Vascular Investigation and Therapy

ORIGINAL ARTICLE
Year
: 2019  |  Volume : 2  |  Issue : 4  |  Page : 88--94

Noninvasive measurement of lower limb outflow resistance andimplications for stenting


A Nicolaides1, O Maleti2, M Lugli2, S Guerzoni2,  
1 Department of Surgery, Medical School, University of Nicosia, Nicosia, Cyprus
2 Department of Cardiovascular Surgery, Hesperia Hospital, Modena, Italy

Correspondence Address:
Prof. A Nicolaides
Department of Surgery, Medical School, University of Nicosia, Nicosia
Cyprus

Abstract

AIMS: Stenting to relieve iliac vein obstruction is now practical and safe. However, the rates of ulcer healing, pain, and edema relief are 76%, 52%, and 42% respectively. In addition, a high ulcer recurrence (up to 86%) has been reported. Currently, patient selection depends on symptoms, imaging methods, and intravascular ultrasound (IVUS) for the assessment of iliac vein stenosis without consideration of the collateral circulation. The aim of this article is to present the results of a noninvasive method of measuring lower limb outflow resistance (LOR) and also to test the hypothesis that LOR is extremely variable in limbs with iliac obstruction and that some patients with iliac stenosis may have a LOR close to that of normal limbs as a result of a well-developed collateral circulation.MATERIALS AND METHODS: LOR was measured at different venous pressures from 60 to 25 mmHg using air-plethysmography in 15 limbs without and 15 limbs with iliac vein obstruction. Reflux in ml/sec (venous filling index [VFI]) and venous clinical severity score (VCSS) were also measured in all limbs. RESULTS: The LOR at 25 mmHg (LOR25) was found to be the most discriminating measurement between the two groups. The area under the receiver operating characteristic curve was 0.973 (95% confidence interval [CI] 0.923–1.000). The range of LOR25in limbs without obstruction was 0.0043–0.038 mmHg/ml/min and 0.0170–0.330 mmHg/ml/min in limbs with obstruction. By plotting VFI against LOR25, a subgroup of limbs was identified with iliac obstruction that had a high VCSS (5–12) and a near-normal LOR25(0.050 mmHg/ml/min) presumably as a result of a well-developed collateral circulation but a high VFI in the range of 5–14 ml/s. Another subgroup of limbs with iliac obstruction and a high VCSS (5–18) had a high LOR25 (0.100–0.330 mmHg/ml/min) presumably from a poorly developed collateral circulation. CONCLUSION: The noninvasive measurement of LOR25provides a quantitative estimation of overall lower LOR. It can indicate which limbs are compensated by the development of a good collateral circulation and which are not. The combination of LOR25with VFI enables the clinician to determine the relative contribution of reflux and obstruction in individual limbs. A low LOR25in the presence of severe iliac stenosis or occlusion is an indication of a well-developed collateral circulation and suggests that stenting would provide little benefit if any. However, this hypothesis needs to be verified by future prospective studies.



How to cite this article:
Nicolaides A, Maleti O, Lugli M, Guerzoni S. Noninvasive measurement of lower limb outflow resistance andimplications for stenting.Vasc Invest Ther 2019;2:88-94


How to cite this URL:
Nicolaides A, Maleti O, Lugli M, Guerzoni S. Noninvasive measurement of lower limb outflow resistance andimplications for stenting. Vasc Invest Ther [serial online] 2019 [cited 2020 Jul 4 ];2:88-94
Available from: http://www.vitonline.org/text.asp?2019/2/4/88/279229


Full Text



 Introduction



The measurement of leg volume changes and volume flow (Q) such as reflux in absolute units, ml, and ml/s, respectively, using air-plethysmograph (ACI) became possible when the instrument was calibrated in the 1980s.[1] By combining air-plethysmographic volume outflow changes with simultaneous venous pressure changes via a needle in a vein in the foot, it became possible to measure limb outflow resistance (LOR) in the 1990s.[2],[3],[4],[5] At that time, the measurement of outflow resistance provided a quantitative measure of outflow obstruction and contributed to a better understanding of the hemodynamic mechanisms of obstruction in patients with the postthrombotic syndrome (PTS), but in the absence of a practical method to treat obstruction, the method remained a research tool.

