Arch Gynecol Obstet (2012) 285:1753–1757

DOI 10.1007/s00404-012-2238-1

REPRODUCTIVE MEDICINE

Appropriate interval between endometrial polyp resection and the proceeding IVF start

Ozlem Gun Eryilmaz · Cavidan Gulerman · Esma Sarikaya · Huseyin Yesilyurt · Fatih Karsli · Nedim Cicek

Received: 15 September 2011 / Accepted: 19 January 2012 / Published online: 1 February 2012© Springer-Verlag 2012

AbstractPurpose Endometrial polyp is the commenest pathologyamong the structural uterine abnormalities. In this study wecompared the IVF outcomes of patients who underwenthysteroscopic endometrial polyp excision with respect tothe time interval between the polyp resection and the subse-quent IVF cycle.Materials and methods A total of 60 patients weredivided into two groups according to the time intervalbetween the hysteroscopic polyp resection and the start ofthe following IVF cycle. Of these, 29 patients had the IVFprocedures <6 months after the polyp resection, whereas in31 patients comprising the comparison group, IVF inter-ventions were done ¸6 months after the resection. The IVFoutcomes were compared.Results DiVerences in the mean number of retrievedoocytes, metaphase II (MII) oocytes and transferredembryos, ratios of MII oocytes, and G1, G2 and G3 embryorates were insigniWcant between the groups (p > 0.05).The fertilization, implantation, and clinical pregnancy rateswere also similar between the groups (p > 0.05).

Conclusion IVF outcomes seem to be unrelated to thetime interval between the hysteroscopic polyp resection andthe initiation of the IVF; the success rates may not be supe-rior if the treatment is started in the Wrst few months post-operatively.

Keywords Endometrial polyp · IVF · Hysteroscopy

Introduction

Embryo quality and uterine receptivity are the two mostimportant factors for a successful implantation in in vitrofertilization (IVF) cycles. Uterine receptivity is the leastunderstood and clariWed point of the two. Hormonal imbal-ance between the uterus and the embryo may be a factor forimplantation failure, while structural uterine abnormalitiesmay be another [1]. While it is not possible to assess themicroenvironment of the uterus, an impaired intrauterinestructure can be evaluated easily by hysteroscopy [2]. Suchstructural abnormalities include congenital fusion defects,leiomyomas, adhesions, and endometrial polyps, the latterof which is the commonest pathology, seen at a rate of1.4–26% in IVF patients [3–5]. Treatment is preferably viahysteroscopic resection, since an increased pregnancy losswas reported in IVF patients continuing with an unresectedpolyp [6]. The positive eVects of hysteroscopic polypec-tomy have been well documented [7, 8]. It is clear thatremoval of an endometrial polyp before the IVF cycle ispreferable, but the most appropriate interval between thehysteroscopic polyp resection and the start of the proceed-ing IVF cycle has not yet been discussed. It remains to bedetermined whether the IVF cycle may be started at thefollowing menstrual period or a few months later, and acomparative study regarding this point is lacking.

O. G. Eryilmaz (&) · C. Gulerman · E. Sarikaya · H. Yesilyurt · F. Karsli · N. CicekDepartment of Reproductive Endocrinology, Zekai Tahir Burak Women’s Education and Research Hospital, Devlet Mah. Subay Loj. no: 19/11 Yenisehir, Ankara, Turkeye-mail: drozlemgun@yahoo.com

C. Gulermane-mail: cgulerman@gmail.com

H. Yesilyurte-mail: dryesilyurt@gmail.com

F. Karslie-mail: fatihkarsli46@gmail.com

N. Ciceke-mail: mncicek@gmail.com

123

1754 Arch Gynecol Obstet (2012) 285:1753–1757

In this study, we aimed to compare the IVF outcomes ofpatients who underwent hysteroscopic endometrial polypexcision with respect to the time interval between the polypresection and the subsequent IVF cycle.

