- Email: [email protected]
Lapse in embryo transfer training does not negatively affect clinical pregnancy rates for reproductive endocrinology and infertility fellows
Lapse in embryo transfer training does not negatively affect clinical pregnancy rates for reproductive endocrinology and infertility fellows Jessica Kresowik, M.D., Amy Sparks, Ph.D., Eyup H. Duran, M.D., and Divya K. Shah, M.D. Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, University of Iowa Hospitals and Clinics, Iowa City, Iowa
Objective: To compare rates of clinical pregnancy (CPR) and live birth (LBR) following embryo transfer (ET) performed by reproductive endocrinology and infertility (REI) fellows before and after a prolonged lapse in clinical training due to an 18-month research rotation. Design: Retrospective cohort study. Setting: Not applicable. Patient(s): All women undergoing in vitro fertilization (IVF) and IVF–intracytoplasmic sperm injection (ICSI) cycles with ET performed by REI fellows from August 2003 to July 2012. Intervention: Eighteen-month lapse in clinical training of REI fellows. Main Outcome Measure(s): CPR and LBR before and after the lapse in clinical training were calculated and compared per fellow and as a composite group. Alternating logistic regression models were used to calculate the odds of clinical pregnancy and live birth following transfers performed before and after the lapse in training. Result(s): Unadjusted odds of clinical pregnancy and live birth were similar between the two time periods both for individual fellows and for the composite group. Alternate logistic regression analysis revealed no significant difference in CPR (odds ratio [OR] 0.94, 95% confidence interval [CI] 0.83–1.07) or LBR (OR 1.05, 95% CI 0.94–1.18) after the lapse in training compared with before. Conclusion(s): A research rotation is common in REI fellowship training programs. This prolonged departure from clinical training does not appear to negatively affect pregnancy outcome Use your smartphone following fellow ET. (Fertil SterilÒ 2015;103:728–33. Ó2015 by American Society for Reproto scan this QR code ductive Medicine.) and connect to the Key Words: Embryo transfer, training, lapse, REI, fellows Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/kresowikj-lapse-embryo-transfer-rei-fellows/
E
mbryo transfer (ET) is a nuanced part of the in vitro fertilization (IVF) process, where success is affected by both provider technique and experience (1, 2). Factors thought
to be associated with implantation following ET include the use of soft versus rigid transfer catheters and provider perception of the ease of transfer (3–5). Avoiding negative
Received September 8, 2014; revised and accepted December 5, 2014; published online December 31, 2014. J.K. has nothing to disclose. A.S. has nothing to disclose. E.H.D. has nothing to disclose. D.K.S. has nothing to disclose. Presented at the 69th American Society for Reproductive Medicine Annual Meeting, Boston, Massachusetts, October 12-17, 2013. Reprint requests: Divya K. Shah, M.D., Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, University of Iowa Hospitals and Clinics, 31324 PFP, 200 Hawkins Drive, Iowa City, Iowa 52242 (E-mail: [email protected]). Fertility and Sterility® Vol. 103, No. 3, March 2015 0015-0282/$36.00 Copyright ©2015 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2014.12.102 728
discussion forum for this article now.*
* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.
pressure when removing the catheter as well as maintaining a controlled velocity when depressing the plunger also have been associated with success (6). It is expected that reproductive endocrinology and infertility (REI) fellowship programs ensure adequate training and competence in ET, although the best training modality and the time required to accomplish this goal are not well understood. The number of ETs required to reach proficiency has been shown to vary greatly between studies as well as VOL. 103 NO. 3 / MARCH 2015
Fertility and Sterility® between individual learners, with estimates ranging from 15 to 100 ETs (7–10). In addition, the implementation of a formalized training program using intrauterine insemination did not appear to improve a learner's time to proficiency compared with those who did not undergo similar training (8). A further complicating factor in the ET training process for many REI fellowship programs in the United States is the requirement for 18 months of dedicated research time. This ‘‘research block’’ is frequently scheduled during the 2nd year of fellowship and removes fellows from participation in patient care. The effect of this prolonged lapse in clinical training on success rates after ET remains unknown. We sought to evaluate the impact of a lapse in training on clinical pregnancy rate (CPR) and live birth rate (LBR) following ET performed by REI fellows.
MATERIALS AND METHODS Study Population Data were retrospectively queried from the University of Iowa's IVF database from August 2003 till July 2012. A total of five fellows started and completed their training during the study time period. All IVF and IVF–intracytoplasmic sperm injection (ICSI) cycles with ET performed by REI fellows during this time period were included. It is institutional practice for 1st-year fellows to observe ETs for 1 month and then perform all subsequent ETs when covering the IVF service. The database does not reliably track the number of ETs initiated by fellows but ultimately taken over by attending physicians, so these transfers were excluded from analysis. Embryo transfers were performed in the IVF procedure room without analgesia. The cervix was gently cleansed with warm saline solution and any excess cervical mucus was aspirated. A mock ET was first performed with the use of a Wallace Trial Catheter (Smiths Medical) to ensure passage through the internal cervical os. If passage of the catheter required significant manipulation, the ‘‘afterload’’ method with retention of the outer sheath was used at the discretion of the transferring provider (11). If a fellow was unable to successfully pass the mock transfer catheter after a reasonable attempt, the transfer was performed by the supervising physician. Once the mock ET was complete, embryos were loaded into either a Wallace Classic Embryo Replacement Catheter (August 2003–June 2004) or Wallace Sure View Embryo Replacement Catheter (June 2004 and subsequently) for transfer. Transfer catheters were loaded with a solid column of culture medium, followed by 5 mL air, 20 mL medium containing the embryo(s), and a final 5 mL air. All ETs (including mock ETs) were performed under transabdominal ultrasound guidance provided either by the supervising faculty physician or a trained sonographer. All patients signed consents for data collection and quality assurance analysis at the start of treatment. The University of Iowa Institutional Review Board deemed the project to be exempt from further review because of the noninterventional and quality assurance nature of the study. VOL. 103 NO. 3 / MARCH 2015
Exposure and Outcome Variables The exposure was an 18-month lapse in clinical training while on a research rotation. Fellow ETs were divided into two groups: those performed before (pre-lapse) and after (post-lapse) the 18-month departure from clinical training. The primary outcome was CPR per ET (defined as the presence of an intrauterine gestational sac on ultrasound). The secondary outcome was LBR per ET (defined as the birth of an infant at or beyond 20 weeks gestation).
