Research Summary - 2
Influence of Early Lactation Behavior on Reproduction and Herd Retention Following Postpartum Hyperketonemia Treatment
| Date/Time: |
8/28/2026
14:30
|
| Author: |
Sara Davis |
| Clinic: |
University of Minnesota |
| City, State, ZIP: |
Cokato, MN 55321 |
Sara Davis, DVM
1
;
Marina Dorella, DVM, MS
1
;
Megan Ruch, BS
1
;
Elise Shepley, MSc, PhD
1
;
Gerard Cramer, DVM, DVSc
1
;
Rafael Bisinotto, MVB, PhD, MS
2
;
Luciano Caixeta, DVM, PhD
1
;
1Department of Veterinary Population Medicine, University of Minnesota, Saint Paul, MN 55321
2Department of Large Animal Clinical Sciences, University of Florida, Gainesville, FL 32610
Introduction:
With increasing demands for labor efficiency, targeting treatments towards cows most likely to benefit is essential. While oral glucogenic supplements are standard therapies for hyperketonemia (HYK), treatment success remains highly variable. Early lactation behavior may help identify cows that require intervention. The objective of this study was to evaluate if early lactation behavioral metrics influence the effect of HYK treatment on reproductive performance and herd retention.
Materials and methods:
In a randomized clinical trial on a Minnesota farm, multiparous cows were screened for HYK (defined as beta-hydroxybutyrate [BHB] ≥ 1.2 mmol/L) at 3 ± 1 and 7 ± 2 DIM using a hand-held BHB meter. Cows with HYK were randomized to receive an oral glucogenic supplement (Propylene Advantage, 300 g for 3 days; TX; n=168) or no treatment (NoTX; n=130). Non-HYK cows (HYK–; n=1,387) served as a control group. Farm staff were allowed to treat cows for clinical HYK as needed (i.e. rescue treatment); these supplemental treatments were recorded and accounted for in the analysis. Cow-level reproductive and herd removal data were sourced from herd management software. Daily rumination times were captured via wearable sensors, averaged from the day of calving to the day prior to HYK diagnosis, and evaluated as continuous variables scaled to 30-minute increments to assess their mediating effects.
We analyzed time-to-event outcomes, 150-day cumulative pregnancy and 60-day herd retention, using multivariable Cox proportional hazards models. We evaluated the first service conception rate (FSCR) via logistic regression. All models incorporated fixed effects for the treatment group, continuous pre-diagnosis average daily rumination time (RT), and their interaction. Models were also adjusted for relevant covariates including parity, season, breed, baseline prepartum rumination, BHB concentration and DIM at HYK diagnosis, and the administration of rescue treatments.
Results:
Overall HYK prevalence was 18.8%. A greater proportion of TX (76%) and NoTX (70%) cows were in their ≥3rd lactation compared to HYK– cows (49%). Additionally, RT was lower for both TX (331 min/day) and NoTX (312 min/day) cows versus HYK– cows (409 min/day).
In fully adjusted models, FSCR was 40.2% (95%CI: 31.7, 48.6) for TX cows, 27.4% (95%CI: 19.1, 35.6) for NoTX cows, and 42.9% (95%CI: 40.1, 45.7) for HYK– cows. Pre-diagnosis rumination positively influenced FSCR across HYK+ groups (NoTX OR = 1.10, 95%CI: 0.97, 1.25; TX OR = 1.08, 95%CI: 0.96, 1.22 per 30-min increase in RT). At 150 DIM, the adjusted cumulative probability of pregnancy was 70.7% (95%CI: 62.6, 78.3) for TX cows, 51.4% (95%CI: 42.4, 59.7) for NoTX cows, and 74.3% (95%CI: 71.8, 76.8) for HYK– herdmates. Furthermore, pre-diagnosis rumination was a predictor of success, increasing pregnancy likelihood by 4% per 30-minute increase in RT (HR = 1.04, 95%CI: 1.01, 1.07).
Regarding 60-day herd retention, neither pre-diagnosis behavior nor therapeutic intervention fully rescued survival rates. The adjusted probability of remaining in the herd at 60 DIM was 81.8% (95%CI: 71.3, 92.3) for TX cows, 88.6% (95%CI: 81.5, 95.7) for NoTX cows, and 91.9% (95%CI: 87.4, 96.4) for HYK– cows.
Significance:
While higher pre-diagnosis rumination is associated with reproductive success, early behavior did not meaningfully alter the efficacy of clinical HYK interventions. Although the percentage of cows pregnant by 150 DIM was higher in the TX group than in the NoTX group, this difference should be interpreted with caution due to underlying survival bias. Because the metabolic tax of HYK dictates early culling regardless of treatment, the observed reproductive recovery likely reflects a resilient sub-population that survived early lactation. It is possible that the differences in reproductive performance are secondary to early culling dynamics, indicating singular interventions cannot outpace the foundational risk of early herd removal.
