Download Cryptosporidium Species: New Insights and Old Challenges

INVITED ARTICLE
FOOD SAFETY
David Acheson, Section Editor
Cryptosporidium Species: New Insights
and Old Challenges
Brett A. Leav, Melanie Mackay, and Honorine D. Ward
Division of Geographic Medicine and Infectious Diseases, Tufts–New England Medical Center, Boston, Massachusetts
Cryptosporidium species are protozoan parasites that cause mainly enteric illnesses in humans and other animals. The mode
of transmission is most commonly waterborne, but other sources of infection, including foodborne and person-to-person
spread, have been documented. The environmental form of the parasite is resistant to most water purification methods,
including chlorination. Cryptosporidium infection usually causes a self-limited diarrheal illness but can be life-threatening
in immunocompromised individuals. There is no effective therapy for cryptosporidiosis.
Cryptosporidium species are intracellular gastrointestinal parasites that were initially thought to cause disease only in animals
[1]. The first case of cryptosporidiosis in a human was described
in 1976 in a 3.5-year-old girl who developed self-limited enterocolitis [2]. The disease subsequently achieved widespread
notoriety in the medical community during the early years of
the AIDS epidemic and later after a massive outbreak in Milwaukee, Wisconsin, in 1993 [3]. The parasite continues to be
a major threat to human health for 2 major reasons. First, the
current methods of water purification are ineffective for its
removal from the public water supply. Second, there is no
effective therapy for cryptosporidiosis. Although most cases of
cryptosporidiosis in humans are self-limited, the consequences
of infection can be devastating in immunocompromised individuals and in children living in underdeveloped countries.
Advances in our understanding of Cryptosporidium species have
been limited because the organism can be propagated only in
live animal hosts. Nevertheless, newer molecular techniques,
such as PCR, have provided new insights into the complex
epidemiology of this parasite.
Received 4 October 2002; accepted 26 November 2002; electronically published 18 March
2003.
Financial support: National Institutes of Health (grants AI51581A [to B.A.L.], AI46299 [to
H.D.W.], AI45194 [to H.D.W.], and D3TW05571A [to M.M.]), US Department of Agriculture
(grant 2000-02247), and American Waterworks Association Research Foundation (grant 2596
[to H.D.W.]).
Reprints or correspondence: Dr. Honorine Ward, Div. of Geographic Medicine and Infectious
Diseases, Tufts–New England Medical Center, Box 041, 750 Washington St., Boston, MA
02111 ([email protected]).
Clinical Infectious Diseases 2003; 36:903–8
2003 by the Infectious Diseases Society of America. All rights reserved.
1058-4838/2003/3607-0012$15.00
EPIDEMIOLOGY
Cryptosporidium infection is a zoonosis with mainly bovine and
human reservoirs. Humans are thought to acquire the parasite
by the ingestion of oocysts, which are shed in the stool of
infected animals or other humans. In developing countries,
cryptosporidiosis is endemic and is one of the most common
causes of persistent diarrhea among children. In the developed
world, cryptosporidiosis mainly occurs in sporadic outbreaks
and epidemics. Immunodeficient people, particularly those
with deficiencies in cell-mediated immunity, are especially susceptible to Cryptosporidium infection and experience the mostsevere consequences of this illness.
Cryptosporidium infection is an important cause of diarrhea
in the developing world [4]. An extensive review of cases of
diarrhea demonstrated that the incidence of Cryptosporidium
infection among immunocompetent patients from developing
countries was 6.1%. Children from underdeveloped countries
appear to be at particular risk for Cryptosporidium infection:
serological evidence of previous infection was present in ∼50%
of children in rural China and nearly all children living in an
urban slum in Brazil. In the United States, there have been
occasional outbreaks of infection in day care centers, foodborne
outbreaks, and outbreaks associated with recreational water
(public swimming pools, lakes, and ponds). The majority of
reported cases of cryptosporidiosis in humans have been associated with contaminated drinking water; the largest such
outbreak occurred in Milwaukee in 1993 [3].
Cryptosporidium infection is probably underdiagnosed as a
cause of community-acquired diarrhea in the United States.