Potential Confounding Variables Patient demographic variables, including female age, body mass index (BMI), race, antral follicle count (AFC), and infertility diagnosis, were noted in the medical record at the time of cycle initiation. Race was classified as white, African-American, Asian, multiracial, or other by the nurse or providing physician. Categories of infertility diagnosis included male factor, tubal factor, ovulatory dysfunction, endometriosis, uterine factor, and diminished ovarian reserve. IVF cycle characteristics, including cycle number (1, 2, 3, R4), use of donor oocytes, use of ICSI, total gonadotropin dosage (IU of FSH or hMG), use of fresh versus frozen embryos, number of embryos transferred (1, 2, R3), and day 3 versus day 5 ET, were recorded in the IVF database on a per-cycle basis. Additional variables reported include numbers of ETs complicated by either retention of one or more embryos in the catheter after initial transfer attempt or the need to temporarily return the catheter to the embryologist because of prolonged difficulty in gaining entry into the uterine cavity.
Statistical Analysis Descriptive statistics were used to assess the demographics of the study population. CPR and LBR per ET performed after the training lapse were compared with CPR and LBR per ET performed before the training lapse by means of the Cochran-Mantel Haenszel method, controlling for the fellow that performed the ET. This same comparison was made for each individual fellow with the use of the Pearson chisquare test, generating unadjusted odds ratios (ORs). A generalized linear model was used to test for the effect of training lapse on CPR after adjusting for potential confounding variables. This model was fitted by the alternating logistic regression algorithm to account for the association of transfer outcome in subjects treated by the same fellow as well for repeated ETs in the same subject. Female age, cycle number, use of donor eggs, fresh versus frozen cycle, and day 3 versus day 5 transfer were included a priori as covariates in the model; other potential confounders were examined but did not change effect estimates by >10% and were therefore not included in the final model.
RESULTS Five fellows completed their training from August 2003 to July 2012. A total of 793 patients and 933 ETs were included in the study. Demographic and IVF cycle characteristics of the study population are listed in Table 1. Patient age, ethnicity, 729
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
TABLE 1 Patient and cycle characteristics for embryo transfers performed by reproductive endocrinology and infertility fellows before and after an 18-mo lapse in clinical training. Characteristic
Pre-lapse group (n [ 378)
Patient Age (y), mean (SD) 33.7 (5.0) White 315 (91) BMI (kg/m2) <20 28 (7) 20–24.9 168 (44) 25–29.9 93 (25) 30–39.9 69 (18) R40 20 (5) Antral follicle count, 19 (13–30) median (IQR) Infertility diagnosis Male factor 157 (42) Tubal factor 84 (22) Ovulatory factor 66 (18) Endometriosis 45 (12) Uterine factor 21 (6) Diminished ovarian 36 (10) reserve Cycle Cycle number 1 174 (46) 2 101 (27) 3 55 (15) R4 48 (13) Use of donor oocytes 34 (9) Use of ICSI 245 (65) Total gonadotropin 24 (0–39) dose (IU) Fresh embryo transfer 259 (69) Transfer day 3 176 (47) 5 201 (53) No. of embryos transferred 1 112 (30) 2 229 (61) R3 37 (10)
Post-lapse group (n [ 555) 33.8 (4.8) 461 (91) 37 (7) 225 (41) 122 (22) 138 (25) 32 (6) 21 (13–30) 224 (40) 99 (18) 98 (18) 65 (12) 50 (9) 50 (9)
227 (41) 143 (26) 87 (16) 98 (18) 41 (7) 333 (60) 22 (0–36) 365 (66) 221 (40) 333 (60) 218 (39) 298 (53) 41 (7)
Note: Values are presented as n (%) except as noted. BMI ¼ body mass index; ICSI ¼ intracytoplasmic sperm injection. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
infertility diagnosis, use of donor oocytes, fresh cycle, use of ICSI, AFC, and gonadotropin dosage were similar between the patients undergoing ET before and after the lapse in clinical training. Patient BMI tended to be higher in the post-lapse group compared with the pre-lapse group. The post-lapse group also included a greater proportion of patients who underwent transfer of one embryo on day 5, coincident with our program's increased use of single-embryo transfer in accordance with practice guidelines from the American Society of Reproductive Medicine. The mean CPR for each fellow before and after the lapse in training are presented in Table 2; the CPR ranged from 53% to 65%. No significant difference in CPR was observed for any fellow when comparing the two time frames. Although this study was designed to have each fellow serve as their own control, the mean CPR for all ETs performed by attending physicians during the corresponding time period are included in the table as a point of reference. Figure 1 730
shows the unadjusted OR of clinical pregnancy and 95% confidence intervals (CIs) for all ETs performed during the post-lapse time frame relative to all ETs performed during the pre-lapse time frame for each individual fellow; no significant differences were seen. The common OR comparing CPR during the pre- and post-lapse time periods, pooling data from all fellows, remained nonsignificant at 0.92 (95% CI 0.70–1.21). In an isolated crude comparison of the final 25 ETs performed pre-lapse and the first 25 ETs performed post-lapse, however, a trend toward higher pregnancy rates pre-lapse and lower pregnancy rates post-lapse was noted (Supplemental Table 1, available online at www.fertstert.org). The adjusted ORs and 95% CIs for both clinical pregnancy and live birth are shown in Figure 2. Adjusting for age, cycle number, use of donor eggs, fresh versus frozen cycle, and day of ET, and accounting for repeated transfers in the same patient, no difference in rates of clinical pregnancy (OR 0.93, 95% CI 0.79–1.09) or live birth (OR 0.94, 95% CI 0.83–1.07) were noted between ETs performed by REI fellows after compared with before the lapse in clinical training. Supplemental Table 2 (available online at www.fertstert.org) reports the percentage of ETs that included retained or reloaded embryos for each fellow, before and after the lapse, as well as an approximate number of ETs that were initiated by a fellow but ultimately taken over by an attending physician. The latter variable was not reliably collected in the dataset, as stated in the Materials and Methods section, and therefore represents an approximation for reference. These difficult ETs were rare, representing fewer than 3% of all ETs performed by any given fellow.
