XML Template (2014) [9.4.2014–6:24am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/PHLJ/Vol00000/140061/APPFile/SG-PHLJ140061.3d
(PHL)
[1–8] [PREPRINTER stage]
Phlebology OnlineFirst, published on May 2, 2014 as doi:10.1177/0268355514531255
Original Article
Haemodynamic changes with the use of neuromuscular electrical stimulation compared to intermittent pneumatic compression
Phlebology 0(0) 1–8 ! The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0268355514531255 phl.sagepub.com
KJ Williams, HM Moore and AH Davies
Abstract Introduction: Enhancement of peripheral circulation has been shown to be of benefit in many vascular disorders, and the clinical effectiveness of intermittent pneumatic compression is well established in peripheral vascular disease. This study compares the haemodynamic efficacy of a novel neuromuscular electrical stimulation device with intermittent pneumatic compression in healthy subjects. Methods: Ten healthy volunteers (mean age 27.1 3.8 years, body mass index 24.8 3.6 kg/m2) were randomised into two groups, in an interventional crossover trial. Devices used were the SCD ExpressTM Compression System, (Covidien, Ireland) and the gekoTM, (Firstkind Ltd, UK). Devices were applied bilaterally, and haemodynamic measurements taken from the left leg. Changes to haemodynamic parameters (superficial femory artery and femoral vein) and laser Doppler measurements from the hand and foot were compared. Results: Intermittent pneumatic compression caused 51% (p ¼ 0.002), 5% (ns) and 3% (ns) median increases in venous peak velocity, time-averaged maximum velocity and volume flow, respectively; neuromuscular electrical stimulator stimulation caused a 103%, 101% and 101% median increases in the same parameters (all p ¼ 0.002). The benefit was lost upon deactivation. Intermittent pneumatic compression did not improve arterial haemodynamics. Neuromuscular electrical stimulator caused 11%, 84% and 75% increase in arterial parameters (p < 0.01). Laser Doppler readings taken from the leg were increased by neuromuscular electrical stimulator (p < 0.001), dropping after deactivation. For intermittent pneumatic compression, the readings decreased during use but increased after cessation. Hand flux signal dropped during activation of both devices, rising after cessation. Discussion: The neuromuscular electrical stimulator device used in this study enhances venous flow and peak velocity in the legs of healthy subjects and is equal or superior to intermittent pneumatic compression. This warrants further clinical and economic evaluation for deep venous thrombosis prophylaxis and exploration of the haemodynamic effect in venous pathology. It also enhances arterial time-averaged maximum velocity and flow rate, which may prove to be of clinical use in the management of peripheral arterial disease. The effect on the microcirculation as evidenced by laser Doppler fluximetry may reflect a clinically beneficial target in microvascular disease, such as in the diabetic foot.
Keywords Intermittent pneumatic compression, neuromuscular, electrical, electrical stimulation, neuromuscular electrical stimulator
Introduction Intermittent pneumatic compression (IPC) has established itself as a potentially useful tool for improving peripheral haemodynamic forces acting on veins, arteries and lymphatics. It is an effective treatment for a variety of circulatory disorders. There is strong clinical evidence demonstrating the efficiency of IPC in the treatment of lymphoedema,1 prophylaxis against deep vein thrombosis,2 enhancing the healing of venous
Academic Section of Vascular Surgery, Imperial College London, London, UK Corresponding author: AH Davies, Section of Surgery, Imperial College London, Charing Cross Hospital, Fulham Palace Road, London, W6 8RF, UK. Email: [email protected]
Downloaded from phl.sagepub.com at University of Manitoba Libraries on June 4, 2015
XML Template (2014) [9.4.2014–6:24am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/PHLJ/Vol00000/140061/APPFile/SG-PHLJ140061.3d
(PHL)
[1–8] [PREPRINTER stage]
2
Phlebology 0(0)
ulcers3 and improving arterial ischemia4; however, the mechanisms of action of these devices are not yet fully understood. Reports have consistently shown that IPC devices enhance venous flow volume and velocity in both the femoral and popliteal veins.5 Significant increases in the popliteal artery flow were also reported during the application of IPC in patients with critical limb ischaemia.6 The haemodynamic changes accompanying IPC use appear to correlate with its beneficial effects.6,7 The limitations of IPC use include improper fitting, inappropriate use of device, peroneal nerve injury, discomfort, and excessive heat and sweating under the inflatable cuffs.8 The size, weight and external power source requirements limit mobility and contribute to poor compliance, which restricts the efficacy of IPC devices.9 Electrical calf muscle pump stimulation has been shown to significantly reduce the incidence of perioperative and postoperative deep venous thrombosis (DVT).10 However, the pain associated with high currents used for stimulation of the muscle limited its practical use. Recent innovation has seen the development of a neuromuscular electrical stimulator (NMES), whereby a smaller energy delivery is needed to initiate an action potential in a subcutaneous motor nerve (such as the common peroneal nerve). Intermittent stimulation results in activation of the muscle pumps of the leg via the anterior and lateral compartment muscles. A previous haemodynamic study has shown significant increases in blood volume flow and velocity and skin capillary blood flow using the same device.11 It is battery operated, disposable and does not tether the wearer to a bedside unit. The NMES device has never been compared to an established method like IPC. We hypothesise that stimulation of calf muscle contraction by neuromuscular electric stimulation (NMES) may achieve similar haemodynamic effects as compared to IPC. The hypothesis of the study is that NMES via the common peroneal nerve will produce an improved haemodynamic effect equal to or greater than application of IPC.
