Reproducibility of Duplex Ultrasonography and ... - Wiley Online Library

Article

Reproducibility of Duplex Ultrasonography and Air Plethysmography Used for the Evaluation of Chronic Venous Insufficiency Akram Mahmoud Asbeutah, MS, Andrea Zdena Riha, MBBS, James Donald Cameron, MD, Barry Patrick McGrath, MD Objective. The purpose of this study was to determine the reproducibility of measurements on duplex ultrasonography (DU) and air plethysmography (APG) in subjects with post-thrombotic syndrome. Methods. Duplex ultrasonography and APG were used to measure indices of lower limb venous reflux in 15 limbs with a history of deep vein thrombosis and evidence of venous insufficiency as diagnosed by ultrasonography. Three limbs were in class 0; 4 were in classes 1 to 3; and 8 were in classes 4 to 6, according to clinical, etiologic, anatomic, and pathophysiologic clinical classification. Duplex ultrasonography was performed 3 times on the same day, and venous diameter, area, peak reflux velocity, reflux flow volume, and reflux duration measurements were obtained. Air plethysmography was performed on 2 days, 7 to 10 days apart, with 1 measurement on the first day and 2 measurements on the second day. Values obtained from APG included outflow fraction, venous filling index, ejection fraction, and residual volume fraction. The measurements were performed by a vascular technologist blinded to the previous test results. One-way analysis of variance, the Student paired t test, and BlandAltman plots were used to examine the statistical differences of the DU and APG parameters for all measurements. Results. The mean coefficient of variation for within-subject measurements of all DU and APG parameters measured was less than 10%. Bland-Altman plots showed that there were no apparent trends with increasing values over a wide range for any of the DU parameters, nor were there any for the APG parameters. Conclusions. Under ideal conditions, when measured by a highly trained technologist, both DU and APG showed satisfactory reproducibility. Key words: air plethysmography; chronic venous insufficiency; duplex ultrasound; reproducibility. Abbreviations ANOVA, analysis of variance; APG, air plethysmography; CEAP, clinical, etiologic, anatomic, and pathophysiologic; CV, coefficient of variation; CVI, chronic venous insufficiency; DU, duplex ultrasonography; DVT, deep vein thrombosis; EF%, ejection fraction; OF%, outflow fraction; RC, repeatability coefficient; RV, residual volume; RVF%, residual volume fraction; VFI, venous filling index

Received August 9, 2004, from the Department of Vascular Sciences and Medicine, Monash University and Southern Health, Melbourne, Victoria, Australia (A.M.A., J.D.C., B.P.M.); and Department of Cardiology, The Wesley Vascular Centre, Brisbane, Queensland, Australia (A.Z.R.). Revision requested September 20, 2004. Revised manuscript accepted for publication November 28, 2004. We thank Kais Hamza, MD, for his kind cooperation in the statistical analysis of this study. This work was supported by the Department of Vascular Sciences and Medicine, Monash University. Address correspondence and reprint requests to Barry McGrath, MD, Department of Vascular Sciences and Medicine, Monash University and Southern Health, Dandenong Hospital, Dandenong, Melbourne, Victoria 3175, Australia. E-mail: [email protected]

D

uplex ultrasonography (DU) and air plethysmography (APG) are 2 noninvasive techniques that are widely used for the evaluation of subjects with chronic venous insufficiency (CVI). Duplex ultrasonography can show reflux in individual veins, whereas plethysmographic methods measure reflux in an entire limb. Plethysmographic measurements also provide an evaluation of the severity of reflux. Previous studies have reported the diagnostic value of both DU and APG in determining the presence of CVI.1 One study has shown that APG is better than photoplethysmography and suggests that the combination of APG and DU provides the best means of assessing venous function.2 Duplex ultrasonography has been used since the early 1980s to diagnose deep vein thrombosis (DVT).3–6 Its use has now been extended to detect obstruction and reflux and to determine their anatomic extent in CVI.7–9 Color flow duplex imaging has decreased examination time and improved accuracy. Among the various functional

© 2005 by the American Institute of Ultrasound in Medicine • J Ultrasound Med 2005; 24:475–482 • 0278-4297/04/$3.50

