Chapter 54
Superficial Thrombophlebitis
Suman Rathbun
Based on a chapter in the seventh edition by Anil P. Hingorani and Enrico Ascher
Superficial venous thrombophlebitis (SVT) has been the focus of increased attention because of recognition of the potential morbidity and mortality associated with it. A global disorder, SVT develops in approximately 125,000 people per year in the United States but is nonetheless underestimated because many cases go unreported.1
Traditional teaching suggested that SVT is a self-limited process of little consequence and of small risk; however, new evidence on the natural history of SVT has led to improvements in evaluation and treatment. A meta-analysis reported a 6% to 44% incidence of deep venous thrombosis (DVT), a 20% to 33% incidence of asymptomatic pulmonary embolism (PE), and a 2% to 13% incidence of symptomatic PE in patients in whom SVT is diagnosed.2 Improved diagnostic evaluation of SVT with duplex scanning, lung scanning, and blood tests has helped identify predisposing risk factors and potential complications.
This chapter examines current data regarding SVT and its management with the goal of improving recognition and treatment of the underlying disorders to prevent recurrence and its life-threatening complications.
Epidemiology and Pathogenesis
Although SVT is a frequently observed condition, its incidence and prevalence have never been adequately assessed. The classic Tecumseh Community Health Study from 1973 reported that the incidence of SVT increases with age in men from 0.05 per 1000 per year in their third decade to 1.8 per 1000 per year in their eighth decade. In women, the incidence similarly rises from 0.31 per 1000 per year to 2.2 per 1000 per year from their third decade to their eighth decade.3 Moreover, other studies have also demonstrated an increased prevalence of SVT in women.4,5 Overall, the incidence of lower extremity SVT has been reported to be 3% to 11% in the general population.
A useful classification is the recognition that SVT may occur in two forms, with and without varicose veins; alternatively, SVT may be primary, involving the vein wall only, or secondary, involving a more systemic inflammatory process. Primary SVT is most common in the saphenous veins and their tributaries, followed by the upper extremity cephalic and basilic veins. The great saphenous vein (GSV) is affected in 60% to 80% of cases, followed by the small saphenous vein (SSV) in 10% to 20% and bilateral lower extremity SVT in 5% to 10%.6 Patients with varicose veins are affected far more frequently than in the general population, with a prevalence of SVT ranging from 4% to 62%.3,5,6
Risk Factors
Risk factors can be classified in accordance with Virchow’s triad: endothelial injury from trauma or insertion of venous catheters; venous stasis as seen in varicosities; and hypercoagulable states such as factor V Leiden, prothrombin G mutation, protein C and protein S deficiency, antithrombin III abnormalities, and malignant neoplasms including both solid tumors and hematologic conditions of Hodgkin’s lymphoma, leukemia, thrombocytosis, polycythemia vera, cryoglobulinemia, and nocturnal paroxysmal hemoglobinuria.6,7 Patients in whom SVT develops without an inciting physical event or varicosities may need to be fully evaluated for the presence of such disorders.