Materials and methods

This retrospective study was conducted between September2009 and August 2011 in Zekai Tahir Burak Women’sEducation and Research Hospital, IVF Department, Ankara,Turkey. All of the patients seen in the IVF Department areroutinely evaluated by oYce hysteroscopy prior to the IVFcycle [2]. A total of 93 patients with a hysteroscopic Wndingof endometrial polyp were diagnosed retrospectively. Theyunderwent polyp resection before the following IVF treat-ment. Of these, 33 patients were excluded as follows: failureto return to our clinic for the following IVF procedurealthough advised to do so (n = 24), no assessment of preg-nancy outcomes (n = 4), existence of multiple polyps (n = 2),and cancellation of IVF cycles (n = 3). The remaining 60patients were included into the study and were divided intotwo groups according to the time interval between the hyste-roscopic polyp resection and the start of the followingIVF cycle. Of these, 29 patients had the IVF procedures<6 months after the polyp resection (Group A, n = 29),whereas in 31 patients comprising the comparison group,IVF interventions were done ¸6 months after the resection(Group B, n = 31).

OYce hysteroscopy was used in the Wrst analysis of theuterine cavity. After diagnosis of an endometrial polyp, allof the patients underwent polyp resection by operativehysteroscopy under general anesthesia. The operative tech-niques included resection at the base for polyps measuring·5 mm and slicing from the free part to the base of thepolyp for those >5 mm (monopolare electrode, Karl Storz).

Diameter of the polyp, duration of induction, requiredtotal gonadotropin dose, number of follicles with a diame-ter ¸16 mm, peak E2 level, endometrial thickness on theday of human chorionic gonadotropin (hCG) injection, fer-tilization failure rate, number of oocytes retrieved, numberof metaphase II (MII) oocytes, grade I (GI), grade II (GII),and grade III (GIII) embryo ratios, fertilization, implanta-tion, and the clinical pregnancy rates were comparedbetween the two groups. The primary outcome was theclinical pregnancy rate.

Stimulation protocol

All patients received a standard gonadotropin-releasinghormone (GnRH) agonist (leuprolide acetate) regimenstarting on Day 21 of a spontaneous menstrual cycle.Recombinant follicle-stimulating hormone stimulation

(150 IU, recFSH) was initiated on the third day of subse-quent withdrawal bleeding. Further recFSH doses weredetermined according to the standard criterion for follicularmaturation assessed by ultrasound and serum E2 measure-ments. 250 �g recombinant hCG (r-hCG) (Ovitrelle, MerckSerono, Italy) was administered when at least three follicleshad reached a diameter of 16 mm. Oocyte retrieval was per-formed 36 h later under transvaginal ultrasound guidanceand intravenous (i.v.) sedation. All patients received lutealphase support of 90 mg/day of vaginally administered pro-gesterone (Crinone gel, Merck Serono, Central Pharma,UK) starting from the day after oocyte retrieval. Embryoquality was assessed before embryo transfer, and a maxi-mum of three embryos were transferred to each patient onthe third to Wfth day of oocyte retrieval using an EdwardsWallace catheter (Simcare Ltd, UK) under transabdominalsonographic guidance.

Luteal phase support was started on the day of oocytepick-up with daily vaginal administration of 90 mg proges-terone (Crinone gel, Merck Serono, Central Pharma, UK).Biochemical pregnancy was established when serum �-hCGlevel was ¸20 IU/L on the 12th day after embryo transfer,and clinical pregnancy was deWned as the presence of agestational sac on ultrasound performed at 6 weeks afterembryo transfer.

Outcome variable and statistical analysis

When endometrial polyp prevalence of 1.4% among theIVF patients [6] and the clinical pregnancy rate of 33.3% inIVF cycles after the polyp resection [9] were taken as thereference, with a 95% conWdence, an alpha of 0.05, and95% power, the sample size was calculated as 28 cases and28 controls; 29 cases and 31 controls were evaluated in thepresent study.

Statistical analysis was carried out using the StatisticalPackage for the Social Sciences software (SPSS, version11.5). Data were presented as either median (range) ormean § SD, 95% conWdence intervals (95%CI) as appro-priate. All variables were tested for normal distributionwith Kolmogorov–Smirnov test, histogram, and p–p plots.Variables were compared with either independent samplest-test or Mann–Whitney U test depending on the normalityof the data. All categorical variables were compared withPearson Chi-square and Fisher’s exact tests. p value <0.05was considered as statistically signiWcant.

Results

Related to the time interval between the hysteroscopicpolyp resection and the clinical pregnancy rates of the IVFtreatment within 6 months (Group A) and after sixth

123

Arch Gynecol Obstet (2012) 285:1753–1757 1755

months (Group B) of the endometrial polyp resection, thecomparison was insigniWcant, 20.5% (95%CI: 0.1–0.4)versus 20.0% (95%CI: 0.1–0.5), respectively (p > 0.05).Figure 1 presented the time intervals (1–17 months)between the hysteroscopic polyp resection and the clinicalpregnancy rates.

