Ovulation side and cycle fecundity: a retrospective analysis of frozen/thawed embryo transfer cycles

I. Järvelä1,3, S. Nuojua-Huttunen2 and H. Martikainen1

1 Department of Obstetrics and Gynaecology, Oulu University Hospital, FIN-90220 Oulu and 2 The Infertility Clinic of Oulu, the Family Federation of Finland, Finland


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The aim of the study was to evaluate a possible lateral difference in ovarian activity and its effect on cycle fecundity. A database was analysed retrospectively which covered 477 cycles in which frozen/thawed embryo transfer had been carried out. The cycles were spontaneous, with no hormonal treatment. Women with ovulation problems as a reason for infertility treatment were excluded. Factors investigated were the side of ovulation, endometrial thickness on cycle days 10–12 and on the day of embryo transfer, and pregnancy rate per embryo transfer. Ovulation was right-sided in 273 of the 477 cycles (57.2%) and left-sided in 204 of the cycles (42.8%) (95% CI 38.3–47.2, P = 0.002). In the age category of 30–37 years, covering 288 cycles, the incidence of left-sided ovulation was 126 (43.7%, 95% CI 38.0–49.5, P = 0.034). In this category, the endometrial thickness (±SD) was significantly greater on the day of embryo transfer (i.e. at time of implantation) when there had been left-sided ovulation compared with right-sided [9.6 mm (2.0) versus 9.1 mm (1.8), P = 0.037]. In addition, the pregnancy rate per embryo transfer was higher when ovulation had been on the left side [32/126 (25.2%) versus right side 24/162 (14.8%), P = 0.035, 95% CI 0.0122–0.199]. In conclusion, right-sided ovulation was more frequent than left-sided in the whole group. This is the first study to report that the side of ovulation has a clinical impact. These data support the hypothesis that the side of ovulation is significant in terms of embryo implantation.

Key words: cycle fecundity/endometrial thickness/frozen-thawed embryo transfer/ovulation side/pregnancy rate


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
It was observed (Potashnik et al., 1987Go) that ovulation in women tends to occur more frequently in the right ovary. In a later study (Check et al., 1991Go), there was a tendency towards more frequent right-sided ovulation, although there are also studies not confirming the issue (Marinho et al., 1982Go; Gougeon and Lefevre, 1984Go). Potashnik et al. (1987) also discovered that ipsilateral ovulation (two consecutive ovulations in the same ovary) is associated with a longer follicular period than contralateral ovulation (the side of ovulation changes between two consecutive ovulations). However, the impact of the side of ovulation on cycle fecundity was not evaluated in the study (Potashnik et al., 1987Go) and the physiological significance of their findings has remained unclear.

Later studies concerning assisted reproduction have shown that contralateral ovulation in the previous menstrual cycle results in better oocyte retrieval, fertilization, cleavage and embryo transfer rates in both spontaneous and stimulated cycles (Fukuda et al., 1996Go, 1998Go). It has been suggested that the corpus luteum secretes factors which negatively affect the follicular health of the responsive cohort of follicles of the next menstrual cycle (Fukuda et al., 1996Go, 1997Go, 1998Go).

If right-sided ovulation takes place more often than left-sided, it suggests that ipsilateral ovulation is more frequent on the right side. The objective of our study was to investigate the side of ovulation and its impact on cycle fecundity in frozen/thawed embryo replacement cycles.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A database was analysed retrospectively concerning frozen/thawed embryo transfer cycles from January 1995 to March 1999. The cycles were spontaneous, with no hormonal treatment. Women with ovulation problems as a reason for infertility treatment were excluded.

In the frozen/thawed embryo transfer protocol used in our clinic, transvaginal ultrasonographic examination was carried out twice during the cycle, first at cycle days 10–12. In these examinations, the side and size of the leading (largest) follicle and endometrial thickness were recorded. Endometrial thickness was measured as the thickest part in longitudinal section, including both endometrial layers. After the first ultrasonographic examination, the women began luteinizing hormone (LH) tests on urine samples every morning to confirm ovulation. Embryo transfer was carried out 48 h after the first positive LH test result. On the day of embryo transfer the second ultrasonographic examination took place. Clinical pregnancy was confirmed by ultrasonography after a positive urinary pregnancy test. No attempt was made to measure interobserver or intra-observer reproducibility in order to determine which side ovulation occurred with transvaginal ultrasonography.

The women were divided into two groups according to the side of ovulation. They were characterized by age, body mass index (BMI), number of embryos transferred and main infertility diagnosis (male factor, tubal factor and endometriosis). The embryos originated either from in-vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI) cycles. The characteristics of the groups are presented in Table IGo.


