The effect of venous thrombus location and extent on the development of post-thrombotic signs and symptoms

Patients and methods 

Patients with a first episode of acute DVT in the lower extremity were included. These patients presented with pain, swelling, or signs and symptoms of PE of <10 days. Diagnosis of DVT was by duplex ultrasound (DUS) imaging. In some patients with inferior vena cava (IVC) and iliac thrombosis, diagnosis was by venography, either by computed tomography (CT) or magnetic resonance imaging, or phlebography, but these were confirmed by DUS. Patients were separated into two groups. Group A patients had isolated thrombosis in one venous segment (ie, in the popliteal vein only). Group B patients had at least two different levels involved (ie, the popliteal and femoral veins). Both groups were treated with heparin and warfarin.

Excluded from the study were patients with previous DVT, thrombolysis, signs and symptoms of CVD before thrombosis, a short life expectancy (<2 years), immobility (unable to walk), major bleeding requiring interruption of anticoagulation, and those with an ankle-brachial index (ABI) <0.8. Patients with CVD were excluded by clinical examination and DUS. All patients had to have at least 1 year of follow-up by clinical examination and DUS.

The signs and symptoms of CVD were recorded using the CEAP system.5 Causes for thrombosis were recorded in detail. This included most of the well-known factors for VTE, such as trauma, surgery, catheterization, thrombophilia, family history, malignancy, and prolonged travel. Family history was positive when at least one first-degree relative had an episode of VTE during traveling when it was uninterrupted for >4 hours.

Symptoms of itching, burning sensation, pain, heaviness, and venous claudication were recorded. Venous claudication was considered to be present when patients were unable to walk more than a block without a pain despite having an ABI of >0.8. These patients had swelling and the limb was feeling tighter after walking.

Distribution and extent of reflux, obstruction, and recurrent DVT was assessed by DUS. Patients with signs and symptoms of PE underwent spiral CT. Recanalization was considered complete when no intraluminal material was detected, nearly complete when wall thickening or a few intraluminal webs were seen, partially complete when intraluminal material was detected, and absent when the whole lumen was occluded. Limbs were separated into normal, those having reflux, obstruction, and those with combined reflux and obstruction.

DUS was performed with multifrequency linear array transducers. Curvilinear array of 2- to 3-MHz was used in obese patients and for the iliocaval veins. Detection of DVT was performed in the supine position. IVC and iliac vein obstruction was tested in the supine and in the lateral decubitus position with the arm placed behind the head ipsilateral to the side of imaging. The infrainguinal veins were examined in the supine position. When the imaging was not adequate, such as the inability clearly delineate a duplicated femoral or popliteal vein, it was performed in the sitting or standing position. The reflux test was performed in the standing position. If the patient was unable to stand for long, the groin area was examined in the standing position and the rest of the examination was in the sitting position.

The cutoff value for reflux was set at 1 second in the common femoral, femoral, and popliteal veins; at 0.5 seconds in the deep femoral, deep calf veins, and superficial veins; and at 0.35 seconds in the perforator veins.6 Manual compression was used in all areas in all venous segments. If this maneuver had a negative result in the proximal femoral and deep femoral veins and higher, the Valsalva maneuver was used as well. In limbs with significant edema, dorsi/plantar flexion was used when the result of manual compression was negative.

The criteria used for the diagnosis of acute and chronic thrombosis have been published elsewhere in detail.7 The diagnosis of recurrent thrombosis was made by using the following criteria: extension of thrombus of >9 cm, noncompressibility of a vein segment that has been previously recanalized, and enlargement of the thrombus thickness by ≥2 mm.8, 9, 10

After the initial DUS at diagnosis, two clinical examinations and a repeat DUS were performed in the first 6 months, and a clinical examination and DUS were repeated at 1 year and then yearly thereafter. Because these patients had known thrombosis, DUS was done for both obstruction and reflux. Clinical examination and DUS were also done whenever the patients had new signs and symptoms of VTE.

