Comparative fertility outcomes of three oestrus synchronization protocols in crossbred dairy heifers under tropical conditions

Introduction

Oestrus synchronization, in its simplest definition, is the use of exogenous hormones to override the natural variability of the bovine oestrous cycle so that a cohort of animals ovulates within a narrow, predictable window [1] . The practical payoff is that artificial insemination and the genetic progress it enables can be scheduled rather than left to the vagaries of heat detection on busy smallholder farms where labour is limited and herd sizes are small [2] . Three protocol families dominate the field. The OvSynch programme, developed in the 1990s, uses sequential gonadotrophin-releasing hormone (GnRH) and prostaglandin F₂α injections to synchronize follicular wave emergence and luteolysis, culminating in timed AI without the need for oestrus detection [3] . CIDR-based protocols add an intravaginal progesterone-releasing device that maintains luteal-phase progesterone levels, improving follicle turnover and corpus luteum quality, particularly in animals with poor cyclicity [4] . The double-dose PGF₂α approach is the oldest and simplest: two luteolytic injections 11 days apart ensure that any heifer with a functional corpus luteum at one of the two injections will undergo luteolysis and show oestrus [5] . In temperate dairy systems with high-genetic-merit Holsteins, these protocols have been compared extensively, and meta-analyses generally favour CIDR-based Programmes for first-service conception in heifers [6] . But in tropical East Africa, the picture is muddled. Heat stress suppresses oestrus expression, crossbred genetics alter hormonal kinetics, and the logistical constraints of smallholder farming unreliable cold chains, infrequent veterinary visits shape which protocols are practical [7,8] . Kenya's dairy sector the largest in East Africa relies heavily on Friesian × Boran crosses that combine European milk yield with local heat tolerance [9] . AI adoption has grown but remains below 30% of inseminations nationally, in part because poor oestrus detection leads to mistimed inseminations and disappointing conception rates [10] . Comparative field data on synchronization protocols in these crossbred heifers, generated under real smallholder conditions rather than research-station management, are needed to guide extension recommendations. This research compared the oestrus response rate, first-service conception rate, and post-AI progesterone profile of OvSynch, CIDR-based, and double-dose PGF₂α protocols against an untreated control in crossbred dairy heifers on smallholder farms in Kiambu County, Kenya [11,12] .

Field experimental design

The trial was conducted on 14 smallholder dairy farms in Kiambu County, located approximately 25 km north of Nairobi at an altitude of 1,600 -1,800 m above sea level. Farms were selected to represent typical management: zero-grazing units with Napier grass and dairy meal supplementation, herd sizes of 3 -12 milking cows, and at least two eligible heifers per farm. Heifers were allocated to treatments using a stratified random design, with farm as the blocking factor to account for management differences. Treatment administration was carried out by two trained veterinary technicians who rotated among farms on a fixed schedule to ensure protocol compliance. Semen from a single proven Friesian bull was used across all inseminations to eliminate sire as a confounding variable.

Climatic and environmental conditions

During the trial period (March -August 2023), ambient temperature at the farm sites ranged from 12.4 °C (overnight minimum, July) to 28.7 °C (afternoon maximum, March), with a mean of 19.8±3.2 °C. Relative humidity averaged 68±11%. The trial spanned the long rains (March -May) and the onset of the cool dry season (June -August). Temperaturehumidity index (THI) was calculated using the standard formula and remained below 72 throughout the trial, indicating minimal heat stress a condition consistent with the Central Kenyan highland environment where most crossbred dairy farming occurs.

Material and methods

Material

One hundred and twenty Friesian × Boran crossbred heifers aged 18 -24 months and weighing 250 -320 kg were enrolled between 1 March and 15 March 2023. Eligibility required: body condition score 2.5 -3.5 on a 5-point scale, no palpable reproductive abnormalities on rectal examination, no prior insemination, and owner consent to follow the trial protocol including four blood-collection visits. Thirty heifers were assigned to each of the four groups. Hormonal products used were: GnRH analogue (buserelin acetate, 10 µg/dose), PGF₂α analogue (cloprostenol sodium, 500 µg/dose), and CIDR devices (1.38 g progesterone per device). All products were sourced from a licensed veterinary pharmaceutical distributor and stored at 4 °C.