Stenting to relieve iliac vein obstruction is now practical and safe as shown by two recent systematic reviews and meta-analyses.[6],[7] However, in these reviews, the primary, assisted primary, and secondary patency rates were 68%, 82%, and 86% at 36 months, respectively; also, the rates of ulcer healing, pain, and edema relief were 76%, 52%, and 42% respectively. In addition, a high ulcer recurrence (up to 86%) has been reported.[8],[9]

Failure to relieve symptoms and heal ulcers after stenting a severe iliac stenosis can be attributed to the presence of venous reflux which is invariably present in patients with PTS or an insignificant contribution of the stent to lowering LOR because it is already near normal due to a well-developed collateral circulation.

Ultrasound examination can determine the presence or absence and anatomic extent of reflux and obstruction but does not provide quantitative measurements of the severity of either. Quantitative measurements of reflux in ml/s and obstruction in terms of LOR in mmHg/ml/min (the latter includes iliac vein and pelvic collaterals) may be helpful in determining which of the two is the most predominant and in deciding whether to stent, deal with reflux, or correct both.

A drawback of the method of measuring outflow resistance as already published[2],[3],[4],[5] had been the cannulation of a vein that was necessary to obtain pressure measurements. However, the method has been modified so that it is now noninvasive. The aim of this article is to present this noninvasive method of measuring lower LOR and also to test the hypothesis that LOR is extremely variable in both normal limbs and limbs with iliac stenosis and that some patients with iliac stenosis may have a LOR close to that of normal limbs as a result of a well-developed collateral circulation.

 Materials and Methods



Patient selection and investigation

Lower LOR was measured in 16 limbs of 16 patients who presented with symptomatic varicose veins (CEAP Clinical Class C1 or C2) and 14 limbs of 14 patients with edema, skin changes, or ulceration (CEAP Clinical Class C3–C6). They were selected from the pool of patients being investigated in the vascular laboratoty so that 15 had iliac vein obstruction and 15 had no evidence of obstruction.

Duplex scanning confirmed the presence of patent deep veins with competent valves without any evidence of outflow obstruction in 15 of the 16 patients in CEAP Class C1 and C2. There was iliac stenosis in one patient although there was no history of deep-vein thrombosis.

Duplex scanning confirmed the presence of axial reflux in the deep veins with the presence of residual old (echogenic) thrombus and severe (>70%) iliac vein obstruction in the 14 patients in the CEAP C3–C6 classes. The presence of iliac vein stenosis or occlusion was confirmed with computed tomography–venography or conventional venography.

The venous clinical severity score (VCSS) was calculated for all patients.

Air-plethysmography was performed to measure reflux (venous filling index [VFI])[1] and LOR as described below. This was part of the routine investigation of patients' assessment.

Patients were investigated in two different vascular laboratories. Patients number 1–18 were studied in Cyprus and the rest 19–30 in Modena. Two different air-plethysmographic systems were used: the APG® from ACI Medical, LIC, San Marcos, California, USA (info@acimedical.com) in the Cyprus laboratory and the air-plethysmographic equipment from Angioflow Microlab, Padova, Italy, in the Modena laboratory.

Measurement of reflux (Venous Filling Index) and limb outflow resistance

Measurement of reflux

The air chamber of the air-plethysmograph was placed around the calf. The chamber extended from the ankle to the knee. The air-plethysmograph was calibrated by injecting 100 ml of air in the air chamber. This provided a means of expressing volume changes of the calf in ml.[1] The 100 ml used for calibration was then removed. A 10 cm wide cuff was placed proximally on the thigh [Figure 1].{Figure 1}

The leg was elevated to 30° to empty the veins. When a plateau was reached, the patient was asked to stand up with the weight on the opposite leg while steadying himself by holding on an orthopedic frame until the volume increase reached a new plateau. The difference between the two plateaus represented the functional venous volume (FVV). The VFI was the average rate of reflux and was calculated by dividing the 90% FVV by the time taken for this level to be reached in seconds. It was expressed in ml/s.

