Introduction Bovine trichomoniasis was first reported in the US in 19321 and continues to cause reproductive failure in US herds. Because this disease transmission under natural conditions is strictly venereal, it should be amenable to control and even eradication. Although eradication has occurred in some cattle populations, the causative organism, Tritrichomonas foetus, continues to circu- late through cattle herds across the US. The most substantial challenge to successfully eliminate this organism from a cattle population is the ability to consistently and accurately identify infected animals for their removal from the breeding herd. The diagnostic process consists of 3 basic phases: preanalytical (sample collection and handling), analytical (performance of the diagnostic test), and postanalytical (test result reporting and interpretation). Examples of preanalytical errors include inappropriate test request, misidentification of the sample, sample collection and handling, and storage and transportation of the sample. Examples of analytical errors include equipment malfunction, sample mix-ups, and undetected quality control failure. Excessive turnaround period, improper data entry, incorrect interpretation, and inappropriate follow up plan are examples of postanalytical errors. Most errors that occur during the entire diagnostic process occur in the pre- and postanalytical phases.2 The purpose of this discussion is to examine the 3 phases of diagnostic testing as they relate specifically to bovine trichomoniasis, with a focus on pre- and postanalytical phases. Preanalytical considerations Sampling location Distribution of T. foetus in the preputial cavity of bulls was Diagnosis of bovine trichomoniasis Jeff Ondrak 6U Ranch, Steele City, NE Abstract Bovine trichomoniasis was first identified in the US in 1930s as a cause of reproductive failure in cattle and its impact continues. Because bovine trichomoniasis transmission is strictly venereal under natural breeding conditions, its control and eradication are feasible. However, failure to consistently and accurately identify infected animals and culling them have been major challenges. Generally, errors in the 3 diagnostic phases (preanalytical, analytical, and postanalytical) prevent correct classification of Tritrichomonas foetus status of tested animals. The purpose of this review is to define these phases of diagnostic testing and discuss potential errors in each phase.. Keywords: Tritrichomonas foetus, cattle, trichomoniasis described in detail in 1943.3 Highest number of organisms were at the midshaft and caudal portion of the free penis, followed by the prepuce surrounding penis and then the galea glandis, and the prepuce near the preputial opening.3 This information has been used as the basis for recommending the collection of preputial samples from the fornix of the prepuce for trichomo- niasis testing. Although this recommendation is still supported4 further work is necessary to unequivocally confirm that fornix is the optimal site for collecting samples. Sampling devices A variety of sampling devices and sampling techniques have been used to collect preputial samples for T. foetus testing including preputial lavage, cotton swabs, washing of artificial vaginas, specialized collection devices, and uterine infusion pipettes.5 However, without strong evidence supporting 1 method over another, the technique of utilizing a uterine infusion pipette and syringe has become the most widely accepted method for T. foetus sample collection in the US. The Pizzle Stick (Lane Manufacturing, Inc., Denver CO) is a recently marketed product that is based on a T. foetus sampling device developed in the Soviet Union and described in 1969.6 The current device is a long plastic hollow rod with circular grooves around the diameter of the rod at the end that appears to provide an atraumatic sample collection option. Although the original device was superior to pipette aspiration for sample collection6,7 the diagnostic sensitivity between these 2 methods was not different.8 A recently developed T. foetus sampling device (TRICHIT, Morris Clinical Theriogenology 2021; 13: 231 Livestock Products, Delavan, WI) has a uterine infusion pipette with a small plastic collection cup affixed to 1 end. The cup is designed to increase the scraping circumference of the pipette while also allowing the sample to be aspirated into the pipette. The purpose of the collection cup is to increase the volume of sample collected while doing so in a less traumatic manner. Efficacy of TRICHIT over other devices remains unknown. Another more recent technique described as a direct swabbing of the extended penis and prepuce with a gauze sponge has been proposed as a means for collecting a sample from a wider area of the penis and prepuce while allowing the collector the opportunity to avoid areas of gross contamination.9 Because there is limited evidence to support any technique or device over another, the best approach is to determine a procedure that works best for the operator and is approved by the laboratory that tests collected samples. Sample quality Determining what constitutes a quality preputial sample for T. foetus testing is a challenge considering the variety of poten- tial contaminates present in the prepuce including soil, feces, urine, semen, blood, and other organisms. Recent reports10-13 emphasized the need to better understand this preanalytical aspect of trichomoniasis testing. An investigation into vaginitis in a group of cows in Switzerland involved testing for T. foetus.10 Although polymerase chain reaction (PCR) testing of vaginal