Abstract
Purpose
To evaluate if assisted reproductive technology (ART) outcomes are different based on whether procedures – oocyte retrieval, insemination, embryo biopsy, or embryo transfer – are performed on a weekday versus weekend/holiday.
Methods
Retrospective cohort study of all patients ≥ 18 years old who underwent oocyte retrieval for in vitro fertilization or oocyte banking (n = 3,197 cycles), fresh or natural-cycle frozen embryo transfers (n = 1,739 transfers), or had embryos biopsied for pre-implantation genetic testing (n = 4,568 embryos) in a large academic practice from 2015–2020. The primary outcomes were as follows: oocyte maturity for oocyte retrievals; fertilization rate for insemination; rate of no result on pre-implantation genetic testing for embryo biopsy; and live birth rate for embryo transfers.
Results
The average number of procedures performed per embryologist per day was higher on weekends/holidays than weekdays. For oocyte retrievals performed on weekdays vs. weekends/holidays, there was no difference in oocyte maturity rate (88% vs 88%). There was no difference in the fertilization rate (82% vs 80%) in cycles that had intracytoplasmic sperm injection performed on weekdays vs. weekends/holidays. No difference was found in the no result rate for embryos biopsied on weekdays vs. weekends/holidays (2.5% vs 1.8%). Finally, there was no difference by weekday vs. weekend/holiday in the live birth rate per transfer among all transfers (39.6% vs 36.1%), or when stratified by fresh (35.1% vs 34.9%) or frozen embryo transfer (49.7% vs. 39.6%).
Conclusion
We found no differences in ART outcomes among women who had their oocyte retrievals, inseminations, embryo biopsies, or embryo transfers performed on weekdays versus weekends/holidays.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10815-023-02872-2.
Keywords: Assisted reproductive technology (ART), In vitro fertilization (IVF), Outcomes, Weekend
Introduction
The first live birth from in vitro fertilization (IVF) in the world occurred in 1978. Since then, the efficiency and success rate of IVF has dramatically improved and access to assisted reproductive technologies (ART) has increased [1]. Although ovarian stimulation that occurs as part of this process is controlled, its duration is not always predictable. Furthermore, many embryo transfers are performed during patient’s natural menstrual cycles [2]. As such, both oocyte retrievals and embryo transfers may occur on any day of the week, including weekends and holidays.
A landmark study in 2001 found that people admitted to hospitals on the weekend had a higher mortality than patients admitted on a weekday [3]. This ‘weekend effect’ has since been explored in multiple fields of medicine across a variety of patient populations and practice patterns, as reviewed in [4]. Within obstetrics, several large retrospective cohort studies have examined the ‘weekend effect.’ One study that examined over 1.3 million births in the United Kingdom between 2010 and 2012 identified small increases in adverse outcomes among deliveries that occurred on weekends as opposed to weekdays, including higher perinatal mortality, puerperal infection, injury to neonate, and 3-day neonatal readmission [5]. In contrast, other studies have not identified an increased risk of neonatal morbidity or mortality on weekends [6, 7]. Thus, there remains uncertainty whether the ‘weekend effect’ is real or artifactual, what may cause it (e.g. staffing differences), and how to counteract it if it does exist, as reviewed by Bray & Steventon [4].
Within ART, many programs have become used to scheduling their cycles for various reasons. In lower-volume clinics, scheduling cycle starts allows for patient batching both to avoid weekends and decrease laboratory and surgical facility utilization time over the course of a month. In higher-volume clinics, this can allow for more equal distribution of work load over the course of a full 7-day week to ensure high quality treatment of each individual patient [8]. Several approaches exist to schedule ovarian stimulation cycles. Knowing that patients on average require 7–12 days of exposure to gonadotropins [9, 10] allows clinics to work backwards from planned retrieval days to schedule stimulation starts. In the classic long GnRH agonist protocols, once suppressed, patients can be continued on the GnRH agonist until they are scheduled to start stimulation. In GnRH antagonist-controlled cycles, oral contraceptive pills are now frequently used to suppress the hypothalamic-pituitary-ovarian axis prior to stimulation start with gonadotropins, without any clear detriment to cycle outcomes [11].
In addition to scheduling cycle starts to minimize weekend procedures, providers may manipulate the day of trigger to avoid weekend oocyte retrievals. Several studies have evaluated whether advancing or delaying the day of trigger to avoid weekend oocyte retrievals changes outcomes, with seemingly no significant detrimental effect [12–14]. Only two studies have looked at whether performing ART procedures on a weekday vs. weekend affects outcomes. One study (n = 327) evaluated whether intracytoplasmic sperm injection (ICSI) outcomes, an embryology lab-driven parameter, are affected by the day on which ICSI was performed, Wednesday vs. Sunday. The authors found no significant differences in fertilization rate, implantation rate, pregnancy rate, or take-home baby rates [15]. A second study (n = 188) found no statistically significant difference in pregnancy rate of fresh embryo transfers performed on weekdays vs. weekends (40.2 vs. 54.8%, p = 0.517), although they were underpowered to detect a difference [16].
