MINIREVIEW

Detection of Intestinal Protozoa in the Clinical Laboratory

Ian H. McHardy,a Max Wu,a Robyn Shimizu-Cohen,a Marc Roger Couturier,b,c Romney M. Humphriesa
Pathology and Laboratory Medicine, UCLA, Los Angeles, CA, USAa; Associated Regional and University Pathologists Laboratories, Institute for Clinical and Experimental
Pathology, Salt Lake City, Utah, USAb; Department of Pathology, University of Utah School of Medicine, Salt Lake City, Utah, USAc

Despite recent advances in diagnostic technology, microscopic examination of stool specimens remains central to the diagnosis
of most pathogenic intestinal protozoa. Microscopy is, however, labor-intensive and requires a skilled technologist. New, highly
sensitive diagnostic methods have been developed for protozoa endemic to developed countries, including Giardia lamblia (syn.
G. intestinalis/G. duodenalis) and Cryptosporidium spp., using technologies that, if expanded, could effectively complement or
even replace microscopic approaches. To date, the scope of such novel technologies is limited and may not include common pro-
tozoa such as Dientamoeba fragilis, Entamoeba histolytica, or Cyclospora cayetanensis. This minireview describes canonical ap-
proaches for the detection of pathogenic intestinal protozoa, while highlighting recent developments and FDA-approved tools
for clinical diagnosis of common intestinal protozoa.

P rotozoan infections significantly contribute to the burden of

gastrointestinal illness worldwide. While the prevalence of
these infections is low in the United States, sporadic outbreaks,
Army Institute of Research, the Naval Medical Research Unit, the
Joint Pathology Center (previously the Armed Forces Institute of
Pathology [AFIP]), and the Centers for Disease Control and Pre-
including the 2013 outbreak of cyclosporiasis in the United States, vention (CDC) DPDx laboratories. Laboratories may also con-
underscore the continued burden of disease these organisms pres- sider pooling resources on a local level, both for training purposes
ent in developed countries. Giardia, Cryptosporidium spp., Dien- and to share specimens for competency. In the authors’ laborato-
tamoeba fragilis, Entamoeba spp. (including nonpathogenic spe- ries, positive specimens are reviewed by all trained technologists to
cies), Blastocystis spp., and Cyclospora cayetanensis are the most maximize staff competency.
common pathogenic protozoa reported in developed settings (1). Long-term solutions to these challenges include lessening lab-
However, accurate determination of the incidence of these infec- oratory reliance on the O&P for the diagnosis of intestinal proto-
tions is hampered by infrequent testing of stool for protozoa when zoa; indeed, some people have already suggested limiting the use
patients present with gastroenteritis (1), by inappropriate test or- of the O&P in routine clinical practice (4). Antigen detection tests
dering by physicians (1, 2, 3), and by the lack of sensitive tech- for Giardia, Cryptosporidium spp., and Entamoeba histolytica have
niques by which to identify pathogenic protozoa in stool speci- been cleared by the U.S. FDA (Table 2) and are associated with
mens. significant improvements in the detection of these organisms in
The microscopic ova and parasite examination (O&P) is the stool. Unfortunately, no FDA-cleared antigen test detects D. fra-
traditional method for stool parasite testing. Although the O&P is gilis, which is a pathogenic protozoa frequently detected in many
labor-intensive and requires a high level of skill for optimal inter- U.S. laboratories (R. M. Humphries and M. R. Couturier, unpub-
pretation, this test remains the cornerstone of diagnostic testing lished observations). Regardless, some have suggested the use of
for the intestinal protozoa. At present, most clinical microbiology algorithmic testing that involves front-line antigen testing for
laboratories in the United States struggle with the ability to pro- Giardia and Cryptosporidium. If the results of such testing are
vide quality O&P results within a clinically significant time frame negative, traditional microscopic approaches are used (5). Suc-
(Table 1). A pressing concern for these laboratories is the shortage cessful implementation of such a system would likely require de-
of skilled technologists capable of reliably evaluating O&P. As the veloping a physician guidance tool to aid in appropriate ordering,
baby boomer generation retires from the workforce, inexperi- as the laboratory very rarely receives the information required to
enced technologists, who in some instances are inadequately determine if the test is requested in the clinical context of gastro-
trained in parasitology, are left to fill the void. Few laboratories in intestinal complaints or as part of the evaluation of a returning
the United States encounter a sufficient number of specimens that traveler, immigrant, or patient prior to transplantation. Further-
harbor intestinal protozoa to maintain technologist proficiency, more, such algorithmic testing delays diagnosis of pathogens for
let alone to allow for robust training of new technologists. As such, which the laboratory has not initially tested.
laboratories may be unable to accurately identify pathogenic pro- There is a pressing need for newer diagnostic test options to
tozoa, differentiate these from nonpathogenic species, and dis-
criminate artifacts on O&P examinations. Further, in many un-
derstaffed laboratories, the labor-intensive O&P is performed Published ahead of print 6 November 2013
only once other laboratory tasks are completed, yielding long Editor: G. V. Doern
turnaround times and limiting this test’s clinical utility. Address correspondence to Romney M. Humphries,
To address competency issues, some laboratories have devel- [email protected].
oped affiliations with organizations that conduct parasitology sur- I.H.M. and M.W. contributed equally to this article.
veillance in regions of disease endemicity around the world and Copyright © 2014, American Society for Microbiology. All Rights Reserved.
have unique access to clinical specimens for teaching and training doi:10.1128/JCM.02877-13
purposes. Examples of such organizations are the Walter Reed

712 jcm.asm.org Journal of Clinical Microbiology p. 712–720 March 2014 Volume 52 Number 3
 Minireview