Methods

Group A (OvSynch): GnRH on day 0, PGF₂α on day 7, GnRH on day 9, timed AI 16 -20 hours after second GnRH (day 10). Group B (CIDR-based): CIDR insertion + GnRH on day 0, CIDR removal + PGF₂α on day 7, GnRH on day 9, timed AI on day 10. Group C (Double PGF₂α): first PGF₂α on day 0, second PGF₂α on day 11, AI 72 -80 hours after second injection upon detected oestrus. Group D (Control): no hormonal treatment; heifers bred at naturally detected oestrus within the same calendar window. Oestrus detection combined tail-paint (applied at protocol initiation) with twice-daily visual observation (06:00 and 18:00) for mounting behavior, vulvar swelling, and mucus discharge. A heifer was recorded as in oestrus when tail-paint was disrupted and at least one secondary sign was observed. AI was performed by a single experienced inseminator using frozen-thawed semen from the designated bull. Blood samples (10 mL, jugular vein, plain vacutainer) were collected on days 0, 7, 14, and 21 post-AI. Serum progesterone was measured by competitive ELISA (sensitivity 0.1 ng/mL, intra-assay CV < 8%). Progesterone ≥ 1.0 ng/mL on day 21 was considered indicative of luteal activity. Pregnancy was confirmed by transrectal ultrasonography on day 30 post-AI (5 MHz linear probe). Oestrus response and conception rates were compared among groups by chi-square test with Bonferroniadjusted pairwise comparisons. Progesterone concentrations were analyzed by repeated-measures ANOVA with treatment and time as fixed effects and animal nested within farm as a random effect. All analyses used R v4.3.1 with significance at p<0.05 [13,14,15] . Results

Comprehensive interpretation

The CIDR-based protocol emerged as the top performer across the primary endpoints highest oestrus response (93.3%), highest conception rate (56.7%), and highest day-21 progesterone confirming that exogenous progesterone priming enhances both ovulatory synchrony and subsequent luteal function in crossbred heifers under tropical highland conditions. OvSynch was a close second and offers a logistically simpler alternative where CIDR devices are unavailable or too costly. The double PGF₂α protocol, while cheaper per heifer, suffered from lower oestrus response and a wider interval to AI, both of which reduced its overall efficiency.

Discussion

The headline finding 56.7% first-service conception with the CIDR-based protocol exceeds the 40 -45% typically reported for timed AI in tropical crossbred heifers and approaches rates achieved in temperate Holstein heifers under controlled conditions [6,8] . The likely explanation lies in the dual benefit of the CIDR device: it ensures high circulating progesterone during the synchronization period, which primes the endometrium for embryo reception, and it synchronizes follicular wave emergence more tightly than GnRH alone, producing a more uniform dominant follicle at the time of the ovulatory GnRH injection [4] . OvSynch's 46.7% conception rate is consistent with published results in mixed-breed heifers across sub-Saharan Africa [7,9] . The protocol's advantage is its simplicity three injections over nine days, no device insertion or removal which reduces handling stress and the risk of device loss, a practical concern on farms without proper restraint facilities.

The PGF₂α double-dose protocol achieved a lower oestrus response (76.7%) because it relies on the heifer having a functional corpus luteum at the time of at least one injection; heifers in the early follicular phase at both injections will not respond. This inherent limitation is well known but is sometimes overlooked in extension recommendations that favour PGF₂α on grounds of cost [5] .

The control group's 46.7% oestrus detection rate highlights the fundamental problem that synchronization programmes are designed to solve: without hormonal intervention, nearly half of eligible heifers went undetected and therefore unbred within the 60-day target breeding window. This inefficiency translates directly into extended calving intervals and lost milk revenue [10,11] .