Determination of pressure/volume relationship for the leg being examined (low recording speed)

The proximal thigh cuff was used to calibrate the leg being studied in terms of pressure and volume. The leg was elevated at 30° with the foot resting on a block and the proximal thigh cuff inflated to 80 mmHg [Figure 1]. When a plateau was reached, the cuff was deflated in steps of 10 mmHg, starting from the pressure of 80 mmHg and ending at the pressure of 10 mmHg. At each step, 1–2 min was required for a new plateau to be defined. For each plateau, corresponding calf volume measurements in ml were obtained (the distance of plateau from the baseline) [Figure 2]. Using these measurements, a pressure/volume curve could be constructed which was specific for the leg being examined [Figure 3]a.{Figure 2}{Figure 3}

Recording of outflow curve (high-speed paper)

The thigh cuff was again inflated to 80 mmHg, and when a plateau was reached, it was deflated suddenly providing a volume outflow curve [Figure 3]b. This was repeated twice so that three outflow curves were available. Curves with small motion artifacts were accepted [Figure 2].

Calculation of limb outflow resistance

Tangents at 0.2 s interval points on the outflow curve provided flow values in ml/min [Figure 3]b. Pressures corresponding to these points were obtained from the pressure/volume curve [Figure 3]a. LOR at every point was calculated by the equation: resistance = pressure/flow. Resistance measurements at pressures of ≤20 mmHg were ignored.

Statistical methods

Initially, a resistance curve was obtained for every limb by plotting the resistance (LOR) obtained at different pressures [Figure 4]a and b].{Figure 4}

Subsequently, boxplots were constructed for the LOR at different pressures (60, 50, 40, 30, and 25 mmHg) for the limbs without and with iliac obstruction [Figure 5]a, [Figure 5]B, [Figure 5]c.{Figure 5}

The areas under the receiver operating characteristic (ROC) curves for identifying patients with iliac stenosis from the LOR were obtained for LOR60, LOR50, LOR40, LOR30, and LOR25.

Scattergrams were obtained by plotting VFI against LOR25 to demonstrate the relationship of these measurements with the presence or absence of iliac stenosis and VCSS [Figure 6].{Figure 6}

Finally, a multivariable linear regression analysis was performed with VCSS as the dependent variable and with VFI and LOR25 as the explanatory variables [Table 1]. Using the regression equation obtained, the predicted VCSS was calculated using the values of VFI and LOR25. A plot of VCSS against the predicted VCSS was constructed, and the correlation coefficient (r) between the two was obtained to demonstrate the strength of the prediction [Figure 7].{Table 1}{Figure 7}

 Results



The presence of iliac stenosis, CEAP class, VCSS and and overall lower limb reflux (VFI) for the 30 patients studied are shown in [Table 2].{Table 2}

[Figure 4]a shows resistance curves at different pressures in five limbs without iliac stenosis and [Figure 4]b shows the resistance curves from five patients with iliac stenosis (only five curves from each group are shown for the sake of clarity). The increasing resistance at the low pressures of 30 mmHg and 25 mmHg is clearly shown.

The boxplots [Figure 5]a-c] show the decreasing overlap of LOR measurements between limbs with iliac stenosis and those without obtained at decreasing pressures of 60, 40, and 25 mmHg. The area under the ROC curve for identifying patients with iliac stenosis from the LOR was 0.850 (95% confidence interval [CI]: 0.664–1.00) at the pressure of 60 mmHg, 0.842 (95% CI: 0.647–1.00) at the pressure of 50 mmHg, 0.942 (95% CI: 0.830–1.0) at the pressure of 40 mmHg, 0.958 (95% CI: 0.869–1.00) at the pressure of 30 mmHg, and 0.973 (95% CI: 0.923–1.0) at the pressure of 25 mmHg.