TABLE 1 Top five challenges faced by the clinical laboratory in the that a single stool specimen submitted for microscopic examina-
detection of intestinal protozoa, as identified by the authors tion will detect 58 to 72% of protozoa present (4, 7). Hiatt and
Challenges colleagues found that evaluating three specimens, as opposed to
1. Reliance on labor-intensive, technically demanding tests (e.g., O&P) one, resulted in an increased yield of 22.7% for E. histolytica,
• O&P testing is left until other laboratory testing is completed, yielding 11.3% for Giardia, and 31.1% for D. fragilis (8). As such, many
long turnaround times, due to the misguided notion this testing is laboratories continue to request 3 specimens be collected and sub-
“less critical” than others mitted for testing; specimen collection is made, optimally, every
• Many laboratories do not have technologists that can reliably identify other day, over a period of up to 10 days (6). However, alternative
pathogens and differentiate these from nonpathogenic species or approaches have been proposed to help curtail unnecessary test-
artifacts ing, including application of an algorithm that requires a negative
2. Reliance on insensitive tests specimen and persistence of symptoms before a second or third
• O&P is associated with a sensitivity of 20 to 90% compared to
specimen is analyzed by the laboratory (4). Specimens may also be
molecular assays
• Some antigen detection tests, e.g., those for Cryptosporidium spp., are
pooled prior to screening based on microscopy. In contrast, the
insensitive enhanced sensitivity of molecular detection methods may require
3. Shortage of clinical specimens positive for intestinal protozoa only 1 specimen for testing to achieve sensitivity equal to, if not
• Limits the opportunities for adequate training greater than, microscopy. One study demonstrated a 14% increase
• Limits ability of technologists to maintain proficiency in yield for gastrointestinal protozoa when a real-time PCR was
• Limits validation of new testing platforms and transport medium performed on a single stool specimen, compared to microscopy
4. Shortage of training programs/resources for parasitology on three specimens (5).
• Confounded by the retirement of experienced technologists who would
otherwise perform training STOOL PRESERVATION
5. Suboptimal physician ordering practices
While visualization of motility in unpreserved specimens may fa-
• Few physicians will order organism-specific tests, even during outbreaks
• Inadequate access to patient information by laboratory prevents
cilitate diagnosis, this technique is impractical for most laborato-
implementation of algorithmic testing ries, as transport of fresh stool to the laboratory for testing is rarely
within the requisite time frame for examination (i.e., 30 to 60
min). A variety of stool fixatives have been developed and modi-
fied in recent decades for use with traditional microscopic exam-
replace the O&P. Such tests should broadly detect most, if not all, ination. Those that remain widely used and commercially avail-
pathogens commonly identified microscopically. Multiplexed able include formalin, sodium acetate-acetic acid-formalin (SAF),
PCR has the potential to meet this need. However, only one such Schaudinn’s fluid, polyvinyl alcohol-containing fixatives (mer-
assay has been cleared by the U.S. FDA to date, the highly multi- cury, copper, or zinc based), and mercury-free/formalin-free fix-
plexed Luminex xTAG GPP, which detects Giardia and Crypto- atives. A two-vial collection system, consisting of one vial contain-
sporidium in addition to numerous bacterial and viral targets. ing 5 to 10% buffered formalin for use in concentrated wet
Such molecular assays, depending on their design, may require a mounts and a second vial containing a polyvinyl alcohol-based
laboratory with proficiency in molecular testing, which would preservative for permanent stained smears, is considered the “gold
limit their use to major academic hospitals and reference labora- standard.” However, concern over working with toxic formalin in
tories. Alternatively, sample-to-answer solutions, which provide the laboratory and the environmental impact and disposal costs
direct diagnosis from unprocessed samples, such as the BioFire associated with the use of mercury-based fixatives have led many
Diagnostics FilmArray platform, could be used in virtually any to consider alternate preservatives and single-tube collection sys-
laboratory setting. tems (9). SAF may be used to achieve this goal, if coupled with iron
Despite the challenges outlined in Table 1, detection of intestinal hematoxylin for the permanent stained smear; however, for labo-
protozoa is still almost exclusively based on O&P microscopic exam- ratories desiring to maintain the trichrome stain, SAF is not a valid
ination. This article will thus review optimal diagnostic approaches option, as poor-quality results have been documented with this
and the microscopic morphology of key pathogenic protozoa. The combination.
pathogenesis of some protozoa discussed is controversial, including Alternative stool preservatives. Zinc- and copper-based poly-
that of Blastocystis hominis and Dientamoeba fragilis. Other common vinyl alcohol (PVA) formulations have been developed and are
protozoa, such as Endolimax nana, are not discussed herein, as less is commercially available to replace the mercury-based fixatives (10,
known about their potential virulence. Antigen and molecular-based 11). In a paired study that evaluated 106 specimens prepared using
detection methods are also summarized. zinc sulfate-PVA versus mercuric chloride-PVA with trichrome
stain, 92.5% overall agreement was reported in the overall mor-
SPECIMEN COLLECTION phology and numbers of organisms detected between the two
Optimal recovery and microscopic identification of protozoa methods (11); in contrast, a study by the same group noted poor
from patients with intestinal infections is dependent on proper preservation of protozoa morphology when a copper-based PVA
collection and preservation of fecal specimens. Well-recognized formulation was evaluated (10). Examples of commercial speci-
factors that influence the sensitivity of parasite examinations in- men collection kits using modification to the mercuric chloride
clude patient medications, specimen collection interval, and the PVA include ProtoFix (AlphaTec, Vancouver, WA), which con-
preservation of stool prior to testing (6). The diagnostic yield of tains no mercury and minimal formalin; EcoFix (Meridian Bio-
the O&P is also significantly impacted by the number of stool science, Cincinnati, OH), which contains neither mercury nor
specimens collected and submitted to the laboratory for testing. formalin; and ParaSafe (Cruinn Diagnostic, Dublin, Ireland),
Many intestinal protozoa are irregularly shed, and data suggest which also does not contain mercury or formalin. A study con-