Method This study has ethical approval from the National Research Ethics Committee (Brent), reference 11/LO/ 1292. It is registered with www.clinicaltrials.gov under reference NCT01939288. Ten healthy volunteers were recruited and screened for evidence of cardiovascular disease. Inclusion criteria were age over 18 years, body mass index (BMI) between 17 and 30 kg/m2, and the ability to give informed consent. Exclusion criteria included a history
of heart disease or respiratory disorder, current pregnancy, history of peripheral vascular disease, varicose vein surgery or thromboembolic event, cardiac pacemaker, history of leg fractures or metal implants in the leg, and long distance travel within one week prior to the study. An ankle-brachial pressure index (ABPI) was taken on all subjects, and subjects were excluded from the study if it was below 0.8. A urinary pregnancy test was taken in women of childbearing age.
Devices IPC device: The IPC device (SCD ExpressTM Compression System, Covidien, Ireland) consists of a controller, non-disposable tubing and medium size knee-length leg garments. The adjustable compression sleeve contains three bladders to deliver circumferential, sequential and gradient compression. The compression cycle delivers 11 s of 45 mmHg pressure at the ankle and 40 mmHg at the calf.12 NMES device: The gekoTM T-1 device (Firstkind Ltd, UK) is a disposable, portable, internally powered, neuromuscular stimulation device (27 mA, 1 Hz, pulse width 70–560 ms). The device is fitted to the skin of the inferolateral aspect of the knee joint. The device stimulation level is set to the minimum level that can achieve upward and outward twitching of the foot when raised from the ground.11
Ultrasound duplex scanning The Phillips iU22 xMATRIX ultrasound machine and L12-5 transducer were used in all ultrasound procedures using a pre-set optimised vascular protocol. Venous compression ultrasonography of the left leg on the day of study confirmed the absence of DVT and significant venous reflux (>0.5 s). Subjects were positioned semi-recumbent on an examination couch, and allowed to adjust position for comfort during the scanning protocol. All ultrasound measurements were taken from the left leg. The left superficial femoral artery and superficial femoral vein were identified and the position of the probe was marked on the skin for consistency of repeat measurements with the leg externally rotated at the hip to 35–40 and flexed at the knee according to subject comfort. An equilibration period of 10 min was observed before the first protocol scans were taken; during this time, the subject was allowed to relax with eyes closed. Real time gated Doppler superimposed on B-mode imaging was used in the evaluation of flow and processed using automated waveformenveloping duplex software, as previously described.13 Five repeated measurements of time-averaged maximum velocity (TAMV) and flow rate were taken,
Downloaded from phl.sagepub.com at University of Manitoba Libraries on June 4, 2015
XML Template (2014) [9.4.2014–6:24am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/PHLJ/Vol00000/140061/APPFile/SG-PHLJ140061.3d
(PHL)
[1–8] [PREPRINTER stage]
Williams et al.