DISCUSSION The results of this study indicate that an 18-month lapse in clinical training had no impact on CPR or LBR following ET performed by REI fellows. The impact of this finding is of particular importance as REI fellowship programs in the United States reevaluate how best to train their fellows in the critical procedure of ET. Skill decay, the loss of a trained or acquired skill following periods of nonuse, is a well described phenomenon. A meta-analysis of surgical skill decay and retention literature found that after 365 days of nonuse or nonpractice, the average participant's performance was reduced by almost a full standard deviation (12). Physical, natural, or speed-based tasks were less susceptible to skill decay than cognitive, artificial, or accuracy-based tasks. Additionally ‘‘overlearning,’’ or practicing a skill past the level of proficiency, is also protective against skill decay. An article examining general surgery residents that left clinical training for 2 years to conduct a research rotation found that residents and faculty perceived nonclinical years to have a slightly deleterious effect on technical skills. However, when their skills were evaluated in a simulation center, all residents performed at proficiency for basic and intermediary knot-tying and suturing on return from their nonclinical years (13). A perception of skill decay may affect the willingness of REI fellowship programs to permit fellows to perform VOL. 103 NO. 3 / MARCH 2015
Fertility and Sterility®
TABLE 2 Clinical pregnancy rates per embryo transfer for REI fellows before and after an 18-mo lapse in clinical training. Pre-lapse Fellow (individual) Fellow 1 2 3 4 5
Post-lapse
Attendings (aggregate)
Fellow (individual)
Attendings (aggregate)
No. ETs
% CPR
No. ETs
% CPR
No. ETs
% CPR
No. ETs
% CPR
55 91 104 96 32
64 65 53 56 59
161 114 111 493 191
56 52 69 61 56
80 168 98 135 74
63 59 57 53 57
170 323 79 151 184
57 55 62 59 59
Note: Results are presented as unadjusted mean CPR per embryo transfer. CPRs are shown for each individual fellow for the pre- and post-lapse time periods. The mean CPRs of all ETs performed by attending physicians during the same time periods are included for reference. CPR ¼ clinical pregnancy rate; ET ¼ embryo transfer. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
ET. A 2007 survey of REI fellowship programs revealed that 44% of fellows graduate without having performed ET during training (14). The Society for Reproductive Endocrinology and Infertility conducted a survey in 2012 of 64 current and former REI fellows and found that 75% of respondents anticipated performing fewer than 20 ETs in their fellowship, with 46.9% performing none (15). Perceived barriers to fellow participation in ET include patient preference, presumed negative effect on pregnancy rates, and competition with private programs without trainees (15). It is also plausible that the prospect of a prolonged absence from clinical practice may increase faculty discomfort with fellows performing ET owing to concerns regarding skill attrition. Our data suggest that an 18-month lapse in clinical training
does not appear to be detrimental to fellow performance of ET. It may be that the physical rather than cognitive nature of the ET procedure makes its less susceptible to skill decay, and that maintenance of ET skill is similar to the aforementioned ability of general surgery residents to maintain proficiency at basic suturing after a 2 year lapse in clinical training (13). Our REI division is able to effectively address this study question because institutional practice dictates fellow participation in all ETs. Our institution is also uniquely positioned to evaluate the effect of a prolonged lapse in clinical training because our fellowship is designed with a clearly defined 18month research block from month 9 to month 27 of REI fellowship. REI fellows are completely removed from clinical
FIGURE 1
Unadjusted odds of clinical pregnancy per embryo transfer for individual fellows after an 18-month lapse in clinical training. Blue dots indicate the unadjusted odds ratio (OR) of clinical pregnancy, comparing the transfers performed in the post-lapse time period with those in the pre-lapse time period for each fellow; the brown lines represent 95% confidence intervals. The red vertical line denotes an OR of 1.0. The common OR (95% confidence interval) is composed of pooled data from all fellows and remained nonsignificant. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
VOL. 103 NO. 3 / MARCH 2015
731
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
FIGURE 2
Adjusted odds of clinical pregnancy and live birth per embryo transfer after an 18-month lapse in clinical training. Blue dots indicate the adjusted odds ratio of clinical pregnancy or live birth for all embryo transfers performed by fellows in the post-lapse time period compared with those in the pre-lapse time period; the brown lines represent 95% confidence intervals. The red vertical line denotes an odds ratio of 1.0. Multivariate analysis adjusting for age, cycle number, use of donor eggs, fresh versus frozen cycle, and day of embryo transfer and accounting for repeated transfers in the same patient noted no difference in the odds of clinical pregnancy or live birth for fellow embryo transfers completed after an 18-month lapse in clinical training. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
practice during this block with no call responsibilities, thereby protecting our analyses from confounding due to occasional performance of ET during the research rotation. A final strength is our paired study design in which individual fellows serve as their own controls for the comparison of pregnancy outcomes after ET before and after the lapse, thus minimizing the impact of individual differences in skill level. Earlier studies have suggested a learning curve for ET requiring a range of 15–100 ETs to gain proficiency (7–10). A crude comparison of the last 50 and 25 ETs immediately before and after the lapse suggests the presence of a learning curve, although our study was not designed to evaluate this hypothesis. A learning curve, if present, may affect our results by increasing pregnancy rates immediately before the lapse and decreasing pregnancy rates immediately after the lapse. When averaged with all transfers, this would result in a greater similarity between the overall pre-lapse and post-lapse pregnancy rates, thus favoring the null hypothesis. Although this remains a possibility, it is difficult to conclude based on comparison of only 25 ETs per fellow, because the small numbers have limited statistical power and may therefore result in spurious findings. This is illustrated by fellow 5, who counterintuitively had the greatest difference between the 25 ETs immediately before and after the lapse, but had gained this level of proficiency after completing only 32 ETs in total during the pre-lapse time period. This issue would be better addressed by including more fellows to increase the pooled number of transfers, thus maintaining the power to adjust for confounders while focusing solely on the ETs immediately before and after the training lapse. 732
Our data would have been further strengthened by a categorization of ‘‘ease of transfer,’’ because this has been shown to affect CPR (5) and may influence the likelihood that a fellow, rather than an attending physician, would complete a given transfer. Unfortunately, this variable is not routinely collected in our institution and was therefore unavailable for analysis. As shown, retained embryos, need to reload embryos because of difficult transfer, and need for staff to take over a transfer are infrequent events and may serve as a proxy to evaluate ease of transfer. Finally, the study design precluded us from commenting on the impact of shorter lapses in clinical care on ET proficiency. A study utilizing learning curve-cumulative summation curves to assess proficiency found lapses of >10 days in embryo transfer to have a negative impact on pregnancy rate (9). This result may be different from our study due to their definition of pregnancy by hCG level rather than clinical intrauterine pregnancy as in our study. It should also be noted that this study included a very unusual patient population in that 40%–70% were the recipients of donor oocytes; this may limit the generalizability of the results to other populations (9). In summary, an 18-month lapse in clinical training appears to have no effect on rates of clinical pregnancy and live birth following ET performed by REI fellows. The findings of this study have implications not only for training programs with an educational structure similar to our own, but also for those that schedule a research block at the end of fellowship, when the lapse in clinical training would fall immediately before a fellow's transition to independent practice. Although our study design precludes us from specifically commenting on the impact of shorter lapses in clinical care on ET proficiency, the absence of skill attrition after an 18month lapse suggests that shorter lapses may also be of limited consequence. The latter situation may be relevant to lower-volume practices where an individual practitioner may not perform ET for a few months, or to academic centers where research commitments may preclude some faculty members from engaging in consistent coverage of the IVF service. The overall absence of skill decay after an 18month lapse in clinical training should be reassuring to training programs attempting to incorporate fellow ET into their curriculum. ET is accepted to be a critical step in the success of IVF cycles, and a concerted effort by REI training programs to educate fellows with hands-on experience should be forthcoming.
