Protocol-based treatment of spontaneous hemorrhage from varicose veins prevents recurrence of bleeding

Abstract

Background

Spontaneous hemorrhage from erosion of varicose veins through the skin is a serious and occasionally fatal complication of varicose vein disease. Various treatments, both acute and delayed, have been advocated. Our two-step clinical protocol was designed to prevent further hemorrhage without delay and to provide durable freedom from recurrent bleeding.

Methods

All patients referred to our surgical vein practice with hemorrhage from varicose veins were entered into a prospective registry. On presentation, all patients underwent diagnostic duplex ultrasound for venous reflux. Immediate treatment consisted of ultrasound-guided polidocanol/CO₂ foam sclerotherapy of the bleeding varicosity and adjacent veins. Subsequent endovenous ablation of underlying incompetent axial veins, with concurrent microphlebectomy when indicated, was scheduled and performed within 8 weeks.

Results

Fifty-nine patients were referred with recent hemorrhage from varicose veins over a period of 4 years. Fifty-six (95%) had an ultrasound-identified tributary varicosity underlying the point of bleeding, and three had a skin-surface erosion only and no ultrasound-identified underlying tributary. Of the 59 study patients, 52 underwent prompt polidocanol/CO₂ foam sclerotherapy, targeted to both the underlying tributary (when present) as well as the cutaneous bleeding varicosity. The remaining seven either declined sclerotherapy or were ineligible. Underlying incompetence of axial (great, small, or anterior saphenous) veins was found in 54 patients (92%). Of these patients with truncal vein incompetence, 48 underwent ablation—5 declined and 1 was too frail for any further intervention. There were no recurrent bleeds in the interval between immediate sclerotherapy and scheduled ablation. During a mean follow-up of 2.2 years, 55 of the 59 patients (93%) had no recurrence of bleeding. Four patients (7%) had late, recurrent hemorrhage: one had failed to return for their scheduled ablation, two were on chronic anticoagulation, and one had severe right heart failure. All four were retreated without further recurrence.

Conclusions

A two-step protocol of immediate ultrasound-guided foam sclerotherapy, followed within 8 weeks by endovenous ablation of incompetent axial veins and concurrent microphlebectomy, provided rapid and efficient treatment with durable freedom from subsequent hemorrhage.

Article Highlights.

• •Type of Research: Single-center retrospective analysis of registry data
• •Key Findings: Hemorrhage from eroded varicose veins in 59 patients was managed with a two-step protocol of immediate ultrasound-guided foam sclerotherapy followed by endovenous ablation of incompetent axial veins. Durable freedom from recurrent hemorrhage was achieved in 55 patients (93%).
• •Take Home Message: Hemorrhage from eroded varicose veins can be managed effectively with immediate ultrasound-guided foam sclerotherapy followed by scheduled endovenous ablation of incompetent saphenous veins.

Spontaneous hemorrhage from lower extremity varicose veins is an infrequent but occasionally fatal event.^1,2 Patients describe sudden, profuse bleeding from previously innocuous raised cutaneous veins that eroded through the skin.³ Bleeding is invariably painless and frequently erupts after toweling, shaving, scratching, or other benign activity.⁴ Patients often remain erect and attempt to manage the hemorrhage with towels or other ineffective means, rather than controlling the cause with direct pressure or elevation of the extremity.⁵ Hemorrhage can thus continue, leading to hypotension, loss of consciousness, and even death, with patients often found surrounded by large amounts of blood.6, 7, 8 Nonfatal bleeding, occurring in 3% to 9% of patients with varicose veins,⁹ is far more common, and provides an opportunity for treatment and prevention of a later fatal event.

Optimal management for the vast majority of patients not succumbing to their initial bleed has not been defined clearly. Historically, open surgery or vein stripping were performed, but improved results with liquid¹⁰ and foam¹¹ sclerotherapy as well as endovenous thermal ablation^12,13 have been reported. It remains unclear whether any of these modalities performed in isolation is adequate.

We hypothesized that a two-stage protocol of immediate ultrasound-guided polidocanol/CO₂ foam sclerotherapy, followed by endovenous ablation of the underlying, diseased axial/saphenous vein, would provide immediate control of hemorrhage as well as long-term freedom from recurrent bleeding.

Methods

Patients

A prospectively maintained registry of all patients presenting to our center with hemorrhage from varicose veins (Fig 1) was begun on July 1, 2019. This registry was reviewed and data for all patients presenting with recent bleeding from varicose veins from registry initiation until June 30, 2023, were collected. Patients with a remote history of bleeding (>4 months prior) were excluded. Their associated medical records were reviewed for demographic information, venous treatments performed, complications, and subsequent office visits. As part of an ongoing continuous quality assurance process, patients were also called by telephone to inquire about any subsequent bleeding events or additional venous interventions. Data were maintained in a firewall-protected registry. This study followed the principles outlined in the Declaration of Helsinki and used only information obtained from the review of medical records. Because of the retrospective nature of the study and the anonymization of the extracted data, institutional review board approval was not required.