Given the limited published data, the goal of this study was to evaluate if ART outcomes are different based on whether a variety of ART procedures are performed on a weekday versus a weekend or holiday in our large, single-institution, academic program. We hypothesized that the outcomes of oocyte retrievals, insemination (ICSI in particular), embryo biopsy, and embryo transfers would be no different between those that were performed on weekdays versus weekends or holidays.
Materials and methods
Ethics approval
This study was approved as exempt by the University of Pennsylvania Institutional Review Board.
Outcomes of interest
For this study, we chose to evaluate whether the outcomes of four time- and resource-intensive ART procedures, including oocyte retrieval (Aim 1), ICSI (Aim 2), embryo biopsy (Aim 3) and embryo transfer (Aim 4), were different if they were performed on weekdays vs weekends or holidays. Procedures performed on Monday-Friday were considered Weekdays, and procedures performed on Saturday-Sunday were considered Weekends. Holidays that fell on a weekday (e.g. Labor Day) were categorized as “Weekends” because they were staffed similarly to a typical weekend.
For Aim 1 (oocyte retrievals), the primary outcome was oocyte maturity (number of mature oocytes / number of follicles ≥ 14 mm on day of trigger). Secondary outcomes included total number of oocytes retrieved and number of mature oocytes [including MII and late-maturing (MI-MII and GV-MII)] retrieved.
For Aim 2 (ICSI), the primary outcome was fertilization rate (number of 2pn zygotes / number of mature oocytes). Secondary outcomes included the number of 2pn zygotes created.
For Aim 3 (embryo biopsy), the outcome was rate of no result on pre-implantation genetic testing (PGT).
For Aim 4 (embryo transfer), the primary outcome was live birth rate (number of live births / transfer cycle). Secondary outcomes included implantation rate (number of gestational sacs seen / number of embryos transferred), clinical pregnancy (intrauterine pregnancy with fetal heartbeat / transfer cycle), and miscarriage rates (number of miscarriages / clinical pregnancy).
Selection criteria
We analyzed all IVF and FET cycles performed at a single large academic practice from May 2015 to December 2020. All data was obtained from an existing clinical database. Missing data was abstracted from patient charts in the electronic medical record system.
For Aims 1 and 2, we included all patients ≥ 18 years old undergoing oocyte retrieval for IVF or oocyte cryopreservation (n = 3,197 cycles). We excluded all donor oocyte cycles, cycles using surgically extracted sperm, cycles where rescue ICSI was performed the day following retrieval, cycles using INVOcell®, and cycles where both oocyte and embryo banking was done.
For Aim 3, we included all patients who had an embryo biopsied for PGT (n = 987 cycles). We excluded embryos that were previously frozen and then thawed for biopsy, embryos derived from donor oocytes or thawed oocytes, and embryos derived from rescue ICSI.
For Aim 4, we included all patients undergoing a fresh embryo transfer or natural (un-medicated) cycle frozen-warmed embryo transfers (FET) (n = 1,739 cycles). We excluded programmed (medicated) FET cycles as they are rarely scheduled on weekends in our practice (16 cycles over 5 years). We also excluded any transfers from egg-thaw cycles, carrier (gestational carrier or co-maternity/reciprocal IVF) cycles, cycles using INVOcell®, and transfers resulting from donor egg or embryos.
Stimulation cycles
A variety of protocols were used for ovarian stimulation; no protocols used at our institution were specifically excluded from this study. In brief, these included the following: gonadotropin-releasing hormone (GnRH) antagonist protocols with or without OCPs or estrace priming; classic luteal-phase (long) GnRH agonist protocols; minimal-stimulation GnRH antagonist protocol with use of clomiphene citrate and low-dose gonadotropin; and microdose leuprolide acetate (LA) flare protocol with high-dose gonadotropin. Patients with a history of estrogen-sensitive tumors or those at high risk of venous thromboembolism had letrozole added to suppress estradiol levels.
Follicular development in all patients was monitored using labs and ultrasound every one to three days. Final follicular and oocyte maturation was triggered using either full-dose hCG (10,000 IU), medium-dose hCG (5,000 IU), co-trigger (1,000 IU hCG with 80 units LA), or LA-only (80 units q12 hours for two doses), depending on risk of ovarian hyperstimulation syndrome.
Oocyte retrievals were performed in the standard fashion transvaginally under ultrasound guidance 36 h after trigger.
Conventional insemination and/or ICSI was performed in standard fashion. Oocyte maturity was assessed after stripping cumulus cells for oocyte cryopreservation cycles or when ICSI was performed, or at the time of fertilization check for conventional insemination cycles. Fertilization was assessed 16-18 h after insemination; normal fertilization was confirmed when two pronuclei were present (2pn).
Embryo biopsy
Embryo biopsy for PGT was performed in standard fashion. Embryos underwent assisted hatching on day 3. Biopsy was then performed on day 5, 6, or 7 depending on when the embryo was deemed of good quality for biopsy and cryopreservation.
