Tel.+49 30 838 62618 Fax.+49 30 838 62620
Acute puerperal metritis is characterized by reddish-brown, fetid, watery vaginal discharge in combination with an elevated rectal temperature of ≥ 39.5°C. Cows with acute puerperal metritis have an impaired reproductive performance, are more likely to be culled and have less milk production in comparison to healthy cows (Fourichon et al., 2000; Wittrock et al., 2011; Machado et al., 2014). Moreover, animal welfare is impaired due to painfulness of the inflamed uterus (Stojkov et al., 2015). The widespread use of 3rd generation cephalosporins for the treatment of acute puerperal needs to be re-evaluated due to the worldwide debate on emerging antibiotic resistances (Chenault et al., 2004; Lima et al., 2014; Machado et al., 2014). Hence, there is the need to encourage prudent use of antibiotics and alternative therapies to antibiotics. Therefore, the overall objective of this work was 1) to evaluate the diagnosis of acute puerperal metritis in order to reduce false positively treated cows and 2) to evaluate an alternative treatment of acute puerperal metritis.
Even though measurement of rectal temperature is part of most fresh cow protocols (Smith and Risco, 2005), this time consuming procedure is only implemented by 34% of farms in Germany (Heuwieser et al., 2010). Therefore, the first study investigated if milk temperature measured by an automatic milking system is a reliable indicator of body temperature of dairy cows and if cows with fever could be detected automatically. The results showed a moderate correlation between vaginal temperature and milk temperature (r = 0.52). Moreover, vaginal temperature was higher (39.1 ± 0.4°C, n = 418) than milk temperature (38.6 ± 0.7°C, n = 418) with a mean difference of 0.5 ± 0.6°C, which might have been caused by cooling effects of the milk tubes. Two different approaches assessed the ability to identify cows with fever by milk temperature. In the first approach, the combination of 39.0°C as a threshold for the milk temperature and 39.5°C for at least 2 h per day as a threshold for vaginal temperature resulted in the highest combination of sensitivity (0.65) and specificity (0.65). In the second approach it was evaluated, if measurement of milk temperature could identify cows with fever at a given milking event. A threshold of milk temperature > 38.7°C delivered the best combination of sensitivity (0.77) and specificity (0.66) when fever was defined as vaginal temperature ≥ 39.5°C. In conclusion, milk temperature should be interpreted with great caution because of high false positive and false negative rates and therefore cannot be recommended as an alternative to measurement of rectal temperature.
Because a lack of objectively measurable criteria for the diagnosis of acute puerperal metritis has been reported (Sannmann et al., 2013a; Burfeind et al., 2014b), the second study evaluated the association between an acute phase protein and acute puerperal metritis. Specifically, it was evaluated if, in addition to acute puerperal metritis, other variables occurring within the first 5 DIM are associated with elevated haptoglobin at 5 DIM. The results showed that primiparous cows had a greater median haptoglobin concentration (2.25 g/L, IQR 1.45−2.50, n = 146) than multiparous cows (1.13 g/L IQR 0.52−2.22; n = 302). Therefore, different haptoglobin thresholds based on references from literature for all cows (1.4 g/L), primiparous (2.49 g/L) and multiparous cows (1.4 g/L) were used for further analysis. Cows with elevated BHBA (≥ 1.2 mmol/L) at 5 DIM had 2.39−2.87 times the odds of elevated haptoglobin depending on parity. Multiparous cows with assisted calving had a 2.46 times higher risk for elevated haptoglobin whereas primiparous cows with assisted calving had no elevated risk for elevated haptoglobin at 5 DIM. Moreover, multiparous cows with retained fetal membranes were 5.51 times as likely to have elevated haptoglobin as cows without retained fetal membranes. Finally, acute puerperal metritis within the first 5 DIM was highly associated with elevated haptoglobin (odds ratio: 2.74−5.01) depending on parity. In summary, different variables occurring at the same time period as acute puerperal metritis were associated with elevated haptoglobin at DIM 5. It is speculated, that the association of calving ease, retained fetal membranes and periparturient metabolic stress could explain the reported moderate sensitivity and specificity (Huzzey et al., 2009; Dubuc et al., 2010; Burfeind et al., 2014a) in the detection of metritic cows by measuring serum haptoglobin concentrations. When using haptoglobin as a diagnostic tool for the detection of acute puerperal metritis, these potential confounders have to be considered.
The third study was designed in order to compare the efficacies of ketoprofen and ceftiofur for the treatment of acute puerperal metritis. The effect on clinical cure, subsequent milk yield, prevalence of endometritis and reproductive performance were evaluated. Cows with acute puerperal metritis were allocated to treatment with ketoprofen or treatment with ceftiofur and cows that showed fever on d 4 to 7 after inclusion received an extended treatment with ceftiofur. Cows treated with ketoprofen had 3.47 times the odds (n = 558) of extended treatment compared to cows treated with ceftiofur. Occurrence of purulent vaginal discharge was similar among treatment groups (ketoprofen: 56%, ceftiofur: 53%). However, cows with extended treatment had 2.00 times the odds (n = 438) of endometritis compared to cows without extended treatment. Treatment group did not affect mean 100 DIM milk yield (ketoprofen group 3531 ± 816 kg, ceftiofur group 3571 ± 743 kg), first artificial insemination pregnancy risk (ketoprofen group: 20% vs. ceftiofur group: 25%; n = 480), time to first artificial insemination (ketoprofen group: 75 d; 95% CI: 73−77 d vs. ceftiofur group: 73 d; 95% CI: 71−75 d; n = 437) and time to pregnancy (ketoprofen group: 133 d; 95% CI: 120−147 d vs. ceftiofur group: 143 d; 95% CI: 132−154 d; n = 437). In conclusion, more than half of the cows initially treated with ketoprofen needed an extended treatment with ceftiofur (61%). However, there is potential of reducing antibiotic use by utilizing ketoprofen for the treatment of acute puerperal metritis as up 39% of cows initially treated with ketoprofen did not need an extended treatment and the occurrence of purulent vaginal discharge, milk yield and reproductive performance was not negatively affected.
Overall, the first two studies on the diagnosis of acute puerperal metritis clearly demonstrated that further research for an improvement of the diagnosis of acute puerperal metritis is still warranted. Measurement of milk temperature by automatic milking systems to detect febrile cows is insufficient, because of high false positive and false negative rates. Although a high association between acute puerperal metritis and serum haptoglobin could be shown in the second study, also other variables impacting the same postpartum period as acute puerperal metritis were associated with elevated haptoglobin and therefore have to be considered when haptoglobin is used as a diagnostic measure. The third study showed that in terms of alternative treatment of acute puerperal metritis, ketoprofen could be used as an initial treatment without negative effects on reproductive performance or milk production. Finally, antibiotic usage could be minimized and moreover, the cow´s welfare could potentially be improved due to the analgesic effect of ketoprofen.