- Email: [email protected]
Hemodynamic effects of intermittent pneumatic compression in patients with critical limb ischemia
From the Society for Clinical Vascular Surgery
Hemodynamic effects of intermittent pneumatic compression in patients with critical limb ischemia Nicos Labropoulos, PhD, DIC, RVT, Luis R. Leon, Jr, MD, RVT, Ahmad Bhatti, MD, RVT, Steven Melton, MD, Steven S. Kang, MD, Ashraf M. Mansour, MD, and Marc Borge, MD, Maywood, Ill Background: Traditional teaching assumes that the distal arterial tree is maximally dilated in patients with critical limb ischemia (CLI). Endovascular or arterial bypass procedures are the commonly used interventions to increase distal perfusion. However, other forms of treatment such as spinal cord stimulation or intermittent pneumatic compression (IPC) have been shown to improve limb salvage rates. This prospective study was designed to determine if the use of IPC increases popliteal, gastrocnemial, collateral arterial, and skin blood flow in patients with CLI. Methods: Twenty limbs with CLI in 20 patients (mean age, 74 years) were evaluated with duplex ultrasound scans and laser Doppler fluxmetry in the semi-erect position before, during, and after IPC. One pneumatic cuff was applied on the foot and the other on the calf. The maximum inflation pressure was 120 mm Hg and was applied for 3 seconds at three cycles per minute. All patients had at least two-level disease by arteriography. Fourteen limbs were characterized as inoperable, and six were considered marginal for reconstruction. Flow volumes were measured in the popliteal, medial gastrocnemial, and a genicular collateral artery. Skin blood flux was measured on the dorsum of the foot at the same time. Results: Significant flow increase during the application of IPC was found in all three arteries (18/20 limbs) compared with baseline values (P < .02). The highest change was seen in the popliteal, followed by the gastrocnemial and the collateral artery. After the cessation of IPC, the flow returned to baseline. This was attributed to the elevation of time average velocity, as the diameter of the arteries remained unchanged. The skin blood flux increased significantly as well (P < .03). In the two limbs without an increase in the arterial or skin blood flow, significant popliteal vein reflux was found. Both limbs were amputated shortly after. Conclusions: IPC increases axial, muscular, collateral, and skin blood flow in patients with CLI and may be beneficial to those who are not candidates for revascularization. Patients with significant venous reflux may not benefit from IPC. This supports the theory that one of the mechanisms by which IPC enhances flow is by increasing the arteriovenous pressure gradient. ( J Vasc Surg 2005;42:710-6.)
Intermittent pneumatic compression (IPC) is widely used as prophylaxis for deep venous thrombosis (DVT).1 In the early 1930s, IPC was shown to have a positive effect on lower-extremity blood flow of the calf and foot.2,3 Later, clinical benefits in patients with various degrees of limb ischemia were demonstrated, including those with critical limb ischemia (CLI).4,5 CLI occurs when the blood supply to the capillary beds is less than that necessary to maintain tissue viability. Arteriolar vasodilation, resulting in a reduction of peripheral vascular resistance, is a compensatory response to CLI.6 Peripheral arterioles in CLI are relatively insensitive to vasodilator stimuli compared with those in control subjects. Consequently, arterioles in CLI patients are thought to be “maximally dilated,” a phenomenon known as vasomotor paralysis, which has become a traditional assumption.7-9 Significant evidence suggests that IPC increases temporarily the arterial blood flow in healthy subjects, in patients with intermittent claudication, and in those with peripheral arterial disease with prior successful surgical revascularization.3,10-14 However, Duplex ultrasound (DU) or laser From the Department of Surgery, Loyola University Medical Center. Competition of interest: none. Correspondence: Nicos Labropoulos, PhD, RVT, DIC, Director, Vascular Laboratory, University of Medicine & Dentistry of New Jersey, 150 Bergen Street, Room D-447, Newark, NJ 07101-1709 (e-mail: [email protected]; [email protected]). 0741-5214/$30.00 Copyright © 2005 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2005.05.051
710
Doppler blood flow (LDF) evaluation of the lower extremities in patients with CLI has not been yet reported. We postulate that the demonstrated clinical effect of IPC on limbs with CLI is due to further increases in the arterial blood flow at all levels, either because of a reversal of vasomotor paralysis or an increase in the lower-extremity arteriovenous gradient caused by venous emptying from intermittent compression. This prospective study was designed to measure the hemodynamic effects of IPC in patients with CLI in the systemic, muscular, and collateral circulation as well as in the foot skin blood flow. PATIENTS AND METHODS Limbs of patients with CLI were evaluated with DU and LDF in the semi-erect position before, during, and after IPC. CLI was defined according to the TransAtlantic Inter-Society Consensus document15 as those patients whose arterial disease has resulted in a breakdown of the skin (ulcer or gangrene) or pain in the foot at rest. It corresponds to stages III and IV of the Fontaine classification. The semi-erect position was chosen to study these limbs for two reasons: it was the natural position adopted by these patients because of severe discomfort while lying supine, and to increase the arterio-venous gradient.16-18 The ArtAssist device (model AA-1000; ACI Medical Inc., San Marcos, California) was used in all cases. Inflatable 12- and 22-cm bladders in a single cuff structure were
JOURNAL OF VASCULAR SURGERY Volume 42, Number 4
applied on the foot and calf, respectively. The maximum inflation pressure of 120 mm Hg was applied for 3 seconds at three cycles per minute. The foot bladder was inflated first, and the calf bladder, 2 seconds later. This is a rapid inflation device, with a time of 300 to 360 milliseconds to reach peak pressure. All patients underwent digital subtraction arteriography to optimize visualization and vessel conspicuity with maximum contrast of the potential distal target arteries. The images were obtained with the target limb gently but securely immobilized to limit patient motion, which would cause misregistration artifact on the digital subtraction images. Whenever possible, the target area was imaged in an anatomic position (either anteroposterior or lateral) to optimize correct identification of distal target vessels. Nonionic and isosmolar contrast agents were used to minimize patient discomfort. Appropriate collimation, filtering and catheter positioning were used to optimize radiographic exposure in the target area. Whenever possible, contrast was injected from the ipsilateral side to maximize contrast opacification. Long, 5- to 10-second infusion times and delay were used to allow for complete filling of any patent channels in the target extremity. Peripheral vasodilators (tolazoline, nitroglycerine) or hot compress warming were often used to maximize peripheral vasodilatation. The images were reviewed by an interventional radiologist (M. B.) and by two vascular surgeons (A. M., S. S. K.). They were classified as inoperable if no target vessels were identified for revascularization and marginal for reconstruction if the target vessels had a small caliber or diffuse disease that made them not suitable for a durable bypass procedure or intervention. For the purpose of this study, patients selected had a patent popliteal artery. Before each measurement, the subjects were acclimatized in supine position on a couch for at least 10 minutes. Flow volumes were measured in the popliteal, medial gastrocnemius, and a genicular collateral artery. Artery blood flow was measured by using a 4- to 7-MHz linear array transducer (ATL-Ultramark 9; Advanced Tech Laboratory, Bothell, Wash). The internal arterial diameter was measured by imaging the vessel longitudinally with real time B-mode. The average of three repeated measurements was obtained. Velocity measurements were performed at an angle of 60°, with the gate of the sample volume matching the internal diameter of the artery. The mean volume flow was calculated from the time average mean velocity of at least five cycles. At the same time, skin blood flux was measured at the dorsum of the foot by using LDF (Laserflo model BPM 403A, TSI Inc, St Paul, Minn) and recorded in flux units. The BPM 403A has a semiconductor gallium-aluminumarsenide diode that emits a laser beam with a 780-nm wavelength. For all studies, a right angle probe was used. The probe was fixed on the skin by means of a double adhesive ring and adhesive tape. LDF readings at rest, before the initiation of the pump, were stabilized over a 5-minute period. The average resting skin blood flow was
Labropoulos et al 711
Table I. Demographic characteristics of the study population before intervention Mean age (range) Males Females Mean ABI (range) Mean TcPO2 (range) Diabetes mellitus Hypertension Hypercholesterolemia Smoking Coronary artery disease Rest pain Tissue loss History of stroke History of prior bypass
74 (65–83) 13 7 0.25 (0.1–0.29) 21.6 (12–31) 10 15 13 12 11 20 9 2 8
ABI, Ankle-brachial index; TcPO2, transcutaneous oxygen pressure measurements.
