Venous reflux in patients with previous venous thrombosis: Correlation with ulceration and other symptoms

Transcription

1 Venous reflux in patients with previous venous thrombosis: Correlation with ulceration and other symptoms deep N. Labropoulos, BSc, M. Leon, MD, A. N. Nicolaides, MS, FRCS, O. Sowade, MSc, MB, BS, N. Volteas, MD, F. Ortega, MD, PhD, and P. Chan, MChir, FRCS, London, United Kingdom Purpose: Deep vein thrombosis (DVT) in many cases leads to chronic symptoms in the damaged leg, even though the affected veins have recanalized. The major hemodynamic defect in such recanalized veins is reflux. The incidence and extent of reflux has been studied in patients with proven DVT and correlated with concurrent symptoms. Methods: Two hundred seventeen limbs in 183 patients were examined by duplex scanning from January 1989 to October All limbs had previous DVT diagnosed by venography. Sites and extent (proximal, distal, or both) of reflux were identified by meticulous duplex scanning of the whole venous system and correlated with presenting symptoms. Results: The patients were classified into nine groups on the basis of the classification of the system involved (superficial, deep, or superficial and deep) and whether the reflux was found proximal or distal to the knee or both. Eighty-one limbs belong to chronic venous insufficiency class 1, 92 belong to class 2, and 38 belong to class 3. Reflux was confined to the deep venous system in 84 limbs (38.7%), to the superficial system in 31 (14.3%) limbs, and to both systems in 102 (47%) limbs. It was confined to proximal veins only in 48 (22.1%) limbs, distal only in 56 (25.8%) limbs and throughout the limb in 113 (52.1%) limbs. The incidence of swelling was increased by distal or a combination of proximal and distal reflux regardless of which system was involved. In limbs with superficial venous insufficiency (SVI) or deep venous insufficiency (DVI) only, the incidence of skin changes was not affected by the extent of reflux. However, in limbs with combined SVI and DVI, it was increased in the presence of reflux throughout the limb. Absence of distal reflux was associated with a low incidence of skin changes even in the presence of DVI. Ulceration increased with an increased extent of reflux in the presence of SVI. Absence of superficial reflux was associated with a low incidence, even in the presence of DVI. Conclusions: The data suggest that as far as the skin changes and ulceration are concerned, distal reflux and reflux in the superficial veins are more harmful than reflux confined to the deep veins, even when such reflux extends throughout the deep venous system. (J VAsc SUgG 1994;20:20-6.) In a number of patients with acute deep venous thrombosis (DVT) postthrombotic syndrome consisting of pain, swelling, skin changes and ulceration subsequently develops. It is largely accepted that From the Irvine Laboratory for Cardiovascular Investigation and Research, Academic Surgical Unit, and Vascular Unit, St. Mary's Hospital Medical School, Imperial College of Science Technology and Medicine, London. Reprint requests: Andrew N. Nicolaides, MS, FRCS, Irvine Laboratory, Academic Surgical Unit, and Vascular Unit, St. Mary's Hospital Medical School, Praed Street, London W2 1NY, United Kingdom. Copyright 1994 by The Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter /94/$ /1/ reflux resulting from valvular destruction 1,2 and obstruction from failure to recanalize: are mainly responsible for the development of postthrombotic sequelae. The relationship between the severity of postthrombotic problems and the sites of reflux has been well documented. In most of these studies, venous valvular incompetence and reflux were largely assessed with photoplethysmography and continuouswave Doppler techniques. 3-6 A reduction in the venous refilling time and an increase in the ambulatory venous pressure are characteristic of the postthrombotic limb with valvular incompetence. However, these physiologic

