building (data not shown), and a concomitant increase in the contamination of internal air. The same genera of fungi were recovered from both environments, suggesting an influence of external factors on internal contamination levels, as has been described in the literature.² There was also an increase in levels of Cladophialophora spp. in all internal collection points when compared to the previous months, and counts of this fungus were increased in the external environment. After the completion of the renovation process, there was a gradual decline, tending towards former levels, especially with respect to the Cladophialophora genus in the surgical center during the afternoon, indicating the importance of environmental monitoring to guide preventive measures. Several studies suggest that the distribution of fungi, both in terms of concentration and of the different genera encountered, vary among geographical areas and is influenced by environmental and seasonal factors.^{6,11,13,17,24}

The levels encountered are considered as high for such environments, especially if compared to the levels reported by Távora et al.²⁴ A complicating factor is the lack of reference values for climate-controlled environments considered as 'special.' Mean levels in the collection points outside the ICUs and surgical center were higher than those found in internal environments, but lower than those found outside the building. The two sectors investigated were climate-controlled, but did not use HEPA filters. However, climate control in the ICUs was less regular, and natural ventilation was used on occasion. A difference in the total number of fungi isolated is evident in the surgical center, probably due to the different routines of the two sectors.

In a study lasting for over 30 weeks carried out during the partial renovation of a hospital in Pittsburgh, USA, Overberger et al¹⁷ (1995) isolated Penicillium spp., Aspergillus spp., and, to a lesser extent, Alternaria spp., commonly isolated fungi, and Cladosporium spp. were the fungi most frequently isolated from the air. The authors reported an increase in levels of viable spores during renovation, and a subsequent decrease to former levels after renovation was concluded, results similar to those of the present study. Likewise, Bouza et al² (2002) found that the demolition of a hospital building was associated with increased colony counts in external and non-HEPA protected internal air, and that counts returned to former levels after 11 days. Aspergillus fumigatus was the fungus most frequently isolated, followed by Penicillium spp., Aspergillus niger, Aspergillus flavus, Aspergillus spp., and Fusarium spp., among other species. On the other hand, Aspergillus spp. were the filamentous fungus most frequently isolated by Panagopoulou et al¹⁸ (2002) in a multicenter hospital study, corresponding to 70,5% of all isolates. Mousa et al¹⁴ (1999), while investigating the correlation between fungi isolated from burn wounds and environmental surfaces of a burn care unit, found a greater proportion of Aspergillus niger in both patients and environments. However, Ulocladium spp. were the most common isolates in the control group (cultures obtained from outside the burn care unit), suggesting that patients may have acquired fungal infections while in the burn care unit.

In the present study, we also monitored the area around the beds and the healthcare professionals that worked inside the surgical center and ICUs. C. albicans, described in the literature as the major agent of nosocomial infections,⁴ was the prevailing species among the samples obtained from the healthcare staff. The presence of yeasts on the hands of these professionals is a source of concern for hospital-infection control teams, since this may be a potential source of contamination and dissemination of microorganisms. In the present survey, yeasts were found in samples obtained from 39.4% of the healthcare staff. Data on this subject in the literature show great disparity, with rates ranging between 17% and 80%.^8,23

Candidiasis and invasive aspergillosis have increased in recent years, especially among immunocompromised patients.^1,25C. albicans is still the major species involved in fungemia, although non-albicans species have increased in frequency, and are often refractory to conventional treatment.^3,4 On the other hand, infections by resistant microorganisms are increasing, resulting in higher mortality rates, increased hospitalization periods, and consequent higher cost to hospital services.²¹

In the present study, yeasts of the Trichosporon genus, considered as emergent agents of hospital infections,¹⁵ were also isolated at substantial levels, both from the healthcare staff and from all environmental surfaces investigated.

Of the samples obtained from the surfaces of furniture, two Candida lusitaniae isolates are worthy of note, since this yeast has mechanisms to develop resistance to Amphotericin B (certain strains being primarily resistant).⁷ One of these samples was isolated from a doorknob in the neonatal ICU, and the other in the following week from a doorknob in the adult ICU, which are physically close to each other. Although a genotypic characterization was not carried out, such a result may suggest that yeasts are carried from one place to the next within critical areas. Although Candida spp. were the yeasts most frequently isolated from healthcare staff and environment, the isolation of Trichosporon spp. was important due to its potential to resist antifungal therapy.

The characterization of environmental fungi from internal environments in critical hospital areas and of the fungal microbiota found on the hands of healthcare professionals is recognized worldwide as an important measure aimed at substantially reducing morbidity, mortality, and high hospital costs.

Environmental sources must hence be monitored, especially in special rooms, which accommodate immunosuppressed patients, susceptible to environmental pathogens exposure. Only then, it will be possible to design appropriate measures for controlling such pathogens and to choose the adequate patients therapies to be instituted.

REFERENCES

Received on 22/12/2003. Reviewed on 15/10/2004. Approved on 2/2/2005.

Financed by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq - Grant n. 400141/1998-1).

Study developed at the Laboratory of Clinical Mycology, Department of Clinical Analysis, Faculdade de Ciências Farmacêuticas of the Universidade Estadual Paulista Júlio de Mesquita Filho.

Part of the master's dissertation presented at the Faculdade de Ciências Farmacêuticas of the Universidade Estadual Paulista, Araraquara, SP, Brazil, 2003.

Presented at the XXII Congresso Brasileiro de Microbiologia, Florianópolis, SC, Brazil, 2003.

