Distribution and quantification of venous reflux in lower extremity chronic venous stasis disease with duplex scanning

Transcription

1 Distribution and quantification of venous reflux in lower extremity chronic venous stasis disease with duplex scanning Michael S. Weingarten, MD, FACS, Charles C. Branas, MPH, RVT, Michael Czeredarczuk, BA, RVT, John David Schmidt, BS, RDMS, and Charles C. Wolferth, Jr., MD, FACS, Philadelphia, Pa. Purpose: The purpose of this study was to use color-flow duplex scanning to identify the anatomic distribution of venous reflux and to quantify venom reflux times in patients with various stages of chronic venom insufficiency (CVI). Methods: Color-flow-assisted duplex scanning was used to identify the anatomic distribution of venous reflux and to quantify reflux times in the deep and superficial venom systems of patients with symptomatic (CVI). Two hundred two patients with class I to III CVI were examined. Results: Only 11% (22 patients) had a documented history of phlebothrombosis. Of the 403 limbs evaluated, 192 had venom ulcers whereas 211 were classified as having class I or II CVI. Nonocclmive venom obstruction was found in only 16 limbs (4%). Venous ulceration was significantly associated with reflux in multiple venom segments as opposed to reflux in isolated venom segments (p < 0.001). Total limb reflux time (Rt) was determined by summing the reflux times of all the venom segments in a limb. The mean R t of patients with venom ulcerations was significantly longer than the mean Rtof limbs with class I and II CVI (p < 0.01). A total limb reflux time of greater than 9.66 seconds was predictive of ulceration. Total limb deep segment reflux time and total limb superficial segment reflux time were also determined by summing the reflux times of the appropriate segments in the limb. The mean deep segment reflux time was prolonged in limbs with venous ulcers when compared with limbs with class I and II CVI disease. The mean superficial segment reflux time of limbs with class I and II CVI and limbs with venom ulcers could not be used to distinguish between the two groups. In assessing the contribution of segments of the deep system to ulceration, reflux times of different segments were compared with wound duration and area. Reflux in the common femoral vein was significantly associated with wound area and duration (p < 0.05) whereas reflux time in the distal posterior tibial vein was associated with wound duration (p < 0.05). (J VASC SURG 1993;18:753-9.) Chronic venous hypertension is believed to lead to progressive venous stasis changes in the lower extremity and, in some patients, to venous stasis ulceration. Factors associated with venous hypertension include primary or secondary valvular insuffi- From the Division of Vascular Surgery, Department of Surgery, The Graduate Hospital, University of Pennsylvania, Philadelphia. Presented at the Fifth Annual Meeting of the American Venous Forum, Orlando, Fla., Feb , Reprint requests: Michael S. Weingarten, M.D., F.A.C.S., The Graduate Hospital, 1800 Lombard Street, Suite 1101, Philadelphia, PA Copyright 1993 by The Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter /93/$ /6/49721 ciency, venous outflow obstruction, or combinations of these entities. Quantitative assessment of venous hypertension and reflux, for the most part, has been based on hemodynamic measurements, involving the entire leg or the limb below the knee.x A distinction can be made between the contribution of the deep and the superficial venous systems to the disease process in many of these examinations. The role of individual venous segments and the degree to which they influence skin changes leading to ulceration cannot be determined by these hemodynamic measurements, however. With the ability to repair, transpose, or transplant individual valves in the venous system, this information is vital both in selecting patients for surgery and in following the long-term results of surgical intervention. 2 Duplex 753

