Cryptosporidiosis outbreak in a recreational swimming pool in minnesota.

Introduction

Cryptosporidium outbreaks attained public prominence after the Milwaukee epidemic of 1993, which was associated with the contamination of the public water supply and involved 403,000 cases of diarrheal illness (MacKenzie et al., 1994). Cryptosporidiosis associated with the use of swimming pools has been reported previously, but it is probably under-recognized (Centers for Disease Control and Prevention [CDC], 1994). This article describes the investigation of a cryptosporidiosis outbreak associated with a swimming pool in Olmsted County, Minnesota. To the authors' knowledge, it is the first published report of such an outbreak in Minnesota.

Description of the Cryptosporidiosis Outbreak

On May 6, 1998, the Olmsted County Public Health Department received a call from the mother of a child who had diarrheal illness. She was concerned that her child had a foodborne illness associated with eating at a banquet on April 26, and reported that others had become ill after eating at the same banquet. After the health department received this notification, it conducted a case control study, and 40 banquet attendees were interviewed. Sixteen of the 40 attendees were ill, with diarrhea as the predominant symptom. Stool samples were requested from symptomatic individuals.

A foodborne outbreak was ruled out when it was discovered that five people had been ill prior to the banquet, and when no banquet foods were found to be associated with the illness. Furthermore, none of the employees from the banquet caterer had reported being ill with similar symptoms, and none of the participants at a second banquet, organized by the same caterer, had fallen ill. The median duration of diarrhea in this initial set of cases was 10 days, longer than that associated with most foodborne illnesses. During the course of the investigation, investigators learned that all of the ill people were frequent swimmers at a local swimming pool. This finding, coupled with the long duration of illness, raised suspicions that Cryptosporidium was a possible agent. The focus then turned toward swimming pool water exposure, and the epidemiologic investigation was expanded to include other groups that used the pool.

Methods

Lists of staff, classes, and other groups of people who swam in the pool were obtained from the pool management. An expanded questionnaire was developed to determine demographic information, symptoms, and exposure to swimming pool water. People on the list were interviewed, and ill people also were asked to provide a stool sample. Local hospitals and clinics were alerted and requested to report suspected cases of cryptosporidiosis. Stool samples were requested for a subset of cases with diarrheal symptoms. The case definition was diarrhea experienced for three or more days after swimming in the pool, or a stool test that was positive for Cryptospordium.

The swimming pool was inspected and the management interviewed. Samples of the pool water and diatomaceous earth filter material from the swimming pool were collected and sent to the Minnesota Department of Health Laboratory for analysis.

Results

Two hundred and six people were interviewed. Sixty-eight of the interviews were excluded from additional analysis because symptoms of gastrointestinal symptoms were present in these individuals but did not meet the case definition of three or more days of diarrhea. Twenty-six of the 138 remaining people who were interviewed had illness that met the case definition, for an attack rate of 19 percent.

Ultimately, swimmers from six different groups were interviewed, and cases were found in four of those groups (Table 1). The most prominent symptom was diarrhea (100 percent), followed by abdominal cramps (81 percent), and nausea (77 percent) (Table 2). The duration of symptoms ranged from 3 to 28 days, with a median duration of 9 days. Ten individuals with case illnesses had sought medical attention at local clinics. Health care providers at these clinics had requested stool specimens for testing in only two of those cases. Cryptosporidium testing was specifically requested for only one individual. The county public health department requested stool specimens from another 12 symptomatic individuals, 10 of which were received for laboratory analysis. Three of these samples were positive for Cryptosporidium, bringing the total number of confirmed cases to four (Figure 1).

The index case, which had an onset of illness on April 10, and six additional cases, which had onsets that spanned 26 days, all occurred in members of a preschool swimming group (Group B) (Figure 1). Three of these seven preschool swimmers had Cryptosporidium in their stools. The largest number of cases occurred from April 26 to April 30, and most of the cases in that cluster occurred in members of a competitive swim group (Group A).

The cases occurred in people who were between 20 months and 29 years of age, with a median age of 14 years (Table 1). The median age for cases from Group A was 14 years. In Groups B and C, the median ages for cases were two and nine years, respectively. Although the median age for the water aerobics group (Group D) was 61, only one adult, 29 years of age, was ill.

