Follicular diameters in conception cycles with and without multiple pregnancy after stimulated ovulation induction

J.R. Richmond1,3, N. Deshpande2, H. Lyall2, R.W.S. Yates2 and R. Fleming2

1 Department of Obstetrics and Gynaecology, Queen Mother's Hospital, Glasgow, G3 8SJ and 2 Department of Obstetrics and Gynaecology, Royal Infirmary, Glasgow, G31 2ER

3 To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, Queen Mother's Hospital, Glasgow, G3 8SJ. Email: jane.richmond{at}blueyonder.co.uk


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Controlled ovulation induction and intrauterine insemination (OI-IUI) is associated with multiple pregnancies, which are a cause of much concern. No reliable datasets have shown clear criteria for predicting multiple pregnancy. The aim of this study was to eliminate a number of variables by examining only conception cycles to determine ultrasound criteria posing risks of multiple pregnancy. METHODS: 112 OI-IUI conception cycles (multiple pregnancy rate 19.6%) were analysed retrospectively to identify factors that may be used to evaluate multiple pregnancy risk. Analyses of ultrasound data on the day of hCG administration allowed study of the role of primary, secondary and tertiary follicle diameters (FD). RESULTS: There were no multiple pregnancies in cases where there was a single FD ≥14 mm, and no higher-order pregnancies where the tertiary follicle measured <14 mm. Follicles with an FD of 15 mm showed an 8% attributable implantation rate. CONCLUSIONS: Revision of the criteria for administration of the ovulatory dose of hCG should include the concept that follicles of 15 mm diameter may yield a pregnancy. We suggest that rigorous application of such criteria (critical FD of 16 mm combined with secondary FD evidence) will not reduce the programme pregnancy rate, but will reduce the incidence of multiple conceptions.

Key words: follicle size/multiple pregnancy/ovulation induction/pregnancy prediction


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Ovulation induction and intrauterine insemination (OI-IUI) using exogenous gonadotrophins is common practice in infertility clinics to treat women with anovulation, polycystic ovary syndrome (PCOS), oligo-ovulation, and unexplained infertility. It is less invasive than other assisted reproductive treatment (ART) procedures, requiring less laboratory commitment and regulation (Homburg, 2003Go; van Santbrink and Fauser, 2003Go). Success rates are reasonable, although multiple pregnancy is a risk, whose importance is becoming increasingly recognised (Lambalk and van Hooff, 2001Go). In countries where the number of embryos transferred in ART is restricted, OI-IUI is possibly the major cause of higher order multiple pregnancy.

Intensive monitoring of ovarian responses, using transvaginal ovarian ultrasonography and assays of circulating estradiol (E2), should be (and usually is) employed in an attempt to reduce the incidence of multiple ovulation and pregnancy. However, there is no universal agreement as to the criteria for the ‘safe’ administration of human chorionic gonadotrophin (hCG) to induce ovulation, and centres employing conventional follicle stimulating hormone (FSH) doses often report high multiple pregnancy rates (van Santbrink and Fauser, 2003Go). The strictness with which designated criteria are applied is variable between centres. In addition, there is currently no agreed accurate predictive model, which can be used on an individual patient basis.

In the UK, the National Institute of Clinical Excellence has issued guidelines (NICE, 2004Go) recommending ultrasound monitoring with no requirement for E2 measurements: the latter deemed not to provide additional information. However, a prospective comparative study using either ultrasound or E2 monitoring has never been performed. The universal acceptance that ultrasonography is best, remains unjustified, as multiple pregnancy has probably not been reduced in programmes dictated solely by ultrasound criteria.

The crucial factor to achieving a pregnancy in ovulation induction is the presence of a mature follicle at the time of hCG administration: one that has the capacity to release a mature oocyte. In the normal ovulatory cycle, pre-ovulatory mature follicle dimensions are between 17 and 25 mm in diameter (Hackeloer et al., 1979Go). There is no evidence to suggest that follicular maturity differs in stimulated cycles. These mature follicles differ from smaller ones biochemically (Thomas et al., 2000Go), and also in their capacity to release viable oocytes (Bergh et al., 1998Go). Administration of hCG when the follicles are of immature dimensions probably leads to retention of the oocyte within the follicle (Stanger and Yovich, 1984Go), or to the ovulation of oocytes with reduced in vivo fertilization and developmental potential. If hCG administration in stimulated cycles (or luteinization) is delayed until there are too many mature follicles present, increased rates of multiple ovulation and pregnancy result. There are a number of reasons why there has been little detailed examination of the follicular circumstances whereby multiple pregnancy can occur. One major problem is that the conception rate of OI-IUI is sufficiently low (10–15 % of the cycles) to cast doubt upon the reliability of data from one cycle. There may be other reasons for failure to conceive, such as inappropriate timing of insemination, or an unidentified problem with the endometrium. Furthermore, there is no agreed definition of a mature-sized follicle.