Transfer cycles
All embryo transfers were performed using a standardized process under ultrasound guidance. Fresh embryo transfers were performed on day 3 or day 5 after oocyte retrieval with standard luteal support. Natural cycle FETs included unmedicated natural cycles for normo-ovulatory patients and cycles induced with clomiphene citrate or letrozole for oligo-ovulatory or anovulatory patients per provider preference. In these natural cycle FETs (with or without hCG trigger), embryo transfer was performed on “day 5,” with day of hCG trigger defined as “day -2” and day of ovulation (follicle collapse on ultrasound and rise in progesterone) defined as “day 0”. Luteal support was provided in the form of vaginal progesterone daily starting on “day 0,” with embryo transfer performed on the 6th day of supplementation. Medicated natural cycles with exogenous administration of clomiphene citrate, letrozole, or gonadotropins to induce ovulation in anovulatory or oligo-ovulatory patients were considered natural cycles.
Staffing
From a laboratory persepective, staffing in the embryology laboratory varied over the years included in the study, ranging from 5 to 7 embryologists (Supplementary Table 1). The cycle to embryologist ratio ranged from 186 to 235 cycles per embryologist per year over this time frame. Any given weekday had at least two embryologists working (one of whom had to be a senior, as defined in [17]), with other embryologists working as well if it was not their day off to account for weekend coverage. All weekends/holidays were covered by at least one senior embryologist. If there were more than three retrievals scheduled, a backup embryologist (senior, embryologist, or junior) was called in for that weekend/holiday day. As cycle volume increased from year to year and the number of tasks to be performed on a daily basis increased, we lowered the threshold to calling in a backup embryologist if more than two retrievals were scheduled. From a scheduling perspective, only the following change was made: oocyte cryopreservation cases are scheduled first on weekdays and last on weekends/holidays to ensure that the oocyte vitrification takes place on time per the protocol.
From a clinical staff perspective, the weekend/holiday call was equally shared between junior and senior attending physicians in our practice. A full complement of nurses worked during the week. At least one nurse worked on weekends/holidays; a second nurse was called in if there is more than one retrieval scheduled, and a third nurse if four or more retrievals scheduled to maintain appropriate nurse-to-patient ratios.
Statistical analysis
Baseline demographics were compared using 2-sample T-test, Wilcoxon Rank Sum, Chi-square, and Fischer’s exact test, as appropriate. To account for individuals undergoing multiple oocyte retrievals, generalized estimating equations were used to compare outcomes for all retrievals and for inseminations, adjusting for patient age, AMH, physician and embryologist performing the retrievals, embryologist performing the insemination, and insemination method (for number of 2pn and fertilization rates). Wilcoxon Rank Sum test was used for retrieval outcomes when restricting to first cycles only. To account for individuals undergoing multiple cycles for embryo biopsy and PGT, logistic regression analysis was used to compare no result rates, adjusting for patient age, day of embryo biopsy, and the embryologist performing the biopsy. To account for individuals undergoing multiple transfer cycles, generalized estimating equations were used to compare outcomes for all transfers, adjusting for patient age, day of embryo transfer (e.g. Day 3 or Day 5), number of embryos transferred, and physician and embryologist performing the transfer. Two-sample T-test and Chi-square tests were used to compare outcomes when restricting to first transfer cycles only. To calculate the average number of procedures per day, only unique weekday and weekend/holiday days were used. To calculate the average number of procedures per embryologist per day, we made the assumption that 1.5 of the 5–7 embryologists were not working on any given weekday on average, and that 1.5 were working on any given weekend/holiday day. Stata Statistical Software: Release 17 was used for all analyses (StataCorp LLC, College Station, TX, 2021).
Results
Aim 1, Oocyte retrieval: A total of 3,197 IVF cycles were included for analysis. Of these, 2,529 oocyte retrievals were performed on weekdays and 668 were performed on weekends/holidays. On average, 2.36 retrievals were performed per day on a weekday and 1.82 retrievals per day on a weekend/holiday. Accounting for the number of embryologists working, this averaged to 0.51 retrievals per embryologist per day on a weekday, as compared to 1.21 retrievals per embryologist per day on a weekend/holiday (Table 1a). Patients who had oocyte retrievals performed on weekdays were on average 0.5 years younger than those whose retrievals occurred on a weekend/holiday (36.0y vs. 36.5y), and their stimulations were one day shorter (11 vs. 12 days). Other baseline demographics were similar between groups based on AMH, BMI, ethnicity, underlying infertility diagnosis, stimulation protocol used, and type of insemination performed (Table 2). No differences were identified in the total number of oocytes retrieved, number of mature oocytes retrieved, or oocyte maturity rate (Table 3). Similarly, no differences were identified in any outcomes when restricting to first oocyte retrieval cycles only (data not shown).
Table 1.
