Eur J Vasc Endovasc Surg 22, 424428 (2001) doi:10.1053/ejvs.2001.1490, available online at http://www.idealibrary.

com on

The Value of the Peak Systolic Velocity Ratio in the Assessment of the Haemodynamic Signicance of Subcritical Iliac Artery Stenoses
S. B. Cof, D. Th. Ubbink, I. Zwiers, A. J. M. van Gurp and D. A. Legemate Department of Vascular Surgery, Academic Medical Center, Amsterdam, The Netherlands
Objectives: iliac stenoses with a PSV ratio between 2.0 and 3.0 in patients with symptomatic arterial obstructive disease of the legs might be misinterpreted when compared with intra-arterial pressure measurements (IAPM). The aim of this study was to compare the value of the PSV ratio with IAPM as the reference standard in the assessment of the haemodynamic signicance of subcritical iliac artery stenoses (iliac stenosis with PSV ratio between 1.5 and 3.5). Design, patients and methods: fty-eight iliac tracts in 53 consecutive patients with symptomatic arterial obstructive disease of the legs with an isolated stenosis with PSV ratio between 1.5 and 3.5 were studied prospectively. The results of those iliac duplex scanning were compared to IAPM. Results: a poor agreement was found between IAPM and PSV ratios. For the PSV ratios [2.0, 2.5 and 3.0 the sensitivities were 74%, 37% and 16%, respectively. The specicities were 70%, 90% and 95%, the positive predictive values 82%, 88% and 86%, respectively, and the negative predictive values 58%, 43% and 37%, respectively. Conclusion: the results of this study show that the PSV ratio parameter is not accurate enough to evaluate the haemodynamic signicance of subcritical iliac artery stenoses. Key Words: PSV ratio; Aortoiliac stenoses; Duplex scanning; Intra-arterial pressure measurement; Subcritical iliac stenoses.

Introduction In the investigation of aortoiliac and the femoropopliteal arteries by means of duplex scanning the peak systolic velocity (PSV) ratio is the most widely accepted diagnostic parameter in assessing the grade of atherosclerotic lesions. It is possible to differentiate between stenoses with less or more than 50% diameter reduction which in turn provides clinically relevant information to formulate optimal treatment strategy. The cut-off point proposed in the literature for the PSV ratio to differentiate between iliac stenoses with less or more than 50% diameter reduction varies between 1.8 and 3.0.1 Most studies included the entire spectrum of lesions, ranging from minimal to subtotal stenoses, and include symptomatic and asymptomatic iliac disease in the analysis. These issues may bias the results, particularly with regard to the accuracy and clinical value of the PSV ratio around the proposed cut-off point. In order to obtain the optimal PSV ratio cut-off point
Please address all correspondence to: D. A. Legemate, Department of Vascular Surgery, G4-111, Academic Medical Center, Meibergdreef 9, 1105 AZ, Amsterdam. 10785884/01/110424+05 $35.00/0 2001 Harcourt Publishers Ltd.

which discriminates between iliac lesions of less or more than 50% diameter reduction, most studies compared the PSV ratio with arteriography as reference standard. However, arteriography provides only morphological information and no functional information.28 Thus, it seems evident that some borderline haemodynamic signicant aortoiliac stenoses will not be detected by angiography. By contrast, intra-arterial pressure measurement (IAPM), is a more accurate assessment of the haemodynamic signicance of aortoiliac lesions, particularly when assessing borderline aortoiliac lesions.911 Previous studies suggest that aortoiliac stenoses with a PSV ratio in the range of 2.03.0 can be misinterpreted when compared with intra-arterial pressure measurements.1214 The aim of this study was to compare the value of the PSV ratio in the assessment of the haemodynamic signicance of subcritical iliac stenoses with IAPM as the reference standard. Subcritical Iliac stenoses were dened here as stenoses with borderline haemodynamic signicance with a PSV ratio between 1.5 and 3.5 (Fig. 1). This range is somewhat larger than 2.03.0 and was chosen because it is expected to contain the entire target population. This study was designed to study the ability of the

The Value of the Peak Systolic Velocity Ratio

425

Ischaemic leg and treatment indicated

Aortoiliac duplex scanning

Isolated iliac lesion

PSV ratio < 1.5, minimal lesion

PSV ratio between 1.5 and 3.5, subcritical lesion

PSV ratio > 3.5, critical lesion

IAPM

Fig. 1. Flow diagram from ischaemic leg to subcritical iliac stenosis and intra-arterial pressure measurement.