3
recorded via an electronic screen shot, and an average of the five measurements was made later. Peak velocity (PV) of the greatest waveform amplitude of each 15s screenshot was analysed at a later date using calibrated ImageJ software (National Institute of Health, USA, version 1.46 R). As the activation cycle of the IPC device is greater than the maximum screenshot time available, it is not possible for the equipment to conduct a single measurement and single calculation. Therefore, a compound equation using two measurements has been used Total volume flowðcc=minÞ tAðsÞ tAðsÞ ¼ Aðcc=minÞ þ 1 Bðcc=minÞ tBðsÞ tBðsÞ where tA and A, respectively, are the time and volume flow during the inflation part of the cycle, and tB and B are the time and volume flow during deflation and rest (Figure 1). The NMES device operates at 60 Hz; therefore, the data capture during 15 s was adequate for direct measurement.
Protocol Patients were randomised by a closed envelope method into two groups to create a crossover interventional trial. Group A had device 1 applied to both legs and
activated first, followed by removal and application of device 2 (Figure 2). In group B, the order was reversed. Devices were applied according to manufacturer’s instruction. After an equilibration rest period of 10 min, baseline haemodynamic measurements were recorded (arrows) and the first device was applied and activated for a 20 min time period before repeat measurements were taken with the device on, off and 10 min postcessation of device. The subject was allowed to rest for a 20 min wash-out period. The second device was activated and, repeat haemodynamic measurements were taken. Tolerability of the device was rated by each user at the end of the protocol, using a verbal reported score (0–5; no pain, to that experienced by inflating a blood pressure cuff to 200 mmHg).14
Reliability In order to determine the accuracy of measurements per subject (intra-subject reliability), analysis was undertaken using one healthy subject. Fifteen Doppler readings each of 15 s were taken on the left leg superficial femoral artery and femoral vein. Data were analysed using Prism5 (GraphPad Software Inc) to derive coefficient of variation, and Bland–Altman analysis was performed on paired consecutive measurements (significance was taken at 0.05, see Figure 3). The same vascular scientist performed all haemodynamic scanning. The coefficient of variation for venous PV, TAMV and flow rate, respectively, was 18.9%, 29.5% and 24.7% and for arterial, 1.0%, 17.8% and 23.7% (n ¼ 15). Due to the variability of results, it was decided to average five consecutive measurements of each variable.
Statistical analysis
Figure 1. Diagrammatic representation of intermittent pneumatic compression cycle for measurement equation.
The results were analysed using Prism5 (GraphPad Software Inc), analysed for normal distribution, and then treated as parametric or non-parametric data as appropriate with significance taken at 0.05.
Figure 2. Study protocol, ultrasound measurements shown with arrows.
Downloaded from phl.sagepub.com at University of Manitoba Libraries on June 4, 2015
XML Template (2014) [9.4.2014–6:24am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/PHLJ/Vol00000/140061/APPFile/SG-PHLJ140061.3d
(PHL)
[1–8] [PREPRINTER stage]
4
Phlebology 0(0)
Figure 3. Bland–Altman diagrams of venous and arterial haemodynamic measurements (bias shown as solid line, 95% confidence interval shown with dotted lines).
Results Demographics Ten healthy recruits were screened and enrolled onto the study (CONSORT diagram shown in Figure 4). There were four males and six females. Their mean (standard deviation) age was 27.1 (3.8) years and BMI 24.8 (3.6) kg/m2. ABPIs all lay between 0.9 and 1.1. Subjects denied any co-morbidities. One female had had two laparoscopies for endometriosis, and two females were on the oral contraceptive pill at the time of trial. All participants completed the study in full. Figure 4. CONSORT diagram.
Venous Both devices caused haemodynamic improvement during activation (Table 1 and Figure 5). IPC caused 51% (p ¼ 0.02, Wilcoxon signed-rank), 5% (ns) and 3% (ns) median percentage increases in PV, TAMV and flow rate, respectively; whilst NMES stimulation caused 103% (p ¼ 0.002), 101% (p ¼ 0.002) and 101% (p ¼ 0.002) increases in the same parameters. The benefit was quickly lost upon deactivation in both cases.
Laser Doppler reading changes taken from the leg were not statistically increased by IPC but were significantly raised by NMES (Table 2, Figure 6). During operation of both individual devices, there was a drop in signal from the ipsilateral hand, which returned to positive values after device cessation.
Arterial
Tolerability
IPC caused a median 8%, 12% and 13% changes in PV, TAMV and flow rates, respectively, which were not statistically significant (Wilcoxon signed-rank). NMES caused 11% (p ¼ 0.002), 84% (p ¼ 0.004) and 75% (p ¼ 0.006) increases in the same parameters.