The demographic data of the women in Group A (IVFtreatment <6 months after hysteroscopic polyp resection)and Group B (IVF treatment ¸6 months after hysteroscopicpolyp resection) are summarized in Table 1. The durationbetween the polyp resection and the beginning of the IVFtreatment was 2.92 § 1.36 (95%CI: 2.39–3.46) months and

7.9 § 2.6 (95%CI: 6.8–8.9) months for Group A and GroupB, respectively (p = 0.0001). The diameters of the polypswere 13.1 § 6.4 (95%CI: 10.6–15.6) mm and 14.7 § 5.9(95%CI: 12.4–17.0) mm for group A and group B, respec-tively (p > 0.05). There were no signiWcant diVerencesbetween the two groups with respect to mean age, bodymass index (BMI), the duration of infertility, total days ofgonadotropin administration, the required total gonadotro-pin dose, peak 17� E2 levels, and the endometrial thicknesson the day of hCG administration (p > 0.05 for all).

Table 2 presented the clinical outcomes of the patients.The mean number of follicles with a diameter ¸16 mm wassigniWcantly diVerent between the IVF treatment within6 months (Group A) and after the sixth months (Group B)of the endometrial polyp resection (4.3 § 2.4, 95%CI: 3.3–5and 5.8 § 2.6, 95%CI: 4.7–6.7; respectively, p < 0.05).DiVerences in the mean number of retrieved oocytes, MIIoocytes, and transferred embryos, G1, G2 and G3 embryorates were insigniWcant between the groups (p > 0.05 forall). The fertilization and the implantation rates were alsosimilar between the groups: 41.3%, 95%CI: 2.3–5.1 versus43.1%, 95%CI: 2.7–5.1 and 20.5%, 95%CI: 0.1–0.4 versus22.5%, 95%CI: 0.1–0.4, respectively (p > 0.05 for all).

Discussion

In determining uterine receptivity, the endometrium anduterine cavity are two important factors that are most suc-cessfully evaluated by hysteroscopy [10, 11]. Endometrialpolyp is the most common pathological Wnding during ahysteroscopic evaluation prior to an IVF attempt. The inci-dence in this population of patients ranges between 1.4and 26% [3–5, 12]. Existence of an endometrial polyp isconsidered to be an uneventful intrauterine condition,and patients are recommended to undergo hysteroscopic

Fig. 1 Time intervals between the hysteroscopic polyp resection andthe clinical pregnancy rates

Time interval (months)

17,0

0

12,0

0

10,0

0

8,00

7,00

6,00

5,00

4,00

3,00

2,00

1,00

12

10

8

6

4

2

0

Pregnancy

Present

Absent

Table 1 Characteristics and outcomes of patients with IVF treatment within 6 months (Group A) or 6 months (Group B) after the endometrialpolyp resection

Values are expressed as mean § SD and 95% conWdence intervals (95%CI)

NS not signiWcant

Group A (n = 29) Group B (n = 31) p value

Age (years) 31.8 § 5.2 (95%CI: 29.8–34.0) 33.0 § 4.6 (95%CI: 30.7–34.5) NS

BMI (kg/m²) 25.4 § 4.6 (95%CI: 23.5–27.2) 24.2 § 4.6 (95%CI: 22.4–25.9) NS

Infertility duration (years) 9.6 § 6.0 (95%CI: 7.2–12.0) 10.3 § 5.7 (95%CI: 7.9–12.4) NS

Interval between the polyp resection and IVF procedure (months)

2.9 § 1.3 (95%CI: 2.3–3.4) 7.9 § 2.6 (95%CI: 6.8–8.9) 0.0001

Polyp daimeter (mm) 13.1 § 6.4 (95%CI: 10.6–15.6) 14.7 § 5.9 (95%CI: 12.4–17.0) NS

Duration of stimulation (days) 10.0 § 2.1 (95%CI: 9.1–10.8) 9.7 § 1.9 (95%CI: 9.0–10.5) NS

Total gonadotropin used (IU) 2,533.8 § 1,570.4 (95%CI: 1,912.6–3,155.1) 2,244.9 § 1,444.3 (95%CI: 1,663.2–2,768.7) NS

