© 2002 American Thoracic Society
To Cohort or Not to CohortHow Transmissible Is Pseudomonas aeruginosa?University of Washington School of Medicine Seattle, Washington Although several studies documenting patient-to-patient transmission of Burkholderia cepacia complex among patients with cystic fibrosis (CF) (1) have led clinicians worldwide to initiate aggressive infection control policies, the risk of Pseudomonas aeruginosa cross-colonization among patients and the most appropriate recommendations for segregation and infection control remain controversial. Molecular methods enabling accurate determination of the genetic relatedness of P. aeruginosa strains, such as pulsed-field gel electrophoresis (2) and random amplified polymorphic DNA analysis (3), have been available only since the late 1980s. Accumulating data from several molecular surveillance studies conducted worldwide (2–10) are beginning to elucidate the natural history of P. aeruginosa infection in the CF host. Two molecular epidemiologic studies of P. aeruginosa from large CF centers in Australia (4) and Canada (5) in the current issue of the AJRCCM (pp. 983–987 and 988–993, respectively) draw very different conclusions about the potential transmissibility of this pathogen. The Australian group (4) reports the extensive spread of a single clonal strain of P. aeruginosa among patients at a pediatric CF clinic. In contrast, the Canadian group (5) found patient-to-patient spread to be limited except with prolonged close contact, such as sibships, and proposed that environmental strains are the primary source of P. aeruginosa acquisition. Both studies examined cohorts of similar size and utilized the same genetic typing methods to determine strain relatedness, that is, random amplified polymorphic DNA analysis and pulsed-field gel electrophoresis. The centers in both studies had comparable infection control policies, whereby patients were not segregated in either the inpatient or outpatient setting. There are clear distinctions, however, between the studies. The Australian study is a cross-sectional study initiated in response to the isolation of a clonal strain of P. aeruginosa from five young patients who died from rapidly fatal lung disease. In contrast, the Canadian study provides a longitudinal perspective based on a repository of sequential P. aeruginosa isolates from patients spanning at least two decades. Among 118 expectorating patients colonized with P. aeruginosa, the Australian group observed a remarkably high prevalence, with 56 (47%) individuals harboring strains with a genetic fingerprint identical to that of the "severe disease" clone. Two surveys of clinics and hospital wards in 1995 and 1999 failed to reveal a common environmental source, suggesting (but not proving) patient-to-patient spread. Increased virulence of the clonal strain was suggested by a greater rate of antibiotic resistance as well as a higher hospitalization rate and slightly lower FEV1 among colonized patients. It is equally plausible, however, that frequently hospitalized patients were more likely to be exposed to this clonal pathogen. The Canadian group found that among 174 patients, the vast majority had unique P. aeruginosa genotypes that were maintained over many years. There were, indeed, a few clusters of common genotypes, the largest identified in 21 patients. With the exception of some sibships, there was no evidence of either geographic or temporal proximity among these patients except when receiving medical care. Prior studies support the positions of both groups. Widespread transmission of unique genotypes has been documented in cross-sectional studies from both Liverpool (6, 7) and Manchester (8). In contrast, two longitudinal studies did not identify evidence of significant risk for cross-infectivity with the exception of prolonged contact (sibships or close friendships) (2, 9). Thus, it appears that both authors are correct. Most patients acquire P. aeruginosa from the environment and will likely not become cross-contaminated with other CF pathogens except with prolonged contact. Nonetheless, more transmissible strains may emerge in certain clinical settings. Are these strains more virulent or simply opportunists that infect patients with severe lung disease? The cross-sectional nature of the studies in which epidemic strains have been identified precludes a full understanding of the impact of these strains on decline in lung function and disease progression. The common finding of all the reported "outbreak" strains has been resistance to multiple antipseudomonal antibiotics. However, antimicrobial resistance has also been observed among many genetically distinct P. aeruginosa strains infecting patients with CF. Thus, understanding the true clinical impact of widespread clonal strains will require more longitudinal surveillance from multiple geographic regions. There are limited published data to determine whether patient segregation or other infection control policies (such as face masks and limiting social contact) affect the rate of cross-infection. In the Manchester study (8), the authors did note that 24 patients with B. cepacia segregated from the remainder of the clinic population did not acquire the clonal strain of P. aeruginosa. Farrell and coworkers (11) also reported slower initial acquisition of P. aeruginosa among newly diagnosed patients who were segregated from P. aeruginosa-colonized patients compared with patients at another outpatient clinic that did not use patient segregation. This study did not use molecular typing methods to demonstrate P. aeruginosa transmissibility. The Danish CF centers have used aggressive segregation policies and cohorting for years and have reduced but not eliminated cross-infection (10). When considering the cost, stigmatization, and psychosocial stress of patient segregation policies, the decision process would be greatly enhanced by more objective data defining the benefit of infection control measures. How do clinicians caring for patients with CF resolve these disparate findings and come to their own conclusions about the best policies to protect their patients? The Cystic Fibrosis Foundation convened a consensus conference in 2001 to devise infection control recommendations. In a soon to be released document (L. Saimen, personal communication), the group recommends standard universal precautions in handling all body fluids and secretions from all patients. They also recommend limiting social interactions with other patients with CF in settings such as camps or schools. In the inpatient and outpatient settings, patients with epidemiologically important pathogens such as B. cepacia complex, methicillin-resistant Staphylococcus aureus, and multiresistant P. aeruginosa should be segregated and follow contact precautions. The group emphasized the important need for ongoing patient/family education and research studies to better understand patient-to-patient transmission of P. aeruginosa and other pathogens. Some clinicians and scientists may find these recommendations too stringent. The key to success of such an infection control policy is that it will be linked to ongoing molecular epidemiologic surveillance studies to measure its impact on rates of transmission of P. aeruginosa and other important pathogens. In addition, the financial and psychosocial impact on patients, families, and caregivers must be considered. Only through such means can the true risk–benefit ratio be ascertained. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
