Tel.+49 30 838 62618 Fax.+49 30 838 62620
Puerperal metritis, clinical endometritis and subclinical endometritis are highly present in dairy cows. To prevent or treat those diseases in an appropriate way the knowledge of risk factors and coherences for their appearance is important.
Various metabolic parameters were identified to be associated with puerperal diseases and bacteria could be isolated from the uterus of cows with clinical signs of illness. The uterine lumen of postpartum dairy cows is commonly contaminated with different bacterial species after parturition. Cows who fail to eliminate the pathogenic species often develop puerperal metritis or later in the puerperium clinical endometritis. Bacterial uterine infections have been shown to negatively affect ovarian function and fertility. The objective of my first study was to investigate relationships between Escherichia coli , Trueperella pyogenes, alpha-hemolytic Streptococci, and coagulase negative Staphylococci and their presence and effect on each other and uterine infections 2 weeks later. Therefore bacteriological samples were collected at 10 ± 1 and 24 ± 1 DIM from the uterine lumen using a cytobrush.
Vaginal examination was performed by vaginoscopy. The vaginal discharge was assigned according to its distribution of pus to identify cows with puerperal metritis (10 ± 1 DIM) or clinical endometritis (24 ± 1 DIM). Bacteria were identified on the basis of the characteristics of the colony and standard tests (e.g. Gram stain, morphology, hemolysis, biochemical profile) and bacterial growth quantified on a 4-point scale. Cows infected with E. coli or T. pyogenes at 10 ± 1 DIM had a higher risk for an infection with the same bacterial species at 24 ± 1 DIM [E. coli relative risk (RR) = 3.7; T. pyogenes RR = 2.9]. Cows infected with T. pyogenes also had a greater risk to be diagnosed with abnormal vaginal discharge at first (RR = 2.4, P < 0.001) and second (RR = 2.4, P < 0.001) examination. In terms of risk factor for clinical endometritis, the risk of being diagnosed with abnormal vaginal discharge at 24 ± 1 DIM increased in cows positive for E. coli (RR = 1.7) or T. pyogenes (RR = 1.7) at 10 ± 1 DIM. These cows also had more days to pregnancy (T. pyogenes 139 d, P < 0.05) and more cows were culled (E. coli 15.8%, P < 0.05) or had significant more days to first artificial insemination (E. coli 70 d, P < 0.001). These results demonstrate that cows infected with or T. pyogenes are prone to persistent infections 2 weeks later and also had higher risks to develop clinical endometritis. Infections with AHS only increased the risk for abnormal vaginal discharge which warrants further research of their role in the uterus.
CNS did not affect the infection with the same or other bacterial species 2 weeks later, as well as AHS, or influenced the vaginal discharge. Recent studies describing uterine bacterial contents have used guarded cotton swabs or biopsy method. The cytobrush technique I used to collect bacteriological samples was originally evaluated in human medicine to collect cell material from the vagina or the uterus to diagnose inflammations by means of polymorph nuclear neutrophil leucocytes.
The technique has been validated for cows to diagnose the SCE. Several recent studies have utilized the CB to harvest endometrial cells and bacteria from uterine lumen simultaneously, however, it never was validated for it.
Therefore my second objectives of the first study were to evaluate the cytobrush for the collection of bacteria from the uterus and to test the agreement of bacteriological findings between 3 different laboratories. Uterine discharge was aspirated with a disposable catheter from 68 cows at 10 ± 1 DIM and transferred into a sterile plastic tube. From these tubes 1 CS and 1 CB each were sent to 3 different laboratories and the positive or negative findings of E. coli, T. pyogenes, AHS and CNS were compared.