View this table:
[in this window]
[in a new window]
 
Table I. Clinical characteristics of the subjects, grouped according to side of ovulation
 
Statistical analyses were carried out using SPSS for MS Windows release 7.0. Departure from a normal distribution was assessed using the Kolmogorov—Smirnov test. The paired t-test was used for normally distributed data. Crosstabs and non-parametric {chi}2 test were used to assess the differences in proportions. Confidence intervals (CI) of 95% were obtained using CIA version 1.0 (Confidence Interval Analysis; Gardner and Altman, 1989). P < 0.05 was considered significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The total number of cycles included in the study was 477 in 345 women. Of these, 250 had one transfer, 68 had two, 19 had three, six had four and two had five frozen/thawed embryo transfers. The groups did not differ from each other in terms of subject characteristics (Table IGo), the mean number of transferred embryos or the cycle day on which the LH test result became positive (Table IIGo). In addition, the total number of blastomeres/transfer was equal between right- and left-sided ovulation (8.1 ± 3.6 and 8.3 ± 4.1 respectively) in the group as a whole as well as in the 30–37 year age group (8.4 ± 3.6 and 7.9 ± 3.8 respectively). Data on fragmentation were incomplete and are therefore not reported.


View this table:
[in this window]
[in a new window]
 
Table II. Incidence of ovulation, endometrial thickness on menstrual cycle day 10–12 (Endom. 1) and on the day of embryo transfer (Endom. 2), cycle day of the first positive LH test and the number of clinical pregnancies with respect to side in 477 frozen/thawed embryo transfer cycles
 
The leading follicle was on the right side in 273 of the 477 cycles (57.2%) and on the left side in 204 of the cycles (42.8%, 95% CI 38.3–47.2, P = 0.002) (Table IIGo).

On the day of embryo transfer the endometrium tended to be thicker when the leading follicle was on the left side but the difference was not significant. There was no difference in endometrial thickness as regards the first measurement on cycle days 10–12.

The pregnancy rate per embryo transfer was 20.9% (57/273) when the leading follicle had been on the right side and 25.0% (51/204) when on the left side (95% CI for the difference between proportions was –0.118 to 0.0353), i.e. not significantly different.

We re-analysed the data according to the age of the subjects. In the age category of 30–37 years, covering 288 cycles (Table IIIGo), the incidence of left-sided ovulation was 126, significantly lower than right-sided ovulation (P = 0.034). In this category, the endometrial thickness (±SD) was significantly greater on the day of embryo transfer when there had been left-sided ovulation (P = 0.037). In addition, the pregnancy rate per embryo transfer was significantly higher when ovulation had been on the left side (95% CI for the difference between proportions was 0.0122–0.199, P = 0.035). In women under 30 years or above 37 years there were no such differences in endometrial thickness or pregnancy rate.


View this table:
[in this window]
[in a new window]
 
Table III. Incidence of ovulation, endometrial thickness on the day of embryo transfer (Endom. 2) and the number of clinical pregnancies with respect to side in 288 frozen/thawed embryo transfer cycles in age category 30–37 years
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The present results confirmed the earlier finding (Potashnik et al., 1987Go) that right-sided ovulation is more frequent than left-sided. In addition, this study provides the first evidence that there are clinically significant differences between right and left ovarian function during the fertile period of a woman's lifespan. In the data overall there was a tendency, albeit non-significant, for the endometrium to become thicker during mid-cycle in the presence of left-sided ovulation. In the age group of 30–37 years, left-sided ovulation resulted in a significantly thicker endometrium at mid-cycle and in statistically significantly better pregnancy rates in comparison with right-sided ovulation. This observed difference in the pregnancy rate cannot be explained by subject characteristics or any other factor. In women under 30 years or above 37 years, such differences in the pregnancy rate were not significant, which may partly be explained by the lower numbers of women in these groups.

The only known difference between the right and left ovary lies in the anatomy of the veins. The left ovarian vein drains to the left renal vein and the right ovarian vein to the inferior vena cava. In the broad ligament there is a pampiniform plexus of veins which drains into both the ovarian and uterine trunks, indicating anastomoses between ovarian and uterine venous drainage.

In males, venous drainage from the testis corresponds to female ovarian drainage. Varicocele is a collection of large veins in the spermatic cord. The predominance of left-sided varicocele (the left side is affected in 70–100% and the right side in 0–9% of cases; Takihara et al., 1991) is considered to be due to the anatomical difference in venous drainage. According to one of the main aetiological hypotheses on the background of varicocele, it is believed that the left renal vein becomes trapped and compressed between the abdominal aorta and the superior mesenteric artery when the person is upright, leading to subsequent dilatation of the veins and even retention of venous blood flow (Takihara et al., 1991Go).

