Assessment of the Clinical Presentation and Treatment of 353 Cases of Laboratory- Confirmed Leptospirosis in Hawaii,

Transcription

1 MAJOR ARTICLE Assessment of the Clinical Presentation and Treatment of 353 Cases of Laboratory- Confirmed Leptospirosis in Hawaii, Alan R. Katz, 1 Vernon E. Ansdell, 2 Paul V. Effler, 3 Charles R. Middleton, 3 and David M. Sasaki 3 1 Department of Public Health Sciences and Epidemiology, John A. Burns School of Medicine, University of Hawaii, 2 Kaiser-Permanente Medical Center, and 3 Epidemiology Branch, Communicable Disease Division, Hawaii State Department of Health, Honolulu Leptospirosis is frequently misdiagnosed as a result of its protean and nonspecific presentation. Leptospirosis, a zoonosis with global distribution, commonly occurs in tropical and subtropical regions; most reported cases in the United States occur in Hawaii. All laboratory-confirmed leptospirosis cases in the State of Hawaii from 1974 through 1998 ( n p 353) were clinically evaluated. The most common presentation involved nonspecific signs or symptoms, including fever, myalgia, and headache. Jaundice occurred in 39% of cases; conjunctival suffusion was described in 28% of these cases. Initiation of antibiotics before the seventh day of symptoms was associated with a significantly shortened duration of illness. Because early recognition and initiation of antibiotic therapy are important, clinicians should familiarize themselves with the clinical presentation of leptospirosis, and when evaluating a patient with a febrile illness, they should obtain exposure and travel histories and entertain the possibility of leptospirosis in the differential diagnosis. Leptospirosis is a zoonosis with protean manifestations and worldwide distribution [1]. It is caused by an infection with a bacterial spirochete belonging to the family Leptospiraceae. The broad group of leptospires with animal host associations has been defined as the genospecies Leptospira interrogans sensu lato [2]. The climatic conditions in tropical and subtropical regions help provide an optimal environment to support the survivability of leptospires [3]. From 1974 through 1994, the State of Hawaii consistently had the highest reported annual incidence rate of leptospirosis in the Received 7 March 2001; revised 25 June 2001; electronically published 23 October Financial support: State of Hawaii Department of Health through a grant administered by the Research Corporation of the University of Hawaii. Reprints or correspondence: Dr. Alan R. Katz, Dept. of Public Health Sciences and Epidemiology, John A. Burns School of Medicine, University of Hawaii, Biomedical Sciences Bldg., Rm. D-104K, 1960 East-West Rd., Honolulu, HI (katz@hawaii.edu). Clinical Infectious Diseases 2001; 33: by the Infectious Diseases Society of America. All rights reserved /2001/ $03.00 United States [4 24]. Although dropped from the list of national notifiable diseases after 1994 [24], leptospirosis remains a reportable illness in the State of Hawaii. Many, if not most, cases of leptospirosis go unrecognized because of the lack of specificity in signs and symptoms. Confirmation of the diagnosis is also difficult because of problems associated with isolating the organism and with serologic testing. A better understanding of the clinical presentation of leptospirosis is needed to enhance its recognition and appropriate treatment. Here, we assess the clinical presentation and treatment of 353 laboratory-confirmed cases of leptospirosis, infected through exposure in the State of Hawaii during a 25-year period. PATIENTS AND METHODS Leptospirosis case reports during the 25-year period of were investigated by epidemiologists from the Hawaii State Department of Health (DOH). Initial contact was with the reporting doctor, and medical records, when available, were abstracted. Attempts were 1834 CID 2001:33 (1 December) Katz et al.

