Blood flow velocity and vascular resistance during passiveleg exercise in the critically ill patientAnneli Thelandersson1, Reinhard Volkmann2,3, and Asa Cider1,3

1Department of Physical Therapy, 2Department of Clinical Physiology, Sahlgrenska University Hospital, Gothenburg, Sweden, and 3Institute of Neuroscience and

Physiology, Physiotherapy, The Sahlgrenska Academy Gothenburg University, Gothenburg, Sweden

Summary

CorrespondenceAnneli Thelandersson, Department of Physical

Therapy, Sahlgrenska University Hospital, 413 45

Gothenburg Sweden

E-mail: anneli.thelandersson@vgregion.seRV died during the preparation of this

manuscript.

Accepted for publicationReceived 28 August 2011;

accepted 09 March 2012

Key words

blood flow velocity; intensive care; passive exercise;

physical therapy; range of motion; resistance index

Passive range of motion exercise is a very common physical therapy treatment forpatients admitted to an intensive care unit. However is the knowledge scarceregarding its impact on blood circulation in the extremities. The objective of thisstudy was therefore to investigate the effect of passive range of motion on arterialperipheral leg blood flow velocity (BFV) and vascular resistance. A cross-sectionalconsecutive study of twelve patients admitted to an intensive care unit and twelvehealthy age- and gender-matched controls was conducted. Passive range ofmotion was performed in one leg by a physical therapist. Blood flow velocitiesand resistance index in the common femoral artery (CFA), blood pressure andheart rate were measured before, directly after and at rest after passive range ofmotion. No changes were seen in BFV or resistance index in the patient group orthe control group. No changes were found in blood pressure or heart rate in thepatient group. In the control group, changes were found in systolic and meanblood pressure, with a higher pressure before passive range of motion. The con-trols had lower BFV and higher resistance index than the patients when compar-ing the groups. The conclusion of this study including twelve patients is thatpassive range of motion does not alter BFV or resistance index in the CFA incomatose and/or sedated critically ill patients.

Introduction

The European Respiratory Society and European Society of

Intensive Care Medicine Task Force on Physiotherapy for Criti-

cally Ill Patients have stated that physical therapy for the criti-

cally ill patient may improve outcome and minimize costs.

They recommend that active and passive mobilization and

muscle training should be introduced early in the treatment of

the critically ill patient (Gosselink et al., 2008).

Physical therapy including positioning, mobilization, chest

physical therapy and active or passive exercises form a large

part of the treatment for patients in the intensive care unit

(ICU) (Zeppos et al., 2007). Range of motion exercises with

or without different devices is a very common physical ther-

apy treatment in the rehabilitation of patients after suffering

from stroke, orthopaedic injuries, sporting accidents and so

forth (Lynch et al., 2005; Kline Mangione et al., 2008; Braun

et al., 2009). It is also a very common treatment for critically

ill patients in intensive care units. The purpose of passive

range of motion (PROM) is for example to maintain or

improve range of motion and reduce the risk of thromboem-

bolism (Stiller, 2000).

On the venous side of the circulation, it has been shown

that passive flexion of the foot increases peak blood flow

velocity (BFV) in the femoral vein (Yamashita et al., 2005).

On the arterial side of the circulation, the effects of passive

range of motion for the critically ill patients are to our knowl-

edge not known. Heart rate (HR), blood pressure (BP), intra

cranial pressure and cerebral perfusion pressure have earlier

been shown not to change to unsafe levels during PROM in

patients admitted to a neurosurgical intensive care unit

(NICU) (Koch et al., 1996; Brimioulle et al., 1997; Thelanders-

son et al., 2010). Neither does cerebral BFV in the middle

cerebral artery nor pulsatility index change during passive

range of motion in severe neurosurgical patients (Thelanders-

son et al., 2010). Despite there being very little research evalu-

ating the effectiveness of specific physical therapy treatments

such as PROM in the ICU (Wiles & Stiller, 2010;), PROM

remains a very common clinical treatment.

To our knowledge, the effect of passive range of motion

exercises according to arterial BFV and peripheral leg vascular

resistance index (RI) has not been investigated earlier in

patients that are critically ill and cared for in an intensive care

unit. The aim of this study was therefore to investigate and

Clin Physiol Funct Imaging (2012) 32, pp338342 doi: 10.1111/j.1475-097X.2012.01132.x

338 2012 The Authors

Clinical Physiology and Functional Imaging 2012 Scandinavian Society of Clinical Physiology and Nuclear Medicine 32, 5, 338342

compare the effect of PROM in comatose and/or sedated

patients with a healthy age- and gender-matched control

group with special reference to BFV and RI in the common

femoral artery (CFA).

