American Cutaneous Leishmaniasis in U.S. Travelers

Methods

Persons were eligible for the study if 1) they were civilian U.S. travelers [that is, U.S. citizens or permanent residents, or persons who had lived in the United States for at least 6 months immediately before traveling]; 2) they acquired localized cutaneous leishmaniasis [that is, not diffuse cutaneous or mucosal leishmaniasis] in Mexico or Central or South America; and 3) SSG was released for them by the CDC Drug Service between 1 January 1985 and 30 April 1990 (in one case, the SSG was not used). Sodium stibogluconate is released under an investigational new drug protocol requiring informed consent from patients. In the spring of 1990, we reviewed CDC Drug Service records and conducted a telephone survey of the study participants. Detailed interviews lasting at least 30 to 60 minutes were conducted by the same person (BLH). In two instances, a relative (mother, ex-wife) was interviewed. The study was approved by CDC's institutional review board and the patients were informed that participation was voluntary.

Results

General Information

During the study period, SSG was released by the CDC Drug Service to treat 129 civilians with leishmaniasis. Most of these persons (105 of 129 [81%]) had localized cutaneous leishmaniasis, which was acquired most often in the Americas (69 of 129 patients [53%]). Of the 69 patients with ACL, 59 (86%) were eligible for the study; 58 (98%) of the 59 eligible patients were interviewed a median of 2.6 years (range, 4 months to 5.8 years) after their return to the United States. They typically had vivid memories of their experiences and commonly refreshed their memories by checking itineraries and diaries. In the case of the one traveler who could not be found even by his family, some relevant information was obtained from his physician.

The median age of the 59 study patients at the time of travel was 34 years (range, 3 to 64 years); 40 (68%) were male. Overall, they acquired ACL in as many as 14 countries (Figure 2); at least 33 patients (56%) acquired ACL in Mexico or Central America. Table 1 provides estimates of the numbers of U.S. travelers who, during the study period, visited the countries where the study patients acquired ACL. The estimates may be most useful for contrasting the extremes: About 1 per 1000 U.S. travelers in Suriname acquired ACL that was treated with SSG in the United States, whereas the rate for U.S. travelers in Mexico was less than 1 per million.

Of the 59 study patients, 27 (46%) were conducting field studies or school projects. These travelers, who will be referred to as researchers, acquired ACL in up to 10 countries, but 8 (30%) worked in Peru. Eleven (19% of the 59 travelers) were studying birds. The researchers' activities commonly were nocturnal and included banding birds, catching alligators and bats, studying frog communication, studying biologic diversity, analyzing rock for gold, and working in a refugee camp. At least four of the researchers worked in areas in which forest was being or recently had been cleared. Of the 19 travelers (32%) who were tourists or visitors, 4 were in organized groups touring Peru (3 tourists) or Belize (1 tourist); at least 11 of the 15 who were not in organized groups acquired ACL in Mexico, Guatemala, or Belize. The other 13 travelers included 4 tour guides (7%), 2 of whom were scientists; 4 missionaries (7%); 3 nature photographers (5%), 2 of whom became infected 1 month apart in Suriname; and 2 landowners (3%).

The 58 interviewees were outside the United States for a median of 54 days (range, 4 days to 3 years): the 19 tourist-visitors for a median of 20 days (range, 4 to 182 days) and the 27 researchers for a median of 76 days (range, 14 to 230 days) (P = 0.003, Wilcoxon two-sample test). Twenty-four persons (41%) first noticed skin lesions while overseas, whereas 10 (17%) did not notice lesions for at least 1 month after returning. The travelers had spent an estimated median of 26 days (range, 0 days to 6 months) in forested areas before first noticing skin lesions. At least 15 persons (26%) had such exposures for 1 week or less, with at least 6 persons (10% of the 58 interviewees) having a maximum exposure of 2 days. For these 15 persons, the maximum possible incubation period ranged from less than 1 day (in persons in whom the wound from the sand-fly bite was noticed and reportedly evolved directly into a persistent skin lesion) to 5 months (median, 30 days). The only person who never ventured into a forested area stayed 9 days in an unscreened home in a rural area in Belize. A tourist in Belize vividly recalled being bitten (where her lesion developed) on her first day in a forested area (Altunha). A tourist in Guatemala became infected during the 1 day (no nights) he spent exploring Mayan ruins in Tikal National Park. On the other hand, at least 10 travelers (17%) had taken many previous trips to forested areas in the Americas before developing clinically evident ACL. A conservationist who rarely used insect repellent had spent 2 to 9 months per year for 20 years in developing countries before acquiring ACL, which was attributed to a 3-week sojourn in Suriname.