Duplex Ultrasonography and Air Plethysmography in Chronic Venous Insufficiency

and imaging tests that are available, DU is regarded as the method of choice because it is accurate and noninvasive.10–12 The severity of reflux with duplex scanning as determined by DU has shown only a weak correlation with plethysmographic quantification. This may be because complex patterns of reflux are frequent in different vein segments.13,14 It would be complicated and laborious to arrive at an estimation of total reflux by adding reflux in superficial, perforator, muscular, and deep veins. Air plethysmography was first used in the early 1960s to study relative volume changes in the lower limb in response to postural alterations and muscular exercise.15 In recent years, it has been widely used to assess patients with CVI and to assess calf muscle pump function and venous reflux.16–18 Also, the evaluations have been extended to assess the effect of treatments on venous reflux and calf muscle pump function16–19; however, Yang et al20 reported that there were significant variations within subjects on retesting and on different days, suggesting that APG is unlikely to be able to show small changes in venous reflux and calf pump function. The reproducibility and accuracy of DU have been documented for normal limbs,10–12 but little information is available on reproducibility measurements for abnormal limbs. Similarly, APG reproducibility has been reported previously in healthy subjects, but few data have been reported in subjects with chronic venous disease.16,17,20 Without reproducibility data for DU and APG in abnormal limbs, it is difficult to interpret many of the published studies. The aim of this study was to test the reproducibility of both DU and APG techniques in a group of subjects with deep venous insufficiency. Reproducibility of APG was also examined in a group of healthy subjects of similar age and gender.

Committee of Southern Health approved the study, and informed consent was obtained from all subjects. Clinical Assessment The same experienced physician (A.Z.R.) classified the venous disease before the DU and APG measurements according to the clinical, etiologic, anatomic, and pathologic (CEAP) clinical classification of the International Consensus Committee of the American Venous Forum.21 The CEAP clinical classification of lower limb venous disease states that subjects with no visible or palpable signs of venous disease are identified as class 0; subjects with telangiectases or reticular veins are class 1; subjects with varicose veins are class 2; subjects with edema are class 3; subjects with skin changes ascribed to venous disease such as pigmentation or lipodermatosclerosis are class 4; subjects with skin changes and healed ulcer are class 5; and subjects with skin changes and active ulcers are class 6. For simplicity, the subjects were divided into 3 classes: class 0 (no venous disease), classes 1 to 3 (mild to moderate venous disease), and classes 4 to 6 (severe venous disease). Three limbs were in class 0; 4 limbs were in classes 1 to 3; and 8 limbs were in classes 4 to 6.

Duplex Ultrasonography An HDI 5000 ultrasonography machine (Philips Medical Systems, Bothell, WA) coupled with a 7.4-MHz linear transducer was used for the examination. Subjects were examined in a temperature-controlled room (22°C). The examination included 40 deep vein segments: 32 proximal vein segments (4 common femoral veins, 13 superficial femoral veins, and 15 popliteal veins) and 8 distal vein segments (8 posterior tibial veins). Proximal venous segments involve popliteal veins and more proximal veins, whereas distal vein segments involve the

Materials and Methods Subjects The characteristics of the research population are summarized in Table 1. Fourteen subjects (15 limbs) with chronic venous disease participated in this study (8 male and 6 female), with an average age of 52.9 years (range, 23–76 years). All subjects had a previous diagnosis of DVT by DU, venography, or both for evidence of chronic venous disease. The Human Research and Ethics 476

