Use of hysteroscopy for the diagnosis of postpartum clinical endometritis in dairy cows

Abstract

CLINICAL endometritis is one of the most common disorders in dairy cows, causing decreased fertility and large economic losses (LeBlanc 8). The condition can be defined as an inflammation of the endometrium with uterine exudate, varying from clear mucus with flecks of pus to purulent discharge in the vagina, 21 days or more postpartum (Sheldon and others 12). The common methods of diagnosis of clinical endometritis, rectal palpation and vaginal examination (LeBlanc and others 9, Sheldon and others 13), provide little information about inflammation of the endometrium itself. Thus, false-positive findings (for example, in cases of vaginitis) or false-negative findings (where endometritis is not detected) may sometimes occur. Subclinical endometritis is defined as the presence of polymorphonuclear neutrophils (PMNs) in the endometrium of cows without uterine discharge (Sheldon and others 12). For the diagnosis of subclinical endometritis, endometrial cells can be obtained with the cytobrush or low-volume flush techniques (Kasimanickam and others 6, Gilbert and others 4, Galvão and others 3). Endoscopy is commonly used as a diagnostic technique in human, equine and small animal medicine. In bovine medicine, several applications for endoscopy have been described (Franz 2). Endoscopy provides an opportunity to take visually guided biopsies as well as carrying out procedures such as the treatment of a displaced abomasum or internal teat surgery (Franz 2). Endoscopic examination of the uterus (hysteroscopy) in cows has been described only by a few authors (Devine and Lindsay 1, Metzner and others 11). Hysteroscopy can be a useful tool for research, for example, for the interpretation of physiological and pathological changes during the postpartum period, and for the validation of other diagnostic tools, such as vaginoscopy or the cytobrush technique. The aim of this study was to demonstrate that hysteroscopy can be used to diagnose clinical and subclinical endometritis in dairy cows. In a field trial conducted on a commercial dairy farm in Brandenburg, Germany, 30 cows between 20 and 35 days postpartum were examined by vaginoscopy, rectal palpation and hysteroscopy. During the examination, the cows were kept in a squeeze chute with a headlock. No sedation or local anaesthesia was used. The diagnosis of endometritis by vaginoscopy was made according to Sheldon and others (2006). Rectal palpation was performed to evaluate the fluid content and symmetry of the uterine horns. Hysteroscopy was performed with a rigid endoscope, 750 mm in length and 6 mm external diameter, specifically manufactured for this study by World of Medicine. It was equipped with two channels, one through which air could be passed to inflate the uterine lumen and the other one through which cytological samples could be taken. The endoscope was introduced into the vagina and controlled by rectal palpation. To keep the endoscope free of cervical mucus, it was covered with a disposable plastic sleeve (Minitüb). After the endoscope had been passed through the cervix, the plastic sleeve was pulled back, and air was pumped in until the lumen was wide enough to enable exploration of the uterus. The presence or absence of fluid in the uterus, and scars and red spots on the endometrium were documented. The presence of mucopurulent fluid or pus in the uterus was regarded as a sign of endometritis. In 20 cows that had no vaginal discharge and no pus visible at hysteroscopic examination, cytological samples were taken from the uterus with a small brush (Karl Storz Endoskop Austria) through the working channel of the endoscope. After collecting the sample, the brush was rolled on to a microscope slide. The slide was fixed on-farm and then stained (LT-SYS; Labor und Technik) in the laboratory (Kaufmann and others 7). A proportion of greater than 5 per cent PMNs was regarded as indicative of subclinical endometritis (Gilbert and others 4). The sensitivity, specificity, positive and negative predictive values, and accuracy of rectal palpation and vaginoscopy in the diagnosis of clinical endometritis were calculated, using hysteroscopy as the reference method (Martin and others 10, Iwersen and others 5). Hysteroscopy was chosen as the reference method because in the present study it was the only direct method used to evaluate the endometrium in cows with clinical endometritis. The prevalence of clinical endometritis was 27 per cent by vaginoscopy, 23 per cent by rectal palpation and 13 per cent by hysteroscopy. Hysteroscopic examinations revealed no significant differences in the presence of fluid in the uterine horns of cows diagnosed positive or negative for clinical endometritis by vaginoscopy (75 v 43 per cent, P<0.12). However, pus was found by hysteroscopy only in cows that were positive for clinical endometritis, and not negative, by vaginoscopy (50 v 0 per cent, P<0.05). Red spots in the endometrium were found by hysteroscopy in 33 per cent of the cows that were diagnosed with clinical endometritis on the basis of vaginoscopy, and in 6 per cent of the cows that were negative by vaginoscopy (P=0.08). The test characteristics for rectal palpation and vaginoscopy compared with hysteroscopy are shown in Table 1. Among the 20 cows that were negative for clinical endometritis by vaginoscopy and had no pus visible by hysteroscopy, the prevalence of subclinical endometritis diagnosed by cytology was 40 per cent. Further abnormalities of the endometrium that may be found in cows with clinical endometritis, that is, a reddish colour or the presence of red spots or scars, were not found in the cows diagnosed with subclinical endometritis, or in the 12 cows that were negative for subclinical endometritis. For the diagnosis of subclinical endometritis, the use of hysteroscopy seems to be limited as there was no difference in the hysteroscopic observations in cows with or without subclinical endometritis. However, the small number of observations made in this study does not allow a firm conclusion to be drawn on this possible application of hysteroscopy. For validation of the cytobrush technique, it could be of interest to study whether cytological samples obtained from specific regions of the endometrium, for example, the caruncular areas, provide similar results to samples obtained from smooth endometrium. For these evaluations, hysteroscopy could be a valuable tool. The authors gratefully acknowledge the support of the study by World of Medicine, Berlin, Germany, for providing all the endoscopic equipment, and the cooperation of the dairy farm in Brandenburg, Germany. LVM's visit to the Freie Universität Berlin was supported by a fellowship from the Deutscher Akademischer Austauschdienst.