March 2014 Volume 52 Number 3 jcm.asm.org 713

714
Minireview

jcm.asm.org
TABLE 2 Sensitivities and specificities of FDA-approved assays for molecular and serologic detection of intestinal protozoan parasitesa
Assay type (% sensitivity/% specificity)
Target organism Stool enzyme immunoassay Immunochromatography Direct fluorescent antibody Other
Entamoeba histolytica TechLab Entamoeba histolytica II (100/94.7); Serum-based EIAs: Bordier Affinity
Cellabs Entamoeba CELISA Path Entamoeba histolytica IgG (100/
(93–100/93–100) 80–96); NovaTec Entamoeba
histolytica IgG (95/95); Sciemedx
Corp. Entamoeba histolytica
antibody detection test (92/100)
E. histolytica (possibly E. dispar)
Remel ProSpect Entamoeba histolytica
(87/99)
Giardia lamblia Remel ProSpecT Giardia (EZ) (96–98/98); Remel Xpect Giardia (97.9/97.1) Cellabs Giardia-CELb (100/100)
Remel ProSpecT Giardia IFU (98–100/
98–100); Medical Chemical Para-Tect
Giardia (85/95.9); Cellabs Giardia-
CELISAb (98–100/100); TechLab Giardia
IIb (100/100)
Cryptosporidium spp. Remel ProSpecT Cryptosporidium (97/96– Cellabs Crypto-CEL (100/100)
10); Medical Chemical Para-Tect
Cryptosporidium (100/97–100)
Cryptosporidium spp. and Giardia Remel ProSpecT Giardia/Cryptosporidium Meridian ImmunoCard Stat Crypto/Giardia Meridian Merifluor Cryptosporidium/Giardiab Multiplex PCR Luminex xTAG
intestinalisc (97.7–99.2/99.6); TechLab (97.3–100/100); Remel Xpect Giardia/ (97–100/94–100); Medical Chemical Para- gastrointestinal pathogen panel
Giardia/Cryptosporidium Chekb Cryptosporidium (95.8–96.4/98.5); Tect Cryptosporidium/Giardiab (100/100); (GPP) (95–100/89–100)
b
(97.6/100) TechLab Giardia/Cryptosporidium Quick Cellabs Crypto/Giardia-CEL (100/100)
b
Chek (98.9/100)
Giardia intestinalis E. histolytica Biosite Diagnostics Triage parasite panel
(possibly E. dispar) (90.5–95.1/85–88.4)
Cryptosporidium parvumc
a
Sensitivity and specificity values were obtained from package inserts of each product. Performance characteristics for these tests in individual laboratories may vary from the data presented in the package inserts.
b
Only detects Giardia cysts.
c
The ranges of sensitivity and specificity for all targets are listed.

Journal of Clinical Microbiology

 Minireview

TABLE 3 Common fixatives used to preserve ova and parasites in stool

Potential for
Preservative Downstream preparations Downstream assaysa single-vial use Notes
Formalin, 5% or 10% Only concentrated wet mount EIA, FA, IC Poor Poor NAT potential, poor trophozoite
preservation
SAF Permanent stained smear and EIA, FA, IC Fair Poor NAT potential, suboptimal
concentrated wet mount trophozoite morphology
Mercury-based fixative with PVA Permanent stained smear and NAT Poor Immunoassays not possible, fixative is
concentrated wet mount highly toxic
(rare)
Modified Schaudinn’s (copper, zinc, Permanent stained smear and NAT Fair Immunoassays not possible,
or other fixative with PVA) concentrated wet mount concentrated wet mounts are
(rare) uncommonly performed,
suboptimal trophozoite
morphology
Single -vial proprietary fixative Permanent stained smear and Some immunoassays are Good Suboptimal trophozoite morphology,
formulations concentrated wet mount possible; most NATs not all immunoassays are possible
a
Abbreviations: EIA, enzyme immunoassay; FA, fluorescent antibody; IC, immunochromographic test; NAT, nucleic acid amplification test.

ducted by the CDC evaluated the performance of these preserva- DETECTION OF SPECIFIC INTESTINAL PROTOZOA
tives head to head with the traditional two-vial set of formalin and Giardia lamblia (syn. Giardia intestinalis and Giardia duode-
mercuric chloride-PVA. This study found EcoFix and ProtoFix, nalis). Giardiasis is a common gastrointestinal parasitic infection
but not ParaSafe, yielded an acceptable morphological quality to associated with diarrhea, stomach cramps, upset stomach, and
the preserved parasites on concentrated wet mounts compared to excessive gas. Annually, roughly 20,000 U.S. cases of giardiasis are
formalin-fixed specimens. EcoFix alone yielded satisfactory pro- reported to the CDC, but these are estimated to comprise as little
tozoan morphology on the permanent stained smears, compared as 1 to 10% of the total infection burden, despite being a nationally
with stool preserved in mercuric chloride-PVA (9). In contrast, a notifiable disease (13). While numerous diagnostic tests are avail-
separate study found significantly (P ⬍ 0.001) reduced recovery of able for Giardia, its highly distinctive morphology facilitates mi-
B. hominis and Endolimax nana in 261 EcoFix-preserved concen- croscopic diagnosis. Giardia cysts can be observed in fresh smears,
trates compared to formalin-fixed stool concentrates (12). Al- on formalin-ethyl acetate or permanent stained smear, although
though the manufacturer of EcoFix has developed a proprietary the latter is associated with a higher sensitivity for identification.
stain, EcoStain, the conventional trichrome stain can be used with Trophozoites are not always found in stool, as encystation begins
EcoFix and has been shown to produce comparable protozoan before passage through the colon. In cases where Giardia is sus-
morphology (12). Total-Fix (Medical Chemical Corporation, pected but not detected in stool, duodenal specimens, such as
Torrance, CA) is a relatively new, FDA-approved mercury-, for- those collected by a string test, may be used for permanent stains
malin-, and PVA-free fixative. Similar to EcoFix, specimens pre- and concentrated wet mounts. Tear drop-shaped trophozoites
pared by using Total-Fix can be used for concentration, perma- range from 10 to 20 ␮m in length, 9 to 12 ␮m in width, and
nent stain, and a variety of immunoassays for detecting protozoa, contain two nuclei, a sucking disk, 4 pairs of flagella, 2 axonemes,
though there have been no published reports describing the per- and 2 median bodies. Cysts contain 4 nuclei, 4 axonemes, and 4
formance of this fixative compared to others to date. Table 3 sum- median bodies and range from 11 to 14 ␮m in length and 7 to 10
marizes many available fixatives used by clinical laboratories and ␮m in width (Fig. 1E).
highlights possible preparations and downstream assays for each. While Giardia cysts are easily recognizable on permanent
A major impediment to replacing the traditional two-vial sys- stained smears, they are shed sporadically, and O&P examinations
tems of laboratories in the United States is the requirement for are often insufficient to demonstrate the presence of this organism
laboratories to perform a verification study to confirm the perfor- (14). Alles and colleagues demonstrated a sensitivity of 66.4% for
mance specifications of these products. Few institutions encoun- the detection of Giardia via a permanent stained smear, albeit
ter a sufficient number of positive clinical specimens to allow ro- chlorazol black stain was used as opposed to the more standard
bust evaluation of these preservatives. Furthermore, in order to trichrome, and the number of specimens tested per patient was
perform a method comparison study, specimens need to be col- not taken into account (15). Regardless, detection of Giardia is
lected in both fixatives, which may require preapproval or exemp- improved through the use of antigen detection assays, several of
tion status by local institutional review boards. Laboratories may which are commercially available and widely used in clinical lab-
thus need to develop creative means by which to evaluate these oratories across the United States. For example, in the aforemen-
fixatives prior to clinical use. A combination of approaches has tioned study by Alleles and colleagues, a sensitivity of 99.2% for
been used in our laboratories, including comparison of the mor- the detection of Giardia was observed via a commercial, direct
phology of white cells present in stool preserved in both fixatives, fluorescent antibody (DFA) test. Both the permanent stained
seeding fresh stool specimens with cultured protozoa, obtaining smear and the DFA were 100% specific for Giardia in the 2,696
veterinary specimens for testing, and consulting the published lit- stool specimens examined by this study (15). In addition to the
erature (if available) on the performance of these products. DFA, which requires laboratory access to a fluorescence micro-