Average number of procedures performed by embryology staff per day
| a | ||||||
| Total Retrievals | Average Retrievals per Day | Average Retrievals per Embryologist per Day | ||||
| Weekday | 2529 | 2.36 | 0.51 | |||
| Weekend | 668 | 1.82 | 1.21 | |||
| b | ||||||
| Total ICSI and ICSI/IVF Cycles | Number of Oocytes ICSI'ed | Average ICSI Cycles per Day | Average # of Oocytes ICSI'ed per Day | Average ICSI Cycles per Embryologist per Day | Average # of Oocytes ICSI'ed per Embryologist per Day | |
| Weekday | 1500 | 15,394 | 1.68 | 17.22 | 0.36 | 3.69 |
| Weekend | 410 | 4363 | 1.52 | 16.22 | 1.02 | 10.81 |
| c | ||||||
| Total Embryos Biopsied | Average Embryos Biopsied per Day | Average Embryos Biopsied per Embryologist per Day | ||||
| Weekday | 3615 | 4.69 | 1.01 | |||
| Weekend | 953 | 3.94 | 2.63 | |||
| d | ||||||
| Total Transfers | Average Transfers per Day | Average Transfers per Embryologist per Day | ||||
| Weekday | 1168 | 1.57 | 0.34 | |||
| Weekend | 571 | 1.60 | 1.07 | |||
Average procedures per day calculated as number of procedures performed divided by the number of unique weekday days or weekend/holiday days on which those procedures were performed (not shown). Average procedures per embryologist per day calculated as average number of procedures divided by average number of embryologists in the lab over the course of the study (6.17) minus 1.5 on weekdays (to account for 1–2 embryologists out on any given weekday day) or divided by 1.5 on weekends (to account for a backup embryologist being called in on some but not all weekend/holiday days)
Table 2.
Demographics of patients who underwent oocyte retrieval with or without insemination
| Weekday (n = 2529) | Weekend (n = 668) | p-value | |
|---|---|---|---|
| Age (years), Mean (SD) | 36.0 (4.5) | 36.5 (4.6) | < 0.01 |
| AMH (ng/mL), Median (IQR) | 2.1 (1.1–4) (n = 2494) | 2.1 (1.0–4.2) (n = 655) | 0.57 |
| BMI (kg/m2), Median (IQR) | 24.0 (21.6–28.2) (n = 2442) | 24.0 (21.6–28.7) (n = 637) | 0.98 |
| Race/Ethnicity, n (%) | 0.73 | ||
| Asian | 307 (13%) | 71 (11%) | |
| Black or African American | 247 (10%) | 63 (10%) | |
| Hispanic or Latino | 96 (4%) | 23 (4%) | |
| White | 1669 (70%) | 456 (72%) | |
| Other | 78 (3%) | 24 (4%) | |
| Infertility diagnosis, n (%) | 0.16 | ||
| Diminished ovarian reserve | 362 (14%) | 121 (18%) | |
| Endometriosis | 83 (3%) | 16 (2%) | |
| Tubal | 152 (6%) | 33 (5%) | |
| Uterine | 56 (2%) | 17 (3%) | |
| Ovulatory | 135 (5%) | 43 (6%) | |
| RPL | 124 (5%) | 42 (6%) | |
| Unexplained | 455 (18%) | 103 (15%) | |
| PGT-M | 137 (5%) | 40 (6%) | |
| Procreative Management | 459 (18%) | 117 (18%) | |
| Male Factor | 311 (12%) | 72 (11%) | |
| Both Female & Male Factors | 253 (10%) | 64 (10%) | |
| Stimulation protocol, n (%) | < 0.01 | ||
| Antagonist | 1535 (61%) | 403 (61%) | |
| LPL | 664 (26%) |
134 (20%) 18 (3%) |
|
| Minimal Stimulation | 30 (1%) | ||
| Microdose Flare | 255 (10%) | 97 (15%) | |
| Letrozole | 43 (2%) | 14 (2%) | |
| Days of stimulation, Median (IQR) | 11 (10–12) | 12 (10–13) | < 0.01 |
| Insemination, n (%) | (n = 2169) | (n = 567) | 0.16 |
| ICSI | 1444 (67%) | 400 (71%) | |
| Conventional | 669 (31%) | 157 (28%) | |
| Split Conventional/ICSI | 56 (3%) | 10 (2%) | |
Table 3.