The contribution of these other factors can be eliminated by examining data from conception cycles only. The aims of this study were to examine the ability of ultrasound criteria (primary and secondary follicle diameters) to predict the outcome (pregnancy and multiple pregnancy) following controlled ovarian stimulation (COS) in a programme of OI-IUI. If we assume that pregnancy in cycles of ovulation induction results from the fertilization and implantation of the oocyte released by the lead follicle, and multiple implantations derive from ovulation of secondary (±tertiary) follicles, then estimates can be made of the subordinate follicle criteria likely to yield multiple pregnancies.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Database origins
A database was constructed from 1461 consecutive cycles in the OI-IUI programme at the Assisted Conception Unit, Glasgow Royal Infirmary (a tertiary referral centre), between January 1997 and December 1998 (762 cycles), and February 2000 and October 2002 (699 cycles). The mean age of the women was 33.2 years (range 21.7–42.1 years). The proportions of the different diagnostic categories of patients treated were: unexplained infertility, 62%; PCOS, 10%; hypogonadotrophic hypogonadism, 1%; male factor, 1%, and donor insemination, 25%.

Treatment
All COS cycles, except those in patients with hypogonadotrophic hypogonadism, were effected using a long course gonadotrophin releasing hormone agonist (GnRH-ag) regime with various exogenous ovarian stimulants. Patients were down regulated with intra-nasal Buserelin (150 µg four times daily: Shire Pharmaceuticals Ltd, Andover, UK) and stimulation commenced when the circulating E2 was <100 pg/ml, combined with a thin endometrium and no ovarian cysts on transvaginal ultrasound scan. Ovarian stimulation was effected with highly purified human FSH (Metrodin High Purity: Serono UK Ltd, Feltham), Gonal-F (Serono UK Ltd), Puregon (Organon UK Ltd, Cambridge), Normagon (Organon UK Ltd), or Menogon (Ferring Pharmaceuticals, Langley). The starting dose of FSH was 150 IU per day in all cases, although this was frequently modified in subsequent cycles. Cycle responses were monitored with serum E2 concentrations taken daily or alternate daily from stimulation day 5 (S5), and with frequent transvaginal ultrasonic assessment of follicular growth. Hormone concentrations (E2, P, LH and FSH) were determined using the Immulite semi-automated assay system (DPC, Los Angeles).

Ultrasound data were recorded following scans, which were performed by eight operators with a Siemens Sonoline Sienna Ultrasound Imaging System using a 7.5 MHz vaginal probe. Follicular diameters were calculated as a mean of the largest diameter and its perpendicular value.

Criteria for administration of hCG
The criteria for pre-ovulatory hCG administration were strictly applied, and intra-muscular hCG (Profasi 10 000 IU: Serono UK Ltd) was administered when there were 1–3 leading follicles with FD≥16 mm observed by ultrasound scan. HCG administration was withheld in over-stimulated cycles, where more than three follicles with follicle diameter≥16 mm were observed, and also in a few cases where there was insufficient follicular development. The patients who over-responded were maintained on GnRH-ag therapy to avoid spontaneous multiple ovulation. IUI was scheduled 26–42 h after hCG injection. Hormone support in the luteal phase was given in the form of vaginal progesterone pessaries (Cyclogest 200 mg bd: Shire Pharmaceuticals Ltd) for 10 days following IUI. Pregnancy was diagnosed by a positive urine test 17 days after hCG, and confirmed by subsequent transvaginal ultrasound scan at 6 weeks gestation.

Database information
Data maintained on the database included patient diagnoses, anthropometric criteria, sequential E2 concentrations from S5 to hCG administration, and the ultrasound data and circulating progesterone concentrations on the day of hCG. Ultrasound data on the day of hCG were recorded as the number of individual follicles defined by follicle diameter (FD). Pregnancy outcomes were independently confirmed by the authors (J.R. and N.D.), and anonymity maintained with respect to use of donor insemination at all times. Ectopic pregnancies and miscarriages were excluded from the subsequent analyses. The possibility of monozygocity in multiple gestations was not taken into account, as the influence of embryo culture was not relevant to this programme.

Definition of lead and secondary follicles
Follicles were categorized according to mean diameter: the lead follicle was considered to be the largest active follicle (FDMax); the secondary follicle (FD2) was defined as the next largest follicle; and the tertiary follicle (FD3) was the third largest.