March 2014 Volume 52 Number 3 jcm.asm.org 715

Minireview

FIG 1 Key microscopic morphology of the enteropathogenic protozoa. Organisms are ordered from largest to smallest, based on average cell size. (A)
Balantidium coli trophozoite unstained, wet mount. (B) Cystoisospora belli oocyst. (C, D, and F) Trophozoite forms are shown stained with trichrome for E.
histolytica (C), D. fragilis (D), and B. hominis (F). (E) Cyst form of Giardia stained with trichrome. (G and H) Cyclospora cayetanensis oocyst (G) and
Cryptosporidium spp. oocyst (H) after modified acid-fast staining.

scope, immunochromatographic (IC) tests, and enzyme immu- skilled technologist, permanent stained smears are only 34% sen-
noassays (EIAs) are commercially available for the detection of sitive compared to molecular methods (21).
Giardia (Table 2). IC tests are optimally suited for laboratories Cryptosporidium spp. Cryptosporidiosis is a gastrointestinal
with lower capacities for diagnostic complexity, while EIA-based infection caused by various species of Cryptosporidium. Fecal-oral
tests may be more appropriate for high-throughput screening in transmission via contaminated food, drinking water, or exposure
high-prevalence areas. A study comparing four EIAs, including in public swimming pools is responsible for most infections. Like
the FDA-approved ProSpecT (Remel, Lenexa, KS) and CELISA all coccidian intestinal parasites, the small and poorly staining
(Cellabs, Brookvale, NSW, Australia) assays, found sensitivities Cryptosporidium oocysts can be easily missed in routine O&P ex-
that ranged from 63 to 91% and specificities of ⱖ95% for all assays ams. Sensitivity of light microscopy is improved by performing
(16). A second study demonstrated 94 to 100% sensitivity and modified acid-fast (MAF) stains, though even this modification
100% specificity when 5 Giardia EIAs were evaluated with 100 has been shown to be associated with a sensitivity of only 54.8%
positive and 50 negative specimens (17). Table 2 provides an over- (15). Furthermore, MAF staining is typically only performed
view of many of the available FDA-approved EIAs and their re- upon physician request, or if the technologist detects structures
spective sensitivities and specificities, as determined by the man- suspicious for Cryptosporidium on the permanent stained smear.
ufacturer, for detection of Giardia either alone or in combination Unfortunately, many physicians assume that testing for Crypto-
with other pathogenic protozoa. sporidium is included with the routine O&P and infrequently or-
Dientamoeba fragilis. Dientamoebiasis is an enteric infection der specialized stains or Cryptosporidium immunoassays, even in
caused by the flagellate D. fragilis. Symptoms associated with in- outbreak situations (3). Upon MAF staining, Cryptosporidium
fection vary dramatically, with some individuals suffering nausea, spp. oocysts appear as bright red spheres (4 to 6 ␮m) containing
vomiting, and diarrhea containing mucous and including abdom- four crescent-shaped sporozoites (which may or may not be seen
inal discomfort, while others are asymptomatic. Accordingly, as in all oocysts) (Fig. 1H). Additionally, oocysts may also occlude
with the case of B. hominis, described below, there is some uncer- stain, resulting in transparent “ghost” cells.
tainty about the pathogenesis of D. fragilis. However, the morbid- As is the case for Giardia, sensitivity of detection is improved
ity associated with some infections justifies its inclusion as a de- when an EIA or DFA is used (Table 2). Multiple studies have
finitive pathogen (18). The prevalence of D. fragilis has been evaluated the sensitivities and specificities of the available kits and
estimated in many studies and ranges from 1.1 to 20% in patients found overall similar performance levels for EIA- and DFA-based
in the developed world with diarrhea, but its prevalence may be methods (sensitivity, ⬎90%; specificity, ⬎95% [17]). Rapid IC-
higher in select populations or if molecular methods are used for based methods are significantly less sensitive, with one multi-in-
detection (19). stitutional study reporting 50.1 to 86.7% sensitivity, dependent on
Despite this relatively high prevalence, no antigen-based, mo- the test manufacturer (22). Because HIV-infected and immuno-
lecular, or serologic diagnostics have been commercially devel- compromised individuals are particularly at risk for severe com-
oped to aid with laboratory identification. As such, detection of D. plications due to infection with these coccidian parasites, physi-
fragilis on the permanent stained smear is the current standard. cians should consider routinely ordering DFA at a minimum and
Unfortunately, D. fragilis is difficult to identify morphologically. molecular-based assays, if available, for patients with suspect
No cyst stage has been observed in humans, although a cyst stage cryptosporidiosis.
has been recently observed in mice (20). Trophozoites range from Giardia and Cryptosporidium spp. are two of the most common
5 to 15 ␮m in length, 9 to 12 ␮m in width, and contain 1 to 2 protozoan infections in the United States, and multiple combined
characteristically fragmented nuclei. While well-preserved speci- tests have been developed to facilitate rapid screening for both
mens might contain cells with the classically described tetrad nu- organisms simultaneously. Such tests include EIAs, IC assays,
clei (Fig. 1D), in general practice nuclei will only have visible holes DFA assays, and multiplex PCR assays. A comparison between
through the center of the nucleus. Given its indistinct appearance, several DFA tests and EIAs for Giardia and Cryptosporidium re-
diagnosis is often only possible by experienced technologists, lead- vealed that (i) DFA tests tend to have slightly higher sensitivity for
ing to many potentially missed infections. Even under ideal con- both organisms, (ii) the Merifluor Cryptosporidium/Giardia test
ditions, with prompt preservation of stool and evaluation by a had the highest sensitivity of the DFAs, and (iii) the specificities of