All patients tolerated both devices and completed the trial. On a verbal reported score, from 0 (painless) to 5 (as painful as a sphygmomanometer cuff inflated to 200 mmHg), IPC was rated a mean of 1.5, and NMES as 2.8 (p ¼ 0.006, paired Student’s t-test).
Downloaded from phl.sagepub.com at University of Manitoba Libraries on June 4, 2015
XML Template (2014) [9.4.2014–6:24am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/PHLJ/Vol00000/140061/APPFile/SG-PHLJ140061.3d
(PHL)
[1–8] [PREPRINTER stage]
Williams et al.
5
Table 1. Overall haemodynamic changes from baseline for both devices.
Venous
Arterial
Change from baseline
Median peak velocity, cm/s (interquartile range)
Median TAMV, cm/s (interquartile range)
Median flow rate, cc/min (interquartile range)
IPC on IPC off 10-min post NMES on NMES off 10-min post IPC on IPC off 10-min post NMES on NMES off 10-min post
6.2* 0.6 2.3 36.7** 1.4 1.0 5.2 1.9 0.8 6.8** 5.1 2.9
0.2 0.6 0.1 2.1** 0.0 0.3 0.6 0.3 0.9 4.2** 0.6 0.9
4.2 23.4 0.4 107.6* 8.7 9.2 20.1 3.9 10.8 90.0** 8.0 24.1
IPC
NMES
IPC
NMES
(0.1–16.5) (3.8–0.0) (4.3–0.7) (18.8–49.1) (3.8–2.6) (1.8–4.3) (8.5–2.2) (11.3–0.4) (12.5–4.3) (3.59.7) (8.0–0.2) (6.2–6.2)
(0.2–1.2) (1.6–0.4) (1.6–0.8) (1.2–3.3) (1.0–1.0) (0.1–0.7) (2.0–1.9) (1.8–0.1) (2.5–1.0) (2.9–7.2) (2.2–1.6) (2.3–1.6)
(6.7–48.9) (108–11.0) (91.7–47.8) (13.5–218.5) (85.2–26.7) (21.0–58.7) (37.4–82.1) (45.7–23.1) (73.8–53.2) (64.6–181.7) (51.6–32.4) (62.8–53.7)
Significance tested using Wilcoxon signed-rank, *p
(PHL)
[1–8] [PREPRINTER stage]
Phlebology OnlineFirst, published on May 2, 2014 as doi:10.1177/0268355514531255
Original Article
Haemodynamic changes with the use of neuromuscular electrical stimulation compared to intermittent pneumatic compression
Phlebology 0(0) 1–8 ! The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0268355514531255 phl.sagepub.com
KJ Williams, HM Moore and AH Davies
Abstract Introduction: Enhancement of peripheral circulation has been shown to be of benefit in many vascular disorders, and the clinical effectiveness of intermittent pneumatic compression is well established in peripheral vascular disease. This study compares the haemodynamic efficacy of a novel neuromuscular electrical stimulation device with intermittent pneumatic compression in healthy subjects. Methods: Ten healthy volunteers (mean age 27.1 3.8 years, body mass index 24.8 3.6 kg/m2) were randomised into two groups, in an interventional crossover trial. Devices used were the SCD ExpressTM Compression System, (Covidien, Ireland) and the gekoTM, (Firstkind Ltd, UK). Devices were applied bilaterally, and haemodynamic measurements taken from the left leg. Changes to haemodynamic parameters (superficial femory artery and femoral vein) and laser Doppler measurements from the hand and foot were compared. Results: Intermittent pneumatic compression caused 51% (p ¼ 0.002), 5% (ns) and 3% (ns) median increases in venous peak velocity, time-averaged maximum velocity and volume flow, respectively; neuromuscular electrical stimulator stimulation caused a 103%, 101% and 101% median increases in the same parameters (all p ¼ 0.002). The benefit was lost upon deactivation. Intermittent pneumatic compression did not improve arterial haemodynamics. Neuromuscular electrical stimulator caused 11%, 84% and 75% increase in arterial parameters (p < 0.01). Laser Doppler readings taken from the leg were increased by neuromuscular electrical stimulator (p < 0.001), dropping after deactivation. For intermittent pneumatic compression, the readings decreased during use but increased after cessation. Hand flux signal dropped during activation of both devices, rising after cessation. Discussion: The neuromuscular electrical stimulator device used in this study enhances venous flow and peak velocity in the legs of healthy subjects and is equal or superior to intermittent pneumatic compression. This warrants further clinical and economic evaluation for deep venous thrombosis prophylaxis and exploration of the haemodynamic effect in venous pathology. It also enhances arterial time-averaged maximum velocity and flow rate, which may prove to be of clinical use in the management of peripheral arterial disease. The effect on the microcirculation as evidenced by laser Doppler fluximetry may reflect a clinically beneficial target in microvascular disease, such as in the diabetic foot.