calculated over 30 seconds. The skin blood flow during the pump application was not measured because of significant motion artifact and was averaged for 10 seconds immediately after. No attempts were made to correlate the hemodynamic findings with transcutaneous oxygen pressure measurements (TcPO2), as they were not performed simultaneously with the other two methods. The degree of discomfort produced by the application of the pump was graded as none, mild, moderate, and severe. In cases of severe discomfort, the pump was discontinued and the patient was excluded from the study. Statistical analysis. Descriptive statistics were used for patient demographic characteristics. A paired t test was used to compare the arterial flows and diameters before, during, and after the application of IPC. Data were expressed as a mean and 95% confidence intervals (CI). The Wilcoxon sum-rank test was used for the LDF data, as their distribution was nonparametric. These data were expressed as median and interquartile range (IQR). The level for statistical significance was set at P ⫽ .05. RESULTS Twenty limbs with CLI in 20 patients (mean age, 74 years) were evaluated. The demographic characteristics are summarized in Table I. An additional patient was excluded because of inability to tolerate the foot component of the pump due to severe pain. All other patients tolerated it well, with mild-to-moderate discomfort that disappeared after a few pump cycles. All of the patients in our study had rest pain, and nine had tissue loss. All had an ankle-brachial index (ABI) of ⬍0.3, and the mean TcPO2 was 21.6. All patients had at least two-level disease by arteriography. Stenosis or occlusion in the aortoiliac segment was found in 17 patients, common femoral in 6, deep femoral in 6, superficial femoral in 18, and all patients had infrapopliteal disease. Fourteen limbs were inoperable, and six were marginal for reconstruction. Significant flow increase during the application of IPC was found in all three arteries (18/20 limbs) compared
JOURNAL OF VASCULAR SURGERY October 2005
712 Labropoulos et al
Table II. Mean and 95% confidence intervals (CI) of arterial blood flow before, during, and after intermittent pneumatic compression Artery Popliteal Gastrocnemial Collateral
Before During After Before During After Before During After
Mean
95% CI
22 45 20 10 17 9 9 14 8
19.7–24.3 39.4–50.6 17.7–22.3 8.6–11.4 15.1–18.9 7.6–10.4 8.06–9.94 12.6–15.4 6.6–9.4
P ⬍ .01 for comparisons before vs during, and during vs after, in all three arteries. The before and after values were comparable.
with baseline values (P ⬍ .02) (Table II). The highest change was seen in the popliteal (104.5%), followed by the gastrocnemial (70%) and the collateral artery (55.6%). After the cessation of IPC, the flow returned to baseline. This was attributed to the elevation of time average velocity as the diameter of the arteries remained unchanged (Fig 1 and Table III). The skin blood flux increased significantly as well (3.5 flux units, IQR 2 to 4.2 vs 5.2 flux units, IQR 3.8 to 6.2; P ⬍ .03) (Fig 2). In the two limbs without an increase in the arterial or skin blood flow, significant popliteal vein reflux was found (Fig 3). This reflux was of high velocity and ⬎5 seconds in duration in both cases. Both limbs were amputated shortly after. The remaining 18 patients did not have popliteal reflux. The rest of the veins were not examined, but none of them had clinically evident varicose veins or a history of venous thrombosis. Two patients were lost at a 6-month follow-up. Four patients underwent amputations at 10, 14, 30, and 60 days after the experiment.
Fig 1. Color flow duplex imaging in the evaluation of volume blood flow in the popliteal artery. A, At rest, the waveform is abnormal because of multiple levels of disease. B, during intermittent pneumatic compression application, the volume flow almost doubled, with significant increase in both the peak systolic and end-diastolic velocities.
DISCUSSION Several trials on the effects of IPC on the arterial system in the lower limbs have been reported. A meta-analysis of studies between 1966 and 2001 identified several limitations.19 There were differences in the type of devices used (sequential vs simultaneous compression), protocols of cycle length, and duration of treatment, severity of limb ischemia, flow measurement techniques, and end points for analysis. Moreover, only three prospective randomized studies were found, all with small sample sizes. Despite those limitations, a clear trend toward a favorable effect on limb hemodynamics was observed. The first prospective, randomized, controlled evidence for an increase in arterial blood flow with IPC in patients with intermittent claudication was reported in 1993,20 and that was followed by three additional trials.5,21,22 They showed significant beneficial effects on walking distance, symptoms, and systolic blood pressure measurements on the upper and lower limbs with therapy. However, studies on CLI and IPC have been limited to clinical outcomes4,23-25 and case reports.26 The case series are mostly retrospective.23,24
Montori et al23 retrospectively analyzed IPC in 107 patients with arterial ulcers. With a median follow-up of 6 months, wound healing was seen in 40% of patients with TcPO2 levels ⬍20 mm Hg. Vella et al24 retrospectively analyzed IPC in 96 patients with ischemic ulcers and found benefit in their treatment at all TcPO2 levels. Van Bemmelen et al25 studied 14 consecutive ischemic legs in a Veterans Administration hospital. The patients were deemed poor candidates for surgery and were treated with IPC for 4 hours per day at home for 3 months. Clinical outcomes and calibrated pulse-volume amplitudes were recorded. Nine legs were salvaged. The only prospective trial was a pilot study of clinical outcomes involving IPC therapy of 33 limbs with CLI.4 At a mean follow-up of 3 months, 58% of legs were saved. Rest pain improved in 40% of patients, 26% of foot ulcers healed, and toe pressures significantly improved. Several studies have shown increased popliteal artery blood flow during IPC application in healthy subjects and in patients with claudication.13,14,32,34 However, our study
JOURNAL OF VASCULAR SURGERY Volume 42, Number 4
Labropoulos et al 713
Table III. Mean vessel diameters (95% confidence interval; range) in millimeters, before, during, and after intermittent pneumatic compression Artery
Before
During
After
Popliteal Gastrocnemial Collateral
4.02 (3.69–4.31; 3–5.7) 1.91 (1.72–2.08; 1.3–2.9) 1.79 (1.60–1.98; 1.1–2.8)
4.06 (3.77–4.35; 3.2–5.7) 1.95 (1.75–2.14; 1.3–3) 1.82 (1.63–2.01; 1.1–2.8)
3.94 (3.65–4.21; 3–5.5) 1.86 (1.66–2.04; 1.2–3) 1.75 (1.56–1.94; 1.1–2.6)
P ⬎ .1 for comparisons before, during, and after in all three arteries
Fig 2. Laser Doppler flux recording at rest, during, and after the application of intermittent pneumatic compression. During intermittent pneumatic compression, there is a motion artifact with a gradual increase of the flux.