2 made to contact and interview each patient. A standardized case investigation form was used to compile demographic, epidemiologic, and clinical data. The case investigation form uses a checklist for signs or symptoms, laboratory results, and activities relating to exposures; it also has questions about occupation, antibiotic use, exposure locations, and pertinent dates. All reported cases were investigated in a similar manner. Epidemiologic data came predominately from patient interviews; clinical information came from interviews and chart reviews; and laboratory results came exclusively from medical records. To assess exposure source and estimate incubation periods, patients were asked about high-risk activities that occurred during the 21 days before the onset of symptoms. These included exposures to animals, animal urine, mud, or potentially contaminated freshwater sources involving occupational activities (e.g., farming, ranching), recreational activities (e.g., freshwater swimming, hiking), or activities around the home (e.g., gardening, trapping rats). If there were continuous or multiple exposures, the incubation period was considered indeterminate. To assess completeness of ascertainment for each independent variable studied, the denominator (total patients for whom information on the specific variable was obtained) is supplied. A confirmed case had a clinically compatible illness with 1 of the following laboratory criteria for confirmation: a 4- fold increase in microscopic agglutination test (MAT) titer between acute-phase and convalescent-phase serum specimens, isolation of Leptospira species from a clinical specimen, or demonstration of Leptospira species in a clinical specimen by use of immunofluorescence [25, 26]. All other cases were classified as either probable (clinically compatible with supportive serologic findings i.e., MAT titer of 1:200 in 1 serum specimens [27]) or suspect (clinically compatible with less supportive laboratory evidence of infection e.g., MAT titer of!1: 200, positive macroscopic slide agglutination test result, or positive result of an indirect hemagglutination assay). Only cases classified as confirmed were included in this study. MAT and direct fluorescent antibody testing were conducted by the Centers for Disease Control and Prevention (CDC) in Atlanta. The DOH laboratory conducted isolation procedures with Ellinghausen-McCullough-Johnson-Harris semisolid media. Serogrouping and serotyping of culture isolates were also conducted by the CDC [25]. Technical assistance was received from the Royal Tropical Institute in Amsterdam, The Netherlands, in 1989, for microscopic agglutination testing and serotyping. Infecting serogroups were definitively identified on culture isolates but only presumptively identified by MAT; definitive identification is not possible with MAT because of cross-agglutination or cross-reactivity between serovars of different serogroups [28]. For cases that demonstrated a 4-fold increase in MAT titer between acute- and convalescent-phase serum specimens, the serogroup showing the highest titer on the convalescent-phase serum specimen was considered to be the presumptive infecting serogroup. If there was 11 serogroup with the same high titer, the presumptive infecting serogroup for the patient was designated indeterminate. Frequencies and x 2 tests for linear trend were calculated by use of EpiInfo, version 6.04 (CDC). Fisher s exact test and mid- P corrected 95% CIs with exact P values for odds ratios (ORs) were calculated using StatXact, version (Cytel Software Company). The Wilcoxon-Mann-Whitney test was used to assess differences in the duration of illness in days, as a function of antibiotic use, and interval between symptom onset and initiation of antibiotic therapy, for patients treated with antibiotics. StatXact, version 4.0.1, was also used to calculate P values associated with the Wilcoxon-Mann-Whitney test statistic. All tests were 2-tailed. A P value of.05 was considered statistically significant. RESULTS In total, 752 cases of leptospirosis were reported to the DOH during the 25-year period of A total of 709 cases were contracted through exposures within the State of Hawaii; 43 cases were related to exposures that occurred outside of the state, including 27 cases from the Federated States of Micronesia, 6 from American Samoa, 4 from Guam, 2 from Costa Rica, and 1 each from Thailand, The Philippines, Mexico, and Utah. Of the 709 cases of leptospirosis acquired via exposures within the state, 353 (50%) were classified as confirmed ; the remainder were classified as probable or suspect. The completeness of the case investigation form varied according to the variable being measured: demographic variables, such as sex, age, and island of exposure, had virtually 100% ascertainment, whereas completeness of information regarding specific laboratory results varied 40% 70%. Men and boys accounted for 325 cases (92%). The median age of affected patients was 33 years (range, 1 78 years). Island of exposure was ascertained for 351 of 353 cases. Of the 351 cases with a known island of exposure, 176 (50%) were on the island of Hawaii, 100 (28%) were on the island of Kauai, and 67 (19%) were on the island of Oahu. Five cases were reported from the island of Maui, and a single case was exposed on each of the following islands: Lanai, Molokai, and Niihau. A comparison of mean annual incidence rates revealed that Kauai had with the highest rate (7.9 cases per 100,000 population), followed by the islands of Hawaii (5.9 cases per 100,000 population), Oahu (0.3 cases per 100,000 population), and Maui (0.2 cases per 100,000 population). Current occupation was ascertained for 328 (93%) of the Leptospirosis in Hawaii CID 2001:33 (1 December) 1835