PSV ratio in the assessment of the haemodynamic signicance and not the diameter reduction of subcritical iliac stenoses. For this reason, intra-arterial pressure measurement was chosen as reference standard in this study.

Patients and Methods Between March 1999 and October 2000 all patients suffering from arterial obstructive disease of the lower limbs referred to our vascular laboratory for aortoiliac duplex investigation were studied according to a xed protocol. To judge the presence or absence of arterial insufciency the ankle-brachial index (ABI) was estimated at rest and after treadmill exercise. The presence of arterial insufciency was dened as an ABI <0.90. In case of an ABI [0.90 at rest, the ABI was reassessed after treadmill exercise. Besides aortoiliac duplex scanning, some patients also had femoropopliteal and/or cruropedal duplex scanning. Patients were included in the study when an isolated iliac lesion with a PSV ratio between 1.5 and 3.5 was found on duplex scanning. An isolated stenosis in the iliac arteries combined with femoropopliteal and/or cruropedal lesions was not a contraindication to inclusion. Patients with aortic stenoses were excluded. The study was approved by the medical ethical committee and written informed consent was obtained from all patients. To diminish the effect of serially connected multiple stenoses on the measurements of the pressure gradient,

which could obscure analysis of the pressure-to-velocity relationship, only those sites with an isolated stenosis on duplex scanning were included.15 The duplex scans were performed with 3.55.0 MHz B-mode linear or curved-linear probes incorporating a 3.5 MHz Doppler probe (Hewlett Packard Sonos 2000, Hewlett Packard, Andover, MA, U.S.A.). Doppler samples where taken at the site of the stenosis and at the site where the lowest pre- or poststenotic PSV value was measured. If a lesion was located at the origin of an arterial segment the post-stenotic PSV was taken. Velocity parameters were measured from the spectral waveforms in m/s. Duplex scanning was performed by either of the two experienced vascular technologists. IAPM were performed in the vascular laboratory within two weeks after the duplex scanning by the junior investigator (SC). These were performed without papaverine and after intra-arterial administration of 50 mg of papaverine via the common femoral artery. The pressure measurements were performed simultaneously in the radial artery and the femoral artery in the groin using a catheter (Arrow, 20G), and needle (Microlance3, 21G) respectively, both connected to a pressure transducer. The radial artery pressure was measured in the arm with the highest systolic blood pressure. In case cannulation of the radial artery was unsuccessful, blood pressure was measured indirectly at the upper arm with a Dinamap Plus pressure monitor (Dinamap Plus Vital Signs Monitor 8720, Johnson & Johnson, Tampa, Florida, U.S.A.). An absolute peak systolic pressure gradient of >20 mmHg at rest and/ or a decrease in femoral/radial pressure index ((FBI) of [15% after the administration of papaverine was considered haemodynamically signicant. This cut-off level preserves the mean between several recommended values in the literature.9,10 Sensitivity, specicity, positive and negative predictive values for the PSV ratios 2.0, 2.5 and 3.0 were calculated to judge the diagnostic power for each of the cut-off points.

Results In the study period of 20 months 554 patients were referred to our vascular laboratory for aortoiliac duplex investigation. Two patients did not consent and 499 patients did not meet the inclusion criterion. Fiftythree consecutive patients (44 male, 9 female), with a median age of 67 (range 3883) years, and 58 iliac segments with an isolated stenosis, were studied prospectively. The median ankle/brachial index of these patients at rest was 0.70 (range 0.261.1). Forty-eight
Eur J Vasc Endovasc Surg Vol 22, November 2001

