Introduction

Chronic pelvic pain in women, caused by pelvic venous congestion (PCS), can be treated with venous embolisation.¹ There are two divergent philosophical approaches to conduct the procedure: the keep-it-simple (KIS) approach—embolising only the refluxing vein(s), typically only the left ovarian vein (LOV), unless the right ovarian vein (ROV), left or right internal iliac vein (IIV) tributaries are also refluxing^{2, 3}—and the extensive (EXT) approach—empirically embolising almost all of the LOV, ROV, and left and right IIV tributaries.^{4, 5}

Previous studies comparing the EXT and KIS approaches have reported similar clinical outcomes.¹ The KIS approach could be associated with shorter procedure times, reduced radiation exposure, fewer coils, and theoretically a lower risk of coil migration. Considering the prevalence of PCS, there is potential for reducing healthcare costs by using the KIS approach. However, a recent systematic review on clinical outcomes seems dominated by the EXT approach, with a large number of ROVs and IIVs embolised to treat PCS.⁶ The aim of this study is to determine if the KIS approach is sufficient to effectively treat PCS, embolising only refluxing veins.

Methods

The study was approved by the local research ethics committee.

Venogram and embolisation protocol

All venograms and embolisation procedures were performed by a single operator who has more than 20-year experience in treating PCS. All procedures were performed under local anaesthetic with conscious sedation (intravenous midazolam and fentanyl), using a right internal jugular vein approach under ultrasound guidance, via a 5Fr sheath and 5Fr diagnostic catheters, such as MP A1 (Cordis Australia). Glyceryl trinitrate 50 mg topical patch was applied 30 min prior to the procedure to prevent venous spasm.

Catheter venography was performed to interrogate the bilateral renal, ovarian and internal iliac veins; external compression of the left renal vein and left common iliac vein was checked for and excluded. With our ‘Keep-It-Simple’ (KIS) protocol, only refluxing ovarian and/or internal iliac vein tributaries connected to the pelvic varicosity were embolised. The Valsalva manoeuvre was used to demonstrate reflux unless it was already evident during quiet respiration.

A ‘Sandwich’ technique similar to that of the testicular vein embolisation described by Reiner⁷ was used to embolise the ovarian vein, deploying a nest of 1–2 pushable coils distally at the pelvic brim level to slow down the flow to near stasis, followed by 1–2 mL of sclerosing foam injected proximal to the coil nest. The catheter was then positioned further proximally, to about 5 cm from its junction with the renal vein on the left, or IVC on the right, and a second nest of 1–2 coils was deployed, followed by 1 mL of sclerosing foam, injected slowly proximal to the coil nest. Any significant ovarian vein branches were either individually embolised or their junctions covered by coils and sclerosing foam. Any refluxing internal iliac vein anterior division tributaries feeding the pelvic varicosity were also embolised, typically using only one nest of coils and optional sclerosing foam just proximal to the coil nest, if deemed safe to do so (Fig. 1).

Pushable coils used were fibred platinum, 14 cm long (Nester Coils; COOK Australia), sized at 20% larger than the measured diameter of the target vein. Retractable coils were rarely used, only when the operator felt unsafe to use pushable coils due to the unfavourable anatomy of the target vein. The sclerosing foam was prepared by mixing 4 mL air with 2 mL 3% sodium tetradecyl sulfate (Fibro-Vein; Australasian Medical and Scientific Limited) or 3% polidocanol (Aethoxysklerol; Getz Healthcare Pty Ltd, Australia), agitated 20 times through a three-way stopcock using two 5 mL syringes, similar to the technique originally described by Tesarri.⁸

Out KIS approach did not involve injecting sclerosing agents into the pelvic veins or use of occlusion balloons.

Results

During the 5-year study period, 154 women underwent diagnostic venogram, 15 women failed to show any refluxing veins to embolise, and 139 women were found to have one or more refluxing veins and were treated with embolisation (Fig. 2 and Table 1). Of these, 102 (73%) required unilateral ovarian vein embolisation only [71% left ovarian vein (LOV) only; 3% right ovarian vein (ROV) only], 20 (14%) required bilateral ovarian vein embolisation, and 17 (12%) required pelvic vein embolisation (Table 2). The average number of coils used per case was six, with most women requiring only four coils.

Six-week follow-up were attended by 128/139 (92%) women. Short-term success was achieved in 114 out of 128 women (89%), who reported being ‘Happy or Very Happy’ with the overall outcome, indicating ‘Complete Resolution’ or ‘Significant Improvement’ of symptoms (Fig. 2).

