Venous Thromboembolism and Adenomyosis: A Retrospective Review

Adenomyosis is a common benign condition of the uterus, characterized by the presence of endometrial glands and stroma in the myometrium. Similar to fibroids, adenomyosis tends to affect women between 40 and 50 years old and is commonly acknowledged to cause dysmenorrhea and heavy menstrual bleeding.[ 1 ] Interestingly, literature search has also yielded isolated case reports describing patients with adenomyosis and thromboembolic events such as cerebral infarcts and deep-vein thrombosis (DVT).[ 2 3 4 ] While a definite association between adenomyosis and venous thromboembolism (VTE) has not been established, several theories have been proposed to explain the potential link between these two conditions.[ 1 5 ] Treatment of these patients also poses a challenging medical dilemma. The presence of both heavy menstrual bleeding and thromboembolism means that the treatment of one aggravates the other. Anticoagulation would predispose the patient to worsening menorrhagia, while hormonal or antifibrinolytic therapy for menorrhagia is contraindicated if there is an active VTE. In this case series, we set out to determine the incidence of VTE in patients who were admitted with adenomyosis in our institute and to describe their clinical characteristics and management.

A total of 41 patients were included in this study. The mean age was 44.1 ± 5.18 years old, while the mean BMI was 27.9 ± 7.50 kg/m 2 . Thirty-seven (87.8%) out of the 41 patients had concurrent abnormal uterine bleeding and 5 (12.2%) out of the 41 patients had concurrent VTE. These five patients' clinical characteristics are shown in Table 1 . Clinical details of the five patients with venous thromboembolism VTE: Venous thromboembolism, PE: Pulmonary embolism, DVT: Deep vein thrombosis, COCP: Combined oral contraceptive pill, GnRH: Gonadotropin-releasing hormone, UAE: Uterine artery embolization, IVC: Inferior vena cava, HB: Hemoglobin Interestingly, all five patients had VTE in the form of pulmonary embolism (PE), while two of them also had a concurrent left lower limb DVT. All five patients had an enlarged uterus and abnormal bleeding due to adenomyosis. All five patients with VTE were referred to a hematologist for management and in all but one patient (patient 4), the VTE was deemed to be provoked by adenomyosis. Patient 4's development of PE was likely related to her underlying condition of chronic thromboembolic pulmonary hypertension, which probably developed as a result of previous undiagnosed PE. This patient did not give a history of a previous acute VTE episode; hence, it was likely to have been a silent event. However, it would have been most likely precipitated by her enlarged adenomyotic uterus and the use of oral contraceptive pills, as those were her only risk factors. Anticoagulation was commenced promptly for all five patients; four of them received warfarin and one was given rivaroxaban, a direct factor Xa inhibitor. Three out of the five patients developed worsening menstrual bleeding on starting anticoagulation and were treated with gonadotropin-releasing hormone (GnRH) analogs, which induced a period of amenorrhea during anticoagulant therapy, while also having the additional benefit of reducing uterine volume. This approach was effective for two patients who were both on warfarin. However, the last patient on rivaroxaban had persistent menorrhagia despite GnRH analogs and uterine artery embolization and eventually required a hysterectomy in the same admission. Four of the five patients eventually underwent a hysterectomy for long-term management of adenomyosis. The remaining patient (patient 4) had opted to return to her home country for further management. A comparison of the clinical characteristics between patients who had adenomyosis with VTE and patients who had adenomyosis without VTE is shown in Table 2 . There were no statistically significant differences between the groups in terms of age, BMI, hemoglobin, hematocrit, or weight of the adenomyotic uterus (for women who had a hysterectomy). There was also no difference in the number of women taking combined oral contraceptive pills or tranexamic acid. Comparison of clinical characteristics between venous thromboembolism and no venous thromboembolism groups using independent samples t -test Z -scores were used to compare percentages. Statistical significance is defined by P <0.05. BMI: Body mass index, VTE: Venous thromboembolism, NA: Not available From our results, it can be concluded that VTE in concurrence with adenomyosis is uncommon and there appears to be a greater occurrence of PE than DVT. However, a direct causation cannot be drawn between both conditions as of now.

