The Accuracy of Ultrasound Shear Wave Elastography in the Diagnosis of Adenomyosis

INTRODUCTION

The presence of ectopic endometrial glands and stroma within the myometrium is known as adenomyosis.¹ It is an inner myometrium illness caused by infiltration of the basal endometrium into the underlying myometrium, resulting in smooth muscle hypertrophy and hyperplasia. Although everyone agrees on the basic criteria, the exact depth of endometrial invasion required to make a diagnosis is still up for debate.² They went on to say that the majority of articles relate to a depth of about 3 mm or one low power field below the basal endometrial layer.

According to several research, the prevalence of adenomyosis ranges from 10% to 70%.³ This large range is owing to a lack of preoperative diagnostic tools. None of the common symptoms associated with uterine adenomyosis, such as abnormal uterine haemorrhage, secondary dysmenorrhoea, and an enlarged painful uterus, are pathognomonic for this illness.² Because adenomyosis is usually combined with other pelvic disorders, it might be difficult to isolate symptoms to this condition.

The majority of uterine adenomyosis diagnoses and treatments are hysterectomy. Until a safe and consistent way of diagnosing this illness is developed, adenomyosis can only be diagnosed surgically by removing the uterus. ²

A more accurate estimate of the prevalence of uterine adenomyosis can help us comprehend the disease's impact, identify women at risk of developing the condition, and promote the development of preventive methods and successful treatment.⁴ In clinical practise, early identification of uterine adenomyosis is critical. Conservative treatment may be ineffective in advanced situations.⁵ Non-invasive or minimally invasive diagnostic approaches are required to avoid needless hysterectomy due to incorrect clinical diagnosis and to study non-surgical alternatives.

The preliminary MRI results for adenomyosis diagnosis were favourable. Togashi et al.⁶ used MRI to correctly diagnose 16 patients of adenomyosis before surgery. According to Agostinho et al.⁷, MRI is an accurate tool for adenomyosis diagnosis, and the most established MRI finding is JZ thickening more than 12 mm. They also mentioned that high-signal intensity myometrial foci on T2- or T1-weighted images are a feature. According to Guarnaccia et al.², the expense of MRI may prevent its widespread use as a diagnostic tool for adenomyosis.

Shear wave elastography is becoming more prominent in the diagnosis of several organs and systemic illnesses.⁸ The growth of adenomyotic foci within the interfacial compartments of connective tissue between the fasicles of hypertrophied smooth muscle cells, accompanied by hyperemia, edoema of perivascular myometrial tissue, and myometrial perifocal hyperplasia around adenomyotic foci, are known to be characteristic changes of the myometrial structures in cases of adenomyosis.

These tissue alterations may alter the stiffness of the myometrium, which can be identified with ultrasonic elastography.³ According to the preliminary findings of Tessarolo et al.⁹, elastosonography could be considered a useful tool in the diagnosis of uterine adenomyosis because it is non-invasive, simple to understand, simple to perform, causes no significant extension of examination time, and has a short learning curve to becoming skilled at the procedure.

This study aims to assess the accuracy of shear wave elastography in the diagnosis of adenomyosis compared to MRI.

PATIENTS AND METHODS

In the current comparative study, 118 premenopausal patients planned for total hysterectomy at El Hussein University Hospital, Obstetrics and Gynecology Department, due to benign pelvic conditions were enrolled.

Inclusion criteria for the study were: women in reproductive age with abnormal uterine bleeding who had failed medical treatment, women with endometriosis and patients having chronic pelvic pain with failed medical treatment for more than six months. Also, some patients included in the study had hysterectomy performed due to CIN 3 or vaginal prolapse. The exclusion criteria were women before age of menarche, or after menopause, fibroid uterus, and women in whom vaginal examination cannot be done (virgin or patient refusal).

Approval of the study was obtained before its initiation. Only the usual consent for hysterectomy was needed. No consent for entering the study.

Six cases were excluded from the study due to postponed surgery; 4 cases were excluded because patients refused operation. They changed their decision and asked for Mirena insertion. 8 more cases were excluded because the operation was done before completing the imaging procedures.

The remaining hundred patients comprised the study group. Each one of the study patients had TVS, SWE and MRI done between day 7 and 14 of the menstrual cycle and the hysterectomy operation was done within 2 months from imaging.

Total hysterectomy was done abdominally in 78 cases and vaginally in 22 cases. One stitch was put at each cornu on anterior uterine wall for orientation, the uterus was bisected longitudinally, fixed in formalin, and sent to pathology department.

TVS and SWE were done in El Hussein University Hospital, Ultrasound and Special Care Unit for the Fetus Department by use of Samsung HS60. Uterine adenomyosis was diagnosed, according to Sun et al., in the presence of one or more of the following sonographic features: subendometrial echogenic linear striations, intramyometrial cysts 2-7 mm in diameter, a heterogenous myometrial echo- texture, poor definition of endometrial myometrial interface, globular uterus or myometrial antero - posterior asymmetry.