Table 1. Subject Characteristics Characteristic

Abnormal Limbs

No. of limbs Age, y, average (range) Sex, male/female Clinical class 0 1–3 4–6

Normal Limbs

15 52.9 (23–76) 6/8

10 52.2 (21–73) 6/4

3 4 8

10 0 0

J Ultrasound Med 2005; 24:475–482

Asbeutah et al

tibial veins but without involvement of the popliteal or more proximal veins. The subjects were examined while in a standing position. They held onto the edge of a couch with their weight mainly on the leg that was not being examined; the leg that was being examined was kept in a relaxed position with the knee slightly flexed. The subjects rested the leg for 15 minutes before the start of the examination to avoid exercise hyperemia. Venous flow augmentation was accomplished by placing a 10-cm pneumatic cuff connected to a Hokanson E20 automatic inflator (Hokanson Inc, Issaquah, WA) immediately distal to the venous segment under examination. To ensure rapid inflation/deflation time, a large inflation/deflation tube was connected to the cuff. A pressure of 100 mm Hg was used. An automatic cuff inflator was used to provide inflation within 3 seconds and, most importantly, to provide deflation within 0.5 seconds. Deep venous reflux was specified as duration of reflux of more than 0.5 seconds.22 If reflux was detected, the vein diameter and area, peak velocity of reflux, and flow at peak velocity of reflux were measured. The testing procedure was similar to that described previously by Vasdekis et al10 and is not outlined here. The description of the duplex measurement is illustrated in Figure 1. The flow at peak velocity of reflux was measured using the following formula: reflux flow (mL/s) = peak velocity (cm/s) × area (cm2). Air Plethysmography After DU, venous hemodynamics were assessed with an APG-1000 air plethysmograph (ACI Medical Inc, San Marcos, CA). The APG apparatus consists of a tubular, polyvinyl chloride (5-L capacity) air chamber cuff that surrounds the whole leg from knee to ankle. This was inflated to 6 mm Hg, ensuring good contact between the cuff and the leg, and connected to a pressure transducer, amplifier, and recorder. The apparatus was then calibrated by introducing 100 mL of air into the air chamber via a graduated syringe. The supine venous volume (milliliters), 1 second outflow volume (milliliters), outflow fraction (OF%), functional venous volume (milliliters), venous filling index (VFI; milliliters per second), 90% venous filling time (seconds), ejection volume (milliliters), ejection fraction (EF%), residual volume (RV; milliliters), and residual volume fraction (RVF%) were calculated as described previously16,18,19,23 and are not outlined here. J Ultrasound Med 2005; 24:475–482

Figure 1. When reflux is detected by the reversal of flow, as it appears on the Doppler trace (A), then peak velocity at reflux and the time of reflux (B) are measured by placing a + on the trace; then the diameter is then measured by placing the marker (calipers) from inner to inner wall of the vein (C), and the software calculates the area, considering that the vein is a circular structure, while the patient is in a standing position.

A

B

C

477


Normal and abnormal values of APG indices have been summarized by Belcaro et al.24 An OF% of less than 40% is considered obstructed venous outflow, and more than 40% is considered normal. A VFI of more than 10 mL/s is considered severe venous reflux; 5 to 10 mL/s is considered moderate venous reflux; 2 to 5 mL/s is considered mild venous reflux; and less than 2 mL/s is considered normal. An EF% of less than 40% is considered severe calf pump dysfunction; 40% to 60% is considered calf pump dysfunction; and more than 60% is considered normal. An RVF% of more than 35% is considered abnormal, and less than 35% is considered normal.

results of repeated measures between and within days were analyzed. The Student paired t test was used to investigate differences between paired measurements for both between- and withinday data sets. Differences were considered significant with P < .05. Repeatability was also assessed as recommended by Bland-Altman analysis and presented graphically as differences versus mean plots.25 Repeatability coefficients (RCs) were calculated as RC = 2√(Σ Di2/n), where Di is the absolute difference between measurements, and n is the number of observations.

Test Sequence In the 14 subjects (15 limbs) with CVI, DU was repeated 3 times on the same day, with intervals of 5 minutes, for the measurement of venous diameter, area, velocity at peak reflux, flow at peak reflux, and the duration of reflux. This was followed by APG assessment. Air plethysmography measurements were performed on 2 days, 7 to 10 days apart; 1 test was performed on the first day, and 2 APG tests were performed on the second day, with an interval of 5 minutes. Repeatability of APG was also assessed in 10 healthy subjects with matching age and sex for comparison. The healthy subjects were not tested with DU because there are no venous reflux indices to be measured in normal veins. The measurements were performed by a highly trained vascular technologist (A.M.A.) who performed both DU and APG measurements. The data were stored on the hard disks of the DU and APG machines and subsequently entered into a Microsoft Access program (Microsoft Corporation, Redmond, WA) for analysis. The data from any former study were not examined before the next study.