Clustering

Some destinations were common to multiple travelers. Eleven of the 12 who traveled in Peru sojourned in forested areas in or near Manu National Park or Puerto Maldonado. Of the 12 travelers who may have acquired ACL in Guatemala, at least 9 visited Tikal National Park. Two researchers became infected 1 year apart at the same field site in Panama, but 13 others who worked at the site on one to four occasions during a 4-year period did not develop clinically evident ACL.

Persons traveling together developed ACL; the 59 study patients represent 55 trips. Two tourists traveling together in Mexico, Guatemala, and Belize acquired ACL. Two of three family members doing everything together for 9 days in Costa Rica developed ACL; the one who did not develop skin lesions spent hours shirtless in a forested area with his brother, who did acquire ACL.

Two college students acquired ACL during a 2-month stay in Guatemala and Belize. Three of the 10 accompanying students also noticed at least one persistent skin ulceration, but only 1 student had his lesion evaluated (no parasites were found in a skin biopsy specimen). Therefore, the attack rate of ACL in this group of 12 students was at least 17% to 42%.

Two persons in a group touring Peru acquired ACL despite spending less than 48 hours in an ACL-endemic area upriver of Puerto Maldonado in the Amazon basin. Their lesions were on skin areas that were uncovered only after sundown while they reclined in hammocks on unscreened porches in a jungle clearing. The patients' spouses, who spent less time (none or half as much) in hammocks, did not note lesions. One of the nine other tourists in the group developed a skin nodule that persisted at least 6 months but was not evaluated. Therefore, the attack rate of ACL in this group of 13 tourists was at least 15% to 23%. The tour agency had sent about 650 persons on similar tours during a 4-year period, but most had stayed at a different lodge.

Overall, 18 of the 58 interviewees (31%) reported that at least one expatriate with whom they had traveled or worked overseas had acquired ACL. In addition to other study participants, they knew at least nine Americans not referred to above and at least nine expatriates from other countries who had acquired ACL during the same period and in the same geographic area as they had. A geologist who analyzed rock for gold in a 1-month-old jungle clearing in French Guiana said that five coworkers had acquired ACL, including one who was in the area only 1 week, for an attack rate among expatriates working in this area of 67%; the five coworkers reportedly were treated in France.

Knowledge of Leishmaniasis and Use of Protective Measures

Thirty-one of the 58 interviewees (53%) had heard of leishmaniasis before traveling, but only 1 attributed such knowledge to pretravel medical advice. Only 21 of these 31 persons had also known they were at risk for leishmaniasis, with 17 (29% of the 58 interviewees) also aware of measures that could decrease that risk; 16 of these 17 patients were researchers or, in one instance, a scientist tour guide. Of the 27 travelers who had not heard of leishmaniasis, 17 had received pretravel medical advice.

Of the 51 travelers who remembered whether they had used insect repellent, 9 (18%) had never used it; only 1 of these 9 persons was 1 of the 17 persons most knowledgeable about leishmaniasis. Some had mistakenly judged their need for repellent by whether mosquitoes were plentiful, and others had been concerned it might affect their experiments or equipment. Forty-three of the 58 interviewees (74%) had sometimes slept outside or in unscreened rooms in rural areas; 21 of these 43 (49%) had never used bednets; this group of 21 included 6 of the 9 patients who had never used repellent.

All lesions of 37 patients (63%) developed on areas of the body that would have been covered by long-sleeved shirts, long pants, socks, and shoes (Figure 3). Some persons who developed arm lesions had worn long-sleeved shirts but had rolled up the sleeves, and others developed lesions on areas of skin that were uncovered only while they bathed outside at night. A researcher who conscientiously applied personal protective measures developed a lesion where she had had a hole in her slacks.