Fig 1.

Patient seen the prior evening in the emergency room for a bleeding varicose vein and treated by suturing with silk. Note the associated prominent blue venulectasias, dilated veins, and other areas of thinning at risk for erosion and bleeding.

Treatment methods

Patient referrals to our center for bleeding from varicose veins were prioritized and given an immediate appointment the same or following day. Initial clinical evaluation included medical history, a physical examination, and functional Duplex ultrasound examination of the lower extremities. Duplex ultrasound examination for reflux was performed in the standing position. Reflux testing was performed by manual compression of the proximal calf, distal calf, and instep. No patient had acute or subacute DVT. Data regarding historical or ultrasound evidence of remote DVT were not recorded in the registry.

Prompt treatment with ultrasound-guided sclerotherapy was then performed, optimally at the same visit or within 48 hours. All patients with pathological reflux in the axial veins (great, small, and/or anterior saphenous) were then scheduled for definitive treatment of their underlying diseased veins with saphenous ablation and concurrent microphlebectomy as needed. Ablation was scheduled as soon as possible, with some short, nonclinical wait times owing to carrier preauthorization requirements, but all patients undergoing ablation had this definitive treatment performed within 8 weeks. Postprocedure Duplex ultrasound examination and clinical follow-up occurred 1 week after ablation, in person in the office. Longer term follow-up was via quality assurance checks by telephone.

Patients were diagnosed, classified, and submitted to insurance carriers as having a venous ulceration (Clinical, Etiological, Anatomical, and Pathophysiological class C6). Delays for conservative management were thereby avoided, and preauthorization, when required, was submitted as urgent. In cases where carrier approval could not be obtained immediately, we nonetheless proceeded with initial sclerotherapy treatment because of the clinical danger of rebleeding.

Ultrasound-guided polidocanol/CO₂ foam sclerotherapy was performed using a ButterflyIQ+ handheld ultrasound unit (Burlington, MA) and a 4:1 mixture of medical-grade CO₂/polidocanol 1% (Asclera; Methapharm, Coral Springs, FL) (Fig 2). The foam was created by the Tessari method, using two nonlubricated specialty syringes (Norm-Ject Luer Lock Solo; Air-Tite Products, Virginia Beach, VA) and delivered through a 1-inch 25G or 27G needle. This treatment was directed into the superficial veins feeding and draining the area of hemorrhage. This resulted in filling of the ulcerated, bleeding vein complex with white foam. When indicated, additional direct injections of foam sclerosant were delivered into other visible veins near the site of bleeding using a 32G needle and 3.5× loupe magnification. Additional concurrent foam sclerotherapy of other protruding venulectasias and telangectasias thought to be at risk for future bleeding was performed during the same session.

Fig 2.

Ultrasound-guided foam sclerotherapy of incompetent vein near the site of earlier hemorrhage, with intentional backbleeding into the tubing confirming correct positioning, to be followed by foam instillation.

After sclerotherapy, 20 to 30 mm Hg below-knee compression stockings were placed on the leg to be kept on for 24 hours, after which daily wear was recommended until the follow-up ablation procedure was performed. This strategy prevented progression of edema in all but two patients with preexisting open venous stasis ulcers; these patients were treated with an Unna boot or multilayer dressing. Insurance preauthorization and scheduling for endovenous radiofrequency ablation of ipsilateral diseased saphenous veins was begun before the patient departed the office.

All indicated ablations were performed within eight weeks in our office-based procedure room using the Venclose RF Ablation catheter (BD, Franklin Lakes, NJ) or the ClosureFast RF Ablation System (Medtronic, Minneapolis, MN) and standard tumescent anesthesia (0.1% lidocaine and 1:200,000 epinephrine with 10 mL 8.4% sodium bicarbonate added per liter). In two patients, cyanoacrylate ablation was chosen and was performed using the Venaseal system (Medtronic). All segments of refluxing axial veins were treated down to the level of previous bleeding when indicated by the presence of reflux. Concurrent microphlebectomy of bulging varicose veins both adjacent and distant to the area of hemorrhage was performed using a disposable ophthalmic cataract scalpel (BVI, Waltham, MA) and a Mueller microphlebectomy hook (Wagner Medical, Middlebourne, WV). Postablation duplex ultrasound examination and follow-up history and physical examination were performed 1 week after the ablation. Patients were then discharged from routine in-person follow-up unless they had open ulcers requiring further observation and treatment.