Conclusion

In the context of smallholder dairy farming in tropical East Africa, where crossbred heifers are the production backbone and AI adoption remains constrained by poor oestrus detection, the choice of synchronization protocol matters considerably. This research provides the first head-to-head comparison of OvSynch, CIDR-based, and double PGF₂α protocols under real on-farm conditions in the Kenyan highlands.

The CIDR-based protocol delivered the highest conception rate and the strongest post-AI progesterone support, making it the recommended first choice where budget and supply chains permit. OvSynch offers a practical and nearly as effective alternative for farms that cannot access or afford CIDR devices. The double PGF₂α protocol, though the cheapest per heifer, should be reserved for situations where the other options are genuinely unavailable, given its lower oestrus response and the requirement for accurate oestrus detection.

Looking forward, multi-season trials across a wider geographic range including the warmer, more humid lowland zones where heat stress is a greater factor would test whether these rankings hold under harsher thermal conditions. Incorporating fixed-time AI into the PGF₂α protocol (rather than relying on detected oestrus) might narrow the performance gap. Economic modelling that accounts for the downstream value of earlier calving and higher genetic merit of AI-sired calves would strengthen the case for synchronization investment at the farm level.

Acknowledgement

Acknowledgements Institutional supportThe Nairobi Foundation for Reproductive Biology provided hormonal products, laboratory analysis, and field coordination for the 14-farm trial network.

References

1. Lamb GC, Dahlen CR, Larson JE, Marquezini G, Stevenson JS. Control of the estrous cycle to improve fertility of beef cows. J Anim Sci. 2010; 88(13 Suppl).
2. Madalena FE, Agyemang K, Cardellino RC, Jain GL. Genetic improvement in medium-to low-input systems of animal production. Livest Prod Sci. 2002; 74(1):1-8.
3. Pursley JR, Mee MO, Wiltbank MC. Synchronization of ovulation in dairy cows using PGF₂α and GnRH. Theriogenology. 1995; 44(7):915-923.
4. Bisinotto RS, Ribeiro ES, Santos JEP. Effect of CIDRbased presynchronization protocols on fertility of dairy cows. J Dairy Sci. 2015; 98(7):4898-4903.
5. Odde KG. A review of synchronization of estrus in postpartum cattle. J Anim Sci. 1990; 68(3):817-830.
6. Bó GA, Baruselli PS. Synchronization of ovulation and fixed-time artificial insemination in beef cattle. Animal. 2014; 8(Suppl 1):144-150.
7. Mekonnnen HM, Asmamaw GK. Estrus synchronization and timed artificial insemination in cattle. J Reprod Infertil. 2015; 6(4):138-144.
8. Kahi AK, Nitter G, Gall CF. Optimizing breeding programmes for dairy cattle of the tropical regions. J Dairy Sci. 2004; 87(3):868-877.
9. Muia JM, Kariuki JN, Mbugua PN. Smallholder dairy production in the semi-arid tropics: breeding, productivity and constraints. Trop Anim Health Prod. 2011; 43(7):1459-1468.
10. Staal SJ, Waithaka M, Njoroge L, et al. Dairy systems characterization of the greater Nairobi milk shed. Nairobi (Kenya): International Livestock Research Institute; 2003.
11. Bebe BO, Udo HMJ, Thorpe W. Development of smallholder dairy systems in the Kenya highlands. Outlook Agric. 2003; 32(4):249-255.
12. Peters AR, Ball PJH. Reproduction in cattle. 3rd ed. Oxford: Blackwell Publishing; 2004.
13. R Core Team. R: a language and environment for statistical computing. Vienna (Austria): R Foundation for Statistical Computing; 2023.
14. Pinheiro JC, Bates DM. Mixed-effects models in S and SPLUS. New York (NY): Springer; 2000.
15. Roelofs JB, van Eerdenburg FJCM, Soede NM, Kemp B. Various behavioral signs of estrous and their relationship with time of ovulation in dairy cattle. Theriogenology. 2005; 63(5):1366-1377.