Apparent increase in Legionnaires' disease in Hong Kong attributed to rapid diagnostic kits

Abstract

We investigated potential common sources of infection and causes of increase in number of cases of Legionnaires’ disease (LD), which was statutorily notifiable to the Centre for Health Protection in Hong Kong, from 11-16 in previous four years to 37 in 2009. We compared clinical and epidemiological characteristics including smoking habit, chronic illnesses, disease severity, residences and travel history of cases recorded in 2009 with those reported in previous four years. We collected environmental samples of possible sources of infection of individual patients and compared the Legionella isolated with those obtained from patients using sequence-based typing. We also described the changes of laboratory diagnostic methods used in 2009. The LD cases reported in 2009 included 33 men and 4 women with age ranging between 38-92 years. Twenty-two of them were current or ex-smokers and 13 were known diabetics. All of them were hospitalized with pneumonic features and five died of the infection. The clinical and epidemiological features of cases reported in 2009 were similar to those of 51 cases in previous four years. There was no obvious geographical clustering of cases by their residences. Legionella pneumophila serogroup 1 strains were isolated from respiratory specimens of 10 cases. They belonged to 7 different types and differed from the environmental isolates; findings not suggesting common source of infection. Comparing with previous four years, the proportion of cases diagnosed by urinary Legionella antigen test (UAT) instead of serology and bacterial culture increased from <60% to >70% in 2009, while the number of specimens tested per year by UAT almost doubled from around 700 to 1350. Nonetheless, the proportion of UAT with positive result did not increase in 2009. The results did not suggest common source of infection. UAT allowed rapid, convenient and accurate diagnosis comparing with conventional methods. The apparent increase was possibly attributable to increased use of UAT in diagnosis and thus increased number of cases being identified.