Original Article

Relationship between medical compression and intramuscular pressure as an explanation of a compression paradox

Phlebology 2015, Vol. 30(5) 331–338 ! The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0268355514527442 phl.sagepub.com

J-F Uhl1, J-P Benigni2, A Cornu-Thenard3, J Fournier4 and E Blin2

Abstract Background: Using standing magnetic resonance imaging (MRI), we recently showed that medical compression, providing an interface pressure (IP) of 22 mmHg, significantly compressed the deep veins of the leg but not, paradoxically, superficial varicose veins. Objective: To provide an explanation for this compression paradox by studying the correlation between the IP exerted by medical compression and intramuscular pressure (IMP). Material and methods: In 10 legs of five healthy subjects, we studied the effects of different IPs on the IMP of the medial gastrocnemius muscle. The IP produced by a cuff manometer was verified by a PicopressÕ device. The IMP was measured with a 21G needle connected to a manometer. Pressure data were recorded in the prone and standing positions with cuff manometer pressures from 0 to 50 mmHg. Results: In the prone position, an IP of less than 20 did not significantly change the IMP. On the contrary, a perfect linear correlation with the IMP (r ¼ 0.99) was observed with an IP from 20 to 50 mmHg. We found the same correlation in the standing position. Conclusion: We found that an IP of 22 mmHg produced a significant IMP increase from 32 to 54 mmHg, in the standing position. At the same time, the subcutaneous pressure is only provided by the compression device, on healthy subjects. In other words, the subcutaneous pressure plus the IP is only a little higher than 22 mmHg—a pressure which is too low to reduce the caliber of the superficial veins. This is in accordance with our standing MRI 3D anatomical study which showed that, paradoxically, when applying low pressures (IP), the deep veins are compressed while the superficial veins are not.

Keywords Medical compression, intramuscular pressure, interface pressure, three-dimensional modeling, leg vein anatomy

Introduction A compression dogma It has generally been accepted1 that very high pressures are required to compress deep veins, while lower pressures are able to reduce the caliber of superficial veins. A blood pressure cuff with a transparent window has been used by several authors2,3 to assess vein compression based on the placement of the compression device and the interface pressure (IP) exerted. In the standing position, a pressure of more than 70 mmHg is required to occlude deep veins.4 In the supine position, an IP of 25 mmHg is sufficient to significantly reduce the caliber

of the saphenous vein. These anatomical data were derived by Duplex ultrasound, but these findings have now also been confirmed by multislice computed tomography in the supine position with compression stockings of differing strengths.5 1

URDIA research unit, EA 4465, Paris Descartes University, France HIA Begin, St Mande´, France 3 Hoˆpital Saint Antoine, Paris, France 4 Zola Orthope´die, Paris, France 2

Corresponding author: J-F Uhl, 113 Avenue Victor Hugo, Paris 75116, France. Email: [email protected]

Downloaded from phl.sagepub.com at Karolinska Institutets Universitetsbibliotek on July 21, 2015

332

Phlebology 30(5)

Parameters of compression Compression therapy has been shown to be an efficient treatment for chronic venous insufficiency (CVI), but its precise mode of action remains unclear. One of the main parameters of compression therapy is the pressure exerted on the leg by the device (bandage or stocking). This is also known as the IP. Choosing the pressure to be exerted by the compression device is just like prescribing a drug: the IP should be adapted to the severity of the disease, as it was suggested by the consensus meeting of the International Compression Club in 2008.6

The compression therapy paradox The belief that very high pressures are required to compress deep veins has been called into question by several studies that have used magnetic resonance imaging (MRI) to assess leg anatomy: 1. Downie et al.7 showed by prone MRI scanning on eight healthy volunteers that a compression of 15–20 mmHg, provided by an appropriately sized, knee-length stocking (MedivenÕ Travel), reduced the mean venous cross-sectional area more in the deep veins (64%) than in the superficial veins (39%). 2. Partsch et al.8 presented a standing MRI study of a varicose vein patient, with and without compression stockings that provided an IP of 22 mmHg, measured with the PicopressÕ device at the B1 point.