Outcomes for oocyte retrievals and inseminations
| Weekday (n = 2529) | Weekend (n = 668) | p-value | |
|---|---|---|---|
| Oocytes retrieved | 12 (7–18) | 12 (7–18) | p = 0.36, *p = 0.91 |
| MII oocytes retrieved | 9 (5–14) | 9 (5–14) | p = 0.72, ^p = 0.58 |
| Total number of MII (MI-MII + GV-MII) oocytes retrieved | 9 (5–14) | 9 (5–15) | p = 0.74, ^p = 0.55 |
| Maturity Rate (MII/Follicles ≥ 14 mm) | 86% (63–108%) | 85% (64–107%) | p = 0.92, ^p = 0.88 |
| Total Maturity Rate (Total MII/Follicles ≥ 14 mm) | 88% (67–110%) | 88% (67–111%) | p = 0.92, ^p = 0.92 |
| Number of 2pn zygotes | 7 (4–11) | 7 (3–12) | p = 0.20, **p = 0.19 |
| Fertilization rate (overall) | 83 (70–94%) | 81 (67–93%) | p = 0.15, **p = 0.40 |
| ICSI | 82 (70–93%) | 80 (67–92%) | p = 0.21, **p = 0.43 |
| Conventional | 83 (71–100%) | 83 (69–100%) | p = 0.75, **p = 0.75 |
| Split Conventional/ICSI | 87 (79–94%) | 79 (74–91%) | p = 0.39, **p = 0.72 |
All values median (IQR). Unadjusted statistics using Wilcoxon Rank Sum. All adjusted statistics using generalized estimating equations to account for multiple cycles per patient
*Adjusting for Patient age, AMH, Days of stimulation, and physician and embryologist performing the retrieval
^Adjusting for Patient age, AMH, Days of stimulation, and physician performing the retrieval
**Adjusting for Patient age, AMH, ± insemination method, ± embryologist performing ICSI
Aim 2, ICSI: Of the 3,197 oocyte retrievals, a total of 2,736 underwent insemination (1,844 via ICSI and 66 via split insemination of both conventional and ICSI) (Table 2). The number of oocytes inseminated via ICSI averaged out to 3.69 per embryologist per day on weekdays vs. 10.81 oocytes per embryologists per day on weekend/holidays (Table 1b). There were no differences found in the fertilization rate or number of 2pn zygotes created in cycles that had ICSI performed on the weekday vs. weekend/holiday (Table 3). Similarly, no differences were found in the fertilization rate overall (any insemination type), for cycles using conventional insemination, or in split insemination cycles. These findings persisted when restricting to first oocyte retrieval cycles only (data not shown).
Aim 3, embryo biopsy: A total of 987 cycles were included for this analysis, with a total of 4,568 biopsied embryos. Of these, 3,615 embryos were biopsied on a weekday and 953 embryos were biopsied on a weekend/holiday. This averaged out to 1.01 vs. 2.63 embryos per embryologist per day on a weekday vs. weekend/holiday, respectively (Table 1c). Embryos biopsied on weekends tended to come from younger patients (35.5y vs. 36.6y, p < 0.01) and were more likely to be biopsied on Day 5 (Supplementary Table 2). There was no difference in the no result rate between embryos biopsied on the weekday (2.5%) as compared to the weekend (1.8%) (p = 0.22, adjusted p = 0.17).
Aim 4, embryo transfer: A total of 1,739 transfers were included for analysis. Of these 1,235 were fresh transfers and 504 were FETs. On average, there were 0.34 vs. 1.07 transfers per embryologist per day on weekdays vs. weekend/holidays, respectively (Table 1d). Among all transfers, those that were performed on the weekends/holidays were more likely to be fresh than frozen (Table 4). Among fresh transfers, patients who had transfers performed on a weekday were on average 0.6 years older than those with weekend/holiday transfers (36.0 ± 4.3 vs. 35.4 ± 4.3y, p = 0.02), but they had similar other baseline demographics (data not shown). Among FETs, patients who had transfers performed on a weekday were one year younger both at time of embryo creation (i.e. oocyte retrieval) (35.5 ± 3.4 vs 36.6 ± 3.6y, p = 0.003) and embryo transfer (34.7 ± 3.5 vs 35.7 ± 3.6y, p = 0.004) than patients with weekend/holiday transfers. In general, there were no differences identified in transfer outcomes (implantation rate, clinical pregnancy rate, miscarriage rate, or live birth rates) when looking at all transfers (Table 5), all transfers stratified by fresh or FET, or when restricting to all first transfer cycles or first transfer cycles stratified by fresh or FET (data not shown). The only statistically significant difference identified was a lower live birth rate among all FETs performed on weekends/holidays (39.6%) vs. weekdays (49.7%), but this difference attenuated after adjusting for patient age at the time of embryo creation, day of embryo transfer, number of embryos transferred, whether it was an embryo that had been biopsied for preimplantation genetic testing, and the embryologist and physician performing the transfer.
Table 4.