One-way ANOVA showed no statistical significance between the mean measurements for any of the DU parameters on the 3 occasions. The mean ± SD and the average of within-subject CVs for repeated measurements for DU parameters in the 15 limbs (14 subjects, 32 proximal and 8 distal deep vein segments) are shown in Table 2. Mean CVs for all of the measured parameters did not exceed 10%. The Bland-Altman plots showed that there were no apparent trends with increasing values over a wide range for any of the indices. There were no differences in the repeatability measurements for either technique or for proximal versus distal venous segments. The mean values for all APG parameters were not significantly different for paired within- and between-day data sets. The mean ± SD and the average of the within-subject CVs for the different APG indices, determined using day-to-day and within-day tests for normal and abnormal limbs, are reported in Table 3. The Bland-Altman plots are presented graphically in Figure 2, A and B. The overall mean CVs for all APG parameters were less than 10%, and there were no appreciable trends to differences between measurements over a wide range.

Statistical Analysis Statistical analysis was performed with SPSS statistical software version 10 (SPSS Inc, Chicago, IL). For the DU parameters, 1-way analysis of variance (ANOVA) was used to examine for differences between mean values on the 3 occasions. For each parameter, the group mean and SD were calculated. For each parameter, in each subject, the coefficient of variation (CV) was calculated, and the average values are reported. The CV was calculated as (SD/mean) × 100. Proximal and distal venous segments examined by DU were analyzed separately. For the APG indices, 478

Results

Discussion Duplex ultrasonography and APG are two widely used noninvasive diagnostic tools for evaluating venous function. Duplex ultrasonography is currently considered the method of choice to assess venous disease because it is accurate and noninvasive, and there have been some studies on technique reproducibility10–12; however, only limited information is available regarding the reproducibility of APG methods.26,27 J Ultrasound Med 2005; 24:475–482

Asbeutah et al

Table 2. Mean ± SD and Mean of CVs of DU Variables Measured on 3 Occasions Duplex Parameter

Diameter, cm Area, cm2 Reflux peak velocity, cm/s Reflux flow volume, mL/s Reflux duration, s

Proximal Venous Segments Mean ± SD CV, %

0.9 0.7 39.2 27.3 3.3

± ± ± ± ±

0.2 0.4 23.2 22.2 1.5

Distal Venous Segments Mean ± SD CV, %

4 7 5 7 7

0.4 0.1 35.8 4.9 2.7

± ± ± ± ±

0.05 0.04 14.1 2.7 1.3

4 4 5 7 8

One-way ANOVA showed no statistical significance for all parameters over a wide range. CV represents mean of withinsubject values.

The results of the present study show that there is good repeatability of indices of venous reflux as measured by each of these techniques in subjects with chronic venous disease. Our study has shown good reproducibility of DU parameters in subjects with deep venous disease, and the CVs of all DU indices are less than 10% and show no appreciable trend for differences between measurements over a wide range. This study has also shown good repeatability for each of the 4 key APG indices, OF%, VFI, EF%, and RVF, extending into the severely abnormal range. Normal and abnormal values for different APG parameters have been reported.1,21,24,28,29 It has been shown that 95% of healthy limbs have a duration of reflux in deep veins or in both deep and superficial veins of less than 0.5 seconds.7,22,30 Araki et al31 found no difference between manual release and pneumatic tourniquet release; however, in our study, pneumatic release was used for standardization of the data. In our study, venous reflux was measured by pulsed Doppler imaging. Color Doppler imaging was limited in use and was used to confirm reflux in certain vein segments. This imaging shows the venous segment with flow toward the heart with distal compression and flow away from the heart after release of compression, indicating reflux. The venous reflux was detected mainly in the deep venous system as a result of a history of

DVT in our subjects. Only a few limbs were diagnosed with superficial or perforator venous reflux in our subjects, and that is why they were not included in this study. Reproducibility of DU in superficial venous reflux has been reported previously with the same technique.10 In our DU reproducibility study, the venous segments were divided into proximal and distal because we thought that there could be a difference in reproducibility due to the difference in vein caliber. Our study showed that the CVs of the venous indices were approximately the same for the proximal and distal venous segments. The study also showed that subjects with a severe form of the venous disease, according to CEAP clinical classification, had approximately similar CVs when compared with limbs with no venous disease or with mild or moderate venous disease. Three limbs were class 0 (no venous disease) when examined by CEAP clinical classification but showed venous reflux with DU. The DU described in our study gives only segmental information applied to the deep venous system, whereas APG provides quantitative information about venous reflux in the whole limb as well as information about the various components of the calf muscle pump.19,32 A recent study reported that APG is better than photoplethysmography and that the combination of APG and DU provides the best means of

