Short report: asymptomatic Cryptosporidium hominis infection among human immunodeficiency virus-infected patients in Tanzania.

Abstract

Few data exist on the relative importance of individual Cryptosporidium species in acquired immunodeficiency syndrome cryptosporidiosis. We characterized 127 inpatients infected with human immunodeficiency virus (HIV) in Tanzania for their CD4 cell count and by stool analysis, including Cryptosporidium immunofluorescence and polymerase chain reaction-restriction fragment length polymorphism. Cryptosporidium was detected in patients both with and without diarrheal symptoms (defined as > or = 3 liquid stools/day, 11 of 61 versus 11 of 66; P = not significant) and was a marker for low CD4 cell count (median = 124/microL versus 212/microL in Cryptosporidium-negative patients; P < 0.04). Cryptosporidium hominis was the predominant species in this region and was associated with a longer duration of symptoms, a higher rate of asymptomatic infection, and a lower CD4 cell count versus C. parvum-infected patients (P < 0.05). This study suggests there may be important differences in the natural history of Cryptosporidium infection in HIV-infected persons depending on parasite species.

Full text

Page 1

SHORT REPORT: ASYMPTOMATIC CRYPTOSPORIDIUM HOMINIS INFECTION
AMONG HUMAN IMMUNDEFICIENCY VIRUS–INFECTED PATIENTS IN TANZANIA
ERIC R. HOUPT,* OLUMA Y. BUSHEN, NOEL E. SAM, ANITA KOHLI, AMON ASGHARPOUR, CHERIE T. NG,
DAVID P. CALFEE, RICHARD L. GUERRANT, VENANCE MARO, SENDUI OLE-NGUYAINE, AND JOHN F. SHAO
Division of Infectious Diseases and International Health, University of Virginia, Charlottesville, Virginia; Kilimanjaro Christian
Medical Centre, Moshi, Tanzania; Division of Infectious Diseases, Mount Sinai School of Medicine, New York, New York
Abstract.
ficiency syndrome cryptosporidiosis. We characterized 127 inpatients infected with human immunodeficiency virus
(HIV) in Tanzania for their CD4 cell count and by stool analysis, including Cryptosporidium immunofluorescence and
polymerase chain reaction−restriction fragment length polymorphism. Cryptosporidium was detected in patients both
with and without diarrheal symptoms (defined as ? 3 liquid stools/day, 11 of 61 versus 11 of 66; P ? not significant) and
was a marker for low CD4 cell count (median ? 124/?L versus 212/?L in Cryptosporidium-negative patients; P < 0.04).
Cryptosporidium hominis was the predominant species in this region and was associated with a longer duration of
symptoms, a higher rate of asymptomatic infection, and a lower CD4 cell count versus C. parvum-infected patients
(P < 0.05). This study suggests there may be important differences in the natural history of Cryptosporidium infection
in HIV-infected persons depending on parasite species.
Few data exist on the relative importance of individual Cryptosporidium species in acquired immunode-
Cryptosporidium is a major enteric pathogen of patients
with acquired immunodeficiency syndrome (AIDS), with in-
fection rates of 8−48% reported among African AIDS pa-
tients with diarrhea.1–4Studies of patients with cryptosporidi-
osis indicate that a majority experience chronic diarrhea,
while less than 15% have transient diarrhea or are asymp-
tomatic.5,6However, this asymptomatic rate may be an un-
derestimate because relatively few AIDS patients without di-
arrhea have been tested for Cryptosporidium infection in such
studies. Additionally, the relative clinical impact of the two
major Cryptosporidium species that infect humans, C. homi-
nis (also referred to as genotype 1) and C. parvum (also re-
ferred to as genotype 2), is poorly defined. A predominance
of C. hominis infection has been observed in HIV-associated
diarrhea patients in Peru7and Thailand,8C. parvum has been
prevalent in Europe,9,10while genotype data from Africa
have been limited to only a few cases.11−13
We performed a cross-sectional study of HIV-suspected
inpatients in the Kilimanjaro region of Tanzania at the Kili-
manjaro Christian Medical Centre, the Mawenzi Government
Hospital, and the Kibong’oto National Tuberculosis Hospital
from July to October 2002. Informed consent was obtained
from all participants and the University of Virginia Human
Investigation Committee and the Kilimanjaro Christian
Medical Centre Ethics Committee reviewed and approved
the project. The hospitals’ standard HIV testing and counsel-
ing procedures of the hospitals were maintained and no pa-
tient was tested for HIV for the purpose of entry into the
study. Patients with previously documented or newly sus-
pected HIV infection were eligible. Enrolled patients were
asked if they were experiencing increased stool frequency
(? 3 liquid stools/day), the duration if present, and a single
blood and stool sample was obtained. This study took place
before the era of widespread availability of anti-retroviral
therapy in this region.