Demographics of patients who underwent embryo transfer
| Weekday (n = 1168) | Weekend (n = 571) | p-value | |
|---|---|---|---|
| Age (years), Mean (SD) | 35.9 (4.1) | 35.7 (4.2) | 0.39 |
| AMH (ng/mL), Median (IQR) | 2.2 (1.1–4.0) (n = 1143) | 2.2 (1.1–3.7) (n = 562) | 0.98 |
| BMI (kg/m2), Mean (SD) | 26.0 (6.0) (n = 1124) | 26.0 (5.8) (n = 541) | 0.91 |
| Race/Ethnicity, n (%) | 0.84 | ||
| Asian | 133 (12%) | 59 (11%) | |
| Black or African American | 121 (11%) | 54 (10%) | |
| Hispanic or Latino | 49 (4%) | 20 (4%) | |
| White | 788 (69%) | 389 (71%) | |
| Other | 48 (4%) | 27 (5%) | |
| Infertility diagnosis, n (%) | 0.49 | ||
| DOR | 158 (14%) | 67 (12%) | |
| Endometriosis | 38 (3%) | 22 (4%) | |
| Tubal | 91 (8%) | 43 (8%) | |
| Uterine | 20 (2%) | 15 (3%) | |
| Ovulatory | 75 (6%) | 31 (5%) | |
| RPL | 34 (3%) | 22 (4%) | |
| Unexplained | 284 (24%) | 147 (26%) | |
| PGT-M | 27 (2%) |
5 (1%) 11 (2%) |
|
| Procreative Management | 24 (2%) | ||
| Male Factor | 283 (24%) | 139 (24%) | |
| Both Female & Male Factors | 133 (11%) | 69 (12%) | |
| Transfer Type, n (%) | 0.02 | ||
| Fresh | 808 (69%) | 427 (75%) | |
| Frozen | 360 (31%) | 144 (25%) | |
| Stimulation protocol (Fresh Transfers only), n(%) | 0.01 | ||
| Antagonist | 427 (53%) | 222 (52%) | |
| LPL | 243 (30%) | 158 (37%) | |
| MinStim | 21 (3%) | 7 (2%) | |
| Microdose Flare | 116 (14%) | 39 (9%) | |
| Letrozole | 1 (0.1%) | 1 (0.1%) | |
| Number of embryos transferred | 0.26 | ||
| Median (IQR) | 1 (1–2) | 1 (1–2) | |
| Mean (SD) | 1.4 (0.6) | 1.4 (0.6) | |
Table 5.
Outcomes of embryo transfers
| Weekday (n = 1168) | Weekend (n = 571) | p-value | |
|---|---|---|---|
| Implantation Rate | 44.8 (41.9–47.6%) | 41.1 (37.1–45.1%) | p = 0.14 |
| Fresh | 38.9 (35.6–42.2%) | 37.8 (33.4–42.2%) | p = 0.69, *p = 0.59 |
| FET | 57.9 (52.7–63.1%) | 50.7 (42.0–59.4%) | p = 0.18, *p = 0.35 |
| Clinical Pregnancy Rate | 45.9 (43.1–48.9%) | 41.3 (37.3–45.3%) | p = 0.07 |
| Fresh | 41.7 (38.2–45.2%) | 39.8 (35.2–44.4%) | p = 0.52, *p = 0.47 |
| FET | 55.6 (50.5–60.6%) | 45.8 (37.7–53.9%) | p = 0.05, *p = 0.12 |
| Miscarriage Rate | 18.8 (15.5–22.0%) | 19.1 (14.1–24.1%) | p = 0.92 |
| Fresh | 21.2 (16.9–25.5%) | 18.5 (12.8–24.2%) | p = 0.45, *p = 0.67 |
| FET | 14.6 (9.6–19.5%) | 20.5 (11.3–29.8%) | p = 0.24, *p = 0.21 |
| Live Birth Rate | 39.6 (36.8–42.5%) | 36.1 (32.2–40.0%) | p = 0.15 |
| Fresh | 35.1 (31.8–38.5%) | 34.9 (30.4–39.3%) | p = 0.93, *p = 0.79 |
| FET | 49.7 (44.5–54.9%) | 39.6 (31.8–47.4%) | p = 0.04, *p = 0.08 |
All statistics using generalized estimating equations
*Adjusting for Patient age, Embryo age (days) at time of transfer, number of embryos transferred, physician and embryologist performing the transfer (and embryo biopsy status for FETs)
Discussion
In this large retrospective cohort, we found no differences in ART outcomes among patients who had their oocyte retrievals, inseminations, embryo biopsies, or embryo transfers performed on weekdays or weekends/holidays, despite a higher average workload per embryologist on weekends/holidays than weekdays.
Our findings are consistent with the two other studies identified in the literature that have also looked at weekday versus weekend outcomes. Setti et al. found no difference in fertilization rates, implantation rates, pregnancy rates, or take-home baby rates between patients who had ICSI performed on Wednesdays (n = 196) vs. Sundays (n = 131) as representative of a weekday vs. weekend [15]. In our larger study (1,444 weekday vs. 400 weekend ICSI cycles), we similarly found no difference in fertilization rate in the time- and resource-intensive ICSI cycles. We expanded on this analysis by evaluating all insemination, conventional insemination, and split (conventional and ICSI) insemination cycles, where we also did not identify differences in fertilization rate.
In turn, Hasdemir et al. looked at clinical pregnancy rate between fresh embryo transfers performed on weekdays (n = 156) or weekends (n = 32) [16]. They identified no statistically significant difference in clinical pregnancy rates (40.2 vs. 54.8%, p = 0.517), although on post-hoc analysis they only had 35% power to detect this 15% difference. The authors postulated that the higher pregnancy rate seen on weekends could be due to Saturday being more relaxing from a workload, crowdedness, and patient mood perspective. They also noted that they would need a three-times higher sample size to be able to reach statistical significance for the 14.6% difference in outcomes. In our cohort of fresh transfers, (which was more than 6 times larger), the clinical pregnancy rates were no different between weekday vs. weekend/holiday transfers (41.8 vs. 39.6%, adjusted p = 0.28). We further expanded on these findings by looking at first fresh transfers only, as well as evaluating both miscarriage and, the most clinically important finding, live birth rates. Reassuringly, we found no differences in outcomes.