3 353 patients. Persons without formal employment (e.g., unemployed, retired, student, homemaker) accounted for 102 (31%) of 328 cases. Farmers or ranchers (including sugarcane workers) accounted for 92 cases (28%). Blue-collar workers (e.g., construction workers, day laborers, carpenters) accounted for 51 cases (16%). Twelve cases (4%) were in active-duty military personnel. Information allowing for an exposure classification was possible for 335 (95%) of 353 cases. Occupationally related exposures were reported for 137 (41%) of 335 cases (e.g., farming, ranching, aquaculture, abattoir, veterinarian); 143 (43%) of 335 cases were exposed through recreational activities, including freshwater swimming, hiking, camping, and hunting. Among confirmed cases, 81 (23%) of 353 were diagnosed by culture isolates from the following clinical specimens: blood samples, for 66 cases; urine samples, for 6 cases; CSF specimens, for 2 cases; blood and urine samples, for 1 case; blood and CSF samples, for 1 case; and an unspecified source, for 5 cases. The 81 culture-confirmed cases included 48 with diagnostic MAT results and 33 diagnosed using culture alone. The diagnoses were made using serologic testing without culture confirmation for 270 cases, and 2 cases demonstrated antibodies to Leptospira species in postmortem tissue specimens by means of a direct fluorescent antibody test. Among confirmed cases, the most frequently reported symptoms were fever, myalgia, and headache. Pertinent laboratory results included evidence of renal abnormalities, hepatic abnormalities, or both in 49% 73% of patients (table 1). Moresevere disease manifestations were associated with infection by the Icterohemorrhagiae serogroup (table 2). Among 353 confirmed cases during this 25-year interval, 5 patients died (casefatality rate, 1.4%). Increasing age was associated with an increase in leptospirosis related mortality (x 2 for linear trend, 5.5; P p.02). Initial clinical impressions were recorded for 312 (88%) of 353 patients. The most common initial diagnoses were leptospirosis (193 cases [62%]), influenza (20 cases [6%]), viral syndrome (19 cases [6%]), hepatitis (19 cases [6%]), fever of unknown etiology (12 cases [4%]), gastroenteritis (10 cases [3%]), and meningitis (9 cases [3%]). An incubation period was estimated for 150 patients for whom exposure and onset dates were recorded. The incubation period ranged in duration from 1 to 21 days (median, 8 days). Information on the duration of illness was obtained from 245 patients and ranged from 2 to 84 days (median, 14 days). A total of 239 (68%) of 349 patients were hospitalized. Information regarding antibiotic therapy was obtained from 327 of 353 patients. Two hundred ninety-four (90%) of the 327 patients received antibiotics. Untreated patients had a minimum 7-day duration of illness, whereas 25% of those treated Table 1. Clinical findings for 353 confirmed cases of leptospirosis, Hawaii, Sign, symptom, and result (definition) Selected sign or symptom No. of patients with data available No. (%) of patients affected Fever (99) Myalgia (91) Headache (89) Chills (87) Anorexia (82) Nausea (77) Vomiting (73) Arthralgia (59) Diarrhea (53) Abdominal pain (51) Backache (51) Jaundice (39) Conjunctival suffusion (28) Nuchal rigidity (27) Oliguria or anuria (26) Hepatomegaly (16) Pneumonia (17) Rash (8) Splenomegaly (4) Laboratory results Renal Elevated blood urea nitrogen (120 mg/dl) (49) Elevated creatinine (11.5 mg/dl) (54) Hematuria (72) Proteinuria (54) Hepatic Elevated alanine aminotransferase level (serum glutamic-pyruvic transaminase; 140 U/L) (73) Elevated total bilirubin level (11.0 mg/dl) (70) Hematologic Elevated WBC count (110,000 cells/mm 3 ) (39) Decreased WBC count (!4300 cells/mm 3 ) (7) Thrombocytopenia (!140,000 cells/mm 3 ) (58) Decreased hematocrit (!34%) (32) with antibiotics recovered within 7 days. However, the observed difference in illness duration for people treated with or without antibiotics was not statistically significant (Wilcoxon-Mann- Whitney test, P p.4). For 210 (71%) of 294 patients who received antibiotics, we had information on date of symptom onset and date of initiation of antibiotic therapy. Of 210 patients who received antibiotics, 205 (98%) had initiated therapy within 9 days of onset 1836 CID 2001:33 (1 December) Katz et al.

