TB in Captive Elephants in the United States

SOURCE: Diagnosis of tuberculosis in three zoo elephants and a human contact — Oregon, 2013. MMWR Morbid Mortal Wkly Rep 2016;64: 1398-1402.

Animal-to-human transmission of tuberculosis (TB) has been well-documented for a number of mammalian species, including deer, dogs, and even cats. Transmission of TB from circus elephants to humans occurred in the United States in the 1990s, and was reported in Infectious Disease Alert many years ago. It has since been recognized that TB is endemic in captive elephants in the United States (~5% are infected with TB based on truncal washings or necropsy), prompting the U.S. Department of Agriculture’s Animal and Plant Health Inspection Service in 1998 to develop guidelines for screening and diagnosis of TB in captive elephants, as well as contact precautions. These included recommendations for three consecutive truncal washings for AFB smear and culture for each elephant every year.

The problem is, no one really knows how to detect latent TB in elephants (imagine the skin test). The use of gamma-interferon assays is controversial and believed to result in too many false-positives. Nor is it clear how to effectively screen those with latent disease for active infection. Data for 2011 indicated that about 270 Asian and 220 African elephants were living in the United States, many of them at refuges for the old or infirm — and all of these are screened with trunk washings on an annual basis.

Evidence suggests, however, this approach may not be sufficiently sensitive for the detection of active TB. From 2004 to 2006, a non-profit reserve in Tennessee, which cares for retired or sick elephants within a 2700-acre reserve, received two Asian elephants with active TB and eight elephants with TB exposure. Infection control and treatment protocols were established, and the sick elephants were quarantined. Sadly, one the elephants died of TB, but the other was cleared of infection at 1 year of treatment.

The remaining elephants were trained to provide their own trunk washing samples (install 30-60 cc of saline, lift and lower, lift and lower, and exhale into a plastic bag) for regular screening. In addition, environmental samples from the elephants and their excrement were cultured. From 2006 to 2009, all trunk washings, as well as environmental samples, were negative. Despite these measures, 13 of 46 employees converted their PPD. Five were elephant caregivers, two were maintenance workers, and several were administrative staff who worked in an adjacent two-story building (which shared a ventilation system). Employees who converted their skin tests admitted to less rigorous use of N95 masks, in part because trunk washings from elephants in the main barn had been consistently culture-negative, yielding a general laxity regarding infection control (IC) precautions.

More recently, public health officials in Portland, OR, have confirmed an outbreak of TB in elephants at the Oregon Zoo, likely resulting in one human case of active TB and seven cases of latent infection, despite appropriate precautions. The investigation, which has been ongoing since 2013, focused on three bull elephants — including Packy, his son Rama, and Tusko (ages 51, 20, and 44 years, respectively) — each of whom developed active TB in 2013-2014. The first elephant TB case was detected in Rama, with a positive annual trunk washing in May 2013. This prompted zoo officials to increase the frequency of trunk washings in the other seven elephants, including monthly washings in those with positive TB antibodies and quarterly washings in those without TB antibodies.

Within 6 months, Rama’s father, Packy, who had positive antibiodies to TB, had a positive trunk culture. And in January 2014, a third male elephant was found with active infection. All sick elephants were appropriately quarantined and treated (elephants with TB are not euthanized).

An initial public health investigation of 19 close contacts began in 2013, and six new latent TB infections and one previously positive PPD were identified, none of whom had evidence of active infection. Close contacts were defined as persons with any presence within the 8300 square foot elephant barn or within 15 feet of any of the eight elephants in the enclosed outdoor space within the previous 12 months. Based on these results, the investigation was expanded to 59 casual contacts, including 20 people who had attended a special event where Rama painted a canvas by spraying paint with his trunk. None of these individuals were positive. Somewhat after the fact, an individual treated for TB in 2012 in Portland was recognized as having volunteered at the facility in 2012. This fact was not initially recognized, but a review of patients’ genotypes found no match for his isolate anywhere in Oregon. Whole genome sequencing at the CDC confirmed that the volunteer’s isolate and Rama’s isolate were a match, indicating likely transmission of TB between the two. While it seems more likely that transmission occurred from the elephant to the volunteer, the direction of transmission is still “up in the air.” Further screening of 18 other individuals who worked with elephants in 2012, before the outbreak was recognized, screened negative.

This information suggests that episodic excretion and transmission of TB from elephants is possible, despite culture-negative trunk washings. The practice of annual trunk washings for screening should be re-evaluated, and IC practices should be more strictly enforced for employees with contact with high-risk elephants.