2 JOURNAL OF VASCULAR SURGERY Volume 20, Number 1 Labropoulos et al. 21 Fig. 1. a, Real time color-flow imaging of saphenofemoral juncfion.blue colorproduced by calf compression indicates flow in cephalad direction, b, Absence of flow in both common femoral and long saphenous veins on release of compression indicates competent valves. This picture is taken from normal subject with competent saphenofemoral junction, c, Red color in saphenofemoral junction on release of compression indicates incompetent valves. There is no reflux in common femoral vein. measurements are often not corrected with surgical procedures that aim to deal with valvular incompetence. 7,s This suggests that our understanding of the relationship between the sites of reflux and severity of the postthrombotic problem still remains inexact. The advent of color-flow duplex imaging (CFDI), which combines B-mode imaging with color-coded Doppler shift, has provided additional insight into this problem because a detailed anatomic and visual assessment of venous reflux has become possible. 9,1 The aim of this study was to determine the relationship between the symptoms of the postthrombotic limb and the anatomic extent of venous reflux. MATERIAL A N D M E T H O D S One hundred eighty-three patients with involvement of 217 limbs who had been referred to the vascular laboratory with a history of DVT were examined with CFDI between January 1989 and October There were 105 male (57.4%) and 78 female (42.6%) patients with a mean age of 56.2 years (range 17 to 81). All patients had previous DVT diagnosed by venography. The time interval between their phlebograms and duplex examination ranged from 6 months to 25 years. At the time of their visit, clinical signs and symptoms ofpostthrombotic sequelae such as swelling, pain, pigmentation, and ulceration were sought by history and clinical examination. The limbs were classified into different chronic venous insufficiency (CVI) classes according to their clinical presentation as it has been described by the Ad Hoc Committee For Reporting Standards of the Society for Vascular Surgery and the North American Chapter of the International Society for Cardiovascular Surgery. 11 CFDI was performed with the ATL ultramark 9 scanner with the 5 MHz linear array probe (Advanced Technology Laboratories, BotheU, Wash.). The saphenofemoral junction, the common femoral vein, and the origin of the deep and superficial femoral veins were examined with the patient in the standing position. The effect of Valsalva maneuver on the blood velocity followed by manual compression of the calf and sudden release was noted. Subsequently the femoropopliteal segment and the axial calf veins (posterior and anterior tibial, peroneal, and gastrocnemial veins) were examined with the patient in the sitting position facing the examiner and with the foot resting on a stool. In this position distal compression of the calf or foot was also undertaken. These positions ensured maximum filling of the veins, as well as allowing reflux to occur on release of the calf compression. The long and short saphenous veins were exam-

3 JOURNAl. OF VASCULAR SURGERY 22 Labropoulos et al july 1994 Fig. 2. a, Real-time imaging of common femoral vein bifurcation. Doppler sample volume is positioned in superficial femoral vein. b, Cephalad flow in vein is obtained during limb compression, and reversal of flow on release indicates incompetent valve. ined along their whole length with intermittent calf compression and sudden release. Foot compression was used when examining their distal ends. In addition compression of the superficial vein examined distal to the probe with sudden release was performed. The thigh portion of the long saphenous vein was examined with the patient in the standing position, whereas the calf portion and short saphenous were examined with the patient in the sitting position. Color-flow imaging was used initially to locate the appropriate artery (red), and then the adjacent veins were identified. On distal compression the color image obtained in the vein was different (blue) from that obtained in the artery, indicating that the flow was in the opposite direction to that of the artery. Absence of color on release of the compression indicated absence of reflux and by inference competent valves (Fig. 1, a and b). The appearance of red color in the vein on release of the compression indicated the presence of reflux (Fig. 1, c). This reflux was documented by recording the Doppler waveform (Fig. 2). Reflux in the vein tinder investigation was considered to be present if the duration of the distal flow was greater than 1 second. Reflux of shorter duration was considered physiologic, that is, the brief distal flow was recorded just before the closure of the valves) 2 The extent of reflux was characterized as proximal, distal, or both. It was considered to be proximal if it was present in the common femoral, superficial femoral, and popliteal veins up to tibial confluence or the long saphenous vein above the popliteal crease. Reflux was considered to be distal if it was