426

S. B. Cof et al.

patients (52 limbs) had intermittent claudication and ve (6 limbs) rest pain. Thirty-three patients (35 limbs) also had femoropopliteal atherosclerotic lesions, seven patients (9 limbs) had both femoropopliteal and cruropedal atherosclerotic lesions. Thirteen patients (14 limbs) had neither femoropopliteal nor cruropedal lesions. Five patients underwent bilateral examinations. All duplex scannings were completed successfully. Median PSV ratio was 2.0 (range 1.53.4). In 15 patients radial artery cannulation was unsuccessful. In those patients brachial blood pressure was measured with a Dinamap Plus monitor. According to the pressure measurements, 38 (65.5%) stenoses were haemodynamically signicant, while 20 (34.5%) stenoses were not. Of the 38 haemodynamically signicant stenoses according to IAPM, 23 were found to be haemodynamically signicant without papaverine and 15 after the administration of papaverine. Of the 15 haemodynamically signicant stenoses after the administration of papaverine 7 stenoses had a PSV ratio between 1.5 and 1.9, 7 stenoses had a PSV ratio between 2.0 and 2.9 and 1 stenosis a PSV ratio between 3.0 and 3.5. To evaluate the results, the PSV ratio between 1.5 and 3.5 was subdivided into 3 categories: 1.51.9, 2.02.9 and 3.03.5. The results are shown in Figure 2. Seventy-nine per cent (27/ 34) of the 34 stenoses with PSV ratio between 2.0 and 3.5 were haemodynamically signicant according to the IAPM. The remainder (7/34) was misinterpreted by duplex scanning. Forty-two percent (10/24) of the stenoses with PSV ratio between 1.5 and 1.9 were found to be haemodynamically signicant. Figure 3 shows a scatterplot of the PSV ratio vs the systolic pressure gradient before vasodilatation and Figure 4 the PSV ratio vs the femoral/radial pressure index ( FBI) after vasodilatation. The reference lines represent the IAPM cut-off values at rest and after vasodilatation. No clear correlation is seen between the PSV ratio and the pressure measurements. In Figure 3, the signicant results below the reference line were found signicant after the administration of papaverine. In Figure 4, those below the reference line were already signicant at rest. Table 1 shows the results of the true positive, true negative, false positive and false negative for the PSV ratios 2.0, 2.5 and 3.5. A considerable number of false negative results are observed. The results of the comparison between the 3 PSV ratio cut-off points and intra-arterial blood pressure measurements are shown in Table 2. Both the sensitivity and the negative predictive value are less satisfactory for the cut-off levels due to the high number of false negative results.
Eur J Vasc Endovasc Surg Vol 22, November 2001

30 25 20
Stenoses

Significant Insignificant

15 10 5 0

1.51.9

2.02.9 PSV ratio

3.03.5

Fig. 2. Signicant and insignicant iliac stenoses according to intraarterial pressure measurements in 3 PSV ratio categories.

100 80 60 40 20 0

Psys before vasodil. (mmHg)

20 1.0

1.5

2.0

2.5

3.0

3.5

PSV ratio
Fig. 3. Results of intra-arterially measured pressure gradient at rest compared with PSV ratio at the site of the stenosis. Overall results of 58 iliac stenoses. ( haemodynamically signicant, haemodynamically insignicant).

Discussion Patients who required lower limb revascularization to correct ischaemic symptoms often are found to have occlusive disease in both the aortoiliac and femoropopliteal arterial segments.4 In these patients, unreliable knowledge of the iliac haemodynamic signicance may lead to misclassication of stenoses and incorrect management.16,17 Although duplex scanning is highly accurate in the discrimination between severe (PSV ratio >3.5) and minimal (PSV ratio <1.5) iliac stenoses, this study shows that the PSV ratio parameter is not accurate enough to differentiate between haemodynamically signicant and insignicant iliac artery stenoses with PSV ratios between 1.5 and 3.5. Our results are less comparable with others since

The Value of the Peak Systolic Velocity Ratio

427

50 40 30
FBI (%)

20 10 0 10 1.0

1.5

2.0

2.5

3.0

3.5

PSV ratio
Fig. 4. Results of the intra-arterially measured pressure gradient after administration of papaverine compared with the PSV ratio at the site of the stenosis. Overall results 58 iliac stenoses. ( haemodynamically signicant, haemodynamically insignicant).