4 Table 2. Association between infecting serogroup (Icterohemorrhagiae vs. non-icterohemorrhagiae) and clinical manifestations of disease serverity among confirmed leptospirosis cases with an identified serogroup (by isolate or, presumptively, by microscopic agglutination test), Hawaii, No. patients affected/total Clinical manifestation (definition) Icterohemorrhagiae cases Non-Icterohemorrhagiae cases OR (mid-p corrected 95% CI) Jaundice 65/133 36/ ( ).0014 Elevated bilirubin (total bilirubin 11 mg/dl) 69/84 45/ ( ).0016 Elevated blood urea nitrogen (120 mg/dl) 76/126 32/ ( )!.0001 Elevated creatinine (11.5 mg/dl) 48/72 24/ ( ) Dialysis 9/109 1/ ( ).0093 Oliguria/anuria 50/125 16/ ( )!.0001 Death a 3/169 0/154 (0.53 ).25 b a Five patients in this series died. One had a serogroup Icterohemorrhagiae culture isolate, an additional 2 had presumptive serogroup Icterohemorrhagiae identified by microscopic agglutination test, and 2 had cases that were diagnosed by positive direct fluorescent antibody testing of specimens obtained at autopsy, without serogroup identification. b Not statistically significant. of symptoms. Differences in duration of illness were examined as a function of how soon after the onset of symptoms antibiotic therapy was initiated. A consistent shortening of illness duration was observed for people for whom antibiotics were initiated earlier rather than later for every day tested after the onset of symptoms, from day 0 through day 8. Patients treated with antibiotics before 7 days after symptom onset experienced a significant shortening in their duration of illness compared with patients treated on or after the seventh day of symptoms (Wilcoxon-Mann-Whitney test, P p.04), with a mean difference of 4.5 days in duration of illness (figure 1). Initiation of therapy within 2 days of symptom onset was associated with a highly significant shortening of illness duration (Wilcoxon- Mann-Whitney test, P p.006). There were no significant differences in length of illness if antibiotic therapy was initiated after the seventh day of symptoms (Wilcoxon-Mann-Whitney test, P p.18). Of the 294 patients who received antibiotic therapy, for 280 (95%), we had information on the specific antibiotics administered. Single-drug antimicrobial therapy was used in 167 (60%) of 280 patients, whereas combination antimicrobials were used in 113 patients (40%). The most commonly prescribed regimen was single-drug penicillin antimicrobials (penicillin, ampicillin, amoxicillin, or ticarcillin); this regimen was prescribed to 88 patients (31%). Combination treatment with antimicrobials that included a penicillin-class antibiotic was the next-most common treatment; it was used in 68 patients (24%). Single-drug tetracycline antimicrobials (tetracycline, doxycycline, or minocycline) were provided for 59 patients (21%). A total of 103 (91%) of 113 combination therapies included either a penicillin or tetracycline antimicrobial (table 3). DISCUSSION This large case series is unique in presenting only confirmed cases that met specified diagnostic laboratory criteria [25, 26]. The majority of case series published elsewhere either contained all case reports in their analysis, including those without laboratory confirmation [29, 30], or utilized a variety of less specific case definitions for the purposes of confirmation [31 36]. The use of consistent laboratory criteria allow for valid comparisons between clinical and epidemiologic patterns of leptospirosis in different populations [2]. This case series reflects the majority of recognized leptospirosis cases within the United States in , and to our knowledge, this is the first large case series and surveillance report in the United States since 1979 [30]. The symptoms reported in this series are consistent with those reported in other large case series, with fever, headache, and myalgia being the most common presentations. Nausea and vomiting are also relatively common, and jaundice occurs in 30% 40% of patients [29 33]. Conjunctival suffusion has been noted as being pathognomonic for leptospirosis [37]. It has been suggested that a careful examination will identify conjunctival suffusion in virtually all patients with leptospirosis [37]. The finding of conjunctival suffusion in 28% of patients in this series may better reflect its true occurrence and indeed, this is consistent with reports from case series published elsewhere [30 34]. The clinical laboratory findings in this series are also similar to descriptions published elsewhere. Thrombocytopenia and polymorphonuclear leukocytosis are common. The findings of urinalysis are frequently abnormal, with hematuria and albu- P Leptospirosis in Hawaii CID 2001:33 (1 December) 1837

