Comparison of ovarian vein embolization alone with pelvic venous plexus embolization in the treatment of pelvic venous disorders at 12-month follow-up

Conception and design: HK, CA Analysis and interpretation: HK Data collection: HK Writing the article: HK, CA Critical revision of the article: HK, CA Final approval of the article: HK, CA Statistical analysis: Not applicable Obtained funding: Not applicable Overall responsibility: HK

None.

This retrospective observational cohort included 72 patients diagnosed with PeVD between April 2022 and April 2024 who subsequently underwent PEE in our clinic. The primary outcome was symptom relief at 12 months assessed by the visual analog scale (VAS). The secondary outcomes were symptom recurrence or complications. The study was conducted at a single institution and approved by the Üsküdar University Non-Interventional Research Ethics Committee (approval number: 61351342/020-54) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained for the procedure; consent for publication was waived due to anonymized data. Demographic and clinical data were retrieved from patient records and the hospital database. Of 89 patients who were diagnosed with PeVD and underwent PEE, 72 patients aged between 20 and 45 years were selected for inclusion. The diagnosis of PeVD was based on both clinical and radiological signs. Pain symptoms in the pelvic region lasting more than 6 months were defined as CPP. 1 All patients underwent a gynecological checkup to rule out nonvascular causes of CPP, and transabdominal or transvaginal Doppler ultrasound examination was performed as the initial imaging modality. Radiological assessment primarily focused on ovarian vein dilation (≥6 mm) and the presence of retrograde venous flow, which were considered the main diagnostic features of PeVD. 8 Reduced flow velocity was considered supportive and interpreted alongside vein diameter, reflux patterns, and clinical presentation. To evaluate for potential upstream venous outflow obstruction, all patients underwent cross-sectional imaging with computed tomography or magnetic resonance venography as part of the diagnostic workup. These examinations were systematically reviewed for morphologic features suggestive of venous compression, including focal luminal narrowing, side-to-side asymmetry, collateral venous pathways, and, for the LRV, aortomesenteric narrowing, beak sign, altered aortomesenteric angle, and abnormal hilar-to-compressed segment diameter ratios. Patients with imaging findings consistent with LRV or iliac vein compression were excluded from the study. In addition to CPP, dysmenorrhea, dyspareunia, and postcoital pain were also assessed with VAS before the procedure and 12 months after treatment. Exclusion criteria included patients with vascular compression syndromes such as LRV and/or left common iliac vein compression, gynecological diseases such as endometriosis, pelvic inflammatory disease, any kind of systemic disease, gastrointestinal diseases, psychosomatic diseases, neuromuscular diseases, a history of prior abdominal or pelvic surgery or radiotherapy, and loss to follow-up. Although several exclusion criteria were prespecified, only vascular compression (n = 3), systemic disease (n = 1), and loss to follow-up (n = 13) were encountered in this cohort ( Fig 1 ). Fig 1 Flowchart of patient inclusion and group allocation in the study. OVEA , Ovarian vein embolization alone; PCS , pelvic congestion syndrome; PVPE , pelvic venous plexus embolization. Flowchart of patient inclusion and group allocation in the study. OVEA , Ovarian vein embolization alone; PCS , pelvic congestion syndrome; PVPE , pelvic venous plexus embolization. All procedures were performed by two vascular surgeons with >10 years of experience. Under local anesthesia, venous access was obtained via the femoral or right brachial vein using a 6F introducer (Terumo). A 0.035-inch-diameter hydrophilic guidewire and 5F guiding catheters (Cobra, Terumo, or Simmons1, Cordis Johnson & Johnson) were used to target relevant venous structures. During venography, the LRV and left common iliac vein were evaluated to further exclude venous outflow obstruction. Pressure gradients were measured across these segments under standardized resting conditions using catheter-based manometry and interpreted as supportive rather than standalone diagnostic criteria. Renocaval pressure gradients <1 mm Hg and iliac venous gradients <2 mm Hg were considered within physiological limits. 9 Patients demonstrating pressure gradients exceeding these thresholds were conservatively excluded unless obstruction was unequivocally ruled out by concordant cross-sectional imaging and venographic findings. Bilateral ovarian veins were evaluated with venography, and segments with reflux were determined. During venography, imaging was supported by applying the Valsalva maneuver to make the reflux more apparent. PEP, both ipsilateral and contralateral, were evaluated if present. PEP were defined as collateral venous connections originating from refluxing ovarian or iliac veins and directing to the parametrial, uterine, or contralateral pelvic venous plexus (PVP). To reduce procedural heterogeneity, the selection of PVPE vs OVEA was based on predefined venographic findings. PVPE was performed when ipsilateral or contralateral PEP and/or PVP segments were clearly opacified on venography and could be selectively catheterized in a reproducible and safe manner. In contrast, OVEA was deliberately chosen when venography demonstrated reflux confined to the gonadal axis without visualization of targetable PEP or when the PVP could not be selectively accessed despite a standardized imaging technique and Valsalva augmentation. The degree of reflux was evaluated with the Hiromura classification. 