Their observation showed that, in comparison with the same calf slice with no stocking, compression significantly reduced the diameter of the deep veins. This was particularly true of the muscular veins of the soleus, which completely disappeared. Paradoxically, on the same slice, no effect was seen on the varicose veins located on the medial aspect of the leg (Figure 1). 3. Using the same two-dimensional data, we recently performed9 three-dimensional modeling of the venous anatomy of the leg to better understand the mechanical effects of compression. This was achieved by a three-dimensional reconstruction of the anatomy of the leg, making it possible to accurately quantify the volume of each vein of the leg. The methodology used was to segment MRI slices. Using Winsurf software10 version 3.5, we were able to create a three-dimensional vector model of the anatomy of the whole leg by a manual segmentation of each anatomical structure of the leg and calf. This tedious work was done slice by slice by tracing the boundaries of each anatomical element. A realistic three-dimensional model of the leg was obtained (Figure 2). This allowed comparison of the leg with a compression stocking producing an IP of 22 mmHg versus no compression. The results of the standing varicose vein patient revealed threedimensional quantification of venous volumes with and without compression of differing strengths that produced increasing pressures at the B1 point (Figures 3 and 4). These data demonstrate that the

Figure 1. Comparison of MRI slices at the mid calf showing the anatomy with an IP of 22 mmHg produced by compression (on the right) and no compression (on the left). This shows that, in the standing position, the deep veins are flattened (red arrows), but not the varix (white arrows). 1: fibular veins; 2: posterior tibial veins; 3: anterior tibial vein; 4: veins of the soleus muscle.

Downloaded from phl.sagepub.com at Karolinska Institutets Universitetsbibliotek on July 21, 2015

Uhl et al.

333

pressure required to compress varicose veins is between 54 and 70 mmHg.

Objectives

points on the leg. These two particular points have been determined to be the optimal locations for measuring the IP.11 The corresponding anatomy for the B1 point is the apex of the calf (Figure 5). For the C point, the corresponding anatomical point is the gastrocnemius muscle at the largest diameter of the calf.12

The objective of this paper is to explain the compression paradox by measuring IMP under a stiff compression device at several different IPs.

Material and methods Ten legs of five healthy volunteers (age 60  7 years, BMI 22  4, mean  SD) were included in this study.

Experimental protocol First, a Duplex ultrasound investigation was performed to check the location of the gastrocnemius vessels so their puncture could be avoided. Landmarks were drawn on the skin of the medial aspect of each calf. Then, after disinfecting the skin, a blood pressure cuff was applied higher on the calf muscle. The small probe of a PicopressÕ device was placed at the C point to continuously check the IP exerted by the cuff. The IP can be easily measured by a pressure probe placed between the skin and the compression device. KikuhimeÕ and PicopressÕ devices are commonly used with a small probe placed over the B1 and C

Figure 3. Quantification of the volume of each vein of the calf in the same patient. Volume reduction (%) due to a compression of 22 mmHg in the standing position. AT: anterior tibial veins: PT: posterior tibial veins; Fib: fibular veins and veins of the soleus; MG: medial gastrocnemius veins; DEEP: all deep veins; VAR: varicose veins.

Figure 2. Three-dimensional reconstruction of the calf from the whole slice dataset of the same MRI. The calf muscles have been made transparent to show the deep veins. The three-dimensional models show even more clearly the difference between the presence and absence of compression. 1: fibular veins; 2: posterior tibial veins; 3: anterior tibial vein; 4: veins of the soleus muscle; 5: varix of the great saphenous vein; 6: small saphenous vein.

Downloaded from phl.sagepub.com at Karolinska Institutets Universitetsbibliotek on July 21, 2015

334

Phlebology 30(5)

Figure 4. Volume reduction (%) of the veins with different pressures exerted by a compression device at the B1 point. It shows that the pressure needed to flatten the varicose veins is between 51 and 83 mmHg. AT: anterior tibial veins: PT: posterior tibial veins; Fib: fibular veins and veins of the soleus; MG: medial gastrocnemius veins; VAR: varicose veins.

Figure 6. Position of the needle below the cuff for IMP measurement.