Results 

From an original 192 consecutive patients, the study included 120 limbs of 105 patients (59 men and 46 women) who were a mean age of 54 years (range, 23-82 years). Groups A and B were comparable in age and sex. Excluded were 45 patients owing to previous DVT (n = 8), CVD before DVT (n = 11), early fatal PE (n = 2), short life expectancy (n = 6), immobility (n = 3), major bleeding (n = 3), and thrombolysis treatment (n = 12). The remaining 42 patients were excluded because they did not complete at least 1 year of follow-up, either because they chose not to come (n = 36) or because they moved to another area (n = 6). The causes for thrombosis are reported in Table I. Thrombophilia alone or combined with cancer had a trend for a higher prevalence in group B (20 of 79) compared with group A (6 of 41; P = .2). At presentation, group A and group B had similar duration of signs (range, 1-9 days vs 1-7 day) and symptoms (range, 2-8 days vs 1-6 days). The median follow-up was 3.4 years (range, 1.2-7 years).

Table I. Causes of thrombosis

	Cause	Group A, No.	Group B, No.
	Family historya	3	6
	Trauma	6	8
	Surgery	8	11
	Thrombophilia	4	17
	Travel >4 hours	2	5
	Extrinsic compression	2	4
	Catheter	4	5
	Tumor	2	3
	Unknown	7	8
	Total	38	67

a Family history was present in 5 additional patients in group A and in 7 in group B that had also one of the other factors present.

The distribution and extent of thrombosis as well as the prevalence of signs and symptoms are reported in Table II, Table III. The extent of thrombosis was longer in group B because these patients had multisegmental thrombosis and also had more symptomatic limbs (21 of 41 vs 71 of 79, P < .001). This was true for all symptoms, whether examined together or separately.

Table II. Location of thrombi and signs and symptoms in patients with isolated vein thrombosis

	Segment	Patients, No.	Limbs, No.
			Signs	Symptoms
	Inferior vena cava	3	4a	2
	Common iliac vein	2	2	1
	External Iliac vein	2	2	1
	Common femoral vein	6	6	4
	Femoral vein	3	1	1
	Popliteal vein	7	6	5
	Calf veins	15	7	7
	Total	38	28	21

The total number of patients was 38 and the limbs 41.

a Patients with inferior vena cava thrombosis could have bilateral signs and symptoms.

Table III. Distribution and extent of thrombosis, signs, and symptoms in patients with multisegmental deep venous thrombosis

	Segments	Patients, No.	Limbs, No.
			No.	Signs	Symptoms
	IF ± IVC	7	9	8	8
	IF + fem-pop	9	10	8	9
	IF + fem-pop + calf	11	13	13	13
	Fem-pop	16	19	14	18
	Fem-pop + calf	24	28	26	23
	Total	67	79	69	71

Fem-pop, Femoral-popliteal; IF, iliofemoral; Pop, popliteal.

The CEAP classification for both groups in relation to the distribution of thrombosis is summarized in Table IV, Table V. More advanced PTS was found in group B (61 of 79) vs group A (26 of 41; P < .001), including the prevalence of skin damage and ulceration (29 of 79 vs 6 of 41, P = .019). Limbs with calf DVT that had focal thrombosis or only one vein involved remained in CEAP class 0 and were most often asymptomatic, and those with thrombosis in multiple calf veins developed symptoms and were most often in CEAP class 3 and, occasionally, in class 4. Venous claudication was exclusively found in group B (7 of 79) and was present only when iliac veins were involved.

Table IV. CEAP classification in patients with isolated vein thrombosis

	Segment	Limbs, No.
		Total	Class 0-1	Class 2-3	Class 4a	Class 5-6
	Inferior vena cava	6	2	3	1	0
	Common iliac vein	2	0	2	0	0
	External iliac vein	2	1	1	0	0
	Common femoral vein	6	1	4	1	0
	Femoral vein	3	2	1	0	0
	Popliteal vein	7	1	4	2	0
	Calf veins	15	8	5	2	0
	Total	41	15	20	6	0

a Four limbs were in CEAP class 4A and 2 were in class 4B.

Table V. CEAP classification in patients with vein thrombosis in multiple segments

	Segment	Limbs, No.
		Total	Class 0-1	Class 2-3	Class 4a	Class 5-6
	IF ± IVC	9	1	4	3	1
	IF + fem-pop	10	2	4	3	1
	IF + fem-pop + calf	13	0	6	5	2
	Fem-pop	19	5	8	4	2
	Fem-pop + calf	28	2	18	6	2
	Total	79	10	40	21	8

Fem-pop, femoral-popliteal; IF, iliofemoral; Pop, popliteal.

a Thirteen limbs were in CEAP class 4A and 8 were in class 4B.

