A novel management strategy for treatment of pelvic venous disorders utilizing a clinical screening score and non-invasive imaging
Abstract
Summary:Background: Treatment of pelvic venous disorders (PVD) including pelvic congestion syndrome (PCS) are often delayed due to its varying clinical manifestations. Patients and methods: Patient referral was based on a literature- and personal experience-derived clinical “PCS screening score” (higher score points indicate greater likelihood with a maximum score of 10 points). We studied consecutive women who were (i) referred for vascular assessment and treatment to the University Hospital Zurich (2017–2021), (ii) had a PCS score ≥3 points, (iii) had evidence of obstructive or non-obstructive PVD by duplex sonography or cross-sectional imaging, and (iv) underwent endovascular therapy. The primary outcome was change in symptom severity after endovascular therapy: (i) freedom from symptoms, (ii) improvement with residual symptoms, (iii) no improvement. Results: We included 43 women (mean age 36 years): 81% had previous pregnancy, 19% endometriosis. The median PCS score was 7 (IQR 5–9) points. Chronic lower-abdominal pain was the leading symptom in 86% patients, followed by recurrent leg (9%) and vulvar (5%) varicosities. The main PVD pathologies were ovarian vein insufficiency (61%), internal iliac vein insufficiency (9%), or a combination of both (30%), whereas 42% had a deep venous obstruction of the inferior vena cava, common iliac or left renal veins. Endovascular therapy included ovarian vein embolization (86%), internal iliac vein embolization (9%), and venous stent placement (35%). After a median of 4 (IQR 1–8) months from endovascular treatment, 40 (93%) patients reported improvement of the leading symptom, and 14 (33%) were symptom-free. Complications included re-intervention for stent stenosis (13%, all post-thrombotic), coil-migration into the left renal vein (7%, all retrieved), and transient pelvic sclerotherapy-induced thrombophlebitis (2%). Conclusions: Endovascular therapy following a diagnostic approach, which included a PCS screening tool and non-invasive imaging, appeared to be highly effective and was associated with a low rate of complications.
Introduction
Pelvic congestion syndrome (PCS) is a common condition, particularly among multiparous women. It is estimated that one third of women with chronic pelvic pain of unknown cause may have PCS [1]. However, it is often underdiagnosed or its diagnosis is delayed [2]. Since the term “PCS” only incompletely characterizes the full pathophysiology and clinical picture of pelvic venous disorders (PVD), the latter will be used throughout the text as recommended by the International Union of Phlebology [3]. The clinical presentation is heterogeneous: the most prevalent symptom is chronic pelvic pain, which is typically dull and aching in nature, and often non-cyclical. It may enhance during long periods of standing, menstruation, or sexual intercourse. Patients may also experience venous leg claudication, varicosities of the lower extremities or vulva, hemorrhoids, obstipation, and micturition urgency.
The pathogenesis is most likely multifactorial. Enlarged gonadal vein and pelvic vein plexuses consequent to pregnancy-related increase in uterine blood flow due to the vasodilator effects of estrogens may be involved. In addition, contextual valve incompetence and obstruction of draining veins play a pathogenetic role. The final common mechanism is a venous reflux causing venous congestion of periovarian, periuterine, and vulvar varicosities. In obstructive PVD, the venous congestion is typically caused by (i) the compression of the left renal vein by the aorta and superior mesenteric artery (nutcracker syndrome) with a reflux in the ovarian vein, and (ii) the compression of the left common iliac vein (May-Thurner syndrome) with a reflux in the internal iliac vein. Similarly, PVD may also occur in patients with post-thrombotic syndrome (PTS) with iliac vein or inferior vena cava obstruction [4].
There are no reliable diagnostic criteria for PVD. The absence of dilated pelvic veins in vascular imaging (magnetic resonance imaging (MRI) venography, computed tomography (CT) venography, or invasive venography) can usually exclude PVD. Its diagnosis relies on the combination of typical symptoms, dilated pelvic veins and reflux, and, ultimately, an improvement of symptoms after endovascular treatment. Endovascular therapy has the goal to reduce venous congestion. This is achieved by (i) sealing of pelvic varicose veins including the origin of reflux by embolization techniques using sclerotherapy and metallic coils, and (ii) decompression or recanalization of venous obstruction using stents, if present. Mini-invasive catheter-based techniques are regarded as the standard of treatment, as opposed to surgical ovarian vein ligation or hysterectomy.