Keywords Intermittent pneumatic compression, neuromuscular, electrical, electrical stimulation, neuromuscular electrical stimulator
Introduction Intermittent pneumatic compression (IPC) has established itself as a potentially useful tool for improving peripheral haemodynamic forces acting on veins, arteries and lymphatics. It is an effective treatment for a variety of circulatory disorders. There is strong clinical evidence demonstrating the efficiency of IPC in the treatment of lymphoedema,1 prophylaxis against deep vein thrombosis,2 enhancing the healing of venous
Academic Section of Vascular Surgery, Imperial College London, London, UK Corresponding author: AH Davies, Section of Surgery, Imperial College London, Charing Cross Hospital, Fulham Palace Road, London, W6 8RF, UK. Email: [email protected]
Downloaded from phl.sagepub.com at University of Manitoba Libraries on June 4, 2015
XML Template (2014) [9.4.2014–6:24am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/PHLJ/Vol00000/140061/APPFile/SG-PHLJ140061.3d
(PHL)
[1–8] [PREPRINTER stage]
2
Phlebology 0(0)
ulcers3 and improving arterial ischemia4; however, the mechanisms of action of these devices are not yet fully understood. Reports have consistently shown that IPC devices enhance venous flow volume and velocity in both the femoral and popliteal veins.5 Significant increases in the popliteal artery flow were also reported during the application of IPC in patients with critical limb ischaemia.6 The haemodynamic changes accompanying IPC use appear to correlate with its beneficial effects.6,7 The limitations of IPC use include improper fitting, inappropriate use of device, peroneal nerve injury, discomfort, and excessive heat and sweating under the inflatable cuffs.8 The size, weight and external power source requirements limit mobility and contribute to poor compliance, which restricts the efficacy of IPC devices.9 Electrical calf muscle pump stimulation has been shown to significantly reduce the incidence of perioperative and postoperative deep venous thrombosis (DVT).10 However, the pain associated with high currents used for stimulation of the muscle limited its practical use. Recent innovation has seen the development of a neuromuscular electrical stimulator (NMES), whereby a smaller energy delivery is needed to initiate an action potential in a subcutaneous motor nerve (such as the common peroneal nerve). Intermittent stimulation results in activation of the muscle pumps of the leg via the anterior and lateral compartment muscles. A previous haemodynamic study has shown significant increases in blood volume flow and velocity and skin capillary blood flow using the same device.11 It is battery operated, disposable and does not tether the wearer to a bedside unit. The NMES device has never been compared to an established method like IPC. We hypothesise that stimulation of calf muscle contraction by neuromuscular electric stimulation (NMES) may achieve similar haemodynamic effects as compared to IPC. The hypothesis of the study is that NMES via the common peroneal nerve will produce an improved haemodynamic effect equal to or greater than application of IPC.
Method This study has ethical approval from the National Research Ethics Committee (Brent), reference 11/LO/ 1292. It is registered with www.clinicaltrials.gov under reference NCT01939288. Ten healthy volunteers were recruited and screened for evidence of cardiovascular disease. Inclusion criteria were age over 18 years, body mass index (BMI) between 17 and 30 kg/m2, and the ability to give informed consent. Exclusion criteria included a history
of heart disease or respiratory disorder, current pregnancy, history of peripheral vascular disease, varicose vein surgery or thromboembolic event, cardiac pacemaker, history of leg fractures or metal implants in the leg, and long distance travel within one week prior to the study. An ankle-brachial pressure index (ABPI) was taken on all subjects, and subjects were excluded from the study if it was below 0.8. A urinary pregnancy test was taken in women of childbearing age.
