Abstract
Congenital uterine anomalies (CUAs) are gaining increasing attention in the field of gynaecological ultrasound for a number of reasons: first, they appear to be of relatively high prevalence in both selected and unselected groups of women ; second, they appear to have a significant impact on reproductive outcomes and, on occasion, in adolescent symptomatology ; third, there has been a recent surge in relevant publications, which has culminated in a new international classification , and also a new international consensus for diagnosis. Three-dimensional (3D) ultrasound is now recommended as the gold standard method for diagnosis, which implies that gynaecologists and/or sonographers may be expected to attain the correct diagnoses and classification of CUA for women presenting to them with, and even without, symptomatology.
Introduction
Congenital uterine anomalies (CUAs) are gaining increasing attention in the field of gynaecological ultrasound for a number of reasons: first, they appear to be of relatively high prevalence in both selected and unselected groups of women [1,2]; second, they appear to have a significant impact on reproductive outcomes and, on occasion, in adolescent symptomatology [3,4]; third, there has been a recent surge in relevant publications, which has culminated in a new international classification [5], and also a new international consensus for diagnosis [6]. Three-dimensional (3D) ultrasound is now recommended as the gold standard method for diagnosis, which implies that gynaecologists and/or sonographers may be expected to attain the correct diagnoses and classification of CUA for women presenting to them with, and even without, symptomatology.
The aim of the current chapter is to cover a number of areas relating to CUA. First, current and previous classifications of CUA will be described; second, the diagnostic accuracy of different diagnostic modalities will be presented; third, the prevalence of CUA will be outlined; fourth, the clinical implications of each CUA will be ascertained; and finally, a practical step-by-step guide to diagnosing CUA using 3D ultrasound will be suggested.
Which Classification Should Be Used?
One cannot set out to diagnose a CUA without knowledge of an appropriate classification system to apply to their findings. The evolution of the classifications for CUA has been interesting to observe. The very first attempts at classifying CUA appear to have originated from Cruveilher, Foerster and von Rokitansky in the mid-nineteenth century [7]. Following a series of further publications describing various patterns of CUA, Buttram and Gibbons presented a classification, which was based on both the anatomy of the anomaly and the degree of embryological failure that occurred at the Mullerian duct system [8]. This classification would later form the backbone of the most well-known classification used to date: the American Fertility Society (AFS; nowadays American Society for Reproductive Medicine (ASRM)) classification [9]. This classification contains a schematic graph of seven different types of anomalies, and has served the research and clinical community for almost 30 years. However, as imaging has become more accurate, new proposals of classifications emerged, which intended to be more comprehensive. This included the ‘VCUAM classification’, which individually described the anatomical anomalies of the vagina, cervix, uterus and associated malformations in order to systematically describe and report all genital anomalies [10]. It also included the ‘Embryological clinical classification for female genitourinary problems’ proposed by Acien and Acien originally in 1992 and subsequently updated in 2011 [7,11], which included as a basis the different embryological pathways and stages of development that have been observed.
Finally, most recently, the European Society for Human Reproduction and Embryology (ESHRE) and the European Society for Gynaecological Endoscopy (ESGE) jointly formed a working group in order to create the newest classification system. This ESHRE/ESGE classification was published jointly in 2013 and was achieved through a structured voting procedure (known as the Delphi procedure), involving a wider number of experts of the field [5]. This classification consists of descriptions for all female genital tract malformations (similar to the VCUAM classification) and also provides schematic guides to allow for diagnosis and differentiation between different anomalies (similar to the AFS classification). It also encourages objective measurements of the uterus, so as to accurately differentiate between subtypes of CUA. In particular, the septate uterus has been defined as the uterus with normal outline and an internal indentation at the fundal midline exceeding 50 per cent of the uterine wall thickness, while the bicorporeal (or bicornuate uterus) has been defined as a uterus with the presence of an external indentation at the fundal midline exceeding 50 per cent of the uterine wall thickness. The ESHRE/ESGE classification for the diagnosis of CUA is shown in Figure 5.1.
However, it is worth noting that although the most modern and updated classification is the ESHRE/ESGE classification, it has received criticism from a group of authors who observed an increase in the diagnosis of the septate uterus when using this classification compared with former classifications. These authors advised caution when using this new classification, as they fear it may cause confusion in terms of clinical management [12].
What Is the Most Accurate Investigation?