Approximately 2% of stools tested (presumably for active enFOOD SAFETY • CID 2003:36 (1 April) • 903
teritis) are positive for Cryptosporidium species [5]. Furthermore, antibodies to Cryptosporidium have been detected in
∼30% of children and adults in the United States [6]. Even
higher seroprevalence rates have been detected in certain populations, such as dairy farmers [6]. Patients with poorly controlled HIV infection and patients receiving immunosuppressive agents are at particular risk for cryptosporidiosis. Other
causes of immunodeficiency, such as solid-organ transplantation or primary immunodeficiencies, x-linked hyper-IgM syndrome, severe combined immunodeficiency, selective IgA deficiency, and possibly IFN-g deficiency, are also thought to
increase the risk of infection with Cryptosporidium species [7].
TRANSMISSION
The best-documented routes of transmission are waterborne,
foodborne, and person-to-person spread. The majority of the
documented outbreaks of waterborne infection in the world
have been attributed to contaminated drinking water supplies,
although contaminated water used for recreational activities has
also been implicated [5]. The first reported waterborne outbreak of cryptosporidiosis occurred in 1984 and was attributed
to fecal contamination of a public artesian well in Texas. In
the spring of 1993, the largest outbreak of waterborne disease
of any kind recorded occurred in Milwaukee. More than
400,000 of ∼1,600,000 people in the greater Milwaukee area
developed cryptosporidiosis after consuming contaminated
drinking water. The onset of illness correlated with an increase
in the turbidity of the treated water from Lake Michigan, despite
use of water purification methods that were standard at the
time [3]. Although the source of contamination remains uncertain, recent genotypic analysis of 4 of the isolates indicates
consistency with a human origin [8].
Unfortunately, despite the public attention garnered by this
massive outbreak, there have been many subsequent outbreaks
of Cryptosporidium infection attributed to contaminated drinking water in the United States and the United Kingdom. This
is in part due to increased surveillance, but it also reflects our
inability to rid the public water supplies of this troublesome
parasite. Concern about the safety of the public water supply
has prompted government authorities to issue standards for
the detection of Cryptosporidium species. These guidelines have
been updated as recently as 2001, and our ability to detect
Cryptosporidium species in the public water supply has unquestionably improved [9].
However, although surveillance has improved, there is still
no consensus between scientists and public officials about the
safe limits of Cryptosporidium species in the water supply. Furthermore, implementation of practical, affordable, and effective
methods for the elimination of the parasite from drinking water
remains an elusive goal. The current strategy for elimination
904 • CID 2003:36 (1 April) • FOOD SAFETY
of Cryptosporidium species from the public water supply involves preventing contamination of water sources, physical removal of the organisms, and chemical or physical disinfection.
However, these methods are only capable of reducing the number of oocysts, not eliminating the parasite from the water
supply. Therefore, it is imperative that patients at risk for Cryptosporidium infection contact their local public water authorities
for advice about the safety of their water supply. More sophisticated approaches, such as filtration, ultraviolet light irradiation, and ozone treatment, have not been widely applied
for various reasons, including financial and health-related concerns [10].
Although waterborne routes of transmission have been the
most notable, foodborne and person-to-person spread have
also been documented. Cryptosporidiosis has been attributed
to ingestion of contaminated apple cider, chicken salad, milk,
and food prepared by an ill food handler [5, 11, 12]. Cryptosporidium species have also been detected in seawater and have
even been found in commercially harvested oysters [5]. Another
potential source of infection may be raw vegetables sold in the
marketplaces in developing countries [5]. However, in the latter
2 cases, a causative relationship has not been established between environmentally detected Cryptosporidium species and
cases of infection in humans. Nevertheless, although waterborne outbreaks of cryptosporidiosis in humans have been documented more frequently, foodborne transmission may be underappreciated. Finally, it should be mentioned that, although
cattle are thought to be the most common animal reservoir of
Cryptosporidium species, other species of animals, including
reptiles, birds, and insects, have been shown to harbor the
parasite [5]. It is not clear what role these other animal hosts
have in the transmission of Cryptosporidium species.
TAXONOMY
Cryptosporidium species, like Cyclospora, Isospora, and Toxoplasma species, are members of the phylum Apicomplexa. Cryptosporidium species were initially distinguished by their size, host
specificity, and site of infection within the animal. Recently, molecular techniques have been used to establish a genetic basis for
these distinctions [13, 14]. There are currently thought to be 10
valid species of Cryptosporidium: Cryptosporidium andersoni,
Cryptosporidium baileyi, Cryptosporidium felix, Cryptosporidium
meleagridis, Cryptosporidium muris, Cryptosporidium nasorum,
Cryptosporidium parvum, Cryptosporidium saurophilum, Cryptosporidium serpentis, and Cryptosporidium wrairi. Cryptosporidium infections have been described in 152 different mammalian
species. Although Cryptosporidium organisms are known to be
zoonotic pathogens, human-to-human transmission seems be
the most common mode of transmission in the United States
[5, 11]. Recent efforts to use PCR to amplify specific genetic loci
have led to a much better understanding of the biological characteristics and transmission of the parasite. These studies have
confirmed that C. parvum, the most common cause of cryptosporidiosis in humans, can be divided into 2 major genotypes.
