Lisa Trommelen

ENGLISH SUMMARY

Part 1
In part one, we discuss the diagnosis of adenomyosis and focus on ultrasound diagnosis. Previously, the international Morphological Uterus Sonographic Assessment (MUSA) consensus group suggested reporting on specific direct and indirect features. Direct features of adenomyosis are signs of ectopic endometrial glands and stroma in the myometrium: subendometrial lines and buds, hyperechogenic islands, and intra-myometrial (micro-)cysts. Indirect features are the consequence of the ectopic endometrial glands: hypertrophic myometrium resulting in globular uterus or asymmetric myometrium; cysts or vessels reflecting echo sound beams resulting in fan-shaped shadowing, and intralesional vascularity, and interrupted or irregular junctional zone (JZ). These features resulted in more consistent research and reporting; however, some questions were still left unanswered. In part 1, we aim to unravel some of these questions by looking into the reason for different appearances on sonography (explained by histological findings or pathogenetic pathways) and by finding a more objective method to assess the severity of adenomyosis.

Chapter 2
The junctional zone is the inner subendometrial layer of the myometrium and is thought to play an essential role in the pathogenesis of adenomyosis. It is, however, not a well-defined entity across different diagnostic disciplines. In histology, the JZ exists out of gradual cellular and vascular changes towards the outer myometrium, while on imaging, a distinction between the JZ and the outer myometrium is found. Additionally, the JZ of the same patient during the same cycle phase is often visualized thicker on MRI than TVS. Given the inconsistencies among the imaging modalities, the thickness of the JZ cannot remain the gold standard for diagnosing adenomyosis. Therefore, the focus should be on markers for adenomyosis visible in all modalities, the so-called direct signs of ectopic endometrium in the JZ, or signs of myometrium affected by adenomyosis. On histology, these are spots of endometrial glands and stroma present in the myometrium, which are visualized on transvaginal sonography (TVS) as sub-endometrial lines and buds and hyperechogenic islands and cysts in the myometrium or as an irregular JZ or JZ interrupted by invaginated endometrium, and on MRI as bright high-intensity foci interrupting the JZ or myometrium. A thickened junctional zone might be explained by smooth muscle cell hypertrophy as an indirect sign of adenomyosis.

Chapter 3
Besides the different appearance of the junctional zone in adenomyosis with different diagnostic modalities, other morphologies, and tissue characteristics are also reported in adenomyosis. We reviewed pathways reported in the pathogenesis and progression of adenomyosis to understand what mechanisms take place, possibly resulting in different tissue characteristics. We hypothesized that the sonographic features reflect four histological appearances: the presence of glandular, microcystic, vascular and fibrotic tissue. These can be present simultaneously or separately and with varying grades of predominance, creating a continuous spectrum of various adenomyosis phenotypes. These morphological characteristics might be better visualized by adding new modalities to conventional 2D- and 3D ultrasonography: slow-flow and elastography. Slow-flow visualizes microvascularity, while elastography visualizes the softness of tissue in comparison to the surrounding tissue. We speculated on the different appearances of ultrasound and histology and the potential impact on symptomatology and responsiveness to minimally invasive or medical therapy.

Incorporating phenotype characterization- using 2D and 3D ultrasound, color Doppler, MVFI, and elastography- into future extensions of the MUSA reporting framework could enhance future research.

Chapter 4
In Chapter 4, we aimed to find a method to report on the sonographic severity of adenomyosis more objectively. We hypothesized that not the number of MUSA features present but the extent of the adenomyotic myometrium is important in assessing severity and correlation with symptomatology (e.g., dysmenorrhea, abnormal uterine bleeding, and subfertility). The MUSA consensus group suggested categorizing the extent of adenomyosis by an estimation into mild (<25% affected), moderate (25%-50% affected), or severe (>50% affected). This categorization was not validated and was very subjective. Therefore, we aimed to find a more objective way to semi-quantify the severity of adenomyosis, which was feasible in use and with good interobserver reliability. We collected 30 uterine volumes with adenomyosis (at least one direct MUSA feature present) and assessed the volumes offline, blinded for the evaluation at the outpatient clinic. We developed six potential offline methods using 5D-viewer. Of these six methods, only three were feasible regarding technique, time, and interpretation. These three methods were consecutively assessed for interobserver reliability. Eventually, one method, which included taking 20 transversal slices out of the uterine volume and evaluating each slice as affected (≥1 direct MUSA feature present) was feasible and had good interobserver reliability: the XI VOCAL counting method. XI VOCAL counting appears promising in future research, making the assessment of adenomyosis more objective and consistent.

Chapter 5
The method developed and evaluated in Chapter 4, XI VOCAL counting, was further utilized in Chapter 5 to investigate the relationship between the sonographic severity and the severity of symptoms. An observational study was performed, and 115 patients were prospectively included at the Amsterdam UMC gynecological outpatient clinic. Dysmenorrhea, chronic abdominal/pelvic pain, and dyspareunia were assessed with a numeric rate scale (NRS: 0-10), and abnormal menstrual bleeding with a three-point Likert scale and scored with a pictorial blood assessment chart (PBAC: 0- ∞). A positive correlation was found between both XI-VOCAL counting and XI-VOCAL categories with the severity of menstrual blood loss. XI-VOCAL counting and XI-VOCAL categories did not correlate with the severity of dysmenorrhea, chronic abdominal/pelvic pain, and dyspareunia. The subjective real-time severity assessed by clinicians at outpatient clinic visits showed a statistically significant correlation with PBAC and subjective blood loss. The applicability of XI VOCAL counting in the outpatient clinic should be further investigated, including assessing the affected layer(s), and uterine volume. Also, the relation between sonographic severity and subfertility and miscarriages is interesting for further research.