One laboratory received a second CS to test the intraassay repeatability. The agreement of bacteriological results of the laboratories were significant for laboratory A + B and A + C for each sampling method, CB and CS (A + B: CB = 86%, CS = 92%; A + C: CB = 82%, CS = 86%). The highest agreement considering the bacterial species was found for E. coli (A + B: CB = 89.7%, CS = 98.5%; A + C: CB = 82.4%, CS = 88.2%). The interlaboratory agreements between CB and CS were significant for each of the 3 laboratories (A = 92%; B = 97%; C = 93%) as was intralaboratory repeatability for laboratory A. Due to a missing gold standard it is impossible to prove which results are more accurate. But these results consider the cytobrush a valuable technique to generate information about infectious and inflammatory properties in the endometrium with a single sample. The subclinical endometritis has a high prevalence in dairy cows but the approaches to diagnose the disease showed a wide variety of cut points for the amount of percent PMNL and sampling time. SCE was proved to have a severe negative impact on fertility.
Recent studies identified a low body condition score, hyperketonemia or increased haptoglobin concentrations as risk factors for their development. My objective of the second study was first, to determine the relationship between aerobic bacteria in the uterus of healthy animals and PMNL in the early postpartum period.
Second I compared 2 classification schemes for the diagnosis of SCE. Cytological and bacteriological samples were taken from 149 cows at 10 ± 1 and 24 ± 1 DIM. Cows were included into the study it they were diagnosed with no or clear vaginal mucus at the second examination. Additionally, 131 cows at 21 to 27 DIM without signs of clinical endometritis were sampled. The cytobrush technique was used for sample collection and vaginal examination was performed by vaginoscopy. The cut points for PMNL in the uterus were set at 5, 10 and 18%. Cows infected with AHS at 10 ± 1 DIM had significant (P = 0.006) higher percent PMNL, more days to pregnancy (Hazard Ratio = 1.7, P = 0.038) and decreased odds of pregnancy by 250 DIM (Odds Ratio = 0.4, P = 0.022) compared to cows not infected with AHS. Anaerobic bacteria were not isolated in this study so their effect cannot be excluded. However, in another study Fusobacterium necrophorum and Prevotella melaninogenica, 2 well known uterine pathogens, did not show any association with percent PMNL at concurrent infection in which enhances the results for AHS as risk factor for SCE. Also T. pyogenes increased the odds about 5.1 to diagnose the same cow with more than 18% PMNL 2 weeks later but without abnormal vaginal discharge. The impairment of the endometrium induced by T. pyogenes and stimulus for the immune system could explain the long period cows need to cure the infection and the persistence of PMNL in the uterus. AHS, T. pyogenes , E. coli or CNS isolated at 24 ± 1 DIM did not affect the distribution of percent PMNL at the same examination day.
For the comparison of different cut points the reproductive performance outcomes were compared when the same cows (24 ± 1 and 21 to 27 DIM) were arranged in 2 different schemes: PMNL1 (< 5, 5 – 18, > 18% PMNL) and PMNL2 (< 10, 10 – 18, > 18% PMNL). Reproductive performance outcomes were numerically better in the intermediate groups (5 – 18 and 10 – 18%) compared to cows with low (< 5/ < 10%) or high (> 18%) PMNL. Considering parity reproductive performance parameters differed significantly. In primiparous cows days to pregnancy were lower in cows of each intermediate group (5 -18: 68 d, P = 0.017; 10 – 18: 70 d, P = 0.039) compared to cows with > 18% PMNL (144.5 d). In multiparous cows the reproductive performance parameters did not differ significantly neither in PMNL1 nor in PMNL2. These results demonstrate that the cut point of 18% PMNL had a high diagnostic value, at least for primiparous cows, at 21 to 27 DIM to identify cows with SCE.
In conclusion, the results of the two trials showed that previous infections with E. coli or T. pyogenes are essential risk factors for CE, whereas both negatively affected reproductive performance parameters and E. coli seemed to pave the way for T. pyogenes. The SCE was shown to be affected by previous infections with T. pyogenes or AHS. Additionally, cows infected with AHS had longer days to pregnancy intervals and showed a decrease in the number of cows pregnant. Hence, more research is needed to prove the pathogenic potential of AHS including anaerobic bacteria.
The bacteriological results of different laboratories showed high agreements between CB and CS. The CB validated for the collection of cells from the uterus is also useful for bacterial analysis.