Abstract
Objective
Proton pump inhibitors (PPIs) can kill some human protozoan parasites in cell culture better than the drug metronidazole. Clinical data showing an antiprotozoal effect for PPIs are lacking. The objective of the study is to determine if PPI use is associated with a reduced risk of having intestinal parasites.
Methods
We obtained electronic medical record data for all persons who received a stool ova and parasite (O & P) examination at our tertiary care academic medical center in Cleveland, Ohio, between January 2000 and September 2014. We obtained the person’s age, whether they were taking a PPI at the time of the O & P examination, and whether the pathology report indicated the presence of any parasites. χ2 with Yates correction was used to determine if PPI use was associated with stool protozoa.
Results
Three intestinal protozoa were identified in 1199 patients taking a PPI (0.3%), and 551 intestinal parasites were identified in the 14,287 patients not taking a PPI (3.9%). There was a statistically significant lower likelihood of finding protozoa in the stool of a person taking a PPI compared with those not taking a PPI (P < .0001).
Conclusions
Patients taking a PPI were statistically less likely to have an intestinal protozoa reported on stool O & P examination compared with those not taking a PPI.
Introduction
Intestinal parasites can be common in persons visiting developing countries with unsanitary conditions and weak public health infrastructure. 1 International travelers who develop prolonged infectious diarrhea are commonly infected with the following protozoa: Giardia lamblia, Cryptosporidium parvum, Entamoeba histolytica, and Cyclospora cayetanensis. 2 Backcountry enthusiasts also are at increased risk for intestinal protozoa, especially G lamblia. 3
Proton pump inhibitors (PPIs) are commonly prescribed medications in the United States. US Food and Drug Administration–approved PPIs include omeprazole (Prilosec) [Procter & Gamble Co.], lansoprazole (Prevacid) [Takeda Pharmaceutical Company Ltd.], rabeprazole (AcipHex) [Eisai Co., Ltd.], pantoprazole (Protonix) [Pfizer Inc.], esomeprazole (Nexium) [AstraZeneca plc.], and dexlansoprazole (Kapidex) [Takeda Pharmaceutical Company Ltd.]. PPIs are structurally similar weak bases that accumulate as prodrugs in acidic environments and undergo acid-catalyzed conversion to the active drug. 4 PPIs are used to treat a variety of medical conditions, including gastroesophageal reflux disease, gastric and duodenal ulcers, non–steroidal-induced gastrointestinal lesions, Zollinger-Ellison syndrome, dyspepsia, and Helicobacter pylori infection. 4 Short-term use of PPIs is well tolerated and has few significant side effects; however, long-term use may be associated with dementia, micronutrient deficiencies, and bone fractures.4,5
PPIs are benzimidazole derivatives that are structurally similar to benzimidazole 2-methylcarbamates, albendazole, and mebendazole, which are able to kill some human protozoans such as G lamblia, E histolytica, and Trichomonas vaginalis, in cell culture. 6 –9 Albendazole and mebendazole exert an antimicrobial effect by adversely affecting β-tubulin polymerization, resulting in changes to the parasite cytoskeleton, microtubules, and microribbons. 9 Quantitative structure activity relationship analyses studies based on comparative molecular field analysis, molecular similarity indices in comparative analysis studies, and structure-activity relationship studies based on structure-activity-similarity maps found that PPIs are expected to have antiparasitic activity based on past experience with benzimidazole 2-methylcarbamates, albendazole, and mebendazole. 6 In vitro cell culture studies show that PPIs kill T vaginalis 2 to 3 times, G lamblia 13 to 78 times, and E histolytica 15 to 135 times better than metronidazole, a drug commonly used to treat these infections. 6 A retrospective chart review of female patients who presented to the emergency department found that persons taking a PPI were significantly less likely to be infected with T vaginalis when controlling for co-infection with gonorrhea and chlamydia. 10
PPIs have antiprotozoal activity in the laboratory, but no clinical investigations have examined their effect on intestinal protozoa. The objective of this pilot study was to examine whether PPI use was associated with presence of protozoa in the stool. Our hypothesis was that PPI use would be associated with the presence of fewer intestinal parasites than in those persons not taking a PPI.