Stool specimens were tested for Cryptosporidium infection
by immunofluorescence (IF) microscopy (MeriFluor™ Cryp-
tosporidium/Giardia; Meridian Bioscience, Inc., Cincinnati,
OH) and by a nested polymerase chain reaction−restriction
fragment length polymorphism (PCR-RFLP) according to the
protocol of Sturbaum and others.14DNA was isolated from
aliquots of frozen stool using the QiAMP DNA stool mini kit
(Qiagen, Valencia, CA) after two washes in sterile phosphate-
buffered saline (PBS) and six cycles of freeze-thaw. For each
PCR, water (no DNA) and positive control DNA from C.
parvum oocysts were used to confirm absence of contamina-
tion and successful amplification, respectively. The PCR assay
exhibited a sensitivity of 94% and a specificity of 96% versus
the reference IF assay15and detected five IF-negative infec-
tions. All positive PCR results were repeated for verification
and a positive PCR or IF result was considered valid. In ad-
dition to Giardia, which was tested by the IF assay, other stool
pathogens were evaluated as follows: an enzyme-linked im-
munosorbent assay (ELISA) for Entamoeba histolytica (E.
histolytica II ELISA assay; Techlab, Blacksburg, VA), culture
for enteric bacteria using Selenite F broth and deoxycholate
citrate and MacConkey agars, and microscopy for ova and
parasites on formalin-ether sediments.
Two hundred forty-five patients were enrolled and com-
plete data were obtained from 156, of which 127 were HIV-
1/2 positive by ELISA (Vironostika Uniform HIV plus
O; Organon Teknika B.V., Boxtel, The Netherlands or
GenScreen HIV 1/2, Bio-Rad Laboratories, Marnes la Co-
quette, France). Quantification of CD4 T cells was performed
on EDTA-containing blood within four hours of collection
using the Coulter Manual CD4 Count kit16(Beckman
Coulter, Hialeah, FL). Cryptosporidium infection was a
marker for HIV infection in that 22 of 127 HIV-positive pa-
tients were infected versus 0 of 29 HIV-suspected but HIV-
negative patients (P ? 0.01). Clinical features of the HIV-
positive population are summarized in Table 1. At least 53%
(67 of 127) of patients met the AIDS surveillance case defi-
nition via a CD4 cell count < 200/?L.17Diarrheal symptoms
(? 3liquidstools/day)werereportedby48%ofthepatientsand
had been present for an average of duration of one month.
Upon examining the relationships between Cryptosporidi-
um infection, diarrheal symptoms, and CD4 cell count (Table
* Address correspondence to Eric R. Houpt, Division of Infectious
Diseases and International Health, University of Virginia, PO Box
801340, MR4 Building Room 2144, Charlottesville, VA 22908-1340.