716 jcm.asm.org Journal of Clinical Microbiology

 Minireview

all tested EIA and DFA tests were 100% (17). However, these as- with intestinal infections by E. histolytica, serological tests are not
says do not detect D. fragilis and, as such, these tests do not replace typically informative in uncomplicated cases because seroconver-
the O&P for routine testing. sion is rare outside the context of extraintestinal involvement.
Cyclospora cayetanensis. Cyclosporiasis is usually a self-lim- Despite their microscopic morphological similarity to Entamoeba
iting gastroenteritis caused by the coccidian C. cayetanensis. Due dispar and Entamoeba moshkovskii, intestinal infections with E.
to poor uptake of most conventional stains by C. cayetanensis histolytica in nonendemic areas are still primarily diagnosed via
oocysts, microscopic detection can be challenging, but it remains microscopy on the permanent stained smear. Organisms may be
the recommended diagnostic method (14). C. cayetanensis oocysts accompanied by clubbed RBCs in cases of dysentery. On the per-
may stain irregularly by trichrome or the MAF stain. As is the case manent stained stool smear, E. histolytica trophozoites are 12 to 60
with Cryptosporidium, not all oocysts will take up these stains in a ␮m in diameter and contain a single, well-defined nucleus (Fig.
single smear, which may lead inexperienced technologists to over- 1C). Spherical cysts measure 12 to 15 ␮m, contain 2 to 4 nuclei,
look the organism. When observed, Cyclospora oocysts in stool are and occasionally have cigar-shaped, cytoplasmic chromatoidal
easily identified as 8- to 10-␮m refractile spheres with a central bars. Nuclei of both forms are surrounded by an obvious nuclear
morula, resembling wrinkled cellophane (Fig. 1G). If Cyclospora membrane, a compact, central karyosome, and evenly distributed
infection is specifically suspected (e.g., during established out- peripheral chromatin. Without evidence of erythrophagocytosis
breaks), use of a modified safranin staining protocol provides con- (which is seen most often in tissue specimens), E. histolytica is
sistent reddish-orange staining of oocysts and thus simplifies indistinguishable from E. dispar and should be annotated as E.
identification (23). In addition to the modified safranin stain, histolytica/dispar on the laboratory report. Ingested RBCs can only
oocysts of C. cayetanensis in a standard concentrated wet mount be definitively identified when concomitant extracellular RBCs
intrinsically autofluoresce white-blue under UV light when a 330- are visible. In cases of chronic amebic infection, ingested RBCs are
to 365-nm excitation filter is used. Less-intense, blue-green auto- infrequently observed, making differentiation from E. dispar dif-
fluorescence can be seen when a 450- to 490-nm excitation filter is ficult.
used. This property aids in the identification of Cyclospora; however, In areas of the world where E. histolytica infection is endemic or
all fluorescent structures should be visualized by light microscopy to if infection is specifically suspected by a physician, antigen-based
verify the morphology (http://www.asm.org/images/PSAB/Cyclospo tests can be performed, though these require unpreserved speci-
raWhitePaper2013.pdf). mens. E. histolytica antigen tests that are specific for E. histolytica
Relman et al. developed a nested PCR assay that targets the 18S employ monoclonal antibodies against the Gal/GalNAc-specific
rRNA gene that has been used in outbreak situations to confirm lectin expressed by E. histolytica. Not all commercially available
Cyclospora (23). Many other molecular techniques have been de- antigen tests differentiate between E. histolyica and E. dispar (Ta-
veloped for the identification of Cyclospora (1), but there are no ble 2). Sensitivity for the E. histolytica antigen detection tests has
FDA-approved or analyte-specific reagents for Cyclospora avail- been shown in several studies to range from 80 to 94% compared
able in the United States. The Biofire (Salt Lake City, UT) Film- to PCR, but one study found the TechLab enzyme-linked immu-
Array GI panel includes C. cayetanensis and is currently available nosorbent assay (ELISA) to be less sensitive than microscopy (29).
in the Unites States with research use only (RUO) status, but it is Examples of FDA-approved EIAs for Entamoeba spp. are included
in clinical trials for the FDA. in Table 2 along with their sensitivities and specificities, as defined
Cystoisospora belli. Cystoisosporiasis is a relatively uncom- in their package inserts.
mon gastroenteritis caused by the coccidian C. belli that can Diagnosis of disseminated amebiasis caused by E. histolytica is
result in cholera-like symptoms in up to 1% of HIV-infected or challenging because stool O&P examinations are almost always
otherwise-immunocompromised individuals (25). Detection of negative for these patients. When such cases are suspected, cecal or
oocysts from stool or duodenal samples is simplified by their dis- colonic endoscopy to look for hallmark lesions followed by endo-
tinctive size and shape. However, C. belli oocysts are only easily scopic biopsy to visualize the presence of E. histolytica trophozo-
recognizable in concentrated wet mounts from O&P exams. Im- ites are quite helpful (30). This algorithm has been shown to be
portantly, oocyst maturation continues postdefecation, and thus effective in differentiating amebic colitis from colon cancer and
morphology depends upon the duration between specimen col- uncomplicated colitis (31, 32). Sigmoidoscopy material may also
lection and preservation. If placed immediately into preservative, be submitted to the laboratory for permanent stained smear eval-
long oval-shaped C. belli oocysts (20 to 33 ␮m in length and 10 to uation. In patients with liver abscesses, serological assays are in-
19 ␮m in width) will contain a single circular immature sporo- formative due to the concomitant systemic exposure to amoebic
blast. If specimens are not quickly preserved, oocysts of roughly antigens (1); 95% of patients with extraintestinal disease will be
the same size and shape will contain 1 to 2 circular sporoblasts. positive by serology. When evaluating patients from areas where
While detection is relatively straightforward from concentrated E. histolytica infection is endemic, it is important to be aware that
wet mounts, modified acid-fast, safranin, or auramine rhodamine modern serological assays, which employ recombinant E. histo-
stains can be used to increase contrast and simplify detection, lytica antigens, will turn negative following abscess treatment ear-
although staining may interfere with sporoblast visualization (Fig. lier than the traditional indirect hemagglutination-based tests,
1B) (26, 27). Similar to Cyclospora, the oocysts of Cystoisopora will which remain positive for at least 6 months following treatment.
autofluoresce under the conditions described above. C. belli Serum and liver abscess aspirates from patients with disseminated
oocysts are not always found in stool, and examination of duode- E. histolytica have been subjected to off-label antigenic testing,
nal specimens collected by biopsy or string test may be necessary. with varying sensitivity.
Entamoeba histolytica. Roughly 50 million worldwide cases of Blastocystis hominis. The pathogenicity of B. hominis is largely
amoebic dysentery and 100,000 deaths are associated with E. his- controversial, given that it is commonly identified in nonsymp-
tolytica annually (28). Despite the extreme morbidity associated tomatic individuals. Some experts hypothesize that B. hominis