At the time of study, 118 women were identified to have had embolisation more than 12 months ago (mean 34 months; range: 12–60 months), 101 (86%) women responded the request to long-term study surveys; 17 (14%) were unable to be reached and lost to follow-up; 7 (6%) declined to participate; and 8 (7%) were unable to complete the surveys online but contacted the research team or clinic, stating that they were unable to complete the survey, citing concurrent/coexisting/alternative medical conditions with overlapping symptoms (fibromyalgia, Crohn's disease, rectal cancer, pudendal neuralgia, pedunculated fibroid, severe adenomyosis, oversized Mirena, non-specific bloating). Completed surveys were available for analysis in 86 women (Fig. 2).

Overall, long-term success was achieved in 72 out of 86 women (84%), who reported being ‘Very Happy/Very Satisfied’ or ‘Happy/Satisfied’ with the overall outcome.

Long-term outcome of ancillary symptoms is summarised in Table 3. Symptom improvements were noted in high proportion of patients treated, notably: back pain 89%, leg pain/heaviness 90%, bladder symptoms in 74%, pain during sex in 79%, heavy menstrual bleeding 76% and period pain in 79%.

Three women required one repeat venogram and additional embolisation, and one woman required two repeat venograms and further embolisation. These four women eventually achieved long-term success.

For those who reported symptomatic relief, the majority (77%) experienced relief within 1 month: 17% reported immediate relief, 23% within a week and 30% within a month.

The mean pain reduction was 5.2 on the visual analogue scale (VAS), decreasing from 7.8 ± 1.7 to 2.7 ± 1.7 (P < 0.001). Notably, 42% of women no longer required pain medications, and 86% reported a reduced requirement for pain medications. Post-procedure, eight (6%) patients experienced mild to moderate pelvic or loin pain, attributed to thrombophlebitis. No worsening of pain or pelvic pressure was recorded. Only one patient reported worsening of leg pain/pressure, one reported worsening of period pain, and one reported worsening of heavy menstrual bleeding. There were no cases of infection/haemorrhage, or readmission for unexpected outcomes. There was no coil migration or dislodgement.

When assessing the impact of covariates influencing long-term clinical success, patients who had pre-procedure pelvic pressure, back pain, leg pain, bladder symptoms, fatigue, period pain and pre-menstrual bloating were associated with treatment success. Upon binary logistic regression analysis, none of the covariates remained significant (Table 4).

When assessing the impact of covariates influencing long-term symptomatic return, patients without fatigue (34% vs 15%, P = 0.032) and patients aged ≥45 (37% vs 11%, P = 0.005) were more likely to experience symptomatic return. Patients with bladder symptoms were also more likely to have symptom return, though this did not meet statistical significance (30% vs 15%, P = 0.091). Upon binary logistic regression analysis, fatigue (P = 0.02) and age ≥45 (P = 0.001) remained significant (Table 5).

Discussion

This study confirmed that the keep-it-simple (KIS) approach was effective in treating pelvic congestion syndrome (PCS), achieving short-term success in 114 out of 128 women (89%) and long-term success in 72 out of 86 women (84%).

Based on our KIS experience, 73% women required unilateral ovarian vein embolisation, only 14% women required bilateral ovarian vein embolisation, and only12% women require internal iliac vein tributary embolisation. None of our patients required embolisation of all four sets of veins. Comparing with the EXT approach empirically embolising all four sets of veins, KIS approach implicate potential savings in procedure time, radiation dose and coil cost.

Our KIS approach results compare favourably with the extensive (EXT) approach reported by Kim and Venbrux⁴ on 127 cases over 5 years (1998–2003) with a mean clinical follow-up of 45 months, achieving a similar 83% improvement. However, in their study, an attempt was made to embolise all four veins in two separate sessions (84% had bilateral ovarian vein embolisation, 85% also had iliac vein embolisation), yet 13% showed no improvement. Furthermore, 4% of women reported worsening of symptoms, possibly due to the occlusion of non-refluxing drainage pathways and the forced injection of sclerosing agent into pelvic veins, causing pelvic thrombophlebitis. Notably, no worsening of pelvic pain or pressure was recorded in our cohort.

In contrast, using our KIS approach, even though only refluxing veins were selectively embolised, only 4.7% of women required a repeat venogram and embolisation to address residual or recurrent. We avoided unnecessary embolisation of antegrade non-refluxing pathways.