is the first to demonstrate significant improvements in the popliteal artery blood flow in patients with CLI. It is also the first study to demonstrate significant flow increases in muscular and collateral arteries. The previous reports and the current study support arteriovenous pressure gradient augmentation as the main mechanism for the flow increase. This has been shown by the reduction of the venous pressure during IPC application.16 In the sitting position, the venous pressure in the foot is about 60 mm Hg. The application of IPC reduces this pressure and therefore increases the arteriovenous pressure gradient. The resting blood flow in the popliteal artery in patients with CLI is reported for the first time in the literature. Compared with previous publications, the resting blood flow in the sitting position in claudicants is significantly higher than in our patients with CLI.13,34 This can be explained by the multiple levels of the occlusive disease as depicted by the angiograms and the very low ABI values. The relative increase of blood flow volumes in patients with CLI was significant for all three arteries during IPC. However, the magnitude of change was significantly less compared with previous studies that evaluated patients with claudication.13,34 With increasing levels of disease, more vasodilatation is likely present, and therefore, the margin for flow enhancement is less. Vasodilatation is one of the mechanisms by which limbs with non-CLI obtain an increase in their blood flow.27 This is through the production of nitric oxide during mechanical stimulation (device inflation). It has been recently proposed that the application of IPC causes a forward pulsatile flow, resulting in emptying of blood at that site. The increase of blood volume and velocity exerts a compressive strain and enhances the shear stress on the venous endothelial cells.27 IPC has also an effect on the arterial system, as the sudden drop in the venous pressure causes an increase in
the arteriovenous gradient that augments the arterial blood velocity. As a consequence, the shear stress on arterial endothelial cells is increased.27 Cell cultures and animal models have shown that the mechanical forces of shear and strain may elicit vasodilatory, antithrombotic, and profibrinolytic responses in endothelial cells.27 Patients with CLI are assumed to have maximal vessel dilatation caused by the critical lack of oxygen.7-9 Our findings bring the possibility of a submaximal level of dilatation, which could be raised with compression. Such vasodilation should occur in the arteriolar level, as we showed that there is no change in the diameter of the popliteal, gastrocnemial, and collateral arteries during IPC. Other studies have also shown no change in the diameter of the popliteal artery in healthy subjects and in patients with claudication during IPC.22,32,33 This is also substantiated by the increased LDF in our patients, which measures flow in the microcirculation. Further research is required to elucidate this matter. Another question is related to the magnitude of the blood flow increase. We hypothesize that collaterals showed the lowest increase in blood flow because vessel dilatation in these arteries is likely almost maximum. The same theory may be applicable to muscular arteries, smaller in size and probably fully dilated because of muscle ischemia. The popliteal artery had the greatest flow increase. This was possibly because of the additive effect of the diminished peripheral resistance in collateral and muscular branches that the main popliteal artery supplies. It is of interest to observe two occurrences of popliteal vein reflux in patients who ultimately underwent limb amputation shortly after the study. No reports have addressed the causes for failure to achieve beneficial arterial hemodynamic effects with IPC. Malone et al28 demonstrated that patients with post-thrombotic venous disease have a compromised venous hemodynamic response to all IPC devices. Furthermore, they showed an increased velocity response to high-pressure, rapid-inflation IPC machines. But this has only been evaluated in the context of DVT prophylaxis. Van Bemmelen et al25 did not see any hemodynamic improvement in limbs that were amputated after IPC therapy. No comments are made as to the reasons for failure and no documentation of reflux made. Conceivably, significant axial vein reflux will impede a clinically significant arteriovenous gradient, limiting flow augmentation in the arterial system, supporting further this theory as the underlying mechanism for the success of IPC devices. The introduction of novel IPC designs could po-
714 Labropoulos et al
Fig 3. Popliteal artery blood flow in a patient with significant popliteal reflux after a femoropopliteal thrombosis. A, Flow at rest has high resistance due to popliteal vein insufficiency. B, High velocity and long duration of reflux in popliteal vein. C, No change in the volume flow is seen during the application of intermittent pneumatic compression.
tentially improve venous emptying and arterial flow in these subjects.29 Commercially available IPC devices have constant cycles of compression and deflation. It has been shown that vein refilling times vary with limb position and with different venous pathologies.29,30 Relatively new systems have been designed with the ability to detect venous volume changes and to respond by initiating the subse-
JOURNAL OF VASCULAR SURGERY October 2005
quent cycle when the veins are again full. So far, this technology has not been tested from the arterial flow point of view but appears feasible and promising. No studies have been designed to identify the ideal compression sequence to enhance arterial blood flow maximally in the lower limbs of patients with ischemia. However, following the premise that IPC acts through enhancement of venous flow and increasing the arteriovenous gradient, we chose our treatment strategy by selecting the optimal way for venous emptying. This was based on a study that showed that foot and calf compression of 120 to 140 mm Hg with a frequency of 3 to 4 impulses per minute and 1-second delay was the optimum stimulus, as it had the maximal reduction in the venous pressure.16 LDF has been used to assess blood flow at the foot skin level in patients with different levels of limb ischemia.12,35 However, foot skin level assessment of blood flow has never been reported in patients with CLI. Our study is the first evidence of an improvement of blood flow at the skin level in this subgroup, which once more is against the presumed clinical dictum of maximal dilatation. This supports further the theory of a submaximal level of dilatation even at the skin capillary level, which is ultimately required for the healing of ulcers caused by arterial disease. Nailfold capillary video microscopy has shown that the arteriolar postural vasoconstriction is abolished in patients with CLI.36 It has also demonstrated that the capillary density increases significantly from the supine to the dependent position. These phenomena explain why patients with CLI prefer to have their limbs in the dependent position. As previously mentioned, the application of IPC is effective in reducing the venous pressure, further increasing the limb blood flow. The increase in LDF at the foot supports these findings and implies significant changes at the arteriolar level. The marked increase of blood flow at the systemic, muscular, collateral, and skin levels provide useful evidence for the application of IPC in this cohort of patients when no other options for limb salvage are available. Only one patient in our series had severe pain due to IPC application. Seven of 107 patients in the Montori et al series23 discontinued the device because of pain experienced with its use. These results suggest that this machine is well tolerated by most patients. Clinically, these devices can still be used on those few patients that experience pain by omitting the bladder that compresses the sensitive area. Later, the bladder may be reintroduced into therapy if tolerated. Limitations. We did not analyze the effect of patient position during the test to see how much of the improvement was due to arteriovenous gradient vs vasodilatation. All of our patients were tested in the semi-erect position because they tolerated the application of IPC better and because it has been proven to be optimal for venous emptying and arterial flow enhancement.13,16 Volume flow measurements were performed by using the internal diameter of the arteries, assuming that these vessels were perfect cylinders. However, they were often
JOURNAL OF VASCULAR SURGERY Volume 42, Number 4
irregular, which further increased the inherent errors of this measurement. Since the location of the measurement was kept constant and each vessel as its own control, this should not have a significant effect in the results of the study. No correlation was made with TcPO2 measurements. Even though all of our patients had this test performed, they were not simultaneously done with DU or LDF. These, added to the inherent variability and limited reliability of this exam, were reasons not to include TcPO2 results in the final analysis. It is important to mention that all 20 patients had a patent popliteal artery; therefore, our findings apply only to CLI patients with such an anatomic configuration. Patients with CLI who have this condition may be selectively advantaged compared with those with popliteal artery occlusion. IPC may be beneficial for this subgroup of patients, however, as the collateral and muscular artery flow were shown to improve as well. Finally, no long-term clinical follow-up was available to address the clinical impact of the hemodynamic changes in this subset of patients. As this was an acute-phase experiment, no detailed follow-up was available other than the amputations performed within 6 months. However, several clinical trials have suggested that this technique is associated with an improved rate of limb salvage, wound healing, and improvement in ischemic symptoms. CONCLUSIONS This is the first evidence to show IPC improvements at all levels: axial, muscular, and collateral arteries as well as in skin blood flow in patients with CLI. This treatment is well tolerated and may be beneficial to those who are not candidates for revascularization. Patients with significant venous reflux may not benefit from IPC. This supports the theory that IPC enhances flow by increasing the arteriovenous pressure gradient. Our findings also support the theory that distal vasodilatation is not maximal in areas of ischemia and that further decreases in their vascular resistance can be achieved with IPC. As one of the most skeptical surgeons have acknowledged,31 the time for a multicenter randomized trial has come. REFERENCES 1. Nicolaides AN, Breddin HK, Fareed J, Goldhaber S, Haas S, Hull R, et al; Cardiovascular Disease Educational and Research Trust and the International Union of Angiology. Prevention of venous thromboembolism. International Consensus Statement. Guidelines compiled in accordance with the scientific evidence. Int Angiol 2001;20:1-37. 2. Herrmann LG, Reid MR. The conservative treatment of arteriosclerotic peripheral vascular diseases. Ann Surg 1934;100:750-60. 3. Landis EM, Gibbon JH. The effects of alternative suction and pressure on blood flow to the lower extremities. Ann Intern Med 1934;8: 282-95. 4. Louridas G, Saadia R, Spelay J, Abdoh A, Weighell W, Arneja AS, Tanner J, Guzman R. The ArtAssist Device in chronic lower limb ischemia. A pilot study. Int Angiol 2002;21:28-35. 5. Delis KT, Nicolaides AN, Wolfe JH, Stansby G. Improving walking ability and ankle brachial pressure indices in symptomatic peripheral vascular disease with intermittent pneumatic foot compression: a prospective controlled study with one-year follow-up. J Vasc Surg 2000; 31:650-61.