E-mail: erh6k@virginia.edu
Am. J. Trop. Med. Hyg., 73(3), 2005, pp. 520–522
Copyright © 2005 by The American Society of Tropical Medicine and Hygiene
520

Page 2

2), we found that Cryptosporidium-infected patients had
lower CD4 cell counts than Cryptosporidium-negative pa-
tients (median ? 124/?L versus 212/?L; P < 0.04). Similar to
the findings of others,1,2we found that diarrheal symptoms
also correlated with a low CD4 cell count (median ? 139/?L
versus 226/?L; P ? 0.02). We were surprised to find that
asymptomatic infection with Cryptosporidium was common
(11 of 66 versus the 11 of 61 symptomatic Cryptosporidium
infection rate; P not significant), particularly given the low
CD4 cell counts (median ? 128/?L) of these patients. Indeed,
this low CD4 cell count was a distinguishing feature of Cryp-
tosporidium-infected asymptomatic patients (median ? 128/
?L versus 336/?L for Cryptosporidium-negative asymptom-
atic patients; P ? 0.02), indicating that Cryptosporidium in-
fection was a marker for advanced lymphopenia independent
of diarrhea. Cryptosporidium loads estimated by the method
of Goodgame and others18were variable and did not signifi-
cantly differ between patients with symptoms and those with-
out symptoms (mean ± SD/mL of sediment ? 15,000 ± 43,000
versus 5,600 ± 12,000; P not significant). Of note, only 1 of 11
Cryptosporidium-infected patients with diarrheal symptoms
was co-infected with other enteric pathogens as tested (one
with a Giardia co-infection and none with E. histolytica,
Strongyloides, Salmonella, or Shigella co-infections).
Since Cryptosporidium was an important marker in this
population for advanced CD4 lymphopenia, we sought to de-
termine whether this risk was influenced by Cryptosporidium
species. Of the 21 stool specimens that were PCR-positive for
Cryptosporidium (one specimen was PCR-/IF+), RFLP analy-
sis showed that 15 were C. hominis and 6 were C. parvum
(Table 3). There was no association between species and di-
arrheal symptoms. However, the duration of symptoms in C.
hominis-infected patients was longer than that of C. parvum-
infected patients, and the rate of asymptomatic C. hominis
infection was higher (14% versus 2% for C. parvum; P ?
0.02). Interestingly, the median CD4 cell count of C. hominis-
infected patients was lower than that of C. parvum-infected
patients, and this was true for both symptomatic and asymp-
tomatic groups. Again, there was no statistical difference in
the average parasite load between C. hominis and C. parvum
(mean ± SD/mL of sediment ? 10,000 ± 37,000 versus 5,000
± 8,000; P not significant).
The most important new findings in this study are the high
rate of asymptomatic C. hominis infection in this population
and the apparent preferential association of C. hominis infec-
tion with a low CD4 cell count. The 17% rate of asymptom-
atic Cryptosporidium carriage in this study is higher than the
1−5% asymptomatic rate previously reported in AIDS inpa-
tients from many developing world sites.2,19,20The high rate
was not due to our use of a PCR-based assay (since 9 of 10
asymptomatic patients were also IF positive) and is surprising
given the low CD4 cell count of this asymptomatic group
(128/?L), since other studies have suggested that chronic
symptoms are the norm in this CD4 range.5We believe there
may be an underappreciation of asymptomatic Cryptosporid-
ium infection, at least in this region of Africa, given that most
clinical data on cryptosporidiosis derives from series of pa-
tients with diarrhea and not asymptomatic controls.3,21−23It
may be relevant that community-based studies from Bolivia24
and Korea25have shown asymptomatic Cryptosporidium car-
riage rates of 8−32% in large numbers of healthy individuals.
Another feature of our patient population was the low Cryp-
tosporidium oocyst loads relative to what others have ob-
served,18which could be contributing to the relative lack of
symptoms, or simply reflect biological variability or labora-
tory differences.
Important limitations of our study are the small sample
size, the inherent risk of recall bias when evaluating diarrheal
symptoms, and the lack endoscopy or serial stool examina-
tion. We are interested in determining if these asymptomatic
Cryptosporidium-infected patients were transiently or chroni-
cally infected, and whether they were forever symptomatic,
went on to become symptomatic in the future, or had symp-
toms in the remote past. One could speculate that the pref-
erential detection of C. hominis in patients with low CD4 cell
count is due to delayed clearance of the organism, perhaps
TABLE 2
Relationships between Cryptosporidium infection, symptoms, and CD4 cell count*
? 3 liquid stools/dayAsymptomaticTotal
Cryptosporidium positive
Cryptosporidium negative
Total
11 (CD4 ? 124)
50 (CD4 ? 150)
61 (CD4 ? 139§)
11 (CD4 ? 128)
55 (CD4 ? 336‡)
66 (CD4 ? 226§)
22 (CD4 ? 124†)
124 (CD4 ? 212†)
* CD4 cell values are reported as median/?L; the value was not available for one Cryptosporidium-positive asymptomatic patient. Statistical analyses were based on the Mann-Whitney test.
