Clinical Significance of Respiratory Involvement in Cryptosporidiosis: Cross-Sectional Study of Children with Diarrhea and Respiratory Symptoms in Uganda.

Respiratory cryptosporidiosis is considered an occasional, late-stage complication of HIV/AIDS. This study aimed to assess the clinical importance of respiratory cryptosporidiosis in children with diarrhea and respiratory symptoms at Mulago Hospital, Kampala, Uganda. Children aged 9 to 36 months presenting with diarrhea and cough or unexplained tachypnea (N = 1,918) were screened for fecal Cryptosporidium using polymerase chain reaction (PCR). Children with positive stool samples were eligible for further diagnostic tests, including sputum induction. Sputum samples were subjected to PCR for Cryptosporidium, as well as routine microbiology (culture and gram stain) and auramine stain for tuberculosis. Regression analyses were used to investigate 1) factors associated with respiratory cryptosporidiosis and 2) whether respiratory cryptosporidiosis was independently associated with hospitalization. Prevalence of enteric cryptosporidiosis was 260/1,918 (13.6%) (>80% Cryptosporidium hominis). Of the 236 children who had sputum available for analysis, 62 (26.3%) had Cryptosporidium in the sputum, only two of whom had HIV infection. Children with Cryptosporidium in the sputum were more likely to have abnormal oxygen saturation at presentation (SpO2 <96%; P = 0.053); no other differences in frequency or severity of respiratory signs were noted. No alternative bacterial cause of respiratory symptoms was identified in 37.7% of children with respiratory cryptosporidiosis, compared with 23.6% of children without (P = 0.04). Sputum-positive children had twice the odds of hospitalization compared with children without Cryptosporidium infection at this site (adjusted odds ratio = 2.08, 95% confidence interval: 1.02-4.22; P = 0.043). Respiratory tract involvement is common in children with intestinal cryptosporidiosis who are experiencing respiratory symptoms. Such children may experience some degree of respiratory compromise and may be at increased risk for hospitalization.

Giardia duodenalis in Ugandan Children Aged 9-36 Months in Kampala, Uganda: Prevalence and Associated Factors.

Giardia duodenalis is a common gastrointestinal pathogen globally that has been associated with growth failure in children. Most of the studies have been done in school-age children, and there is a paucity of data in pre-school children. We determined the prevalence and factors associated with G. duodenalis infection in children aged 9-36 months presenting to Mulago Hospital with diarrhea or cough. Demographic and socio-economic characteristics, animal ownership, medical history, and physical examination findings were recorded. Stool was tested for G. duodenalis using real-time quantitative polymerase chain reaction (qPCR), and additional tests included stool microscopy and qPCR for Cryptosporidium. The overall prevalence of G. duodenalis infection was 6.7% (214/3,173). In children with diarrhea the prevalence was 6.9% (133/1,923), whereas it was 6.5% (81/1,250) in those with cough as the main symptom. Of 214 children with G. duodenalis infection, 19 (8.9%) were co-infected with Cryptosporidium. Older children (25-36 months) were more likely to have G. duodenalis infection (adjusted odds ratio [aOR]: 2.93, 95% CI: 1.93-4.43). Use of an unimproved toilet (aOR: 1.38, 95% CI: 1.04-1.83) and the wet season (aOR: 1.33, 95% CI: 1.00-1.77) were associated with increased infection. Other factors associated with infection were recurrent diarrhea (aOR: 2.46, 95% CI: 1.64-3.70) and passing of mucoid stool (aOR: 2.25, 95% CI: 1.08-4.66). Having a ruminant at the homestead was also associated with infection (aOR: 1.83, 95% CI: 1.20-2.79). Giardia duodenalis infection occurred in 1 of 15 children aged 9-36 months with diarrhea or cough in Kampala, Uganda. Further studies are needed to clarify the zoonotic significance of G. duodenalis infection in this setting.

Seasonality of Cryptosporidium oocyst detection in surface waters of Meru, Kenya as determined by two isolation methods followed by PCR.

Meru, Kenya has watersheds which are shared by wildlife, humans and domesticated animals. These surface waters can be contaminated by the waterborne pathogen Cryptosporidium. To quantify the seasonality and prevalence of Cryptosporidium in Meru regional surface waters, we used a calcium carbonate flocculation (CCF) and sucrose floatation method, and a filtration and immunomagnetic bead separation method, each of which used PCR for Cryptosporidium detection and genotyping. Monthly water samples were collected from January through June in 2003 and 2004, bracketing two April-May rainy seasons. We detected significant seasonality with 8 of 9 positive samples from May and June (p<0.0014), which followed peak rainy season precipitation and includes some of the subsequent dry season. Six of 9 positive samples revealed C. parvum, and 3 contained C. andersoni. None contained C. hominis. Our results indicate that Meru surface waters are Cryptosporidium-contaminated at the end of rainy seasons, consistent with the timing of human infections reported by others from East Africa and contrasting with the onset of rainy season peak incidence reported from West Africa.

Expectoration of Cryptosporidium Parasites in Sputum of Human Immunodeficiency Virus-Positive and -Negative Adults.

Respiratory cryptosporidiosis is thought to be a rare, end-stage complication of HIV. Few studies have systematically examined the frequency of such infection in adults. Sputum specimens submitted for tuberculosis (TB) testing at Mulago Hospital, Uganda, were anonymously retested for Cryptosporidium using real-time polymerase chain reaction (PCR). Visual confirmation using immunofluorescence confocal microscopy was performed for a subset of PCR-positive samples. Of 824 sputum samples tested, 24 (2.9%) were Cryptosporidium positive. Prevalence in sputum ranged between 0% and 10% in each month of the study and exceeded TB prevalence in some months. In this referral population, respiratory Cryptosporidium prevalence was lower in people with HIV (1.3% versus 4.4% without HIV, P = 0.028) and higher in those with TB (6.8% versus 2.6% without TB, P = 0.086). The weak association between respiratory Cryptosporidium infection and TB persisted after controlling for HIV (odds ratio = 3.2, 95% confidence interval: 0.9, 11.8; P = 0.080). This is the first study to document adult respiratory tract cryptosporidiosis in a referral population with presumed TB. These findings 1) confirm that Cryptosporidium respiratory infection occurs in HIV-negative and -positive adults; 2) suggest there is potential for Cryptosporidium to be disseminated or transmitted by coughing or expectoration; and 3) identify possible synergy between Cryptosporidium and TB in the respiratory tract.