2 754 Weingarten et al. November 1993 scanning, augmented by color-flow Doppler scanning, can provide a combination of anatomic information and segmental hemodynamic data not obtainable by other means) Duplex ultrasonography, augmented by color-flow Doppler scanning, provides visualization of refluxing, partially obstructed, or completely obstructed venous segments. Quantification of reflux times can then be obtained with spectral analysis. + The purpose of this study was to use color-flow duplex scanning to identify the anatomic distribution of reflux and to quantify reflux times in patients diagnosed with various stages of symptomatic chronic venous insufficiency (CVI), including long-standing venous stasis ulceration. MATERIAL AND METHODS The study sample consisted of 202 patients diagnosed with venous stasis ulceration of at least one lower extremity or symptomatic CVI. Patients with evidence of significant arterial obstruction were excluded. Patients without palpable pedal pulses were included if ankle-brachial indexes greater than 0.75 or toe-brachial indexes greater than 0.6 were obtained. There were 106 men and 96 women in the study group. The mean age of the group was 60.2 years with a range of 27 to 88 years. Twenty-two (11%) patients had a history of phlebitis in either the deep or superficial system. Fourteen of the patients had undergone previous greater saphenous vein stripping and ligation in the ulcerated leg. Two patients had had the saphenous vein in the involved leg harvested previously for coronary artery bypass grafting. One patient in the study had undergone a May-Husni procedure in the ulcerated limb several years before the evaluation and was admitted with recurrent ulceration. The mean duration of the venous ulcer before evaluation was 41.2 months with a range of 1 to 828 months. Mean ulcer area was mm 2 with a range of 3 to 21,441 mm 2. Color-flow duplex scanning was performed with either the Acuson 128 or 128XP duplex scanner (Mountain View, Calif.). Scanning of the deep system and the superficial system was first performed with the patient in the supine position and by use of standard techniques, s'6 Patients then underwent scanning while standing upright and without bearing weight on the leg being studied. Cuffs were placed around the leg distal to the venous segment being studied and were rapidly inflated and then deflated with an automatic cuff deflator (Parks, Aloha, Ore.). Keflux was first observed in the longitudinal plane with color-flow scanning, and then a pulsed-wave Doppler spectral analysis was performed. The initial use of color-flow scarming allowed the technologist to screen for reflux before performing spectral analysis. This shortened the time of the examination. A pilot study performed before this study indicated that all patients with reflux by spectral analysis had reflux demonstrable with color-flow scanning. Postprocessing spectral analysis was used to measure the duration of retrograde flow from the onset of distal cuff deflation to the cessation of flow. In all patients, the common femoral vein (CFV), popliteal vein, and mid posterior tibial veins (PTV) of the deep system were examined. Examination of the superficial venous system consisted of interrogating the greater saphenous vein (GSV) just distal to the fossa ovalis and the lesser saphenous vein (LSV) just beyond its takeoff from the popliteal vein. Before beginning the study, eight volunteers with normal lower extremities were examined to determine baseline reflux times in normal venous systems for the laboratory. The three technologists involved in the study followed strict protocols in performing the measurements to minimize interrater error. The results from this pilot study and the techniques used were described by these authors. 7 Information regarding wound size and duration was obtained during the initial evaluation of the the patient, ha patients with more than one extremity wound, the largest wound was used in all calculations. RESULTS Of the 403 limbs examined in this study, 192 were diagnosed with venous stasis ulceration and were therefore classified as having class 3 CVI. Two hundred eleven limbs had clinical findings consistent with mild or moderate class 1 or 2 CVI. s All patients diagnosed with unilateral venous ulceration had evidence of class 1 or 2 CVI in the contralateral limb. Of the 192 ulcerated limbs, 148 (77%) had detectable reflux. Of the group of ulcerated limbs with detectable reflux, 81 (55%) had reflux involving both the portions of the superficial and deep system evaluated by the study. Only 37 (25%) had reflux in just the deep system segments. Thirty-seven ulcerated limbs (25%) had reflux in isolated venous segments. In 23, the isolated segments involved the deep system segments studied (seven CFV, 16 popliteal vein). In the remaining 14 limbs, isolated segments were found in the superficial venous segments that were studied (4 LSV, 10 GSV) (Fig. 1). Of the 211 limbs with class 1 and 2 CVI examined, 90 (43%) had demonstrable reflux. Seventy limbs in this group (78%) had reflux in only one venous segment (6 CFV, 21 GSV, 26 popliteal, 4 PTV, 13 LSV) (Fig. 2). With chi-squared contingency analysis, venous ulceration was also more often associated with reflux in multiple venous segments than with reflux in isolated venous segments (p < 0.001) (Fig. 3).

