Gastrointestinal nematodes (GIN) cause significant disease and economic losses in grazing animals (1). Haemonchus contortus is one of the most damaging GIN species affecting the health and welfare of small ruminant livestock worldwide (2, 3). Warm, moist conditions favour the survival of H. contortus larvae; however, larvae can also survive relatively cold and dry conditions by suspending development in their host during winter (termed hypobiosis) and continuing their development and transmission in the following spring (8). In Australia, H. contortus is most prevalent in south-eastern Queensland, and northern New South Wales (9), although the distribution of H. contortus has been expanding to more southern parts of the country (10). In favourable conditions, female H. contortus are capable of laying up to 10,000 eggs daily, resulting in rapid and extensive pasture contamination, thus increasing the flock exposure to the parasite (4, 5). Animals infected with H. contortus can suffer blood loss of up to 30 mL daily in severe cases (6, 7). Blood loss in sheep typically occurs from blood ingested by the nematode and blood leakage into the sheep gut due to damage sustained through nematode attachment (7).
Slaughter of sheep to quantify nematodes is the gold standard method for the precise estimation of parasite burden in sheep. This method is also often required to demonstrate anthelmintic efficacy for the registration of new control products, but the loss of life makes it an impractical method for use on farms (11). Faecal worm egg count (FWEC) is the method currently recommended for GIN monitoring by the World Association for the Advancement of Veterinary Parasitology and is used by many sheep producers (12, 13). The most commonly used FWEC technique, the McMaster method, is known to have relatively low analytical sensitivity of 10–50 eggs per gram of faeces (epg) and high variability between subsamples (14–17). It has been reported that the presence of blood in faeces from infected sheep can be detected as early as 11 days after initial infection, whereas H. contortus eggs begin to occur in faeces at approximately 18 days post-infection (2, 6). Therefore, for new infections, the detection of blood or haemoglobin (Hb) in the faeces may enable earlier diagnosis of H. contortus infections in sheep than with FWEC.
Several tools based on blood loss have been developed for the diagnosis of H. contortus infections. The Haemonchus Dipstick Test, which is no longer commercially available, quantified the amount of faecal occult blood (FOB) present in sheep faeces (18, 19). The FOB reagent strips, Hemastix® (Bayer HealthCare LLC), were used to detect the presence of Hb peroxidase activity in faeces as an index of the level of worm infection present. Given that peroxidase is also found in plant materials, a strict boiling time of 20 minutes was necessary to ensure accuracy of the test (18, 20). The dipstick test used a score range of 1–5, with the manufacturer’s manual indicating that a score of 1 is a negative result, and score of 5 represents 200 blood cells/µL or higher, suggesting the presence of a heavy H. contortus infection requiring immediate anthelmintic treatment. Detailed interpretation of the test result was challenging, however, as each score corresponded to a wide range of Hb concentrations (21). Additionally, FOB test kits such as Hemastix® have been found to be less sensitive in the presence of mixed-parasite infected samples (21).
The FAMACHA© method is a five-point scoring system that can be used at sheep yards by a trained observer to assess the level of anaemia of the sheep (22). Relatively good sensitivity of the FAMACHA© method has been reported in various regions for detecting anaemic animals (> 50% for sheep and > 89% for goats) (23–27). Despite the relative ease of use and low cost of FAMACHA©, a drawback of this system is that it relies heavily on the operator's experience, which can lead to inconsistent results both within and across sheep management systems (23). Furthermore, production losses can occur before scores indicative of the need for anthelmintic treatment are obtained, resulting in economic losses for sheep producers (27–29).
Previous studies evaluating the accuracy of FOB test kits were performed based on the measurement of sheep blood collected through jugular venepuncture, rather than direct measurement of FOB found in sheep faeces (11, 21). In these studies, results from packed cell volume and Hb concentrations obtained from whole blood samples for sheep infected with H. contortus were compared with the predictions from FOB test kits. It can be challenging to evaluate the accuracy of Hb assessed using FOB test kits, as there may not be a direct relationship between the changes in haematocrit levels in the host blood and the presence of blood in the faeces (21). Furthermore, there is no qualitative method available for the direct measurement of Hb concentrations in sheep faeces.
Faecal analyses using near-infrared (NIR) spectroscopy have been previously used to evaluate diet quality in ruminants and to monitor their health and welfare (30–34). Faecal NIR spectroscopy has also been applied to estimate tick burdens in cattle and horses through the prediction of stress (35) and internal parasite burdens through the prediction of FWEC (34). The use of NIR spectroscopy for chemical analysis in animals has continued to increase due to its wide range of applications, low processing time, low-cost, non-invasiveness, bulk-sampling capacity and the ability to measure samples under different conditions (36–38). Importantly, it is possible to use a single NIR spectral scan of a sample to determine several chemical attributes simultaneously. However, spectroscopists must structure and design calibration models carefully to focus on specific chemical attributes related to their particular topic of interest.
The ultimate aim of a test to determine the need for anthelmintic treatment is to estimate the total worm count of H. contortus present in the abomasum of sheep. This is most commonly done using FWEC, but as noted, FWEC may be underestimating the worm burden, given that nematodes may be present for up to seven days without producing eggs (39, 40). Furthermore, methods involving the assessment of anaemia may be affected by the experience of operators (23, 24). Given that there are currently no tests available to directly measure sheep blood content in faeces, it is particularly challenging to determine reference values for the validation of a prediction model, in our case, the validation of a calibration model based on blood content of the faeces. For these reasons, the calibration models used here to predict Hb in faeces were based on faecal samples spiked with known concentrations of sheep blood.
We previously reported the potential of vis-NIR spectroscopy in detecting blood in sheep faeces, as an indicator of H. contortus infection (41). We developed calibration models within the wavelength region of 400–600 nm to measure blood in faeces collected from a single sheep. In this study, the prediction precision was lowest with the lower Hb concentrations, which was likely due to lack of physical, chemical and environmental variation in the faecal samples in the calibration model. Here we investigated the predictability of calibration models built using faeces collected from various locations free from H. contortus and spiked with known amounts of blood, for the prediction of Hb concentrations in both blood-spiked and naturally infected sheep faeces.