March 2014 Volume 52 Number 3 jcm.asm.org 717

Minireview

should be split into multiple species, some of which are more pathogens, including Giardia and Cryptosporidium spp. This assay
pathogenic than others, though few studies have been performed is the first molecular method approved by the FDA for the detec-
to confirm this hypothesis (33). The continuing uncertainty is tion of pathogenic protozoa. The analyte-specific reagents (ASRs)
primarily due to the fact that all isolates of Blastocystis are mor- for the xTAG assay were recently evaluated; while the overall num-
phologically similar and are occasionally found in combination ber of positive specimens was low in this study (5 to 20 positives),
with other protozoan infections. However, in the absence of anti- the ASRs were highly sensitive and specific for Cryptosporidium
gen detection or molecular diagnostics, the standard method for (95% sensitivity and 99% specificity), Giardia (95% sensitivity
detection is still microscopy. While B. hominis is visible on wet and 99% specificity), and E. histolytica (100% sensitivity and 89%
mounts, definitive identification is easier with permanent stained specificity) (36). The FDA-approved version of the assay does not
smears. B. hominis is typically 6 to 40 ␮m in diameter with a large include E. histolytica, but the reagents for this analyte are available
central body surrounded by up to six small nuclei (Fig. 1F). The for research use only.
large central body often stains a characteristic red, green, or blue in BioFire Diagnostics has in development a sample-to-answer
trichrome-stained samples. Development of nonmicroscopic and gastrointestinal pathogen panel that includes detection of Gi-
molecular strategies for diagnosis will likely hinge on whether ardia, Cryptosporidium, E. histolytica, and Cyclospora cayentensis.
studies can effectively differentiate pathogenic versus nonpatho- Whether the company will be able to collect sufficient numbers of
genic strains (33). When observed on routine O&P, B. hominis specimens positive for each target to garner FDA clearance or if
should be reported, along with a semiquantitative assessment. some will remain RUO remains to be seen. Like the Luminex
Balantidium coli. Balantidiasis is an intestinal parasitic disease panel, this platform does not include detection of D. fragilis, which
that is associated with ciliated B. coli trophozoites, which typically is one of the most commonly encountered protozoa in the United
only affect immunocompromised or malnourished individuals States.
and have a worldwide distribution (34). Like many other intesti- One major critique for these multiplex panels is the cost per
nal protozoa, no established molecular or serologic tests are avail- test, which is many times higher than the reagents associated with
able for B. coli. Instead, microscopic diagnosis is facilitated by its performing the O&P. However, if an assay were to replace the
distinctive size and morphology on concentrated wet mounts; di- O&P examination, the savings in labor, from our perspective,
agnosis from permanent stains is not recommended, because tro- would far outweigh the cost associated with performing a multi-
phozoites absorb large amounts of dye, which can mask its char- plex commercial test.
acteristic features.
B. coli is the largest infectious intestinal protozoan, at 50 to 100 SUMMARY
␮m in length and 40 to 70 ␮m in width. Trophozoites have fine, In summary, adequate diagnosis of intestinal protozoa by the clinical
visible cilia and a large, kidney-bean-shaped macronucleus (Fig. laboratory is limited by many factors (Table 1). There is increasing
1A). A single, polar cystosome, or oral groove, can also be detected demand for low-complexity, high-throughput, and cost-effective
on some cells. The cyst form also has a visible macronucleus, but is complements to (or replacements for) the labor-intensive microsco-
smaller (50 to 70 ␮m long, 40 to 60 ␮m wide) and rounder than py-based approaches to protozoan diagnosis. While efforts in this
the trophozoites. Cysts have a thick cyst wall and often do not have regard have been slow to come, many diagnostic manufacturers are
visible cilia. While molecular or serologic-based diagnostics might rising to the challenge, including Luminex and BioFire. These efforts
improve detection sensitivity compared to microscopic diagnosis, may restore or enhance the abilities of laboratories to identify these
development of such tests has been a low priority due to the rela- pathogens, yielding increased knowledge on the present state of these
tive simplicity of microscopic detection and infrequency of infec- diseases in the United States and other countries.
tion in the United States.
ACKNOWLEDGMENT
IMPLICATIONS FOR FUTURE DIAGNOSTICS
We thank David A. Bruckner for his insightful review of the manuscript.
As discussed above and documented in recent studies, multiplex
PCR assays are both more sensitive and specific than microscopy REFERENCES
for the detection and identification of pathogenic protozoa (35).
1. Fletcher SM, Stark D, Harkness J, Ellis J. 2012. Enteric protozoa in the
However, despite a rapidly growing field of molecular and genetic developed world: a public health perspective. Clin. Microbiol. Rev. 25:
technologies for the clinical microbiology laboratory, diagnostic 420 – 449. http://dx.doi.org/10.1128/CMR.05038-11.
developments for intestinal protozoan parasites have remained 2. Centers for Disease Control and Prevention. 2013. Notes from the field:
relatively stagnant. Challenges associated with developing a re- use of electronic messaging and the news media to increase case finding
during a Cyclospora outbreak—Iowa, July 2013. MMWR Morb. Mortal.
placement test for the O&P includes coverage of all pathogenic Wkly. Rep. 62:613– 614. http://www.cdc.gov/mmwr/preview/mmwrhtml
species and the potential for long-term, residual detection of pre- .mm6230a4.htm.
vious infections. Furthermore, while analyte-specific approaches 3. Polage CR, Stoddard GJ, Rolfs RT, Petti CA. 2011. Physician use of
may yield enhanced sensitivity for pathogenic protozoa, docu- parasite tests in the United States from 1997 to 2006 and in a Utah Cryp-
tosporidium outbreak in 2007. J. Clin. Microbiol. 49:591–596. http://dx
mentation of the presence of human cells (white blood cells and
.doi.org/10.1128/JCM.01806-10.
erythrocytes), Charcot-Leyden crystals, and nonpathogenic pro- 4. Branda JA, Lin TY, Rosenberg ES, Halpern EF, Ferraro MJ. 2006. A
tozoa is lost. In particular, some physicians interpret the presence rational approach to the stool ova and parasite examination. Clin. Infect.
of nonpathogenic protozoa as indicative of patient exposure to Dis. 42:972–978. http://dx.doi.org/10.1086/500937.
contaminated food or water, although there are no studies that 5. Bruijnesteijn van Coppenraet LE, Wallinga JA, Ruijs GJ, Bruins MJ,
Verweij JJ. 2009. Parasitological diagnosis combining an internally con-
have clearly demonstrated this to be fact. trolled real-time PCR assay for the detection of four protozoa in stool
The Luminex xTAG gastrointestinal pathogen panel has re- samples with a testing algorithm for microscopy. Clin. Microbiol. Infect.
ceived FDA approval and can simultaneously detect 14 entero- 15:869 – 874. http://dx.doi.org/10.1111/j.1469-0691.2009.02894.x.