• 1. Alberti C, Bouakline A, Ribauad P, Lacroix C, Rousselot P, Leblanc T et al. Relationship between environmental fungal contamination and the incidence of invasive aspergillosis in haematology patients. J Hosp Infect 2001;48:198-206.
• 2. Bouza E, Peláez T, Pérez-Molina J, Marín M, Alcalá L, Padilla P et al. Demolition of a hospital building by controlled explosion: the impacto on filamentous fungal load in internal and external air. J Hosp Infect 2002;52:234-42.
• 3. Colombo AL, Nucci M, Salomão R, Branchini ML, Richtmann R, Derossi A et al. High rate of non-albicans candidemia in Brazilian tertiary care hospitals. Diagn Microbiol Infect Dis 1999;34:281-6.
• 4. Colombo AL. Epidemiology and treatment of hematogenous candidiasis: a Brazilian perspective. Braz J Infect Dis 2000;4:113-8.
• 5. Cooley JD, Wong WC, Jumper CA, Straus DC. Correlation between the prevalence of fungi and sick building syndrome. Ocupp Environ Med 1998;55:579-84.
• 6. Gambale W, Purchio A. Influência de fatores abióticos na dispersão aérea de fungos na cidade de São Paulo. Rev Microbiol 1983;14:204-14.
• 7. Hawkins JL, Baddour LM. Candida lusitaniae infections in the era of fluconazole availability. Clin Infect Dis 2003;36:14-8.
• 8. Horn WA, Larson EL, Mcginley KJ, Leyden JJ. Microbial flora on the hands of health care personnel: differences in composition and antibacterial resistance. Infect Control Hosp Epidemiol 1988;(9):189-93.
• 9. Kurtzman CP, Fell JW. The Yeasts: a taxonomic study. 4Ş ed. New York: Elsevier; 1998. p. 605.
• 10. Lacaz CS, Porto E, Heins-Vaccari E. Guia para identificação: fungos, actinomicetos e algas de interesse médico. São Paulo: Sarvier; 1998.
• 11. Lacaz CS, Porto E, Martins JEC, Heins-Vaccari EM, Melo NT. Tratado de Micologia Médica Lacaz. São Paulo: Sarvier; 2002. p. 1104.
• 12. Mendes-Giannini MJS, Melhem MSC. Fungos. In: Ferreira AW, Ávila SLM, coordenadores. Diagnóstico laboratorial das principais doenças infecciosas e auto-imunes. Rio de Janeiro: Guanabara-Koogam; 2001. p. 333-403.
• 13. Mezzari A, Perin C, Santos Jr AS, Bernd LAG, Gesu GD. Fungos anemófilos e sensibilização em indivíduos atópicos em Porto Alegre, RS. Rev Assoc Med Bras 2003;49(3):270-3.
• 14. Mousa HAL, Al-Bader SM, Hassan DA. Correlation between fungi isolated from burn wounds and burn care units. Burns 1999;25:145-7.
• 15. Moretti-Branchini M, Fukushima K, Schreiber AZ, Nishimura K, Papaiordanou PMO, Trabasso P et al. Trichosporon species infection in bone marrow transplanted patients. Diagn Microbiol Infect Dis 2001;39:161-4.
• 16. Nesa D, Lortholary J, Bouakline A, Bordes M, Chandenier J, Derouin F et al. Comparative performance of impactor air samples for quantification of fungal contamination. J Hosp Infect 2001;47:149-55.
• 17. Overberger PA, Wadowsky RM, Schaper MM. Evaluation of airborne particulates and fungi during hospital renovation. Am Ind Hyg Assoc 1995;56:706-12.
• 18. Panagopolou P, Filioti J, Petrikkos G, Giakouppi P, Anatoliotaki M, Farmaki E et al. Enviromental surveillance of filamentous fungi in three tertiary care hospitals in Greece. J Hosp Infect 2002;52:185-91.
• 19. Pfaller MA. Nosocomial candidiasis: emerging species, reservoirs, and modes of transmission. Clin Infect Dis 1996;22(Suppl 2):89-94.
• 20. Raad I, Tarrand J, Hanna H, Albitar M, Janssen E, Boktour M et al. Epidemiology, molecular mycology, and environmental sources of Fusarium infection in patients with cancer. Infect Control Hosp Epidem 2002;23:532-7.
• 21. Rentz AM, Halpern MT, Bowden R. The impact of candidemia on length of hospital stay, outcome and overall cost of illness. Clin Infect Dis 1998;27:781-8.
• 22. Silva MG, Moreira YK, Cisalpino IO. Flora fúngica do ar e do piso no Hospital das Clínicas da Universidade de Minas Gerais, Belo Horizonte, Brasil. Rev Microbiol 1983;14:215-22.
• 23. Strausbaugh LJ, Sewell DL, Ward TT, Pfaller MA, Tjoelker RC, Heitzman T. High frequency of yeast carriage on hands of hospital personnel. J Clin Microbiol 1994;32:2299-300.
• 24. Távora LGF, Gambale W, Heins-Vacari EM, Arriagada GLH, Lacaz CS, Santos CR et al. Comparative performance of two air samples for monitoring airborne fungal propagules. Braz J Med Biol Res 2003;36:613-6.
• 25. Wenzel RP. Nosocomial candidemia: risk factors and attributable mortality. Clin Infect Dis 1995;20:1531-4.

Publication Dates

• Publication in this collection
  30 June 2005
• Date of issue
  June 2005

History

• Received
  22 Dec 2003
• Reviewed
  15 Oct 2003
• Accepted
  02 Feb 2005