Five large counties from different parts of California (Alameda, Sacramento, Orange, Riverside, and San Diego) participated in the survey. Counties chosen served large populations and had multiple nail salons with whirlpool footbaths. In each county, a team including the regional investigator of the California Bureau of Barbering and Cosmetology and a local public health professional visited selected nail salons. They assessed footspa equipment, cleaning solutions, and cleaning techniques and frequencies. Swab samples were also collected.

In each participating county, a convenience sample of >3 different nail salons equipped with whirlpool footbaths located in the town’s main business section was randomly selected for the survey. Salon managers were questioned about cleaning and disinfection regimens of their footspas. Pedicure equipment time in service within the salon and make and model numbers of whirlpool pedicure equipment were noted. For each salon, 2 separate footspas were sampled, unless that salon only had 1 footspa, in which case only 1 swab was collected. Using a screwdriver, investigators removed the grate or filter screen covering the recirculation port in each footspa basin and inspected the area behind the screen for debris. A sterile, cotton-tipped culturette was used to swab this area and placed in standard transport medium.

At the California Microbial Disease Laboratory, each swab was removed from the transport medium, placed into a 50-mL tube containing 5 mL of sterile water, and the contents vortexed. The swab was then removed from the tube, and the remaining suspension was decontaminated with an equal volume of N-acetyl-L-cysteine-sodium hydroxide for 15 minutes, followed by neutralization with phosphate buffer and concentration by centrifugation (9). The sediment was spread onto Middlebrook 7H10 and Middlebrook 7H11/Mitchison 7H11 selective agar plates, Lowenstein-Jensen slants, Bactec 12B, and Bactec Mycobacteria Growth Indicator Tube 960 system (Becton-Dickinson, Sparks, MD, USA) liquid media.

The mycobacteria isolated were identified by >1 of the following methods: rapid DNA probes using nucleic acid hybridization (10), high performance liquid chromatography that produces mycolic acid patterns (11), and biochemical tests (9). M. simiae and M. lentiflavum were differentiated by urease activity and photochomogenicity; for both of these, M. simiae is positive and M. lentiflavum is negative (12). M. smegmatis group organisms were not differentiated to the species level. M. mageritense was identified by polymerase chain reaction restriction analysis at a Mayo Clinic laboratory and by DNA sequencing at the University of Texas Health Center in Tyler.

Thirty-one swabs were collected from 30 whirlpool footbaths in 18 nail salons from the 5 California counties. Twelve salons had >2 footspas; 6 had only 1 footspa. Of these 30 footspas, nontuberculous mycobacteria were cultured from 29 (97%). From 15 (50%), >1 mycobacterium species were isolated. No mycobacteria or other acid-fast organisms were isolated from 1 footspa that had only been in use for 11 days, whereas the positive footbaths had been in use for an average of 22 months (range 3–84 months).

Isolated from the whirlpool footbaths were 10 species of mycobacteria, 6 of which were RGM: M. fortuitum, M. mucogenicum, M. smegmatis group, M. mageritense, M. neoaurum-like RGM, and a pigmented unidentified nontuberculous mycobacteria (Table). M. fortuitum was the most frequently isolated mycobacterium, found in 14 (47%) of the 30 footspas surveyed and from all 5 counties. Rapid growers, including M. fortuitum, were found in 23 (76%) of the footspas. Slow-growing mycobacteria species were also recovered, including M. avium complex, M. gordonae, M. simiae, and M. lentiflavum. These species were less frequent than the rapid growers, except for M. avium complex, which was found in 5 (17%) of the footspas.

Mycobacterial species vary in their ability to survive the selective NaOH decontamination step that was used in this study (13). Some solid media cultures that grew M. fortuitum had only a few colonies of this species; others had nearly confluent growth; and still other cultures grew in broth only, not solid media, making it impossible to determine the quantity of growth. For these reasons, quantitative information about the number of colonies present on solid media is not reported.

The whirlpool footbaths sampled came from 3 manufacturers. Disinfectants reportedly used included a variety of brand name products and chlorine bleach, used at intervals of 1 to 14 days. Five (17%) footspas reportedly did not go through any disinfectant process. Twenty-five (83%) of the surveyed footbaths had collected visible debris or slime behind the recirculation screen cover, either on the screen itself, on the tub surface, or both. Fifteen (50%) of footspa operators reported never having cleaned behind this screen. One footspa had no screen or visible debris; nevertheless, it tested positive for mycobacteria.

Mycobacteria were isolated from virtually all pedicure spas surveyed, the sole exception being the footspa that had only been in service for 11 days. Mycobacteria were recovered whether or not disinfectants were reportedly used and whether or not debris was visible behind the recirculation screen.

RGM, M. fortuitum in particular, were the most frequently isolated mycobacteria. Our survey suggests that potentially pathogenic mycobacteria are widespread in these footspas across California. These organisms most likely were introduced into the footspas through the municipal water supply, where they colonized parts of the spas and probably the plumbing. Given that these whirlpool footbaths are widespread in California but similar infections known to date are rare, the presence of such mycobacteria alone may not be sufficient to cause pedicure customers to get cutaneous infections from using these spas. Our 2000 outbreak investigation noted an unusually large amount of debris behind the footspa recirculation screens, which might have provided a niche for mycobacteria to colonize and proliferate to large numbers. In that outbreak, customers who shaved their legs before using these implicated footspas were at higher risk for furunculosis than those who did not (1). However, some customers in that outbreak were infected even though they reportedly did not shave their legs before using the pedicure spas. Thus, while we documented the widespread presence of potentially pathogenic mycobacteria in footspas, the risk for infection remains unclear.

A limitation of this study is our inability to quantify the risk for cutaneous infection to pedicure customers despite finding widespread presence of RGM. We could not quantify reliably the amount of mycobacteria in each footspa with a positive culture. Furthermore, what we found in these footbaths may not be representative of other California counties or other states.

Nonetheless, our findings document the ubiquitous presence of potentially pathogenic mycobacteria among footspas of nail salons in California. The 2000 outbreak might have been a warning of what can happen again if this emerging infection is not adequately addressed. In 2004, a case report documented 2 cases of M. mageritense furunculosis associated with using footbaths at a nail salon in Georgia (14).

The California Board of Barbering and Cosmetology adopted new regulations in May 2001 requiring nail salons to follow specific cleaning and disinfection procedures to ensure that their footspa equipment is properly cleaned and maintained (15). Since our survey was conducted before these new regulations were implemented, further monitoring and research are needed to determine whether complying with the regulations will decrease the potential risk for mycobacterial cutaneous infections among pedicure customers.