Labropoulos et al 715
6. Lowe G. Pathophysiology of critical ischaemia. In: Dormandy JA, London SG, editors. Critical leg ischaemia. Berlin: Springer-Verlag, 1990; p. 21-32. 7. DeMey JGR, Uitendaal MP, Boonen HCM, Schiffers PMH, Fazzi G. Structural and functional remodeling of poststenotic arteries in the rat. Adv Exp Med 1991;346:283-90. 8. Ludbrook J. Collateral artery resistance in the human lower limb. J Surg Res 1966;6:423-34. 9. Strandness DE, Sumner DS. Hemodynamics for surgeons. New York: Grune & Stratton; 1975. 10. Morris RJ, Woodcock JP. Effects of supine intermittent compression on arterial inflow to the lower limb. Arch Surg 2002;137:1269-73. 11. Delis KT, Husmann MJ, Nicolaides AN, Wolfe JH, Cheshire NJ. Enhancing foot skin blood flux in peripheral vascular disease using intermittent pneumatic compression: controlled study on claudicants and grafted arteriopaths. World J Surg 2002;26:861-6. 12. Abu-Own A, Cheatle T, Scurr JH, Coleridge Smith PD. Effects of intermittent pneumatic compression of the foot on the microcirculatory function in arterial disease. Eur J Vasc Surg 1993;7:488-92. 13. van Bemmelen PS, Mattos MA, Faught WE, Mansour MA, Barkmeier LD, Hodgson KJ, et al. Augmentation of blood flow in limbs with occlusive arterial disease by intermittent calf compression. J Vasc Surg 1994;19:1052-8. 14. Eze AR, Comerota AJ, Cisek PL, Holland BS, Kerr RP, Veeramasuneni R, et al. Intermittent calf and foot compression increases lower extremity blood flow. Am J Surg 1996;172:130-4. 15. Dormandy JA, Rutherford RB. Management of peripheral arterial disease (PAD). TASC Working Group. TransAtlantic Inter-Society Concensus (TASC). J Vasc Surg 2000;31:S1-S296. 16. Delis KT, Azizi ZA, Stevens RJ, Wolfe JH, Nicolaides AN. Optimum intermittent pneumatic compression stimulus for lower-limb venous emptying. Eur J Vasc Endovasc Surg 2000;19:261-9. 19. Labropoulos N, Wierks C, Suffoletto B. Intermittent pneumatic compression for the treatment of lower extremity arterial disease: a systematic review. Vasc Med 2002;7:141-8. 20. Mehlsen J, Himmelstrup H, Himmelstrup B, Winther K, Trap-Jensen J. Beneficial effects of intermittent suction and pressure treatment in intermittent claudication. Angiology 1993;44:16-20. 21. Kakkos S, Geroulakos G, Nicolaides A. Improvement of the walking ability in intermittent claudication with supervised exercise and pneumatic foot and calf compression: results at six months of a randomized controlled trial. Presented at the 18th Annual Meeting of the European Society for Vascular Surgery, Innsbruck, Austria; September 17-19, 2004. 22. Delis KT, Nicolaides AN. Effect of intermittent pneumatic compression of foot and calf on walking distance, hemodynamics, and quality of life in patients with arterial claudication: a prospective randomized controlled study with 1-year follow-up. Ann Surg 2005;241:431-41. 23. Montori VM, Kavros SJ, Walsh EE, Rooke TW. Intermittent compression pump for nonhealing wounds in patients with limb ischemia. The Mayo Clinic experience (1998-2000). Int Angiol 2002;21:360-6. 24. Vella A, Carlson LA, Blier B, Felty C, Kuiper JD, Rooke TW. Circulator boot therapy alters the natural history of ischemic limb ulceration. Vasc Med 2000;5:21-5. 25. van Bemmelen PS, Gitlitz DB, Faruqi RM, Weiss-Olmanni J, Brunetti VA, Giron F, et al. Limb salvage using high-pressure intermittent compression arterial assist device in cases unsuitable for surgical revascularization. Arch Surg 2001;136:1280-5. 26. van Bemmelen P, Char D, Giron F, Ricotta JJ. Angiographic improvement after rapid intermittent compression treatment (ArtAssist) for small vessel obstruction. Ann Vasc Surg 2003;17:224-8. 27. Chen AH, Frangos SG, Kilaru S, Sumpio BE. Intermittent compression devices—physiologic mechanisms of action. Eur J Vasc Endovasc Surg 2001;21:383-92. 28. Malone MD, Cisek PL, Comerota AJ Jr, Holland B, Eid IG, Comerota AJ. High-pressure, rapid-inflation pneumatic compression improves venous hemodynamics in healthy volunteers and patients who are post-thrombotic. J Vasc Surg 1999;29:593-9. 29. Kakkos SK, Szendro G, Griffin M, Daskalopoulou SS, Nicolaides AN. The efficacy of the new SCD response compression system in the prevention of venous stasis. J Vasc Surg 2000;32:932-40.
716 Labropoulos et al
30. Kakkos SK, Szendro G, Griffin M, Sabetai MM, Nicolaides AN. Improved hemodynamic effectiveness and associated clinical correlations of a new intermittent pneumatic compression system in patients with chronic venous insufficiency. J Vasc Surg 2001;34:915-22. 31. Porter JM. Pneumatic limb compression: a free lunch? J Vasc Surg 2000;31:821-2. 32. Labropoulos N, Watson WC, Mansour MA, Kang SS, Littooy FN, Baker WH. Acute effects of intermittent pneumatic compression on popliteal artery blood flow. Arch Surg 1998;133:1072-5. 33. Delis KT, Nicolaides AN, Wolfe JH. Peripheral sympathetic autoregulation in arterial calf inflow enhancement with intermittent pneumatic compression. Eur J Vasc Endovasc Surg 2001;22:317-25.
JOURNAL OF VASCULAR SURGERY October 2005
34. Delis KT, Nicolaides AN, Labropoulos N, Stansby G. The acute effects of intermittent pneumatic foot versus calf versus simultaneous foot and calf compression on popliteal artery hemodynamics: a comparative study. J Vasc Surg 2000;32:284-92. 35. van Bemmelen PS, Weiss-Olmanni J, Ricotta JJ. Rapid intermittent compression increases skin circulation in chronically ischemic legs with infra-popliteal arterial obstruction. Vasa 2000;29:47-52. 36. Ubbink DT, Jacobs MJ, Slaaf DW, Tangelder GJ, Reneman RS. Capillary recruitment and pain relief on leg dependency in patients with severe lower limb ischemia. Circulation 1992;85:223-9. Submitted Apr 2, 2005; accepted May 31, 2005.