Statistics
Group statistics were analysed using contingency table analyses (Chi square).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Of the 1461 cycles in which FSH stimulation was commenced, 213 cycles resulted in a positive pregnancy test (14.6%). There were complete data sets for 112 cycles: where follicle diameters on the day of hCG and the number of positive fetal cardiac activities were recorded. The 101 incomplete datasets were excluded from the analyses. They were randomly distributed through the database, and were excluded mainly because the ultrasound data on the day of hCG was not available as primary source evidence, or because the pregnancy outcome information was incomplete.

Of the 112 cycles with complete data sets, 90 were recorded to have resulted in single fetal heart activity. There were 18 cases of twin cardiac activity, three triplet pregnancies and one quadruplet pregnancy. The multiple pregnancy rate was 19.6%, and the rate of higher order pregnancies (triplet/quadruplet) was 3.6%.

In the whole programme of ovulation induction, the cancellation rates were as follows: excessive response, 6.8% of cycles started; and poor response, 4.3%. This yields a total cancellation rate of 11.1%.

Follicular diameters, implantation and multiple pregnancy
We make the single presumption that each implantation derives from a follicle observed on the day of hCG, and that the largest follicle represents the most mature follicle: that most likely to be the source of the oocyte which is subsequently fertilized and implants.

Table I shows the analyses of cycle outcomes depending upon the number of implantations and the FDs of the follicles recorded on the day of hCG administration. The FDMax value has little relevance to the analyses of multiple pregnancy, because it is the single most important criterion determining administration of hCG. FDMax is therefore self-fulfilling. On the other hand, FD2, FD3 and FD4 are dependent only upon the ovarian responses to treatment. It is clear from the secondary follicles in the multiple pregnancies that follicles of diameter 16 mm and 15 mm are capable of contributing to the implantation. Follicles with FD=14 mm also appear capable of yielding an implantation.


View this table:
[in this window]
[in a new window]
 
Table I. Implantations following ovulation induction according to the follicle diameters of the leading follicles on the day of hCG administration

 
Table II shows the numbers of follicles observed in each diameter category on the day of HCG and the number of attributable implantations deriving from them in these conception cycles. Again, assuming that the larger follicles are those responsible for the implantation, it is clear that follicles with FD >16 mm were highly effective as were follicles with FD=16 mm: 31% resulted in a conception, whether they were described as FDMax or not. Follicles with diameter of 15 mm showed an 8% attributable implantation rate. This was a statistically significant reduction (P<0.001) in efficacy. Follicles <15 mm were able to yield an attributable implantation only rarely. However, this does demonstrate that a follicle measured at FD=15 mm has considerable potential to yield an implantation in a pregnancy cycle.


View this table:
[in this window]
[in a new window]
 
Table II. Comparison of the attributable contributions of the different follicle sizes to the cohort of implantations in the series of conception cycles

 
It is noteworthy to record that there were no triplet or quadruplet pregnancies when the tertiary follicle measured <14 mm.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
These analyses demonstrate that incremental FD profiles showed negligible predictive power for multiple pregnancies, in parallel with other studies. However, there were specific ovarian ultrasound criteria that showed significant potential for clinical application. There was no multiple pregnancy when there was a single FD≥14 mm, and no triplet/quadruplet pregnancy when the tertiary follicle was <14 mm. The results indicate that the current criteria of hCG administration, i.e. the presence of 1–3 leading follicles with diameter >16 mm observed by ultrasound scan, may be excessively ‘liberal’. The data suggest that follicles measured with FD >16 mm and FD=16 mm show similar potential to yield pregnancies, while follicles with FD=15 mm also demonstrate potential. A more rational approach to the timing of hCG administration in ovulation induction can be proposed, using the critical follicular diameter of 16 mm combined with the secondary FD evidence. The secondary FD evidence should take into account the fact that, in fecund cycles, follicles with a diameter measured at 15 mm have an 8% chance of producing an implantation.

Limitations of this study include the observer error in recording FDs, and also the basic assumption that the largest follicle is that responsible for the implantations observed. However, the former problem is a practical issue, universal in any ovulation induction protocol, and the latter assumption has been used to determine minimum ‘attributability’ to the smaller follicles. The reality of the situation is that a follicle with FD=16 mm, in these conception cycles, resulted in a minimum implantation rate of 31%. This indicates that follicles with FD=16 mm probably show similar efficacy to FD > 16 mm. However, the reduction in potential seen with FD <15 mm indicates a low implantation potential from this point.

