Scott L. Hill

Ovulation timing and conception risk after automated activity monitoring in lactating dairy cows1

Using 1 market-available activity monitor, 3 experiments were conducted in dairy cows to determine timing of ovulation, compare within-herd conception risk of cows inseminated based on activity monitors versus timed artificial insemination (AI), and determine conception risk of cows inseminated at various intervals after achieving an activity threshold. In experiment 1, ovaries were scanned every 3h by transrectal ultrasonography to determine the time of ovulation beginning 14 ± 0.5 h after the achieved activity threshold (n=132) or first standing event (n=59), or both (n=59). Progesterone at the first ovarian scan (0.1 ± 0.01 ng/mL) and ovarian structures [1 or 2 preovulatory-sized follicles (16.5 ± 0.2 mm)] confirmed that 88.6% of cows identified by activity were in estrus. The remaining 15 cows (11.4%) with a corpus luteum and elevated progesterone concentration (5.3 ± 0.5 ng/mL) were classified as false positives. The average interval from first standing event to ovulation (n=59) differed slightly from the interval after the achieved threshold (26.4 ± 0.7 vs. 24.6 ± 0.7 h, respectively). In 97 cows fitted with activity monitors, that interval was 25.7 ± 0.4 h. In experiment 2, the conception risk in 394 cows in 1 herd fitted with activity monitors was compared with that of 413 cows submitted to a timed AI program through 3 AI services. Days to first AI were reduced in cows fitted with activity monitors, and conception risk after activity threshold was less than that for timed AI at first service because of differing days in milk at first AI. Both median and mean days to pregnancy, however, were reduced in activity-group cows by 10 and 24 d, respectively, compared with timed AI cows. In experiment 3, 4,019 cows in 19 herds were inseminated after achieving the activity threshold. Conception risk was determined for cows inseminated at various intervals after the achieved activity threshold. A curvilinear conception risk curve peaked at 47.9% for primiparous cows inseminated between 13 and 16 h, whereas conception risk in multiparous cows was steady at 34% through 12 h and decreased thereafter. These experiments demonstrate that time of ovulation after activity threshold closely resembles the time of ovulation after first standing estrus. Time of insemination up to 12h after the activity threshold produced similar conception risks for multiparous cows, whereas intervals shorter than 13 and greater than 16 h in primiparous cows seemed to compromise their conception risk. Although conception risk may not be improved at individual inseminations after achieving an activity threshold, the rate of achieving pregnancy is hastened. Activity monitors can accurately predict ovulation and time of AI.

Expression of estrus before fixed-time AI affects conception rates and factors that impact expression of estrus and the repeatability of expression of estrus in sequential breeding seasons

Expression of estrus after PG and before fixed-time AI has been reported to change the uterine environment, increase accessory sperm numbers, fertilization rates, and overall embryo survival. Thus, expression of estrus can strongly impact overall pregnancy success. Because of variation in percentage of beef females detected in estrus and number of animals per study, it can be difficult to detect a significant effect of estrus on pregnancy success. Thus, a meta-analysis was conducted using data from 10,116 beef females in 22 studies that utilized variations of the 5 most common fixed-time AI protocols (CO-Synch, 7-day CO-Synch+CIDR, 5-day CIDR, PG 6-day CIDR, and the 14-day CIDR protocols) to examine the effect of detection in standing estrus on subsequent fixed-time AI pregnancy success. A random-effects model was used to combine the studies/herds. The overall model indicated a positive effect of estrus on conception rates with cows detected in estrus before fixed-time AI having a 27% greater (P<0.05; 95% CI=22-32%) conception rate compared with those not detected in estrus. Next we determined factors that influenced expression of estrus. Data were available on 547 cows synchronized with a CIDR based fixed-time AI protocols and observed for estrus before AI during 2-4 breeding seasons. Analysis of these cows indicated that days postpartum (P=0.22) did not impact estrous expression. In contrast, BCS influenced estrous expression (P=0.04) with cows in a BCS of ≤4 (51±5%) having decreased expression of estrus compared to cows with a BCS>4 (≥70±4%). Initiation of estrous cycles before the breeding season also influenced estrous expression (P=0.03), with anestrous cows having greater expression of estrus compared with estrus-cycling cows (78±5% vs. 70±5%, respectively). In conclusion, among all currently recommended fixed-time AI protocols, cows detected in estrus before fixed-time AI had improved conception rates, with BCS and estrus-cycling status having the greatest influence on expression of estrus.

