By Carol A. Kemper, MD, FACP

Dr. Kemper reports no financial relationships related to this field of study.

Water Birth Death

Fristchel E, et al. Fatal legionellosis after water birth, Texas, USA, 2014. Emerging Infectious Diseases 2015;21(1):130-132.

Although statistics are few, home water births appear to be increasingly popular in the United States and the United Kingdom, and are making an appearance in Australia, France, and in other parts of the modern world. Celebrated in the 2003-2005 British television series “William and Mary,” Mary, who was herself a certified midwife in the U.K., delivers Martin Clune’s child  out of wedlock  and in a spa tub. Yet, concerns have been raised about the adequacy of the training for midwives and appropriate maintenance of equipment for home water deliveries.

This report documents the sad demise of a 6-day-old baby boy admitted to hospital in January 2014 with sepsis, respiratory failure, and loose stool in Texas. The infant immediately required extracorporeal membrane oxygenation. Initial cultures were negative for the usual organisms leading to post-partum sepsis, but astute critical care staff, knowing about the home water birth and suspecting something different, obtained tracheal aspirates and legionella urinary antigen, both of which were positive for Legionella pneumophila at day 4 of hospitalization. Despite aggressive measures, the infant died after 19 days of hospitalization. The nine-month pregnancy had been uncomplicated, and the infant had appeared to be a normal healthy baby boy.

Investigation by Texas state health authorities revealed several areas of concern. Well water was used to fill the home spa tub and, in preparation for the delivery, allowed to sit and circulate at approximately 37 degrees for two weeks prior to delivery. The water was exchanged two days prior to delivery. The water was treated with enzyme tablets but not chlorine. The tub was a recreational spa tub with internal piping and jets, rendering it difficult to sterilize. This type of tub is not licensed for use as medical equipment. The new mom was transferred to a regular bathtub following delivery, which was also filled with well water, and allowed to hold the infant.

Both well water and swabs of the tub failed to yield Legionella, although by the time the investigation was proceeding, the spa tub had been cleaned, disinfected, and put in storage. No comment was made in this report whether the home hot water heater or tub was tested.

Cases of Legionella related to home water baths and whirlpools may occur, although cases in infants are rare. Further, inadequately chlorinated home water spas and hot tubs may be a source for Pseudomonas, non-tuberculous mycobacteria, and fungi. I remember well a regional outbreak of pedicure-associated non-tuberculous mycobacterial folliculitis. Although the source for the outbreak was not ever identified, the whirlpool chairs used to soak feet were dismantled by our public health officer for inspection. Despite previous “cleaning,” the internal jets and piping were coated in slime, debris, and hair. I can just imagine what a spa tub used for home delivery might look like on the inside. Yikes.

This case led to more formal public health recommendations regarding the appropriate type and maintenance of equipment used for home deliveries, specific requirements for water treatment, as well as recommendations for education and training requirements for midwifery certification.

Out, Damned Spore!

Landelle C, et al. Contamination of healthcare workers’ hands with Clostridium difficile spores after caring for patients with C difficile infection. Infection Control and Hosp Epidemiol 2014;35(1):10-15.

Despite our best efforts as Infection Control staff, routine surveillance of all high-risk admission using PCR, strict isolation and contact precautions, private rooms, and amped up daily cleaning and terminal bleach cleaning of all C. difficile rooms, and a recent hand hygiene campaign, our hospital continues to experience a modest level of hospital-acquired C. difficile infection (CDI). While some of these cases were no doubt occult carriers on admission, and later developed symptoms of CDI in hospital, some patients likely acquired CD in the hospital  either from environmental contamination or from hands. Hands remain the most likely suspect.

Studies suggest that, even despite efforts to cleanse hands, both vegetative and spore forms of CD persist on hands of 14% to 59% of health care workers providing care to patients with CD. These spores may resist routine disinfection in the hospital and persist in the environment. These investigators developed a technique for destroying the vegetative forms of CD on hands, and recovering only the spores for a colony count determination.

Health care workers were observed at a 950-bed university hospital on six different wards (critical care, one surgical ward, and four medical wards) over a period of three months. Health care workers were divided into those who were providing care for a CDI patient and those who were not. Every contact was documented, and risk stratified by the duration of contact and the type of contact (high-risk activities were those involving exposure to feces such as bathing the perineum or diaper changes). A CD case was defined as a patient with diarrhea and a positive toxin test. All such patients were immediately placed in strict contact isolation in a private room until 48 hours after diarrhea had resolved. Infection control precautions included the use of gown and gloves on entry to rooms, alcohol gel before donning gloves, before any aseptic task, and hand washing with soap and water followed by alcohol gel after glove removal, as well as daily room cleaning with a hypochlorite solution.

Results of hand sampling for 66 “exposed” health care workers providing care to 7 CDI patients were compared with a control group of 44 unexposed health care workers, who provided care to 16 non-CDI patients. CD spores were recovered from 16/66 (24%) of the exposed group compared with none of the unexposed health care worker group. On average, two spores were found per positive hand (range, 1-6). For the exposed group, 30/386 (7.8%) of contacts with the patient and/or room were without gloves. Seven of the 16 health care workers with contaminated hands had at least one contact without gloves. This also means that 10.6% of health care workers failed to use gloves when caring for a CDI patient, despite every reason to do so.

In bivariate analysis, hand contamination was associated with high-risk activities, a longer duration of contact, a higher number of contacts, as well as contact without gloves.

Nursing assistants were more likely to have hand contamination (42%) than physicians (23%) or nurses (19%), consistent with the nature of their duties. Despite the fact that physicians had fewer high-risk contacts (only 4% of their exposures were considered high risk), they proportionately had a greater frequency of hand contamination. Logistic regression confirmed that two factors were highly correlated with hand contamination: high-risk activities and contact without gloves.

Infection preventionists know that, despite strict policies, signage, education, and ongoing surveillance, health care workers sometimes fail to don gloves when caring for a patient in isolation. This problem needs to be confronted in a more pecuniary way.