Figure 5. Optimal locations for IP measurement and their corresponding anatomy (three-dimensional reconstruction by MSCT). C is the point where the calf has the largest diameter; B is 3 cm above the malleoli; and B1 is located at the apex of the calf (the inferior end of the gastrocnemius).

injected. It was necessary to wait for 5–10 s to read the stable pressure value. Then, the cuff manometer was inflated under IP control in 10 mmHg increments from 0 to 50 mmHg, and the IMP was measured at each step: 0, 10, 20, 30, 40, and 50 mmHg. In the prone position, the subjects were completely still. In the standing position, the subjects were asked to remain immobile, standing equally on their both legs.

Results An intramuscular needle was connected to a manometer (StrykerÕ quick pressure monitor). Zero pressure was verified prior to insertion of the needle. The needle was bluntly inserted 4–5 cm into the medial gastrocnemius muscle in a slightly promixal direction, just below the cuff, at an angle of about 30 . This allowed the needle tip to be placed below the center of the cuff (Figure 6). Prior to each measurement of the IMP, a very small quantity of normal saline was

In the prone position, Figure 7 shows the IMPs according to the different IPs applied around the calf. We observed that an IP of less than 20 mmHg did not significantly change the IMP. This could be explained by the IMP at rest being higher than the IP (average IMP ¼ 11.4 mmHg with no compression, standard deviation ¼ 6). On the contrary, a perfect linear correlation with the IMP (r ¼ 0.99) was observed when the IP ranged from 20 to 50 mmHg.

Downloaded from phl.sagepub.com at Karolinska Institutets Universitetsbibliotek on July 21, 2015

Uhl et al.

335

Figure 7. Correlation of IMP vs. IP in the prone position (10 healthy legs). IMP: intramuscular pressure; IP: interface pressure.

Figure 8. Correlation of IMP vs. IP in standing position (10 healthy legs). IMP: intramuscular pressure; IP: interface pressure.

Figure 8 demonstrates that we also found a linear correlation between the IP and the IMP in the standing position. Without any compression, the average IMP was 34 mmHg  4.6 SD. These results provide a logical explanation for the compression paradox demonstrated in our MRI study of patients in the standing position. In fact, as shown in Figure 9, the IP of 22 mmHg exerted by the stocking should be added to the standing IMP to have the resulting pressure inside the muscular compartment. In other words, when standing: 22 þ 34 ¼ 56 mmHg. This pressure is able to reduce the caliber of the deep veins but has no effect on the superficial veins inside the subcutaneous tissue. This is because the subcutaneous pressure by itself is indeed very low (near 0), and so the resulting subcutaneous pressure with compression should be only about 22 mmHg. In the supine position, the same IP of 22 mmHg would produce little effect on the deep veins: 22 þ 11 ¼ 33 mmHg. The same is true for the superficial veins: 22 þ 0 ¼ 22 mmHg. This is in accordance with our three-dimensional anatomical study using standing MRI which showed that, paradoxically, low pressures can provide a significant reduction of the caliber of the deep veins, while the superficial veins and varicose veins are not compressed.

of the deep veins and 17% volume reduction of the varicose veins [unpublished data]. It should be also noticed that our study has been performed using a stiff compression device (cuff manometer) which produces a homogenous pressure compared to more elastic medical compression. Thus, these results should be confirmed with different kinds of medical compression. In addition, the intramuscular pressure (IMP) may vary slightly depending on the depth within the leg at which it is measured according to a recent numerical study.13

Discussion Limitations of our demonstration of the compression paradox Only one case of varicose veins has been reported to illustrate the compression paradox. This phenomenon is confirmed, however, by a series of five cases in the supine and standing positions which showed similar results; that is an average of 47% volume reduction

Limitations of the three-dimensional MRI modeling of the calf The first limitation is related to the long static MRI acquisition time for a very limited segment of the calf. As a result, only a small number of slices can be obtained with the patient(s) standing still. In reality, calf pump activity is a dynamic process and here the complex anatomical deformations produced by the compression device are limited to one static position. This limitation is unavoidable due to the restrictions placed by available MRI technology. More detailed anatomy of the whole leg with a shorter acquisition time will be available in the near future. The second limitation of our technique of threedimensional modeling is the variability of the morphology and anatomy of the leg. The phenomenon of the compression paradox requires validation in normal subjects as well as in those who have CVI. The anatomical effects of different IPs and stiffness need to be compared. The last limitation is the lack of hemodynamic data that exists concerning venous flow and velocity compared to the anatomical and morphological variations and abnormalities. Unfortunately, these kinds of data are impossible to acquire during MRI acquisition.