Hemodynamic effects of intermittent pneumatic compression in patients with critical limb ischemia Nicos Labropoulos, PhD, DIC, RVT, Luis R. Leon, Jr, MD, RVT, Ahmad Bhatti, MD, RVT, Steven Melton, MD, Steven S. Kang, MD, Ashraf M. Mansour, MD, and Marc Borge, MD, Maywood, Ill Background: Traditional teaching assumes that the distal arterial tree is maximally dilated in patients with critical limb ischemia (CLI). Endovascular or arterial bypass procedures are the commonly used interventions to increase distal perfusion. However, other forms of treatment such as spinal cord stimulation or intermittent pneumatic compression (IPC) have been shown to improve limb salvage rates. This prospective study was designed to determine if the use of IPC increases popliteal, gastrocnemial, collateral arterial, and skin blood flow in patients with CLI. Methods: Twenty limbs with CLI in 20 patients (mean age, 74 years) were evaluated with duplex ultrasound scans and laser Doppler fluxmetry in the semi-erect position before, during, and after IPC. One pneumatic cuff was applied on the foot and the other on the calf. The maximum inflation pressure was 120 mm Hg and was applied for 3 seconds at three cycles per minute. All patients had at least two-level disease by arteriography. Fourteen limbs were characterized as inoperable, and six were considered marginal for reconstruction. Flow volumes were measured in the popliteal, medial gastrocnemial, and a genicular collateral artery. Skin blood flux was measured on the dorsum of the foot at the same time. Results: Significant flow increase during the application of IPC was found in all three arteries (18/20 limbs) compared with baseline values (P < .02). The highest change was seen in the popliteal, followed by the gastrocnemial and the collateral artery. After the cessation of IPC, the flow returned to baseline. This was attributed to the elevation of time average velocity, as the diameter of the arteries remained unchanged. The skin blood flux increased significantly as well (P < .03). In the two limbs without an increase in the arterial or skin blood flow, significant popliteal vein reflux was found. Both limbs were amputated shortly after. Conclusions: IPC increases axial, muscular, collateral, and skin blood flow in patients with CLI and may be beneficial to those who are not candidates for revascularization. Patients with significant venous reflux may not benefit from IPC. This supports the theory that one of the mechanisms by which IPC enhances flow is by increasing the arteriovenous pressure gradient. ( J Vasc Surg 2005;42:710-6.)
Intermittent pneumatic compression (IPC) is widely used as prophylaxis for deep venous thrombosis (DVT).1 In the early 1930s, IPC was shown to have a positive effect on lower-extremity blood flow of the calf and foot.2,3 Later, clinical benefits in patients with various degrees of limb ischemia were demonstrated, including those with critical limb ischemia (CLI).4,5 CLI occurs when the blood supply to the capillary beds is less than that necessary to maintain tissue viability. Arteriolar vasodilation, resulting in a reduction of peripheral vascular resistance, is a compensatory response to CLI.6 Peripheral arterioles in CLI are relatively insensitive to vasodilator stimuli compared with those in control subjects. Consequently, arterioles in CLI patients are thought to be “maximally dilated,” a phenomenon known as vasomotor paralysis, which has become a traditional assumption.7-9 Significant evidence suggests that IPC increases temporarily the arterial blood flow in healthy subjects, in patients with intermittent claudication, and in those with peripheral arterial disease with prior successful surgical revascularization.3,10-14 However, Duplex ultrasound (DU) or laser From the Department of Surgery, Loyola University Medical Center. Competition of interest: none. Correspondence: Nicos Labropoulos, PhD, RVT, DIC, Director, Vascular Laboratory, University of Medicine & Dentistry of New Jersey, 150 Bergen Street, Room D-447, Newark, NJ 07101-1709 (e-mail: [email protected]; [email protected]). 0741-5214/$30.00 Copyright © 2005 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2005.05.051
710
Doppler blood flow (LDF) evaluation of the lower extremities in patients with CLI has not been yet reported. We postulate that the demonstrated clinical effect of IPC on limbs with CLI is due to further increases in the arterial blood flow at all levels, either because of a reversal of vasomotor paralysis or an increase in the lower-extremity arteriovenous gradient caused by venous emptying from intermittent compression. This prospective study was designed to measure the hemodynamic effects of IPC in patients with CLI in the systemic, muscular, and collateral circulation as well as in the foot skin blood flow. PATIENTS AND METHODS Limbs of patients with CLI were evaluated with DU and LDF in the semi-erect position before, during, and after IPC. CLI was defined according to the TransAtlantic Inter-Society Consensus document15 as those patients whose arterial disease has resulted in a breakdown of the skin (ulcer or gangrene) or pain in the foot at rest. It corresponds to stages III and IV of the Fontaine classification. The semi-erect position was chosen to study these limbs for two reasons: it was the natural position adopted by these patients because of severe discomfort while lying supine, and to increase the arterio-venous gradient.16-18 The ArtAssist device (model AA-1000; ACI Medical Inc., San Marcos, California) was used in all cases. Inflatable 12- and 22-cm bladders in a single cuff structure were
JOURNAL OF VASCULAR SURGERY Volume 42, Number 4
applied on the foot and calf, respectively. The maximum inflation pressure of 120 mm Hg was applied for 3 seconds at three cycles per minute. The foot bladder was inflated first, and the calf bladder, 2 seconds later. This is a rapid inflation device, with a time of 300 to 360 milliseconds to reach peak pressure. All patients underwent digital subtraction arteriography to optimize visualization and vessel conspicuity with maximum contrast of the potential distal target arteries. The images were obtained with the target limb gently but securely immobilized to limit patient motion, which would cause misregistration artifact on the digital subtraction images. Whenever possible, the target area was imaged in an anatomic position (either anteroposterior or lateral) to optimize correct identification of distal target vessels. Nonionic and isosmolar contrast agents were used to minimize patient discomfort. Appropriate collimation, filtering and catheter positioning were used to optimize radiographic exposure in the target area. Whenever possible, contrast was injected from the ipsilateral side to maximize contrast opacification. Long, 5- to 10-second infusion times and delay were used to allow for complete filling of any patent channels in the target extremity. Peripheral vasodilators (tolazoline, nitroglycerine) or hot compress warming were often used to maximize peripheral vasodilatation. The images were reviewed by an interventional radiologist (M. B.) and by two vascular surgeons (A. M., S. S. K.). They were classified as inoperable if no target vessels were identified for revascularization and marginal for reconstruction if the target vessels had a small caliber or diffuse disease that made them not suitable for a durable bypass procedure or intervention. For the purpose of this study, patients selected had a patent popliteal artery. Before each measurement, the subjects were acclimatized in supine position on a couch for at least 10 minutes. Flow volumes were measured in the popliteal, medial gastrocnemius, and a genicular collateral artery. Artery blood flow was measured by using a 4- to 7-MHz linear array transducer (ATL-Ultramark 9; Advanced Tech Laboratory, Bothell, Wash). The internal arterial diameter was measured by imaging the vessel longitudinally with real time B-mode. The average of three repeated measurements was obtained. Velocity measurements were performed at an angle of 60°, with the gate of the sample volume matching the internal diameter of the artery. The mean volume flow was calculated from the time average mean velocity of at least five cycles. At the same time, skin blood flux was measured at the dorsum of the foot by using LDF (Laserflo model BPM 403A, TSI Inc, St Paul, Minn) and recorded in flux units. The BPM 403A has a semiconductor gallium-aluminumarsenide diode that emits a laser beam with a 780-nm wavelength. For all studies, a right angle probe was used. The probe was fixed on the skin by means of a double adhesive ring and adhesive tape. LDF readings at rest, before the initiation of the pump, were stabilized over a 5-minute period. The average resting skin blood flow was
Labropoulos et al 711
Table I. Demographic characteristics of the study population before intervention Mean age (range) Males Females Mean ABI (range) Mean TcPO2 (range) Diabetes mellitus Hypertension Hypercholesterolemia Smoking Coronary artery disease Rest pain Tissue loss History of stroke History of prior bypass
74 (65–83) 13 7 0.25 (0.1–0.29) 21.6 (12–31) 10 15 13 12 11 20 9 2 8
ABI, Ankle-brachial index; TcPO2, transcutaneous oxygen pressure measurements.
