We collected demographic and epidemiologic information to describe characteristics of the 9 reported patients infected with C. fetus subsp. testudinum subsp. nov. Food preferences and limited information about exposures were available for 5 patients. Four patients could not be interviewed because they could not be located or had died. However, some information was available from their original case reports. We summarize our findings in the Table.

Patients resided in Colorado, Louisiana, Massachusetts, and New York, and had onset of illness during 1991–2010. All patients were men (median age 73 years, range 20–90 years). Five of 6 patients were of Asian origin (4 were Chinese and 1 was either Chinese or Vietnamese), and the non-Asian patient had a Chinese spouse. Last names of the remaining 3 patients did not suggest that they were of Asian origin.

C. fetus subsp. testudinum subsp. nov. was isolated from blood (4 patients), feces (2 patients) pleural fluid (1 patient), hematoma (1 patient), and bile (1 patient). Of 5 patients with available information, all had underlying illness. Clinical symptoms varied. One patient had fever, cough, and epigastric pain; another had fever, chills, rigors, cough, and diarrhea; and a third had bloody feces, pulmonary edema, and pleural effusion. One patient sought care for dizziness and mental confusion after a fall, and C. fetus subsp. testudinum subsp. nov. was isolated from a subdural hematoma. For another patient, C. fetus subsp. testudinum subsp. nov. was isolated from blood after cellulitis developed from a leg wound; no gastrointestinal symptoms were reported. All 6 patients for whom outcomes were available were hospitalized, and 1 died of leukemia.

All 5 patients of Asian origin and the 1 patient with an Asian spouse shopped or ate at restaurants in Chinese (Chinatown) areas in Massachusetts and New York. A limited food and travel history was available for 4 patients. All 4 reported eating traditional Chinese dishes. In addition, 1 patient ate eel, 1 ate eel and frog, 1 ate turtle soup, and 1 denied eating turtle or frog. Three patients did not report any recent travel, and 1 reported frequent travel, including trips to Europe and Hong Kong. Food and travel histories were not available for the non-Asian patients. However, 1 patient reported contact with a turtle that had diarrhea. This patient did not appear to have had a systemic infection; his isolate was obtained from feces, and he reported a 16-day history of diarrhea.

C. fetus subsp. testudinum subsp. nov. is a newly proposed subspecies that appears to have originated in reptile species. Although information is limited, our data suggest that humans may contract this subspecies though exposure to reptiles, possibly by ingestion or by contact with feces or the environment. Reptiles, particularly small turtles, are a well-known source of Salmonella spp. infections in humans (11). A recent study in Taiwan reported C. fetus subsp. testudinum subsp. nov. in feces of 12 (6.7%) of 179 reptile feces samples; prevalence was highest in turtles (10 [9.7%] of 103) (12). Turtle is an ingredient in some traditional Asian dishes and turtles are frequently sold in Asian specialty markets; this association might partly explain the predominance of Asian race among reported patients. A review of GenBank found 5 submissions of 16S rDNA sequences representing C. fetus subsp. testudinum subsp. nov. from China (accession nos. DQ997044, HQ450384, HQ681195, JN585921, and JN585922).

Although C. fetus is more common among men (13,14), it is unusual that all of the recognized patients infected with C. fetus subsp. testudinum subsp. nov. were men. This finding might indicate that men have a predisposition to infection with this subspecies or that they are more likely to be exposed to sources of contamination. Nearly all patients with C. fetus subsp. testudinum subsp. nov. infections were >60 years of age or had immunocompromising conditions. This finding indicates that C. fetus subsp. testudinum subsp. nov., like C. fetus subsp. fetus, are opportunistic pathogens that might lead to severe disease. Most patients had primary bacteremia; only 4 of 7 patients had diarrhea associated with their illnesses.

The actual number of C. fetus subsp. testudinum subsp. nov. illnesses is unknown. C. fetus infections are likely to be underdiagnosed and underreported. C. fetus is susceptible to cephalosporins, which are commonly included in media used for isolation of Campylobacter spp., making C. fetus isolation from feces unlikely. Campylobacter spp. infection is not a nationally reportable disease in the United States, and in most states, isolates are not routinely sent to state public health laboratories for confirmation of identification. Although some clinical and state laboratories identify Campylobacter spp., this identification is not conducted routinely, and few laboratories use molecular methods to identify strains to the species level. For C. fetus subsp. testudinum subsp. nov., additional identification methods have to be performed. Therefore, we encourage laboratories that identify the Campylobacter spp. to forward isolates of C. fetus to the Campylobacter Reference Laboratory at the Centers for Disease Control and Prevention for confirmation and screening for C. fetus subsp. testudinum subsp. nov. In addition, when interviewing persons with Campylobacter spp. infections, public health personnel should ask about exposure to live reptiles and traditional Asian dishes made with turtles or other reptiles.

In summary, our data show that C. fetus subsp. testudinum subsp. nov. can cause invasive infection. All known cases have occurred in men, most of whom were of Asian origin, and infection may be related to exposure to traditional Asian foods or reptiles. Persons who are immunocompromised should avoid eating undercooked reptiles, exposure to live reptiles, and their environments. Enhanced public health surveillance and laboratory testing and surveys of the prevalence of the organism in reptiles are needed to better understand the epidemiology and incidence of C. fetus subsp. testudinum subsp. nov. and to recommend additional prevention measures.