Using Molecular Characterization to Support Investigations of Aquatic Facility–Associated Outbreaks of Cryptosporidiosis — Alabama, Arizona, and Ohio, 2016

Alabama

On August 12, 2016, the Alabama Department of Public Health received a report of 35 persons who developed gastrointestinal symptoms after visiting an Alabama aquatic facility. Case-finding efforts identified 23 outbreak-associated cases. Three (13%) patients had laboratory-confirmed Cryptosporidium infection; molecular characterization by CryptoNet of one Cryptosporidium specimen identified it as the C. hominis IfA12G1R5 subtype. Data collected using an outbreak-specific questionnaire completed for 15 patients indicated the median incubation period was 8 days (range = 5–17 days) after visiting the aquatic facility on July 31. The limited number of completed questionnaires provided insufficient statistical power to determine outbreak risk factors. On August 15, investigators collected filter backwash and water samples directly from the facility’s aquatic venues; on August 16, facility operators hyperchlorinated the aquatic venues, raising the free available chlorine concentration for a prolonged period to achieve 3-log₁₀ (99.9%) Cryptosporidium inactivation. Microscopy and real-time polymerase chain reaction (PCR) testing (3) did not detect Cryptosporidium in the filter backwash or water samples. An inspection found facility operation and maintenance in compliance with local standards, including water disinfectant concentration and pH standards. No fecal incident was known to have occurred in any of the facility’s aquatic venues before the outbreak. However, it was recommended the facility develop policies on fecal incident response and maintain a fecal/vomit incident response log.*

Arizona

On August 2, 2016, the Arizona Department of Health Services was notified of a cluster of gastrointestinal illness among players on a Coconino County Little League team and family members; 36 (71%) of 51 persons became ill 6–7 days after visiting a Maricopa County aquatic facility on July 22. Molecular characterization by CryptoNet of four Cryptosporidium specimens from the Little League cohort identified all four as the C. hominis IfA12G1R5 subtype. Maricopa County Department of Public Health simultaneously detected increased laboratory reporting of cryptosporidiosis as of mid-July. Multiple patients reported visits to the same Maricopa County aquatic facility. To determine the magnitude of the outbreak, the counties interviewed cryptosporidiosis patients, focusing on possible risk factors, particularly recreational water exposures. During July 1–October 31, 2016, a total of 352 laboratory-confirmed cryptosporidiosis cases were detected statewide, compared with a median of 46 total cases (range = 42–62) detected annually during 2011–2015. Among 317 interviewed patients, 204 (64%) reported recreational water exposure at 86 public aquatic venues, 74 (86%) of which were in Maricopa County. Environmental health practitioners of affected counties worked with facility operators to hyperchlorinate identified aquatic venues. Among 247 Maricopa County patients interviewed, 43 (17%) reported swimming while symptomatic at a median of one venue (range = 1–3).

Ohio

During 2012–2015, the Ohio Department of Health and local public health partners detected a median of 399 cryptosporidiosis cases annually statewide (range = 324–571). In 2016, annual incidence increased nearly fivefold to 1,940 cases. Ten (42%) of 24 cryptosporidiosis outbreaks detected in Ohio in 2016^† were associated with aquatic venues. Assessing patients’ recreational water exposures to determine the magnitude of individual recreational water–associated outbreaks was complicated by individual patients reporting multiple exposures during their incubation period. Among six Cryptosporidium specimens from patients affected by these outbreaks, all were identified by CryptoNet as the C. hominis IdA19 subtype, which has rarely been identified in the United States. Five specimens were from a university sports team’s members; the sixth specimen was from a patient with no epidemiologic link to the university sports team except visiting the same waterpark. The matching subtype and epidemiologic link led the Ohio Department of Health to classify the 26 cases in the university sports team as part of the waterpark–associated outbreak; these cases previously had been thought to be associated with a different outbreak.

Discussion

This report highlights cryptosporidiosis outbreaks associated with aquatic facilities in three states in 2016. CryptoNet genotyping (18S PCR-RFLP) to determine Cryptosporidium species and subtyping (gp60 PCR and sequencing) to determine subtype (4) supported and strengthened the Alabama, Arizona, and Ohio outbreak investigations. First, molecular characterization identified or confirmed epidemiologic links among individual outbreak-associated cases. Second, and perhaps more importantly, C. hominis was repeatedly identified as the outbreak etiology. Given that individual Cryptosporidium species can have unique host ranges, identifying the Cryptosporidium species can provide insight into possible exposures and outbreak sources. Identifying C. hominis as the etiology of these outbreaks indicates a human source of contamination and underscores the need to engage swimmers and parents of young swimmers in efforts to prevent and control aquatic facility–associated cryptosporidiosis outbreaks.