samples indicated a subs- tantial number of T. foetus positive cows, further investigation revealed these to be false-positive tests due to a cross reaction with Simplicimonas-like organisms. An earlier report also identified false-positive PCR results due to this trichomonad in samples from bulls.11 Bacteria in samples may also provide diagnostic challenges to correctly identify T. foetus infected bulls. Bacterial contamination, if not inhibited in medium, had a negative impact on diagnostic sensitivity.12 Among biological materials (blood, semen, and urine) urine had a negative effect on the correct classification of the T. foetus status of the sample.13 These reports not only highlighted the need to understand the limitations of the tests used to analyze T. foetus samples but also recognized the importance of reducing and eliminating contamination. Sample handling Once an adequate sample has been collected, the conditions under which it is transported to a diagnostic laboratory for testing can have a substantial impact on the results of the test. Despite several studies examining the influence of cold temperatures on both culture and PCR testing, not much is known regarding the effects of higher temperatures on T. foetus survival and its relationship to test accuracy. Although temperatures above 98.6°F affected the ability to culture T. foetus positive samples, they had no negative influence on the ability of PCR to identify the organism, even though the growth rate of the organism was affected.14 Exposure of inoculated pouches (with fewer organisms) to higher temperatures (39, 68, 98.6, and 107.6°F for 24, 48, and 72 hours, respectively) were positive to culture and PCR tests except for 107.6°F.15 Authors attributed this discrepancy to a difference in the number of organisms inoculated and the cutoff values used for the PCR. However, both studies supported the recommendation to protect all T. foetus samples from temperatures above 98.6°F. Analytical considerations Testing samples for PCR was described in 1997 that provided a substantial improvement over traditional culture-based testing protocols.16 Despite advances made in PCR techniques (improved workload management for laboratories, decreased turnaround times for reporting results, and potentially improved analytical sensitivity and specificity versus the traditional culture tech- niques), it has not necessarily improved the overall diagnostic process for bovine trichomoniasis.17-19 The apparent lack of diagnostic improvement for trichomoniasis may have more to do with our expectations of a diagnostic test than the assay itself. An excellent review20 of diagnostic test terminology summarized the following. Analytical sensitivity is an assay’s ability to detect a very low concentration of a given substance or organism in a biological sample, whereas diagnostic sensitivity is the assay’s ability to detect an individual with the condition in the population. For trichomoniasis, the PCR technique has been recognized to have extremely high analytical sensitivity with the theoretical ability to detect as few as 2 T. foetus organisms per ml sample.18 However, the ability of a test to detect a T. foetus-infected animal in a population is our priority and that is an example of diagnostic sensitivity. The diagnostic sensitivity of PCR is ~ 98%.21 Analytical specificity is an assay’s ability to exclusively identify a target substance or organism in a sample, whereas diagnostic specificity is the ability of the assay to correctly identify an individual in a population who does not have the disease in question. The diagnostic specificity of PCR for trichomoniasis testing is ~ 98%.17,18 Postanalytical considerations While a variety of errors may occur in the postanalytical phase, the focus of the discussion for this phase of the diagnostic process will be on interpretation of results. Diagnostic test sensitivity and specificity receive much attention, but they are characteristics of the test itself. Positive and negative predic- tive values are clinically relevant information related to the Clinical Theriogenology 2021; 13: 232 diagnostic process. Positive predictive value is the probability that an individual that tests positive has the disease of interest. Negative predictive value is the probability that an individual that tests negative does not have the disease of interest.20 In other words, predictive values give some indication as to how strongly the results of the test can be trusted to represent the true infection status of an animal. Predictive values are greatly influenced by the prevalence of the disease in a population tested, as described in the following examples using these assumptions: the prevalence of trichomo- niasis in the test population is 1%, the diagnostic sensitivity of the test is 98%, and the diagnostic specificity of the test is 98%. If 10,000 bulls fit the description of a test population, then we would expect to have 100 (10,000 x 0.01) infected bulls in the 10,000 bulls tested. With a diagnostic sensitivity of 98%, 98 (100 x 0.98) of the 100 infected bulls would result in a positive test and 2 (100 minus 98) of the 100 infected bulls would result in a negative test. Because 100 of the 10,000 bulls are infected, then 9,900 (10,000 minus100) of the bulls would be uninfected. With a diagnostic sensitivity of 98%, 9,702 (9,900 x 0.98) of 9,900 uninfected bulls would result in a negative test and 198 (9,900 minus 9702) of the 9,900 uninfected bulls would result in a positive test. The positive predictive value is calculated by dividing the number of infected bulls which tested positive by