TB Skin Testing and IGRA: An Ongoing Source of Confusion

SOURCE: Collins LF, et al. Diagnosis of latent tuberculosis infection: Too soon to pull the plug on the tuberculin skin test. Ann Intern Med 2016;164:122-124.

Variations in results between TB skin testing (TST) and gamma-interferon-release assays (IGRA) continue to stump providers, who are unsure which test to believe and how to proceed. It is increasingly recognized that TST and IGRAs are not uniformly correlated, and they measure different “arms” of the immune system response. To confuse matters further, the three IGRAs currently approved for use in the United States (the QuantiFERON-TB test, the QuantiFERON TB Gold In-Tube test, and the T-Spot.TB test) each employ different antigens, with different interpretative cut-offs, and the interpretative cut-offs used in published studies often differ from those used by the FDA for approval.

In low-risk populations, e.g., healthcare workers (HCWs), discordance between TST and IGRA results occurs at least 17% of the time. In household contacts with recent exposure, discordant results may occur up to 21% of the time. In part, this is related to the vagaries of IGRA testing in those at lower risk — 52% to 65% of HCWs at low risk for TB who had an initially positive IGRA result “reverted” to a negative result on repeat testing. For those with low-level positive test results, 75% to 80% reverted their test on repeat testing. For this reason, in those at low risk with low positive results, it is becoming commonly accepted practice to repeat a positive IGRA test in 2-8 weeks. In contrast, documented conversion of a TST is generally accepted as a true positive. In individuals at high risk, both tests can be used, and a positive for either test should prompt appropriate chemoprophylaxis.

Discordance seems to occur more commonly with newly exposed individuals and for those individuals with low-intensity exposure. While the generally accepted window period between exposure and a measurable immune system response is 8 weeks, some individuals may take longer to convert their skin test or IGRA. This may be the result of a dose-dependent relationship between the intensity of the exposure and the immune system response.

One Ugandan study found that while 41% of household exposures had skin test conversion by 3 months, an additional 10% of conversions occurred at 6 months or later. Compared to those with a baseline positive TST, those with TST conversions at 3 months or later had slowly increasing gamma-interferon production that took 3-12 months following exposure to reach detectable levels. In a group of soldiers with TB exposure, 78% had positive skin tests within 1 month, but only 63% had a positive IGRA at that time. Serial IGRA testing showed that conversions continued to occur up to 14-22 weeks after exposure.

On the basis of these studies, Collins and colleagues have proposed a hypothetical model for TST and IGRA responses. With high-intensity exposure, some individuals may have already developed a positive TST response with baseline testing (within 2 weeks after a known exposure), while few would have a positive IGRA. Over the next 2-10 weeks, conversion of the IGRA is delayed relative to the TST, although both tests will capture the majority of infected persons by 10 weeks post-exposure. However, with low-intensity exposure, conversion of both tests may be delayed beyond 10 weeks, with even further delay of the IGRA test compared with the TST.

Interpretation of IGRA results, therefore, may require not only risk stratification but consideration of exposure history. And skin testing still may prove to be more reliable and efficient than IGRA in evaluating household exposures or other exposures, especially low-intensity exposures.


LASIK, Humidifiers, and Mycobacterium chelonae Ocular Infections

SOURCE: Notes from the field: Mycobacterium chelonae eye infections associated with humidifier use in an outpatient LASIK clinic — Ohio, 2015. Morbid Mortal Wkly Rep 2015;64:1177.

Approximately 600,000 LASIK surgeries are performed annually in the United States. In February 2015, public health authorities were notified of the occurrence of four cases of eye infection secondary to M. chelonae in persons who had undergone LASIK surgery at an ambulatory surgery clinic in Toledo, OH. Two infections had occurred in six patients undergoing LASIK in January, and an initial investigation was unrevealing. Two additional infections occurred in February, prompting further investigation.

It turns out that the manufacturers of LASIK equipment recommend the use of humidified air to maintain 40-50% relative humidity during procedures. The clinic employed two consumer-grade cold air humidifiers for this purpose, which were filled with tap water and located near the patient procedures. The humidifiers each had an internal reservoir, and one provided a cold air mist. Samples taken from the misting humidifier reservoir were positive for M. chelonae, which was identical to three of four patient isolates and closely related to the fourth (> 95%) by pulsed-field gel electrophoresis.

Currently, the CDC environmental infection control guidelines prohibit the use of reservoir-style humidifiers in healthcare facilities, and state that only steam humidification should be used with appropriate precautions. The clinic subsequently upgraded its ventilation system to provide appropriately humidified central air.