present in the tibioperoneal trunk, anterior and posterior tibial, peroneal, gastrocnemial, and short saphenous veins, or the long saphenous vein below the popliteal crease. RESULTS All limbs had thrombosis cxtcnding above the kncc on vcnography performed at the timc of the acute DVT. The clinical status of our patients with regards to chronic venous insufficiency before the onset of thcir DVT was as follows: 184 limbs (84.8%) out of the 217 did not have any symptoms, varicose veins, or skin changes (class 0), and the remaining 33 limbs (15.2%) had symptoms and signs of CVI. Twcnty-onc of these limbs had mild CVI (class 1), and the remaining 12 had skin changes (class 2) but no ulceration. CFDI examination of the

4 Volume 20, Number 1 Labropoulos et al. 23 Table I. Incidence of reflux in the superficial and deep venous systems Location No. % DVI only SVI only SVI + DVI Total 217 Table II. Distribution of reflux in the proximal and distal venous segments Location No. of limbs % Proximal Distal Proximal and distal 11 ~ Total 217 subjects showed that reflux was present somewhere in all the limbs. Reflux was confined to the deep venous system in 84 limbs (38.7%), to the superficial system in 31 (14.3%) limbs, and was present in both superficial and deep systems in 102 (47%) limbs (Table I). The sites of reflux were proximal only in 48 limbs (22.1%), isolated distal reflux in 56 limbs (25.8%), and combined proximal and distal reflux in 113 limbs (52.1%) as shown in Table II. Symptoms and signs found on examination of the limbs are listed in Table III. Distribution of reflux in the superficial and deep system and its extent, whether proximal or distal, are shown in Table IV. It can be seen that limbs are classified into nine groups. The incidence of symptoms (swelling, skin changes, and ulceration) in relation to the anatomic extent of reflux, that is, each one of the nine groups of Table IV is shown on Tables V to VII. The incidence of swelling was increased by distal or a combination of proximal and distal reflux regardless of which system was involved (Table V). In limbs with superficial venous insufficiency (SVI) or deep venous insufficiency (DVI) only, the incidence of skin changes was not affected by the extent of reflux. However, in limbs with combined SVI and DVI, it was increased in the presence of proximal and distal reflux. Absence of distal reflux was associated with a low incidence of skin changes even in the presence of DVI (Table VI). Ulceration increased with an increased extent of reflux in the presence of SVI. Absence of superficial reflux was associated with a low incidence even in the presence of DVI (Table VII). Seven patients (7 limbs) had venous claudication caused by proximal obstruction, and all of these had associated reflux. Five of these patients had occlusion of the iliofemoral segment. One patient had combined iliofemoral and inferior vena cava occlusion, and the remaining patient was admitted with femoropopliteal occlusion and associated occlusion of the deep femoral vein. Only one of these patients had reflux confined to deep veins. Table III. Incidence of symptoms in the limbs Symptoms No. of limbs % Swelling (class 1-3) Skin changes (class 2) Varicose veins (class 1-3) Ulceration (class 3) Claudication (class 1-2) Class 1, n = 88; class 2, n = 91; class 3, n = 38. DISCUSSION Various studies have shown that about two thirds of patients in whom postthrombotic symptoms develop have evidence of a clinically or phlebographically proven DVT. 2,1s,1. However, our study population was a select group of patients who had phlebographically proven DVT and were subsequently admitted with signs and symptoms of a "postthrombotic limb." The time interval between the phlebogram and their CFDI ranged from 6 months to 25 years. It is not surprising that symptoms were present in our patients as early as 6 months after their episode of DVT. Markel et al.ls had shown that the incidence of reflux was 17% in the first week after DVT, 37% after the first month, and 69% at the end of the first year. Our study differs from previous reports-in which venous reflux was assessed with photoplethysmography and continuous-wave Doppler scanning- in the use of CFDI.36 Duplex ultrasonography has been shown to be superior in assessing the degree and distribution of venous reflux than descending venography, 16 and the addition of CFDI shortens the time of examination and provides easier visualization of the deep calf veins. ~7 There were 33 patients who had symptoms of CVI before their DVT. Clearly reflux can exist independently of DVT, and symptoms in this group may not entirely be due to postthrombotic damage. It has also been shown that 15% of "normal people" have some degree of venous reflux, 18 mostly superficial. We have not analyzed this group separately,