Results

During the 4-year study period, 59 patients presented to our center with a history of spontaneous hemorrhage from erosion of varicose veins. Their average age was 67 ± 12 years (range, 31-87 years). There were 32 females (54%) and 27 males (45%). All points of bleeding were below the knee. Initial management and treatment of the bleeding was by the patient, family member, or friend in 23 (39%) and by a primary care provider in 5 (8%). The remaining 31 patients (53%) were treated by emergency services (urgent care center, emergency medical services, and/or emergency department). Two of these patients were admitted to the hospital, one of whom received multiple transfusions and intensive care unit treatment.

Of these 59 consecutive patients entered into the registry, 56 (95%) had an identifiable tributary varicosity (diameter 0.5-3.0 mm) seen on ultrasound examination and associated with the point of hemorrhage, and 3 exhibited no ultrasound abnormality at the site of hemorrhage. This tributary was the primary target of ultrasound-guided injection sclerotherapy. The three patients without this typical, underlying, ultrasound-identified tributary were still treated, but with visual, skin surface sclerotherapy alone, targeting the visible blebs and varicosities responsible for the bleeding.

Our protocol called for prompt polidocanol/CO₂ foam sclerotherapy as the first step of treatment. However, two patients presented to us in delayed fashion, could not recall exactly where their bleeding had occurred, and had no discernible ulceration or bleeding point identifiable. These two were kept in the registry and were moved onward to the ablation portion of the protocol, because both had ipsilateral saphenous incompetence.

Four other patients had bled from points within large, open venous stasis ulcers or from eroded, giant varicosities. All four had saphenous incompetence. These patients were felt not to be good candidates for sclerotherapy, but rather to require prompt saphenous ablation and concurrent microphlebectomy, which was performed without delay.

The remaining 52 patients underwent polidocanol/CO₂ foam sclerotherapy, to include ultrasound-guided foam sclerotherapy tributary injection (in all but the three without the typical underlying tributary) as well as visual injection into blebs and surface varicosities identified as the points of hemorrhage (Fig 3, step 1). No patient had complications from ultrasound-guided foam sclerotherapy.

Fig 3.

Treatment algorithm with patient totals undergoing each step. USG, ultrasound-guided foam.

Duplex ultrasound examination demonstrated underlying incompetence of truncal (great, small, or anterior saphenous veins) in 54 of the 59 patients (92%) (Fig 3, step 2). Of these, 48 underwent endovenous ablation of their diseased, refluxing axial veins (Fig 3, step 3). This left six patients for whom ablation was indicated but not performed: five had failed to return for their ablation appointment or declined ablation (one of these five returned 23 months later with new bleeding and underwent the previously recommended ablation without further hemorrhage). The sixth patient was judged too frail for ablation and was managed with local sclerotherapy alone.

Ipsilateral bulging varicose veins were considered a contributing factor for symptoms and risk of recurrent hemorrhage. Forty-one of the 48 patients undergoing ablation exhibited bulging varicosities and therefore underwent concurrent microphlebectomy.

At an average patient follow-up of 2.2 ± 1.3 years (median, 2.8 years), 55 patients (93%) had no further bleeding episodes after treatment. No patient experienced bleeding in the interval between immediate sclerotherapy and scheduled ablation. Four patients suffered late recurrent bleeding, not at the same site but all in the same extremity. Three had undergone their prescribed treatment of ultrasound-guided foam sclerotherapy followed by axial ablation, but had rebleeding at a mean time period of 1.3 years. The fourth patient had refused the recommended ablation after the initial sclerotherapy. These four patients were retreated after their recurrent hemorrhage and had no subsequent bleeding at a mean follow-up time of 1 year. Interestingly, two of these four patients had been on long-term anticoagulation, and a third had 4+ tricuspid regurgitation on echocardiography.

Discussion

Our results demonstrate that patients surviving initial hemorrhage from eroded varicose veins can expect a low risk of recurrence when treated with a two-stage approach of foam sclerotherapy followed by endovenous ablation of the underlying diseased saphenous veins.