10 This classification was used solely to standardize the anatomical extent and distribution of ovarian vein reflux during venographic assessment and was not applied as a diagnostic criterion or as a surrogate for symptom severity. The target veins were selectively cannulated with a microcatheter (Renegade, Boston Scientific, or Rebar, Medtronic), and the embolization procedure was initiated. Reflux originating from the internal iliac venous system was defined by selective venography demonstrating retrograde opacification of the PVP or parametrial veins via internal iliac venous tributaries, with or without cross-pelvic filling. Isolated internal iliac venous reflux without demonstrable PEP was not observed in this cohort and therefore was not treated separately from PVP involvement. Balloon occlusion testing was not routinely performed; embolization decisions were based on fixed venographic reflux patterns and reproducible visualization of pelvic escape pathways rather than temporary flow interruption maneuvers. In the PVPE approach, platinum-based detachable coils (Interlock, Boston Scientific, or Concerto, Medtronic) of appropriate sizes, depending on the targeted veins, were placed starting from distal to proximal, first to occlude all ipsilateral or contralateral PEP detected and subsequently to embolize the distal and midportions of the ovarian vein to ensure complete elimination of reflux ( Fig 2 ). In some cases, in patients in whom complete occlusion could not be achieved with coils or to reduce coil burden, complementary embolization was performed with a liquid embolizing agent (Onyx; Medtronic). Fig 2 (A) Venography demonstrating pelvic venous congestion. (B) Control venography after pelvic venous plexus embolization (PVPE) showing complete occlusion. (A) Venography demonstrating pelvic venous congestion. (B) Control venography after pelvic venous plexus embolization (PVPE) showing complete occlusion. In the OVEA approach, only the distal and middle segments of the ovarian vein were embolized with platinum-based detachable coils from distal to proximal. Onyx was used for complementary embolization when necessary. After each embolization, control venography was performed under the same acquisition conditions as the initial assessment to confirm occlusion of the embolized venous segments and the absence of immediate retrograde flow within the treated axis. Postembolization syndrome (pelvic pain, tenderness, or fever) was treated with nonsteroidal anti-inflammatory drugs. 11 The patients were kept under observation for a few hours and then discharged the same day. A structured symptom assessment was performed before PEE (T0), recording VAS scores (0-10) for CPP, dysmenorrhea, dyspareunia, and postcoital pain. All patients were called to the outpatient clinic for control purposes at 1 (T1), 3 (T3), 6 (T6), and 12 (T12) months after the procedure. During these visits, the general clinical condition of the patients, changes in their symptoms, possible complications that may arise after the procedure, and newly developing complaints, if any, were evaluated and recorded in detail. Although patients were routinely examined at T1, T3, T6, and T12, VAS scores were specifically recorded at T0 and T12 to evaluate midterm outcomes. Bladder symptoms (frequency, urgency, and nocturia) were recorded as present or absent. Recurrence was defined as the reappearance or worsening of symptoms during the 12-month follow-up at any time point. Patients presenting with persistent or worsening symptoms underwent transvaginal Doppler ultrasound examination to investigate findings suggestive of reflux or pelvic congestion. Evaluations were conducted by the treating vascular team using a standardized follow-up form. No additional medical therapy was administered. Treatment efficacy was compared using VAS at T12. Statistical analyses were performed using IBM SPSS Statistics for Windows, version 25.0 (Statistical Package for the Social Sciences; IBM Corp). Descriptive statistics were presented as n (%) for categorical variables and mean ± SD and median (min-max) for continuous variables. The normality assumption was evaluated with the Kolmogorov-Smirnov test ( P < .05), and the Mann-Whitney U test was used for paired group comparisons. Study groups were comparable at baseline, rendering unadjusted tests unnecessary. The Pearson χ 2 test and the Fisher exact test were used for comparing categorical variables. All 72 patients completed 12-month follow-up; therefore, no missing data were present. P < .05 was considered statistically significant.