The highest illness attack rates were found in the competitive and preschool swimming groups (61 percent and 39 percent, respectively) (Table 1). Sixteen of the 26 cases were from Group A, and 10 were from Groups B, C, and D. Swimmers from Group A had a significantly higher likelihood of being affected than did the other swimmers (Fisher's exact test, p < .0001).

Swimming Pool Assessment

The swimming pool involved in the outbreak holds 530.000 gallons of water with a six-hour turnover. It uses a diatomaceous earth filter, which is changed weekly. No lapses in pool operation were identified during the inspection or through review of the pool operation records. Throughout April and early May, free-chlorine levels were adequate. Cryptosporidium was not identified in the samples of swimming pool water or in material from the diatomaceous earth filter.

Discussion

Cryptosporidiosis transmission occurs with ingestion of water contaminated with Cryptosporidium oocysts (Guerrant, 1997), which are highly resistant to chemical disinfectants, including chlorine (Korich, Mead, Madore, Sinclair, & Sterling, 1990). Infection is spread by the fecal-oral route via the ingestion of oocysts excreted in human or animal feces (Marshall & LaMont, 1997). The Cryptosporidium oocysts infect the gastrointestinal tract of a susceptible host and have an incubation period of up to two weeks (Meinhardt, Casemore, & Miller, 1996). Infection can occur even when small quantities of oocysts are ingested (DuPont et al., 1995; Guerrant, 1997). Cryptosporidium infection occurs frequently in young children, with a peak incidence in those between one and five years of age (Casemore, 1990). Symptoms can range from nausea and watery diarrhea to malabsorption and weight loss in severe cases. In immunocompetent persons, the disease is usually self-limiting, with full recovery after 10 to 14 days. People with a compromised immune system may have a more prolonged period of diarrhea, leading to malnutrition and increased morbidity and mortality (Marshall & LaMont, 1997). To date, no effective pharmacotherapy exists, and treatment is largely supportive.

[FIGURE 1 OMITTED]

Several features of the cryptosporidiosis outbreak support the conclusion that the swimming pool was the source. Cryptosporidium was isolated in four of the affected swimmers. The common exposure in all the affected cases was the swimming pool. Infected individuals had clinical manifestations consisting mainly of watery diarrhea, abdominal cramps, and nausea. The median duration of illness was nine days, which is longer than that associated with most bacterial or viral agents

Foodborne illness from the banquet was ruled out in the initial case control study. Specifically, five people with cases of the illness had been ill prior to eating at the banquet. Other pertinent factors included the absence of ill employees from the banquet caterer and the lack of subsequent outbreaks from food prepared by the caterer. The index case was a preschool-aged swimmer. The sporadic distribution of illness onset dates over a long period of time suggests intermittent release of organisms from more than one individual into the pool (Figure 1). It is not surprising that Cryptosporidium was not detected in samples of swimming pool water and diatomaceous earth filter. This form of testing is known to have poor sensitivity (MacKenzie, Kazmierczak, & Davis, 1995). Outbreaks also have been known to occur in the absence of detectable oocysts in water (Haas, 2000).

Competitive swimmers from Group A generally spent a lot more time in the pool than did other swimmers, increasing the opportunity for water ingestion and thus increasing their risk of acquiring cryptosporidiosis. In water aerobics (Group D), there is less potential for water ingestion because total head submersion usually does not occur. Only one person from Group D was affected. Unlike the other water aerobics participants, this person also was involved in other swimming activities. Swimmers from groups E and F generally were older and more experienced swimmers, and may have been less likely to ingest swimming pool water. The child swimmers of Group C did not swim as frequently as the competitive swimmers, so were likely to have had less opportunity for pool water exposure. On the other hand, preschool swimmers (Group B) may not have been adequately toilet trained, and could have aided in the propagation of the illness through fecal contamination of the swimming pool. They were also more likely to ingest swimming pool water during head submersion. This factor may explain the high attack rate in this group.

Outbreak Interventions

The swimming pool was closed on May 13, one day after Cryptosporidium was detected in the initial stool samples. Before it was reopened, the pool was superchlorinated to 60 parts per million for 24 hours, which provided enough time for the water to circulate throughout the system, exceeding the recommended CT value of 9600 (CDC, 2001b). (CT refers to concentration [C] of free available chlorine, in milligrams per liter (mg/L), or ppm, multiplied by time [T], in minutes.) The pool equipment was disinfected with a solution containing sodium hypochlorite. To limit the transmission of cryptosporidiosis, county public health officials also implemented the following steps: 1) A press release was issued to the public to reduce the risk of continued transmission. 2) Notice was given to all pool operators in the county describing the outbreak and ways to prevent additional cases. 3) Signs were placed at all county swimming pools requesting people not to enter the pool for up to two weeks following an episode of diarrheal illness.

