Definition of post-thrombotic syndrome of the leg for use in clinical investigations: a recommendation for standardization

Authors


Susan R. Kahn, Division of Internal Medicine, Center for Clinical Epidemiology and Community Studies, Jewish General Hospital, McGill University, 3755 Cote Ste. Catherine Rm. A-114, Montréal, QC, H3T 1E2, Canada.
Tel.: +1 514 340 7587; fax: +1 514 340 7564.
E-mail: susan.kahn@mcgill.ca

Abstract

Summary.  The post-thrombotic syndrome (PTS) is increasingly recognized to be a common and important complication of deep venous thrombosis (DVT). Because there is no ‘gold standard’ objective test to establish its presence, PTS is diagnosed primarily on the basis of the presence of typical symptoms and clinical signs in a limb that was affected by DVT. As a wide variety of definitions of PTS have been used by researchers, it is difficult to compare data across studies and to formally combine data in meta-analyses. In a step towards standardization of the measurement of PTS in clinical studies, available scales and evidence to support their utility to diagnose PTS and to classify its severity were reviewed and discussed at the Control of Anticoagulation Subcommittee of the International Society on Thrombosis and Haemostasis (Vienna, July 2008).

Introduction

The post-thrombotic syndrome (PTS) is the most common complication of venous thromboembolism (VTE), occurring despite optimal anticoagulant therapy in 20%–40% of patients within the first 1–2 years after deep venous thrombosis (DVT) of the lower limb [1,2]. Severe PTS, which may include ulceration, occurs in 5%–10% of DVT patients. PTS leads to suffering and disability and is costly to society [3,4]. Recent research has shown that quality of life among patients with PTS is poorer than that among patients of similar age with arthritis, chronic lung disease or diabetes, while patients with severe PTS report quality of life that is comparable to persons with angina, cancer or congestive heart failure [5,6].

In spite of its high incidence and long-term burden, PTS has received little attention from researchers. Indeed, among the numerous published large clinical trials that have evaluated the effectiveness of various anticoagulant regimens to treat DVT, none has evaluated PTS as a prespecified study endpoint. While it is encouraging that in recent years a number of trials have evaluated non-pharmacologic ways to prevent or manage PTS [7–12], these have usually been small and, overall, there is a paucity of good quality evidence on the epidemiology, prevention and management of PTS.

Rationale for standardizing the definition of PTS

An important barrier to research on PTS has been a lack of standardization of its definition. Lack of standardization of PTS may also partly explain the wide-ranging differences in rates of PTS reported in various studies. A standardized way to define and measure the severity of PTS would facilitate investigation of this important condition, permit between-study comparisons of findings and enable data from individual studies to be combined in meta-analyses.

Methods

The objective of this ISTH SSC statement is to propose a uniform definition of PTS for use in clinical studies. The definition we propose is based on discussions held at a symposium on the standardization of assessment tools for PTS that was part of the Control of Anticoagulation Subcommittee at the 54th Scientific and Standardization Committee Meeting of the International Society on Thrombosis and Haemostasis (5 July 2008, Vienna, Austria). This process involved a review of existing PTS measures, how PTS has been defined in past and ongoing studies, and a discussion of evidence evaluating the psychometric properties of these measures for diagnosing and classifying the severity of PTS.

Presenters at this session included S. Kahn, H. Partsch, S. Vedantham and C. Kearon. P. Prandoni contributed his expertise to this symposium but was unable to attend in person. C. Kearon and W. Ageno were co-chairs of the session. During the discussion period, attendees at the session contributed comments and suggestions. A draft definition was circulated to co-chairmen of the subcommittee and the text was revised to incorporate their input.

Brief description of PTS measures reviewed

Six clinical scales have been used to define PTS in clinical studies. Of these, three were developed for chronic venous disease generally (Widmer classification [13], CEAP (clinical, etiologic, anatomic, pathophysiologic) classification [14] and VCSS (venous clinical severity score) [15]), while three were developed specifically to diagnose PTS after objectively-diagnosed DVT (Brandjes scale [16], Ginsberg measure [8] and Villalta scale [17]). In brief, the Widmer scale [13] classifies patients into three descriptive categories of chronic venous disease according to the presence of various clinical signs. In addition, five symptoms are described: pain, heaviness, heat, tension, and tiredness of limb. The CEAP classification, originally introduced in 1994 [14] and revised in 2004 [18], categorizes patients into one of eight clinical classes of increasing severity based on clinical signs, with modifiers that reflect the underlying etiology (congenital, primary, or secondary to DVT), anatomic distribution (superficial, deep, or perforating veins) and pathophysiologic condition (reflux, obstruction, or both). Although the scale classifies patients as symptomatic or asymptomatic, it does not specify individual symptoms or grade their severity. The venous clinical severity score [15] incorporates elements of CEAP and other criteria (number, size and duration of ulcers, use of compressive therapy, and a single symptom, pain) to yield a numeric score that can be used to categorize the presence and severity of chronic venous disease [19]. The Brandjes’ scale [16] uses separate scales to categorize patients as having no, mild-to-moderate or severe PTS that include items on symptoms, signs and differences in calf circumference. Points for the items are summed into a total score, and cut-off values are used to classify the presence of mild-to-moderate PTS and severe PTS, respectively. The Ginsberg measure [8] diagnoses PTS if the patient reports leg pain and swelling of at least 1 month duration that is typical in character (worse at end of day or with prolonged sitting or standing, and better after a night’s rest and leg elevation) and that occurs 6 months or more after the acute DVT; a global rating scale that grades whether symptoms have improved or worsened since the last assessment can be co-administered to monitor change in PTS severity and response to therapy. The Villalta scale [17] classifies patients as having or not having PTS, and rates its severity, based on a score that is the sum of severity ratings of five venous symptoms and six clinical signs. More detailed descriptions and discussions of these scales can be found in the original reports [8,13–19] as well as in recent reviews [20–22].