This is a question that has puzzled clinicians for several years. One of the reasons is that there are few anatomical organs that can be assessed with so many different investigative modalities. For the female genital tract this includes: 2D ultrasound, 3D ultrasound, saline infusion sonography (SIS), magnetic resonance imaging (MRI), hysterosalpingography (HSG), hysteroscopy and laparoscopy. Until recently, the methods used were primarily dependent on local availability and clinician preference, rather than rates of diagnostic accuracy [1]. However, recently an international consensus (coined the Thessaloniki Consensus) was published, which included a systematic review of 38 studies examining the diagnostic accuracy of all the aforementioned methodologies against the historical gold standard of combined hysteroscopy and laparoscopy [6]. The most accurate methods in descending order (presented here as mean accuracy (95 per cent confidence interval)) were found to be: 3D ultrasound 97.6 per cent (94.3–100), SIS 96.5 per cent (93.4–99.5), HSG 86.9 per cent (79.8–94.0) and 2D ultrasound 86.6 per cent (81.3–91.8). Although no studies were found examining MRI as a screening tool (which is understandable due to the cost, frequent lack of availability and time required for assessment), in direct comparisons with 3D ultrasound it was considered to be at least as accurate [13]. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and overall accuracy for each methodology are presented in Table 5.1.
Diagnostic method | Studies (n) | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | Accuracy (%) |
---|---|---|---|---|---|---|
Hysteroscopy Laparoscopy | Used as gold standard | |||||
MRI | At least as accurate as 3D US | |||||
3D US | 11 | 98.3 (95.6–100) | 99.4 (98.4–100) | 99.2 (97.6–100) | 93.9 (84.2–100) | 97.6 (94.3–100) |
SIS | 13 | 95.8 (91.1–100) | 97.4 (94.1–100) | 97.8 (93.3–100) | 94.6 (87.6–100) | 96.5 (93.4–99.5) |
2D US | 9 | 67.3 (51.0–83.7) | 98.1 (96.0–100) | 94.6 (89.4–99.8) | 86.0 (73.7–98.3) | 86.6 (81.3–91.8) |
HSG | 16 | 84.6 (74.4–94.9) | 89.4 (80.0–100) | 83.6 (74.6–92.6) | 89.1 (79.7–98.5) | 86.9 (79.8–94.0) |
In view of these findings, the Thessaloniki Consensus concluded that 3D ultrasound should be considered as the gold standard for diagnosis of female genital anomalies, supplemented by MRI and/or hysteroscopy and laparoscopy for complex or inconclusive cases [6]. This should not come as a surprise, as in addition to the high diagnostic accuracy rates, 3D ultrasound offers other significant advantages such as being non-invasive, time efficient, cost-effective, readily available, objective and highly reproducible [14].
What Is the True Prevalence?
Until recently, the true prevalence of CUA had not been precisely determined. The reported prevalence in different populations was in fact so inconsistent that it was difficult to even ascertain whether it was a common or rare problem [1]. This was predominantly due to investigators using different methodologies with different diagnostic accuracy rates, as described above. However, recent meta-analyses have now managed to control for this bias, and at present, the prevalence (mean (95 per cent CI)) appears to be in the order of 5.5 per cent (3.5–8.5) in the unselected population, 8.0 per cent (5.3–12) in the infertile population, 13.3 per cent (8.9–20.0) in the recurrent miscarriage population and 24.5 per cent (18.3–32.8) in the combined recurrent miscarriage and infertility population [2]. The prevalence of CUA according to subtypes is presented in Table 5.2.
Population | Total | Arcuatea | Septate | Bicorporeal | Didelphysb | Unicornuate | Others |
---|---|---|---|---|---|---|---|
(%) | (%) | (%) | (%) | (%) | (%) | (%) | |
General | 5.5 (3.5–8.5) | 3.9 (2.1–7.1) | 2.3 (1.8–2.9) | 0.4 (0.2–0.6) | 0.3 (0.1–0.6) | 0.1 (0.1–0.3) | 0.1 (0–2.2) |
Infertile | 8.0 (5.3–12.0) | 1.8 (0.8–4.1) | 3.0 (1.3–6.7) | 1.1 (0.6–2.0) | 0.3 (0.2–0.5) | 0.5 (0.3–0.8) | 0.9 (0.4–1.8) |
Recurrent miscarriage | 13.3 (8.9–20) | 2.9 (0.9–9.6) | 5.3 (1.7–16.8) | 2.1 (1.4–3) | 0.6 (0.3–1.4) | 0.5 (0.3–1.1) | 0.9 (0.1–12.6) |
Recurrent miscarriage and infertility | 24.5 (18.3–32.8)* | 6.6 (2.8–15.7) | 15.4 (12.5–19) | 4.7 (2.9–7.6) | 2.1 (1.4–3.2) | 3.1 (2–4.7) | 0.3 (0–2.3) |
a The arcuate uterus does not exist as an entity in the ESHRE/ESGE classification; therefore these anomalies would be expected to fall within either the normal group or the septate group.