5 24 Labropoulos et al July 1994 Table IV. Distribution and extent of reflux System Proximal Distal Proximal + Distal Total SVI only DVI only DVI + DVI Total Z ~2 217 Table V. Incidence of swelling in relation to location and extent of reflux (n = 168 limbs) Extent of reflux System Proximal Distal Proximal + Distal Total SVI only 2/5 (40%) 6/9 (66.7%) 10/17 (58.8%) 18/31 (58.1%) DVI only 10/14 (71.4%) 13/21 (61.9%) 39/49 (79.6%) 62/84 (73.8%) SVI + DVI 19/29 (65.5%) 25/26 (96.1%) 44/47 (93.6%) 88/102 (86,8%) Total 31/48 (64.5%) 44/56 (78.6%) 93/113 (82.3%) 168/217 (77.4%) Table VI. Incidence of skin changes in relation to location and extent of reflux (n = 91 limbs) ~t~.t of r~ux System Proximal Distal Proximal + Distal Total svi only o/5 (0.0%) 3/9 (33.3%) DVI only 3/14 (21.4%) 7/21 (33.3%) SVI + DVI 5/29 (17.2%) 15/26 (57.6%) Total 8/48 (16.7%) 25/56 (44.6% 6/17 (35.3%) 13/49 (26.5%) 39/47 (82.9%) 58/113 (51.3%) 9/31 (29.0%) 23/84 (27.4%) 59/102 (57.8%) 91/217 (41.9%) Table VII. Incidence of ulceration in relation to location and extent of reflux Extent of reflux System Proximal Distal Proximal + Dirtal Total SVI only 0/5 (0.0%) 2/9 (22.2%) 10/17 (58.8%) 12/31 (38.7%) DVI only 0/14 (0.0%) 1/21 (4.8%) 2/49 (4.1%) 3/84 (3.6%) SVI + DVI 1/29 (3.4%) 4/26 (15.4%) 18/47 (38.3%) 23/102 (22.5%) Total 1/48 (2.1%) 7/56 (12.5%) 30/113 (26.5%) 38/217 (17.5%) because the vast majority of our patients were known to be symptom-free before their episode of DVT. Half of our patients with symptoms of postthrombotic limb, had combined SVI and DVI (47%). Incompetence of the deep venous system with or without superficial venous incompetence was present in 85.7% of our study population. Most studies have shown that patients in whom reflux develops after DVT have incompetence of the deep venous system, and this is consistent with our results) s,19,2 In this study, the superficial system was also examined thoroughly, and we were able to show that 14.3 % of our patients had reflux in the superficial venous system alone. Some of this may be attributed to the existence of reflux before their episode of DVT. Other studies have tended not to examine the superficial system. Most of our patients were admitted with swellhag, and a significant number were admitted with skin changes and varicose veins. Sixteen percent of the limbs examined had ulceration. This represents a typical sample of patients with postthrombotic symptoms.2,6,21, 22 A small number of patients were admitted with claudication (3.2%). Venous claudication generally arises from occlusion ha the proximal veins 23, and this

6 Volume 20, Number 1 Labropoulos et al. 25 was seen in all seven cases, with most (6 of 7) arising in the iliofemoral segment. In addition, all the patients with venous claudication demonstrated reflux, five of them with combined SVI and DVI. One patient had incompetence of the deep venous system alone, and two had "pure" superficial venous incompetence. In most of the limbs the veins had been completely recanalized and in some the recanalization was partial. The vast majority of the limbs with recanalization had reflux. However, the effect of the residual obstruction was not studied because its location and extent were not recorded. Also it was not possible to correlate the site of thrombosis with the sites of reflux because the original venogram was not available, and detailed information was not in the report in at least a quarter of the patients. In patients with ulceration SVI was present in 31.6% and deep reflux alone was present in 7.9% of the limbs. These data are in agreement with Hanrahan et al.,24 but other studies report a higher incidence of SVI in venous ulceration. 