5 Figure 1. Duration of illness (in days) for 210 laboratory-confirmed cases of leptospirosis treated with antibiotics as a function of how soon after symptom onset antibiotic therapy was initiated (treatment within 6 days of symptom onset versus treatment on or after day 7), Hawaii, minuria occurring during the early phase of the illness [2]. Elevations of both blood urea nitrogen and creatinine levels are frequently found; in severe illness, this is related to renal failure, but in mild disease, this may reflect a prerenal pattern related to dehydration [2]. During the initial leptospiremic phase, even in mild cases of disease, there is typically some hepatic involvement with elevated levels of bilirubin and mild increases (up to 200 U/L) of aminotransaminases [38]. The clinical presentation of leptospirosis may be highly variable and may be related to different infecting serovars, amount of inoculum, host factors, or a combination of these [2]. Two classic forms have been described: an anicteric febrile illness, and the more serious icteric leptospirosis (or Weil s disease), with hepatic, renal, and vascular involvement [3]. Among 150 United States military personnel in South Vietnam hospitalized with leptospirosis during a 2-year period, only 1.5% had jaundice, and most presented with nonspecific symptoms of fever with abrupt onset, chills, and headache. There were no deaths in this series [39]. In contrast, among 326 patients with leptospirosis in an urban outbreak of infection in Salvador, Brazil, 91% manifested jaundice; the case-fatality rate was 15% [40]. Patients from recently reported epidemics in Korea and Nicaragua have presented with a febrile illness of sudden onset and severe pulmonary hemorrhage; jaundice and renal manifestations were not common [41, 42]. In the outbreak of infection in Salvador, Brazil, 173 (90%) of the 193 patients who had cases that were defined as confirmed (4-fold increase in MAT titer between paired serum specimens; minimum MAT titer of 11:800 in 1 serum specimen; or culture isolation) or probable (minimum MAT titer of 11:100 in 1 serum specimen) were infected with serogroup Icterohemorrhagiae (identified presumptively or by isolates) [40]. In contrast, 9 (6%) of the 150 patients in the Vietnam series (minimum MAT titer of 1:400 or culture isolation) were infected with serogroup Icterohemorrhagiae (identified presumptively or by isolates) [39]. Fifty-five (63%) of 88 patients in the Korean epidemic (minimum MAT titer of 1:80 or 4-fold increase in MAT titer between paired serum specimens) were presumptively infected with serogroup Icterohemorrhagiae [41]. The Nicaraguan epidemic was associated with the Canicola serogroup [42]. Although infection with serovars from the Icterohemorrhagiae serogroup has been reported to be associated with increased disease severity [35, 43], a number of published case series have questioned the relationship between infecting serogroups and disease severity [32, 44, 45]. In the current series, a clear association between infection with Icterohemorrhagiae and increased severity of illness was demonstrated (table 2). Although no significant differences could be detected between use or nonuse of antibiotics and duration of illness, 90% of patients in this series received antibiotics. The finding of shortened duration of illness associated with earlier initiation of therapy is consistent with the well-described biphasic nature of leptospirosis. The initial leptospiremic phase starts abruptly, lasting 4 7 days with fever and myalgias. The immune phase follows 1 3 days later with recurrent fever and myalgia [2]. A number of studies have reported shortened duration of illness when appropriate antibiotic therapy was administered during the initial phase of the illness (within 2 4 days) [3, 46, 47]. A review that evaluated antibiotic effectiveness in leptospirosis treatment concluded that, because of the small number of published randomized clinical trails, there was insufficient evi CID 2001:33 (1 December) Katz et al.

6 Table 3. Prescribed antimicrobial treatment regimens for 280 laboratory-confirmed cases of leptospirosis receiving antibiotic therapy, Hawaii Regimen No. (%) of patients who received therapy Single antimicrobial regimen 167 (60) Penicillins 88 (31) Penicillin 71 Ampicillin a 10 Amoxicillin b 6 Ticarcillin with clavulanic acid 1 Tetracyclines 59 (21) Doxycycline 44 Tetracycline 14 Minocycline 1 Cephalosporins 17 (6) Ceftriaxone 8 Cefazolin 4 Cephalexin 3 Cephalothin 1 Unspecified 1 Quinolones 2 (1) Erythromycin 1 (!1) Combination antimicrobial regimen 113 (40) Including penicillin-class antibiotic 68 (24) Including tetracycline without penicillin 35 (13) Including erythromycin without penicillin or tetracycline 4 (1) Cephalosporin combinations without penicillin, tetracycline, or erythromycin 4 (1) Trimethoprim and sulfamethoxazole 1 (!1) Ofloxacin plus amantadine 1 (!1) a Including 2 with sulbactam. b Including 1 with clavulanic acid. dence