calculated over 30 seconds. The skin blood flow during the pump application was not measured because of significant motion artifact and was averaged for 10 seconds immediately after. No attempts were made to correlate the hemodynamic findings with transcutaneous oxygen pressure measurements (TcPO2), as they were not performed simultaneously with the other two methods. The degree of discomfort produced by the application of the pump was graded as none, mild, moderate, and severe. In cases of severe discomfort, the pump was discontinued and the patient was excluded from the study. Statistical analysis. Descriptive statistics were used for patient demographic characteristics. A paired t test was used to compare the arterial flows and diameters before, during, and after the application of IPC. Data were expressed as a mean and 95% confidence intervals (CI). The Wilcoxon sum-rank test was used for the LDF data, as their distribution was nonparametric. These data were expressed as median and interquartile range (IQR). The level for statistical significance was set at P ⫽ .05. RESULTS Twenty limbs with CLI in 20 patients (mean age, 74 years) were evaluated. The demographic characteristics are summarized in Table I. An additional patient was excluded because of inability to tolerate the foot component of the pump due to severe pain. All other patients tolerated it well, with mild-to-moderate discomfort that disappeared after a few pump cycles. All of the patients in our study had rest pain, and nine had tissue loss. All had an ankle-brachial index (ABI) of ⬍0.3, and the mean TcPO2 was 21.6. All patients had at least two-level disease by arteriography. Stenosis or occlusion in the aortoiliac segment was found in 17 patients, common femoral in 6, deep femoral in 6, superficial femoral in 18, and all patients had infrapopliteal disease. Fourteen limbs were inoperable, and six were marginal for reconstruction. Significant flow increase during the application of IPC was found in all three arteries (18/20 limbs) compared
JOURNAL OF VASCULAR SURGERY October 2005
712 Labropoulos et al
Table II. Mean and 95% confidence intervals (CI) of arterial blood flow before, during, and after intermittent pneumatic compression Artery Popliteal Gastrocnemial Collateral
Before During After Before During After Before During After
Mean
95% CI
22 45 20 10 17 9 9 14 8
19.7–24.3 39.4–50.6 17.7–22.3 8.6–11.4 15.1–18.9 7.6–10.4 8.06–9.94 12.6–15.4 6.6–9.4
P ⬍ .01 for comparisons before vs during, and during vs after, in all three arteries. The before and after values were comparable.
with baseline values (P ⬍ .02) (Table II). The highest change was seen in the popliteal (104.5%), followed by the gastrocnemial (70%) and the collateral artery (55.6%). After the cessation of IPC, the flow returned to baseline. This was attributed to the elevation of time average velocity as the diameter of the arteries remained unchanged (Fig 1 and Table III). The skin blood flux increased significantly as well (3.5 flux units, IQR 2 to 4.2 vs 5.2 flux units, IQR 3.8 to 6.2; P ⬍ .03) (Fig 2). In the two limbs without an increase in the arterial or skin blood flow, significant popliteal vein reflux was found (Fig 3). This reflux was of high velocity and ⬎5 seconds in duration in both cases. Both limbs were amputated shortly after. The remaining 18 patients did not have popliteal reflux. The rest of the veins were not examined, but none of them had clinically evident varicose veins or a history of venous thrombosis. Two patients were lost at a 6-month follow-up. Four patients underwent amputations at 10, 14, 30, and 60 days after the experiment.
Fig 1. Color flow duplex imaging in the evaluation of volume blood flow in the popliteal artery. A, At rest, the waveform is abnormal because of multiple levels of disease. B, during intermittent pneumatic compression application, the volume flow almost doubled, with significant increase in both the peak systolic and end-diastolic velocities.
DISCUSSION Several trials on the effects of IPC on the arterial system in the lower limbs have been reported. A meta-analysis of studies between 1966 and 2001 identified several limitations.19 There were differences in the type of devices used (sequential vs simultaneous compression), protocols of cycle length, and duration of treatment, severity of limb ischemia, flow measurement techniques, and end points for analysis. Moreover, only three prospective randomized studies were found, all with small sample sizes. Despite those limitations, a clear trend toward a favorable effect on limb hemodynamics was observed. The first prospective, randomized, controlled evidence for an increase in arterial blood flow with IPC in patients with intermittent claudication was reported in 1993,20 and that was followed by three additional trials.5,21,22 They showed significant beneficial effects on walking distance, symptoms, and systolic blood pressure measurements on the upper and lower limbs with therapy. However, studies on CLI and IPC have been limited to clinical outcomes4,23-25 and case reports.26 The case series are mostly retrospective.23,24
Montori et al23 retrospectively analyzed IPC in 107 patients with arterial ulcers. With a median follow-up of 6 months, wound healing was seen in 40% of patients with TcPO2 levels ⬍20 mm Hg. Vella et al24 retrospectively analyzed IPC in 96 patients with ischemic ulcers and found benefit in their treatment at all TcPO2 levels. Van Bemmelen et al25 studied 14 consecutive ischemic legs in a Veterans Administration hospital. The patients were deemed poor candidates for surgery and were treated with IPC for 4 hours per day at home for 3 months. Clinical outcomes and calibrated pulse-volume amplitudes were recorded. Nine legs were salvaged. The only prospective trial was a pilot study of clinical outcomes involving IPC therapy of 33 limbs with CLI.4 At a mean follow-up of 3 months, 58% of legs were saved. Rest pain improved in 40% of patients, 26% of foot ulcers healed, and toe pressures significantly improved. Several studies have shown increased popliteal artery blood flow during IPC application in healthy subjects and in patients with claudication.13,14,32,34 However, our study
JOURNAL OF VASCULAR SURGERY Volume 42, Number 4
Labropoulos et al 713
Table III. Mean vessel diameters (95% confidence interval; range) in millimeters, before, during, and after intermittent pneumatic compression Artery
Before
During
After
Popliteal Gastrocnemial Collateral
4.02 (3.69–4.31; 3–5.7) 1.91 (1.72–2.08; 1.3–2.9) 1.79 (1.60–1.98; 1.1–2.8)
4.06 (3.77–4.35; 3.2–5.7) 1.95 (1.75–2.14; 1.3–3) 1.82 (1.63–2.01; 1.1–2.8)
3.94 (3.65–4.21; 3–5.5) 1.86 (1.66–2.04; 1.2–3) 1.75 (1.56–1.94; 1.1–2.6)
P ⬎ .1 for comparisons before, during, and after in all three arteries
Fig 2. Laser Doppler flux recording at rest, during, and after the application of intermittent pneumatic compression. During intermittent pneumatic compression, there is a motion artifact with a gradual increase of the flux.