Table 3. Mean ± SD and Mean CV for APG Variables for Between- and Within-Day Measurements for Normal and Abnormal Limbs APG Parameter

OF% VFI, mL/s EF% RVF%

Between-Day Measurements Mean ± SD CV, %

45.6 3.8 58.3 21.4

± ± ± ±

2.1 0.9 3.9 3.5

3 6 5 8

Within-Day Measurements Mean ± SD CV, %

45.5 3.8 58.2 21.4

± ± ± ±

2.1 0.9 3.9 3.5

2 6 4 8

The Student paired t test showed no statistical significance for all parameters over a wide range. CV represents mean of within-subject values.


479


Figure 2. Bland-Altman plots of differences between measurements versus mean values for all the APG indices between days (1–2) and within days (2–3) are shown. The OF% is shown in A and B; VFI is shown in C and D; EF% is shown in E and F; and RVF is shown in G and H. The values for RC are shown for each measurement. Filled circles indicate normal limbs; and open circles, abnormal limbs. Horizontal lines are drawn at the mean difference ± 2 SD.

A

B

C

D

E

F

G

H

480


Asbeutah et al

assessing venous function.2 Thus, to achieve global assessment of the lower limb venous system, examinations by both DU and APG should be performed. The combined effect of venous reflux, obstruction, and ejection capacity is evaluated by measuring the RV and RVF; the latter shows a linear relationship with measurements of ambulatory venous pressure. Yang et al20 reported variations for RV and RVF between 7.5% and 27% and concluded that the variations were such that APG was unlikely to be able to detect small changes in venous reflux and calf pump function. This variability could be due to interobserver variations and the nature of this part of the APG examination, which requires a lot of subject movement. In contrast, the results of this study show that a single experienced operator using standardized procedures can obtain very satisfactory repeatability of RV and RVF values. For other APG parameters (OF%, VFI, and EF%), when less subject movement is required, the variability of these indices is lower. It is acknowledged that this study did not examine interobserver variability, and it cannot be assumed that 2 different technologists could obtain the same repeatability. We urge all technologists who report results using these techniques to examine their own repeatability, and this study provides appropriate guidelines for this. In conclusion, under ideal conditions, when measured by a single highly trained technologist, both DU and APG showed satisfactory reproducibility and can be used to track changes in venous indices.

References 1.

2.

3.

Nicolaides AN, Christopoulos DG. Quantification of venous reflux and outflow obstruction with air plethysmography. In: Bernstein EF (ed). Vascular Diagnosis. 4th ed. St Louis, MO: CV Mosby Co; 1993:915–921. Bays RA, Healy DA, Atrip RG, Neumyer M, Thiele BL. Validation of air plethysmographic measurements for the evaluation of chronic venous disease. J Vasc Surg 1994; 20:721–727. Talbot S. Use of real time imaging in identifying deep venous obstruction: a preliminary report. Bruit 1982; 6:41–45.


4

Sullivan ED, Peter DJ, Cranley JJ. Real time B-mode venous ultrasound. J Vasc Surg 1984; 1:465–471.

5.

Hannan LJ, Stedjie KJ, Skorez MJ. Venous imaging of the extremities: our twenty-five hundred cases. 1986; 10:29–32.

6.

Kalodiki E, Matson R, Volteas N, et al. The combination of liquid crystal thermography and duplex scanning in the diagnosis of deep vein thrombosis. Eur J Vasc Surg 1992; 6:311–316.

7.

Szendo G, Nicolaides AN, Zukowski AJ, et al. Duplex scanning in the assessment of deep venous incompetence. J Vasc Surg 1986; 4:237–242.

8.

Rollins DL, Semrow C, Friendell ML, Buchbinder D. The use of ultrasonic venography in the evaluation of venous valve function. Am J Surg 1987; 154:189–191.

9.

Kalodiki E, Calahoras L, Nicolaides AN. Make it easy: duplex examination of the venous system. Phlebography 1993; 8:17–21.