718 jcm.asm.org Journal of Clinical Microbiology

 Minireview

6. Garcia LS. 2009. Practical guide to diagnostic parasitology. ASM Press, and real-time PCR for the detection of Dientamoeba fragilis in clinical
Washington, DC. stool samples. Eur. J. Clin. Microbiol. Infect. Dis. 29:411– 416. http://dx
7. Nazer H, Greer W, Donnelly K, Mohamed AE, Yaish H, Kagalwalla A, .doi.org/10.1007/s10096-010-0867-4.
Pavillard R. 1993. The need for three stool specimens in routine labora- 22. Agnamey P, Sarfati C, Pinel C, Rabodoniriina M, Kapel N, Dutoit E,
tory examinations for intestinal parasites. Br. J. Clin. Pract. 47:76 –78. Garnaud C, Diouf M, Garin JF, Totet A, Derouin F, Cryptosporidium
8. Hiatt RA, Markell EK, Ng E. 1995. How many stool examinations are National Network ANOFEL. 2011. Evaluation of four commercial rapid
necessary to detect pathogenic intestinal protozoa? Am. J. Trop. Med. immunochromatographic assays for detection of Cryptosporidium anti-
Hyg. 53:36 –39. gens in stool samples: a blind multicenter trial. J. Clin. Microbiol. 49:
9. Pietrzak-Johnston SM, Bishop H, Wahlquist S, Moura H, Da Silva ND, 1605–1607. http://dx.doi.org/10.1128/JCM.02074-10.
Nguyen-Dihn P. 2000. Evaluation of commercially available preservatives 23. Visvesvara GS, Moura H, Kovacs-Nace E, Wallace S, Eberhard ML.
for laboratory detection of helminths and protozoa in human fecal spec- 1997. Uniform staining of Cyclospora oocysts in fecal smears by a modi-
imens. J. Clin. Microbiol. 38:1959 –1964. http://dx.doi.org/10.1128/JCM fied safranin technique with microwave heating. J. Clin. Microbiol. 35:
.38.5.1959-1964.2000. 730 –733.
10. Garcia LS, Shimizu RY, Brewer TC, Bruckner DA. 1983. Evaluation of 24. Relman DA, Schmidt TM, Gajadhar A, Sogin M, Cross J, Yoder K,
intestinal parasite morphology in polyvinyl alcohol preservative: compar- Sethabutr O, Echeverria P. 1996. Molecular phylogenetic analysis of
ison of copper sulfate and mercuric chloride bases for use in Schaudinn Cyclospora, the human intestinal pathogen, suggests that it is closely re-
fixative. J. Clin. Microbiol. 17:1092–1095. lated to Eimeria species. J. Infect. Dis. 173:440 – 445.
11. Garcia LS, Shimizu RY, Shum A, Bruckner DA. 1993. Evaluation of 25. Lindsay DS, Dubey JP, Blagburn BL. 1997. Biology of Isospora spp. from
intestinal protozoan morphology in polyvinyl alcohol preservative: com- humans, nonhuman primates, and domestic animals. Clin. Microbiol.
parison of zinc sulfate- and mercuric chloride-based compounds for use Rev. 10:19 –34.
in Schaudinn’s fixative. J. Clin. Microbiol. 31:307–310. 26. DPDx. 2009. Key points for laboratory diagnosis of cystoisosporiasis.
12. Garcia LS, Shimizu RY. 1998. Evaluation of intestinal protozoan mor- Centers for Disease Control and Prevention, Atlanta, GA. http://dpd.cdc
phology in human fecal specimens preserved in EcoFix: comparison of .gov/dpdx/HTML/Cystoisosporiasis.htm.
Wheatley’s trichrome stain and EcoStain. J. Clin. Microbiol. 36:1974 – 27. Cuomo MJN, Lawrence B, White DB. 2012. Diagnosing medical para-
1976. sites: a public health officers guide to assisting laboratory and medical
13. Feng Y, Xiao L. 2011. Zoonotic potential and molecular epidemiology of officers. USAF Air Education and Training Command, Randolph, TX.
Giardia species and giardiasis. Clin. Microbiol. Rev. 24:110 –140. http://dx http://www.phsource.us/PH/PARA/.
.doi.org/10.1128/CMR.00033-10. 28. WHO/PAHO/UNESCO. 1997. WHO/PAHO/UNESCO report. A con-
14. Clinical and Laboratories Standards Institute. 1997. Procedures for the sultation with experts on amoebiasis. Mexico City, Mexico, 28 –29 Janu-
recovery and identification of parasites from the intestinal tract; approved ary, 1997. Epidemiol. Bull. 18:13–14.
guideline, 2nd ed. CLSI document M28-A2. Clinical and Laboratories 29. Fotedar R, Stark D, Beebe N, Marriott D, Ellis J, Harkness J. 2007.
Standards Institute, Wayne, PA. Laboratory diagnostic techniques for Entamoeba species. Clin. Microbiol.
15. Alles AJ, Waldron MA, Sierra LS, Mattia AR. 1995. Prospective com- Rev. 20:511–532. http://dx.doi.org/10.1128/CMR.00004-07.
parison of direct immunofluorescence and conventional staining meth- 30. Blumencranz H, Kasen L, Romeu J, Waye JD, LeLeiko NS. 1983. The