is the first to demonstrate significant improvements in the popliteal artery blood flow in patients with CLI. It is also the first study to demonstrate significant flow increases in muscular and collateral arteries. The previous reports and the current study support arteriovenous pressure gradient augmentation as the main mechanism for the flow increase. This has been shown by the reduction of the venous pressure during IPC application.16 In the sitting position, the venous pressure in the foot is about 60 mm Hg. The application of IPC reduces this pressure and therefore increases the arteriovenous pressure gradient. The resting blood flow in the popliteal artery in patients with CLI is reported for the first time in the literature. Compared with previous publications, the resting blood flow in the sitting position in claudicants is significantly higher than in our patients with CLI.13,34 This can be explained by the multiple levels of the occlusive disease as depicted by the angiograms and the very low ABI values. The relative increase of blood flow volumes in patients with CLI was significant for all three arteries during IPC. However, the magnitude of change was significantly less compared with previous studies that evaluated patients with claudication.13,34 With increasing levels of disease, more vasodilatation is likely present, and therefore, the margin for flow enhancement is less. Vasodilatation is one of the mechanisms by which limbs with non-CLI obtain an increase in their blood flow.27 This is through the production of nitric oxide during mechanical stimulation (device inflation). It has been recently proposed that the application of IPC causes a forward pulsatile flow, resulting in emptying of blood at that site. The increase of blood volume and velocity exerts a compressive strain and enhances the shear stress on the venous endothelial cells.27 IPC has also an effect on the arterial system, as the sudden drop in the venous pressure causes an increase in
the arteriovenous gradient that augments the arterial blood velocity. As a consequence, the shear stress on arterial endothelial cells is increased.27 Cell cultures and animal models have shown that the mechanical forces of shear and strain may elicit vasodilatory, antithrombotic, and profibrinolytic responses in endothelial cells.27 Patients with CLI are assumed to have maximal vessel dilatation caused by the critical lack of oxygen.7-9 Our findings bring the possibility of a submaximal level of dilatation, which could be raised with compression. Such vasodilation should occur in the arteriolar level, as we showed that there is no change in the diameter of the popliteal, gastrocnemial, and collateral arteries during IPC. Other studies have also shown no change in the diameter of the popliteal artery in healthy subjects and in patients with claudication during IPC.22,32,33 This is also substantiated by the increased LDF in our patients, which measures flow in the microcirculation. Further research is required to elucidate this matter. Another question is related to the magnitude of the blood flow increase. We hypothesize that collaterals showed the lowest increase in blood flow because vessel dilatation in these arteries is likely almost maximum. The same theory may be applicable to muscular arteries, smaller in size and probably fully dilated because of muscle ischemia. The popliteal artery had the greatest flow increase. This was possibly because of the additive effect of the diminished peripheral resistance in collateral and muscular branches that the main popliteal artery supplies. It is of interest to observe two occurrences of popliteal vein reflux in patients who ultimately underwent limb amputation shortly after the study. No reports have addressed the causes for failure to achieve beneficial arterial hemodynamic effects with IPC. Malone et al28 demonstrated that patients with post-thrombotic venous disease have a compromised venous hemodynamic response to all IPC devices. Furthermore, they showed an increased velocity response to high-pressure, rapid-inflation IPC machines. But this has only been evaluated in the context of DVT prophylaxis. Van Bemmelen et al25 did not see any hemodynamic improvement in limbs that were amputated after IPC therapy. No comments are made as to the reasons for failure and no documentation of reflux made. Conceivably, significant axial vein reflux will impede a clinically significant arteriovenous gradient, limiting flow augmentation in the arterial system, supporting further this theory as the underlying mechanism for the success of IPC devices. The introduction of novel IPC designs could po-
714 Labropoulos et al
Fig 3. Popliteal artery blood flow in a patient with significant popliteal reflux after a femoropopliteal thrombosis. A, Flow at rest has high resistance due to popliteal vein insufficiency. B, High velocity and long duration of reflux in popliteal vein. C, No change in the volume flow is seen during the application of intermittent pneumatic compression.
tentially improve venous emptying and arterial flow in these subjects.29 Commercially available IPC devices have constant cycles of compression and deflation. It has been shown that vein refilling times vary with limb position and with different venous pathologies.29,30 Relatively new systems have been designed with the ability to detect venous volume changes and to respond by initiating the subse-
JOURNAL OF VASCULAR SURGERY October 2005
quent cycle when the veins are again full. So far, this technology has not been tested from the arterial flow point of view but appears feasible and promising. No studies have been designed to identify the ideal compression sequence to enhance arterial blood flow maximally in the lower limbs of patients with ischemia. However, following the premise that IPC acts through enhancement of venous flow and increasing the arteriovenous gradient, we chose our treatment strategy by selecting the optimal way for venous emptying. This was based on a study that showed that foot and calf compression of 120 to 140 mm Hg with a frequency of 3 to 4 impulses per minute and 1-second delay was the optimum stimulus, as it had the maximal reduction in the venous pressure.16 LDF has been used to assess blood flow at the foot skin level in patients with different levels of limb ischemia.12,35 However, foot skin level assessment of blood flow has never been reported in patients with CLI. Our study is the first evidence of an improvement of blood flow at the skin level in this subgroup, which once more is against the presumed clinical dictum of maximal dilatation. This supports further the theory of a submaximal level of dilatation even at the skin capillary level, which is ultimately required for the healing of ulcers caused by arterial disease. Nailfold capillary video microscopy has shown that the arteriolar postural vasoconstriction is abolished in patients with CLI.36 It has also demonstrated that the capillary density increases significantly from the supine to the dependent position. These phenomena explain why patients with CLI prefer to have their limbs in the dependent position. As previously mentioned, the application of IPC is effective in reducing the venous pressure, further increasing the limb blood flow. The increase in LDF at the foot supports these findings and implies significant changes at the arteriolar level. The marked increase of blood flow at the systemic, muscular, collateral, and skin levels provide useful evidence for the application of IPC in this cohort of patients when no other options for limb salvage are available. Only one patient in our series had severe pain due to IPC application. Seven of 107 patients in the Montori et al series23 discontinued the device because of pain experienced with its use. These results suggest that this machine is well tolerated by most patients. Clinically, these devices can still be used on those few patients that experience pain by omitting the bladder that compresses the sensitive area. Later, the bladder may be reintroduced into therapy if tolerated. Limitations. We did not analyze the effect of patient position during the test to see how much of the improvement was due to arteriovenous gradient vs vasodilatation. All of our patients were tested in the semi-erect position because they tolerated the application of IPC better and because it has been proven to be optimal for venous emptying and arterial flow enhancement.13,16 Volume flow measurements were performed by using the internal diameter of the arteries, assuming that these vessels were perfect cylinders. However, they were often
JOURNAL OF VASCULAR SURGERY Volume 42, Number 4
irregular, which further increased the inherent errors of this measurement. Since the location of the measurement was kept constant and each vessel as its own control, this should not have a significant effect in the results of the study. No correlation was made with TcPO2 measurements. Even though all of our patients had this test performed, they were not simultaneously done with DU or LDF. These, added to the inherent variability and limited reliability of this exam, were reasons not to include TcPO2 results in the final analysis. It is important to mention that all 20 patients had a patent popliteal artery; therefore, our findings apply only to CLI patients with such an anatomic configuration. Patients with CLI who have this condition may be selectively advantaged compared with those with popliteal artery occlusion. IPC may be beneficial for this subgroup of patients, however, as the collateral and muscular artery flow were shown to improve as well. Finally, no long-term clinical follow-up was available to address the clinical impact of the hemodynamic changes in this subset of patients. As this was an acute-phase experiment, no detailed follow-up was available other than the amputations performed within 6 months. However, several clinical trials have suggested that this technique is associated with an improved rate of limb salvage, wound healing, and improvement in ischemic symptoms. CONCLUSIONS This is the first evidence to show IPC improvements at all levels: axial, muscular, and collateral arteries as well as in skin blood flow in patients with CLI. This treatment is well tolerated and may be beneficial to those who are not candidates for revascularization. Patients with significant venous reflux may not benefit from IPC. This supports the theory that IPC enhances flow by increasing the arteriovenous pressure gradient. Our findings also support the theory that distal vasodilatation is not maximal in areas of ischemia and that further decreases in their vascular resistance can be achieved with IPC. As one of the most skeptical surgeons have acknowledged,31 the time for a multicenter randomized trial has come. REFERENCES 1. Nicolaides AN, Breddin HK, Fareed J, Goldhaber S, Haas S, Hull R, et al; Cardiovascular Disease Educational and Research Trust and the International Union of Angiology. Prevention of venous thromboembolism. International Consensus Statement. Guidelines compiled in accordance with the scientific evidence. Int Angiol 2001;20:1-37. 2. Herrmann LG, Reid MR. The conservative treatment of arteriosclerotic peripheral vascular diseases. Ann Surg 1934;100:750-60. 3. Landis EM, Gibbon JH. The effects of alternative suction and pressure on blood flow to the lower extremities. Ann Intern Med 1934;8: 282-95. 4. Louridas G, Saadia R, Spelay J, Abdoh A, Weighell W, Arneja AS, Tanner J, Guzman R. The ArtAssist Device in chronic lower limb ischemia. A pilot study. Int Angiol 2002;21:28-35. 5. Delis KT, Nicolaides AN, Wolfe JH, Stansby G. Improving walking ability and ankle brachial pressure indices in symptomatic peripheral vascular disease with intermittent pneumatic foot compression: a prospective controlled study with one-year follow-up. J Vasc Surg 2000; 31:650-61.