Downloaded from phl.sagepub.com at Karolinska Institutets Universitetsbibliotek on July 21, 2015

336

Phlebology 30(5)

Figure 9. Explanation of the compression paradox. Segmentation of the MRI slices to show the muscular compartments of the leg. IP: interface pressure.

For the future: A numerical approach and simulation by finite element analysis In order to simulate the deformation of the vessel resulting from compressive loads, finite element analysis is the best way to study anatomical and morphological changes of the leg compartments. A realistic simulation was achieved by Dai et al.14 A two-dimensional finite element analysis was performed to simulate venous collapse as a function of (1) venous pressure and (2) the magnitude and spatial distribution of skin IP. Circumferentially symmetric and asymmetric compression was compared to examine pressure distribution along the limb. The results showed that asymmetric compression produces greater vessel collapse and generated higher blood flow velocities and shear stresses than circumferential compression. Using MRI, Avril et al.15 and Bouten and Drapier16 proposed a mixed experimental and numerical approach to characterize the biomechanical response of the human leg under elastic compression. Using a finite element analysis with two- and three-dimensional simulation, they found discrepancies of more than 35% from one location to another, showing that the same compression garment cannot be applied for treating deficiencies of the deep venous system and simultaneously, superficial veins as well. Their final conclusion was that the morphology of the leg plays an important role. Recently, Rohan et al.13 developed a finite element model of a human leg with a varicose vein to compute

the stress distribution around the vein wall and analyze the biomechanical response. It showed that medical compression reduces the trans-mural pressure and narrows the vein. Lurie et al.17 used MRI and Duplex ultrasound to clarify the mode of action of intermittent pneumatic compression, which is more complex to assess than permanent compression. They found a measurable decrease in the volume of subcutaneous tissue under the garment. Surprisingly, no measurable changes occurred in the sub-fascial volume of the calf. The strongest single predictor of venous flow increase was change in subcutaneous vein volume. Narracott et al.18 validated subject-specific finite element models of the calf by using MRI cross sections. It was shown that both symmetric and asymmetric compression reduced the caliber of the posterior tibial veins (81% and 89%), but greater flattening of the anterior tibial veins was achieved with symmetric compression. They also found that external measurements of calf tissue deformation do not accurately predict deep vessel collapse. Wang et al.19 used a combination of MRI and computational fluid dynamics analysis. They found that reduction of deep vein volume was an important consequence of external compression and led to a higher blood velocity in the fibular veins. The large individual variability observed in their study suggests that the selection of a suitable compression stocking for a given individual needs to be refined.

Downloaded from phl.sagepub.com at Karolinska Institutets Universitetsbibliotek on July 21, 2015

Uhl et al.