Interestingly, several studies have examined whether intentionally avoiding ART procedures on weekends has a detrimental effect on ART outcomes. One study evaluated outcomes of Monday (potentially delayed), Tuesday-Thursday (weekend non-constrained), and Friday (potentially premature) oocyte retrievals. The authors found no difference in outcomes, including number of eggs, mature eggs, fertilization rate, or pregnancy rate (per embryo transfer), although the pregnancy rate trended to be lower in the Friday (potentially premature) oocyte retrieval day [21.8% vs. 27.6% (Monday) vs. 27.4% (Tuesday-Thursday)] [13]. To see if it could be the delay or advancement of trigger that affected outcomes, another group specifically looked at that question in GnRH antagonist cycles [14]. They found that compared to an ideal Monday retrieval (n = 221), non-ideal Monday retrievals (delay from Sunday, n = 251) had two more oocytes retrieved (9.88 vs 7.78, p < 0.0001) and one more embryo created (5.78 vs 4.78, p < 0.004), but this did not translate into higher biochemical pregnancy (39.04 vs. 38.91%), implantation (29.54 vs. 32.15%), or live birth rates (33.06 vs. 30.76%). In contrast, non-ideal Friday retrievals (i.e. ideally Saturday retrievals, n = 165) compared to ideal Friday retrievals (n = 207) had one less oocyte retrieved (7.95 vs 9.04, p = 0.02) and one less embryo created (4.45 vs 5.32, p = 0.01), although this similarly did not have a detrimental effect on biochemical pregnancy (35.8 vs. 40.0%), implantation (31.72 vs. 34.60%), or live birth rates (32.12 vs. 35.74%) in the fresh transfers. As the authors point out, though, the one less embryo may translate into a lower cumulative pregnancy rate when analyzing follow-up FETs. Of note, that study was not powered to detect the observed differences in pregnancy outcomes.
Unlike some large programs which can fully staff their laboratories seven days a week, we do not have enough embryologists to have a full team present in the laboratory on the weekends and holidays. The minimum number of staff necessary to perform all required procedures are scheduled, but they are usually performing more tasks on the weekend than they would on a typical weekday, as we show in Table 1. To account for this, our cycle start schedule is designed to maximize the number of oocyte retrievals that occur on weekdays and minimize weekend/holiday retrievals. Nevertheless, length of ovarian stimulation is often unpredictable and we do not systematically advance or delay trigger to avoid a weekend retrieval. As in other ART laboratories, our embryologists are extensively trained and supervised until they are independently able to perform procedures at the highest level. Only once are they independent on enough different protocols (e.g. ICSI, transfers, embryo biopsies, or oocyte vitrification) do they enter the rotation to cover weekends/holidays. Always maintaining at least the minimum number of embryologists needed and the high standards for embryologist proficiency may help explain why no differences were identified in the laboratory-specific outcomes.
Our study has several strengths. First, it is, to our knowledge, the largest study comparing ART outcomes by weekday vs. weekend/holidays procedures. Second, this is the first study that has evaluated PGT biopsy outcomes and live birth rates following embryo transfers if performed on a weekday vs. weekend/holiday. Nevertheless, several limitations should be noted. First, these results are based on a single institution, which could limit generalizability of our findings. Second, the embryology staffing calendar has been maintained by a combined system of paper, electronic calendars, and a separate time-clock software; because there is no easy way to merge these data, we could only estimate the number of embryologists working on any given weekday or weekend. Furthermore, as the same embryologist may have grown from junior to senior over the course of the study, we could not adjust our data for the rank of the embryologist performing a procedure. Finally, limitations of our electronic medical record system make some of our demographic data, including BMI and infertility diagnoses, less reliable, potentially introducing some information bias. This limitation should not affect our findings about transfer outcomes, though, as day of transfer was not considered in our practice when deciding on trigger day (weekday retrievals will often have their fresh transfers on weekends) and day of transfer cannot be predicted ahead of time for natural cycle FETs.
In conclusion, our study adds to the literature on performing ART procedures throughout the entire week versus weekdays only. In combination with existing literature, it appears that both approaches do not compromise patient outcomes, and either one could be chosen based on clinic volume and staff availability. Patients, embryologists, and providers should be reassured that having ART procedures performed on weekends or holidays does not appear to have a detrimental impact on their outcomes.
Supplementary Information
Below is the link to the electronic supplementary material.
Data Availability
Data were collected from a prospectively maintained clinical database at Penn Fertility Care at the University of Pennsylvania. Due to patient privacy concerns, de-identified data may be available by reasonable request under a data use agreement.