As a result of intensified surveillance, another outbreak of cryptosporidiosis associated with an Olmsted County swimming pool was detected in 1998. This outbreak, involving seven cases, would not have been recognized without enhanced physician testing and reporting. In 1999, eight cases of cryptosporidiosis were associated with swimming at an Olmsted County manufactured-home park. There have even been prior reports of cryptosporidiosis related to recreational water use in Minnesota. In 1993, 26 residents of a manufactured-home park had diarrhea caused by Cryptosporidium and Giardia associated with swimming in neighborhood wading pools. Elsewhere in Minnesota, a Cryptosporidium outbreak was associated with a fountain at a zoo (CDC, 1998).

A diagnosis of cryptosporidiosis is reportable by law to health departments in 41 states, including Minnesota (Minnesota Department of Health, 2003). Nevertheless, waterborne Cryptosporidium infection is probably under-recognized and under-reported (CDC, 1998). Cryptosporidium oocysts are too small to be detected on routine stool ova and parasite examinations. Specialized testing, including acid-fast staining and immunofluoresence antibody techniques, are required to detect Cryptosporidium oocysts with certainty (Marshall & LaMont, 1997). Contrary to what many physicians think, most laboratories do not specifically test for Cryptosporidium during routine stool examination (CDC, 1998). This large outbreak might have gone undetected if the parent had not reported the illness to public health officials, and if the stool sample had not been specifically analyzed for Cryptosporidium.

Prevention and Treatment of Recreational Water-Associated Cryptosporidiosis Outbreaks

The detection and control of swimming pool-associated Cryptosporidium outbreaks presents several challenges. Ill people often do not seek health care for diarrheal illness (CDC, 2001a; Hellard et al., 2000; MacKenzie et al., 1994). When they do see their physician, they are unlikely to be tested for Cryptosporidium infection. The risk of water contamination increases when people with diarrheal illness continue to swim. People with gastrointestinal symptoms may be mistakenly diagnosed with viral gastroenteritis or "intestinal flu" without further investigation (Mackenzie et al., 1994). Furthermore, the perception that cryptosporidiosis is a self-limiting illness for which there is not an effective treatment may deter testing. These challenges need to be overcome through better physician and public awareness.

Stool sampling for Cryptosporidium is recommended if there is an index of suspicion for cryptosporidiosis. Apart from ascertaining the diagnosis, testing has other benefits. It allows for patient education, facilitates early recognition of outbreaks, and provides the opportunity to institute preventive measures against further spread of the disease (MacKenzie et al., 1994). Testing for Cryptosporidium oocysts is crucial in children presenting with diarrheal illness in the setting of a recent exposure to recreational water or, in the absence of exposure to recreational water, diarrhea lasting more than three days. A study conducted in Minnesota revealed that Cryptosporidium was the third most common parasite recovered from patients under 13 years of age (Lally & Woolfrey, 1998).

The pool management and the public can pursue certain measures to limit the spread of Cryptosporidium and other diarrheal illnesses in swimming pools (see sidebar at right). Cryptosporidium oocysts are particularly resistant to regular swimming pool chlorination (Korich et al., 1990). Also, oocysts attach readily to biologic particles found in swimming pools, such as sweat, hair, skin cells, and urine, which may further delay their inactivation by chlorine (Carpenter, Fayer, Trout, & Beach, 1999). Although sand filtration can effectively remove oocysts under laboratory conditions (Chapman & Rush, 1990), low-grade rapid sand filters used in swimming pools may not be as effective (CDC, 1994; MacKenzie et al., 1995). With diatomaceous earth filtration systems, complete pool drainage may not be necessary following a fecal accident as long as the pool is closed to allow sufficient time for complete filtration following disinfection procedures (Sorvillo et al., 1992). Swimming pool fecal accidents have to be regarded as a potential health hazard and treated appropriately. CDC has provided guidelines for treating swimming pool fecal contamination (CDC, 2001b). Public education is essential in implementing stringent poolside hygiene. Well-placed signs posted around the swimming pool (warning people with recent diarrhea not to swim for at least two weeks, for example) may provide effective reinforcement.