Part 2
In part 2, we investigated uterine artery embolization as an alternative to hysterectomy for patients with symptomatic adenomyosis when conservative therapy fails. This study was part of the multicenter QUality of life after Embolization versus hySTerectomy for symptomatic Adenomyosis (QUESTA) study, a two-year prospective cohort study in 12 centers in the Netherlands that initially started as a randomized controlled trial.

By comparing UAE and hysterectomy, we aimed to answer whether UAE is a promising and less-invasive alternative for hysterectomy because comparative data has been lacking up to now. In Chapter 6, we reported on the primary objective after one year: non-inferiority of embolization versus hysterectomy regarding health-related quality of life (HRQOL). HRQOL is assessed using validated questionnaires SF-12 and WHO-QoL-Bref. The WHO-QoL-Brev exists out of four domains: physical, psychological, social relations, and environment. The SF-12 is divided into a mental component summary score (MCS) and a physical component summary score (PCS). In chapter 7, the peri-procedural, short-term, and recovery outcomes were discussed. In Chapter 8, we described the myometrial change on MRI six months after UAE and explored baseline imaging variable as potential predictors for this change, which might be used in counseling patients at the outpatient clinic.

Chapter 6
After a one-year follow-up, neither non-inferiority nor inferiority of UAE versus hysterectomy for HRQOL in symptomatic adenomyosis could be established within the predefined margins. Both UAE and hysterectomy significantly improved HRQOL for patients with symptomatic adenomyosis. Nevertheless, adjusted means of the SF-12 physical and mental component summary score (PCS and MCS) at one year were statistically significantly higher in the hysterectomy group compared to the UAE group; the other HRQOL questionnaire (WHO-QOLBrev) resulted in comparable outcomes between groups. Both UAE and hysterectomy improved sexual activity and pain, although hysterectomy had a more significant effect on pain reduction. UAE performed better regarding symptoms of urinary frequency and incontinence. More participants were satisfied after hysterectomy. Even though hysterectomy performs better in several domains, UAE is a good alternative in adenomyosis patients. It can be offered to patients who do not want or are not suitable to undergo radical surgery or who want to preserve their uterus. For patients seeking a definitive solution, hysterectomy remains the treatment of choice.

Chapter 7
The periprocedural and short-term complication rates were minor and comparable between groups. Hospital stay after hysterectomy was shorter than after UAE (1.40 versus 1.85 days; p=.008); hysterectomy patients scored better on the recovery index (hysterectomy: 39.4 versus UAE: 34.5, p=.009), and reflected more positively on their perceived procedure. UAE favored faster procedure time (60 minutes versus 109 minutes, p<.001) and less blood loss (0 ml versus 50 ml, p<.001), and patients returned faster to their daily activities and work. For example, patients returned to work 17.5 days after UAE and 43.8 days after hysterectomy (p<.001). Patient-controlled intravenous analgesia for UAE resulted in shorter hospital stay than patient-controlled epidural analgesia and may be considered standard clinical practice. This study assists patients with symptomatic adenomyosis and their doctors in evidence-based decision-making regarding UAE and hysterectomy by weighing short-term outcomes. Chapter 8 This study analyzed the 50 participants undergoing UAE as part of the QUESTA study. It focused on post-procedural MR imaging characteristics, predictors for MRI change, and the relation with improvement in quality of life. UAE resulted in a significant decrease in uterine volume at six months compared to baseline (from 254.2 cm3 to 203.8 cm3, p<.001), adenomyosis volume (from 75.4cm3 to 31.4 cm3, p<.001), and JZ thickness (from 21mm to 17mm, p=.0013). Uterine volume at baseline positively impacted the rate of infarction (exp(β) = 1.0074, 95% CI 1.0001 - 1.0148). High-intensity foci (representing myometrial cysts) were related to a higher residual uterine volume (exp(β) 1.351, 95% CI: 0.947 – 1.912) and a higher residual adenomyosis volume (exp(β) 1.885, 95% CI: 0.967 – 3.572) at six-months. In our small sample, a correlation between infarction rate, reduction of uterine volume, adenomyosis volume or JZ thickness, and health-related quality of life at 6 months follow-up could not be demonstrated. General conclusion This thesis contributes to the further development of both the diagnosis and treatment of adenomyosis, with an emphasis on more personalized and evidence-based care. The results show that adenomyosis should not be considered a binary condition, but rather a continuous spectrum of different phenotypes, characterized by diverse histological and ultrasound features such as glandular, microcystic, vascular, and fibrotic patterns. This approach can contribute to a better classification of adenomyosis, more uniform study populations, and ultimately more targeted therapies. Additionally, the development of the 'XI VOCAL counting' method offers a promising, semi-quantitative way to more objectively determine the severity of adenomyosis. In the therapeutic field, the QUESTA study shows that both hysterectomy and uterine artery embolization (UAE) lead to a clear improvement in the quality of life in patients with symptomatic adenomyosis. Although hysterectomy showed better results in several domains, UAE is a valuable less-invasive alternative for patients who wish to preserve their uterus or want to avoid surgery. Good counseling, tailored to individual preferences and expectations, remains essential here. Future research should focus on further validation of adenomyosis phenotypes, application of advanced imaging techniques, integration of artificial intelligence, and the development of new therapeutic strategies. International cooperation and larger study populations are necessary to take further steps towards personalized care for patients with adenomyosis.