Delays in Diagnosing Leishmaniasis and Initiating Treatment

For the 58 interviewees, the temporal relations among events leading up to the diagnosis and treatment of ACL are shown in box plots (Figure 4). A median of 112 days (range, 13 to 1022 days) elapsed from when they first noticed their lesions until CDC released SSG. This period was a median of only 55 days (range, 13 to 144 days) for 13 persons (22%) who were researchers or, in one instance, a scientist tour guide and who were unusually knowledgeable about leishmaniasis; they not only had heard of it before traveling, knew of their risk, and were aware of preventive measures but they also were the first to consider and even suggest the diagnosis to their physicians. Overall, 28 persons (48%) claimed that they or a relative or acquaintance were the first to consider the possibility of ACL; 2 of these persons got this idea while perusing textbooks. At least 25 of the 28 suggested to their physicians that ACL be considered, but 7 reportedly were received skeptically.

In 30 of the 58 cases (52%), medical professionals were the first to consider the diagnosis of ACL; their specialties were dermatology (for 14% of 58 patients); infectious disease or tropical medicine (12%); pathology (12%); internal medicine, pediatrics, or general or family practice (7%); unknown (3%); laboratory technology (2%); or occupational medicine (2%). Overall, the 58 interviewees consulted a mean of 2.1 physicians (range, 1 to 7 physicians), taking into account the physician who, in each case, diagnosed ACL; 19 patients (33%) consulted at least 3 physicians. (Pathologists and clinicians to whom patients were referred for second opinions or diagnostic procedures were not counted separately from the referring physician.) The 13 patients unusually knowledgeable about leishmaniasis consulted a mean of 1.7 physicians (range, 1 to 3 physicians). Whereas researchers consulted a mean of 1.9 physicians (range, 1 to 4 physicians), tourist-visitors consulted a mean of 2.5 (range, 1 to 7 physicians) (P = 0.1, Wilcoxon two-sample test). The 58 patients ultimately were treated with SSG by 45 physicians, 1 of whom treated 9 patients and was consulted about 4 others (9 of these 13 patients were researchers or, in one instance, a scientist tour guide).

Being a knowledgeable patient and consulting a knowledgeable physician were complementary. Seven of the 13 patients who were unusually knowledgeable about leishmaniasis were among 16 patients (28% of the 58 interviewees) for whom SSG was released at most 60 days (median, 48 days) after they first noticed lesions; these 16 patients, 11 (69%) of whom were researchers or, in one instance, a scientist tour guide, generally consulted physicians exceptionally knowledgeable about infectious and tropical diseases. In fact, 7 of the 16 patients consulted authorities on leishmaniasis; for 2 others of the 16, ACL was diagnosed while they were still overseas.

In addition to the time spent being shuttled among physicians, delays of variable duration occurred because of treatment of alternative diagnoses (50% of patients), patient convenience (36%), the wait to return home (16%), alternative treatment of ACL (9%), and physician convenience (5%). Other diagnoses that were considered included bacterial or fungal infection, syphilis, skin cancer, pseudolymphoma, insect bite, and auricular chondritis. A patient whose ACL was diagnosed within 3 months of when she noticed lesions began SSG therapy 2.5 years later, after moving, changing physicians, and receiving two treatment trials of ketoconazole.

Fifty-seven patients remembered the appearance of their lesions when they first consulted medical personnel; the 41 patients who apparently had classic ulcerative lesions and the 16 patients with persistently nodular or otherwise atypical lesions had similar median periods from first awareness of lesions to release of SSG (132 compared with 118 days; P = 0.8, Wilcoxon two-sample test). Because we did not have medical records from all consulted physicians, we could not quantify the delays that occurred because of suboptimal diagnostic strategies. However, whereas few patients mentioned having had tissue scrapings or aspirates obtained, at least 48 recalled having had at least one skin biopsy; these included at least 8 patients who had excisional biopsies. For at least six patients (10%), parasites were initially overlooked in biopsy specimens and were later noted on re-examination. One such patient consulted six physicians, endured two skin grafts, and paid $6600 in medical bills before ACL was considered. Six months after he consulted his first physician, a woman from his tour group in Peru who also had acquired ACL suggested he might have ACL. He then consulted his seventh physician, who had slides from a tissue specimen obtained 6 months earlier reviewed again; previously overlooked parasites were noted.