Current gaps of knowledge concerning PVD call for better awareness and an improvement of diagnostic standards. In this clinical series, we report on the prevalence of symptoms at baseline among PVD patients and clinical outcomes of a novel management strategy including a clinical screening score, non-invasive imaging, and endovascular therapy.
Patients and methods
This study was conducted as a retrospective analysis of medical charts of women with suspected PVD. A clinical score was previously derived based on data from the literature and personal experience. This served as part of a gynecologist- and general practitioner-targeted awareness campaign to aid them in screening patients with a symptomatology that may be consistent with PVD (Table I). Typically, patients summing three or more points are referred to our clinic for further vascular assessment. For this study, we included all women (age 18–60 years) who were (i) referred to the University Hospital Zurich (Oct 2017 to Nov 2021), (ii) had a PCS score ≥3 points, (iii) had imaging evidence of obstructive or non-obstructive PVD by duplex sonography or cross-sectional imaging, i.e. CT or MRI venography, and (iv) underwent endovascular therapy. The study was approved by the local Ethic Committee; all patients included provided a signed general consent.
Data was collected from medical reports, discharge letters and interventional reports and included (i) demographic information, (ii) relevant comorbidities, (iii) symptoms at baseline and after endovascular treatment; (iv) diagnostic information from pelvic venography with fluoroscopy, and (v) procedural details on embolization and venous decompression. PVD was classified according to Greiner’s classification: Type 1: Valvular or parietal venous anomalies without pelvic or supra pelvic obstruction to venous outflow responsible for a reflux pathology (non-obstructive type) [5]. Type 2: Stenosis or obstruction in a draining vein responsible for symptomatic substitute collaterals (obstructive type) [5]. Type 3: Venous anomalies secondary to a local extrinsic cause [5]. In addition, the PVD main pathology was categorized into (i) ovarian vein incompetence, (ii) internal iliac vein incompetence, or (iii) a combination of both. The primary outcome was the change in symptom severity following endovascular therapy. We categorized it as follows: complete freedom from PVD symptoms, improvement with residual symptoms, and no improvement of symptoms at latest follow-up. Secondary outcomes included the prevalence of baseline symptoms related to PVD. The safety outcome was the prevalence of peri-interventional complications.
Diagnostic subtraction venography was performed from internal jugular vein (IJV) access or femoral access, while embolization of gonadal veins was always performed in sandwich technique via IJV access. Prior to embolization, each patient underwent venographic evaluation of deep venous obstruction (typical or atypical iliac vein compression, nutcracker compression) and reflux of both ovarian and internal iliac veins with help of Valsalva maneuver. Deep venous obstruction was evaluated by biplane venography or, if equivocal, by intravenous ultrasound (IVUS). A compression (or obstruction) of >50% with the presence of either reflux into the internal iliac vein or presence of collateral flow was considered relevant. For decompression of May-Thurner lesions, the Sinus-Obliquus© stent (Optimed, Germany) was used. Initially we used pushable coils (MReye© Embolization Coils, Nester© Embolization Coils, Cook, Bloomington, IN, USA), which were replaced by semi-detachable soft coils (Interlock Detachable Embolization Coils, Boston Scientific, Marlborough, MA, USA) due to issues of migration.
Normally distributed data were presented as means with standard deviations, ordinal scales as medians with interquartile ranges. Categorical outcomes were presented as numbers with percentages. Statistical analysis was performed with SPSS (IBM, SPSS Statistics Version 26).
Results
Baseline characteristics
We included 43 consecutive women suspected of PVD referred to our clinic for further assessment, who consented to share their data for research purposes. The baseline characteristics are displayed in Table II. In brief, mean age was 36±8 years, 35 (81%) patients had previous pregnancies (median number of parities: 2, IQR 1–2), and 8 (19%) had previous endometriosis treated with hormones or laparoscopic surgery. Symptoms at initial presentations are displayed in Table III. The most prevalent symptoms were chronic lower abdominal pain (93%), dysmenorrhea (77%), orthostatic pain (74%), dyspareunia (54%) and leg varicosis (51%). Chronic lower abdominal pain was the leading symptoms in 35 (86%) patients, followed by recurrent painful leg varicosities (4 patients, 9%) and vulvar varicosities (2 patients, 5%). Leg and vulvar varicosities were more frequent among patients with non-obstructive (60%) versus obstructive (50%) PVD.