REFERENCES 1. 2.
3.
4. 5.
Mains L, Van Voorhis BJ. Optimizing the technique of embryo transfer. Fertil Steril 2010;94:785–90. Hearns-Stokes RM, Miller BT, Scott L, Creuss D, Chakraborty PK, Segars JH. Pregnancy rates after embryo transfer depend on the provider at embryo transfer. Fertil Steril 2000;74:80–6. Abou-Setta AM, Al-Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Hum Reprod 2005;20:3114–21. Buckett WM. A review and meta-analysis of prospective trials comparing different catheters used for embryo transfer. Fertil Steril 2006;85:728–34. Tomas C, Tikkinen K, Tuomivaara L, Tapanainen JS, Martikainen H. The degree of difficulty of embryo transfer is an independent factor for predicting pregnancy. Hum Reprod 2002;17:2632–5. VOL. 103 NO. 3 / MARCH 2015
Fertility and Sterility® 6.
Grygoruk C, Ratomski K, Kolodziejczyk M, Gagan J, Modlinski JA, Gajda B, et al. Fluid dynamics during embryo transfer. Fertil Steril 2011;96:324–7. 7. Dessolle L, Freour T, Barriere P, Jean M, Ravel C, Darai E, et al. How soon can I be proficient in embryo transfer? Lessons from the cumulative summation test for learning curve (LC-CUSUM). Hum Reprod 2010;25:380–6. 8. Shah DK, Missmer SA, Correia KF, Racowsky C, Ginsburg E. Efficacy of intrauterine inseminations as a training modality for performing embryo transfer in reproductive endocrinology and infertility fellowship programs. Fertil Steril 2013;100:386–91. 9. Lopez MJ, Garcia D, Rodriguez A, Colodron M, Vassena R, Vernaeve V. Individualized embryo transfer training: timing and performance. Hum Reprod 2014;29:1432–7. 10. Papageorgiou TC, Hearns-Stokes RM, Leondires MP, Miller BT, Chakraborty P, Cruess D, et al. Training of providers in embryo transfer:
VOL. 103 NO. 3 / MARCH 2015
11.
12. 13.
14. 15.
what is the minimum number of transfers required for proficiency? Hum Reprod 2001;16:1415–9. Neithardt AB, Segars JH, Hennessy S, James AN, McKeeby JL. Embryo afterloading: a refinement in embryo transfer technique that may increase clinical pregnancy. Fertil Steril 2005;83:710–4. Perez RS, Skinner A, Weyhrauch P, Niehaus J, Lathan C, Schwaitzberg SD, et al. Prevention of surgical skill decay. Mil Med 2013;178:76–86. Willis Ross E, Van Sickle Kent R, Peterson Richard M. Impact of non-clinical years on surgery residents' technical skills: evaluation of a technical skills refresher curriculum. Surg Sci 2013;4:131–4. Wittenberger MD, Catherino WH, Armstrong AY. Role of embryo transfer in fellowship training. Fertil Steril 2007;88:1014–5. Bishop L, Brezina PR, Segars J. Training in embryo transfer: how should it be done? Fertil Steril 2013;100:351–2.