The shear wave elastography mode was triggered after the transvaginal sonographic examination. The process was carried out as suggested by Acar et al.³. The transducer was placed vaginally at a depth of three centimetres. Scanning was performed without the use of additional compression by moving the hand and transducer. Color mapping and determining the Young's modulus value were performed in a 5 mm diameter area of interest (AOI). In the absence of suspected pathology, the ultrasonography area of interest was placed in the anterior uterine wall. Adenomyosis was diagnosed if the Young's modulus value was greater than the cut off value of 34.6 Kpa proposed by Acar et al.

Adenomyosis was identified by magnetic resonance imaging when one or more of the following criteria were present: JZmax equal or greater than 12 mm, JZdiff greater than 5 mm, and JZmax / total myometrial thickness greater than 40%.7 The MRI protocol used was that proposed by Agostinho et al. 7. T2-WI sagittal, axial, and coronal planes were used in this protocol, as well as T1 3D fast suppressed axial and coronal planes. Contrast, as proposed by Acar et al., was not required for suspected adnexal lesions.

The histopathologist was asked to check for uterine adenomyosis without knowing the imaging results (no comment was needed for extent or severity of the lesion). Fundus, anterior wall, posterior wall, left lateral wall, right lateral wall, and macroscopically questionable myometrial region histopathological sections were investigated. If ectopic endometrium was detected at 2.5 mm or greater from the basal endometrium, uterine adenomyosis was identified.¹⁰

The histopathological findings were utilised to determine the prevalence of uterine adenomyosis in the study population as well as the values of risk factors previously identified (age, gravidity, parity, history of uterine surgery, smoking, depression, oral contraceptive usage, and IUD insertion).

The results of histological examination were also utilised as a reference to investigate the diagnostic usefulness of the procedures used to diagnose uterine adenomyosis: TVS, SWE, and MRI.

The elastogram demonstrates a soft adenomyotic lesion (red) surrounded by rigid myometrial tissue (blue).

Statistical Analysis:

Data were analysed using IBM c SPSS c Statistics version 26 (IBM c Corp., Armonak, NY) and Version 2016.02.28451 (Addinsoft c, Paris, France). Diagnostic accuracy of TVS, MRI or SWE is calculated using histopathology as the gold - standard for diagnosis. The following diagnostic indices are calculated: sensitivity, specificity, positive and negative predictive values, positive and negative likelihood ratios and correct classification and misclassification rates. The McNemar test is used to compare sensitivities or specificities of different modalities.  Inter - method agreement is examined using Cohen Kappa coefficient (k), Scott's bias - adjusted kappa coefficient (BAK) and Bennett's prevalence and bias - adjusted Kappa coefficient (PABAK). The coefficients of agreement (k, BAK and PABAK) are interpreted as follows: values less than 0.5 are indicative of poor agreement, values between 0.5 and 0.75 indicate moderate agreement, values between 0.75 and 0.9 indicate good agreement and values greater than 0.9 indicate excellent agreement. P < 0.05 is considered statistically significant.

RESULTS

*. Mean ± SD (minimum to maximum) = 46.8 ± 3.1 (42 to 52 years).

Table 1: Characteristics of the study population.

Table (1) shows the clinical characteristics of the study population. The clinical presentations (AUB, chronic pelvic pain and infertility) which are known to be present in cases with uterine adenomyosis. Chronic pelvic pain was present in 44% AUB was present in 34% and infertility was present in zero percent. The patients without any of these complaints comprises 46% of the study population.

Data are mean ± SD or number (%). NA = test not applicable.

*. Chi-squared test unless otherwise indicated. †. Independent samples t-test.

‡. Linear by linear association.

Table 2: Comparison of clinical variables in patients with or without adenoyosis

The clinical variables in patients with adenomyosis versus those without are shown in table (2). The only variables that have significant differences are AUB, chronic pelvic pain and asymptomatic cases.

Table 3: Results of TVUS, MRI, SWE and histopathology.

Table 4: Diagnostic accuracy of TVUS evaluated versus histopathology as the gold-standard test.

Table (4) shows the diagnostic accuracy of TVS evaluated versus histopathological diagnosis of uterine adenomyosis as the gold standard test. Sensitivity is 78%, specificity is 96%, PLR is 17.9 and NLR is 0.23.

Table 5. Sensitivity is 89%, specificity is 87%, PLR is 6.8 and NLR is 0.13.

Table 5: Diagnostic accuracy of MRI evaluated versus histopathology as the gold-standard test

The diagnostic accuracy of MRI evaluated versus histopathological examination is shown in

Table 6: Diagnostic accuracy of SWE evaluated versus histopathology as the gold-standard test.

Table (6) shows the diagnostic accuracy of SWE versus histopathological examination to diagnose uterine adenomyosis. Sensitivity is 93%, specificity is 91%, PLR is 10.6 and NLR is 0.08.

Fig. 1: Diagnostic accuracy of TVUS, MRI and SWE evaluated versus histopathology as the gold-standard test. Error bars represent the 95% confidence limits.

The diagnostic accuracy of TVS, SWE and MRI evaluated versus histopathological examination are shown in figure (1). The diagnostic accuracies of the three modalities are comparable.