Progesterone status, parity, body condition, and days postpartum before estrus or ovulation synchronization in suckled beef cattle influence artificial insemination pregnancy outcomes

Our objective was to assess the effects of progesterone before initiating an estrus- or ovulation-synchronization program in addition to the influence of parity, BCS, and days postpartum on resulting pregnancy rates per AI. Experimental data were combined from 73 herd-year studies consisting of more than 8,500 suckled beef cows exposed to variants of the CO-Synch program. Blood was harvested from samples collected at 10 and 0 d before the onset of CO-Synch, and progesterone concentrations of the samples were determined. The progesterone environment preceding synchronization was assessed in 3 ways on the basis of progesterone concentrations measured in the 2 defined blood samples. All binomial logistic regression models used procedure GLIMMIX in SAS and included the fixed effects of program duration, inclusion of progesterone via an intravaginal insert, parity, days postpartum at AI, BCS, and appropriate interactions. In addition, model 1 included 3 categories of progesterone concentrations (low [<1 ng/mL], medium [1.00 to 3.99 ng/mL], and high [≥4.00 ng/mL] concentrations) at 10 and 0 d before synchronization and their interaction. Model 2 included 4 categories defining the stage of the estrous cycle (late diestrus, early diestrus, and proestrus-estrus-metestrus) or anestrus, at which cows started the synchronization program. Model 3 defined cows as cycling or noncycling at the onset of the program. Significant effects of progesterone supplementation, which hormone was used to initiate the timed AI program, parity, BCS, days postpartum, and progesterone status assessed in 3 ways were consistent in nearly all models. Progesterone status at the onset of synchronization was not important to pregnancy outcomes in multiparous cows, whereas pregnancy rate per AI was suppressed in primiparous cows that began in a low-progesterone environment (proestrus, estrus, metestrus, or anestrus). A significant 3-way interaction of parity, BCS, and days postpartum in 2 models reinforced the importance of these factors to AI pregnancy outcomes. Ancillary analyses identified the significant effects of cycling status and BCS as well as days postpartum on luteolytic response to PGF(2α). Pregnancy loss of 2.7% to 4.2% was detected to occur between a positive pregnancy diagnosis at 35 d post-AI and later stages of pregnancy. We concluded that progesterone status at the onset of the synchronization program is critical to pregnancy outcomes in primiparous but not multiparous cows.

Gonadotropin-releasing hormone increased pregnancy risk in suckled beef cows not detected in estrus and subjected to a split-time artificial insemination program

We hypothesized that GnRH would increase pregnancy risk (PR) in a split-time AI program for cows in which estrus was not detected. A total of 1,236 suckled beef cows at 12 locations in 3 states (Colorado, Kansas, and North Dakota) were enrolled. Before applying the fixed-time AI program, BCS was assessed. Cows were treated on d -7 with a progesterone insert concurrent with 100 μg GnRH and on d 0 with 25 mg PGF plus removal of the insert. Estrus-detection patches were affixed to cows at insert removal. Estrus was defined to have occurred when an estrus-detection patch was >50% colored (activated). Cows in estrus by 65 h ( = 758; 61.3% of all cows) were randomly allocated to 2 treatments: 1) 100 μg GnRH and early + GnRH (E+G; = 373) or 2) AI only at 65 h (early - no GnRH [E-G]; = 385). The remaining cows were randomly allocated to 2 treatments: 1) 5(L+G; = 252) or 2) AI only at 84 h (late no GnRH [L-G]; = 226). Pregnancy was determined 35 d after AI via transrectal ultrasound. Pregnancy risk did not differ ( = 0.68) between E+G and E-G cows (61.9 vs. 60.4%, respectively). Conversely, for cows inseminated at 84 h, PR was greater ( = 0.01) in cows that received GnRH (L+G) compared with their herd mates not receiving GnRH (L- G; 41.7 vs. 30.8%, respectively). Of those cows not detected in estrus by 65 h, 42.1% were detected by 84 h, for a total expression of estrus by all cows of 77.6%. Administration of GnRH increased ( < 0.01) PR in cows not detected in estrus by 84 h (+GnRH = 33.4% [ = 146] vs. no GnRH = 15.0% [ = 128]) but had no effect in cows expressing estrus by 84 h (+GnRH = 65.3% [ = 103] vs. no GnRH = 61.7% [ = 97]). Neither estrus expression by 65 or 84 h nor PR was influenced by BCS, parity, or days postpartum at AI. Cows had greater PR when they had been detected in estrus before AI, and PR was improved by administration of GnRH at 65 h after insert removal in cows that were not detected in estrus and inseminated at 84 h.