Recommended definition of PTS

As PTS is a syndrome that produces a range of symptoms and clinical signs, there is wide agreement that categorization of the presence and severity of PTS should incorporate the assessment of both patient symptoms and clinical signs. Indeed, in a recent international ad-hoc conference organized by the European Venous Forum, PTS was defined as ‘chronic venous symptoms and/or signs secondary to deep vein thrombosis’ [23]. Of the previously noted scales, our group, representing the SSC of the ISTH, judged that the Villalta scale was the scale most suitable for defining the presence and severity of PTS after objectively diagnosed DVT (see Table 1). The basis for this recommendation is that (i) the Villalta PTS scale takes into account the range and severity of both subjective symptoms and physical signs of PTS; (ii) the validity, inter-rater reliability, responsiveness to clinical change and acceptability (i.e. ease of use) of this scale have been systematically studied and found to be high [24]); (iii) physician-nurse reliability has been demonstrated, hence the measure can be used by properly trained non-physician research staff; (iv) the scale has the advantage that it can be used as a binary, categorical or continuous outcome measure; and (v) the scale has successfully been used in clinical trials and longitudinal cohort studies of PTS in a number of countries [1,2,7,9,25,26].

Table 1.   Villalta’s PTS scale [17]
Symptoms and Clinical signsNoneMildModerateSevere
  1. Points are summed into a total score (range 0–33). PTS is defined by a total score of ≥5 or the presence of a venous ulcer. PTS is classified as mild if the Villalta score is 5–9, moderate if the Villalta score is 10–14, and severe if the Villalta score is ≥15 or a venous ulcer is present. To use the Villalta score as a continuous measure, it is recommended that patients who meet criteria for severe PTS based solely on the presence of an ulcer (i.e. total Villalta score is <15) be assigned a score of 15 [2].

Symptoms
 Pain0 points1 point2 points3 points
 Cramps0 points1 point2 points3 points
 Heaviness0 points1 point2 points3 points
 Paresthesia0 points1 point2 points3 points
 Pruritus0 points1 point2 points3 points
Clinical signs
 Pretibial edema0 points1 point2 points3 points
 Skin induration0 points1 point2 points3 points
 Hyperpigmentation0 points1 point2 points3 points
 Redness0 points1 point2 points3 points
 Venous ectasia0 points1 point2 points3 points
 Pain on calf  compression0 points1 point2 points3 points
Venous ulcerAbsentPresent

It is recommended that the following criteria are used to define and classify the severity of PTS using the Villalta PTS scale (see Table 1):

  • 1PTS definition (present/absent): PTS is present if the Villalta score is ≥5, or a venous ulcer is present, in a leg with previous DVT.
  • 2PTS severity as a category (none, mild, moderate, severe): PTS is classified as mild if the Villalta score is 5–9, moderate if the Villalta score is 10–14, and severe if the Villalta score is ≥15 or if a venous ulcer is present, regardless of the Villalta score.
  • 3PTS severity as a continuous measure: The Villalta score serves as a continuous measure of PTS severity, ranging from 0 to 33. A limitation of the Villalta scale as a continuous measure of PTS severity is that no points are attached to the presence of an ulcer. To circumvent this problem, patients with a venous ulcer after DVT who have a Villalta score of <15 are assigned a score of 15 [2].

If assessments of PTS (i.e. Villalta scale measurements) are performed in a study at a number of predefined time points (e.g. every 3 or 6 months), we recommend that the frequency of PTS be expressed as a cumulative incidence with specification of the interval between testing and the duration of follow-up after DVT (e.g. cumulative incidence of 20% at 2 years with assessments performed every 6 months). We recommend that the presence of PTS on a single assessment is used to diagnose PTS.

Acute symptoms of DVT may take months to resolve and, indeed, acute symptoms may transition to chronic symptoms without a symptom-free interval [2]. We recommend that PTS is not diagnosed before 3 months to avoid inappropriately attributing acute symptoms and signs to the PTS. This is particularly important when the cumulative incidence of PTS is being measured as, once patients are diagnosed with PTS, those patients will be counted as having PTS even if symptoms and clinical signs subsequently resolve and do not recur.