† P < 0.04, comparing the median CD4 cell count of Cryptosporidium-positive vs. negative patients.
‡ P < 0.04, comparing the median CD4 cell count of Cryptosporidium-negative asymptomatic patients versus positive/asymptomatic, negative/symptomatic, or positive/symptomatic groups.
§ P ? 0.02, comparing the median CD4 cell count of patients with vs. without symptoms.
TABLE 3
Comparison of Cryptosporidium hominis versus C. parvum infection*
C. hominisC. parvum
Total
Association with symptoms
Duration of symptoms (months)
Association with asymptomatic
infection
CD4 cell count
15 (71%)
6/61 (10%)
4†
9/66‡ (14%)
6 (29%)
5/61 (8%)
0.3†
1/66‡ (2%)
55§256§
* CD4 cell counts/?L and duration of symptoms are reported as median values; the CD4
cell count was not available for one asymptomatic C. hominis-infected patient.
† P ? 0.009 comparing duration of symptoms of C. hominis vs. C. parvum-infected pa-
tients (by Mann-Whitney test).
‡ P ? 0.02 comparing rates of asymptomatic infection with C. hominis vs. C. parvum (by
Fischer’s exact test).
§ P < 0.003 comparing CD4 cell counts of C. hominis vs. C. parvum-infected patients (by
Mann-Whitney test).
TABLE 1
Clinical features of the human immunodeficiency virus–infected
study population (n ? 127) Kilimanjaro, Tanzania
Feature
Age, years (median, range)
Sex (M:F)
Cryptosporidium infection
CD4 cell count (median/?L)
Symptoms of ? 3 liquid stools/day
35 (18–65)
47:80
22 (17%)
179
61 (48%);
median duration ? 1 month
ASYMPTOMATIC CRYPTOSPORIDIOSIS WITH AIDS
521

Page 3

from symptomatic infections acquired during the previous
wet season when transmission and new infection are com-
mon.26Highly powered, prospective longitudinal studies are
needed to confirm our findings of asymptomatic C. hominis
carriage with low CD4 cell count, and to determine whether
such a relationship is cause versus effect.
Received June 23, 2004. Accepted for publication November 11,
2004.
Acknowledgments: We thank Edward Mushi, Sr., Ruwaichi Uiso,
and the other nurses involved in this study; Stanislaus Siriwa, Richard
Tarimo, and Eline Ngomuo and the other clinical laboratory techni-
cians; and all the patients who participated in this work. We also
thank Dr. William Petri (University of Virginia) for helpful discus-
sions and David Lyerly (Techlab, Inc.) for the E. histolytica II ELISA
kits.
Financial support: This study was supported by the National Insti-
tutes of Health (grant U19 AI056872-01) and the Virginia Common-
wealth Technology Research Fund.
Disclosure: None of the authors has a commercial or other associa-
tion that might pose a conflict of interest.
Authors’ addresses: Eric R. Houpt, Oluma Y. Bushen, Anita Kohli,
Amon Asgharpour, Cherie T. Ng, and Richard L. Guerrant, Division
of Infectious Diseases and International Health, University of Vir-
ginia, PO Box 801340, MR4 Building Room 2144, Charlottesville, VA
22908-1340, Telephone: 434-243-9326, Fax: 434-924-0075, E-mails:
erh6k@virginia.edu, ob3d@vurginia.edu, ak7t@virginia.edu,
aa3z@virginia.edu, ctn8d@virginia.edu, and rlg9a@virginia.edu. Noel
E. Sam, Venance Maro, Sendui Ole-Nguyaine, and John F. Shao,
Kilimanjaro Christian Medical Centre, Moshi, Tanzania, E-mails:
nsam@kcmc.ac.tz, venmaro@yahoo.co.uk, solengudr@yahoo.com,
and jshao@kcmc.ac.tz. David P. Calfee, Division of Infectious Dis-
eases, Mount Sinai School of Medicine, New York, NY 10029, E-mail:
David.Calfee@msnyuhealth.org.