Conclusion

The outbreak reported here illustrates the importance of considering waterborne exposure to agents that cause enteric disease, especially during the summer months. Cryptosporidium infection is not an uncommon occurrence, and it is associated with fecal accidents in swimming pools. A high index of suspicion for Cryptosporidium infection is necessary during outbreaks of gastroenteritis associated with exposure to swimming pools or recreational-water parks. No effective therapy for Cryptosporidium infection is known. Hence, early detection and control of cryptosporidiosis outbreaks are important, especially for preventing significant morbidity and mortality in immunocompromised people. Prevention of cryptosporidiosis outbreaks in swimming pools requires diligence on the part of the pool management and the swimmers in maintaining a hygienic swimming pool environment. Specific measures to limit the transmission of Cryptosporidium infection include installing a fine-grade diatomaceous-earth filtration system and adherence to strict protocols when fecal accidents occur. Public awareness--including an understanding that one should refrain from the use of pools during diarrheal illness--and the practice of proper hygiene are vital in preventing cryptosporidiosis outbreaks associated with swimming pools.

TABLE 1 Cryptosporidiosis Cases, by Swimming Group

Group  Description      Number       Median Age  Number    Median Age
       of Swimmers      Interviewed  of Group    of Cases  of Cases

A      Competitive      26           14          16        14
B      Preschool        18            2           7         2
C      Children         61            9           2         9
D      Water aerobics   22           61           1        29
E      Lap               8           79           0        --
F      Casual            3           16           0        --
Total                  138           10          26        14

Group  Description     Attack Rate
       of Swimmers

A      Competitive     61%
B      Preschool       39%
C      Children         3%
D      Water aerobics   4%
E      Lap              0%
F      Casual           0%
Total                  19%

TABLE 2 Symptom Profile of Cases

Symptom           Number  Percentage

Diarrhea          26      100%
Abdominal cramps  21       81%
Nausea            20       77%
Headache          12       46%
Fever             12       46%
Sweats/chills     11       42%
Muscle aches      11       42%

Acknowledgements: The following people were involved in the outbreak investigation: Rich Peter. M.S., Olmsted County Public Health; Scott Fryer, R.S., Olmsted County Public Health; Jeff Bender, D.V.M., M.S., University of Minnesota (formerly with the Minnesota Department of Health); and interviewers from the Olmsted County Public Health Services.

www.neha.org

Journal of Environmental Health

NEHA Training

Bookstore

Credentials

Continuing Education 

Online Training

Research & Development

Radon Proficiency Credential

Position Papers

Calendar of Events

Affiliates

Links

Information and opportunities abound behind the research and development button on NEHA's homepage. Have you visited this section lately? www.neha.org/research/index.html contains information on all of NEHA's federally funded grant programs, many of which include free or low-cost training and education opportunities. You'll find information about NEHA's

* Terrorism Response Program, which includes an online resource center;

* Food Safety Program, which includes a link to extensive reviews of food safety training materials for all types of audiences;

* Foodborne Disease Outbreak Investigation Program, which includes free regional 2.5-day Epi-Ready Workshops; and

* Radon and Indoor Air Quality Program, which includes free 3-day workshops.

In addition, you can learn about work NEHA is currently doing to develop an Onsite Wastewater System Installer Credential.

Check it out--we're sure you'll be pleased with what you find!

REFERENCES

Carpenter, C., Fayer, R., Trout, J., & Beach, M.J. (1999). Chlorine disinfection of recreational water for Cryptosporidium parvum. Emerging Infectious Diseases, 5(4), 579-584.

Casemore, D.P. (1990). Epidemiological aspects of human cryptosporidiosis. Epidemiology and Infection, 104(1), 1-28.

Centers for Disease Control and Prevention. (1994). Cryptosporidium infections associated with swimming pools--Dane County, Wisconsin, 1993. Morbidity and Mortality Weekly Report, 43(31), 561-563.

Centers for Disease Control and Prevention. (1998). Outbreak of cryptosporidiosis associated with a water sprinkler fountain--Minnesota, 1997. Morbidity and Mortality Weekly Report, 47(40), 856-860.

Centers for Disease Control and Prevention. (2001a). Protracted outbreaks of cryptosporidiosis associated with swimming pool use--Ohio and Nebraska, 2000. Morbidity and Mortality Weekly Report, 50(20), 406-410.

Centers for Disease Control and Prevention. (2001b). Responding to fecal accidents in disinfected swimming venues. Morbidity and Mortality Weekly Report, 50(20), 416-417.