In women, retrograde phlebography has revealed even reflux and retrograde flow in the ovarian veins, resulting first in congestion of ipsilateral pampiniform plexus and finally also on the contralateral side due to anastomoses between the left and right plexus (Giacchetto et al., 1989Go, 1990Go). This reflux has been found to be predominantly present on the left side (Giacchetto et al., 1989Go). The reflux seems to result in pelvic varicosities and congestion, which have been found to be related to chronic pelvic pain in women with no other pathological findings concerning their gynaecological status (Beard et al., 1984Go).

According to the theory above, it may be that higher venous pressure in the left ovarian vein directs venous drainage from the left ovary more toward the uterus, enabling better absorption of hormones from the veins to the adjacent arteries supplying the uterus (Einer-Jensen, 1990Go), finally resulting in higher hormone concentrations in the uterus than does drainage from the right side. Higher hormone concentrations in the uterus might explain the thicker endometrium and better pregnancy rates observed in our study.

The results of previous studies suggest that intra-ovarian factors secreted locally, e.g. by the corpus luteum, affect neighbouring follicular growth negatively during the luteal phase of the menstrual cycle (Fukuda et al., 1997Go). If the venous flow from the left ovary is slower, it might result in a longer clearance period of these factors and subsequently diminish the possibility of ipsilateral left-sided ovulation in the next menstrual cycle, explaining our finding of lower ovulation frequency on the left side.

The results of this study indicated that right-sided ovulation is more frequent than left-sided. In addition, this study gives support to the hypothesis that the side of ovulation has an impact on implantation of the embryo. The left ovary appears to act more effectively than the right one in women between 30 and 37 years of age, as reflected in endometrial thickness at mid-cycle and the pregnancy rate following frozen/thawed embryo transfer.


    Notes
 
3 To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, Oulu University Hospital, FIN-90220 Oulu, Finland. E-mail: ijarvela{at}cc.oulu.fi Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Beard, R.W., Highman, J.H., Pearce, S. et al. (1984) Diagnosis of pelvic varicosities in women with chronic pelvic pain. Lancet, ii, 946–949.

Check, J.H., Dietterich, C. and Houck, M.A. (1991) Ipsilateral versus contralateral ovary selection of dominant follicle in succeeding cycle [see comments]. Obstet. Gynecol., 77, 247–249.[Abstract]

Einer-Jensen, N. (1990) Functional and therapeutic aspects of counter current transfer in the female adnexa. Med. Hypoth., 33, 125–127.[ISI][Medline]

Fukuda, M., Fukuda, K., Andersen, C.Y. et al. (1996) Contralateral selection of dominant follicle favours pre-embryo development. Hum. Reprod., 11, 1958–1962.[Abstract]

Fukuda, M., Fukuda, K., Yding Andersen, C. et al. (1997) Does corpus luteum locally affect follicular growth negatively? Hum. Reprod., 12, 1024–1027.[ISI][Medline]

Fukuda, M., Fukuda, K., Andersen, C.Y. et al. (1998) Contralateral ovulation shortens follicular phase length and favours pre-embryo development during ovarian stimulation with clomiphene citrate. Hum. Reprod., 13, 1590–1594.[Abstract]

Gardner, M.J. and Altman, D.G. (eds) (1989) Statistics with Confidence. British Medical Journal, London.

Giacchetto, C., Catizone, F., Cotroneo, G.B. et al. (1989) Radiologic anatomy of the genital venous system in female patients with varicocele. Surg. Gynecol. Obstet., 169, 403–407.[ISI][Medline]

Giacchetto, C., Cotroneo, G.B., Marincolo, F. et al. (1990) Ovarian varicocele: ultrasonic and phlebographic evaluation. J. Clin. Ultrasound, 18, 551–555.[ISI][Medline]

Gougeon, A. and Lefevre, B. (1984) Histological evidence of alternating ovulation in women. J. Reprod. Fertil., 70, 7–13.[Abstract]

Marinho, A.O., Sallam, H.N., Goessens, L. et al. (1982) Ovulation side and occurrence of mittelschmerz in spontaneous and induced ovarian cycles. Br. Med. J., 284, 632.[ISI][Medline]

Potashnik, G., Insler, V., Meizner, I. et al. (1987) Frequency, sequence, and side of ovulation in women menstruating normally. Br. Med. J., 294, 219.

Takihara, H., Sakatoku, J. and Cockett, A.T. (1991) The pathophysiology of varicocele in male infertility. Fertil. Steril., 55, 861–868.[ISI][Medline]

Submitted on November 26, 1999; accepted on March 6, 2000.