These 2 genotypes, human and bovine, are also referred to as
genotype 1 and genotype 2, respectively [13]. With rare exceptions, C. parvum genotype 2 is capable of infecting humans and
animals, whereas genotype 1 infects only humans [13]. This classification may be an oversimplification, however, because several
recent studies have shown significant genetic polymorphisms between isolates of the same genotype. We and others have shown
tremendous heterogeneity among genotype 1 isolates at the
Cpgp40/15 (also known as “gp60/45/15”) locus, which encodes
surface glycoproteins present on invasive stages of the parasite
[15, 16]. Furthermore, these studies suggest that Cryptosporidium
organisms may undergo genetic recombination within and between the 2 prototypal genotypes. Recently, Feng et al. [17] provided experimental evidence for genetic recombination after infection of a murine host with 2 different genotype 2 isolates via
microsatellite analysis.
Although the clinical significance of these genotypes is not
completely clear, some patterns are beginning to emerge. Most
infections in humans appear to be caused by genotype 1 [5].
However, some studies have shown that genotype 2 infections
are more common, particularly during the spring and in sporadic outbreaks in rural agricultural communities [11]. Infections in humans due to genotype 1 isolates have been reported
to be associated with increased quantity and duration of oocyst
shedding [18, 19].
Finally, it should be noted that other species and genotypes
of Cryptosporidium in addition to C. parvum might infect humans. Several recent reports have described the isolation of C.
meleagridis, C. felis, and the “dog” genotype from humans [5,
11]. The relative importance of these other species of Cryptosporidium in infections in humans is currently uncertain. Furthermore, because the taxonomy of Cryptosporidium is in a
state of flux, the validity of some of the more unusual species
may not endure.
PATHOGENESIS
Infection with Cryptosporidium begins when the ingested oocysts release sporozoites, which subsequently attach to and invade the intestinal epithelial cell. The parasite has a particular
predilection for the jejunum and terminal ileum. This point
deserves emphasis, because diagnostic evaluations, such as endoscopy, may miss the site of infection. In patients with AIDS,
other sites within the gastrointestinal tract may be involved,
including the stomach, duodenum, and colon, as well as the
biliary tract [7]. The histopathological features of cryptosporidiosis include a minimal inflammatory infiltrate and blunting
of the villus. More-pronounced inflammatory changes, such as
disruption of the epithelial cell barrier and more-extensive infiltration of the lamina propria with inflammatory cells, are
seen in immunodeficient patients [20].
The mechanism by which Cryptosporidium infection causes
diarrhea remains elusive. The diarrhea is typically noninflammatory and is often profuse. The parasite does elicit a local
inflammatory response, and increased production of prostaglandins and several cytokines, particularly IFN-g, has been
described [21]. It is possible that these inflammatory mediators
may consequently alter solute transportation in the intestinal
epithelial cell, leading to osmotic diarrhea. Cryptosporidium infection has also been shown to inhibit apoptosis in infected
epithelial cells as well as promote it in adjacent epithelial cells
in vitro [21]. This could theoretically prolong parasite survival
and impair absorption in the intestinal mucosa. The presence
of an enterotoxin has also been hypothesized but never conclusively demonstrated [21].
Whatever the mechanisms by which Cryptosporidium infection causes disease may be, attachment to and invasion of host
cells are crucial primary events in pathogenesis. However, little
is known about specific parasite and host molecules involved
in these processes. Knowledge of such molecules is essential
for understanding the pathogenic mechanisms used by this
parasite. Increasing recognition of Cryptosporidium species as
emerging pathogens in humans has led to the identification of
proteins, including CSL, GP900, p23/27, TRAP C1, gp40/45,
cp47, and gp15/Cp17, which are implicated in mediating these
interactions [22]. However, progress in establishing the functional role of these proteins has been hindered by the inability
to propagate C. parvum in vitro and the lack of suitable systems
for genetic manipulation of the parasite.