Method

Design overview

This study has a cross-sectional design with consecutive inclu-

sion of patients admitted to the Neurosurgical Intensive Care

Unit at Sahlgrenska University Hospital. In addition, we also

investigated an age- and gender-matched healthy control

group. Our protocol followed the declaration of Helsinki and

was approved by the Regional Ethics Committee of Gothen-

burg, Sweden. Informed consent for study participation was

obtained from the patients next of kin and the healthy con-

trols, respectively.

Setting and participants

Setting for the study was an eight-bed NICU at Sahlgrenska

University Hospital Gothenburg, Sweden. A total of 12

patients, 11 men and 1 woman (median age 43, ranging

between 21 and 74 years) were included. Inclusion criteria

were admittance to the NICU, mechanical ventilation, a paren-

chymal or intraventricular catheter for intra cranial pressure

measurements and inability to actively move their limbs due

to coma and/or sedation. Their diagnoses included traumatic

brain injury (n = 9), intracerebral infarction (n = 2) and sub-dural haematoma (n = 1). None of the patients were paralysedat any time and sedation and medication were unchanged

during the study period. Exclusion criteria were fractures or

other injuries making PROM impossible. An age- and gender-

matched control group of 12 healthy volunteers (median age

43, ranging between 19 and 75 years) was also enrolled. The

data of the participants are presented in Table 1.

Intervention

Passive exercise in one leg was performed by a physical thera-

pist according to the clinical guidelines for the NICU at

Sahlgrenska University Hospital with the subjects lying in the

supine position. The passive exercise performed was flexion/

extension of toe and ankle, flexion of the hip at the same time

as flexion of the knee, abduction and rotation of the hip. Every

movement was repeated seven times on each subject. This pro-

cedure follows the standard PROM technique for the leg as is

performed in the clinic. In the patient group, BP (invasive

measurements) and HR were measured every minute and in

the control group every second minute for ten minutes prior

to PROM, once again directly after PROM and then for ten

minutes at rest after PROM. BFV and RI were measured in CFA

twice before, once directly after and twice at rest after the

PROM in both groups. The probe was placed over the CFA in

the groin. RI was calculated as systolic BFV minus diastolic

BFV divided by systolic BFV. The measurements were obtained

with the patients and the healthy controls leg lying flat on the

bed. BP and HR in the patient group were monitored by Datex

Ohmeda S/5, (Datex-Ohmeda Division Instrumentarium Corp,

Helsinki, Finland). For the healthy control group, BP and HR

were measured with Datex cardiocap II (Datex Instrumen-

tarium Corp, Helsinki, Finland). The patients were ventilated

with Maquets Servoi, with settings according to Table 1

(Maquet, Solna, Sweden). BFV in the CFA was monitored by

PMD 100 (Spencer Technologies, Seattle, WA, USA).

Statistical analysis

Comparisons between measurements were made by Wilcoxon

signed rank test. When comparing the two groups, a Mann

Whitney U-test was used. All calculations were performed

with the Statistical Package Software Systems (SPSS) for Win-

dows version.150 (Chicago, IL, USA). To avoid mass signifi-cance related to the relatively large number of statistical tests,

the level of significance was set at P 001. Data are given asmean standard deviation (SD) if not otherwise stated.

Results

A total of 12 patients met the criteria for inclusion during the

study period and none were excluded because of the exclusion

Table 1 Data of the participants

Patient group Control group

Number of participants 12 12Male/Female 11/1 11/1Age (year) 43 19 44 19Diagnosis (n)TBI 9INF 2SDH 1SedationPropofol 20 mg ml1

(ml h1) (n = 4)55 31

Midazolam 5 mg ml1

(ml h1) (n = 10)28 14

Leptanal 50 lg ml1

(ml h1) (n = 12)24 11

Pentothal 25 mg ml1

(ml h1) (n = 5)49 2

Ventilator settingsVentilator setting VC-PR 12Minute volume (l min1) 103 12Frequency 17 3FiO2 035 01PEEP (cmH2O) 9 2

TBI, traumatic brain injury; INF, intracerebral infarction; SDH, sub-dural haemorrhage; VC-PR, volume controlled-pressure regulated;FiO2, fraction of inspired oxygen; PEEP, positive end-expiratory pres-sure.Data in mean SD or number of participants (n).

2012 The AuthorsClinical Physiology and Functional Imaging 2012 Scandinavian Society of Clinical Physiology and Nuclear Medicine 32, 5, 338342

Blood flow during passive range of motion, A. Thelandersson et al. 339

criteria set prior to study start. No differences were found in

BFV or RI in the patient group or the control group. When

comparing the results of the two groups, a significantly lower

BFV and higher RI were found in the control group compared

to the critically ill patients, Table 2.