Other secondary causes for the development of SVT include the use of oral contraceptives, hormonal replacement therapy, pregnancy, obesity, prolonged immobilization, recent surgery, trauma, sclerotherapy, history of venous thromboembolism (VTE), some drugs (e.g., diazepam, amiodarone, vancomycin, heroin, some chemotherapy), and intravenous catheter use with or without bacterial infection.3,5–8 In addition, patients with autoimmune disorders including systemic lupus erythematosus and vasculitis, such as Behçet’s disease and Buerger’s disease, have also been identified as being susceptible to the development of SVT. A 2006 review of 2319 patients with Behçet’s disease found that 14.3% of these patients had vascular involvement. Of these 332 patients, 53.3% had SVT and 29.8% had DVT.9 Patients with Buerger’s disease appear to have an increased incidence of SVT because the inflammatory process involves small arteries and veins of the extremities. It can be diagnosed from biopsy findings of acute superficial thrombophlebitis showing the characteristic acute-phase lesion—inflammation of all three layers of the vessel wall with occlusive cellular thrombosis.10
Involvement of Deep Veins
The main concern related to SVT is the likelihood that the thrombus will extend into the deep veins, causing DVT and potential PE. Several older studies have evaluated this risk especially in relation to the proximity of GSV SVT to the deep veins. Chengelis and colleagues observed 263 patients with isolated SVT and performed follow-up ultrasound at 2 to 10 days (mean, 6.3 days).11 Thirty (11%) had progression to DVT while not receiving anticoagulation. The most common site was propagation of the SVT in the GSV into the common femoral vein. In a small retrospective review, 185 patients with SVT and 370 age- and sex-matched controls were evaluated.12 A minority (13%) received any nonsteroidal anti-inflammatory drug (NSAID) or rarely low-molecular-weight heparin (LMWH) therapy. At 6 months, overall 2.7% had developed DVT and 0.5% PE. SVT conferred a 10 times increased risk for development of DVT compared with controls without SVT. This study represents an estimate of the natural history of SVT because few patients received pharmacologic management. Another focused study evaluated the incidence of PE in 21 patients with isolated SVT in the thigh.13 Nuclear perfusion lung scans were performed within 3 hours of SVT diagnosis and demonstrated PE in seven patients (33%), including one symptomatic patient. Of note, there were no significant differences in the distance of the SVT from the common femoral vein or the presence of common risk factors for thrombosis compared with those with SVT who did not have PE. Although the study was small, these findings highlight the potential serious consequences of SVT and suggest the need for anticoagulant therapy to prevent these complications.
Two large studies have evaluated the natural history of SVT in patients, most of whom received medical therapy. In the POST (Prospective Observational Superficial Thrombophlebitis) trial, Decousus and colleagues prospectively observed a cohort of 844 consecutive patients with symptomatic SVT of the lower limbs confirmed by ultrasound testing.14 Patients with surgery within 10 days or SVT due to sclerotherapy in the previous 30 days were excluded. Patients were initially assessed for concomitant DVT, and then those with isolated SVT (634 patients) were observed with ultrasound again at 10 days and then at 3 months. A secondary outcome was overall mortality at 3 months. DVT or PE was confirmed in 24.9% of patients with SVT (82 patients with proximal DVT, 33 patients with PE). In 41.9% of patients with DVT, both proximal and distal, the DVT was not contiguous with the SVT. Of 634 patients with SVT, 90.5% received one or more anticoagulant medications, mostly LMWH administered at therapeutic doses (62.9%). Sixty patients (10.2%) had venous surgery. Fourteen patients were lost to follow-up. Of the remaining 586 patients, 58 (10.2%) had thromboembolic complications, including 7 with symptomatic proximal DVT, 3 with PE, and 5 with extension of SVT to DVT. Multivariate analysis showed that male sex, history of DVT or PE, previous cancer, and no varicose veins were independent risk factors for symptomatic VTE at 3 months, including recurrence or extension of SVT.
The OPTIMEV study evaluated 788 patients with SVT of 8256 patients who were referred for VTE.15 The median age of these patients was 65 years; 36% were men, and 16% were inpatients. Similar to the POST study, 29% were found to have concomitant DVT. Patients with both SVT and DVT had risk factors of presence of non-varicose veins involved, age older than 75 years, inpatient status, and active cancer compared with isolated SVT, which was independently associated with anticoagulant treatment at inclusion and pregnancy or postpartum state. Compared with SVT of varicose veins, SVT in non-varicose veins was a strong risk factor for concurrent DVT but did not confer a higher risk of 3-month adverse outcomes of death, VTE recurrence, and bleeding. In this study, 76% of those with isolated SVT were treated with anticoagulants, 92.5% with full-dose LMWH.