In one patient, minimal ovarian vein rupture occurred during control venography after PEE. One patient experienced a localized hematoma in the right groin that did not require intervention and was treated conservatively. No migration was observed ( Table I ). Table I Data on demographic and clinical characteristics Variables Values Age, years  Mean ± SD 34.54 ± 5.43  Median (min-max) 35.0 (22-44) BMI, kg/m 2  Mean ± SD 27.28 ± 4.72  Median (min-max) 26.45 (19.5-39.5) Parity, No. (%)  1 14 (19.4)  2 15 (20.8)  3 23 (31.9)  4 14 (19.4)  5 6 (8.3) Increase in CPP at the end of the day, No. (%)  Yes 53 (73.6) Increase in CPP with physical activity, No. (%)  Yes 55 (76.4) Increase in CPP at the premenstrual period, No. (%)  Yes 42 (58.3) Dysmenorrhea, No. (%)  Yes 22 (30.6) Dyspareunia, No. (%)  Yes 28 (38.9) Postcoital pain, No. (%)  Yes 43 (59.7) Bladder symptoms, No. (%)  Yes 9 (12.5) Chronic venous insufficiency (lower limbs), No. (%)  Yes 38 (52.8) Reflux grade (Hiromura classification), No. (%)  Grade 2 36 (50.0)  Grade 3 36 (50.0) Venous access site, No. (%)  Right femoral vein 53 (73.6)  Right brachial vein 17 (23.6)  Left femoral vein 2 (2.8) Embolization material, No. (%)  Coil + Onyx 50 (69.4)  Coil 22 (30.6) Embolization approach, No. (%)  PVPE 34 (47.2)  OVEA 38 (52.7) Complications, No. (%)  Yes 2 (2.8) Hematoma, No. (%)  Yes 1 (1.4) Rupture, No. (%)  Yes 1 (1.4) Recurrence, No. (%)  Yes 11 (15.3) Postembolization syndrome, No. (%)  Yes 21 (29.2) Symptom duration until diagnosis, months  Mean ± SD 34.98 ± 9.03 BMI , Body mass index; CPP , chronic pelvic pain; OVEA , ovarian vein embolization alone; PVPE , pelvic venous plexus embolization; SD , standard deviation. Data on demographic and clinical characteristics BMI , Body mass index; CPP , chronic pelvic pain; OVEA , ovarian vein embolization alone; PVPE , pelvic venous plexus embolization; SD , standard deviation. Technical success rate was 100%. We observed that the VAS scores regarding CPP, dyspareunia, and postcoital pain at T12 were statistically significantly lower in the PVPE group ( P = .001, P = .008, P = .005, respectively). Moreover, recurrence rates were found to be lower in the PVPE group ( P = .036) ( Table II ). Table II Comparison of demographic and clinical variables between the OVEA and PVPE groups Variables OVEA (n = 38) PVPE (n = 34) P Age, years  Median (min-max) 34.0 (22-43) 35.5 (23-44) .910 a BMI, kg/m 2  Median (min-max) 26.7 (22.1-35.4) 28.4 (19.7-39.5) .131 a Parity, No. (%)  1 9 (23.7) 5 (14.7) .085 b  2 10 (26.3) 5 (14.7)  3 9 (23.7) 14 (41.2)  4 6 (15.8) 8 (23.5)  5 4 (10.5) 2 (5.9) Increase in CPP at the end of the day, No. (%)  Yes 27 (71.1) 26 (76.5) .603 c Increase in CPP with physical activity, No. (%)  Yes 29 (76.3) 26 (76.5) .988 c Increase in CPP at the premenstrual period, No. (%)  Yes 23 (60.5) 19 (55.9) .690 c CPP, VAS at T0  Median (min-max) 7.5 (5.0-8.0) 7.6 (3.0-8.0) .940 a CPP, VAS at T12  Median (min-max) 2.0 (0-5.0) 1.5 (0-3.0) .001 a Dysmenorrhea, No. (%)  Yes 10 (26.3) 12 (35.3) .409 c Dysmenorrhea, VAS at T0  Median (min-max) 0.0 (0.0-6.0) 0.0 (0.0-6.0) .459 a Dysmenorrhea, VAS at T12  Median (min-max) 0.0 (0.0-6.0) 0.0 (0.0-6.0) .398 a Dyspareunia, No. (%)  Yes 12 (31.6) 16 (47.1) .178 c Dyspareunia, VAS at T0  Median (min-max) 5.0 (0.0-8.0) 0.0 (0.0-8.0) .764 a Dyspareunia, VAS at T12 5.0 (0.0-7.0) 0.0 (0.0-7.0) .008 a  Median (min-max) Postcoital pain, No. (%)  Yes 23 (60.5) 20 (58.8) .833 c Postcoital pain, VAS at T0  Median (min-max) 5.0 (0.0-8.0) 4.0 (0.0-6.0) .488 a Postcoital pain, VAS at T12  Median (min-max) 4.0 (0.0-8.0) 0.0 (0.0-7.0) .005 a Presence of bladder symptoms at T0, No. (%) 4 (11.8) 5 (13.2) .727 b Presence of bladder symptoms at T12, No. (%) 3 (8.8) 4 (10.5) .700 b Chronic venous insufficiency (lower limbs), No. (%)  Yes 22 (57.9) 16 (47.1) .358 c Ovarian vein diameter, mm  Median (min-max) 9.2 (6.5-14.6) 8.8 (6.4-13.2) .680 a Reflux grade, No. (%)  Grade 2 22 (57.9) 14 (41.2) .157 c  Grade 3 16 (42.1) 20 (58.8) Access site, No. (%)  Right femoral vein 29 (76.3) 24 (70.6) .889 b  Right brachial vein 8 (21.1) 9 (26.5)  Left femoral vein 1 (2.6) 1 (2.9) Embolization material, No. (%)  Coil + Onyx 25 (65.8) 25 (73.5) .477 c  Coil 13 (34.2) 9 (26.5) Complication, No. (%)  Yes 1 (2.6) 1 (2.9) .725 b Hematoma, No. (%)  Yes 0 (0) 1 (2.9) .472 b Rupture, No. (%)  Yes 1 (2.6) 0 (0) .528 b Recurrence, No. (%)  Yes 9 (23.7) 2 (5.9) .036 c Postembolization syndrome, No. (%)  Yes 14 (36.8) 7 (20.6) .130 c BMI , Body mass index; CPP , chronic pelvic pain; OVEA , ovarian vein embolization alone; PVPE , pelvic venous plexus embolization; T0 , preprocedural baseline; T12 , 12 months after the procedure; VAS , visual analog scale. a Mann-Whitney U test. b Fisher exact test, P < .05 statistically significant. c Pearson χ 2 test. Comparison of demographic and clinical variables between the OVEA and PVPE groups BMI , Body mass index; CPP , chronic pelvic pain; OVEA , ovarian vein embolization alone; PVPE , pelvic venous plexus embolization; T0 , preprocedural baseline; T12 , 12 months after the procedure; VAS , visual analog scale. Mann-Whitney U test. Fisher exact test, P < .05 statistically significant. Pearson χ 2 test.