A difference in a FD measurement of 1 mm, at this diagnostic range, may be too demanding for the technology and skills available to many centres. However, in reality, the decision to administer hCG or not relies on clinical markers, making the definition of cut-off levels necessary. Further limitations of the study derive from its retrospective nature resulting in potential confounding factors. The study population was a heterogeneous group of women and further work is needed to facilitate the analysis of different patient groups according to their diagnostic criteria.

More than two thirds of higher order multiple births can now be attributed to infertility treatments of which programmes of OI are a major component. Of 156 triplet and higher multiple births studied, 31% were conceived spontaneously, 34% had ovarian stimulation, 24% had in vitro fertilization (IVF), and 11% gamete intrafallopian transfer (GIFT) (Levene et al., 1992Go). Higher order pregnancies are associated with an increased risk of prematurity with resultant higher perinatal morbidity and mortality, increased cost and greater use of neonatal resources (Wilcox et al., 1996Go). Furthermore, even among twins, those deriving from infertility treatment have poorer outcomes defined by some classical criteria (Lambalk and van Hooff, 2001Go; Lambert, 2003Go).

In the UK, the Royal College of Obstetricians and Gynaecologists has issued guidelines stressing that measures should be taken to reduce the risks of multiple pregnancy (RCOG, 1998Go). Advice includes discontinuation of stimulation and the ovulatory dose of hCG withheld if there are more than three follicles with FD≥16 mm, although the evidence base for this recommendation is not strong. However, even with these precautions being followed, multiple pregnancies will still occur in an excessive proportion of cycles, as demonstrated in this audit.

Ovarian stimulation for infertility treatment is a complex process requiring close observation and individualized decision making. Examination of the relevant literature on the techniques of OI with IUI reveals that most reports are of uncontrolled studies involving retrospective analyses of cycle programmes, which vary in their patient inclusion criteria, methods of monitoring and treatment protocols. Few programmes used GnRH-ag suppression of premature luteinization. Whilst the use of GnRH-ag control during ovulation induction in women with functional pituitaries may not improve pregnancy potential overall (Dodson and Haney, 1991Go), it may display a number of clinical advantages over conventional dose administration of human menopausal gonadotrophin (hMG). These include an avoidance of premature luteinization, and its consequent decline in the E2 concentration, which confuses the relationship between follicular growth and circulating estradiol concentrations. Continued GnRH-ag treatment also avoids spontaneous ovulation when excess follicles are seen to develop. In summary, GnRH-analogs may allow more multi-follicular development, but they avoid the confounding effects of spontaneous LH surges.

The critical factor in the development of multiple pregnancies is the criteria used for administration of hCG. The strictly applied practice of withholding hCG when there were more than three follicles with diameter ≥16 mm observed (current study), contrasts with a ‘guideline approach’ which only ‘recommends’ the cancellation of cycles with six or more ‘preovulatory’ follicles of diameter ≥16 mm (Gleicher et al., 2000Go). Of the 3137 completed cycles reported in the latter single centre study, 15.8% had a raised hCG level. Of their pregnancies, 8.9% were of three embryos or more (0.5% sextuplets), with 20% being as a result of twinning. The incidence of multiple pregnancies was associated with higher E2 concentrations and a greater total number of follicles. The data presented above suggest that tighter restrictions of hCG criteria using assessments of secondary follicle diameters (based upon a critical diameter of 15 mm) may lower the multiple pregnancy rate further. This is backed by previous studies (Dickey et al., 2001Go) which found that withholding hCG or IUI in hMG cycles when six follicles are ≥12 mm may reduce triplet and higher-order implantations by 67% without significantly reducing pregnancy rates for patients under 35 years of age.

Other studies have retrospectively examined different variables affecting OI-IUI outcome (Dodson et al., 1988Go; Tomlinson et al., 1996Go; Nuojua-Huttunen et al., 1999Go) but been unable to find clear positive predictive factors of multiple pregnancy.

More recently, debate has centred on the rationale of using stimulated IUI to treat unexplained infertility versus IUI alone or IVF (Stewart, 2003Go). While OI-IUI is currently more cost effective than IVF per pregnancy, ovarian stimulation results in multifolliculogenesis with the attendant risk of multiple pregnancy. Perhaps in future, the move towards a two-embryo transfer limit may allow IVF to become a safer and cheaper form of treatment for these couples.