A study by Laborda et al., for ultrasound-proven PCS treated patients with the EXT approach embolisation, had a total of 179 patients (88.6%) complete the 5-year follow-up period, achieving a VAS drop of 6.6 from 7.34 to 0.78. However, there was a long lead time (13.5 ± 1.9 months) for patients with severe pain to see clinical improvement.⁵ In contrast, with our KIS approach, a VAS drop of 5.2 was achieved, and 72% reported symptom relief within 1 month. Immediate relief of pelvic congestion symptoms is expected after embolisation of refluxing veins, unless procedure-induced pain masks and delays the expected improvement. These divergent results between KIS and EXT could be due to the latter contributing to extensive pelvic thrombophlebitis and occlusion of antegrade non-refluxing drainage pathways.

Laborda et al.⁵ also reported four cases of coil migration (1.9%), considered major complications, with two coils migrating to the right external iliac vein and two to the pulmonary arteries. In comparison, with our KIS approach, only a small percentage (12.2%) of women required iliac vein coil embolisation, which carries a greater risk of coil migration compared to ovarian vein embolisation. We hypothesise that embolising fewer iliac veins and having a more targeted approach for refluxing veins minimises the number of coils and the risk of complications. When we embolised the tributaries of the anterior division of the internal iliac vein (IIV), coils were placed as distally as possible to avoid coil migration. Furthermore, using the ‘Sandwich’ technique for ovarian vein embolisation, we did not need to deploy coils within the proximal 5 cm of the ovarian vein, further reducing the risk of coil mal-deployment. We also used glyceryl trinitrate (GTN) topical patches to prevent venous spasm, which could predispose improper coil formation in the desired location.

Our study compares similarly to other previous smaller cohort studies utilising the KIS approach to treat PCS. D'Archambeau et al.² reported a cohort of 66 patients with a pain scale drop of VAS 5.74 (from 7.88 to 2.15) by using only an average of 4–5 coils. Kwon et al.³ reported a cohort of 67 patients achieving pain reduction (completely gone or significant reduction) in 82% of cases. Similarly, Kwon also noted relatively early symptom relief within 2–3 months for most patients. This contrasts with Laborda's EXT approach, which required 13.5 months to provide appreciable pain relief.⁵ Kwon's observations support, and our results suggest, that pain symptom relief following embolisation should be almost immediate,³ unless complicated by extensive pelvic thrombophlebitis due to indiscriminate embolisation of pelvic veins using the EXT approach.

We propose that PCS in females is a functional analogue to varicocoele in males. We believe women's pelvic varicosities should be treated similarly to men's pampiniform plexus. The common pathophysiology is the incompetence of the gonadal vein, resulting in increased pressure and reversed flow, congesting the pelvic venous plexus in females and the pampiniform plexus in males. When treating male varicocele, one would not contemplate injecting sclerosants into the varicocele itself, as this would cause painful thrombophlebitis. Similarly, the logical approach to treat PCS is to address the root cause of the congestion (the refluxing gonadal veins). This approach will lead to the collapse of the varicosities, which, no longer being subject to the pressure of the refluxing veins, will decompress and will no longer be a cause of symptoms.

Incompetence of iliac vein tributaries can contribute to PCS in females, but published studies utilising the KIS technique suggest that this is uncommon.^{2, 3}

Venous embolisation by using ethylene vinyl alcohol copolymer (Onyx)⁹ and Amplatzer Vascular Plug (AVP)¹⁰ have been described. There is no comparison study to show that these more complicated and expensive devices has a better outcome than pushable coils.

This present study has some limitations. First, our follow-up is relatively short, ranging between 12 and 60 months. However, in common with Kwon,³ we have noted that most patients report relief within 1–3 months, and therefore a follow-up study of treated PCS patients for a minimum of 12 months should be sufficient. Second, results from this single-centre, single-operator cohort study may not translate to worldwide practice. However, with a consistent technique and strict adherence to protocol, the results can be a useful reference benchmark for future studies. Finally, our outcome survey questionnaire has not been validated, and retrospectively collected data suffers inherent biases. Ideally, to formally validate the KIS approach, a prospective randomised comparison study comparing KIS and EXT approaches would be required.

In conclusion, the keep-it-simple approach, embolising only the refluxing ovarian and/or iliac veins, can achieve a successful clinical outcome for pelvic congestion syndrome. The extensive approach of empirically embolising all ovarian veins and internal iliac veins may not be necessary. This carries implications for potential savings in procedure time, cost and radiation dose.

Funding

There was no financial support from outside organisations.

Declaration

We declare that all procedures performed were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Consent to participate

Informed consent was obtained from all individual participants involved in the study.