IMMUNE RESPONSE
Cell-mediated immunity appears to be the major component
of the immune response to Cryptosporidium infection. Although there is a prominent humoral response to infection, the
exact role of antibodies in host defense to Cryptosporidium
infection is unclear. The correlation between the decreased
number of CD4⫹ T cells and the risk of Cryptosporidium infection is evidence of the critical role that immune cells play
in immunity [23]. Studies that use murine models in which
CD4⫹ T cells were lacking confirmed the central role of these
cells in adaptive immunity [23]. The cytokines IFN-g, IL-12,
and TNF-a have also been shown to be protective against Cryptosporidium infection in laboratory models [24]. CD4⫹ T cells
appear to be the source of IFN-g in the adaptive immune
response, and these cells have been shown to limit experimental
infection [23]. Innate immune mechanisms may also be important in resistance to Cryptosporidium infection, because mice
FOOD SAFETY • CID 2003:36 (1 April) • 905
Figure 1. Cryptosporidium parvum oocysts in feces visualized by a Kinyoun modified acid-fast stain (A) and immunofluorescence with use of a
monoclonal antibody (Cellabs) to the oocyst wall (B).
that lack adaptive immune cells are initially resistant to infection. Furthermore, mutations of the mannose-binding lectin
gene, a component of the complement cascade, were associated
with an increased risk of cryptosporidiosis in patients with
AIDS [24]. Therefore, it seems that, although adaptive immunity is necessary for the resolution of cryptosporidiosis, innate immune mechanisms may be important in resistance to
infection.
CLINICAL MANIFESTATIONS
Essentially all humans are susceptible to Cryptosporidium infection, although people with serological evidence of previous
infection seem to be more resistant [25]. The parasite is highly
infectious, with an ID50 ranging from 9 to 1042 oocysts, depending on the isolate [11]. The presentation is highly variable
but is typically characterized by watery diarrhea. The diarrhea
is often accompanied by abdominal pain and occasionally systemic symptoms such as fever, malaise, nausea, vomiting, and
loss of appetite. The symptoms begin ∼5 days after the ingestion
of oocysts but can begin up to 2 weeks later. The duration of
illness in immunocompetent hosts is variable, lasting from several days to 5 weeks [26]. It is worth noting that many exposed
patients are asymptomatic. In a study of experimental infection
in humans, one-third of the subjects who ingested oocysts had
no diarrhea [27].
Although the disease is self-limited in most patients, in several other situations, cryptosporidiosis can cause significant
morbidity and death. Before the widespread use of HAART,
cryptosporidiosis was a relatively common opportunistic infection in patients with advanced HIV infection. The risk of
acquiring cryptosporidiosis was correlated with lower numbers
of CD4⫹ T cells (⭐200 cells/mm3) [7]. In patients with advanced HIV infection, the illness has a highly variable presentation. In a cohort of HIV-infected patients studied in Baltimore
during 1985–1995, four patterns were described: transient infection, intermittent relapsing infection, cholera-like infection,
906 • CID 2003:36 (1 April) • FOOD SAFETY
and chronic infection [28]. Another well-described manifestation of Cryptosporidium infection in patients with AIDS is
cholangitis, which is seen more often with more-severe immunosuppression, defined as a CD4⫹ T cell count of ⭐50 cells/
mm3. These patients have a poor prognosis and often present
without the typical symptoms of abdominal pain [21]. Gastric
cryptosporidiosis and pancreatitis have also been described in
patients with AIDS. Other causes of immunodeficiency mentioned above have also been associated with these more recalcitrant forms of cryptosporidiosis [7].
Although Cryptosporidium infection does not have a predilection for any particular age group, it is common among children in developed and underdeveloped nations. Several published reports have shown that cryptosporidiosis may have
devastating long-term effects in children in poorer countries.
In Brazil, cryptosporidiosis and persistent diarrhea among children have been correlated with subsequent impairments in
physical fitness and diminished cognitive function [29]. Sadly,
it appears that these children with persistent diarrhea are caught
in a vicious cycle of malnutrition, placing them at risk for
further diarrheal illness, which in turn leads to progressively
worse nutritional status [30].
