To the Editor: Mycoplasma spp. are well-known pathogens in human and veterinary medicine. Mammals, especially primates and including humans, share similar or even identical Mycoplasma spp., which might be commensal or pathogenic (1). Additionally, sporadic infections of immunocompromised persons with Mycoplasma spp. that originated from domestic animals have been reported (1); susceptibility in this human population is increased (2,3). M. phocicerebrale is the only Mycoplasma pathogen of animals that regularly infects humans, causing a disease called seal fingers (1,4). However, we report a human infection with an avian Mycoplasma organism.

A clinical trial to investigate the capability of M. lipofaciens (strain ML64) (5) to spread horizontally between infected and noninfected turkey poults in an incubator demonstrated airborne transmission of the pathogen within 24 hours (6). During the trial, the veterinarian conducting the study, a 36-year-old man, was monitored for infection. Each day, 2 swabs were taken from both nostrils, starting from the day before the infected poults hatched (day 0) through day 7 after the poult hatching date. When handling eggs and poults, the veterinarian wore gloves but not a protective mask. Two days after the poults hatched (day 3), the veterinarian reported throat pain and a slight rhinitis, which indicated a respiratory disease. The next day only the rhinitis with minor nasal pain was present.

One nasal swab from each sampling day was used for Mycoplasma spp. culture (5). Isolated Mycoplasma organisms were subjected to an immunobinding assay (6) with antiserum against M. lipofaciens, M. buteonis, M. falconis, M. gypis, M. gallisepticum, M. meleagridis, M. synoviae, and M. iowae, selected because the veterinarian regularly handled these isolates. The second nasal swab from each sampling day was taken to detect Mycoplasma DNA by PCR (5). The samples for PCR testing were stored at –20°C and tested only when attempts to isolate Mycoplasma spp. failed.

Six weeks after the infection, a serum sample from the patient was examined for specific antibodies against M. lipofaciens by 3 different methods: modified immunobinding assay (6) (which used patient’s serum and peroxidase-conjugated goat-antihuman serum [STAR90P, Serotec Ltd, Oxford, UK]), a growth inhibition test, and antibody titers of serum dilutions. The first 2 methods used M. lipofaciens (strain ML64) from a turkey trial (6) and the reference strains of M. buteonis (Bb/T2g) and M. falconis (H/T1). For the third method, to determine the titer, 2-fold serial dilutions of the patient’s serum samples were prepared and incubated with a bacterial suspension (M. lipofaciens strain ML64) containing 3.2 × 10² CFU. The antibody titer was based on the highest serum dilution capable of reducing 50% of the mean CFU count.

Before the infected poults hatched, attempts to isolate Mycoplasma organisms and demonstrate Mycoplasma DNA from the nasal swabs were unsuccessful. However, from the day of hatching (day 1) until 3 days later (day 4), Mycoplasma organisms were isolated from the nasal swabs and identified as M. lipofaciens. On day 5, only Mycoplasma DNA was demonstrated (Table). Specific antibodies against M. lipofaciens (strain ML64) were detected by using the immunobinding assay and growth inhibition test. Antibodies against other Mycoplasma spp. were not detected. Antibody titer against M. lipofaciens was 128.

M. lipofaciens have been reported from a chicken, turkey, duck (7,8), and a raptor egg (5). Strain ML64 is highly pathogenic for chicken and turkey embryos and can be transmitted by air (6,9). The veterinarian handling the infected poults was free of nasal Mycoplasma organisms a day before contact. His infection occurred concurrent with demonstration of airborne transmission among poults. The isolation of M. lipofaciens from his nares for 4 days demonstrates the infectivity and reproductive capability of this Mycoplasma strain in humans: as a pure contaminant, isolation for several days would be unlikely. Christensen et al. (10) have reisolated different avian Mycoplasma strains (M. gallisepticum, M. synoviae, M. iowae) from a human nose, from 12 hours through 1 day after artificial infection, demonstrating differences between avian strain abilities to survive on human mucosa. M. lipofaciens invasiveness for humans is underscored by finding specific antibodies against this species 6 weeks after infection. Cross-reactivity to other Mycoplasma spp. cannot be excluded but seems unlikely.

This study suggests that M. lipofaciens (strain ML64) can be transmitted successfully to humans and may cause clinical symptoms; the study documents nonartificial human infection with an avian Mycoplasma sp. These findings should be considered especially for humans highly susceptible to Mycoplasma infections, including children and persons with congenital or acquired immunodeficiencies (2,3).