Diagnosis and treatment
Non-invasive imaging prior to pelvic venography was performed in all patients: 38 (88%) patients had a duplex ultrasound exam and 36 (84%) had cross-sectional imaging (31 patients had MRI venography and 8 had CT venography). All patients underwent pelvic venography with fluoroscopy. Findings from pelvic venography with fluoroscopy and therapeutic details are displayed in Table IV. Based on the pelvic venography, PVD type 1 (non-obstructive type) was diagnosed in 25 (58%) patients, while 18 (42%) patients were diagnosed with PVD type 2 (obstructive type). The primary cause of PVD was ovarian vein reflux and internal iliac vein reflux in 26 (61%) and 4 (9%) patients, respectively; in 13 (30%) patients it was a combination of both. Ovarian vein embolization, internal iliac vein embolization and venous stent implantation was performed in 37 (86%), 4 (9%) and 15 (35%) patients, respectively. Among the 18 patients with obstructive PVD, 9 underwent staged interventions: 6 underwent embolization prior to stent placement, 3 underwent stent placement prior to embolization. None of the 4 (9%) patients with nutcracker compression received venous stents or surgical transposition.
Follow-up and clinical outcome
Median follow-up was 4.0 (IQR 1.4–8.3) months. At follow-up, 40 (93%) patients reported improvement of the leading symptom, with 14 (33%) having complete resolution of symptoms. Complete resolution of symptoms occurred in 28% and 39% of patients with non-obstructive and obstructive PVD, respectively.
In 3 (7%) patients, endovascular intervention did not improve symptoms. One patient had chronic lower abdominal pain and underwent left-sided ovarian vein embolization and venous stent implantation due to severe common iliac vein compression. She had multiple previous surgical procedures including abdominoplasty, ovarian cystectomy and laparoscopy for removal of endometriosis. The second patient had chronic perianal and gluteal pain and underwent bilateral ovarian vein embolization. She had seen various specialists and was unsuccessfully treated for piriformis syndrome and spasms of the puborectal sling. The third patient had disabling, chronic lower abdominal pain and underwent left-sided ovarian vein embolization. She had multiple gynecological and urological exams, which were all unremarkable.
Complications related to the procedure occurred in 5 (12%) patients: 3 (7%) patients had partial coil migration into the left renal vein, which were all retrieved by snaring techniques during the index procedure. One patient (2%) had a non-flow limiting mural hematoma of the left renal vein. Another (2%) patient had symptoms of pelvic thrombophlebitis due to ovarian vein embolization with polidocanol. Among the 15 patients with venous stents, two (13%) patients underwent reintervention due to stent restenosis. One patient complained of intermenstrual bleeding during anticoagulation therapy after stent placement, which resolved after changing non-Vitamin-K dependent oral anticoagulants to aspirin.
Discussion
Endovascular therapy following a diagnostic approach, which included a PCS screening tool and non-invasive imaging, appeared to be effective to improve PVD symptoms and was associated with a low rate of complications. In our opinion, this is the first study reporting clinical outcomes of endovascular treatment using a novel management strategy with a PCS screening tool used by primary care physicians and gynecologists for referral. Our findings add on the limited clinical data available on outcomes of PVD patients, in particular those with underlying deep venous obstruction who are treated with endovascular techniques.
The most prevalent symptoms consisted of chronic lower abdominal pain, intensified while standing in upright position, and dysmenorrhea, which was present in three quarter of women. The most prevalent clinical findings were leg (51%) and vulva (47%) varicosities. In three patients without pelvic pain, pelvic vein insufficiency was the primary cause for leg or vulva varicosities. The most common pathophysiological form of PVD was an isolated non-obstructive left ovarian vein insufficiency, which was present in more than 50% of patients. However, obstructive PVD, often consisting of combined ovarian and internal iliac vein reflux, was considerably frequent, and required a more complex approach involving decompression or recanalization of deep veins by stent implantation (35%) followed by staged embolization, or vice versa. Clinical outcomes were overall positive: 93% and 33% of patients reported on an improvement or a complete resolution of symptoms, respectively; 7% did not benefit from endovascular therapy. Periinterventional complications were mild and without long-term consequences, and mostly consisted of coil migration during the detachment process; all were immediately retrieved.