Table 1. Overall diagnostic accuracy of PSV ratio in 58 iliac segments according to pressure measurements. True positive (TP), true negative (TN), false positive (FP) and false negative (FN) for PSV ratio 2.0, 2.5 and 3.0. PSV ratio [2.0 [2.5 [3.0 TP 28 14 6 TN 14 18 19 FP 6 2 1 FN 10 24 32 Diagnostic accuracy (%) 73 55 43

most studies have included the entire spectrum of iliac stenoses, ranging from minimal to subtotal stenoses.1 The disadvantage of including the entire spectrum of stenoses in the analysis is that this might yield a cutoff level for the PSV ratio that has good sensitivity and specicity, but not accurate enough to assess the haemodynamic signicance of borderline iliac stenoses. From Figure 2 it becomes clear that the PSV ratio is not accurate enough to differentiate between signicant and insignicant iliac stenoses with PSV ratio in the range of 1.5 and 3.5, in particular those between 1.5 and 1.9. Another conclusion from this study is that the haemodynamic signicance of some

subcritical stenoses can only be assessed during increased blood ow after the administration of papaverine, and not by pressure measurements at rest or by duplex scanning using the PSV ratio criterion. From a theoretical point of view it has been shown that the PSV ratio is not a ow-related parameter and provides only information directly related to changes in the cross-sectional area of the lumen.1821 Since moderate iliac stenoses may only be of haemodynamic signicance under circumstances of increased ow, it would be benecial to use a criterion which is not only related to the change in cross-sectional area, but also to the amount of ow. Hence to improve the assessment of the haemodynamic signicance of iliac stenoses some authors have postulated to use parameters which are not only related to morphological changes but also to changes in ow.19,22 Some additional explanations for the discrepancy between duplex scanning and pressure measurement and potential limitations of this study should be mentioned. Due to technical aspects of duplex scanning and the haemodynamics of blood ow, duplex scanning may show considerable errors in velocity measurements.23,24 This is also reected in the moderate observer agreement of the assessment of iliac stenoses with duplex scanning as found in a previous study.25 In this study, the number of false negatives are probably an illustration of the failure of the PSV ratio to differentiate between haemodynamically signicant and insignicant iliac lesions. Although difcult to prove, it is debatable whether IAPM is always the best standard reference for the assessment of the haemodynamic signicance of the aortoiliac tract. Intra-arterial pressure measurements may be adversely affected by kinetic energy of blood ow on the catheter. Air trapping, length of tubing system and the pressure transducer itself are potential sources of error in the measurement of intra-arterial blood pressure. This study did not address the relation between IAPM, the choice of treatment and the nal clinical outcome, which may put the value of duplex scanning and pressure measurement for subcritical iliac stenoses in a better perspective. The results of this study may

Table 2. Sensitivity (Sens), specicity (Spec), positive predictive value (PPV) and negative predictive value (NPV) for PSV ratio cut-off points of 2.0, 2.5 and 3.0, including 95% condence intervals. Overall results of 58 iliac stenoses. PSV ratio [2.0 [2.5 [3.0 Sens % 74 (0.700.87) 37 (0.220.54) 16 (0.060.31) Spec % 70 (0.460.88) 90 (0.680.99) 95 (0.751.00) PPV % 82 (0.660.93) 88 (0.620.98) 86 (0.421.00) NPV % 58 (0.370.78) 43 (0.280.59) 37 (0.240.52)

Eur J Vasc Endovasc Surg Vol 22, November 2001

428

S. B. Cof et al.

not affect the overall clinical utility of iliac duplex scanning. However, assessment of the haemodynamic signicance of iliac stenoses by means of PSV ratio should be used with caution in the interpretation of patient test results. The results of this study indicate that in cases of borderline iliac stenoses in combination with clinical symptoms of the patient, assessment may benet from intra-arterial pressure measurement.

Acknowledgement
We gratefully thank the Netherlands Heart Foundation for the nancial support of this study, (grant 98.164).

References
1 Koelemay MJW, Hartog D den, Prins MH et al. Diagnosis of arterial disease of the lower extremities with duplex ultrasonography. Br J Surg 1996; 83: 404409. 2 Archie Jr JP, Feldtman RW. Intraoperative assessment of the hemodynamic signicance of iliac and profunda femoris artery stenosis. Surgery 1981; 90: 876880. 3 Barber GG, Fong H, McPhail NV et al. Hemodynamic assessment of the aortoiliac segment: A prospective study. Can J Surg 1980; 23: 542544. 4 Brener BJ, Brief DK, Alpert J et al. The usefulness of intraarterial pressure. Measurements in occlusive arterial disease: Part I. Vasc Dis Ther 1982; 3: 3743. 5 Brener BJ, Raines JK, Darling RC et al. Measurement of systolic femoral artery pressure during reactive hyperemia: An estimate of aortoiliac disease. Cardiovasc Surg 1974; 4950: 259267. 6 Brewster DC, Waltman AC, OHara PJ et al. Femoral artery pressure measurement during aortography. Cardiovasc Surg 1979; 60: 120124. 7 Moore WS, Hall AD. Unrecognized aortoiliac stenosis: A physiologic approach to the diagnostic. Arch Surg 1971; 103: 633638. 8 Udoff EJ, Barth KH, Harrington DP et al. Hemodynamic signicance of iliac artery stenosis: pressure measurements during angiography. Radiologie 1979; 132: 289293. 9 Flanigan DP, Ryan TJ, Williams LR et al. Aortofemoral or femoropopliteal revascularization? A prospective evaluation of papaverine test. J Vasc Surg 1984; 1: 215223. 10 Baker AR, Macpherson DS, Evans DH, Bell PRF. Pressure studies in arterial surgery (review). Eur J Surg 1987; 1: 273283.