733
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
SUPPLEMENTAL TABLE 1 Clinical pregnancy rate per embryo transfer for REI fellows based on the total number of embryo transfers performed immediately before and after an 18-month lapse in clinical training. All ETs
Last 50 ETs
Last 25 ETs
Fellow Pre-lapse Post-lapse Pre-lapse Post-lapse Pre-lapse Post-lapse 1 2 3 4 5
64 65 53 56 59
63 59 57 53 57
66 64 50 52 59a
58 56 54 48 52
68 68 56 56 60
64 56 56 48 40
Note: Results are presented as percent unadjusted mean CPR per embryo transfer. CPR ¼ clinical pregnancy rate; ET ¼ embryo transfer. a Fellow performed only 32 transfers before the lapse. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
733.e1
VOL. 103 NO. 3 / MARCH 2015
Fertility and Sterility®
SUPPLEMENTAL TABLE 2 Quantification of embryo transfer difficulty for REI fellows before and after an 18-month lapse in clinical training. Pre-lapse Fellow 1 2 3 4 5
Post-lapse
No. ETs
Retained
Reloaded
Attending took over
No. ETs
Retained
Reloaded
Attending took over
55 91 104 96 32
1 0 3 2 0
0 1 1 3 0
1 0 0 2 0
80 168 98 135 74
1 0 2 1 0
0 2 1 2 0
1 0 1 1 1
Note: Retained ¼ transfers in which one or more embryos were noted to be retained in the catheter after initial attempt; reloaded ¼ transfers in which embryos were temporarily returned to the embryologist owing to prolonged difficulty in gaining entry into the uterine cavity; attending took over ¼ transfers initiated by a fellow but ultimately completed by an attending physician. Abbreviations as in Supplemental Table 1. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
VOL. 103 NO. 3 / MARCH 2015
733.e2
Objective: To compare rates of clinical pregnancy (CPR) and live birth (LBR) following embryo transfer (ET) performed by reproductive endocrinology and infertility (REI) fellows before and after a prolonged lapse in clinical training due to an 18-month research rotation. Design: Retrospective cohort study. Setting: Not applicable. Patient(s): All women undergoing in vitro fertilization (IVF) and IVF–intracytoplasmic sperm injection (ICSI) cycles with ET performed by REI fellows from August 2003 to July 2012. Intervention: Eighteen-month lapse in clinical training of REI fellows. Main Outcome Measure(s): CPR and LBR before and after the lapse in clinical training were calculated and compared per fellow and as a composite group. Alternating logistic regression models were used to calculate the odds of clinical pregnancy and live birth following transfers performed before and after the lapse in training. Result(s): Unadjusted odds of clinical pregnancy and live birth were similar between the two time periods both for individual fellows and for the composite group. Alternate logistic regression analysis revealed no significant difference in CPR (odds ratio [OR] 0.94, 95% confidence interval [CI] 0.83–1.07) or LBR (OR 1.05, 95% CI 0.94–1.18) after the lapse in training compared with before. Conclusion(s): A research rotation is common in REI fellowship training programs. This prolonged departure from clinical training does not appear to negatively affect pregnancy outcome Use your smartphone following fellow ET. (Fertil SterilÒ 2015;103:728–33. Ó2015 by American Society for Reproto scan this QR code ductive Medicine.) and connect to the Key Words: Embryo transfer, training, lapse, REI, fellows Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/kresowikj-lapse-embryo-transfer-rei-fellows/
E
mbryo transfer (ET) is a nuanced part of the in vitro fertilization (IVF) process, where success is affected by both provider technique and experience (1, 2). Factors thought
to be associated with implantation following ET include the use of soft versus rigid transfer catheters and provider perception of the ease of transfer (3–5). Avoiding negative
Received September 8, 2014; revised and accepted December 5, 2014; published online December 31, 2014. J.K. has nothing to disclose. A.S. has nothing to disclose. E.H.D. has nothing to disclose. D.K.S. has nothing to disclose. Presented at the 69th American Society for Reproductive Medicine Annual Meeting, Boston, Massachusetts, October 12-17, 2013. Reprint requests: Divya K. Shah, M.D., Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, University of Iowa Hospitals and Clinics, 31324 PFP, 200 Hawkins Drive, Iowa City, Iowa 52242 (E-mail: [email protected]). Fertility and Sterility® Vol. 103, No. 3, March 2015 0015-0282/$36.00 Copyright ©2015 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2014.12.102 728
discussion forum for this article now.*
* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.
pressure when removing the catheter as well as maintaining a controlled velocity when depressing the plunger also have been associated with success (6). It is expected that reproductive endocrinology and infertility (REI) fellowship programs ensure adequate training and competence in ET, although the best training modality and the time required to accomplish this goal are not well understood. The number of ETs required to reach proficiency has been shown to vary greatly between studies as well as VOL. 103 NO. 3 / MARCH 2015
Fertility and Sterility® between individual learners, with estimates ranging from 15 to 100 ETs (7–10). In addition, the implementation of a formalized training program using intrauterine insemination did not appear to improve a learner's time to proficiency compared with those who did not undergo similar training (8). A further complicating factor in the ET training process for many REI fellowship programs in the United States is the requirement for 18 months of dedicated research time. This ‘‘research block’’ is frequently scheduled during the 2nd year of fellowship and removes fellows from participation in patient care. The effect of this prolonged lapse in clinical training on success rates after ET remains unknown. We sought to evaluate the impact of a lapse in training on clinical pregnancy rate (CPR) and live birth rate (LBR) following ET performed by REI fellows.
MATERIALS AND METHODS Study Population Data were retrospectively queried from the University of Iowa's IVF database from August 2003 till July 2012. A total of five fellows started and completed their training during the study time period. All IVF and IVF–intracytoplasmic sperm injection (ICSI) cycles with ET performed by REI fellows during this time period were included. It is institutional practice for 1st-year fellows to observe ETs for 1 month and then perform all subsequent ETs when covering the IVF service. The database does not reliably track the number of ETs initiated by fellows but ultimately taken over by attending physicians, so these transfers were excluded from analysis. Embryo transfers were performed in the IVF procedure room without analgesia. The cervix was gently cleansed with warm saline solution and any excess cervical mucus was aspirated. A mock ET was first performed with the use of a Wallace Trial Catheter (Smiths Medical) to ensure passage through the internal cervical os. If passage of the catheter required significant manipulation, the ‘‘afterload’’ method with retention of the outer sheath was used at the discretion of the transferring provider (11). If a fellow was unable to successfully pass the mock transfer catheter after a reasonable attempt, the transfer was performed by the supervising physician. Once the mock ET was complete, embryos were loaded into either a Wallace Classic Embryo Replacement Catheter (August 2003–June 2004) or Wallace Sure View Embryo Replacement Catheter (June 2004 and subsequently) for transfer. Transfer catheters were loaded with a solid column of culture medium, followed by 5 mL air, 20 mL medium containing the embryo(s), and a final 5 mL air. All ETs (including mock ETs) were performed under transabdominal ultrasound guidance provided either by the supervising faculty physician or a trained sonographer. All patients signed consents for data collection and quality assurance analysis at the start of treatment. The University of Iowa Institutional Review Board deemed the project to be exempt from further review because of the noninterventional and quality assurance nature of the study. VOL. 103 NO. 3 / MARCH 2015
Exposure and Outcome Variables The exposure was an 18-month lapse in clinical training while on a research rotation. Fellow ETs were divided into two groups: those performed before (pre-lapse) and after (post-lapse) the 18-month departure from clinical training. The primary outcome was CPR per ET (defined as the presence of an intrauterine gestational sac on ultrasound). The secondary outcome was LBR per ET (defined as the birth of an infant at or beyond 20 weeks gestation).