Most Cryptosporidium species are indistinguishable by traditional diagnostic tests (microscopy or immunoassays); only molecular diagnostic methods, such as those used by CryptoNet, can distinguish these species and their subtypes. C. hominis IfA12G1R5 subtype was identified as the etiology in the Alabama and Arizona outbreak investigations. This subtype was initially identified in small numbers of specimens from sporadic (i.e., not outbreak-associated) cryptosporidiosis cases in the United Kingdom and Australia (5–7). In the United States, it was first seen in specimens from patients with acquired immunodeficiency syndrome (AIDS) and was responsible for a 2009 Oregon cryptosporidiosis outbreak associated with the care of an AIDS patient. Since 2013, it has emerged as the dominant C. hominis subtype among sporadic and outbreak-associated cases with Cryptosporidium subtyping data; 107 (36.6%) of 292 Cryptosporidium specimens from sporadic cases in 2016 were identified as the C. hominis IfA12G1R5 subtype.

To better understand the implication of identifying this subtype, molecular characterization of Cryptosporidium specimens needs to shift from predominantly supporting outbreak investigations to becoming nationally systematic. In 2010, CDC launched CryptoNet (https://www.cdc.gov/parasites/crypto/cryptonet.html), the first U.S. molecularly based surveillance system for a parasitic disease. Formal collaborations with state public health partners began in mid-2015. The objectives are to efficiently integrate CryptoNet into existing infrastructure when possible (e.g., merging the CryptoNet BioNumerics infrastructure into that of PulseNet) and to regularly analyze molecular characterization and epidemiologic data for each nationally notified case of cryptosporidiosis to further elucidate Cryptosporidium chains of transmission and cryptosporidiosis epidemiology (e.g., geographic and temporal changes in the distribution of Cryptosporidium species and their subtypes and associated exposures). Achieving these objections requires overcoming barriers to successful molecular characterization and sharing epidemiologic data by 1) increasing the positive predictive value of rapid diagnostic tests (i.e., decreasing the frequency of false positive results) (8), 2) shifting away from fixing specimens in formalin (which precludes molecular characterization), 3) advancing molecular diagnostics from single-gene to multilocus or whole-genome sequencing (which will increase discriminatory power), and 4) increasing state capacity to collect and share epidemiologic data with CDC.

The emergence of Cryptosporidium as the leading etiology of aquatic facility–associated outbreaks results from the parasite’s extreme chlorine tolerance. Free available chlorine inactivates most infectious pathogens within minutes at CDC-recommended concentrations of at least 1 ppm^§; however, Cryptosporidium oocysts can survive for days (9). As the Alabama outbreak investigation indicates, even properly operated and maintained aquatic venues can be sites of Cryptosporidium transmission. In addition, cyanuric acid (a stabilizer added to prevent chlorine depletion by the sun’s ultraviolet light) has been found to substantially delay chlorine inactivation of Cryptosporidium (9). Consequently, in July 2016, CDC issued revised recommendations for hyperchlorination (https://www.cdc.gov/healthywater/swimming/aquatics-professionals/fecalresponse.html) when responding to diarrheal incidents in public aquatic venues (i.e., high-risk Cryptosporidium contamination events) and aquatic facility–associated cryptosporidiosis outbreaks. These recommendations are also included in CDC’s 2016 Model Aquatic Health Code (https://www.cdc.gov/mahc/editions/current.html). This national guidance can be adopted voluntarily by state and local jurisdictions and aquatic facilities to minimize the risk for public aquatic facility–associated illness and injury, particularly cryptosporidiosis.

Preventing Cryptosporidium contamination of water in an aquatic venue would prevent Cryptosporidium transmission more efficiently than remediating actions once contamination occurs. This means that public health agencies and the aquatics sector need to collaborate on engaging swimmers, who are the source of contamination, in prevention efforts. Young swimmers aged <5 years are more likely to contaminate the water because they are more likely to have inadequate toileting and hygiene skills; therefore, prevention efforts should focus on their parents. As the Arizona outbreak investigation demonstrated, patients continue to swim while symptomatic. The key healthy swimming message to the public to prevent contamination is “Don’t swim or let your kids swim if sick with diarrhea.”Health care providers should also instruct cryptosporidiosis patients not to go back into the water until they have been diarrhea-free for 2 weeks.^¶ Healthy swimming promotion campaigns conducted before the summer swim season could reduce the risk for outbreaks caused by Cryptosporidium and other enteric pathogens (10), while optimizing the health benefits of water-based physical activity (https://www.cdc.gov/healthywater/swimming/swimmers/health_benefits_water_exercise.html).