REFERENCES
1. Tarimo DS, Killewo JZ, Minjas JN, Msamanga GI, 1996. Preva-
lence of intestinal parasites in adult patients with enteropathic
AIDS in north-eastern Tanzania. East Afr Med J 73: 397–399.
2. Gassama A, Sow PS, Fall F, Camara P, Gueye-N’diaye A, Seng
R, Samb B, M’Boup S, Aidara-Kane A, 2001. Ordinary and
opportunistic enteropathogens associated with diarrhea in
Senegalese adults in relation to human immunodeficiency vi-
rus serostatus. Int J Infect Dis 5: 192–198.
3. Gumbo T, Sarbah S, Gangaidzo IT, Ortega Y, Sterling CR, Car-
ville A, Tzipori S, Wiest PM, 1999. Intestinal parasites in pa-
tients with diarrhea and human immunodeficiency virus infec-
tion in Zimbabwe. AIDS 13: 819–821.
4. Colebunders R, Lusakumuni K, Nelson AM, Gigase P, Lebughe
I, van Marck E, Kapita B, Francis H, Salaun JJ, Quinn TC, Piot
P, 1988. Persistent diarrhoea in Zairian AIDS patients: an en-
doscopic and histological study. Gut 29: 1687–1691.
5. Blanshard C, Jackson AM, Shanson DC, Francis N, Gazzard BG,
1992. Cryptosporidiosis in HIV-seropositive patients. QJM 85:
813–823.
6. Manabe YC, Clark DP, Moore RD, Lumadue JA, Dahlman HR,
Belitsos PC, Chaisson RE, Sears CL, 1998. Cryptosporidiosis
in patients with AIDS: correlates of disease and survival. Clin
Infect Dis 27: 536–542.
7. Cama VA, Bern C, Sulaiman IM, Gilman RH, Ticona E, Vivar
A, Kawai V, Vargas D, Zhou L, Xiao L, 2003. Cryptosporid-
ium species and genotypes in HIV-positive patients in Lima,
Peru. J Eukaryot Microbiol 50 (Suppl): 531–533.
8. Tiangtip R, Jongwutiwes S, 2002. Molecular analysis of Crypto-
sporidium species isolated from HIV-infected patients in Thai-
land. Trop Med Int Health 7: 357–364.
9. Guyot K, Follet-Dumoulin A, Lelievre E, Sarfati C, Rabodoni-
rina M, Nevez G, Cailliez JC, Camus D, Dei-Cas E, 2001.
Molecular characterization of Cryptosporidium isolates ob-
tained from humans in France. J Clin Microbiol 39: 3472–3480.
10. Alves M, Matos O, Pereira Da Fonseca I, Delgado E, Lourenco
AM, Antunes F, 2001. Multilocus genotyping of Cryptosporid-
ium isolates from human HIV-infected and animal hosts.
J Eukaryot Microbiol Suppl: 17S–18S.
11. Gatei W, Greensill J, Ashford RW, Cuevas LE, Parry CM, Cun-
liffe NA, Beeching NJ, Hart CA, 2003. Molecular analysis of
the 18S rRNA gene of Cryptosporidium parasites from pa-
tients with or without human immunodeficiency virus infec-
tions living in Kenya, Malawi, Brazil, the United Kingdom, and
Vietnam. J Clin Microbiol 41: 1458–1462.
12. Leav BA, Mackay MR, Anyanwu A, RM OC, Cevallos AM,
Kindra G, Rollins NC, Bennish ML, Nelson RG, Ward HD,
2002. Analysis of sequence diversity at the highly polymorphic
Cpgp40/15 locus among Cryptosporidium isolates from human
immunodeficiency virus-infected children in South Africa. In-
fect Immun 70: 3881–3890.
