The mechanism by which L. interrogans is transmitted sporadically in cities in the absence of occupational and recreational exposure is not known. We report the epidemiology of three confirmed cases of leptospirosis that occurred in Baltimore, Maryland, and we show a hitherto unrecognized population at risk for acquiring L. interrogans.

Methods

Top Methods Results Discussion Author & Article Info References

Our study was approved by the Joint Committee on Clinical Investigation of the Johns Hopkins Medical Institutions. Sprague-Dawley rats (Rattus norvegicus) (Jackson Laboratory, Bar Harbor, Maine) were used as controls for the polymerase chain reaction (PCR) and microagglutination assays. Feral rats (R. norvegicus) were trapped alive in the alleys where the patients probably became infected. The rats were anesthetized with methoxyflurane and ketamine before blood and tissue samples were obtained. Separate sterile surgical instruments were used to obtain specimens from the kidney and brain. Human serum and cerebrospinal fluid samples were obtained from clinical specimens taken for routine clinical purposes from patients suspected of having leptospirosis and from patients with unrelated illnesses (negative controls).

Polymerase Chain Reaction

Samples of DNA were extracted from the brains and kidneys of rats and from pellets of centrifuged human cerebrospinal fluid and serum using the QIAGEN QIAAmp Tissue Kit 29306 (QIAGEN, Chatsworth, California). Special precautions were used to avoid cross-contamination, which can cause false-positive results; these precautions included the use of aerosol-filtered pipette tips and different locations for DNA preparation and handling of PCR products. Polymerase chain reaction was done on L. interrogans by using a published primer set and cycling protocol [9]. Primers G1 (5' CTGAATCGCT GTATAAAAGT) and G2 (5' GGAAAACAAA TG GTCGGAAG) are specific for pathogenic leptospiraceae [9]. The L. interrogans serovar autumnalis (American Type Culture Collection, catalog 23476, Rockville, Maryland) was used as a positive control in the PCR experiments. The PCR mixtures were subjected to electrophoresis in 3% 3:1 NuSieve (FMC Bioproducts, Rockland, Maine) agarose gels in Tris-acetate-ethylenediaminetetraacetic acid buffer. Bands of DNA of the appropriate size were purified and cloned as described elsewhere [10]. The Biopolymer Laboratory of the Howard Hughes Medical Institute at the Johns Hopkins School of Medicine sequenced DNA with the dye-terminator method using an automated DNA sequencer (Applied Biosystems Model 373, Perkin-Elmer Corp., Foster City, California) with Taq polymerase (Boehringer Mannheim, Indianapolis, Indiana).

Serologic Testing

The presence of antibodies to L. interrogans was determined by the microagglutination test. Live spirochetes were used as antigen [11], and testing was done at the CDC. The panel of antigens included the following 20 serovars: ballum, canicola, icterohaemorrhagiae, bataviae, grippotyphosa, pyrogenes, autumnalis, pomona, wolffi, australis, tarassovi, georgia, alexi, cynopteri, mankarso, celledoni, djasiman, borincana, javanica, and bratislava. The serovars icterohaemorrhagiae and mankarso are members of the serogroup Icterohaemorrhagiae. The titers reported here are the highest dilution that produced agglutination of at least 50% of the leptospires relative to the phosphate-buffered saline control. Titers of 1:100 or greater were considered positive.

Results

Top Methods Results Discussion Author & Article Info References

Clinical Manifestations on Initial Presentation

Three patients with confirmed leptospirosis were seen in Baltimore, Maryland, between November 1993 and October 1995 (Table 1). All had abrupt onset of symptoms, which were typically fever and severe leg pain and weakness. Accompanying manifestations included meningitis, hemorrhagic diathesis, jaundice, acute renal insufficiency with nephritis, and thrombocytopenia.