Potential Confounding Variables Patient demographic variables, including female age, body mass index (BMI), race, antral follicle count (AFC), and infertility diagnosis, were noted in the medical record at the time of cycle initiation. Race was classified as white, African-American, Asian, multiracial, or other by the nurse or providing physician. Categories of infertility diagnosis included male factor, tubal factor, ovulatory dysfunction, endometriosis, uterine factor, and diminished ovarian reserve. IVF cycle characteristics, including cycle number (1, 2, 3, R4), use of donor oocytes, use of ICSI, total gonadotropin dosage (IU of FSH or hMG), use of fresh versus frozen embryos, number of embryos transferred (1, 2, R3), and day 3 versus day 5 ET, were recorded in the IVF database on a per-cycle basis. Additional variables reported include numbers of ETs complicated by either retention of one or more embryos in the catheter after initial transfer attempt or the need to temporarily return the catheter to the embryologist because of prolonged difficulty in gaining entry into the uterine cavity.
Statistical Analysis Descriptive statistics were used to assess the demographics of the study population. CPR and LBR per ET performed after the training lapse were compared with CPR and LBR per ET performed before the training lapse by means of the Cochran-Mantel Haenszel method, controlling for the fellow that performed the ET. This same comparison was made for each individual fellow with the use of the Pearson chisquare test, generating unadjusted odds ratios (ORs). A generalized linear model was used to test for the effect of training lapse on CPR after adjusting for potential confounding variables. This model was fitted by the alternating logistic regression algorithm to account for the association of transfer outcome in subjects treated by the same fellow as well for repeated ETs in the same subject. Female age, cycle number, use of donor eggs, fresh versus frozen cycle, and day 3 versus day 5 transfer were included a priori as covariates in the model; other potential confounders were examined but did not change effect estimates by >10% and were therefore not included in the final model.
RESULTS Five fellows completed their training from August 2003 to July 2012. A total of 793 patients and 933 ETs were included in the study. Demographic and IVF cycle characteristics of the study population are listed in Table 1. Patient age, ethnicity, 729
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
TABLE 1 Patient and cycle characteristics for embryo transfers performed by reproductive endocrinology and infertility fellows before and after an 18-mo lapse in clinical training. Characteristic
Pre-lapse group (n [ 378)
Patient Age (y), mean (SD) 33.7 (5.0) White 315 (91) BMI (kg/m2) <20 28 (7) 20–24.9 168 (44) 25–29.9 93 (25) 30–39.9 69 (18) R40 20 (5) Antral follicle count, 19 (13–30) median (IQR) Infertility diagnosis Male factor 157 (42) Tubal factor 84 (22) Ovulatory factor 66 (18) Endometriosis 45 (12) Uterine factor 21 (6) Diminished ovarian 36 (10) reserve Cycle Cycle number 1 174 (46) 2 101 (27) 3 55 (15) R4 48 (13) Use of donor oocytes 34 (9) Use of ICSI 245 (65) Total gonadotropin 24 (0–39) dose (IU) Fresh embryo transfer 259 (69) Transfer day 3 176 (47) 5 201 (53) No. of embryos transferred 1 112 (30) 2 229 (61) R3 37 (10)
Post-lapse group (n [ 555) 33.8 (4.8) 461 (91) 37 (7) 225 (41) 122 (22) 138 (25) 32 (6) 21 (13–30) 224 (40) 99 (18) 98 (18) 65 (12) 50 (9) 50 (9)
227 (41) 143 (26) 87 (16) 98 (18) 41 (7) 333 (60) 22 (0–36) 365 (66) 221 (40) 333 (60) 218 (39) 298 (53) 41 (7)
Note: Values are presented as n (%) except as noted. BMI ¼ body mass index; ICSI ¼ intracytoplasmic sperm injection. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
infertility diagnosis, use of donor oocytes, fresh cycle, use of ICSI, AFC, and gonadotropin dosage were similar between the patients undergoing ET before and after the lapse in clinical training. Patient BMI tended to be higher in the post-lapse group compared with the pre-lapse group. The post-lapse group also included a greater proportion of patients who underwent transfer of one embryo on day 5, coincident with our program's increased use of single-embryo transfer in accordance with practice guidelines from the American Society of Reproductive Medicine. The mean CPR for each fellow before and after the lapse in training are presented in Table 2; the CPR ranged from 53% to 65%. No significant difference in CPR was observed for any fellow when comparing the two time frames. Although this study was designed to have each fellow serve as their own control, the mean CPR for all ETs performed by attending physicians during the corresponding time period are included in the table as a point of reference. Figure 1 730
shows the unadjusted OR of clinical pregnancy and 95% confidence intervals (CIs) for all ETs performed during the post-lapse time frame relative to all ETs performed during the pre-lapse time frame for each individual fellow; no significant differences were seen. The common OR comparing CPR during the pre- and post-lapse time periods, pooling data from all fellows, remained nonsignificant at 0.92 (95% CI 0.70–1.21). In an isolated crude comparison of the final 25 ETs performed pre-lapse and the first 25 ETs performed post-lapse, however, a trend toward higher pregnancy rates pre-lapse and lower pregnancy rates post-lapse was noted (Supplemental Table 1, available online at www.fertstert.org). The adjusted ORs and 95% CIs for both clinical pregnancy and live birth are shown in Figure 2. Adjusting for age, cycle number, use of donor eggs, fresh versus frozen cycle, and day of ET, and accounting for repeated transfers in the same patient, no difference in rates of clinical pregnancy (OR 0.93, 95% CI 0.79–1.09) or live birth (OR 0.94, 95% CI 0.83–1.07) were noted between ETs performed by REI fellows after compared with before the lapse in clinical training. Supplemental Table 2 (available online at www.fertstert.org) reports the percentage of ETs that included retained or reloaded embryos for each fellow, before and after the lapse, as well as an approximate number of ETs that were initiated by a fellow but ultimately taken over by an attending physician. The latter variable was not reliably collected in the dataset, as stated in the Materials and Methods section, and therefore represents an approximation for reference. These difficult ETs were rare, representing fewer than 3% of all ETs performed by any given fellow.