10. Vasdekis S, Clarke GH, Nicolaides AN. Quantification of venous reflux by means of duplex scanning. J Vasc Surg 1989; 10:670–677. 11. Labropoulos N, Leon M, Nicolaides AN, Giannoukas AD, Volteas N, Chan P. Superficial venous insufficiency: correlation of anatomic extent of reflux with clinical signs and symptoms. J Vasc Surg 1994; 20:953–958. 12. Thibault PK. Duplex examination. Dermatologica 1995; 21:77–82. 13. van Bemmelen PS, van Ramshort B, Eike’boom BC. Photoplethysmography reexamined: lack of correlation with duplex scanning. Surgery 1992; 112:544– 548. 14. Rutgers PA, Kitslaar PS, Ermers EJ. Photoplethysmography in the diagnosis of superficial venous valvular incompetence. Br J Surg 1993; 80:351–353. 15. Allan JC. Volume changes in the lower limb in response to postural alterations and muscular exercise. S Afr J Surg 1964; 2:75–90. 16. Katz ML, Comerota AJ, Kerr R. Air plethysmography (APG): a new technique to evaluate patients with chronic venous insufficiency. J Vasc Tech 1991; 15: 23–27.

481


17. Christopoulos DG, Nicolaides AN, Szendo 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; 5:148–159. 18. Cordts PR, Hartono C, Lamorte WW, Menzoian JO. Physiological similarities between extremities with varicose veins and with chronic venous insufficiency utilizing air plethysmography. Am J Surg 1992; 164:260–264. 19. Christopoulos DG, Nicolaides AN. Noninvasive diagnosis and quantification of popliteal reflux in the swollen and ulcerated leg. J Cardiovasc Surg (Torino) 1988; 29:535–539. 20. Yang D, Vandongen YK, Stacey MC. Variability and reliability of air plethysmography measurements for the evaluation of chronic venous disease. J Vasc Surg 1997; 26:638–642.

raphy in the diagnosis of chronic venous insufficiency. J Vasc Surg 1998; 27:660–670. 29. Comerota AJ, Harada RN, Eze AR, Kartz ML. Air plethysmography: a clinical review. Int Angiol 1995; 14:45–52. 30. Sarin S, Sommerville K, Farrah J, Scurr JH, ColeridgeSmith PD. Duplex ultrasonography for assessment of venous valvular function of the lower limb. Br J Surg 1994; 81:159–165. 31. Araki CT, Back TL, Padberg FT Jr, Thompson PN, Duran WH, Hobson RW II. Refinements in the ultrasonic detection of popliteal vein reflux. J Vasc Surg 1993; 18:742–748. 32. Christopoulos DG, Nicolaides AN, Szendro G. Venous reflux: quantification and correlation with the clinical severity of chronic venous disease. Br J Surg 1988; 75:352–356.

21. Porter JM, Moneta GL. International Consensus Committee on Chronic Venous Disease: reporting standards in venous disease: an update. J Vasc Surg 1995; 2:635–645. 22. van Bemmelen PS, Bedford GR, Beach KW, Strandness DE. Quantitative segmental evaluation of venous valvular reflux with duplex ultrasound scanning. J Vasc Surg 1989; 10:425–431. 23. van Bemmelen PS, Mattos MA, Hodgson KJ, et al. Does air plethysmography correlate with duplex scanning in patients with chronic venous insufficiency? J Vasc Surg 1993; 18:796–807. 24. Belcaro G, Labropoulos N, Christopher D, et al. Noninvasive tests in venous insufficiency. J Cardiovasc Surg (Torino) 1993; 34:3–11. 25. Bland JM, Altman D. Statistical methods for assessing agreement between methods of clinical measurement. Lancet 1986; 8:307–310. 26. Katz ML, Comerota AJ, Kerr RP, Caputo GC. Variability of venous hemodynamics with daily activity. J Vasc Surg 1994; 19:361–365. 27. Booke TW, Nicolaides AN, Clement D. Laboratory assessment of circulation in limbs. In: Clement DL, Shepherd JT (eds). Vascular Disease in the Limbs: Mechanisms and Principles of Treatment. 1st ed. St Louis, MO: CV Mosby Co; 1993:103–131. 28. Criado E, Farber MA, Marston WA, Daniel PF, Burnham CB, Keagy BA. The role of air plethysmog-

482