Declarations
Competing interests
The authors have no competing interests to declare that are relevant to the content of this article.
Footnotes
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Centers for Disease Control and Prevention Assisted Reproductive Technology Fertility Clinic and National Summary Report. US Dep Health Hum Serv. 2019;2021:112. [Google Scholar]
- 2.Lee JC, Calzada-Jorge NS, Hipp HS, Kawwass JF. Natural cycle frozen embryo transfer: a survey of current assisted reproductive technology practices in the U.S. J Assist Reprod Genet. 2023;40:891–9. [DOI] [PMC free article] [PubMed]
- 3.Bell CM, Redelmeier DA. Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Engl J Med. 2001;345:663–668. doi: 10.1056/NEJMsa003376. [DOI] [PubMed] [Google Scholar]
- 4.Bray BD, Steventon A. What have we learnt after 15 years of research into the ‘weekend effect’? BMJ Qual Saf. 2017;26:607–610. doi: 10.1136/bmjqs-2016-005793. [DOI] [PubMed] [Google Scholar]
- 5.Palmer WL, Bottle A, Aylin P. Association between day of delivery and obstetric outcomes: observational study. BMJ. 2015;351:h5774. doi: 10.1136/bmj.h5774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Reif P, Pichler G, Griesbacher A, Lehner G, Schöll W, Lang U, et al. Do time of birth, unit volume, and staff seniority affect neonatal outcome in deliveries at ≥34+0 weeks of gestation? BJOG Int J Obstet Gynaecol. 2018;125:884–891. doi: 10.1111/1471-0528.15000. [DOI] [PubMed] [Google Scholar]
- 7.Gould JB, Qin C, Marks AR, Chavez G. Neonatal Mortality in Weekend vs Weekday Births. JAMA. 2003;289:2958–2962. doi: 10.1001/jama.289.22.2958. [DOI] [PubMed] [Google Scholar]
- 8.Garcia-Velasco JA, Fatemi HM. To pill or not to pill in GnRH antagonist cycles: that is the question! Reprod Biomed Online. 2015;30:39–42. doi: 10.1016/j.rbmo.2014.09.010. [DOI] [PubMed] [Google Scholar]
- 9.Taylor HS, Pal L, Seli E. Speroff’s clinical gynecologic endocrinology and infertility, 9th edn. Philadelphia, Pennsylvania: Lippincott Williams & Wilkins (LWW); 2019.
- 10.Stout S, Dabi Y, Dupont C, Selleret L, Touboul C, Chabbert-Buffet N, et al. Stimulation Duration in Patients with Early Oocyte Maturation Triggering Criteria Does Not Impact IVF-ICSI Outcomes. J Clin Med. 2022;11:2330. doi: 10.3390/jcm11092330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Li J, Sun Y, Mo S, Wang S, Luo W. Effects of oral contraceptive for different responder women before GnRH antagonists: a systematic review and meta-analysis. Gynecol Endocrinol. 2021;37:977–986. doi: 10.1080/09513590.2021.1918664. [DOI] [PubMed] [Google Scholar]
- 12.Pratap D, Kumar P, Adiga SK. Advancing or postponing the day of human chorionic gonadotropin does not matter for the outcome in assisted reproductive technology. J Hum Reprod Sci. 2014;7:107–110. doi: 10.4103/0974-1208.138868. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ben-Chetrit A, Senoz S, Greenblatt EM. In vitro fertilization programmed for weekday-only oocyte harvest: analysis of outcome based on actual retrieval day. J Assist Reprod Genet. 1997;14:26–31. doi: 10.1007/BF02765748. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tremellen KP, Lane M. Avoidance of weekend oocyte retrievals during GnRH antagonist treatment by simple advancement or delay of hCG administration does not adversely affect IVF live birth outcomes. Hum Reprod Oxf Engl. 2010;25:1219–1224. doi: 10.1093/humrep/deq059. [DOI] [PubMed] [Google Scholar]
- 15.Setti AS, de Braga DPAF, de Figueira RCS, Laconelli A, Jr, Borges E, et al. Do weekend oocyte retrievals have an impact on the intracytoplasmic sperm injection cycles outcome? Arch Med Sci AMS. 2012;8:368. doi: 10.5114/aoms.2012.28569. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hasdemir PS, Kamali MB, Calik E, Ozcakir HT. A Comparison of Success Rates of Embryo Transfer on Weekdays and Weekends. Int J Fertil Steril. 2016;10:48. doi: 10.22074/ijfs.2016.4768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Practice Committees of the American Society for Reproductive Medicine (ASRM) and the Society for Reproductive Biologists and Technologists (SRBT). Electronic address: asrm@asrm.org. Comprehensive guidance for human embryology, andrology, and endocrinology laboratories: management and operations: a committee opinion. Fertil Steril. 2022;117:1183–202. [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data were collected from a prospectively maintained clinical database at Penn Fertility Care at the University of Pennsylvania. Due to patient privacy concerns, de-identified data may be available by reasonable request under a data use agreement.