3 Volume 18, Number 5 Weingarten et al % [] POP 27% PTV [] GSV LSV Fig. 1. Anatomic distribution of isolated venous reflux among limbs with class 3 CVI. 9% [] CFV I [] POP I 30% 37% PTV I [] GSV I LSV I 6% Fig. 2. Anatomic distribution of isolated venous reflux among class 1 and 2 CVI. Nonocclusive chronic venous thrombosis was found in 16 limbs. Eleven of the 16 limbs were diagnosed with venous ulceration whereas five had signs and symptoms of moderate class 2 CVI. All patients diagnosed with nonocclusive venous thrombosis had evidence of reflux by duplex scanning. No patient in this series was diagnosed with totally occlusive venous thrombosis. In the patient admitted after having undergone May-Husni repair, the superficial femoral vein had evidence of nonocclusive chronic thrombus and reflux of the saphenous vein bypass was noted. Reflux times in limbs with isolated venous segment reflux in the ulcerated and nonulcerated group were compared. No significant association was found between reflux time and the presence or absence of ulceration. Within each limb the reflux time (Rt) for all segments was obtained by summing the R t of each individual segment examined in this study, which resulted in the total limb R v The total limb R t was then averaged for all ulcerated and nonulcerated limbs, giving a mean R t. The mean R t in the patients with venous ulceration was 9.66 _ seconds. The mean Rt in patients without ulceration was seconds. The mean total limb R t in patients diagnosed with ulceration was significantly longer than the group without ulceration, as determined by the Student's t test (p < 0.01) (Fig. 4). These data sets, although highly dispersed, did maintain an acceptable degree of power (power = 0.90). Total limb deep segment reflux times (R,d), total limb superficial segment reflux times (R~), and the respective means were then calculated

4 756 Weingarten et al. November 1993 c~ 0.6 Percentage of limbs demonstrating multiple segment reflux nonulcerated ulcerated Fig. 3. Percentage of limbs demonstrating multiple segment reflux in the nonulcerated and ulcerated categories. 1 p. Total limb reflux time (Rt) nonulcerated ulcerated Fig. 4. Total limb reflux time (in sec.) in ulcerated and nonulcerated limbs. for each limb in a similar manner, again summing individual R t in the various segments of the deep and superficial system examined in this study. In the patients with venous ulceration, the mean Rta time was seconds whereas the mean Rtd in the nonulcerated population was _ 3.23 seconds. Mean total limb deep segment reflux duration was significantly associated with venous ulceration by the Student's t test (p < 0.05). A corresponding power analysis was also completed for this comparison (power = 0.49). The mean Rt~ in the ulcerated limbs was seconds. The mean R~ in the group of nonulcerated limbs was 6.32 _ seconds. Mean total limb superficial segment reflux duration, on the other hand, was not associated with venous ulceration (as determined by the Student t test) and could not be used to distinguish the two patient groups (Fig. 5). Reflux time in the CFV was significantly associated with wound duration and area as determined by a standard linear regression model (p < 0.05). Reflux time in the PTV was significantly associated with duration of the wound through a similar analysis (p < 0.05). DISCUSSION Quantification of the severity of venous disease has been based on hemodynamic measurements or invasive imaging techniques. Ambulatory venous pressure (AVP) measurements, first performed in the

5 Volume 18, Number 5 Weingarten et al <%1 ~D ~D 6- Total limb reflux time (Rtd and Rte) 4- nonutcerated BB ulcerated 2- deep superficial Fig. 5. Total limb reflux time in deep and superficial systems (in sec.) in ulcerated and nonulcerated limbs. 1940s, have been considered the gold standard in the development of noninvasive techniques assessing the severity of CVI. 9 AVP measurements, although indicative of the severity of the venous abnormality in the limb, do not predict the cause of the venous disorder or the magnitude of the clinical problem. 1 '~2 Raju and Fredericks as have combined AVP measurements with foot venous pressure measurements after the Valsalva maneuver to better quantify the degree of reflux that the leg is subjected to. The anatomic distribution of reflux can be determined by descending venography 14 but limitations in complete visualization of the venous system are found) s Photoplethysmography provides a noninvasive technique of assessing venous reflux in the limb 16 but limitations in this technique have been reported) 7,18 Air plethysmography provides hemodynamic data based on whole limb arterial inflow and venous outflow. 