In conclusion, the critical secondary follicle diameter appears to be 15 mm, below which multiple pregnancy is unlikely to occur. This observation suggests that secondary FD evidence and a more conservative approach to hCG administration may help to reduce the multiple pregnancy rate in OI-IUI programmes. In reality, there is probably an unfortunate and inevitable link between an increased pregnancy rate and increased multiple pregnancy rate, but a more conservative approach should reduce the latter without significantly compromising the former. In the knowledge that, in fecund cycles, follicles with a diameter of 15 mm have an 8% chance of producing a viable implantation, and that 14 mm follicles show a much reduced incidence (4%), we would recommend that cycles be cancelled when there are more than two follicles present whose diameters are ≥15 mm at the time of hCG administration. This should be tested prospectively.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Bergh C, Broden H, Lundin K and Hamberger L (1998) Comparison of fertilization, cleavage and pregnancy rates of oocytes from large and small follicles. Hum Reprod 13, 1912–1915.[Abstract]

Dickey RP, Taylor SN, Lu PY, Sartor BM, Rye PH and Pyrzak R (2001) Relationship of follicle numbers and estradiol levels to multiple implantation in 3,608 intrauterine insemination cycles. Fertil Steril 75, 69–78.[CrossRef][ISI][Medline]

Dodson WC, Hughes CL Jr and Haney AF (1988) Multiple pregnancies conceived with intrauterine insemination during superovulation: an evaluation of clinical characteristics and monitored parameters of conception cycles. Am J Obstet Gynecol 159, 382–385.[ISI][Medline]

Dodson WC and Haney AF (1991) Controlled ovarian hyperstimulation and intrauterine insemination for treatment of infertility. Fertil Steril 55, 457–467.[ISI][Medline]

Gleicher N, Oleske DM, Tur-Kaspa I, Vidali A and Karande V (2000) Reducing the risk of higher-order multiple pregnancy after ovarian stimulation with gonadotrophins. N Engl J Med 343, 2–7.[Abstract/Free Full Text]

Hackeloer BJ, Fleming R, Robinson HP, Adam AH and Coutts JRT (1979) Correlation of ultrasonic and endocrinological assessment of follicular development. Am J Obstet Gynecol 135, 122–128.[ISI][Medline]

Homburg R (2003) The case for initial treatment with intrauterine insemination as opposed to in vitro fertilization for idiopathic infertility. Hum Fertil 6, 122–124.

Lambalk CB and Van Hooff M (2001) Natural versus induced twinning and pregnancy outcome: a Dutch nationwide survey of primiparous dizygotic twin deliveries. Fertil Steril 75, 731–736.[CrossRef][ISI][Medline]

Lambert RD (2003) Safety issues in assisted reproductive technology: aetiology of health problems in singleton ART babies. Hum Reprod 18, 1987–1991.[Abstract/Free Full Text]

Levene MI, Wild J and Steer P (1992) Higher multiple births and the modern management of infertility in Britain. The British Association of Perinatal Medicine. B J Obstet Gynecol 99, 607–613.

National Institute for Clinical Excellence guideline (2004) Fertility: assessment and treatment for people with fertility problems. NICE.

Nuojua-Huttunen S, Tomas C, Bloigu R, Tuomivaara L and Martikainen H (1999) Intrauterine insemination treatment in subfertility: an analysis of factors affecting outcome. Hum Reprod 14, 698–703.[Abstract/Free Full Text]

Royal College of Obstetricians and Gynaecologists evidence-based clinical guidelines (1998) The management of infertility in secondary care. RCOG Press.

Stanger JD and Yovich JL (1984) Failure of human oocyte release at ovulation. Fertil Steril 41, 827–832.[ISI][Medline]

Stewart JA (2003) Stimulated intra-uterine insemination is not a natural choice for the treatment of unexplained subfertility: should the guidelines be changed. Hum Reprod 18, 903–914.[Abstract/Free Full Text]

Thomas N, Goodacre R, Timmins EM, Gaudoin M and Fleming R (2000) Fourier transform infrared spectroscopy of follicular fluids from large and small antral follicles. Hum Reprod 15, 1667–1671.[Abstract/Free Full Text]

Tomlinson MJ, Amissah-Arthur JB, Thompson KA, Kasrraie JL and Bentick B (1996) Prognostic indicators for intrauterine insemination (IUI): statistical model for IUI success. Hum Reprod 11, 1892–1896.[Abstract]

Van Santbrink EJP and Fauser BCJM (2003) Is there a future for ovulation induction in the current era of assisted reproduction? Hum Reprod 18, 2499–2502.[Abstract/Free Full Text]

Wilcox LS, Kiely JL, Melvin CL and Martin MC (1996) Assisted reproductive technologies: estimates of their contribution to multiple births and newborn hospital days in the United States. Fertil Steril 13, 361–366.

Submitted on September 9, 2004; accepted on November 25, 2004.