DIAGNOSIS
Cryptosporidiosis can be diagnosed via invasive or noninvasive
techniques. Intestinal analysis via biopsy with demonstration
of intracellular forms of the parasite is specific, but the diagnosis
may be missed, because most common sites of infection are
less accessible endoscopically. The “gold standard” and perhaps
most widely used test for the detection of Cryptosporidium
oocysts in stool remains the modified acid-fast or Kinyoun stain
(figure 1A). The test should be specifically requested, because
it will not be performed as part of a routine examination for
ova and parasites. Interpretation of the stained smear requires
experience, because other organisms in the stool may stain acid
fast. Several immunofluorescent assays (figure 1B) and EIA kits
have become commercially available and show promising sensitivity and specificity. These tests use antibodies against Cryptosporidium antigens to detect the parasite in stool specimens.
One of these kits, the ColorPAC Cryptosporidium/Giardia rapid
assay (Becton-Dickinson), was recently recalled because of a
cluster of false-positive results [31]. PCR-based techniques also
accurately detect the parasite in environmental samples and
stool specimens but have yet to be standardized for routine
clinical use and are not yet commercially available.
3.
4.
5.
6.
7.
THERAPY
8.
There is no effective antimicrobial agent for treatment of cryptosporidiosis in humans, and, consequently, the US Food and
Drug Administration has not approved a medication for this
purpose. Paromomycin and, recently, nitazoxanide have been
reported to have some clinical efficacy. The results of clinical
trials of these 2 drugs have been mixed. Azithromycin in combination with paromomycin appeared to have a demonstrable
effect in one uncontrolled trial involving patients with AIDS
who have cryptosporidiosis. Finally, resolution of cryptosporidiosis has been demonstrated after reconstitution of the immune system when HAART is used to treat advanced HIV
infection [21].
PREVENTION
Patients with underlying immune system weaknesses are at
risk for the more severe complications of Cryptosporidium
infection, as mentioned above. In the absence of effective,
specific therapy against infection with this parasite, preventative measures are of great importance among this patient
population. Such measures include extensive hand washing,
avoiding direct contact with stool from animals or humans,
avoiding the accidental ingestion of water used in recreational
activities, and taking measures to ensure the safety of the
drinking water. It should be noted that the quality of the local
drinking water is regionally and seasonally variable. Local
public health and municipal water authorities can provide
specific information about the safety of the water supply.
Cryptosporidium species can be removed from drinking water
by either boiling for 1 min or by filtering the water through
a filter with a pore size of !1 mm. These recommendations
are well summarized on the Centers for Disease Control
and Prevention’s Web site (http://www.cdc.gov/ncidod/dpd/
parasites/cryptosporidiosis/factsht_crypto_prevent_ci.htm#12).
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
References
1. Tyzzer EE. A sporozoon found in the peptic glands of the common
mouse. Proc Soc Exp Biol Med 1907; 5:12–3.
2. Nime FA, Burek JD, Page DL, Holscher MA, Yardley JH. Acute en-
25.
terocolitis in a human being infected with the protozoan Cryptosporidium. Gastroenterology 1976; 70:592–8.
MacKenzie WR, Schell WL, Blair KA, et al. Massive outbreak of waterborne cryptosporidium infection in Milwaukee, Wisconsin: recurrence of illness and risk of secondary transmission. Clin Infect Dis
1995; 21:57–62.
Guerrant RL. Cryptosporidiosis: an emerging, highly infectious threat.
Emerg Infect Dis 1997; 3:51–7.
Fayer R, Morgan U, Upton SJ. Epidemiology of Cryptosporidium: transmission, detection and identification. Int J Parasitol 2000; 30:1305–22.
Griffiths JK. Human cryptosporidiosis: epidemiology, transmission,
clinical disease, treatment, and diagnosis. Adv Parasitol 1998; 40:37–85.
Hunter PR, Nichols G. Epidemiology and clinical features of Cryptosporidium infection in immunocompromised patients. Clin Microbiol
Rev 2002; 15:145–54.
Peng MM, Xiao L, Freeman AR, et al. Genetic polymorphism among
Cryptosporidium parvum isolates: evidence of two distinct human transmission cycles. Emerg Infect Dis 1997; 3:567–73.
Quintero-Betancourt W, Peele ER, Rose JB. Cryptosporidium parvum
and Cyclospora cayetanensis: a review of laboratory methods for detection of these waterborne parasites. J Microbiol Methods 2002; 49:
209–24.
Rose JB, Huffman DE, Gennaccaro A. Risk and control of waterborne
cryptosporidiosis. FEMS Microbiol Rev 2002; 26:113–23.
Dillingham R, Lima A, Guerrant R. Cryptosporidiosis: epidemiology
and impact. Microbes Infect 2002; 4:1059–66.