19,20 Quantitative data obtained with air plethysmography correlates with the severity of CVI but cannot provide information on the contribution of individual venous segments to the disease process. 21,z2 Duplex scanning provides both anatomic and functional data regarding the contribution of the deep and superficial venous system and individual segments within each system to CVI. The technique is augmented by color-flow Doppler scanning. In a study by Hanrahan et al.,2s 95 limbs diagnosed with venous ulceration were examined by duplex scanning. Valvular incompetence was defined as the presence of bidirectional flow in a venous segment elicited by a Valsalva maneuver or proximal or distal compression by the technician. Sixty-three (66.3%) of the ulcerated limbs had multiple incompetent segments in the deep, superficial, and perforator system. 2s Duplex scanning of the perforating veins has revealed significant differences in size between competent and incompetent perforators in ulcerated limbs. 24 Duplex evaluation of flow patterns in perforating veins of the calf have demonstrated bidirectional perforator flow in normal limbs and limbs with CVI. In patients with significant CVI, however, perforator blood flow in the relaxation phase after distal compression occurred more often than in normal limbs. 2s Spectral analysis of blood flow in a segment of vein with reflux allows quantification of the extent of reflux. In a study of patients with various stages of CVI, Vasdekis et al.26 used spectral analysis to determine the mean velocity at the peak of reflux in different venous segments. By then measuring the diameter of the vein segment being sampled, the crosssectional area of the veins could be determined. Reflux flow at peak reflux in ml/sec could then be calculated. Flow at peak reflux of greater than 10 ml/sec was associated with a high incidence of lipodermatosclerosis and ulceration. 26 Duration of reflux in various venous segments in normal limbs was studied by van Bemmelen et al.,27 and baseline reflux times were determined. Valve closure was elicited in this study most reliably by distal cuff deflation. In a previous report the authors stressed the importance of the patient's being in the upright position during reflux evaluation and the inadequacy of the Valsalva maneuver in causing distal valve closure. 2s Quantitative duplex scanning has recently been compared with descending phlebography, the gold standard in

6 758 Weingarten et al. November 1993 assessing the contribution of reflux to CVI. With two different anatomic scoring scales for reflux, results obtained by duplex scanning correlated both with the clinical severity of CVI and the hemodynamic measurements taken at the same time. Scores based on the results of descending phlebography failed to demonstrate this correlation. 29 The findings in this study that venous ulceration was significantly associated with reflux in multiple venous segments of the deep and superficial systems confirms the findings of others who used this technique. 23 However, only 14 of the 148 limbs diagnosed with class 3 CVI had reflux in either the GSV or LSV, a finding that contrasts with reports from other examiners. 3 The striking incidence of class 1 and 2 CVI in the contralateral limb in this group of patients is not surprising given the high incidence of occult venous abnormalities found by others in supposedly normal contralateral limbs. 3~ The absence of totally occluded venous segments and the small number (8%) of nonocclusive thrombotic segments in the ulcerated limbs in this group of patients was unexpected, given the prolonged duration of the ulceration in many of the patients. Results in other series vary from 5% to 28%. 32'33 The finding that single segment venous reflux was more often associated with class 1 and 2 CVI, whereas the fact that the minority of patients with venous ulcer had single segment reflux suggests that progression of valvular reflux parallels the progression of CVI, as suggested by others. 