The purpose of this trial is to evaluate the morphological characteristics of the uterus with three-dimensional ultrasound (3DUS) at the time of IVF/ICSI treatment and correlate it to clinical outcomes. The uterus will be assessed in the 3D coronal plane and measurements will be performed to assess whether they correlate with clinical pregnancy rate, miscarriage rate and live birth rate.

INTRODUCTION

There has been a lot of increased interest in the recent years in the prevalence and diagnosis of congenital uterine anomalies (Chan et al 2011; Saravelos et al 2008). This has been further intensified by the introduction of several new classifications for these malformations, including the most recent ESHRE-ESGE classification (Grimbizis et al, 2013).

Most of the clinical interest, and what affects daily clinical practice is the diagnosis and treatment of the septate uterus, which is the commonest anomaly amenable to treatment (Homer and Li, 2000). Since the new ESHRE-ESGE classification, which has introduced a new objective method for diagnosis of this anomaly, by measuring the percentage of fundal indentation and considering to be a septate uterus when it is >50%.

To our knowledge there have been no prospective studies as yet assessing women undergoing ART in order to demonstrate whether indeed the 3D US morphological characteristics (including the percentage of fundal indentation) are correlated to either clinical pregnancy or clinical miscarriage. Indeed, this is of upmost importance, as for example the decision to operate on a so-called septate uterus should be based on whether it has a true impact on clinical outcomes. This is all the more most topical in the present era of IVF/ICSI, and the new classifications for uterine anomalies, as certain authors have criticized the new ESHRE-ESGE classification as significantly over-diagnosing the septate uterus which risks causing unnecessary interventions (Ludwin et al, 2015). Meanwhile other 3D US morphological characteristics such as characterisation of the subendometrial layers/junctional zone and Cesarean scan niches my also prove to be correlated with clinical outcomes.

AIMS Aim: To determine whether and to what extent different 3D US morphological characteristics of the uterus impact the clinical outcomes of women undergoing IVF/ICSI.

PLAN OF INVESTIGATION Subject recruitment: Women with infertility will be recruited from the Assisted Reproductive Technology Unit of Hammersmith Hospital, London. It is often our routine clinical practice to offer 3D US to patients undergoing treatment and we are anticipating that over 90% of patients undergoing treatment will be willing to have this evaluation as part of their routine treatment.

Study design: Prospective cohort observational study.

Inclusion criteria: Described Elsewhere

Exclusion criteria: Described elswhere

Power calculation: To our knowledge no previous prospective study has been performed within a similar context, therefore a power calculation is not easily applicable. However, given that most recent similar retrospective study comparing arcuate versus normal uteri in women undergoing IVF/ICSI included 161 subjects, we will also aim to collect over 150 subjects.

Procedure: All ultrasound examinations will be performed using the GE Voluson series ultrasound machines, with a standard 3D transvaginal probe (model RIC5-9-D; 6.6 MHz central image frequency). The examination with 3DUS will be identical to the routine 2D examination including an acquisition of a 3D volume which takes a few seconds. The images will be shown instantly to the operator and can be stored digitally onto the hard disc attached to ultrasonography machine for subsequent analysis. For the purpose of homogeneity and to reduce operator bias, no more than three experienced operators will perform all 3DUS examinations.

Outcome measures: Described elsewhere

Data processing and analysis: The researchers will ensure the confidentiality of sensitive data by minimizing the number of personnel who handles subject data. The computer data will be encrypted as required to maximize security. Data processing for statistical analysis will be performed using the Statistical Packages of Social Sciences for Windows (SPSS, Inc). Data will be presented by percentage, mean and standard deviation, median and range where appropriate. Comparisons between groups will be carried out by Student T test for continuous variables, ANOVA for mean value comparisons, Chi-square/Fisher's exact test for categorical data. Stepwise logistic regression analysis may be performed as part of systematic analysis for primary outcomes. P-values of <0.05 will be considered significant.

Monitoring and interim analysis: As the study is perceived to be safe and poses no additional hazard, a Data Monitoring Committee and interim analysis is not considered necessary. There are no known adverse effects of routine ultrasound. No extra pain or discomfort is associated with the 3D US.

Consent: All subjects will be asked to consent for the performance of 2D/3D US which forms part of routine care in our unit.

Inclusion Criteria:

• Women undergoing IVF/ICSI in either fresh or frozen cycles.
• Women whose endometrial cavity can be visualised adequately via US

Exclusion Criteria:

• Women aged >42 years
• Women whose endometrial cavity cannot be visualised adequately via US, such as large fibroids, adenomyosis, BMI>35, significant uterine retroflexion/retroversion
• Previous uterine surgery (e.g. septum resection, adhesiolysis)

Women with infertility undergoing IVF/ICSI at the Assisted Reproductive Technology Unit, Hammersmith Hospital, London.