Devices IPC device: The IPC device (SCD ExpressTM Compression System, Covidien, Ireland) consists of a controller, non-disposable tubing and medium size knee-length leg garments. The adjustable compression sleeve contains three bladders to deliver circumferential, sequential and gradient compression. The compression cycle delivers 11 s of 45 mmHg pressure at the ankle and 40 mmHg at the calf.12 NMES device: The gekoTM T-1 device (Firstkind Ltd, UK) is a disposable, portable, internally powered, neuromuscular stimulation device (27 mA, 1 Hz, pulse width 70–560 ms). The device is fitted to the skin of the inferolateral aspect of the knee joint. The device stimulation level is set to the minimum level that can achieve upward and outward twitching of the foot when raised from the ground.11
Ultrasound duplex scanning The Phillips iU22 xMATRIX ultrasound machine and L12-5 transducer were used in all ultrasound procedures using a pre-set optimised vascular protocol. Venous compression ultrasonography of the left leg on the day of study confirmed the absence of DVT and significant venous reflux (>0.5 s). Subjects were positioned semi-recumbent on an examination couch, and allowed to adjust position for comfort during the scanning protocol. All ultrasound measurements were taken from the left leg. The left superficial femoral artery and superficial femoral vein were identified and the position of the probe was marked on the skin for consistency of repeat measurements with the leg externally rotated at the hip to 35–40 and flexed at the knee according to subject comfort. An equilibration period of 10 min was observed before the first protocol scans were taken; during this time, the subject was allowed to relax with eyes closed. Real time gated Doppler superimposed on B-mode imaging was used in the evaluation of flow and processed using automated waveformenveloping duplex software, as previously described.13 Five repeated measurements of time-averaged maximum velocity (TAMV) and flow rate were taken,
Downloaded from phl.sagepub.com at University of Manitoba Libraries on June 4, 2015
XML Template (2014) [9.4.2014–6:24am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/PHLJ/Vol00000/140061/APPFile/SG-PHLJ140061.3d
(PHL)
[1–8] [PREPRINTER stage]
Williams et al.
3
recorded via an electronic screen shot, and an average of the five measurements was made later. Peak velocity (PV) of the greatest waveform amplitude of each 15s screenshot was analysed at a later date using calibrated ImageJ software (National Institute of Health, USA, version 1.46 R). As the activation cycle of the IPC device is greater than the maximum screenshot time available, it is not possible for the equipment to conduct a single measurement and single calculation. Therefore, a compound equation using two measurements has been used Total volume flowðcc=minÞ tAðsÞ tAðsÞ ¼ Aðcc=minÞ þ 1 Bðcc=minÞ tBðsÞ tBðsÞ where tA and A, respectively, are the time and volume flow during the inflation part of the cycle, and tB and B are the time and volume flow during deflation and rest (Figure 1). The NMES device operates at 60 Hz; therefore, the data capture during 15 s was adequate for direct measurement.
Protocol Patients were randomised by a closed envelope method into two groups to create a crossover interventional trial. Group A had device 1 applied to both legs and
activated first, followed by removal and application of device 2 (Figure 2). In group B, the order was reversed. Devices were applied according to manufacturer’s instruction. After an equilibration rest period of 10 min, baseline haemodynamic measurements were recorded (arrows) and the first device was applied and activated for a 20 min time period before repeat measurements were taken with the device on, off and 10 min postcessation of device. The subject was allowed to rest for a 20 min wash-out period. The second device was activated and, repeat haemodynamic measurements were taken. Tolerability of the device was rated by each user at the end of the protocol, using a verbal reported score (0–5; no pain, to that experienced by inflating a blood pressure cuff to 200 mmHg).14
Reliability In order to determine the accuracy of measurements per subject (intra-subject reliability), analysis was undertaken using one healthy subject. Fifteen Doppler readings each of 15 s were taken on the left leg superficial femoral artery and femoral vein. Data were analysed using Prism5 (GraphPad Software Inc) to derive coefficient of variation, and Bland–Altman analysis was performed on paired consecutive measurements (significance was taken at 0.05, see Figure 3). The same vascular scientist performed all haemodynamic scanning. The coefficient of variation for venous PV, TAMV and flow rate, respectively, was 18.9%, 29.5% and 24.7% and for arterial, 1.0%, 17.8% and 23.7% (n ¼ 15). Due to the variability of results, it was decided to average five consecutive measurements of each variable.
Statistical analysis
Figure 1. Diagrammatic representation of intermittent pneumatic compression cycle for measurement equation.
The results were analysed using Prism5 (GraphPad Software Inc), analysed for normal distribution, and then treated as parametric or non-parametric data as appropriate with significance taken at 0.05.
Figure 2. Study protocol, ultrasound measurements shown with arrows.