Labropoulos et al 715
6. Lowe G. Pathophysiology of critical ischaemia. In: Dormandy JA, London SG, editors. Critical leg ischaemia. Berlin: Springer-Verlag, 1990; p. 21-32. 7. DeMey JGR, Uitendaal MP, Boonen HCM, Schiffers PMH, Fazzi G. Structural and functional remodeling of poststenotic arteries in the rat. Adv Exp Med 1991;346:283-90. 8. Ludbrook J. Collateral artery resistance in the human lower limb. J Surg Res 1966;6:423-34. 9. Strandness DE, Sumner DS. Hemodynamics for surgeons. New York: Grune & Stratton; 1975. 10. Morris RJ, Woodcock JP. Effects of supine intermittent compression on arterial inflow to the lower limb. Arch Surg 2002;137:1269-73. 11. Delis KT, Husmann MJ, Nicolaides AN, Wolfe JH, Cheshire NJ. Enhancing foot skin blood flux in peripheral vascular disease using intermittent pneumatic compression: controlled study on claudicants and grafted arteriopaths. World J Surg 2002;26:861-6. 12. Abu-Own A, Cheatle T, Scurr JH, Coleridge Smith PD. Effects of intermittent pneumatic compression of the foot on the microcirculatory function in arterial disease. Eur J Vasc Surg 1993;7:488-92. 13. van Bemmelen PS, Mattos MA, Faught WE, Mansour MA, Barkmeier LD, Hodgson KJ, et al. Augmentation of blood flow in limbs with occlusive arterial disease by intermittent calf compression. J Vasc Surg 1994;19:1052-8. 14. Eze AR, Comerota AJ, Cisek PL, Holland BS, Kerr RP, Veeramasuneni R, et al. Intermittent calf and foot compression increases lower extremity blood flow. Am J Surg 1996;172:130-4. 15. Dormandy JA, Rutherford RB. Management of peripheral arterial disease (PAD). TASC Working Group. TransAtlantic Inter-Society Concensus (TASC). J Vasc Surg 2000;31:S1-S296. 16. Delis KT, Azizi ZA, Stevens RJ, Wolfe JH, Nicolaides AN. Optimum intermittent pneumatic compression stimulus for lower-limb venous emptying. Eur J Vasc Endovasc Surg 2000;19:261-9. 19. Labropoulos N, Wierks C, Suffoletto B. Intermittent pneumatic compression for the treatment of lower extremity arterial disease: a systematic review. Vasc Med 2002;7:141-8. 20. Mehlsen J, Himmelstrup H, Himmelstrup B, Winther K, Trap-Jensen J. Beneficial effects of intermittent suction and pressure treatment in intermittent claudication. Angiology 1993;44:16-20. 21. Kakkos S, Geroulakos G, Nicolaides A. Improvement of the walking ability in intermittent claudication with supervised exercise and pneumatic foot and calf compression: results at six months of a randomized controlled trial. Presented at the 18th Annual Meeting of the European Society for Vascular Surgery, Innsbruck, Austria; September 17-19, 2004. 22. Delis KT, Nicolaides AN. Effect of intermittent pneumatic compression of foot and calf on walking distance, hemodynamics, and quality of life in patients with arterial claudication: a prospective randomized controlled study with 1-year follow-up. Ann Surg 2005;241:431-41. 23. Montori VM, Kavros SJ, Walsh EE, Rooke TW. Intermittent compression pump for nonhealing wounds in patients with limb ischemia. The Mayo Clinic experience (1998-2000). Int Angiol 2002;21:360-6. 24. Vella A, Carlson LA, Blier B, Felty C, Kuiper JD, Rooke TW. Circulator boot therapy alters the natural history of ischemic limb ulceration. Vasc Med 2000;5:21-5. 25. van Bemmelen PS, Gitlitz DB, Faruqi RM, Weiss-Olmanni J, Brunetti VA, Giron F, et al. Limb salvage using high-pressure intermittent compression arterial assist device in cases unsuitable for surgical revascularization. Arch Surg 2001;136:1280-5. 26. van Bemmelen P, Char D, Giron F, Ricotta JJ. Angiographic improvement after rapid intermittent compression treatment (ArtAssist) for small vessel obstruction. Ann Vasc Surg 2003;17:224-8. 27. Chen AH, Frangos SG, Kilaru S, Sumpio BE. Intermittent compression devices—physiologic mechanisms of action. Eur J Vasc Endovasc Surg 2001;21:383-92. 28. Malone MD, Cisek PL, Comerota AJ Jr, Holland B, Eid IG, Comerota AJ. High-pressure, rapid-inflation pneumatic compression improves venous hemodynamics in healthy volunteers and patients who are post-thrombotic. J Vasc Surg 1999;29:593-9. 29. Kakkos SK, Szendro G, Griffin M, Daskalopoulou SS, Nicolaides AN. The efficacy of the new SCD response compression system in the prevention of venous stasis. J Vasc Surg 2000;32:932-40.
716 Labropoulos et al
30. Kakkos SK, Szendro G, Griffin M, Sabetai MM, Nicolaides AN. Improved hemodynamic effectiveness and associated clinical correlations of a new intermittent pneumatic compression system in patients with chronic venous insufficiency. J Vasc Surg 2001;34:915-22. 31. Porter JM. Pneumatic limb compression: a free lunch? J Vasc Surg 2000;31:821-2. 32. Labropoulos N, Watson WC, Mansour MA, Kang SS, Littooy FN, Baker WH. Acute effects of intermittent pneumatic compression on popliteal artery blood flow. Arch Surg 1998;133:1072-5. 33. Delis KT, Nicolaides AN, Wolfe JH. Peripheral sympathetic autoregulation in arterial calf inflow enhancement with intermittent pneumatic compression. Eur J Vasc Endovasc Surg 2001;22:317-25.
JOURNAL OF VASCULAR SURGERY October 2005
34. Delis KT, Nicolaides AN, Labropoulos N, Stansby G. The acute effects of intermittent pneumatic foot versus calf versus simultaneous foot and calf compression on popliteal artery hemodynamics: a comparative study. J Vasc Surg 2000;32:284-92. 35. van Bemmelen PS, Weiss-Olmanni J, Ricotta JJ. Rapid intermittent compression increases skin circulation in chronically ischemic legs with infra-popliteal arterial obstruction. Vasa 2000;29:47-52. 36. Ubbink DT, Jacobs MJ, Slaaf DW, Tangelder GJ, Reneman RS. Capillary recruitment and pain relief on leg dependency in patients with severe lower limb ischemia. Circulation 1992;85:223-9. Submitted Apr 2, 2005; accepted May 31, 2005.