Using estrus detection patches to optimally time insemination improved pregnancy risk in suckled beef cows enrolled in a fixed-time artificial insemination program

A multilocation study examined pregnancy risk (PR) after delaying AI in suckled beef cows from 60 to 75 h when estrus had not been detected by 60 h in response to a 7-d CO-Synch + progesterone insert (CIDR) timed AI (TAI) program (d -7: CIDR insert concurrent with an injection of GnRH; d 0: PGF injection and removal of CIDR insert; and GnRH injection at TAI [60 or 75 h after CIDR removal]). A total of 1,611 suckled beef cows at 15 locations in 9 states (CO, IL, KS, MN, MS, MT, ND, SD, and VA) were enrolled. Before applying the fixed-time AI program, BCS was assessed, and blood samples were collected. Estrus was defined to have occurred when an estrus detection patch was >50% colored (activated). Pregnancy was determined 35 d after AI via transrectal ultrasound. Cows ( = 746) detected in estrus by 60 h (46.3%) after CIDR removal were inseminated and treated with GnRH at AI (Control). Remaining nonestrous cows were allocated within location to 3 treatments on the basis of parity and days postpartum: 1) GnRH injection and AI at 60 h (early-early = EE; = 292), 2) GnRH injection at 60 h and AI at 75 h (early-delayed = ED; = 282), or 3) GnRH injection and AI at 75 h (delayed-delayed = DD; = 291). Control cows had a greater ( < 0.01) PR (64.2%) than other treatments (EE = 41.7%, ED = 52.8%, DD = 50.0%). Use of estrus detection patches to delay AI in cows not in estrus by 60 h after CIDR insert removal (ED and DD treatments) increased ( < 0.05) PR to TAI when compared with cows in the EE treatment. More ( < 0.001) cows that showed estrus by 60 h conceived to AI at 60 h than those not showing estrus (64.2% vs. 48.1%). Approximately half (49.2%) of the cows not in estrus by 60 h had activated patches by 75 h, resulting in a greater ( < 0.05) PR than their nonestrous herd mates in the EE (46.1% vs. 34.5%), ED (64.2% vs. 39.2%), and DD (64.8% vs. 31.5%) treatments, respectively. Overall, cows showing estrus by 75 h (72.7%) had greater ( < 0.001) PR to AI (61.3% vs. 37.9%) than cows not showing estrus. Use of estrus detection patches to allow for a delayed AI in cows not in estrus by 60 h after removal of the CIDR insert improved PR to TAI by optimizing the timing of the AI in those cows.

Effects of administration of prostaglandin F at initiation of the seven-day CO-Synch+controlled internal drug release ovulation synchronization protocol for suckled beef cows and replacement beef heifers