21 In the patients with ulceration and single segment reflux, 43% had reflux in the popliteal vein and 10.8% had reflux in the lesser saphenous system. These results emphasize the importance of disease in the veins below the knee in the generation of venous ulcers. Of the 192 limbs with venous ulceration studied, 44 (23%) had no demonstrable reflux by duplex scanning. Similar problems with detection of reflux in patients with ulceration by duplex scanning was reported by Neglen and Raju. 29 Of the 211 limbs diagnosed with class 1 and 2 CVI, 121 limbs (57%) had no demonstrable reflux by duplex scanning. The diagnosis of venous ulceration in these patients was made on the basis of the physical findings of venous stasis in the limb and the absence of other conditions associated with ulcerations of the lower extremity. The absence of detectable reflux in these groups of patients can be explained by the design of this study. Although the study did include the posterior tibial veins, reflux studies in the peroneal and anterior tibial veins were not performed. Likewise, profunda femoris reflux was not measured because of difficulties in consistently imaging this vessel. Proximal obstruction or reflux above the level of the femoral vein could also not be consistently ruled out because of difficulties visualizing the iliac vessels. In this study, although perforating veins were examined, measuring reflux times was found to yield inconsistent data and was not reported. Given the findings of others with regard to bidirectional flow, this could be expected) 2"2s In this study, increasing duration of reflux as determined by the total limb reflux time was associated with ulceration. In a patient with a total limb reflux time of 9.66 seconds or more, venous ulceration could be expected. The large standard deviations of the data were similar to the large standard deviations reported by van Bemmelen et al. 27 in the original description of the technique used in this study. In fact,the standard deviations in this study were smaller. This dispersion of the data, common to both studies, may reflect potential errors inherent in quantifying venous reflux by duplex scanning. These potential sampling problems include variations in cuff deflation time, both intratechnologist and intertechnologist differences, and sampling errors during spectral analysis. A power analysis of these data confirms that, on the basis of sample size, the differences between total limb reflux time between the ulcerated and nonulcerated limbs, as determined in this study, was significant. The contribution of the deep and superficial venous segments to ulceration was assessed by determining the total limb deep segment and total limb superficial segment reflux time for each limb. Prolonged total limb deep segment reflux time was associated with venous ulceration whereas total limb superficial segment reflux time was virtually the same in the ulcerated and nonulcerated limbs but the large standard deviations and the smaller sample size do not allow us to draw any firm conclusions. Finally, the study suggests that, in assessing the role of individual segments in the deep system in the production of ulceration, reflux duration in the CFV, followed by reflux duration in the PTV, are predictive of the severity of ulceration, as measured by ulcer area and duration. More proximal reflux would be expected to raise ambulatory venous pressure throughout the limb, resulting in larger venous ulcers less likely to heal. The anatomic information obtained by duplex scanning, when combined with quantitative reflux data, provides a useful tool for evaluating patients with CVI. Multiple segment reflux in the deep and superficial venous systems correlated with the severity of CVI. Total limb reflux time was predictive of venous ulceration in this patient population. In separating the deep and superficial components of total limb reflux, it appeared that reflux in the deep system played a more significant role in venous

7 Volume 18, Nurnbcr 5 Weingarten et al. 759 ulceration than the superficial system but a larger sample size is necessary to confirm this. Specifically, reflux time in the femoral veins and PTV did correlate with the severity of ulceration. Duplex scanning with quantitative reflux determination has proven to be a useful tool in the assessment of the chronically ulcerated limb. In the patient with multiple factors leading to limb ulceration, the presence and severity of venous disease can be determined noninvasively. This study focused on specific segments of the deep and superficial venous system. Future investigation should focus on additional segments and correlation with hemodynamic testing. REFERENCES 1. Belcaro G, Christopoulos D, Nicolaides AN. Lower extremity venous hemodynamics. Ann Vasc Surg 1991;5: Kismer RL. Late results of venous valve repair. In: Yao JST, Pearce WH, eds. Long-term results in vascular surgery. Norwalk, Conn: Appelton & Lange, 1993: Nicolaides AN, Christopoulous DC. Methods ofquantitation of chronic venous insufficiency. In: Bergen JJ, Yao JST, eds. Venous disorders. Philadelphia: WB Saunders, 1991: 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: Hobson RW, Mintz BL, Jamil Z, Breitbart GB. Current status of duplex ultrasonography in the diagnosis of acute deep venous thrombosis. In: Bergen JJ, Yao JST, eds. Venous disorders. Philadelphia: WB Saunders, 1991: Sumner DS, Londrey GL, Spadone DP, Hodgson KJ, Leutz DW, Stauffer ES. Study of deep venous thrombosis in high-risk patients using color flow Doppler. In: Bergen JL Yao JST, eds. Venous disorders. Philadelphia: WB Saunders, 1991: Czeredarczuk M, Branas CC, Weingarten MS. Duplex imaging and distal cuff deflation to measure venous reflux time. J Vasc Tech 1991;15: Porter JM, Rutherford RB, Clagett GP, et al. Reporting standards in venous disease. J VASC SURG 1988;8: Pollack AA, Wood EH. Venous pressure in the saphenous vein in man during exercise and changes in posture. J Appl Physiol 1949;1: Coleridge Smith PD, Scurr JH. Current views on the pathogenesis of venous ulceration. In: Bergen JJ, Yao JST, eds. Venous disorders. Philadelphia: WB Saunders, 1991: Schanzer H, Pierce EC. Pathophysiologic evaluation of chronic venous stasis with ambulatory venous pressure studies. Angiology 1982;33: Raju S, Fredricks R. Valve reconstruction procedures for nonobstructive venous insufficiency: rationale, techniques, and results in 107 procedures with 2-8 year follow-up. J VASC SURG 1988;7: Raju S, Fredricks R. Hemodynamic basis of stasis ulceration-a hypothesis. J VASC SUV, G 1991;13: Herman RJ, Neiman HL, Yao JST, Egan TJ, Bergan J], Malave SR. Descending venography: a method of evaluating lower extremity venous valvular function. Radiology 1980; 137: Moore Dj ~, Himmel PD, Sumner DS. Distribution of venous valvular incompetence in patients with postphlebitic syndrome. J VASC SURG 1986;3: Satin S, Shields DA, Scurr JH, Coleridge Smith PD. Photoplethsmyography: a valuable noninvasive tool in the assessment of venous dysfunction? J VASC SURG 1992;16: Raju S, Fredricks R. Evaluation of methods for detecting venous reflux. Arch Surg 1990;125: van Bemmelen PS, Bergan JJ. Photoplethysmography and LRR in quantitative measurement of venous incompetence. Austin: RG Landes Co, 1992: Christopoulos D, Nicolaides AN, Szendro G, Irvine AT, Bull ML, Eastcott HHG. Air plethysmography and the effect of elastic compression on venous hemodynamics of the leg. J VASC SURG 1987;5: Welkie JF, Kerr RP, Katz ML, Comerota AJ. Can noninvasive venous volume determinations accurately predict venous pressure? J Vasc Tech 1991;15: Welkie JF, Comerota AJ, Katz ML, Aldridge SC, Kerr RP, White JV. Hemodynamic deterioration in chronic venous disease. J VAsc SURG 1992;16: van Bemmelen PS, Bergen JJ. Air plethysmography. In: van Bemmelen PS, Bergen JJ, eds. Quantitative measurement of venous incompetence. Austin: RG Landes Co, 1992: Hanrahan LM, Araki CT, Rodriguez AA, Kechejian GJ, LaMorte WW, Menzoian JO. Distribution of valvular incompetence in patients with venous ulceration. ]" Vase SURG 1991;13: Hanrahan LM, Araki CT, Fisher JB, Rodriguez AA, Walker TG, Woodson J, LaMorte WW, Menzoian JO. Evaluation of the perforating veins of the lower extremity using high resolution duplex scanning. J Cardiovasc Surg 1991;32: Satin S, Scurr JH, Coleridge Smith PD. Medial calf perforators in venous disease: the significance of outward flow. J VASC SURG 1992;16: Vasdekis SN, Heather Clarke G, Nicolaides AN. Quantification of venous reflux by means of duplex scanning. J VASC SURG 1989;10: van Bemmelen PS, Bedford G, Beach K, Strandness DE. Quantitative segmental evaluation of venous valvular reflux with duplex ultrasound scanning. J VASC SUV, G 1989; 10: van Bemmelen PS, Beach K, Bedford G, Strandness DE. The mechanism of venous valve closure. Arch Surg 1990;125: Neglen P, Raju S. A comparison between descending phlebography and duplex Doppler investigation in the evaluation of reflux in chronic venous insufficiency: a challenge to phlebography as the "gold standard." J VASC SURG 1992;16: Shami SK, Satin F, Cheatle TR, Scurr JH, Coleridge Smith PD. Venous ulcers and the superficial system. J VASC SURG 1993; 17: Stacey MC, Burnard KG, Pattison M, Lea Thomas M, Layer GT. Changes in the apparently normal limb in unilateral venous ulceration. Br J Surg 1987;74: McEnroe CS, O'Donnell TF, Mackey WC. Correlation of clinical findings with venous hemodynamics in 386 patients with chronic venous insufficiency. Am J Surg 1988; 156: Raju S, Fredricks R. Venous obstruction: an analysis of one hundred thirty-seven cases with hemodynamic, venographic, and clinical correlations. J VASC SURG 1991;14: Submitted April 7, 1993; accepted June 30, 1993.