337

Hach et al.20 found in the standing position an average pressure in the deep compartment of 30 mmHg in 10 healthy controls compared to 62 mmHg in 14 patients with venous compartment syndrome (p ¼ 0.003). Fullana et al.21 created a venous return simulator to try to predict the effects of compression of the veins of the leg. Finally, Murthy et al.,22 using both inelastic and elastic leggings, which provided a pressure of 14–27 mmHg at the calf, compared IMPs in 11 subjects in 1994. They also used soft catheters which enabled movement, including walking and running. IMP measurement was performed in the three deep compartments of the leg: soleus, gastrocnemius, and anterior tibial. The leggings had no effect on IMP during walking and running, but there was a significant increase in IMP in the standing position for low IPs (15–27 mmHg). The main limitation, however, of Murthy’s results was their improper use of CircaidÕ devices which, in our opinion, were applied too loosely in their study. Our study, like Murthy’s, only studied healthy subjects. It would be interesting to include patients with CVI and post-thrombotic syndrome. In fact, variations of IMP based on CVI have been shown by Christenson and coworkers.23,24 Subcutaneous pressures increase from 8 to 16 mmHg and from 12 to 34 mmHg for IMP with CVI severity with the highest pressures being for severe obstructive syndromes. In healthy subjects, subcutaneous tissue pressure is very low, less than 5 mmHg. So, the effects of compression seem to be located mainly on the deep system, and less on the superficial veins. The increase of the IMP acting on the muscular veins has a direct effect on the calf pump function and enhances calf pump ejection. This supports the idea that compression should be focused on the calf muscles, and that graduated compression along the leg is not mandatory.25,26 It also makes sense that higher IMPs, provided by the working pressure of inelastic compression devices, work more efficiently on the calf pump.27–29

Conclusion The mode of action of compression therapy on the anatomy and physiology of leg veins is not fully understood. In fact, the response of internal tissues, muscular compartments, and veins to the external pressure is still partially unknown. Our previous study using 3D modeling and quantification of venous anatomy by standing MRI shows that a compression stocking exerting a low pressure (22 mmHg) is able to reduce the caliber of the deep veins but, paradoxically, does not affect the superficial venous system.

This study produces a possible explanation of this paradox: the IMP of 34 mmHg in the standing position added to the IP of the stocking (22 mmHg) is able to flatten the deep veins inside the muscular compartment, while the subcutaneous pressure is only 22 mmHg, which is not sufficient to compress the superficial system. However, further investigation is needed to improve our knowledge about the complex anatomical and hemodynamic responses to compression. This would lead to a better understanding of the biomechanical mechanisms of compression therapy and, thus, to better clinical efficacy for CVI and assist in the prevention of deep vein thrombosis in the future. Acknowledgments We would thank Ted King, MD (Chicago) and Mark Malouf, MD (Sydney) for their advice and re-reading the paper.

Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest The authors declare that there is no conflict of interest.

References 1. Flour M, Clark M, Partsch H, Mosti G, Uhl J-F, et al. Dogmas and controversies in compression therapy: report of an International Compression Club (ICC) meeting, Brussels, May 2011. Int Wound J 2012; 10: 516–526. 2. Partsch H, Menzinger G, Borst-Krafek B, et al. Does thigh compression improve venous hemodynamics in chronic venous insufficiency? J Vasc Surg 2002; 36: 948–952. 3. Couzan S, Prufer M, Ferret JM, et al. Un nouveau concept de contention-compression: apport de l’e´cho-doppler couleur avec prise des pressions veineuses et de l’IRM. Phle´bologie 2002; 55: 159–171. 4. Partsch B and Partsch H. Calf compression pressure required to achieve venous closure from supine to standing positions. J Vasc Surg 2005; 42: 734–738. 5. Uhl J-F. 3D multislice CT to demonstrate the effects of compression therapy. Int Angiol 2010; 29: 411–415. 6. Partsch H, Flour M, Coleridge-Smith P, et al. Indications for compression therapy in venous and lymphatic disease. Consensus based on experimental data and scientific evidence under the auspices of the IUP. Int Angiol 2008; 27: 193–219. 7. Downie SP, Firmin DN, Wood NB, et al. Role of MRI in investigating the effects of elastic compression stockings on the deformation of the superficial and deep veins in the lower leg. J Magn Reson Imaging 2007; 26: 80–85. 8. Partsch H, Mosti G and Mosti F. Narrowing of leg veins under compression demonstrated by magnetic resonance imaging (MRI). Int Angiol 2010; 29: 408–410.