Two experiments were conducted to determine the effect of administering PGF at the initiation of the 7-d CO-Synch+controlled internal drug release (CIDR) fixed-timed AI (TAI) protocol on pregnancy rates of suckled beef cows and replacement heifers. Within location, cows were stratified by days postpartum (DPP), BCS, and parity (Exp. 1; = 1,551) and heifers were stratified by BCS (Exp. 2; = 999) and randomly assigned to 1 of 2 treatments: 1) CO-Synch+CIDR (100-μg injection of GnRH at CIDR insertion [d -10] with a 25-mg injection of PGF at CIDR removal [d -3] followed by injection of GnRH and TAI on d 0) or 2) PG-CO-Synch+CIDR (a 25-mg injection of PGF on d -10 of the CO-Synch+CIDR protocol). Follicle diameter and corpus luteum (CL) development were assessed on d -10 and -3, and pregnancy status was determined on d 30 to 35. Blood was collected on d -20, -10, -3, and 0 relative to TAI to determine concentrations of progesterone (P4). In Exp. 1, TAI pregnancy rates did not differ ( = 0.667) between treatments and were affected by BCS ( = 0.003) and DPP ( = 0.006). Concentrations of P4 were greater ( < 0.0001) on d -3 for CO-Synch+CIDR than for PG-CO-Synch+CIDR (4.1 ± 0.2 and 3.4 ± 0.2 ng/mL, respectively). Follicle diameter on d -3 differed ( = 0.05) between PG-CO-Synch+CIDR (13.4 ± 0.3 mm) and CO-Synch+CIDR (12.5 ± 0.3 mm) treatments. Cows with P4 > 2.5 ng/mL on d -10 had greater ( = 0.024) pregnancy rate to TAI (56.5%) compared with cows with 2.5 ng/mL < P4 > 1 (43.0%), whereas cows with P4 < 1 ng/mL were intermediate (51.6%). Cows with a CL on d -10 had greater ( = 0.012) pregnancy rates to TAI than cows without a CL (66.3 vs. 39.4%, respectively). In Exp. 2, TAI pregnancy rates did not differ ( = 0.316) between treatments. Concentrations of P4 differed ( < 0.0001) on d -3 with greater concentrations of P4 for CO-Synch+CIDR than for PG-CO-Synch+CIDR (3.75 ± 0.20 ng/mL and 3.60 ± 0.21 ng/mL, respectively). Follicle diameter was similar ( = 0.749) between treatments on d -10 and -3. Regardless of treatment, cyclic status tended ( = 0.062) to improve pregnancy rates to TAI (55 vs. 45%, for cycling and noncycling heifers, respectively). We concluded that addition of PGF to the 7-d CO-Synch+CIDR protocol decreased concentrations of P4 in cows and heifers and increased follicle diameter at CIDR removal in cows but failed to increase TAI pregnancy rates.

Short communication: Change in dose delivery of prostaglandin F2α in a 5-day timed artificial insemination program in lactating dairy cows1

We hypothesized that 50mg of prostaglandin F2α (PG) on d 6 would induce luteolysis in a traditional 5-d Ovsynch-72 program [GnRH 5 d before (d 0) and 72 h after (d 8) 25-mg PG doses (d 5 and 6 after GnRH); timed artificial insemination (AI) on d 8]. Experiment 1 monitored luteal regression of original and GnRH-induced luteal tissue (corpus luteum, CL) by transrectal ultrasonography and blood serum concentrations of progesterone after both 25-mg doses of PG (d 5 and 6; control; n=31) or a single 50-mg dose of PG on d 6 (n=30). Estrous cycles were presynchronized (GnRH 7d before 25 mg of PG); 11 d later, cows were enrolled in a 5-d Ovsynch-72 program (62 to 71 d in milk) and treatments were administered. Blood was sampled for progesterone analysis and luteal structures were measured on d 0 (original CL) and d 5 through 9 to monitor original and new GnRH-induced CL. Control PG reduced luteal tissue area of original CL on d 6 and 7 compared with PG administered only on d 6, but no difference between treatments was detected by d 9. In contrast, no differences were detected in luteal tissue area of the induced CL on d 5 through 9. Serum progesterone on d 5 through 9 differed only on d 6 for control and the 50-mg dose. Luteolysis occurred in all 31 controls, but luteolytic failure occurred in 2 of 30 cows receiving 50mg, in which no CL were present on d 0 but 1 or 3 new CL were present on d 5 in these 2 cows. Pregnancy outcome 32 d after AI was 14 of 30 (40%) compared with 15 of 30 (50%) for control versus 50-mg dose, respectively. Experiment 2 monitored luteolysis in nonpregnant repeat-service cows subsequently treated with the same 2 treatments as in experiment 1. Serum progesterone in 63 cows (serum progesterone ≥1 ng/mL on d 5) on d 5, 6, and 8 differed only on d 6 for control and the 50-mg dose. Luteolysis occurred in 32 of 34 controls and in 29 of 29 cows treated with 50mg. Pregnancy outcome 32 d after AI was 17 of 33 (52%) compared with 13 of 29 (45%) for control versus 50-mg dose, respectively. We concluded that the single 50-mg dose was equivalent to the control based on actual luteal tissue regression and decreased progesterone.