To minimize variability in the assessment of PTS using the Villalta scale and to reduce the frequency of missing data, the following method of administering the Villalta scale is recommended:

  • 1During the initiation phase of a study, assessors (e.g. study coordinator) should receive training on the administration of the Villalta scale. A visual guide to aid with rating of the severity of clinical signs is available on-line.
  • 2Study subjects should be asked to not wear compression stockings on the day of an assessment, and should be examined in the afternoon to facilitate manifestations of symptoms and signs of PTS.
  • 3The subject should be instructed not to disclose which leg had DVT to the assessor of clinical signs.
  • 4The subject should be seated facing the assessor in a well-lit room, with his/her legs unclothed to the mid-thigh.
  • 5First, the subject records his/her ratings for each of the five symptoms required for the Villalta PTS scale on a PTS Symptoms Form (see Table 1). The subject should not have access to his/her ratings from prior study visits.
  • 6Subsequently, and without access to the subject’s symptom ratings, the assessor rates each of the six clinical signs on a PTS Clinical Signs Form (see Table 1). The assessor should not have access to clinical sign ratings from prior study visits.
  • 7Before the subject leaves, study staff (e.g. the assessor of signs) should verify that all symptom items have been rated. If there are missing symptom ratings, the subject should be encouraged to complete the form (at the time of analysis, missing ratings for a symptom can be imputed by taking the mean of the rated symptoms).

It is recommended that the non-DVT leg is also assessed. This ensures that there are data available for the appropriate leg if an error was made about which leg had the DVT, facilitates blind assessment of clinical signs by the assessor, and provides a comparator to aid with interpretation of Villalta scale measurements.

Discussion

A number of points should be emphasized in relation to the above definitions. First, it is recognized that PTS, like other chronic diseases, is a not a static condition but can wax and wane over time [2,27]. Second, it is acknowledged that the symptoms and signs assessed by the Villalta PTS scale (also true for the other scales described above) are non-specific and may have causes other than PTS. Hence, the finding of a change in Villalta PTS score over time is expected to be more informative than a single PTS score at one assessment point. Also, lack of specificity of the symptoms and signs of PTS is expected to result in an overestimate of the proportion of patients who are categorized as having PTS [20]. This potential for overdiagnosis of PTS is less of a concern in the setting of a randomized trial than in an observational study, as in a randomized trial the potential for overdiagnosis should be present in all groups that are being compared. Finally, a limitation of the Villalta scale is that a healed ulcer (i.e. previous ulceration) is not considered independently of the six clinical signs that are assessed as part of the scale. Unlike for the Villalta scale, a healed ulcer is used in the CEAP scale’s description of chronic venous disease. As previously noted, the CEAP classification also documents important information (obtained by Duplex examination) about the anatomic distribution of the original DVT and whether there is residual thrombosis with obstruction or the presence of venous reflux. Use of the Villalta scale is not designed to capture this information; rather it is intended, in patients with a previously documented DVT, to identify the presence of PTS and quantify its severity. Consequently, we suggest that components of the CEAP classification provide information that is complimentary to the Villalta scale assessment.

We acknowledge that there is an arbitrary component to some of our recommendations, particularly those relating to: (i) estimating the frequency of PTS using cumulative incidence; (ii) diagnosing PTS based on a single Villalta assessment rather than requiring two consecutive assessments that correspond to PTS (i.e. Villalta score ≥5); and (iii) assigning a Villalta score of 15 to patients with a venous ulcer who have a calculated Villalta score of <15. The requirement for two consecutive Villalta scores of ≥5 to diagnose PTS, which has been used in some previous studies [1,6,7,25], can be problematic if there are missed assessments, and it introduces challenges when comparing results between studies, as the interval between assessments may vary from study to study.

These recommendations are designed to improve standardization in the diagnosis of PTS. However, we acknowledge that, even when the Villalta scale is being used to assess the presence and severity of PTS, depending on the question that the study is trying to answer, it may be preferable to define the presence of PTS differently (e.g. two consecutive Villalta scores of ≥5), or to select a primary outcome other than cumulative incidence (e.g. point prevalence of PTS at one or more time points). In these situations, we advise that consideration is given to reporting the PTS measurements that we have recommended as secondary assessments of PTS.

We hope that these recommendations will serve as a launching point for discussions about criteria for diagnosing the presence and severity of PTS with other groups that have an interest in PTS (e.g. dermatology, vascular surgery, phlebology and interventional radiology societies), and ultimately facilitate cross-disciplinary standardization of these criteria and collaborative PTS research.

Acknowledgements

S. R. Kahn is a recipient of a Senior Clinical Research Scientist Award from the Fonds de la Recherche en Santé du Québec. C. Kearon is supported by the Heart and Stroke Foundation of Canada and a Canadian Institutes of Health Research Team Grant in Venous Thromboembolism (FRN 79846).

Disclosure of Conflict of Interests

The authors state that they have no conflict of interest.

Ancillary