Quiroz ES, Bern C, MacArthur JR, et al. An outbreak of cryptosporidiosis linked to a foodhandler. J Infect Dis 2000; 181:695–700.
Widmer G, Lin L, Kapur V, Feng X, Abrahamsen M. Genomics and
genetics of Cryptosporidium parvum: the key to understanding cryptosporidiosis. Microbes Infect 2002; 4:1081–90.
Morgan UM, Xiao L, Fayer R, Lal AA, Thompson RC. Variation in
Cryptosporidium: towards a taxonomic revision of the genus. Int J
Parasitol 1999; 29:1733–51.
Strong WB, Gut J, Nelson RG. Cloning and sequence analysis of a highly polymorphic Cryptosporidium parvum gene encoding a 60-kilodalton glycoprotein and characterization of its 15- and 45-kilodalton
zoite surface antigen products. Infect Immun 2000; 68:4117–34.
Leav BA, Mackay MR, Anyanwu A, et al. Analysis of sequence diversity
at the highly polymorphic Cpgp40/15 locus among Cryptosporidium
isolates from human immunodeficiency virus–infected children in
South Africa. Infect Immun 2002; 70:3881–90.
Feng X, Rich SM, Tzipori S, Widmer G. Experimental evidence for
genetic recombination in the opportunistic pathogen Cryptosporidium
parvum. Mol Biochem Parasitol 2002; 119:55–62.
Widmer G, Tzipori S, Fichtenbaum CJ, Griffiths JK. Genotypic and
phenotypic characterization of Cryptosporidium parvum isolates from
people with AIDS. J Infect Dis 1998; 178:834–40.
Xiao L, Bern C, Limor J, et al. Identification of 5 types of Cryptosporidium parasites in children in Lima, Peru. J Infect Dis 2001; 183:
492–7.
Lumadue JA, Manabe YC, Moore RD, Belitsos PC, Sears CL, Clark
DP. A clinicopathologic analysis of AIDS-related cryptosporidiosis.
AIDS 1998; 12:2459–66.
Chen XM, Keithly JS, Paya CV, LaRusso NF. Cryptosporidiosis. N Engl
J Med 2002; 346:1723–31.
Tzipori S, Ward H. Cryptosporidiosis: biology, pathogenesis and disease. Microbes Infect 2002; 4:1047–58.
Riggs M. Recent advances in cryptosporidiosis: the immune response.
Microbes Infect 2002; 4:1067–80.
Lean IS, McDonald V, Pollok RC. The role of cytokines in the pathogenesis of Cryptosporidium infection. Curr Opin Infect Dis 2002; 15:
229–34.
Chappell CL, Okhuysen PC, Sterling CR, Wang C, Jakubowski W,
Dupont HL. Infectivity of Cryptosporidium parvum in healthy adults
with pre-existing anti–C. parvum serum immunoglobulin G. Am J Trop
Med Hyg 1999; 60:157–64.
FOOD SAFETY • CID 2003:36 (1 April) • 907
26. Jokipii L, Jokipii AM. Timing of symptoms and oocyst excretion in
human cryptosporidiosis. N Engl J Med 1986; 315:1643–7.
27. DuPont HL, Chappell CL, Sterling CR, Okhuysen PC, Rose JB, Jakubowski W. The infectivity of Cryptosporidium parvum in healthy volunteers. N Engl J Med 1995; 332:855–9.
28. Manabe YC, Clark DP, Moore RD, et al. Cryptosporidiosis in patients
with AIDS: correlates of disease and survival. Clin Infect Dis 1998; 27:
536–42.
29. Guerrant DI, Moore SR, Lima AA, Patrick PD, Schorling JB, Guerrant
RL. Association of early childhood diarrhea and cryptosporidiosis with
908 • CID 2003:36 (1 April) • FOOD SAFETY
impaired physical fitness and cognitive function 4–7 years later in a
poor urban community in northeast Brazil. Am J Trop Med Hyg
1999; 61:707–13.
30. Lima AA, Moore SR, Barboza MS Jr, et al. Persistent diarrhea signals
a critical period of increased diarrhea burdens and nutritional shortfalls: a prospective cohort study among children in northeastern Brazil.
J Infect Dis 2000; 181:1643–51.
31. Manufacturer’s recall of rapid assay kits based on false positive Cryptosporidium antigen tests—Wisconsin, 2001–2002. MMWR Morb Mortal Wkly Rep 2002; 51:189.