Pregnancy outcomes after change in dose delivery of prostaglandin F2α and time of gonadotropin-releasing hormone injection in a 5-day timed artificial insemination program in lactating dairy cows1

We demonstrated that 50mg of PGF₂α on d 6 successfully induced luteolysis in lactating dairy cows enrolled in a traditional 5-d Ovsynch-72 program [GnRH injection 5 d before (d 0; GnRH-1) and 56 (p.m. on d 7; GnRH-2) or 72 h (d 8; GnRH-2) after a 25-mg injection of PGF₂α (d 5 and 6 after GnRH injection); timed artificial insemination (AI) on d 8]. Our current objective was to determine pregnancy outcomes in lactating dairy cows after a 50-mg injection of PGF₂α on d 6 or a 25-mg injection of PGF₂α on d 5 and 6 in a 5-d Ovsynch program. Cows in herd 1 diagnosed not pregnant between 30 and 36 d since last AI were enrolled to receive either a 50-mg injection of PGF₂α on d 6 (1 × 50; n=134) or a 25-mg injection of PGF₂α on d 5 and 6 (2 × 25; n=139) after GnRH-1 (d 0), with GnRH-2 at 72 h after PGF₂α injection (d 5), concurrent with timed AI (d 8). Cows in herd 2 diagnosed not pregnant between 34 and 40 d were treated similarly: even-tagged cows received the 2 × 25 (n=422) treatment, and odd-tagged cows received the 1 × 50 (n=450) treatment, except that GnRH-2 was administered at 56 h. Blood collected from cows in herd 1 at d 0, 5, 6, and 8 was assayed for progesterone. Luteolysis was defined to occur when progesterone concentration was ≥1 ng/mL on d 5, and 72 h later (d 8) was either <0.5 ng/mL or <1 ng/mL. Progesterone concentrations did not differ between treatments on pretreatment d 0 and 5, but were greater in 1 × 50 than 2 × 25 cows on d 6 (4.7 ± 0.2 vs. 1.1 ± 0.2 ng/mL) and d 8 (0.43 ± 0.04 vs. 0.19 ± 0.04 ng/mL), respectively. Luteolysis was greater in the 2 × 25 versus 1 × 50 treatment when the cut point was 0.5 ng/mL, whereas no difference was detected when the cut point was <1 ng/mL on d 8. Lack of complete luteolysis was greater in cows classified as early cycle on d 0 or having a new corpus luteum after d 0 because progesterone concentration was greater on d 5 and 6 than for cows classified as late cycle on d 0 or cows having low progesterone on d 0 and 5. Pregnancy per AI at 30 to 40 d did not differ between 2 × 25 and 1 × 50 cows having luteolysis by d 8 or in all cows (37.2 vs. 33.3%) in herd 1, respectively, but differed in herd 2 (24.7 vs. 19.5%; no treatment by herd interaction). We conclude that incomplete luteolysis by d 8 was greater in 1 × 50 cows using a cut point of <0.5 ng/mL at AI. The difference in pregnancy outcome in herd 2 may have resulted from insufficient time for complete luteolysis before GnRH-2 at 56 h compared with GnRH-2 at 72 h (at AI) in herd 1.

Increasing estrus expression in the lactating dairy cow.

Using an activity monitoring system (AMS) equipped with an accelerometer, 2 experiments were conducted to test the hypotheses that (1) enhancing progesterone before inducing luteolysis or (2) exposing cows to estradiol cypionate (ECP) or testosterone propionate (TP) after luteolysis would increase occurrence and intensity of estrus. Our goal was to determine if more cows could be detected in estrus by an AMS compared with other estrus-detection aids. In experiment 1, cows (n=154) were fitted with both an AMS collar and a pressure-sensitive, rump-mounted device (HeatWatch; HW) and assigned to 3 treatments: (1) no CL + progesterone insert (CIDR) for 5d, (2) CL only, or (3) CL + 2 CIDR inserts for 5d to achieve a range in concentrations of progesterone. Prostaglandin F2α was administered to all cows upon CIDR insert removal or its equivalent. Progesterone concentration up to 72h posttreatment was greatest in CL + 2 CIDR, followed by CL only, and no CL + CIDR cows. Estrus occurred 14 to 28h earlier in no CL + CIDR compared with CL-bearing cows. Estrus intensity was greater for CL + 2 CIDR than for CL-only cows. The AMS and HW detected 70 and 59% of cows defined to be in estrus, respectively. In experiment 2, cows (n=203) were equipped with both an AMS and a friction-activated, rump-mounted patch (Estrotect patch) and assigned to receive 1mg of ECP, 2mg of TP, or control 24h after PGF2α. Concentrations of estradiol 24 and 48h after treatment were greater in ECP cows compared with controls. Estrus expression detected by AMS or patches in cows defined to be in estrus tended to be greater or was greater for ECP compared with controls, respectively. Compared with controls and in response to TP or ECP, estrus occurred 8 to 18h earlier and was of greater intensity for ECP cows, respectively. The AMS and patches determined 73 and 76% of cows defined to be in estrus, respectively. Of cows exposed to the AMS, HW, or patches, 70, 61, and 75%, respectively, were detected in estrus and more than 93% of these subsequently ovulated. In contrast, of the residual cows not detected in estrus, 62 to 77% ovulated in the absence of detected estrus. Only ECP was successful in inducing more expression and intensity of estrus, and proportions of cows detected in estrus exceeded 80%. Given the large proportion of cows equipped with AMS collars ovulating in the absence of estrus, further research is warranted to determine if more pregnancies can be achieved by inseminating those cows not detected in estrus at an appropriate time when PGF2α is administered to induce luteolysis.