Several previous studies have primarily focused on the comparison of embolization of the main venous structures (such as ovarian and internal iliac veins) or the comparison of unilateral and bilateral embolization, with a notable lack of scientific data. Our study aims to address this gap by comparing the 12-month outcomes of two different PEE approaches: OVEA vs PVPE. To the best of our knowledge, this study is among the few to specifically evaluate the outcomes of the embolization of all ipsilateral and/or contralateral PEP originating from PVP. Our analyses demonstrated that, in this selected cohort with reflux-dominant PeVD, the PVPE approach was associated with greater symptom relief and lower recurrence rates compared with OVEA. In patients who underwent PVPE, there was a more prominent and statistically significant decrease in VAS scores related to CPP, dyspareunia, and postcoital pain. In addition, recurrence rates were significantly lower in the PVPE group compared with the OVEA group. The results of this study suggest that a more comprehensive PEE approach, addressing PEP, may contribute to improved outcomes and reduced recurrence rates in selected patients. However, PeVD represents a heterogeneous spectrum in which symptoms may also be driven by venous outflow obstruction, particularly iliac vein compression, constituting a distinct therapeutic phenotype. Several studies have demonstrated that endovenous stenting can lead to significant improvement in CPP, even in selected patients in whom ovarian vein reflux was not directly treated. 12 , 13 These findings emphasize that reflux is not necessarily the dominant mechanism in all symptomatic cases and underscore the importance of careful phenotypic differentiation. In the present study, patients with suspected venous outflow obstruction were intentionally excluded; therefore, our findings should be interpreted within the context of nonobstructive, reflux-dominant PeVD. Concomitant lower extremity chronic venous insufficiency was present in 52.8% of patients, underscoring the frequent coexistence of pelvic and lower limb venous disorders in clinical practice. Although lower limb symptoms were predominantly right-sided in the majority of these patients, those with left-sided or bilateral involvement were evaluated with particular care. In such cases, cross-sectional imaging and venography were meticulously reviewed for findings suggestive of left iliac vein compression, including focal luminal flattening, segmental caliber reduction, delayed contrast transit, collateral venous filling, and characteristic compression patterns such as the pancake appearance. Although such coexistence may initially prompt consideration of an obstructive component, careful interpretation of imaging and venographic findings remains essential for distinguishing obstruction-driven from reflux-dominant PeVD. Recently, there have been significant changes in the paradigm of PeVD. Although it is still not sufficiently addressed by gynecologists, both in the literature and in clinical practice, PeVD is increasingly gaining ground in the differential diagnosis of CPP, moving away from its previously neglected position. 14 The increasing recognition of PeVD has prompted clinicians to seek effective and safe treatment modalities. PEE has gained prominence in recent years due to the limitations of medical therapies and the invasive nature of surgical interventions. Numerous studies demonstrated that PEE is a safe and effective treatment modality for PeVD. 15 , 16 , 17 Although no conclusive clinical evidence has been established, PEE is recommended with a IIaB level of evidence according to the guidelines of the Society for Vascular Surgery and the American Venous Forum. 18 In this study, we observed a decline in the VAS scores related to CPP, dyspareunia, and postcoital pain, regardless of the PEE approach applied. Similar results were obtained from numerous studies. 19 , 20 The observed symptom improvement following PEE suggests that hemodynamic mechanisms play an important role in symptom generation in reflux-dominant PeVD. Considering the results of our study and previous studies, it can be concluded that PEE is an effective method in the treatment of PeVD, not only for symptom relief but also for improving patients' QoL. Some previous studies demonstrated symptomatic relief regarding dysmenorrhea. 21 In our cohort, we did not observe a statistically significant change in the VAS scores related to dysmenorrhea ( P = .398). This may reflect the predominantly prostaglandin-mediated pathophysiology of dysmenorrhea rather than hemodynamic factors. 22 PeVD is a clinical entity fraught with uncertainty at virtually every stage of the condition, including diagnostic criteria, the PEE approach, and the type of materials used in the treatment. Regarding the PEE approach, there is no consensus on which veins should be embolized in the treatment process; therefore, there is no definitive treatment algorithm defined for PeVD. Many PEE approaches have been described in the literature. Some clinicians argue that embolization of the ovarian vein alone is sufficient. 