Discussion

Although the clinical courses of various study patients have been reported [14, 15, 20, 21], our study is the first systematic epidemiologic investigation of ACL in civilian U.S. travelers. Not surprisingly, researchers with extended stays and intensive nocturnal exposures in forested areas acquired ACL; perhaps ACL should now be called ornithologist's ulcer rather than chiclero's ulcer. Substantial attack rates for persons with intensive exposures have been reported previously [3, 22]. Tourists and other study patients with brief exposures in ACL-endemic areas also developed ACL, but this also has precedent [8]. Although the average tourist may be at less risk for ACL than the average researcher, tourists have the disadvantages of being less likely to have heard of leishmaniasis, to know others who have had ACL, and therefore to have established networks for receiving expeditious medical care. They also may be less likely to associate their skin lesions with their sojourns, especially if they travel for a short time and do not notice lesions until weeks or months after returning home.

Because many U.S. physicians are unfamiliar with leishmaniasis, they may not consider ACL even if their patients have traveled to ACL-endemic areas and have classic lesions. Perhaps not all study patients told their physicians where they had traveled, but some physicians reportedly were reluctant to consider ACL even when it was suggested. Practitioners in the United States commonly obtain biopsy specimens from skin lesions of unknown cause, but this is not the optimal way to diagnose ACL. Although histopathologic examination is useful for excluding other diagnoses, its sensitivity for ACL reportedly is only 14% to 18% [23, 24]; finding scarce amastigotes (the tissue form of the parasite) and identifying their characteristic extranuclear kinetoplast are difficult tasks. If a skin biopsy specimen is obtained, a Giemsa-stained impression smear should also be examined [25], and some of the biopsy specimen should be cultured for Leishmania organisms; DNA [26] and monoclonal antibody analyses [27] and animal inoculation can also be considered. Culture medium (Novy-MacNeal-Nicolle) is available through the Parasitic Diseases Branch, CDC (telephone: 404-488-4050). We recommend obtaining Giemsa-stained thin smears of dermal scrapings of ulcerative lesions and cultures of lesion aspirates [23, 24]. Serologic tests should not be relied on for a definitive diagnosis of ACL.

Although localized cutaneous leishmaniasis generally does not result in as much morbidity as mucosal leishmaniasis or in as many deaths as visceral leishmaniasis, delays in diagnosing ACL and initiating therapy can have consequences. These include increased difficulty in diagnosing ACL in patients with older lesions [16, 24], bacterial superinfection, lesion growth, unsightly scarring, development of mucosal leishmaniasis, and added frustration and expense for the patient. Whether and with what antileishmanial agent to treat ACL [16, 17, 28] can be discussed with a physician of the Parasitic Diseases Branch, CDC. Excisional biopsy should not be considered a form of therapy.

Preventing ACL is clearly preferable to treating it. Many study patients did not recall having heard of ACL despite receiving pretravel medical advice. Although chemoprophylaxis and vaccines for ACL are not available, many study patients' lesions might have been prevented if the patients had consistently minimized the amount of exposed skin. Unfortunately, this measure is often impractical during long stays in steamy climates [29]. Because all study patients developed ACL, we could not evaluate whether rudimentary knowledge of leishmaniasis and consistent use of personal protective measures (for example, protective clothing, repellent, and screening and bednets of sufficiently fine mesh to keep out sand flies, which are about one third the size of mosquitoes) decrease the risk for ACL. However, some study patients noted that companions with essentially the same exposures and practices did not develop clinically evident ACL. Although some such companions may have had subclinical infections, others may have had the good fortune of avoiding sand flies, which have a limited flight range, or of being bitten only by uninfected flies.

The number of study patients (n = 59) during the 5.3-year study period represents about one thousandth of the estimated annual cases of ACL (n = 59 300) [1] and an unknown proportion of all cases of ACL in U.S. travelers. Although the problem of ACL in U.S. travelers is small compared with the global problem of leishmaniasis, our study highlights problems that arise when tropical diseases are imported into the United States. The limited knowledge of tropical medicine among U.S. physicians is relevant not only for travelers with ACL but also for persons with other leishmanial syndromes (for example, soldiers from Operation Desert Storm who contracted viscerotropic leishmaniasis [30]) and other tropical diseases; this limited knowledge affects not only the health of persons wealthy enough to travel for business or pleasure but also that of immigrants and migrant workers from developing countries. The medical problems of travelers returning from the back of the beyond should heighten our awareness of the need to enhance the capability of physicians in developed countries to prevent and treat tropical diseases.