No changes were found in the patient group regarding BP,

but in the control group systolic blood pressure was higher

before PROM than directly after and after at rest, and MAP

was significantly higher before PROM than after at rest,

Table 2. No changes were found in HR in the two groups.

When comparing the two groups BP and HR, no differences

were found. Systolic BFV and RI for the patient group are pre-

sented in Figs 1 and 2.

Discussion

In this study including twelve patients, we demonstrated that

PROM does not alter BFV and RI in CFA in critically ill

patients that are sedated and/or comatose and therefore

unable to actively move their limbs. We also concluded that

PROM does not alter BFV and RI in CFA in severely head-

injured patients with an otherwise fragile circulation. Neither

does BFV and RI change in a healthy age- and gender-matched

control group during PROM.

We found no changes in BP in the patient group and no

changes in HR in the patient group or the control group. We

did, however, find a decrease in SBP and MAP in the control

group during and after PROM compared to before. This

higher BP before PROM in the controls might be explained by

the fact that they were initially stressed because they were par-

ticipating in a study and did not know what to expect. How-

ever, once the PROM commenced, the controls relaxed and

their blood pressure dropped. Many of them also told us that

they found PROM very calming and pleasant which may also

explain the results.

The definition of PROM is repeated movements of a joint

within the range of that joint (Morris, 2007). PROM at our

NICU is performed 5 days a week, one set per movement in a

joint and between 5 and 10 repetitions per movement in

sedated patients without increased muscle tone. Since the

duration of this PROM is so short, it might be too short to

allow a cardiovascular response.

We found no changes in BFV or RI neither in the patient

group nor the controls. These findings are supported by the

Table 2 Blood pressure, heart rate, blood flow velocity and resistance index in common femoral artery (CFA) before, directly after and after pas-sive range of motion exercises (PROM)

Patients Controls

Before PROM Directly after PROM After PROM Before PROM Directly after PROM After PROM

SBP mmHg 134 19 137 20 134 21 133 15 129 14* 129 13*MAP mmHg 91 10 92 11 90 12 94 12 91 11 91 10*DBP mmHg 69 7 71 8 69 8 75 11 72 10 72 10HR 71 13 70 12 70 12 61 11 59 11 59 10CFA/Vs cm/s 80 27 80 24 75 20 60 10 55 10 60 8CFA/Vm cm/s 20 12 21 10 19 9 9 3 9 3 7 1RI 128 015 128 014 127 017 143 006 141 008 148 006

Data are mean SD. SBP, systolic blood pressure; MAP, mean blood pressure; DBP, diastolic blood pressure; HR, heart rate; CFA/Vs, systolicvelocity; CFA/Vm, mean velocity; RI, resistance index.*P

findings of Woerds et al. (2006). They measured blood flow

and vascular resistance after passive movements done by a

physical therapist and after passive cycling in subjects with

spinal cord injury and in an uninjured control group and did

not find any changes in leg blood flow or vascular resistance.

Physical therapists performing PROM often observe and feel

the patients foot becoming warmer and gaining a much bet-

ter colour during treatment. In this study, no differences were

found in BFV and RI when measured at the level of CFA. The

results may, however, have been different had we measured

at the level of a more peripheral vessel. The loss of increase in

BP and HR and thereby no change in energy consumption

might explain why there were no changes in BFV and RI in

the femoral artery. Flow velocity changes have, however, been

reported in studies of the veins during PROM. Manual foot

exercise performed by a nurse for 5 min and intermittent

pneumatic foot compression increases peak blood flow veloci-

ties in the femoral vein (Yamashita et al., 2005). These veloc-

ity changes are probably caused by the activation of the

skeletal muscle pump, not necessarily combined with an

increased flow of extra blood volume. When comparing the

two groups BFV and RI, we found that the patient group had

higher BFV and lower RI than the controls. This difference

might be explained by the fact that the patients were coma-

tose and/or sedated thus fully relaxed in their limbs while the

controls were not due to the testing situation.

Even though no differences were found in BFV and RI dur-

ing passive exercise in the critically ill patients, one should

not forget other possible and positive effects that PROM might

give these patients e.g. preserved joint mobility, muscle

strength and relaxation. The important role physical therapists

have in the treatment/rehabilitation of ICU patients should

also be noted; assessing muscular and neurological function,

treatment to prevent physical deconditioning and related com-

plications as well as treatment to improve respiratory function

all of which contribute to the patients overall well-being. If

the physical therapy treatments are appropriately administered,

they might even improve outcome, reduce risks and minimize

costs for the patients in ICUs (Gosselink et al., 2008).

