CRYPTOSPORIDIUM is a coccidian parasite that infects a broad range of vertebrates including mammals, birds, reptiles, amphibians, and fish. To date, 15 valid species are recognized within the genus Cryptosporidium (Fayer, Santin, and Xiao 2005; Ryan et al. 2004; Xiao et al. 2004). Differentiation between Cryptosporidium species is important such as distinguishing the anthroponotic C. hominis from C. parvum in outbreak situations. Traditional criteria such as host range and distinct morphological features used for identifying species are not fully reliable for identifying Cryptosporidium at the species level. As molecular methods can reveal greater differences than morphology, these approaches are now preferred for identifying Cryptosporidium species and/or genotypes. Human cryptosporidiosis is zoonotic with reservoirs is mainly found in cattle and humans. The parasite is endemic in many developing countries but it can also appear in outbreaks and epidemics in developed countries. Immunosuppressed adults and children are more susceptible to the infection and most of the reported cases are associated with contaminated water (Leav, Mackay, and Ward 2003). Cryptosporidium hominis and C. parvum are the most prevalent species associated with human disease (McLauchlin et al. 2000; Xiao et al. 2001) but other Cryptosporidium species have recently been identified in stools of both immunosuppressed and immunocompetent patients (Xiao et al. 2004). Identification of species other than C. hominis and C. parvum in humans is important to better understand the epidemiology of cryptosporidiosis. Human cryptosporidiosis is usually associated with acute diarrhea illness. In immunocompetent hosts, it is usually self-limiting, whereas in immunocompromised individuals, the infection is often chronic and can be severe and life threatening. In HIV-positive patients with diarrhea, prevalence rates vary between 24% in developing countries to 13.8% in developed countries. In contrast, prevalence in HIV-negative patients presenting with diarrhea is about 6.1% in developing countries and 2.1% in developed countries (Dillingham, Lima, and Guerrant 2002). There is limited data available on Cryptosporidium species infecting humans in Colombia. Previous studies demonstrated the presence of the parasite in rivers and in drinking water in the city of Bogotá (Alarcon et al. 2005). In human serum samples obtained from subjects in the Andean region of Colombia, anti-Cryptosporidium antibodies were detected in approximately 83.3% of the study group (Vergara et al. 2000), and Cryptosporidium sero-positive samples were detected in only 3.6% of specimens obtained from immunosuppressed patients in Antioquia (Botero et al. 2003). To evaluate the prevalence of intestinal cryptosporidiosis and to identify the species and genotypes responsible for infections in HIV-positive patients in Medellín, a prospective and descriptive study of a selected population of HIV-positive patients was carried out at the “Grupo Interdisciplinario para el Estudio de las Parasitosis Intestinales” laboratory. Samples were obtained from institutions treating the HIV-positive patients of Medellín (Colombia) between August 2001 and September 2002. These samples were also evaluated for the presence of microsporidia and other parasites (Botero et al. 2004). Fecal samples and microscopy. One hundred and three HIV-positive patients between the ages of 2 and 74 years were sent to the “Grupo Interdisciplinario para el Estudio de las Parasitosis Intestinales” laboratory. They were evaluated and data in a clinical and epidemiological survey were recorded. Serial stool samples were obtained and each sample was immediately concentrated upon collection. Feces consistency was abnormal in all cases, watery in 81.8% and loose in 13.6% of the patients. CD4 cell levels were determined in 17 of the patients: 1–50 cells/mm3 (three patients), 51–200 cells/mm3 (nine patients), 200–400 cells/mm3 (three patients) and >400 cells/mm3 (two patients). Clinical symptoms were recorded for 21 patients (Table 1). Oocysts in fecal samples were concentration by the formalin–ethyl acetate protocol and smears were prepared. The samples were screened for the presence of Cryptosporidium oocysts using the Kinyoun acid-fast stain and confirmed by direct immunofluorescence (Crypto-Giardia DFA, Meridian