ods for detection of Giardia and Cryptosporidium spp. in human fecal role of endoscopy in suspected amebiasis. Am. J. Gastroenterol. 78:15–18.
specimens. J. Clin. Microbiol. 33:1632–1634. 31. Abe T, Kawai N, Yasumaru M, Mizutani M, Akamatsu H, Fujita S,
16. Maraha B, Buiting AG. 2000. Evaluation of four enzyme immunoassays Nishida T, Iijima H, Tsujuii M, Tsujimoto M. 2009. Ameboma mim-
for the detection of Giardia lamblia antigen in stool specimens. Eur. icking colon cancer. Gastrointest. Endosc. 69:757–758. http://dx.doi.org
J. Clin. Microbiol. Infect. Dis. 19:485– 487. http://dx.doi.org/10.1007 /10.1015/j.gie.2008.12.065.
/s1009600000286. 32. Upadhyay R, Gupta N, Gogia P, Chandra S. 2012. Poached egg appear-
17. Garcia LS, Shimizu RY. 1997. Evaluation of nine immunoassay kits (en- ance in intestinal amebiasis. Gastrointest. Endosc. 76:189 –190. http://dx
zyme immunoassay and direct fluorescence) for detection of Giardia lam- .doi.org/10.1016/j.gie.2012.03.171.
blia and Cryptosporidium parvum in human fecal specimens. J. Clin. 33. Tan KS. 2008. New insights on classification, identification, and clinical
Microbiol. 35:1526 –1529. relevance of Blastocystis spp. Clin. Microbiol. Rev. 21:639 – 665. http://dx
18. Johnson EH, Windsor JJ, Clark CG. 2004. Emerging from obscurity: .doi.org/10.1128/CMR.00022-08.
biological, clinical, and diagnostic aspects of Dientamoeba fragilis. Clin. 34. Schuster FL, Ramirez-Avila L. 2008. Current world status of Balantidium
Microbiol. Rev. 17:553–570. http://dx.doi.org/10.1128/CMR.17.3.55-570 coli. Clin. Microbiol. Rev. 21:626 – 638. http://dx.doi.org/10.1128/CMR
.2004. .00021-08.
19. Barratt JL, Harkness J, Marriott D, Ellis JT, Stard D. 2011. A review of 35. Stark D, Al-Qassab SE, Barratt JL, Stanley K, Roberts T, Marriott D,
Dientamoeba fragilis carriage in humasn: several reasons why this organ- Harkness J, Ellis JT. 2011. Evaluation of multiplex tandem real-time PCR
ism shoul be condired in the diagnosis of gastrointestinal illness. Gut for detection of Cryptosporidium spp., Dientamoeba fragilis, Entamoeba
Microbes 2:3–12. http://dx.doi.org/10.4161/gmic.2.1.14755. histolytica, and Giardia intestinalis in clinical stool samples. J. Clin. Micro-
20. Munasinghe VS, Vella NG, Ellis JT, Windsor PA, Stark D. 2013. Cyst biol. 49:257–262. http://dx.doi.org/10.1128/JCM.01796-10.
formation and faecal-oral transmission of Dientamoeba fragilis: the miss- 36. Navidad JF, Griswold DJ, Gradus MS, Bhattacharyya S. 2013. Evalua-
ing link in the life cycle of an emerging pathogen. Int. J. Parasitol. 43:879 – tion of Luminex xTag gastrointestinal pathogen analyte-specific reagents
883. http://dx.doi.org/10.1016/j.ijpara.2013.06.003. for high-throughput, simultaneous detection of bacteria, viruses, and par-
21. Stark D, Barratt J, Roberts T, Marriott D, Harkness J, Ellis J. 2010. asites of clinical and public health importance. J. Clin. Microbiol. 51:
Comparison of microscopy, two xenic culture techniques, conventional 3018 –3024. http://dx.doi.org/10.1128/JCM.00896-13.

Continued next page

March 2014 Volume 52 Number 3 jcm.asm.org 719

Minireview

Romney M. Humphries, Ph.D., D(ABMM), is

an Assistant Professor in the Department of Pa-
thology and Laboratory Medicine in the David
Geffen School of Medicine at the University of
California, Los Angeles. She is also the Section
Chief for Clinical Microbiology at the UCLA
Health System, and program codirector for
UCLA’s CPEP fellowship program. Dr. Rom-
ney received her undergraduate degree in bio-
chemistry from the University of Lethbridge,
Canada. She completed her Ph.D. in bacterial
pathogenesis in the laboratory of Glen Armstrong, at the University of Cal-
gary, Canada. There, she investigated novel anti-infective strategies for the
prevention of bacterial gastroenteritis caused by enteropathogenic Esche-
richia coli. Dr. Romney subsequently completed a CPEP fellowship in Med-
ical and Public Health Microbiology at the University of California, Los
Angeles, under the direction of Michael Lewinski. One of Dr. Romney’s
research focuses is the detection and characterization of pathogens that cause
gastrointestinal diseases.

720 jcm.asm.org Journal of Clinical Microbiology