There are a few limitations with this study. Firstly, PROM

was not performed using a device that could be standardized

for speed, force and angle of the movements. The PROM was,

however, performed manually by a skilled physical therapist

with many years experience of PROM in this category of

patients and according to the guidelines at the NICU at

Sahlgrenska University Hospital in Gothenburg, Sweden. We

therefore strongly believe that the PROM performed was stan-

dardized to a very good level. We also believe that it is

important that studies are performed in a clinical setting to

obtain realistic results. Secondly, the results obtained directly

after PROM were based on only one measurement compared

to five and ten for BP and HR and two for BFV and RI calcu-

lated as mean at rest before and after PROM. The reason for

this is that it was impossible to hold the probe fixed over the

CFA during exercise and we wanted to get a measurement as

close to the exercise as possible. Finally, the gender distribu-

tion of 11 men and 1 woman makes it hard to generalize the

results for women. This result was unavoidable as the partici-

pants were included consecutively. We also believe that the

response to this type of PROM does not differ between men

and women so the distribution of men and woman is proba-

bly of no concern.

The conclusion of this study including twelve patients is

that passive range of motion exercise does not have any effect

on BFV and resistance index in CFA in the critically ill sedated

and/or comatose patient and that passive range of motion

does not alter the fragile circulation of the severely head-

injured patient.

Authors contribution

All authors designed the study. RV performed the measure-

ments and calculations and AT performed the passive exercise.

AT and AC included the healthy persons. AT and AC drafted

the manuscript, read and approved it. RV died before the

manuscript was completed. RV was a splendid scientist and a

great source of inspiration for us. He will be greatly missed.

Acknowledgments

This study was supported by grants from the Research and

Development Council of Goteborg and Bohuslan.

References

Braun S, Kokmeyer D, Millett P. ShoulderInjuries in the Throwing Athlete. J Bone Joint

Surg Am (2009); 91: 966978.

Brimioulle S, Moraine J, Norrenberg D, KahnR. Effects of positioning and exercise on

intracranial pressure in a Neurosurgicalintensive care unit. Phys Ther (1997); 77:

16821689.Gosselink R, Bott J, Johnson M, Dean E, Nava

S, Norrenberg M, Schonhofer B, Stiller K,

van de Leur H, Vincent JL. Physiotherapy

for adult patients with critical illness: rec-ommendations of the European respiratory

Society and European Society of Intensivecare Medicine Task Force on Physiotherapy

for Critically Ill Patients. Intensive Care Med(2008); 34: 11881199.

Kline Mangione K, Lopopolo R, Neff N, CraikR, Palombaro K. Interventions Used by

Physical Therapists in Home Care for PeopleAfter Hip Fracture. Phys Ther (2008); 88:

199210.

Koch S, Fogarty S, Signorino C, Parmley L,

Mehlhorn U. Effect of passive range ofmotion in intracranial pressure in neurosur-

gical patients. J Crit Care (1996); 11: 176179.

Lynch D, Ferraro M, Krol J, Trudell C, Chris-tos P, Volpe B. Continuous passive motion

improves joint integrity following stroke.Clin Rehabil (2005); 19(6): 594599.

Morris P. Moving Our Critically Ill Patients:mobility Barriers and Benefits. Crit Care Clin

(2007); 23: 120.

2012 The AuthorsClinical Physiology and Functional Imaging 2012 Scandinavian Society of Clinical Physiology and Nuclear Medicine 32, 5, 338342

Blood flow during passive range of motion, A. Thelandersson et al. 341

Stiller K. Physiotherapy in Intensive Care:

towards an evidence-Based Practice. Chest(2000); 118: 18011813.

Thelandersson A, Cider A, Volkmann R. Cere-brovascular and systemic haemodynamic

parameters during passive exercise. Adv Phys-iother (2010); 12: 5863.

Wiles L, Stiller K. Passive limb movement forpatients in an intensive care unit: a survey

of physiotherapy practice in Australia. J Crit

Care (2010); 25: 501508.Woerds W, De Groot P, Kuppevelt van D,

Hopman M. Passive Leg Movements andPassive Cycling Do Not Alter Arterial Leg

Blood Flow in Subjects With Spinal CordInjury. Phys Ther (2006); 86: 636645.

Yamashita K, Yokoyama T, Kitaoka N, Nishiy-ama T, Manabe M. Blood flow velocity of

the femoral vein with foot exercise com-

pared to pneumatic foot compression. J ClinAnesth (2005); 17: 102105.

Zeppos L, Patman S, Berney S, Adsett J,Bridson J, Paratz J. Physiotherapy interven-

tions in intensive care is safe: an observa-tional study. Aust J Physiother (2007); 53:

279283.

2012 The AuthorsClinical Physiology and Functional Imaging 2012 Scandinavian Society of Clinical Physiology and Nuclear Medicine 32, 5, 338342

Blood flow during passive range of motion, A. Thelandersson et al.342