2s All patients with previous DVT who are diagnosed with varicose veins have superficial reflux (combined with deep reflux in 72.2%). None of these dilated veins therefore represent valved conduits, which act as effective collateral veins in the upright position. Whether they are effective in the supine position depends on the site and extent of the residual obstruction. However, our study did not include physiologic tests of obstruction. Correction of valvular reflux by surgical reconstruction of deep veins has been investigated as a corrective therapy for postthrombotic limb symptoms, particularly for ulceration. Most deep venous surgery has been directed at the femoral valves. However, when ulceration is associated with prior DVT, isolated proximal reflux occurs in 2.6% of patients. Therefore a distal antireflux procedure, such as perforator ligation or popliteal vein reconstruction may represent a necessary adjunct to restore normal hemodynamics to the fimb and merits further investigation. It has previously been suspected that the successes of proximal venous reconstruction in alleviating ulceration were due, at least in part, to such adjunctive procedures. 26 Our data indirectly support this idea. REFERENCES 1. Browse NL. The aetiology of venous ulceration. World J Surg 1986;10: Strandness DE Jr, Langlois YE, Cramer MM, et al. Long term sequelae of acute venous thrombosis. JAMA 1983;250: ShuU KC, Nicolaides AN, Fernandes e Fernandes J, et al. Significance of pophteal reflux in relation to ambulatory 4. venous pressure and ulceration. Arch Surg 1979;114: Pearce WH, Ricco JB, Queral LA, Flinn WR, Yao JST. Hemodynamic assessment of venous problems. Surgery 1983;93: McEuroe CS, O'Donnell TF, Mackay WC. Correlation of clinical findings with venous hemodynamics in 386 patients with chronic venous insufficiency. Am J Surg 1988; 156: Lindhagen A, Bergquivst D, Hallbook T. Deep-vein thrombosis after postoperative thrombosis diagnosed with 125 I-labeled fibrinogen uptake test. Br J Surg 1984;7: Johnson ND, Queral LA, Flinn WR, Yao JST, Bergan JJ. Late objective assessment of venous valve surgery. Arch Surg 1981;116: O'Donnell TF, Mackay WC, Shepard AD, Callow A. Clinical, phlebographic and haemodynamic assessment of popliteal vein transplant in postphlebitic syndrome. Presented at the Second International Vascular Symposium; 1986, London, United Kingdom. 9. Rollins DL, Semrow CM, Freidell ML, Buchbinder D. Use of ultrasonic venography in the evaluation of venous valve function. Am J Surg 1987;154: Szendro G, Nicolaides AN, Zukowski, et al. Duplex scanning in the assessment of deep venous incompetence. J VASC SUaG 1986;4: Porter JM, Rutherford R_B, Clagett GP, et al. Reporting standards in venous disease. J Vase SURG 1988;8: Van Bemmelen PS, Bedford G, Beach K, Strandness DE. Quantitative segmental evaluation of venous valvular reflux with duplex ultrasound scanning. J Vasc SURG 1989;10: Lindner DJ, Edwards JM, Phinney ES, Taylor LM, Porter JM. Long-term hemodynamic and clinical sequelae of lower extremity deep venous thrombosis. J VASC SURG 1986;4: Browse NL, Clemenson G, Lea Thomas M. Is the postphlebitic leg always postphlebitic? Relation between the phlebographic appearance of deep-vein thrombosis and late sequelae. Br Med J 1980;281: Markel A, Manzo RA, Bergelin RO, Strandness DE Jr. Valvular reflux after deep vein thrombosis: incidence and time of occurrence. J VAsc SURG 1992;15: Neglen P, Raju S. Should duplex Doppler scanning replace descending phlebography as the "gold standard" in evaluation of venous reflux? [Abstract] J VASC SURG 1992;15: Mattos MA, Londrey GL, Leutz DW, et al. Color-flow duplex scanning for the surveillance and diagnosis of acute deep venous thrombosis. J VASC SURG 1992;15: Raju S. Venous insufficiency of the lower limb and stasis ulceration. Ann Surg 1983;197: Lindhagen A, Bergquivst D, Hallbook T, Efsing HO. Venous function five to eight years after clinically suspected deep venous thrombosis. Acta Med Scand 1985;217: Heldal M, Seem E, Sandset PM, Abildgaard U. Deep vein thrombosis: a 7-year follow-up study. J Int Med 1993;234: 71-5.

7 26 Labropoulos et a/. July KiUewich LA, Martin R, Cramer M, Beach K, Strandness DE Jr. An objective assessment of the physiological changes in the postthrombotic syndrome. Arch Surg 1985;20: Raju S, Frederick R. Venous obstruction in analysis of 137 cases with haemodynamic venographic and clinical correlation. J VASC SURG 1991;14: Strandness DE Jr. Deep venous thrombosis and the post thrombotic syndrome. In: Strandness DE Jr, ed. Duplex scanning in vascular disorders. New York: Raven Press 1990: Hanrahan L, Araki C, Rodriguez A, Kechejian G, Larnorte W, Menzoian J. Distribution of valvular incompetence in patients with venom stasis ulceration. J VASC SURG 1991;13: Shami SK, Sarin S, Cheade TR, Scurr JH, Coleridge Smith PD. Venom ulcers and the superficial system. J VASC SURG 1993; 17: Gooley NA, Sumner MD. Relationship of venous reflux to the site of venous valvular incompetence: implications for venous reconstructive surgery. J VASC SURG 1988;7:50-9. Submitted Sept. 22, 1993; accepted Dec. 13, 1993.