Chapman, P.A., & Rush, B.A. (1990). Efficiency of sand filtration for removing Cryptosporidium oocysts from water. Journal of Medical Microbiology, 32(4), 243-245.

DuPont, H.L., Chappell, C.L., Sterling, C.R., Okhuysen, P.C., Rose, J.B., & Jakubowski, W. (1995). The infectivity of Cryptosporidium parvum in healthy volunteers. New England Journal of Medicine, 332(13), 855-859.

Guerrant, R.L. (1997). Cryptosporidiosis: An emerging, highly infectious threat. Emerging Infectious Diseases, 3(1), 51-57.

Haas, C.N. (2000). Epidemiology, microbiology, and risk assessment of waterborne pathogens including Cryptosporidium. Journal of Food Protection, 63(6), 827-831.

Hellard, M.E., Sinclair, M.I., Fairley, C.K., Andrews, R.M., Bailey, M., Black, J., Dharmage, S.C., & Kirk, M.D. (2000). An outbreak of cryptosporidiosis in an urban swimming pool: why are such outbreaks difficult to detect? Australian and New Zealand Journal of Public Health, 24(3), 272-275.

Korich, D.G., Mead, J.R., Madore, M.S., Sinclair, N.A., & Sterling, C.R. (1990). Effects of ozone, chlorine dioxide, chlorine, and monochloramine on Cryptosporidium parvum oocyst viability. Applied and Environmental Microbiology, 56(5), 1423-1428.

Lally, R.T., & Woolfrey, B.F. (1988). Recovery of Cryptosporidium oocysts from stool samples submitted for ova and parasite examination in a Minnesota pediatric population. Pediatric Infectious Disease Journal, 7(3), 200-201.

MacKenzie, W.R., Hoxie, N.J., Proctor, M.E., Gradus, M.S., Blair, K. A., Peterson, D.E., Kazmierczak, J. J., Addiss, D.G., Fox, K.R., Rose, J.B., & Davis, J.P. (1994). A massive outbreak in Milwaukee of Cryptosporidium infection transmitted through the public water supply. New England Journal of Medicine, 331(3), 161-167.

MacKenzie, W.R., Kazmierczak, J.J., & Davis, J.P. (1995). An outbreak of cryptosporidiosis associated with a resort swimming pool. Epidemiology and Infection, 115(3), 545-553.

Marshall, A.T., & LaMont, J.T. (1997). Cryptosporidiosis and public health. Hospital Practice (Office Edition), 32(7), 11, 15-16, 23.

Meinhardt, P.L., Casemore, D.P., & Miller, K.B. (1996). Epidemiologic aspects of human cryptosporidiosis and the role of waterborne transmission. Epidemiologic Reviews, 18(2), 118-136.

Minnesota Department of Health. (2003). Diseases reportable to the Minnesota Department of Health, http://www.health.state.mn.us/divs/idepc/newsletters/dcn/sum02/reportable.html (18 Sept. 2003).

Sorvillo, F.J., Fujioka, K., Nahlen, B., Tormey, M.P., Kebabjian, R., & Mascola, L. (1992). Swimming-associated cryptosporidiosis. American Journal of Public Health, 82(5), 742-744.

RELATED ARTICLE: Measures to Limit Cryptosporidium Transmission in Swimming Pools With a Fine-Grade Diatomaceous Earth Filtration System

* Regularly change swimming pool filters.

* Maintain an adequate flow of pool water through the filters.

* Promptly treat swimming pool fecal accidents.

* Implement stringent poolside hygiene via public education:

-- Discourage people recovering from an acute diarrheal illness from using the pool for at least two weeks.

-- Discourage ingestion of pool water.

-- Discourage parents from bringing diapered children into the pool.

-- Discourage changing of diapers by the poolside.

-- Encourage showering prior to use of the pool.

-- Encourage handwashing after use of the toilet or contact with fecal material (e.g., through changing of diapers).

-- Encourage frequent restroom breaks for children.

Lionel S. Lim, M.D., M.P.H.

Prathibha Varkey, M.D., M.P.H.

Peter Giesen, M.S.

Larry Edmonson, M.P.H.

Corresponding Author: Larry Edmonson, Director, Olmsted County Public Health Services, Division of Disease, Prevention, and Control. 2100 Campus Dr., SE, Rochester, MN 55904. E-mail: [e-mail].