Downloaded from phl.sagepub.com at Karolinska Institutets Universitetsbibliotek on July 21, 2015

338

Phlebology 30(5)

9. Partsch H, Mosti G and Uhl J-F. Unexpected venous diameter reduction by compression stocking of deep, but not of superficial veins. Veins Lymphatics 2012; 1: e3. 10. Moody D and Lozanoff S. SURFdriver: a practical computer program for generating 3D models of anatomical structures. In: 14th annual meeting of the American Association of Clinical Anatomists, Honolulu, Hawaii, 8–11 July 1997. 11. Partsch H, Clark M, Bassez S, et al. Measurement of lower leg compression in vivo: recommendations for the performance of measurements of interface pressure and stiffness. Dermatol Surg 2006; 32: 224–233. 12. Uhl J-F, Benigni J-P and Cornu-The´nard A. Where should stiffness be measured in vivo? Veins Lymphatics 2013; 2: e5. 13. Rohan C-PY, Badel P, Lun B, et al. Biomechanical response of varicose veins to elastic compression: a numerical study. J Biomech 2013; 46: 599–603. 14. Dai G, Gertler JP and Kamm RD. The effects of external compression on venous blood flow and tissue deformation in the lower leg. J Biomech Eng 1999; 121: 557–564. 15. Avril S, Bouten L, Dubuis L, et al. Mixed experimental and numerical approach for characterizing the biomechanical response of the human leg under elastic compression. J Biomech Eng 2010; 132: 031006. 16. Bouten L and Drapier S. In-vivo identification of soft biological tissues using MR imaging. Eur J Comput Mech 2009; 18: 21–32. 17. Lurie F, Scott V, Hyo-Chun Y, et al. On the mechanism of action of pneumatic compression devices: combined magnetic resonance imaging and duplex ultrasound investigation. J Vasc Surg 2008; 48: 1000–1006. 18. Narracott AJ, John GW, Morris RJ, et al. A validated model of calf compression and deep vessel collapse during external cuff inflation. IEEE Trans Biomed Eng 2009; 56: 273–280. 19. Wang Y, Pierce I, Gatehouse P, et al. Analysis of flow and wall shear stress in the peroneal veins under external compression based on real-time MR images. Med Eng Phys 2012; 34: 17–27.

20. Hach W, Pra¨ve F, Hach-Wunderle V, et al. The chronic venous compartment syndrome. Vasa 2000; 29: 127–132. 21. Fullana J-M, Cros F, Becker F, et al. The venous return simulator: an effective tool for investigating the effects of external compression on the venous hemodynamics—first results after thigh compression. Vasa 2005; 34: 19–23. 22. Murthy G, Ballard RE, Breit GA, et al. Intramuscular pressures beneath elastic and inelastic leggings. Ann Vasc Surg 1994; 8: 543–548. 23. Christenson JT, Prins C and Gemayel G. Subcutaneous fasciotomy and eradication of superficial venous reflux for chronic and recurrent venous ulcers: mid-term results. Phlebology 2011; 26: 197–202. 24. Christenson JT. Relationship between tissue pressures and severity in venous disease. In: Abstracts of AVF. 2013, p.35. 25. Mosti G and Partsch H. High compression pressure over the calf is more effective than graduated compression in enhancing venous pump function. Eur J Vasc Endovasc Surg 2012; 44: 332–336. 26. Mosti G and Partsch H. Compression stockings with a negative pressure gradient have a more pronounced effect on venous pumping function than graduated elastic compression stockings. Eur J Vasc Endovasc Surg 2011; 42: 261–266. 27. Mosti G and Partsch H. The progressive medical compression stockings increase more the capacity of the venous pump of the calf than conventional graduated compression elastic stockings. Phlebology 2012; 65: 13–18. 28. Spence RK and Cahall E. Inelastic versus elastic leg compression in chronic venous insufficiency: a comparison limb size and venous hemodynamics. JVS 1996; 24: 783–787. 29. Mosti G, Mattaliano V and Partsch H. Inelastic compression increases venous ejection fraction more than elastic bandages in patients with superficial venous reflux. Phlebology 2008; 23: 287–294.

Downloaded from phl.sagepub.com at Karolinska Institutets Universitetsbibliotek on July 21, 2015

Relationship between medical compression and intramuscular pressure as an explanation of a compression paradox.

Using standing magnetic resonance imaging (MRI), we recently showed that medical compression, providing an interface pressure (IP) of 22 mmHg, signifi...
646KB Sizes 0 Downloads 3 Views