In the 30 limbs studied, the lowest LOR recorded at a pressure of 25 mmHg (LOR25) was 0.0043 mmHg/ml/s and the highest 0.330 mmHg/ml/s which represents an increase of 77 times. The mean LOR25 was 0.014 ± 0.009 mmHg/ml/min in the limbs without obstruction and 0.111 ± 0.099 mmHg/ml/min in the limbs with obstruction (P < 0.0001). The range of LOR25 in the limbs without obstruction was 0.0043–0.038 mmHg/ml/min and 0.0170–0.330 mmHg/ml/min in limbs with obstruction.

The mean VFI was 2.64 ± 1.57 ml/s in the limbs without obstruction and 7.97 ± 3.47 ml/s in the limbs with obstruction (P < 0.0001). The range of VFI in the limbs without obstruction was 0.72–5.20 ml/s and 3.80–15.13 ml/s in limbs with obstruction.

[Figure 6]a is a scattergram of reflux (VFI) against LOR25, showing the large range of measurements in the patients without (blue) and patients with iliac stenosis (red). In 9 of the 15 limbs with stenosis (red dots in a dotted parallelogram in [Figure 6]a), the LOR is close to 0.05 mmHg/ml/min, suggesting that the stenosis is of mild hemodynamic significance. [Figure 6]b is the same scattergram showing the VCSS for each limb.

In a multivariable linear regression analysis with VCSS as the dependent variable and with VFI and LOR25 as the explanatory variables, both VFI and LOR25 were independent predictors of VCSS [Table 1]. By substituting VFI and LOR25 in the regression equation shown in [Table 1], the predicted VCSS was calculated for every patient. [Figure 7] is a scatterplot of the VCSS against the predicted VCSS showing the regression line and the correlation (r = 0.710; P < 0.0001) between them.

 Discussion



The results of this study show that both the overall lower limb reflux (VFI) and LOR can be obtained noninvasively using air-plethysmography. They also show that LOR is not the same at different pressures. It increases with decreasing pressure, especially in patients with iliac obstruction [Figure 4]. This is presumably the result of vein diameter reduction at pressures lower than 40 mmHg. At pressures lower than 25 mmHg, the resistance curves shown in [Figure 4] become almost vertical and make the resistance difficult to measure. This is because at pressures lower than 25 mmHg, the slope of the plethysmographic outflow curve is very small [Figure 3]b.

The results also demonstrate that the most discriminant LOR value is the one at 25 mmHg (LOR25) [Figure 5]a-c]. This is also shown by the high area under the ROC curve of 0.973 (95% CI 0.923–1.0) for the detection of iliac obstruction by LOR25.

Reflux, obstruction, or both contribute to the development of a high VCSS. However, until now, it was impossible to determine the relative hemodynamic contribution of each in the development of symptoms or signs. The plot of VFI against LOR25 is one way of visually determining the relative contribution of each parameter [Figure 6]. It appears that limbs with VFI <5 ml/s and LOR2525 from 0.017 to 0.33 [Figure 6] irrespective of the VCSS. In fact, [Figure 6] shows that a subgroup of patients with iliac obstruction (limbs enclosed in dotted rectangular) have LOR25 close to 0.05 mmHg/ml/min which is borderline abnormal and two limbs that have LOR25 in the normal range. Our new hypothesis is that limbs in this subgroup are the ones that if stented would show little improvement. These limbs have a high VCSS because of high reflux (VFI >5 ml/s) and are more likely to benefit from the valve or neovalve reconstruction.[10] In contrast, limbs with LOR25> 0.100 mmHg/ml/min are the ones likely to have a poor collateral circulation and are more likely to benefit from stenting. This hypothesis needs to be tested in prospective studies.

