Reporting of leptospirosis to the DDC in Thailand is voluntary. During a review of the national surveillance system for leptospirosis in 2 northeastern provinces, interviewed physicians said the national case definition was difficult to understand and apply (3). Investigators concluded that the lack of a standardized case definition for leptospirosis; the infrequent use of confirmatory laboratory testing; and the inability to link clinical, epidemiologic, and laboratory data hindered the system’s utility (3). These results imply that both underreporting and diagnostic inaccuracy of reported cases may be occurring. We conducted a prospective multicenter study to define the accuracy of clinical diagnoses of suspected leptospirosis in Thailand and to describe the geographic distribution of laboratory-confirmed cases.

Microscopic agglutination test (MAT) was performed at the World Health Organization (WHO)/United Nations Food and Agriculture Organization (FAO)/World Animal Health Organisation (OIE) Collaborating Center for Reference and Research on Leptospirosis, Brisbane, Queensland, Australia (4). A positive MAT was defined as a single titer of >1:400 or a 4-fold rise in titer between acute and convalescent phase samples. For Leptospira culture, 100 μL of whole blood, 500 μL of plasma, and 500 μL of serum were each injected into 3 mL of Ellinghausen, McCullough, Johnson, and Harris (EMJH) medium and supplemented with 3% rabbit serum and 0.1% agarose, then incubated aerobically at room temperature (25°C–30°C) for 6 months and examined every week for 2 months, every 2 weeks during months 3 and 4, and once a month during months 5 and 6. Examination was done by placing 1 drop of culture onto a microscopic glass slide and viewing by dark-field microscopy at 200× magnification. Positive cultures were referred to the WHO/FAO/OIE Collaborating Center for Reference and Research on Leptospirosis for identification by using the cross-agglutination absorption test (4).

A total of 700 patients with a clinical diagnosis of leptospirosis were recruited during the study period. All patients had blood samples collected at the hospital for leptospire culture and serologic testing; convalescent-phase serum samples were obtained during follow-up for 509 (73%) patients.

The median age of patients with suspected leptospirosis was 38 years (range 2–95 years, interquartile range (IQR) 28–49 years); 504 (72%) were men. The number of clinically diagnosed leptospirosis cases by month in the north, northeast, central, and southern regions is shown in Figure 1. Most cases (597, 85%) were recorded in 4 provinces in the north or northeast (Table). Cases were predominantly identified during the rainy season (June–October) in the north and northeast in 2003, with a second peak in the northeast, but not the north, during the rainy season of 2004. Little variation occurred over time in the central and southern regions.

Of the 700 patients who received a clinical diagnosis of leptospirosis, 143 (20%) received a confirmed diagnosis of leptospirosis based on Leptospira isolation, MAT testing, or both (Table). The median age of patients with confirmed leptospirosis was 35 years (range 10–68 years, IQR 27–45 years); 121 (85%) were men. The diagnosis was confirmed after isolation of leptospires from 15 (11%) patients; the geographic distribution is shown in the Table. The serovars of cultured Leptospira were: L. interrogans serovar (sv.) Autumnalis (7), L. interrogans sv. Bataviae (2), L. interrogans sv. Pyrogenes (2), L. borgpetersenii sv. Javanica (1), L. interrogans sv. Hebdomadis (1), L. interrogans sv Grippotyphosa (1), and an unidentified serovar (1). An additional 128 patients with culture-negative samples had been exposed to Leptospira as determined by MAT; results for 96 (75%) were based on a 4-fold rising titer and for 32 (25%), on a single raised titer of >1:400.

The geographic distribution of the 143 laboratory-confirmed cases is summarized in the Table. Most of these patients (124, 87%) lived in the 4 provinces found in the north and the northeast (Table). The month of diagnosis for confirmed cases is shown in Figure 2; most were during the rainy season.

The finding that both clinical and confirmed cases of leptospirosis were more common in the north and northeast is consistent with DDC reports. Increased incidence in this region may have resulted from >1 events, such as an increase in the rodent population, a natural reservoir for this pathogen, and a population in which around one third are positive for Leptospira in northeast Thailand (5). Alternatively, 1 clone or a small number of bacterial clones may have become adapted for persistence at greater numbers within the natural host or in the environment. These factors could increase the leptospire count in contaminated water. It is also possible that 1 clone or a small number of clones have become adapted for enhanced invasion of the human host. The most prevalent serovar isolated was L. interrogans serovar Autumnalis (7/15 [47%] isolates), 6 of which were from cases in the north or northeast. Further genomic analysis is required to determine whether clonality exists among these isolates.