Epidemiology and Potential Exposure to Leptospira interrogans

All three patients lived in inner-city Baltimore. None had recently traveled outside of Baltimore, had been exposed to farm animals, or had pets. After extensive questioning, the patients described clear exposure to feral urban rats. Patient 1 had walked barefoot in inner-city alleys where she had seen rats. She recalled stepping on glass 10 days before the onset of illness, and a healing laceration was seen on the plantar surface of her right foot. Patient 2 remembered cutting his hand on glass in a rat-infested alley about 10 days before the onset of illness. Patient 3 reported that he had walked barefoot in alleys near the harbors in downtown Baltimore, where he had previously seen rats.

Diagnosis

In all three patients, leptospirosis was confirmed by microagglutination testing. The results of leptospiral cultures done on specimens from the three patients were negative; all patients had received antibiotics before specimens were taken.

Cerebrospinal fluid from patient 1 (drawn on day 7 of illness) and serum from patient 2 (drawn on day 3 of illness) were analyzed by PCR (both specimens were obtained before the administration of antibiotics). Primers G1/G2 generated the predicted 285-base pair fragment specific for L. interrogans in the cerebrospinal fluid of patient 1 and the serum of patient 2 (data not shown). Serum from patient 3, taken several days after the initiation of antibiotic therapy, was negative by PCR. Cerebrospinal fluid and serum from five randomly selected patients who did not have leptospirosis (negative controls) were negative by PCR. Sequencing of the DNA in the positive bands confirmed that the DNA was from L. interrogans (data not shown).

Epidemiology of the Rat Vectors of Leptospirosis

Rats were trapped in the alleys that patients 1 and 2 reported were the likely location of their exposure to rat urine. Between January 1995 and March 1995, 18 rats were trapped from the alleys identified by patient 1; in April 1995, 3 rats were trapped from the alleys identified by patient 2. Polymerase chain reaction indicated that all 18 rats from the first site and 1 of the rats from the second site carried L. interrogans in the kidney or brain Table 2. Two laboratory-reared rats serving as negative controls were negative by PCR. Sequencing of the 285-base pair DNA fragment from several of these rats' tissues confirmed that the DNA was from L. interrogans (data not shown).

Microagglutination testing confirmed that the rats had been exposed to L. interrogans: Of the 18 rats trapped from the first site, 8 had antibodies to L. interrogans; the highest titer was against the serovar icterohaemorrhagiae (Table 2). Two laboratory-reared rats that served as negative controls were seronegative by microagglutination testing.

Discussion

Top Methods Results Discussion Author & Article Info References

In the United States, leptospirosis is rarely reported [12] and is typically related to occupational and recreational exposures [2-5]. The disease is thought to be underdiagnosed [2-6], probably because of the protean nature of its clinical presentations, the difficulty of distinguishing leptospirosis from other undifferentiated febrile illnesses, and delayed results from clinical testing. Seroepidemiologic studies suggest that unrecognized leptospirosis may be common in U.S. cities. Of 1150 patients from whom serum samples were drawn at a Baltimore sexually transmitted disease clinic for reasons unrelated to leptospirosis, 185 (16%) were positive for leptospiral exposure by enzyme-linked immunoassay; 1 had a substantial IgM titer that indicated recent infection [13]. A study from Detroit [14] showed that inner-city children had a significantly higher incidence of antileptospiral hemagglutinating antibodies than did suburban children, yet none of the inner-city children had received a diagnosis of leptospirosis. Unrecognized urban cases are probably related to exposures to rat or dog urine.

We described three sporadic urban cases of leptospirosis. These patients acquired L. interrogans at different times and locations in an inner city; the only apparent risk factors were probable percutaneous exposure to rat urine in alleys. This mode of transmission is consistent with the fact that rats are known carriers of L. interrogans.