In a previous meta-analysis by Mahmoud et al. [6], including 648 patients from 17 non-controlled studies, short-term outcomes were similar to ours with 88% of patients reporting clinical improvement following endovascular therapy. Despite these promising results, 6–32% of patients might still not experience any meaningful symptom relief [7]. Data from previous studies on initial signs and symptoms of pelvic vein insufficiency, with or without pelvic pain, as well as clinical outcomes following endovascular therapy are inconsistently reported and only available for inhomogeneous populations (Table V) [8–20].
Among the few studies that reported specific symptoms or clinical signs of pelvic venous insufficiency, its prevalence varied significantly: 10–37% for vulva varices, 4–100% for leg varices, 13–90% for dysmenorrhea, 15–89% for dyspareunia and 43–100% for orthostatic pain [8–18]. Regarding the pathophysiology of PVD, only very few studies investigated the presence of deep venous obstruction as a potential cause for pelvic venous insufficiency [9, 10, 12, 16]. Hartung et al. found ilio-caval lesions, nutcracker compression, or a combination of both in 24%, 3%, and 7% of women who had venography for suspected pelvic venous insufficiency [9]. In our study, venous obstruction was frequently observed, and interpreted as the underlying cause of venous congestion in 42% of patients. We hypothesize that the obstructive form of PVD (type 2) is underdiagnosed, which might lead to treatment failure and recurrence of symptoms in a significant proportion of women.
Taking the results of our study in consideration with previous studies, one can conclude that contemporary endovascular techniques are principally safe and effective for the management of women with PVD. As indicated in current practice guidelines, endovascular therapy should be chosen over a less effective medical (psychotropic, hormonal, or venoactive drugs) and a more invasive surgical (ovarian vein ligation, hysterectomy) approach [3]. Severe complications are rare but include potential harmful migration of coils or sclerosants into the pulmonary circulation [13, 18, 19]. Of note, 13% of patients with venous stents developed an in-stent stenosis and required an endovascular reintervention (all had the PTS with long stent segments involving the inferior vena cava, iliac and femoral veins). We did not observe in-stent stenosis in patients with non-thrombotic iliac veins lesions.
Currently, the management of women with chronic pelvic pain is suboptimal, with only 40% of affected women being referred to specialists [2]. On the one hand, raising PVD awareness, particularly among family practitioners and gynecologists, may accelerate diagnosis and potentially improve quality of life in affected women. On the other hand, vascular specialists who see multiparous women with (recurrent) leg or vulva varicosities should inquire whether chronic pelvic pain exists and extend the routine duplex sonographic exam over assessing the morphology und function of gonadal and pelvic veins. If clinical suspicion is strong, MRI venography should be performed. Simultaneously, vascular specialists should collaborate with primary care physicians and gynecologists to rule out the numerous other, potentially life threatening, causes of chronic lower abdominal pain. Of note, we observed a high prevalence of co-existing endometriosis in our study (19%). Most of these women experienced improvement of lower abdominal pain after endovascular treatment. This information may be important for gynecologists who may consider the PVD as alternative diagnosis for women who have treatment failure for endometriosis.
In the absence of large cohort studies and global cooperation, a formal PCS screening tool was not developed and cannot be developed or validated any time soon. The literature- and experience-based screening tool we propose, may assist general practitioners and gynecologists in the initial screening to identify women for vascular specialist referral. Our data suggest that the specificity of this screening tool is appropriate. Table I displays potential items, which may be useful to discriminate pelvic pain due to PVD from other causes. In other words, one could consider referring to the vascular specialist not only all women with chronic pelvic pain, but also those with varicosities of unclear origin, who may benefit anyway of a vascular assessment, and those with a combination of less typical PVD symptoms. A pilot study of women with suspected PVD from our group to validate this PCS screening tool is currently under way.