Calf thrombosis in one vein segment resolved faster and produced no signs and symptoms of CVD. Patients with multiple calf vein involvement presented with swelling and, in two occasions, skin changes. The presence of calf vein thrombosis in patients with proximal DVT increased the likelihood they would present with post-thrombotic signs and symptoms (proximal veins, 30 of 38 vs proximal and calf veins, 39 of 41; P = .04). Only two of 28 patients with calf vein involvement in group B were in class 0 but were symptomatic, with itching and burning sensation in one and pain in the other; whereas in eight of 38 limbs with proximal DVT alone in class 0, four patients were symptomatic (P = .17).

Recurrent thrombosis had a trend for a higher prevalence (P = .39) in group B (16 of 79) compared with group A (5 of 41). The total number of recurrent thrombotic events was 20, because two patients had two recurrent episodes and another patient had three recurrent episodes. Clinical symptoms of PE were found in four patients (9.7%) in group A and in 16 (20.3%) of group B (P = .2). On spiral CT, PE was present in one of four in group A and in six of 16 in group B (P = .6 and P = .2 among all patients in both groups). Because of the study design, two additional patients with early fatal PE in group B were excluded from analysis. One fatal PE occurred at 42 months in group B.

Venous pathology was more common in group B (61 of 79) compared with group A (15 of 41; P < .0001). In group B, reflux was found in 23 limbs, obstruction in 6, and a combination of reflux and obstruction in 32; whereas in group A, reflux was found in 6, obstruction in 2, and a combination of reflux and obstruction in 6. Five of the six limbs with skin damage (CEAP classes 4 to 6) in group A had reflux and obstruction, and 24 of the 29 limbs with skin damage in group B had reflux and obstruction; therefore, limbs with both reflux and obstruction were more likely to develop skin damage in either group (group A, 5 of 6 vs 1 of 35, P < .0001; group B, 24 of 29 vs 5 of 50, P < .0001).

All eight limbs in which an ulcer developed had a multisegmental thrombosis (group B), of which four had a recurrent DVT and one was in a patient who was obese. All but one of these patients were aged >55 years. One ulcer developed at 2 years, one at 3 years, and six at >3 years. Of the limbs that developed skin changes (6 in group A and 21 in group B), 2 appeared in the first year, 3 in the second year, 9 at 3 years, and 13 at >3 years.

Discussion 

The current study has a well-defined patient population in which the location and extent of thrombosis at baseline were established, coupled with a rigorous follow-up in which signs, symptoms, and severity of PTS, venous claudication, recurrent VTE, and clinically relevant PE were documented. The use of the CEAP classification system in a prospective fashion has allowed the study of the natural history of VTE in a consistent manner with a more valid comparison among the different groups.

The patients were selected to have a reasonable life expectancy such that enough time would elapse to document the frequency and severity of PTS. Patients with impaired mobility were excluded because this by itself would have been a significant risk factor for developing CVD. Patients with previous thrombosis, CVD, and an ABI <0.8 were also excluded to permit the study of the DVT impact on an unaffected extremity. The signs and symptoms of VTE were present for <10 days, allowing a uniform time effect of thrombosis in the involved extremity. Recurrent VTE was well documented and reported in relation to previous distribution of DVT.

Patients with symptomatic DVT are at an increased risk for recurrence.11 It has been shown that PTS occurs in approximately 30% to 60% of patients 5 to 10 years after the initial DVT1, 2, 12 and is approximately six times more likely to occur in patients with ipsilateral recurrent DVT.1 In addition, Johnson et al12 showed that abnormal findings in PTS patients are more common in the deep venous segments of the lower extremities compared with extremities without PTS.12 In the current study, skin damage developed in 35 limbs (29%) at a mean follow-up of 3.4 years. The prevalence of skin damage was more common in group B, and the time taken for its development was >3 years in most limbs. Mild skin changes (CEAP class 4A) were noted in 15 limbs.

Author contributions