E2 level on hCG day (pg/mL) 2,957.2 § 3,726.7 (95%CI: 1,483.0–4,431.5) 2,358.3 § 1,011.5 (95%CI: 1,980.5–2,750.9) NS

Endometrium thickness on hCG day 9.6 § 1.7 (95%CI: 8.9–10.2) 10.0 § 1.8 (95%CI: 9.3–10.8) NS

123

1756 Arch Gynecol Obstet (2012) 285:1753–1757

polypectomy before an IVF cycle [13]. In a randomizedcontrolled trial by Pérez-Medina et al. [13], endometrialpolyp resection was recommended because of the increasedpregnancy success observed among patients who under-went the polyp resection procedure prior to intrauterineinsemination.

The exact timing for the IVF attempt after resection ofan endometrial polyp is not a well-studied point. The ques-tion of whether the interval between polyp resection and theIVF attempt has any importance with respect to the preg-nancy rate was the main subject of our study, and to ourknowledge, this is the Wrst time this issue has beendiscussed.

In our study, pregnancy outcomes were found to beunrelated to the time interval between the hysteroscopicpolyp resection and the subsequent IVF trial. Irrespective ofwhether the IVF treatment was started within 6 months ofthe polyp resection or after more than 6 months, implanta-tion and clinical pregnancy rates were similar. These resultsshowed that a corrected intrauterine environment did notnecessitate a speciWc time interval for proper endometrialreceptivity. Lass et al. [6] assessed the pregnancy outcomesof patients after endometrial polyp resection as 22.4%1 month after the resection, in which the IVF was under-taken with the frozen embryos of the previous cycle. Thepregnancy rate of our study in the time interval of less than6 months, which also included the one-month period, wassimilar to that of Lass’s study, at 20.5%. Additionally, thepregnancy outcomes of the two diVerent time intervals(IVF within 6 months or 6 months or more following resec-tion) did not diVer in the subsequent IVF trial. It was shownthat recovery of the intrauterine environment from a hyste-roscopic trauma may not require a particular time period toestablish the proper endometrial receptivity.

Al-Jefout et al. [14] studied the cellular basis of theintrauterine microenvironment in the presence of an endo-metrial polyp. He found that the number of mast cellswere about sevenfold higher in the uterine cavity with apolyp compared with the normal uterine cavity. Inagakiet al. [15] also supported these Wndings with their resultsof increased cytokine levels in the uterine cavity with apolyp. These data suggest that the disturbed receptivity ofan endometrium with a polyp may be explained via theseinXammatory mediators and likely allergic basis. Thus,resection of the polyp may resolve these problems, andthe restoration of the intracavitary microenvironment mayincrease the pregnancy success. Similar pregnancy resultsobserved among patients who underwent IVF at varioustime points after endometrial polyp resection suggest thatthe endometrium may recover from the adverse eVects ofthe polyp at any time. Furthermore, this healthy stateappears to be preserved for periods of up to severalmonths.

In conclusion, endometrial polyp is one of the com-mon problems related to uterine receptivity in infertilepatients, and patients are advised to undergo polypresection prior to an IVF procedure. With respect to themost suitable time for the IVF trial following the polypresection, the pregnancy rates may not change regardlessof whether the IVF is done within a few months or manymonths later. IVF outcomes seem to be unrelated to thetime interval between the hysteroscopic polyp resectionand the initiation of the IVF; the success rates may notbe superior if the treatment is started in the Wrst fewmonths postoperatively.

ConXict of interest We declare that we have no conXict of interest.

Table 2 Oocyte maturity, embryo score and pregnancy outcomes in patients with IVF treatment within 6 months (Group A) or 6 months after(Group B) the endometrial polyp resection

Values are expressed as mean § SD and 95% conWdence intervals (95%CI). The embryos were scored according to Veeck’s criteria [16]

MII metaphase II, NS not signiWcant

Group A (n = 29) Group B (n = 31) p value

No. of follicles ¸16 mm 4.3 § 2.4 (95%CI: 3.3–5.2) 5.8 § 2.6 (95%CI: 4.7–6.7) 0.49

No. of retrieved oocytes 8.6 § 5.3 (95%CI: 6.5–10.7) 8.9 § 4.3 (95%CI: 7.4–10.7) NS