to provide clear guidelines for practice. However, suggestive evidence supported the use of penicillin and doxycycline [48]. Current internal medicine textbooks advise early antibiotic therapy [38, 49]. It is important to note that, in patients with severe disease, late administration of antibiotics has also demonstrated clinical efficacy and reduction in mortality rates [50]. Because the findings of the current study are based on data collected through surveillance case reports, retrospective interviews, and chart reviews, conclusions regarding the efficacy of early antibiotic therapy are not fully warranted. Randomized clinical trials are necessary to definitively assess the clinical efficacy of early antibiotic intervention in the treatment of leptospirosis. The case-fatality rate in this series (1.4%) is lower than that reported from earlier national case series (5% 7%) [29 31] and case series from India [36], the Seychelles [34], Barbados [45], and New Caledonia [32]. The lower case-fatality rate in this series may be explained by increased awareness of leptospirosis, resulting in earlier recognition and diagnosis and earlier initiation of appropriate therapy, including placement in closely monitored intensive care units. This finding may also be explained by better surveillance and a greater number of patients detected with less serious disease. Other reported case series may be biased toward only recognizing and including the most severely ill or hospitalized patients, thus leading to higher case-fatality rates. The finding of an increased case-fatality rate with increasing age has been described elsewhere [30, 31, 36, 51]. With the protean manifestations of leptospirosis, clinicians must keep a high index of suspicion for leptospirosis in order to make the correct diagnosis. The DOH has strongly emphasized education, regularly updating doctors in the state regarding the presentation, diagnosis, and treatment of leptospirosis. A statewide Leptospirosis Ad Hoc Committee comprising epidemiologists, doctors, vector-control specialists, environmental health workers, and health educators from the military, public, and private sectors has met monthly since 1987 to discuss prevention aspects of leptospirosis and other related zoonotic illnesses, to investigate disease occurrence, and to develop and disseminate educational materials for doctors, other health professionals, and the general public. Although leptospirosis has historically been categorized as an occupational illness of farmers, ranchers, and military personnel, national surveillance has demonstrated a temporal shift since the 1970s, with decreasing occupational exposures and a concurrent increase in exposures related to recreational activities [3]. In support of these observations are the findings of this series in which 10 of 12 military personnel identified with leptospirosis were epidemiologically linked to recreational exposures [52]. Additional challenges to the detection and confirmation of cases of leptospirosis include problems associated with isolating the organism and with serologic testing. Culture isolation requires special laboratory techniques and media, and may take several weeks. The MAT is currently considered to be the reference serologic test for leptospirosis infection. However, methodologic complexities limit its use [25]. In addition, MAT confirmation requires a 4-fold titer increase in paired acute- and convalescent-phase serum samples, yet numerous factors impede collection of the convalescent-phase serum samples, even when the diagnosis of leptospirosis has been entertained. A limitation of the present study involves the manner in which patients were identified and exposure information was ascertained. Case identification was based on physicians reports to the DOH. It is likely that patients with less severe illness never sought medical care and, therefore, their illnesses Leptospirosis in Hawaii CID 2001:33 (1 December) 1839

7 went undiagnosed and unrecognized. This would bias our series to one that portrayed more severe case presentations instead of a representative series of all leptospirosis cases. In addition, because exposure data were obtained retrospectively from patient interviews, there is potential for recall bias. Also, it should be noted that Hawaii, although part of the United States, is not climatically representative of the other states, with the exception of Florida. Therefore, these findings may be more comparable with those of other tropical and subtropical island nations than to areas with temperate climates. It is important to recognize that the 353 cases reported in this series reflect only cases exposed and diagnosed in the State of Hawaii. Hawaii is a major tourist destination, with an estimated 6.7 million visitors in 1998 alone [53], including 4 million visitors from the