Five-day resynch programs in dairy cows including controlled internal drug release at two stages post-artificial insemination

Two experiments were conducted to assess pregnancy outcomes after a 5-day Ovsynch-56 Resynch (RES; gonadotropin-releasing hormone injection 5 days before [GnRH-1; d 0] and 56 hours (GnRH-2) after PGF2α [PG] injections on day 5 and 6, timed artificial insemination [TAI] on day 8) with and without a progesterone-releasing intravaginal controlled internal drug release (CIDR) 5-day insert. In Exp. 1, nonpregnant cows were enrolled on day 34 postAI: day 34 RES-CON (n = 528) or day 34 RES-CIDR (n = 503). Blood was collected for progesterone assay. Pregnancy per AI (P/AI) was diagnosed by uterine palpation per rectum at 34 and 62 days post-TAI. Only 76% of 1,031 cows had high progesterone (≥1 ng/mL) on day 34 at the nonpregnant diagnosis. No differences in P/AI were detected between treatments. The day-34 RES-CIDR cows with low (<1 ng/mL) progesterone, however, had greater (P = 0.036) P/AI than day-34 RES-CON cows (37.7 vs. 29.4%), whereas day-34 RES-CIDR cows with high progesterone had lesser P/AI than day-34 RES-CON (27.4 vs. 34.3%). In Exp. 2, cows were enrolled on day 31 post-AI after a nonpregnant diagnosis: (1) day 31 PG-3-G (n = 102): Pre-PG on day 31, Pre-GnRH on day 34, and RES on day 41 (n = 102); (2) day 41 RES-CON (n = 108) as Exp. 1, but on day 41; and (3) day 41 RES-CIDR (n = 101) as Exp. 2, but on day 41. Blood was collected for progesterone assay and ovarian structures were mapped by ultrasonography on days 31, 34, 41, 46, and 48. Pregnancy was diagnosed by ultrasonography on days 31 and 59 post-TAI. The proportion of cows with high progesterone on day 31 was 70.6%. More (P < 0.001) cows ovulated after Pre-GnRH on day 31 PG-3-G (60.4%) than for day 41 RES-CON (12.5%) or day 41 RES-CIDR (17.1%). More (P < 0.001) PG-3-G cows had luteolysis after Pre-PG on day 31 than other treatments (73.7 vs. < 11%). The proportion of cows with high progesterone on day 41 at GnRH-1 tended (P = 0.10) to be greater for PG-3-G (75.6%) than for other treatments (65 to 70%). The P/AI was greater in cows starting RES on day 41 when progesterone was low (44%) than when it was high (33%), but no treatment differences were detected 31 days after TAI (PG-3-G = 33.3%; d 41 RES-CON = 38.9%; d 41 RES-CIDR = 35.6%). We concluded that improved P/AI for cows initiating the 5-day RES on day 34 without a corpus luteum is progesterone-dependent because addition of the CIDR insert to the RES treatment improved P/AI in cows with low progesterone (Exp. 1). Although day-31 PG-3-G increased luteolysis and produced greater ovulation rates before the onset of RES, no increase in P/AI was detected compared with RES started on day 41 with or without a CIDR insert.