No. of MII oocytes 4.7 § 3.9 (95%CI:3.1–6.2) 5.2 § 3.5 (95%CI: 3.9–6.6) NS

No. of embryos transferred 1.4 § 0.8 (95%CI: 1.1–1.7) 1.4 § 0.7 (95%CI: 1.1–1.6) NS

Grade I embryo (%) 15/39 (38.5%) (95%CI: 0.2–0.9) 15/40 (37.5%) (95%CI: 0.2–0.8) NS

Grade II embryo (%) 20/39 (51.3%) (95%CI: 0.4–1.0) 14/40 (35.0%) (95%CI: 0.2–0.7) NS

Grade III embryo (%) 4/39 (10.2%) (95%CI: 0.0–0.2) 9/40 (22.5%) (95%CI: 0.0–0.6) NS

Fertilization rate (%) 101/244 (41.3%) (95%CI: 2.3–5.1) 110/255 (43.1%) (95%CI: 2.7–5.1) NS

Implantation rate (%) 8/39 (20.5%) (95%CI: 0.1–0.4) 9/40 (22.5%) (95%CI: 0.1–0.4) NS

Clinical pregnancy rate (%) 8/39 (20.5%) (95%CI: 0.1–0.4) 8/40 (20.0%) (95%CI: 0.1–0.5) NS

123

Arch Gynecol Obstet (2012) 285:1753–1757 1757

References

1. Norwitz ER, Schust DJ, Fisher SJ (2001) Implantation and the sur-vival of early pregnancy. N Engl J Med 345:1400–1408

2. Rama Raju GA, Shashi Kumari G, Krishna KM, Prakash GJ,Madan K (2006) Assessment of uterine cavity by hysteroscopy inassisted reproduction programme and its inXuence on pregnancyoutcome. Arch Gynecol Obstet 274(3):160–164

3. Karayalcin R, Ozcan S, Moraloglu O (2010) Results of 2,500 oYce-based diagnostic hyteroscopies before IVF. RBM 20:689–693

4. Bozdag G, Guldeniz Aksan, Esinler I (2008) What is the role ofoYce hysteroscopy in women with failed in vitro fertilizationcycles. RBM 17(3):410–415

5. Doldi N, Persico P, Sebastiano FD, Marsiglio E (2005) PathologicWndings in hysteroscopy before in vitro fertilization-embryo trans-fer (IVF-ET). Gynecol Endocrinol 21(4):235–237

6. Lass A, Williams G, Abusheikha N (1999) The eVect of endome-trial polyps on the outcomes of in vitro fertilization cycles. J AssistReprod Genet 16:410–415

7. Basak S, Dubanchet S, Zourbas S et al (2002) Expression of pro-inXammatory cytokines in mouse blastocysts during implantation:modulation by steroid hormones. Am J Reprod Immunol 4:2–11

8. Barash A, Dekel N, Fieldust S (2003) Local injury to the endome-trium doubles the incidence of successful pregnancies in patientsundergoing in vitro fertilization. Fertil Steril 79:1317–1322

9. AWW K, Anand S, Nallapeta S et al (2010) Management ofendometrial polyps in subfertile women: a systematic review. EurJ Obstet Gynecol Reprod Biol 151:117–121

10. Taylor E, Gomel V (2008) The uterus and fertility. Fertil Steril89:1–16

11. Bettocci S, Nappi L, Ceci O (2004) OYce hysteroscopy. ObstetGynecol Clin North Am 31:641–654

12. VanBogaert LJ (1988) Clinic pathologic Wndings in endometrialpolyp. Obstet Gynaecol 71:771–773

13. Pérez-Medina T, Bajo-Arenas J, Salazar F (2005) Endometrialpolyps and their implication in the pregnancy rates of patientsundergoing intrauterine insemination: a prospective, randomizedstudy. Hum Reprod 20(6):1632–1635

14. Al-Jefout M, Black K, Schulke L (2009) Novel Wnding of highdensity of activated ast cells in endometrial polyps. Fertil Steril92(3):1104–1106

15. Inagaki N, Ung L, Otani T et al (2003) Uterine cavity matrixmetalloproteinases and cytokines in patients with leiomyoma, ade-nomyosis or endometrial polyp. Eur J Obstet Gynecol Reprod Biol111:197–203

16. Veeck LL (1988) Oocyte assessment and biological performance.Ann N Y Acad Scien 541:259–274

123