13. Morgan U, Weber R, Xiao L, Sulaiman I, Thompson RC, Ndiritu
W, Lal A, Moore A, Deplazes P, 2000. Molecular character-
ization of Cryptosporidium isolates obtained from human im-
munodeficiency virus-infected individuals living in Switzer-
land, Kenya, and the United States. J Clin Microbiol 38: 1180–
1183.
14. Sturbaum GD, Reed C, Hoover PJ, Jost BH, Marshall MM, Ster-
ling CR, 2001. Species-specific, nested PCR-restriction frag-
ment length polymorphism detection of single Cryptosporidi-
um parvum oocysts. Appl Environ Microbiol 67: 2665–2668.
15. Garcia LS, Shimizu RY, 1997. Evaluation of nine immunoassay
kits (enzyme immunoassay and direct fluorescence) for detec-
tion of Giardia lamblia and Cryptosporidium parvum in human
fecal specimens. J Clin Microbiol 35: 1526–1529.
16. Carella AV, Moss MW, Provost V, Quinn TC, 1995. A manual
bead assay for the determination of absolute CD4+ and CD8+
lymphocyte counts in human immunodeficiency virus-infected
individuals. Clin Diagn Lab Immunol 2: 623–625.
17. Centers for Disease Control and Prevention, 1993. 1993 revised
classification system for HIV infection and expanded surveil-
lance case definition for AIDS among adolescents and adults.
JAMA 269: 729−730.
18. Goodgame RW, Genta RM, White AC, Chappell CL, 1993. In-
tensity of infection in AIDS-associated cryptosporidiosis. J In-
fect Dis 167: 704–709.
19. Cranendonk RJ, Kodde CJ, Chipeta D, Zijlstra EE, Sluiters JF,
2003. Cryptosporidium parvum and Isospora belli infections
among patients with and without diarrhoea. East Afr Med J 80:
398–401.
20. Wuhib T, Silva TM, Newman RD, Garcia LS, Pereira ML,
Chaves CS, Wahlquist SP, Bryan RT, Guerrant RL, Sousa Ade
Q, de Queiroz TRBS, Sears CL, 1994. Cryptosporidial and
microsporidial infections in human immunodeficiency virus-
infected patients in northeastern Brazil. J Infect Dis 170: 494–
497.
21. Kelly P, Baboo KS, Ndubani P, Nchito M, Okeowo NP, Luo NP,
Feldman RA, Farthing MJ, 1997. Cryptosporidiosis in adults in
Lusaka, Zambia, and its relationship to oocyst contamination
of drinking water. J Infect Dis 176: 1120–1123.
22. Gomez Morales MA, Atzori C, Ludovisi A, Rossi P, Scaglia M,
Pozio E, 1995. Opportunistic and non-opportunistic parasites
in HIV-positive and negative patients with diarrhoea in Tan-
zania. Trop Med Parasitol 46: 109–114.
23. Cegielski JP, Ortega YR, McKee S, Madden JF, Gaido L,
Schwartz DA, Manji K, Jorgensen AF, Miller SE, Pulipaka
UP, Msengi AE, Mwakyusa DH, Sterling CR, Reller LB, 1999.
Cryptosporidium, enterocytozoon, and cyclospora infections in
pediatric and adult patients with diarrhea in Tanzania. Clin
Infect Dis 28: 314–321.
24. Esteban JG, Aguirre C, Flores A, Strauss W, Angles R, Mas-
Coma S, 1998. High Cryptosporidium prevalences in healthy
Aymara children from the northern Bolivian Altiplano. Am J
Trop Med Hyg 58: 50–55.
25. Yu JR, Lee JK, Seo M, Kim SI, Sohn WM, Huh S, Choi HY, Kim
TS, 2004. Prevalence of cryptosporidiosis among the villagers
and domestic animals in several rural areas of Korea. Korean
J Parasitol 42: 1–6.
26. Newman RD, Zu SX, Wuhib T, Lima AA, Guerrant RL, Sears
CL, 1994. Household epidemiology of Cryptosporidium par-
vum infection in an urban community in northeast Brazil. Ann
Intern Med 120: 500–505.
HOUPT AND OTHERS
522