mainland United States. The median duration of stay for tourists is 7 days. Leptospirosis has a usual incubation period of 7 12 days, but can range from 2 to 20 days or longer [2, 3]. It is clear that many cases of leptospirosis contracted by visitors will not manifest until the tourist returns home. It is also clear that the majority of these cases may go unrecognized and misdiagnosed. Although a travel history to tropical destinations, such as Hawaii, Tahiti, Thailand, Indonesia, the Caribbean, or Costa Rica, should heighten one s index of suspicion [54], leptospirosis is increasingly being uncovered within urban settings in the mainland United States [55]. Recent outbreaks associated with recreational exposures among triathletes in Wisconsin and Illinois [56], and among participants returning from the Eco-Challenge competition in Borneo [57], as well as occupational exposures among Missouri pig farmers [58], have refocused attention on this underrecognized disease. The ubiquitous nature of this illness, together with its nonspecific and variable manifestations, require that clinicians at least entertain the inclusion of leptospirosis in the differential diagnosis of all patients with febrile illness. Acknowledgments We thank Harry Domen, Henry Higa, Henri Minette, Robert Worth, Richard Vogt, Mitsuto Sugi, Chester Wakida, and Sally Jo Manea, from the Hawaii State Department of Health (Honolulu); Arnold Kaufmann, Catherine Sulzer, Faye Rogers, Sandra Bragg, and Robbin Weyant, from the Centers for Disease Control and Prevention (Atlanta); Wiepko Terpstra and Hans Korver, from the Royal Tropical Institute (Amsterdam); Linda Odello, from the University of Hawaii School of Public Health (Honolulu); and F. DeWolfe Miller, John S. Grove, and Virginia M. Tanji, from the University of Hawaii John A. Burns School of Medicine (Honolulu). References 1. World Health Organization. Leptospirosis worldwide, Wkly Epidemiol Rec 1999; 74: Faine S, Adler B, Bolin C, et al. Leptospira and leptospirosis. 2d ed. Melbourne: MediSci, Farr RW. Leptospirosis. Clin Infect Dis 1995; 21: Center for Disease Control. Reported morbidity and mortality in the United States, 1974: annual supplement, MMWR Morb Mortal Wkly Rep 1975; 23(53):7, Center for Disease Control. Reported morbidity and mortality in the United States, 1975: annual supplement summary, MMWR Morb Mortal Wkly Rep 1976; 24(54):7, Center for Disease Control. Reported morbidity and mortality in the United States, 1976: annual summary, MMWR Morb Mortal Wkly Rep 1977; 25(53):8, Center for Disease Control. Reported morbidity and mortality in the United States: annual summary, MMWR Morb Mortal Wkly Rep 1978; 26(53):9, Center for Disease Control. Reported morbidity & mortality in the United States: annual summary, MMWR Morb Mortal Wkly Rep 1979; 27(54):8, Center for Disease Control. Reported morbidity & mortality in the United States: annual summary, MMWR Morb Mortal Wkly Rep 1980; 28(54):4, Centers for Disease Control. Reported morbidity & mortality in the United States: annual summary, MMWR Morb Mortal Wkly Rep 1981; 29(54):4, Centers for Disease Control. Reported morbidity & mortality in the United States: annual summary, MMWR Morb Mortal Wkly Rep 1982; 30(54):4, Centers for Disease Control. Reported morbidity & mortality in the United States: annual summary, MMWR Morb Mortal Wkly Rep 1983; 31(54):4, Centers for Disease Control. Reported morbidity & mortality in the United States: annual summary, MMWR Morb Mortal Wkly Rep 1984; 32(54):4, Centers for Disease Control. Reported morbidity and mortality in the United States: annual summary, MMWR Morb Mortal Wkly Rep 1986; 33(54):4, Centers for Disease Control. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1985; 34(54):4, Centers for Disease Control. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1986; 35(55):4, Centers for Disease Control. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1987; 36(54):4, Centers for Disease Control. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1988; 37(54):4, Centers for Disease Control. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1989; 38(54):4, Centers for Disease Control. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1990; 39(53):4, Centers for Disease Control. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1991; 40(53):4, Centers for Disease Control and Prevention. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1992; 41(55):4, Centers for Disease Control and Prevention. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1993; 42(53):4, Centers for Disease Control and Prevention. Summary of notifiable diseases, United States, MMWR Morb Mortal Wkly Rep 1994; 43(53):4, Weyant RS, Bragg SL, Kaufmann AF. Leptospira and Leptonema. In: Murray PR, Baron EJ, Pfaller MA, Tenover FC, Yolken RH, eds. Manual 1840 CID 2001:33 (1 December) Katz et al.