DISCUSSION The results of this study indicate that an 18-month lapse in clinical training had no impact on CPR or LBR following ET performed by REI fellows. The impact of this finding is of particular importance as REI fellowship programs in the United States reevaluate how best to train their fellows in the critical procedure of ET. Skill decay, the loss of a trained or acquired skill following periods of nonuse, is a well described phenomenon. A meta-analysis of surgical skill decay and retention literature found that after 365 days of nonuse or nonpractice, the average participant's performance was reduced by almost a full standard deviation (12). Physical, natural, or speed-based tasks were less susceptible to skill decay than cognitive, artificial, or accuracy-based tasks. Additionally ‘‘overlearning,’’ or practicing a skill past the level of proficiency, is also protective against skill decay. An article examining general surgery residents that left clinical training for 2 years to conduct a research rotation found that residents and faculty perceived nonclinical years to have a slightly deleterious effect on technical skills. However, when their skills were evaluated in a simulation center, all residents performed at proficiency for basic and intermediary knot-tying and suturing on return from their nonclinical years (13). A perception of skill decay may affect the willingness of REI fellowship programs to permit fellows to perform VOL. 103 NO. 3 / MARCH 2015
Fertility and Sterility®
TABLE 2 Clinical pregnancy rates per embryo transfer for REI fellows before and after an 18-mo lapse in clinical training. Pre-lapse Fellow (individual) Fellow 1 2 3 4 5
Post-lapse
Attendings (aggregate)
Fellow (individual)
Attendings (aggregate)
No. ETs
% CPR
No. ETs
% CPR
No. ETs
% CPR
No. ETs
% CPR
55 91 104 96 32
64 65 53 56 59
161 114 111 493 191
56 52 69 61 56
80 168 98 135 74
63 59 57 53 57
170 323 79 151 184
57 55 62 59 59
Note: Results are presented as unadjusted mean CPR per embryo transfer. CPRs are shown for each individual fellow for the pre- and post-lapse time periods. The mean CPRs of all ETs performed by attending physicians during the same time periods are included for reference. CPR ¼ clinical pregnancy rate; ET ¼ embryo transfer. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
ET. A 2007 survey of REI fellowship programs revealed that 44% of fellows graduate without having performed ET during training (14). The Society for Reproductive Endocrinology and Infertility conducted a survey in 2012 of 64 current and former REI fellows and found that 75% of respondents anticipated performing fewer than 20 ETs in their fellowship, with 46.9% performing none (15). Perceived barriers to fellow participation in ET include patient preference, presumed negative effect on pregnancy rates, and competition with private programs without trainees (15). It is also plausible that the prospect of a prolonged absence from clinical practice may increase faculty discomfort with fellows performing ET owing to concerns regarding skill attrition. Our data suggest that an 18-month lapse in clinical training
does not appear to be detrimental to fellow performance of ET. It may be that the physical rather than cognitive nature of the ET procedure makes its less susceptible to skill decay, and that maintenance of ET skill is similar to the aforementioned ability of general surgery residents to maintain proficiency at basic suturing after a 2 year lapse in clinical training (13). Our REI division is able to effectively address this study question because institutional practice dictates fellow participation in all ETs. Our institution is also uniquely positioned to evaluate the effect of a prolonged lapse in clinical training because our fellowship is designed with a clearly defined 18month research block from month 9 to month 27 of REI fellowship. REI fellows are completely removed from clinical
FIGURE 1
Unadjusted odds of clinical pregnancy per embryo transfer for individual fellows after an 18-month lapse in clinical training. Blue dots indicate the unadjusted odds ratio (OR) of clinical pregnancy, comparing the transfers performed in the post-lapse time period with those in the pre-lapse time period for each fellow; the brown lines represent 95% confidence intervals. The red vertical line denotes an OR of 1.0. The common OR (95% confidence interval) is composed of pooled data from all fellows and remained nonsignificant. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
VOL. 103 NO. 3 / MARCH 2015
731
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
FIGURE 2
Adjusted odds of clinical pregnancy and live birth per embryo transfer after an 18-month lapse in clinical training. Blue dots indicate the adjusted odds ratio of clinical pregnancy or live birth for all embryo transfers performed by fellows in the post-lapse time period compared with those in the pre-lapse time period; the brown lines represent 95% confidence intervals. The red vertical line denotes an odds ratio of 1.0. Multivariate analysis adjusting for age, cycle number, use of donor eggs, fresh versus frozen cycle, and day of embryo transfer and accounting for repeated transfers in the same patient noted no difference in the odds of clinical pregnancy or live birth for fellow embryo transfers completed after an 18-month lapse in clinical training. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
practice during this block with no call responsibilities, thereby protecting our analyses from confounding due to occasional performance of ET during the research rotation. A final strength is our paired study design in which individual fellows serve as their own controls for the comparison of pregnancy outcomes after ET before and after the lapse, thus minimizing the impact of individual differences in skill level. Earlier studies have suggested a learning curve for ET requiring a range of 15–100 ETs to gain proficiency (7–10). A crude comparison of the last 50 and 25 ETs immediately before and after the lapse suggests the presence of a learning curve, although our study was not designed to evaluate this hypothesis. A learning curve, if present, may affect our results by increasing pregnancy rates immediately before the lapse and decreasing pregnancy rates immediately after the lapse. When averaged with all transfers, this would result in a greater similarity between the overall pre-lapse and post-lapse pregnancy rates, thus favoring the null hypothesis. Although this remains a possibility, it is difficult to conclude based on comparison of only 25 ETs per fellow, because the small numbers have limited statistical power and may therefore result in spurious findings. This is illustrated by fellow 5, who counterintuitively had the greatest difference between the 25 ETs immediately before and after the lapse, but had gained this level of proficiency after completing only 32 ETs in total during the pre-lapse time period. This issue would be better addressed by including more fellows to increase the pooled number of transfers, thus maintaining the power to adjust for confounders while focusing solely on the ETs immediately before and after the training lapse. 732
Our data would have been further strengthened by a categorization of ‘‘ease of transfer,’’ because this has been shown to affect CPR (5) and may influence the likelihood that a fellow, rather than an attending physician, would complete a given transfer. Unfortunately, this variable is not routinely collected in our institution and was therefore unavailable for analysis. As shown, retained embryos, need to reload embryos because of difficult transfer, and need for staff to take over a transfer are infrequent events and may serve as a proxy to evaluate ease of transfer. Finally, the study design precluded us from commenting on the impact of shorter lapses in clinical care on ET proficiency. A study utilizing learning curve-cumulative summation curves to assess proficiency found lapses of >10 days in embryo transfer to have a negative impact on pregnancy rate (9). This result may be different from our study due to their definition of pregnancy by hCG level rather than clinical intrauterine pregnancy as in our study. It should also be noted that this study included a very unusual patient population in that 40%–70% were the recipients of donor oocytes; this may limit the generalizability of the results to other populations (9). In summary, an 18-month lapse in clinical training appears to have no effect on rates of clinical pregnancy and live birth following ET performed by REI fellows. The findings of this study have implications not only for training programs with an educational structure similar to our own, but also for those that schedule a research block at the end of fellowship, when the lapse in clinical training would fall immediately before a fellow's transition to independent practice. Although our study design precludes us from specifically commenting on the impact of shorter lapses in clinical care on ET proficiency, the absence of skill attrition after an 18month lapse suggests that shorter lapses may also be of limited consequence. The latter situation may be relevant to lower-volume practices where an individual practitioner may not perform ET for a few months, or to academic centers where research commitments may preclude some faculty members from engaging in consistent coverage of the IVF service. The overall absence of skill decay after an 18month lapse in clinical training should be reassuring to training programs attempting to incorporate fellow ET into their curriculum. ET is accepted to be a critical step in the success of IVF cycles, and a concerted effort by REI training programs to educate fellows with hands-on experience should be forthcoming.