Methods
The University Hospital Cleveland Medical Center institutional review board approved this study. The hospital information technology group obtained data from the electronic medical records for persons treated at our tertiary care academic medical center in Cleveland, Ohio, between January 2000 and September 2014. Information technology provided the number of patients aged 18–40, 41–65, and 66–100 years who received a stool ova and parasite (O & P) examination, whether the person was taking a PPI at the time of the stool O & P, and whether the pathology report indicated the presence of any parasites. We were not provided any clinical information about the patients or details on how the stool O & P pathology report was generated. χ2 with Yates correction was used to determine whether PPI use was associated with the presence of intestinal protozoa.
Results
A stool O & P examination was done on 1199 out of 17,422 patients (6.9%) taking a PPI and 14,287 out of 110,327 patients (12.9%) not taking a PPI. In our dataset of those who underwent a stool O & P, 92% were not taking a PPI.
There were 557 parasites reported from 15,486 encounters involving a stool O & P. We received aggregate data; therefore, some patients may have had multiple parasites on a single examination. There were 551 intestinal protozoa identified: Blastocystis hominis (n=322), G lamblia (n=69), Dientamoeba fragilis (n=56), Entamoeba coli (n=50), Cryptosporidium (n=21), Entamoeba hartmanni (n=17), E histolytica (n=8), and Entamoeba dispar (n=8). There were 6 helminthes reported, including 3 Enterobius vermicularis and 1 Trichuris trichiura found in persons aged 18–40 years not taking a PPI and 2 E vermicularis found in persons aged 41–65 years not taking a PPI. There were no cestodes or trematodes reported in individuals taking a PPI or not taking a PPI. The Table summarizes the number and types of parasites found on O & P.
Protozoa observed on stool O & P in individuals taking a PPI and not taking a PPI
O & P, ova and parasite; PPI, proton pump inhibitor.
There were 3 intestinal protozoa identified in 1199 patients taking a PPI (0.3%); 151 persons aged 18–40 years had 0 protozoa (0%), 494 persons aged 41–65 years had 1 protozoa (0.2%), and 554 persons aged 66–100 years had 2 protozoa (0.4%).
There were 551 intestinal parasites identified in the 14,287 patients not taking a PPI (3.9%); 4600 persons aged 18–40 years had 251 protozoa (5.5%), 5881 persons aged 41–65 years had 213 protozoa (3.6%), and 3806 persons aged 66–100 years had 87 protozoa (2.3%). There was a statistically significantly lower likelihood of finding protozoa in the stool of a person taking a PPI compared with those not taking a PPI (P < .0001).
χ2 was used to show a statistically significant difference in the rates of intestinal protozoa based on the patient’s age (P < .001). In the 18–40-year age group, 251 protozoa were found in 5002 persons (5%). In those aged 41–65 years, 214 protozoa were found in 6589 persons (3.2%). In those aged 66–100 years, 89 protozoa were found in 4449 persons (2%).
Discussion
Laboratory studies have shown that PPIs have antiprotozoal activity. 6 –9 The pilot data presented here revealed that patients taking a PPI are statistically less likely to have intestinal protozoa reported on stool O & P examination compared with those not taking a PPI, an observation that has not been reported previously. Most intestinal protozoal are nonpathogenic commensals or mildly pathogenic, but G lamblia and E histolytica are 2 pathogenic intestinal protozoa that were identified 69 and 8 times, respectively, in those persons not taking a PPI, but neither was seen in persons taking a PPI. Overall, intestinal protozoa are rare and were identified in less than 4% of persons undergoing stool O & P examination at the study site; however, younger persons were more likely to have intestinal protozoa than older persons.
If the present findings are confirmed, PPI having clinically relevant activity against intestinal protozoa could have important implications for the management of and prophylaxis against these parasitic infections. Potentially, a short course of a PPI would be safer and have fewer side effects than a course of metronidazole for giardiasis or intestinal amebiasis. The findings reported will need to be confirmed in a larger, multicenter retrospective review that could include international locations with a higher prevalence of stool protozoa in the general population. Future studies should also control for recent antimicrobial use and the different methods of obtaining the stool O & P data.
Limitations
The data are from a retrospective review of a single academic medical center from a population with a low baseline prevalence of intestinal protozoa. The specific diagnostic tests used to identify the stool protozoa and recent antimicrobial use by the study patients were not controlled for in this study. Lastly, some patients may have had multiple intestinal protozoa identified on the same O & P examination or may have undergone multiple O & P examinations.
Conclusions
PPI use is associated with a reduced risk of having intestinal protozoa. Additional studies are needed to further delineate any clinically relevant antiprotozoal activity possessed by PPIs.
Financial/Material Support: None.
Disclosures: None.