Limitations
Our study has several limitations. First, sample size was small. Second, the follow-up was short, and it is unclear whether the treatment effect is durable. Conversely, treatment benefit in some patients might be underestimated since symptom relief may not occur immediately and may be delayed by 4–8 weeks [21]. Third, all patients were recruited from one vascular center leading to a significant selection bias. Prevalence of leg varicosities, vulvar varicosities or deep vein obstruction among women with PVD may be significantly lower in a cohort recruited outside a vascular center. Last, the validity of the results is limited due to the retrospective nature of this study. However, randomized-controlled trials on management and treatment strategies of PVD do not exist, and our data may provide estimates for the future planning of trials.
Conclusions
Endovascular therapy following a diagnostic approach, which included a PCS screening tool, duplex sonography, and cross-sectional imaging, appeared to be effective to improve PVD-related symptoms in most patients with an overall low rate of complications. Raising PVD awareness will help to accelerate diagnosis, enhance therapy, and potentially improve quality of life in affected women. We propose a PCS clinical screening tool helping to identify women with suspected PVD who may benefit from vascular assessment and treatment.
References
1 . A randomized controlled trial of goserelin and medroxyprogesterone acetate in the treatment of pelvic congestion. Hum Reprod. 2001;16:(5):931–9.
2 . Prevalence of chronic pelvic pain among women: an updated review. Pain Physician. 2014;17:(2):E141–7.
3 Diagnosis and treatment of pelvic congestion syndrome: UIP consensus document. Int Angiol. 2019;38:(4):265–83.
4 Post-thrombotic syndrome. VASA. 2021;50:(5):331–40.
5 , How does the patho-physiology influence the treatment of pelvic congestion syndrome and is the result long-lasting? Phlebology. 2012;27:(Suppl 1):58–64.
6 Efficacy of endovascular treatment for pelvic congestion syndrome. J Vasc Surg Venous Lymphat Disord. 2016;4:(3):355–70.
7 . Clinical outcome after treatment of pelvic congestion syndrome: sense and nonsense. Phlebology. 2015;30:(Suppl 1):73–80.
8 . A randomized trial of endovascular embolization treatment in pelvic congestion syndrome: fibered platinum coils versus vascular plugs with 1-year clinical outcomes. J Vasc Interv Radiol. 2018;29:(1):45–53.
9 . Embolization is essential in the treatment of leg varicosities due to pelvic venous insufficiency. Phlebology. 2015;30:(Suppl 1):81–5.
10 . Venous angioplasty and stenting improve pelvic congestion syndrome caused by venous outflow obstruction. J Vasc Surg Venous Lymphat Disord. 2015;3:(3):283–9.
11 . Safety, efficacy, and prognostic factors in endovascular treatment of pelvic congestion syndrome. Int J Gynaecol Obstet. 2014;125:(1):65–8.
12 Evaluation of the efficacy of endovascular treatment of pelvic congestion syndrome. Diagn Interv Imaging. 2014;95:(3):301–6.
13 . Endovascular treatment of pelvic congestion syndrome: visual analog scale (VAS) long-term follow-up clinical evaluation in 202 patients. Cardiovasc Intervent Radiol. 2013;36:(4):1006–14.
14 . Ovarian vein incompetence: a potential cause of chronic pelvic pain in women. Eur J Obstet Gynecol Reprod Biol. 2008;139:(2):215–21.
15 Pelvic vein embolisation in the management of varicose veins. Cardiovasc Intervent Radiol. 2008;31:(6):1159–64.
16 . The pelvic congestion syndrome: role of the “nutcracker phenomenon” and results of endovascular treatment. JBR-BTR. 2004;87:(1):1–8.
17 . Comparison of treatments for pelvic congestion syndrome. Tohoku J Exp Med. 2003;201:(3):131–8.
18 Pelvic congestion syndrome (pelvic venous incompetence): impact of ovarian and internal iliac vein embolotherapy on menstrual cycle and chronic pelvic pain. J Vasc Interv Radiol. 2002;13:(2 Pt 1):171–8.
19 . Transcatheter ovarian vein embolization using coils for the treatment of pelvic congestion syndrome. Cardiovasc Intervent Radiol. 2007;30:(4):655–61.
20 . Embolotherapy for pelvic congestion syndrome: long-term results. J Vasc Interv Radiol. 2006;17:(2 Pt 1):289–97.
21 . Embolization to treat pelvic congestion syndrome and vulval varicose veins. Int J Gynaecol Obstet. 2012;118:(3):227–30.