REFERENCES 1. 2.
3.
4. 5.
Mains L, Van Voorhis BJ. Optimizing the technique of embryo transfer. Fertil Steril 2010;94:785–90. Hearns-Stokes RM, Miller BT, Scott L, Creuss D, Chakraborty PK, Segars JH. Pregnancy rates after embryo transfer depend on the provider at embryo transfer. Fertil Steril 2000;74:80–6. Abou-Setta AM, Al-Inany HG, Mansour RT, Serour GI, Aboulghar MA. Soft versus firm embryo transfer catheters for assisted reproduction: a systematic review and meta-analysis. Hum Reprod 2005;20:3114–21. Buckett WM. A review and meta-analysis of prospective trials comparing different catheters used for embryo transfer. Fertil Steril 2006;85:728–34. Tomas C, Tikkinen K, Tuomivaara L, Tapanainen JS, Martikainen H. The degree of difficulty of embryo transfer is an independent factor for predicting pregnancy. Hum Reprod 2002;17:2632–5. VOL. 103 NO. 3 / MARCH 2015
Fertility and Sterility® 6.
Grygoruk C, Ratomski K, Kolodziejczyk M, Gagan J, Modlinski JA, Gajda B, et al. Fluid dynamics during embryo transfer. Fertil Steril 2011;96:324–7. 7. Dessolle L, Freour T, Barriere P, Jean M, Ravel C, Darai E, et al. How soon can I be proficient in embryo transfer? Lessons from the cumulative summation test for learning curve (LC-CUSUM). Hum Reprod 2010;25:380–6. 8. Shah DK, Missmer SA, Correia KF, Racowsky C, Ginsburg E. Efficacy of intrauterine inseminations as a training modality for performing embryo transfer in reproductive endocrinology and infertility fellowship programs. Fertil Steril 2013;100:386–91. 9. Lopez MJ, Garcia D, Rodriguez A, Colodron M, Vassena R, Vernaeve V. Individualized embryo transfer training: timing and performance. Hum Reprod 2014;29:1432–7. 10. Papageorgiou TC, Hearns-Stokes RM, Leondires MP, Miller BT, Chakraborty P, Cruess D, et al. Training of providers in embryo transfer:
VOL. 103 NO. 3 / MARCH 2015
11.
12. 13.
14. 15.
what is the minimum number of transfers required for proficiency? Hum Reprod 2001;16:1415–9. Neithardt AB, Segars JH, Hennessy S, James AN, McKeeby JL. Embryo afterloading: a refinement in embryo transfer technique that may increase clinical pregnancy. Fertil Steril 2005;83:710–4. Perez RS, Skinner A, Weyhrauch P, Niehaus J, Lathan C, Schwaitzberg SD, et al. Prevention of surgical skill decay. Mil Med 2013;178:76–86. Willis Ross E, Van Sickle Kent R, Peterson Richard M. Impact of non-clinical years on surgery residents' technical skills: evaluation of a technical skills refresher curriculum. Surg Sci 2013;4:131–4. Wittenberger MD, Catherino WH, Armstrong AY. Role of embryo transfer in fellowship training. Fertil Steril 2007;88:1014–5. Bishop L, Brezina PR, Segars J. Training in embryo transfer: how should it be done? Fertil Steril 2013;100:351–2.
733
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
SUPPLEMENTAL TABLE 1 Clinical pregnancy rate per embryo transfer for REI fellows based on the total number of embryo transfers performed immediately before and after an 18-month lapse in clinical training. All ETs
Last 50 ETs
Last 25 ETs
Fellow Pre-lapse Post-lapse Pre-lapse Post-lapse Pre-lapse Post-lapse 1 2 3 4 5
64 65 53 56 59
63 59 57 53 57
66 64 50 52 59a
58 56 54 48 52
68 68 56 56 60
64 56 56 48 40
Note: Results are presented as percent unadjusted mean CPR per embryo transfer. CPR ¼ clinical pregnancy rate; ET ¼ embryo transfer. a Fellow performed only 32 transfers before the lapse. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
733.e1
VOL. 103 NO. 3 / MARCH 2015
Fertility and Sterility®
SUPPLEMENTAL TABLE 2 Quantification of embryo transfer difficulty for REI fellows before and after an 18-month lapse in clinical training. Pre-lapse Fellow 1 2 3 4 5
Post-lapse
No. ETs
Retained
Reloaded
Attending took over
No. ETs
Retained
Reloaded
Attending took over
55 91 104 96 32
1 0 3 2 0
0 1 1 3 0
1 0 0 2 0
80 168 98 135 74
1 0 2 1 0
0 2 1 2 0
1 0 1 1 1
Note: Retained ¼ transfers in which one or more embryos were noted to be retained in the catheter after initial attempt; reloaded ¼ transfers in which embryos were temporarily returned to the embryologist owing to prolonged difficulty in gaining entry into the uterine cavity; attending took over ¼ transfers initiated by a fellow but ultimately completed by an attending physician. Abbreviations as in Supplemental Table 1. Kresowik. Lapse in embryo transfer training. Fertil Steril 2015.
VOL. 103 NO. 3 / MARCH 2015
733.e2