the total number of test positive bulls. In this example, the positive predictive value is 0.331 (98/296) or the probability of a test positive bull being infected with T. foetus is 33%. The negative predictive value is calculated by dividing the number of uninfected bulls that tested negative by the total number of test negative bulls. For this example, the negative predictive value is rounded to 1.00 (9702/9704), which means that the probability of a test nega- tive bull not being infected with T. foetus is very close to 100%. If the same assumptions are made regarding diagnostic sensitivity and diagnostic specificity, but the prevalence is changed to 20% as might be expected in trichomoniasis infected herds, then the positive predictive value is 0.925 and the negative predictive value is 0.995. In other words, the level of confidence in a positive test is lower in low prevalence situations, but much higher in high prevalence situations. In both scenarios the confidence in a negative test was high. These examples highlight the need for carefully interpreting trichomoniasis test results, regardless of the purported diagnostic sensitivity and diagnostic specificity. Conclusion Eradication of T. foetus from the US cattle population has proven elusive for a variety of reasons with consistent, accurate identification of infected animals as 1 of these contributing factors. Although the process of diagnosing trichomoniasis may be imperfect, paying close attention to details in the steps of diagnostic process reduces preanalytical, analytical, and postanalytical errors and ultimately will enhance the prospects for control and perhaps eradication of this disease in the future. Conflict of interest None to report. References 1. Emmerson MA: Trichomoniasis in cattle. J Am Vet Med Assoc 1932;81:636-640. 2. Plebani M: The detection and prevention of errors in laboratory medicine. Ann Clin Biochem 2010;47:101-110. 3. Hammond DM, Bartlett DE: The distribution of Trichomonas foetus in the preputial cavity of infected bulls. Am J Vet Res 1943;4:143-149. 4. Rhyan JC, Willson KL, Wagner B, et al: Demonstration of Tritrichomonas foetus in the external genitalia and of specific antibodies in preputial secretions of naturally infected bulls. Vet Pathol 1999;36:406-411. 5. Ondrak JD: Tritrichomonas foetus prevention and control in cattle. Vet Clin North Am Food Anim Pract 2016;32:411-423. 6. Sutka P, Katai PL: Rapid demonstration of bull trichomonadosis in unstained smear preparations from preputial scrapings. Acta Vet Hung 1969;19:385-389. 7. Tedesco LF, Errico F, Del Baglivi LP: Diagnosis of Tritrichomonas foetus infection in bulls using two sampling methods and a transport medium. Aust Vet J 1979;55:322-324. 8. Parker S, Campbell J, Ribble C, et al: Comparison of two sampling tools for diagnosis of Tritrichomonas foetus in bulls and clinical inter- pretation of culture results. J Am Vet Med Assoc 1999;215:231-235. 9. Dewell GA, Phillips PE, Dohlman TM, et al: Validation of a gauze sponge sampling methodology to detect Tritrichomonas foetus by real- time PCR. J Vet Diagn Invest 2016;28:595-598. 10. Frey CF, Müller N, Stäuber N, et al: Simplicimonas-like DNA in vaginal swabs of cows and heifers cross-reacting in the real-time PCR for T. foetus. Vet Parasitol 2017;237:30-36. 11. Schommer S, Younger S, Fales WH: Cross-reaction of Simplicimonas spp. trichomonads in Tritrichomonas foetus assays. Proc Annu Conf Am Assoc Vet Lab Diagn 2011; p. 34. 12. Clothier KA, Villanueva M, Torain A, et al: Effects of bacterial conta- mination of media on the diagnosis of Tritrichomonas foetus by culture and real-time PCR. Vet Parasitol 2015;208:143-149. 13. Clothier KA, McNabb B, Torain A, et al: Effects of biological materials and collection media on PCR detection of Tritrichomonas foetus. Open J Anim Sci 2019;9:121-128. 14. Davidson JM, Ondrak JD, Anderson AA, et al: Evaluation of effects of high incubation temperature on results of protozoal culture and real- time PCR testing for Tritrichomonas foetus inoculated in a commercially available self-contained culture media system. J Am Vet Med Assoc 2011;239:1589-1593. 15. Clavijo A, Erol E, Sneed L, et al: The influence of temperature and simulated transport conditions of diagnostic samples on real-time polymerase chain reaction for the detection of Tritrichomonas foetus DNA. J Vet Diagn Invest 2011;23:982-985. 16. Felleisen RSJ: Comparative sequence analysis of 5.8S rRNA genes and internal transcribed spacer (ITS) regions of trichomonadid protozoa. Parasitology 1997;115:111-119. Clinical Theriogenology 2021; 13: 233 17. Cobo ER, Favetto PH, Lane VM, et al: Sensitivity and specificity of culture and PCR of smegma samples of bulls experimentally infected with Tritrichomonas foetus. Theriogenology 2007;68:853-860. 18. Mukjufhi N, Irons PC, Michel A, et al: Evaluation of a PCR test for the diagnosis of Tritrichomonas foetus infection in bulls: effects of sample collection method, storage and transport medium on the test, Theriogenology 2003;60:1269-1278. 19. Ondrak, JD, Keen JE, Rupp GP, et al: Repeated sampling and testing by culture and PCR to detect Tritrichomonas foetus carrier bulls in an infected Nebraska herd. J Am Vet Med Assoc 2010;237:1068-1073. 20. Saah AJ, Hoover DR: “Sensitivity” and “specificity” reconsidered: the meaning of these terms in analytical and diagnostic settings. Ann Intern Med 1997;126:91-94. 21. Mutto AA, Giambiaggi S, Angel SO: PCR detection of Tritrichomonas foetus in preputial bull fluid without prior DNA isolation. Vet Parasitol 2006;136:357-361. Clinical Theriogenology 2021; 13: 234 Diagnosis of bovine trichomoniasis