The incidence of hemorrhage from eroded varicose veins is serious enough that a definitive algorithm for treatment is desirable. The annual incidence of fatal hemorrhage from eroded varicose veins has been estimated at 0.5 to 0.7 per million, which would equate to 175 to 245 deaths per year in the United States.^2,7 However, a more recent study found a substantially higher incidence, with death owing to hemorrhage from varicose veins occurring in one of every 1000 autopsies.¹⁴

The risk factors for death from spontaneous varicose hemorrhage include congestive heart failure,¹³ dementia,⁷ alcohol or drug use,¹⁵ or simply living alone.^2,5,14,16 Exposure to warm water via its vasodilator effect is also thought to provoke a bleeding event in susceptible individuals.¹²

Nonfatal hemorrhage is a much more common occurrence and represents an opportunity for the treatment and prevention of recurrence and, potentially, death.⁹ However, bleeding varicose veins often present a conundrum to emergency and primary care providers. Treatment is often with simple bandaging or suturing, but recurrence is common and not all patients are referred for definitive treatment by a venous specialist after the initial bleed. In our experience and that of others, some patients present for specialty care after substantial delay, which further increases their risk. This problem was addressed systematically by the UK's National Institute of Health which formally recommended urgent referral to a vascular specialist of all patients with bleeding varicose veins.¹⁷

Even with referral to a physician dedicated to the treatment of vein disease, there is no clear consensus for management. Treatment has evolved, with liquid sclerotherapy first reported in 1993¹⁰ and found superior to suture ligation in a 1996 report.³ Foam sclerotherapy with polidocanol foam for bleeding varicose veins in five patients was first reported in 2011 with no rebleeding over 17 months.¹¹ Treatment using endovenous radiofrequency ablation specifically for bleeding varicose veins was not reported until 2020.¹² Sclerotherapy was eschewed “because of the required prior authorization and lengthy approval process required by carriers.” Instead, the 32 patients in this series first underwent weekly Unna boot placement. After healing of the ulcerated bleeding site, elective ablation of the responsible diseased saphenous vein was recommended and eventually performed in 20 patients. There were no rebleeds, but follow-up was short. Finally, a 2022 retrospective review of 13 patients with bleeding varicose vein used neither Unna boots nor sclerotherapy. All underwent scheduled radiofrequency ablation, and two (15%) suffered rebleeding over 2.2 years of follow-up.¹³

Our algorithm used foam sclerotherapy because of its effectiveness in instantaneously occluding and rendering harmless most bleeding varicosities. Insurance authorization, when required, can be obtained swiftly through separate urgent pathways with the benefit of immediate bleeding control and the avoidance of weeks of Unna boot therapy. Allowing urgent referral for same- or next-day evaluation and immediate foam sclerotherapy, requiring only a standard sclerotherapy setup and handheld ultrasound unit, has not diminished our efficiency appreciably.

However, sclerotherapy alone is not enough.^9,10 Without treatment of the underlying ambulatory venous hypertension caused by refluxing saphenous veins, the risk of future recurrent hemorrhage persists. Similarly, endovenous saphenous ablation alone may not be timely, and patients risk another bleed while awaiting approval and scheduling of the more time- and resource-intensive ablation procedure. Insurance coverage for our patients' ablation procedures was invariably approved when our submissions included the clearly justified codes for hemorrhage and ulceration.

Our series of 59 patients constitute the largest reported group of patients treated for hemorrhage from varicose veins and is the first report of sequential use of foam sclerotherapy and endovenous ablation. Using this two-stage protocol we observed no early rebleeds and a low incidence of recurrent hemorrhage (7%) at 2 years of follow-up. Other notable findings included a high prevalence of an identifiable tributary varicosity (94%) associated with the point of hemorrhage, as well as a 92% prevalence of underlying saphenous vein reflux. These findings support the advisability of treatment with foam sclerotherapy and endovenous ablation, respectively. There were four instances of later recurrent hemorrhage. Three patients were on therapeutic anticoagulation and one had severe right heart pathology. Although no statistical inferences could be drawn, it seems reasonable to conclude that anticoagulated patients and those with elevated right heart pressures might best be followed more closely after treatment for hemorrhage.

This study has certain limitations. Patients were not randomized to different treatment arms, nor was either sclerotherapy or ablation alone attempted to determine whether single therapy would be as efficacious. Although this clinical series was larger than others previously reported, the sample size was insufficient to determine the clinical risk factors associated with later bleeding in 7% of the patients.

Conclusions

A two-step protocol of immediate ultrasound-guided polidocanol/CO₂ foam sclerotherapy, followed within 8 weeks by endovenous ablation of incompetent axial veins and concurrent microphlebectomy, provided effective control of bleeding with durable longer-term freedom from recurrent hemorrhage.

Author Contributions

Conception and design: RB

Analysis and interpretation: RB, DF, SB, JB

Data collection: RB, DF, SB, GE, KWF

Writing the article: RB

Critical revision of the article: RB, DF, SB, GE, KWF, JB

Final approval of the article: RB, DF, SB, GE, KWF, JB

Statistical analysis: RB, SB, JB

Obtained funding: Not applicable

Overall responsibility: RB

Disclosures

None.