The patients we describe had complex systemic illnesses typical of leptospirosis. The most notable aspect of these cases—the presence of the disease in urban residents—created confusion among the physicians caring for the patient: Until the diagnosis was established, most of the clinicians caring for the first patient were unaware that leptospirosis could develop in urban environments. The first case was correctly diagnosed by the house officers caring for the patient on the day of admission on the basis of the typical clinical and laboratory features of leptospirosis. Because the attending physician and several faculty consultants (specialists in infectious disease, nephrology, and hepatology) rejected the diagnosis of leptospirosis in favor of possible diagnoses related to the patient's sexual and substance abuse behaviors, the patient had extensive testing, including liver biopsy. Because of the institution's experience with the first patient, the second and third cases were correctly diagnosed early in the course of illness and were treated appropriately. A fourth case of infection with urban-acquired L. interrogans was reported to the Maryland State Department of Health (and confirmed by microagglutination testing at the CDC) late in October 1995: A 47-year-old homeless man acquired L. interrogans from swimming in the Jones Falls, a waterway in central Baltimore (Goodman J. Personal communication).

We showed by PCR and serologic testing that most of the rats trapped in alleys where the patients reported their probable exposures were infected with L. interrogans. Rat populations in Baltimore and other large cities probably have a high prevalence of L. interrogans infection. No previous study has shown that persons whose only risk factor for acquiring L. interrogans is living in an inner city are at risk for acquiring the organism. Thus, the epidemiology of human infection by L. interrogans may include sporadic contact with rat urine; this could also represent one of the unidentified risk factors for leptospirosis in urban dwellers suggested by an Italian study [6].

We also found that PCR is useful for early, rapid diagnosis of leptospirosis. The first two patients in our study had PCR-detectable L. interrogans DNA in body fluids early in the course of illness. With appropriate clinical suspicion, PCR can be used to establish a diagnosis early so that specific therapy can be initiated and invasive and expensive testing can be avoided. The use of PCR to diagnose of leptospirosis may eliminate the need for microagglutination testing and culture, which are cumbersome and delay diagnosis. However, before PCR can be adopted as the diagnostic method of choice, further prospective study is needed [15].

In our study, more rats were positive for leptospiral infection by PCR than by microagglutination testing, suggesting that some PCR results could have been false-positive because of cross-contamination by amplified leptospiral DNA. However, a similar discrepancy between rat seronegativity and culture positivity of rat tissues has been reported previously and suggests that rats chronically infected with L. interrogans can become seronegative. Feigin and colleagues [16] trapped rats while investigating an outbreak of leptospirosis in humans and canines. Of nine rats that were positive by culture of either brain or kidney, only three were positive by microagglutination assay. Other studies have had similar results [17, 18]. These observations, together with our data indicating that the kidney and brain specimens obtained from negative control rats were negative for L. interrogans DNA by PCR, argue strongly for the validity of the PCR assay used in our study.

Recent changes in the recognized patterns of infectious disease epidemiology have raised public awareness of emerging diseases (such as Lyme borreliosis, the hantavirus pulmonary syndrome, and ehrlichiosis) [19] and reemerging diseases (such as bartonellosis) [20, 21]. As with the recent cases of trench fever in urban homeless persons and alcoholics [20, 21], newly applied diagnostic methods can alert clinicians and public health officials to previously unrecognized diseases [22]. Many inner cities continue to decay socially and economically, which allows the establishment or expansion of reservoir hosts of diseases (such as rats) and potentially leads to the establishment of an endemic substrate for the spread of disease. Our results suggest that leptospirosis may be yet another reemerging infectious disease in urban areas of the United States.

Presented in part at the American College of Physicians' Annual Session, Associates' Poster Competition, March 1995.

Dr. Glass: Department of Molecular Microbiology and Immunology, Johns Hopkins School of Public Health and Hygiene, 615 North Wolfe Street, Room 4307-A, Baltimore, MD 21205.

Dr. Flexner: Johns Hopkins School of Medicine, 600 North Wolfe Street, Osler 524, Baltimore, MD 21287.

Dr. Mueller: Mayo Clinic, 200 First Street SW, Rochester, MN 55905.

Dr. Kaslow: National Institutes of Health, Building 4, Room B1-31, Bethesda, MD 20892-0425.