8 of clinical microbiology. 7th ed. Washington, DC: American Society for Microbiology, 1999: Faine S. Leptospirosis. In: Hausler WJ Jr, Sussman M, eds. Bacterial infections. Vol 3 of Collier L, Barrows A, Sussman M, eds. Topley and Wilson s microbiology and microbial infections. 9th ed. London: Arnold, 1998: Centers for Disease Control and Prevention. Case definitions for infectious conditions under public health surveillance. MMWR Morb Mortal Wkly Rep 1997; 46(RR-10). 28. Sulzer CR, Jones WL. Leptospirosis: methods in laboratory diagnosis. Revised edition. Publication CDC Atlanta: US Department of Health, Education, and Welfare, Public Health Service, Kaufmann AF. Epidemiologic trends of leptospirosis in the United States, In: Johnson RC, ed. The biology of parasitic spirochetes. New York: Academic Press, 1976: Martone WJ, Kaufmann AF. From the Centers for Disease Control: leptospirosis in humans in the United States, J Infect Dis 1979; 140: Heath CW Jr, Alexander AD, Galton MM. Leptospirosis in the United States: analysis of 483 cases in man, N Engl J Med 1965; 273:857 64, Perrocheau A, Perolat P. Epidemiology of leptospirosis in New Caledonia (South Pacific): a one-year study. Eur J Epidemiol 1997; 13: Merien F, Perolat P. Public health importance of human leptospirosis in the South Pacific: a five-year study in New Caledonia. Am J Trop Med Hyg 1996; 55: Yersin C, Bovet P, Mérien F, et al. Human leptospirosis in the Seychelles (Indian Ocean): a population-based study. Am J Trop Med Hyg 1998; 59: Shenberg E, Gerichter CB, Lindenbaum I. Leptospirosis in man, Israel, Am J Epidemiol 1982; 115: Kuriakose M, Eapen CK, Paul R. Leptospirosis in Kolenchery, Kerala, India: epidemiology, prevalent local serogroups and serovars and a new serovar. Eur J Epidemiol 1997; 13: Watt G. Leptospirosis. In: Strickland GT, ed. Hunter s tropical medicine and emerging infectious diseases. 8th ed. Philadelphia: WB Saunders, 2000: Speelman P. Leptospirosis. In: Fauci AS, Braunwald E, Isselbacher KJ, et al., eds. Harrison s principles of internal medicine. 14th ed. New York: McGraw-Hill, 1998: Berman SJ, Tsai C-C, Holmes K. Sporadic anicteric leptospirosis in South Vietnam. Ann Intern Med 1973; 79: Ko AI, Galvão Reis M, Reibeiro Dourado CM, et al. Urban epidemic of severe leptospirosis in Brazil. Lancet 1999; 354: Park S-K, Lee S-H, Rhee Y-K, et al. Leptospirosis in Chonbuk Province of Korea in 1987: a study of 93 patients. Am J Trop Med Hyg 1989; 41: Trevejo RT, Rigau-Pérez JG, Ashford DA, et al. Epidemic leptospirosis associated with pulmonary hemorrhage Nicaragua, J Infect Dis 1998; 178: Faine S. Leptospirosis. In: Evans AS, Brachman PS, eds. Bacterial infections of humans: epidemiology and control. 3d ed. New York: Plenum, 1998: Dupont H, Dupont-Perdrizet D, Perie JL, et al. Leptospirosis: prognostic factors associated with mortality. Clin Infect Dis 1997; 25: Everard COR, Edwards CN, Everard JD, et al. A twelve-year study of leptospirosis on Barbados. Eur J Epidemiol 1995; 11: McClain JB, Ballou WR, Harrison SM, et al. Doxycycline therapy for leptospirosis. Ann Intern Med 1984; 100: Münnich D, Lakatos M. Treatment of human leptospirosis with Semicillin (ampicillin) or with Amoxil (amoxycillin). Chemotherapy 1976; 22: Guidugli F, Castro AA, Atallah AN. Antibiotics for treating leptospirosis. Cochrane Database Syst Rev 2000; 4:CD Petri WA Jr. Leptospirosis. In: Goldman L, Bennett JC, eds. Cecil textbook of medicine. 21st ed. Philadelphia: WB Saunders, 2000: Watt G, Padre LP, Tuazon ML, et al. Placebo-controlled trial of intravenous penicillin for severe and late leptospirosis. Lancet 1988; 1(8583): Chin J, ed. Control of communicable diseases manual. 17th ed. Washington, DC: American Public Health Association, Katz AR, Sasaki DM, Mumm AH, Escamilla J, Middleton CR, Romero SE. Leptospirosis on Oahu: an outbreak among military personnel associated with recreational exposure. Mil Med 1997; 162: State of Hawaii Department of Business, Economic Development and Tourism. The State of Hawaii data book: 1998, a statistical abstract. Honolulu: State of Hawaii, van Creel R, Speelman P, Gravekamp C, et al. Leptospirosis in travelers. Clin Infect Dis 1994; 19: Vinetz JM, Glass GE, Flexner CE, et al. Sporadic urban leptospirosis. Ann Intern Med 1996; 125: Centers for Disease Control and Prevention. Update: leptospirosis and unexplained acute febrile illness among athletes participating in triathlons Illinois and Wisconsin, MMWR Morb Mortal Wkly Rep 1998; 47: Centers for Disease Control and Prevention. Update: outbreak of acute febrile illness among athletes participating in Eco-Challenge-Sabah 2000 Borneo, Malaysia, MMWR Morb Mortal Wkly Rep 2001; 50: Campagnalo ER, Warwick MC, Marx HL Jr, et al. Analysis of the 1998 outbreak of leptospirosis in Missouri in humans exposed to infected swine. J Am Vet Med Assoc 2000; 216: Leptospirosis in Hawaii CID 2001:33 (1 December) 1841