Downloaded from phl.sagepub.com at University of Manitoba Libraries on June 4, 2015
XML Template (2014) [9.4.2014–6:24am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/PHLJ/Vol00000/140061/APPFile/SG-PHLJ140061.3d
(PHL)
[1–8] [PREPRINTER stage]
4
Phlebology 0(0)
Figure 3. Bland–Altman diagrams of venous and arterial haemodynamic measurements (bias shown as solid line, 95% confidence interval shown with dotted lines).
Results Demographics Ten healthy recruits were screened and enrolled onto the study (CONSORT diagram shown in Figure 4). There were four males and six females. Their mean (standard deviation) age was 27.1 (3.8) years and BMI 24.8 (3.6) kg/m2. ABPIs all lay between 0.9 and 1.1. Subjects denied any co-morbidities. One female had had two laparoscopies for endometriosis, and two females were on the oral contraceptive pill at the time of trial. All participants completed the study in full. Figure 4. CONSORT diagram.
Venous Both devices caused haemodynamic improvement during activation (Table 1 and Figure 5). IPC caused 51% (p ¼ 0.02, Wilcoxon signed-rank), 5% (ns) and 3% (ns) median percentage increases in PV, TAMV and flow rate, respectively; whilst NMES stimulation caused 103% (p ¼ 0.002), 101% (p ¼ 0.002) and 101% (p ¼ 0.002) increases in the same parameters. The benefit was quickly lost upon deactivation in both cases.
Laser Doppler reading changes taken from the leg were not statistically increased by IPC but were significantly raised by NMES (Table 2, Figure 6). During operation of both individual devices, there was a drop in signal from the ipsilateral hand, which returned to positive values after device cessation.
Arterial
Tolerability
IPC caused a median 8%, 12% and 13% changes in PV, TAMV and flow rates, respectively, which were not statistically significant (Wilcoxon signed-rank). NMES caused 11% (p ¼ 0.002), 84% (p ¼ 0.004) and 75% (p ¼ 0.006) increases in the same parameters.
All patients tolerated both devices and completed the trial. On a verbal reported score, from 0 (painless) to 5 (as painful as a sphygmomanometer cuff inflated to 200 mmHg), IPC was rated a mean of 1.5, and NMES as 2.8 (p ¼ 0.006, paired Student’s t-test).
Downloaded from phl.sagepub.com at University of Manitoba Libraries on June 4, 2015
XML Template (2014) [9.4.2014–6:24am] //blrnas3/cenpro/ApplicationFiles/Journals/SAGE/3B2/PHLJ/Vol00000/140061/APPFile/SG-PHLJ140061.3d
(PHL)
[1–8] [PREPRINTER stage]
Williams et al.
5
Table 1. Overall haemodynamic changes from baseline for both devices.
Venous
Arterial
Change from baseline
Median peak velocity, cm/s (interquartile range)
Median TAMV, cm/s (interquartile range)
Median flow rate, cc/min (interquartile range)
IPC on IPC off 10-min post NMES on NMES off 10-min post IPC on IPC off 10-min post NMES on NMES off 10-min post
6.2* 0.6 2.3 36.7** 1.4 1.0 5.2 1.9 0.8 6.8** 5.1 2.9
0.2 0.6 0.1 2.1** 0.0 0.3 0.6 0.3 0.9 4.2** 0.6 0.9
4.2 23.4 0.4 107.6* 8.7 9.2 20.1 3.9 10.8 90.0** 8.0 24.1
IPC
NMES
IPC
NMES
(0.1–16.5) (3.8–0.0) (4.3–0.7) (18.8–49.1) (3.8–2.6) (1.8–4.3) (8.5–2.2) (11.3–0.4) (12.5–4.3) (3.59.7) (8.0–0.2) (6.2–6.2)
(0.2–1.2) (1.6–0.4) (1.6–0.8) (1.2–3.3) (1.0–1.0) (0.1–0.7) (2.0–1.9) (1.8–0.1) (2.5–1.0) (2.9–7.2) (2.2–1.6) (2.3–1.6)
(6.7–48.9) (108–11.0) (91.7–47.8) (13.5–218.5) (85.2–26.7) (21.0–58.7) (37.4–82.1) (45.7–23.1) (73.8–53.2) (64.6–181.7) (51.6–32.4) (62.8–53.7)
Significance tested using Wilcoxon signed-rank, *p