A direct link between VTE and adenomyosis has not yet been established and it is not known if there is a causal relationship between both conditions. Due to its rare occurrence, literature search has produced only isolated case reports with a wide variety of clinical manifestations, as shown in Table 3 . Hence, the estimated incidence is not yet known. From our knowledge, this is the first case series on this topic. Summary of case reports of venous thromboembolism in association with adenomyosis COCP: Combined oral contraceptive pill, GnRH: Gonadotropin-releasing hormone, DIC: Disseminated intravascular coagulation, DVT: Deep vein thrombosis Several theories have been proposed to explain a link between adenomyosis and VTE. Conventionally, it was thought that extrinsic pressure on the inferior vena cava caused by an enlarged uterus resulted in venous stasis and, hence, increased risk of lower limb DVT.[ 5 ] In addition, women with adenomyosis commonly also have menorrhagia, which has been shown to induce a hypercoagulable state by causing polycythemia and reactive thrombocytosis.[ 1 ] However, there was no significant difference in terms of the weight of the uterus or hematocrit between the patients in our study who had VTE and those who did not. Furthermore, it is interesting to note that all five of our patients with adenomyosis and VTE had PE, while only two of them had DVT. There were otherwise no other risk factors for VTE in these patients. Hence, the development of PE in these patients cannot be sufficiently explained by the theories listed above. Alternative theories have been proposed by other authors. A number of case reports have described the development of disseminated intravascular coagulation in association with adenomyosis. This manifested as end-organ injury in the form of cerebrovascular thrombosis, acute renal failure, and nonbacterial thrombotic endocarditis in patients with no other thrombotic risk factors.[ 10 11 12 ] This combination of disseminated intravascular coagulation with menorrhagia during menstruation in some patients with adenomyosis can be attributed to concurrent dysfunctional coagulation and fibrinolysis. Yamanaka et al . described how magnetic resonance imaging detected multiple signal hyperintense spots inside adenomyotic tissues, which likely represented ectopic endometrium or microhemorrhage in the myometrium. They speculated that this then alters the coagulation system through local inflammation and microthrombi formation, which can embolize into the systemic circulation to cause embolic phenomena.[ 13 ] The treatment of these patients with concurrent problems of bleeding and thrombosis is also challenging. In the setting of an active VTE, antifibrinolytic therapy and hormonal therapy, common management options for menorrhagia, are usually contraindicated. GnRH analogs were chosen for our patients, as a single dose could achieve a sustained period of amenorrhea, hence, preventing the recurrence of menorrhagia until anticoagulation was established and the thrombus resolved. They were effective in all of our patients except for one (patient 2), who was also the only patient on rivaroxaban. This corresponds with literature, where rivaroxaban has been shown to result in more cases of abnormal bleeding than low-molecular-weight heparin or Vitamin K antagonists.[ 14 ] An alternative treatment option is the levonorgestrel intrauterine system. However, rates of spontaneous expulsion are higher in women with an enlarged uterus due to adenomyosis or fibroids, and hence, this may not be effective in a significant proportion of patients with adenomyosis. One of the limitations of this study is the small number of patients included, as we have only included patients with adenomyosis who were admitted, while those managed in the outpatient setting were not included in this study. Hence, although there is no previously quoted figure to compare this to, the incidence of VTE in association with adenomyosis at our institution (12.2%) is likely an overestimation. Further studies are needed to determine if there is an association between VTE and adenomyosis. Nevertheless, as the first case series on this topic, we hope that this collection of cases would help to increase the awareness of a potential link between VTE and adenomyosis, the challenges in managing these patients and also effective treatment approaches.

We conclude that the occurrence of VTE in patients with adenomyosis is uncommon and is not well understood for now; more studies are needed to determine if there is a causal association. Nil. There are no conflicts of interest.

This study was performed at National University Hospital, a tertiary hospital in Singapore, and ethics board approval was obtained (DSRB 2016/01170, approval date February 9, 2017). Waiver of informed consent was approved by the ethics board. A retrospective review of the medical records of all patients who were admitted to the gynecology ward between January 2015 and August 2016 was conducted and all patients who had adenomyosis were included in this study. Adenomyosis was defined by sonographic findings of an adenomyoma, coarsened echotexture of the myometrium, asymmetrical myometrial thickening, or myometrial or subendometrial cysts. Clinical details that were evaluated included age, parity, body mass index (BMI), significant comorbidities, size of the uterus on physical examination, size of the adenomyoma (if present) on pelvic ultrasonography, hemoglobin and hematocrit on admission, any associated VTE, and treatment given. Descriptive statistics were performed with continuous variables being presented as means and standard deviations. Categorical data were compared using Z-scores. Between-group means were compared using independent samples t -test. Means with multiple groups were compared using the Kruskal–Wallis test. Statistical significance was set at P < 0.05. All statistical analyses were performed with SPSS version 21 (IBM Inc., Armonk, New York, USA).