Diagnostics, Cincinnati, OH). Aliquots of these samples were preserved in ethanol at 4°C until processing. DNA extraction and PCR amplification. Oocyst disruption and DNA purification were performed using the Fast DNA kit and the FP120 disrupter (QBioGene Inc., Vista, CA), as described previously (da Silva et al. 1999). Extracted DNA was stored at 4°C until analysis by PCR amplification. Polymerase chain reaction of the small subunit ribosomal RNA (SSrRNA) and the Cryptosporidium oocyst wall protein (COWP) gene fragments were performed. The fragment of the 18S rRNA region was analyzed using the nested-PCR described by Tiangtip and Jongwutiwes (2002). Cry15 and Cry9 (region coding the N-terminal domain) and Cry12 and Cry14 (region coding the C-terminal domain), were used as the COWP gene primers (da Silva et al. 1999). The PCR products were analyzed by agarose gel electrophoresis and visualized after the gel was stained with ethidium bromide. DNA sequence analysis. Amplicons obtained were purified using the StrataPrep PCR Purification kit (Stratagene Inc., La Jolla, CA) and then sequenced using the BigDye Terminator kit (Applied Biosystems, Foster City, CA) on an ABI 3100 automated sequencer (Applied Biosystems). Sequencing was performed in both strands of the amplified DNA fragments and sequence accuracy was confirmed by re-sequencing a new PCR product when necessary. Alignments of the DNA sequences for identification of the nucleotide substitutions were done using the SeqMan II program package (DNASTAR Inc., Madison, WI), which was also used for identification of Cryptosporidium at the species level by comparing assembled contigs with the GenBank database. Intestinal parasites were present in 39.8% of the 103 patients evaluated in this study; 6.8% had polyparasitic infections. Prevalence of intestinal parasites were: Entamoeba histolytica/dispar (9.7%), Giardia duodenalis (4.0%), Strongyloides stercoralis (3.9%), microsporidia (3.9%), hookworms (2.9%), Isospora belli (1.9%), Ascaris lumbricoides (1.9%), and Trichuris trichiura (1.9%; Botero et al. 2004). Cryptosporidium sp. was found in one patient with E. histolytica/dispar, with E. bieneusi in the second patient, and with hookworms in the third. Another was co-infected with both E. histolytica/dispar and G. intestinalis. DNA analysis was successful in only six of the isolates studied, which was probably due to long sample storage times. Three isolates were identified as C. hominis: one sequence corresponded with GenBank #AF093489 for the 18S-rDNA, three sequences corresponded with GenBank #XM661099 for the COWP N-terminus, and two sequences for the C-terminus. Two isolates were identified as C. parvum: two sequences corresponded with GenBank #Z22537 for the COWP N-terminal region. One isolate was identified as C. felis: sequence corresponded with GenBank #AF266263 for the COWP N-terminal region. Several studies on the prevalence of Cryptosporidium in Central and South America usually involved children and hospitalized people (Table 2). The prevalence rate observed in our study was 21.4%, which is relatively high compared with other studies carried out in these regions of the world. As in earlier studies, we found chronic diarrhea, anorexia, abdominal pain, and fever as the main symptoms presented by infected individuals. Similar studies of HIV-positive patients gave variable results from 43% prevalence in Zulia, Venezuela (Chacin-Bonilla et al. 1992) to 3.6% in Antioquia, Colombia (Botero et al. 2003). The Antioquia study included immunocompromised patients other than those with AIDS (Table 2) whereas the present study was performed on samples only from AIDS patients. We believe that this is the first report that identified Cryptosporidium in Colombia at the species level and also identified a human C. felis infection in this country. We are indebted to Linda Hamalainen for helpful revision of the manuscript. We also acknowledge the “Comité para el Desarrollo de la Investigación de la Universidad de Antioquia,” Colombia, the “Fundación San Pablo-CEU,” and the “Dirección General de Relaciones Externas y Cooperación de la Generalitat Valenciana” Spain for financing this project. We are grateful to the “Centros de Atención en Salud de pacientes positivos para VIH,”“Fundación Positivos por la Vida,”“Fundación Eudes,” and all the patients that generously collaborated in this project.