11 Flanigan DP, Williams LR, Schwartz JA, Schuler JJ, Gray B. Hemodynamic evaluation of the aortoiliac system based on pharmacologic vasodilatation. Surgery 1983; 93: 709714. 12 Legemate DA, Teeuwen C, Hoeneveld H, Eikelboom BC. Value of duplex scanning compared with angiography and pressure measurement in the assessment of aortoiliac arterial lesions. Br J Surg 1991; 78: 10031008. 13 Kohler TR, Nicholls SC, Eugene Zieler R, Beach KW, Schubart PJ. Assessment of pressure gradient by Doppler ultrasound: experimental and clinical observations. J Vasc Surg 1987; 6: 460469. 14 Smet AAEA de, Kitslaar PJEHM. A duplex criterion for aortoiliac stenosis. Eur J Vasc Surg 1990; 4: 275278. 15 Flanigan DP, Tullis JP, Streeter VL et al. Multiple subcritical arterial stenosis: effect on poststenotic pressure and ow. Ann Surg 1977; 186: 663668. 16 Summer DS, Strandness DE Jr. Aortoiliac reconstruction in patients with combined iliac and supercial femoral arterial occlusion. Surgery 1978; 84: 348355. 17 Kita MJ, Flanigan DP, Bishara RA et al. Long-term follow-up of patients having infrainguinal bypass performed below stenotic but hemodynamically normal aortoiliac vessels. J Vasc Surg 1987; 5: 319328. 18 Burns PN. Hemodynamics. In: Taylor KJW, Burns PN, Wells PNT eds. Clinical applications of Doppler ultrasound. New York: Raven Press, 1988: 5657. 19 Legemate DA, Teeuwen C, Hoeneveld H, Eikelboom BC. How can the assessment of the haemodynamic signicance of aortoiliac arterial stenosis by duplex scanning be improved? A comparative study with intra-arterial pressure measurements. J Vasc Surg 1993; 17: 676684. 20 Elsman BHP, Legemate DA, Lapham R, Mali WPThM, Eikelboom BC. Colour ultrasound duplex scanning after physical exercise: a new model to assess the hemodynamic signicance of stenoses in the aortoiliac tract. J Vasc Inves 1995; 1: 99103. 21 Coffi SB, Ubbink DTh, Zwiers I, Gurp JM van, Legemate DA. Assessment of the hemodynamic signicance of subcritical aortoiliac stenoses by means of hyperemic duplex scanning. J Vasc Tech 2001; 25: 2124. 22 Elsman BHP, Legemate DA, Vos HJ de, Mali WPThM, Eikelboom BC. Hypaeremic colour duplex scanning for the detection of aortoiliac stenoses. A comparative study with intraarterial pressure measurement. Eur J Vasc Endovasc Surg 1997; 14: 462467. 23 Gill RW. Measurement of blood ow by ultrasound: accuracy and sources of error. Ultrasound Med & Biol 1985; 11: 625641. 24 Phillips DJ, Beach KW, Primozich J, Strandness DE. Should results of ultrasound studies be reported in units of frequency or velocity? Ultrasound Med & Biol 1989; 15: 205212. 25 Ubbink DTh, Fidler M, Legemate DA. Interobserver variability in aortoiliac and femoropopliteal duplex scanning. J Vasc Surg 2001; 33: 540545. Accepted 29 July 2001

Eur J Vasc Endovasc Surg Vol 22, November 2001