Finally, VFI and LOR25 are independent predictors of VCSS. The predicted VCSS which is calculated from the values of VFI and LOR25 is highly correlated with the actual VCSS (r = 0.710; P < 0.0001) [Figure 7]. These results confirm our previous finding using the invasive method of measuring LOR.[5]

A limitation of the study is that it is based on a relatively small number of patients. The findings need to be validated in a larger series. Another limitation is that the duration of the reflux or obstruction has not been considered. It is well known that skin changes become worse with time, even in the presence of low-grade reflux. In the absence of any consideration or correction for the duration the pathology was present, the correlation found between the VCSS and predicted VCSS calculated from VFI and LOR25 is remarkable.

 Conclusion



The noninvasive measurement of LOR25, which is a measurement that accounts for the effect of both iliac lesions and the collateral venous development, provides a quantitative measurement of overall lower LOR. It can indicate which limbs are compensated by the development of a good collateral circulation and which are not. The combination of LOR25 with VFI enables the clinician to determine the relative contribution of reflux and obstruction in individual limbs. A low LOR25 in the presence of severe iliac stenosis or occlusion is an indication of a well-developed collateral circulation and suggests that stenting would provide little benefit if any. However, this hypothesis needs to be verified by future prospective studies. Such studies should include measurements of VFI and LOR25 before and after stenting followed by correlation of these hemodynamic parameters with clinical outcomes so that eventually we can refine the indications for stenting.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1Christopoulos D, Nicolaides AN. Noninvasive diagnosis and quantitation of popliteal reflux in the swollen and ulcerated leg. J Cardiovasc Surg (Torino) 1988;29:535-9.
2Nicolaides AN, Sumner DS. Investigation of patients with deep vein thrombosis and chronic venous insufficiency. London, Los Angeles, Nicosia: Published by Medorion Publishing Company; 1991. p. 58-59
3Nicolaides AN, Cardiovascular Disease Educational and Research Trust, European Society of Vascular Surgery, The International Angiology Scientific Activity Congress Organization, International Union of Angiology, Union Internationale de Phlebologie at the Abbaye des Vaux de Cernay. Investigation of chronic venous insufficiency: A consensus statement (France, March 5-9, 1997). Circulation 2000;102:E126-63.
4Labropoulos N, Volteas N, Leon M, Sowade O, Rulo A, Giannoukas AD, et al. The role of venous outflow obstruction in patients with chronic venous dysfunction. Arch Surg 1997;132:46-51.
5Nicolaides A, Clark H, Labropoulos N, Geroulakos G, Lugli M, Maleti O. Quantitation of reflux and outflow obstruction in patients with CVD and correlation with clinical severity. Int Angiol 2014;33:275-81.
6Razavi MK, Jaff MR, Miller LE. Safety and effectiveness of stent placement for iliofemoral venous outflow obstruction: Systematic review and meta-analysis. Circ Cardiovasc Interv 2015;8:e002772.
7Qiu P, Zha B, Xu A, Wang W, Zhan Y, Zhu X, et al. Systematic review and meta-analysis of iliofemoral stenting for post-thrombotic syndrome. Eur J Vasc Endovasc Surg 2019;57:407-16.
8de Wolf MA, de Graaf R, Kurstjens RL, Penninx S, Jalaie H, Wittens CH. Short-term clinical experience with a dedicated venous nitinol stent: initial results with the sinus-venous stent. Eur J Vasc Endovasc Surg 2015;50:518-26.
9Yin M, Shi H, Ye K, Lu X, Li W, Huang X, et al. Clinical assessment of endovascular stenting compared with compression therapy alone in post-thrombotic patients with iliofemoral obstruction. Eur J Vasc Endovasc Surg 2015;50:101-7.
10Maleti O, Lugli M. Neovalve construction in postthrombotic syndrome. J Vasc Surg 2006;43:794-9.