4 , 5 , 6 In our study, we observed that the changes in VAS scores regarding CPP, dyspareunia, and postcoital pain were statistically significantly more prominent in the PVPE group compared with the OVEA group ( P = .001, P = .008, P = .005, respectively). Moreover, lower recurrence rates were observed in the PVPE group ( P = .036) compared with the OVEA group. Liang et al 5 argued that the keep-it-simple embolization approach provided sufficient symptomatic relief. In contrast to our aspect, embolization of only refluxing veins may not be the permanent or long-lasting solution for pathophysiologic problems of PeVD, which is a “living” disease in the complex anatomy of PVP. Pelvic venous drainage occurs through multiple pathways, including the ovarian and iliac veins and their tributaries and PVP. In the absence of venous outflow obstruction, reflux originating from the PVP and internal iliac vein tributaries may be explained by primary valvular incompetence, hormonally mediated venodilation, and long-standing venous remodeling associated with multiparity. In particular, the PVP plays an important role in PEP. However, these structures were not systematically evaluated in the aforementioned study. Failure to address veins that do not exhibit reflux on imaging may limit treatment effectiveness, as these veins can still contribute to symptoms. In addition, such veins may develop reflux after embolization of the main venous structures, further compromising outcomes. In addition, the claim that symptoms largely resolve within the first month may not fully align with the slow-responding nature of PeVD. Given the complexity and uncertainties of the pelvic venous system, simplified embolization strategies may be insufficient to address the full hemodynamic complexity of reflux-dominant PeVD in selected patients. Anatomical variations are more common in the venous system than in the arterial system. The PVP is characterized by multiple inflows and outflows. It is a structure with rich anastomoses and numerous alternative connections. Therefore, embolizing the ovarian or internal iliac veins alone may not drain the venous reservoir and may not result in ultimate venous decompression. We believe that the success of PEE does not depend solely on occlusion of major veins but also on considering hemodynamic factors. Although venous reflux represents an important mechanism of pain generation in reflux-dominant PeVD, venous outflow obstruction may also contribute to pelvic hypertension and symptom development within the broader spectrum of PeVD. Local venous distension, increased wall tension, tissue edema, and increased sensitivity of perivascular nerve endings, whether driven by venous reflux or outflow obstruction, also contribute to pain. 23 Embolization of major veins such as the ovarian and internal iliac veins may stop dynamic reflux, but it may not eliminate congestion in the PVP, which acts as a static reservoir, particularly if alternative drainage pathways remain intact. Embolization of the ovarian veins may reduce dynamic reflux by blocking the flow, whereas embolization of the PVP aims to occlude PEP and facilitate decompression of the pelvic venous reservoir. The drainage of volume overload of the PVP may lead to better venous decompression, greater symptomatic relief, and reduced recurrence rates. To sum up, we believe that clinicians involved in the treatment of PeVD should consider the fact that the challenge is not only anatomic but also hemodynamic. Another important finding in our study was the prolonged time to diagnosis (34.98 ± 9.03 months), indicating that PeVD still ranks low in the differential diagnosis of CPP. Kashef et al 24 defined PeVD as “a common but often underdiagnosed condition” and emphasized that delay in diagnosis has a negative impact on patients' QoL. The delay in diagnosis not only has a negative impact on the patient's comfort and QoL but also significantly increases health care costs. 25 The lack of definitive clinical and radiological diagnostic criteria and the fact that the disorder requires a multidisciplinary approach make it difficult to recognize PeVD. The primary limitations of this study include the relatively small sample size and its single-center, retrospective design. The retrospective nature of the study carries the potential for selection bias, as treatment strategy and embolization extent were determined by clinical and venographic findings rather than by randomization. Another limitation of this study is the absence of formal QoL assessment, as preprocedural measurements were not available due to its retrospective design. Symptom monitoring was performed using subjective VAS scores, and although patients were routinely followed during intermediate visits, VAS scores were systematically recorded only at baseline and at 12 months, which precluded detailed analysis of symptom evolution over time. Because the follow-up period was limited to 12 months, it is not possible to draw conclusions about longer-term durability or delayed recurrence. Although venous outflow obstruction was evaluated using cross-sectional imaging, venography, and pressure gradient measurements, the absence of intravascular ultrasound examination represents a limitation, as intravascular ultrasound imaging is considered a reference imaging modality for the detection and characterization of clinically significant iliac venous obstruction and plays a critical role in assessing its hemodynamic relevance. In conclusion, PeVD is a complex clinical condition that still contains many controversies and uncertainties in terms of diagnosis, treatment approaches, and long-term effects of the treatment approaches. Our study suggests that, in patients with reflux-dominant PeVD, the PVPE approach may offer improved symptom control and lower recurrence rates compared with OVEA. Although we acknowledge that conclusions should not be overgeneralized from a single study, we believe that our study will increase clinicians' awareness of both ipsilateral and contralateral PEP and PVPE, thereby improving treatment success and patients' QoL. Long-term results of this approach should be supported by further studies to reach a more solid scientific basis for the management of PeVD.

None.

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.