The objective of this study was to evaluate the effects of subclinical hypocalcemia (SCHCa) patterns on reproductive performance in dairy cows. In a prospective observational study 916 cows from 9 herds were blood sampled on DIMs 1, 2, 4 and 8; Ca concentration was measured with atomic absorption and SCHCa was defined as Ca ≤ 2.09 mmol/L. Cluster analysis revealed 2 normocalcemic (NORM and HIGH) and five hypocalcemic (SCH-1 to 5) clusters. Cows in cluster NORM (n = 151) had mean day-to-day serum Ca (DIMs 1-8) between 2.24 and 2.41 mmol/L, and cows in HIGH (n = 167) between 2.42 and 2.53 mmol/L. Cows in SCH-1 (n = 130) and SCH-2 (n = 102) had transient and mostly mild SCHCa on day 1 and day 2, respectively. Cows in SCH-3 (n = 123) had mostly severe SCHCa on days 1 and 2, extending to day 4. Cows in SCH-4 (n = 120) and SCH-5 (n = 145) had SCHCa which culminated on days 4 and 8, respectively. Information on reproductive outcomes including pregnancy status by 120 and 200 DIM and days open were retrieved from farm records. Median (±SE) days to 1st artificial insemination (AI) until 120 DIM estimated with Kaplan-Meier analysis for cows in SCH-3, SCH-4 and SCH-5 were 100.0 ± 7.2, 109.0 ± 6.6 and 120.0 ± 3.2, respectively, and were statistically significantly higher from those in NORM and HIGH (91.0 ± 3.4 and 87 ± 3.7, respectively). Mean days (±SE) to detected pregnancy until 200 DIM estimated with Kaplan-Meier analysis for cows in SCH-3 were 161.2 ± 4.8 and were statistically significantly higher from those in clusters NORM and HIGH (155.5 ± 4.2 and 151.6 ± 4.3, respectively). The association of Ca clusters with the odds of detected pregnancy for the 1st insemination until 120 and 200 DIM was evaluated with Linear Mixed Models. Odds for pregnancy by 120 DIM did not differ among clusters, however, cows in clusters SCH-3 and SCH-5 had lower odds for pregnancy by 200 DIM compared to HIGH (OR = 0.55, P = 0.02, and OR = 0.49, P = 0.004, respectively). Covariate adjusted survival curves generated by multivariable Cox proportional hazards model revealed that: a) clusters SCH-3 and SCH-5 had the highest (36% and 38%, respectively), while NORM and HIGH the lowest (16% and 17%, respectively) proportion of cows not inseminated for the 1st time by 120 DIM, b) compared to HIGH, cows in SCH-5 had a lower hazard of pregnancy by 120 DIM (HR = 0.42, P = 0.005), c) the proportion of open cows by 200 DIM for NORM and HIGH were 57% and 47%, respectively, while for SCH-3 and SCH-5 66% and 68%, respectively, and d) compared to HIGH, SCH-3 (HR = 0.54, P = 0.008) and SCH-5 (HR = 0.50, P = 0.001) presented the lowest hazard for pregnancy by 200 DIM. In conclusion, cows that during the entire first week after calving were continuously normocalcemic had the best reproductive performance, while those of SCH-3 and SCH-5 the worst.
Keywords: Cox's proportional hazard survival analysis; Pregnancy hazard; Reproduction; Subclinical hypocalcemia patterns.
Copyright © 2022 Elsevier Inc. All rights reserved.