Thrombophilia
More recently, there has been greater awareness of the presence of hypercoagulable states in patients with SVT. These patients have a higher probability for development of DVT and recurrent SVT and may require long-term anticoagulation to prevent complications. Milio and colleagues evaluated 107 patients with unprovoked SVT for common thrombophilic conditions.16 The patients underwent duplex evaluation at baseline and every 48 hours for 8 days with notation of whether the SVT occurred in varicose or non-varicose veins. In the overall cohort, factor V Leiden was detected in 22.4%, MTHFR in 17.7%, and factor II G20210A mutation in 8.4%. Patients with thrombophilia and SVT in non-varicose veins had a higher rate of extension of thrombus to deep veins. However, because all patients were treated with either LMWH or NSAIDs and the results were not analyzed by treatment group, it is difficult to make definitive conclusions about the role of thrombophilia and DVT extension. Similar studies have supported the presence of acquired and inherited thrombophilic disorders (such as factor V Leiden and prothrombin G20210A gene mutations; deficiencies of antithrombin, heparin cofactor 2, protein C, and protein S; lupus anticoagulant; anticardiolipin antibodies; and abnormal fibrinolytic activity) as being a risk factor for the development of SVT.3,5,9,10,17–19
Even though a great deal of literature has described the pathophysiologic mechanism and various changes that take place in leukocyte–vessel wall interactions, cytokines/chemokines, and various other factors involved in the development and resolution of DVT, data on changes in patients with SVT were not identified. Although some authors have alluded to the observation that the underlying pathologic processes of SVT and DVT may be analogous, this viewpoint remains mostly unsupported to date.
Clinical Presentation
Evaluation for SVT by physical examination is based on the presence of erythema and tenderness in the distribution of the superficial veins, with the thrombosis being suspected by a palpable cord. Pain, erythema, and swelling are the most common symptoms.20 A number of conditions have unique risk factors for clinical presentations of SVT.
Superficial Thrombophlebitis with Varicose Veins
The most common predisposing risk factor for the development of SVT is varicose veins. It has been reported that DVT will develop in only 3% to 20% of SVT patients with varicose veins compared with 44% to 60% of SVT patients without varicose veins.21,22 Therefore, it appears that the pathophysiologic mechanism in patients with varicose veins may be different from that in patients without varicose veins.
It is essential to address patients with SVT involving varicose veins. This type of SVT may remain localized to the cluster of tributary varicosities or may extend into the GSV.5,6,10,23,24 SVT of varicose veins themselves may occur without antecedent trauma. SVT is frequently found in varicose veins in conjunction with venous stasis ulcers. This diagnosis should be confirmed by duplex ultrasound because the degree of SVT may be much greater than that based solely on clinical examination. SVT in varicosities may be manifested as tender nodules with localized induration and erythema.
Traumatic Thrombophlebitis
Traumatic SVT is often seen in individuals using drugs or undergoing drug therapy in a hospital or outpatient setting. It is associated with direct endothelial injury from the intravenous catheter used for the infusion of medications and irritating solutions, particularly when the indwelling catheter has been in place for long periods. Its onset is usually heralded by the development of pain, tenderness, and erythema at the site of catheter insertion or infusion. Treatment usually consists of cessation of the infusion, removal of the offending access device, and sometimes anticoagulation, depending on the severity of symptoms and underlying hypercoagulable condition. The induration may take up to weeks to months to resolve.
Septic and Suppurative Thrombophlebitis
Suppurative SVT is also associated with the use of an intravenous cannula; however, it may cause additional morbidity because of its association with septicemia. Signs and symptoms of suppurative SVT include pus at an intravenous site, fever, leukocytosis, and local intense pain.25 Aerobic, anaerobic, and mixed infections have been reported. Organisms associated with suppurative SVT include Staphylococcus aureus, Pseudomonas spp, Klebsiella spp, Enterococcus spp, Fusobacterium spp, and recently fungi such as Candida spp.24,25 Treatment begins with removal of the foreign body and intravenous administration of antibiotics. Excision of the vein is rarely needed to clear the infection.
Migratory Thrombophlebitis
Migratory thrombophlebitis was first described by Jadioux in 1845 as an entity characterized by repeated thrombosis developing in superficial veins at varying sites but most commonly in the lower extremity.24 This entity may be associated with carcinoma (Trousseau’s syndrome) and may precede diagnosis of the carcinoma by several years. Consequently, evaluation for occult malignant disease is warranted when the diagnosis of migratory thrombophlebitis is made. Migratory thrombophlebitis also occurs in the presence of some forms of vasculitis, such as Behçet’s disease, Buerger’s disease, and polyarteritis nodosa.7
