Effects of Continuous

Passive Motion and

Elevation on
Hand Edema
Marcia L. Giudice
Key Words: edema. hemiplegia.
rehabilitation, hand

The purpose of this study was to evaluate the effi

cacy of the use of continuous passive motion (CPM)
of the digits in combination with limb elevation to
reduce hand edema. The effects of 30 min of CPM of
the digits with the Limb elevated were compared
with the effects of 30 min of Limb elevation alone.
Each of 16 subjects with hand edema of varied etiol
ogy received both treatments, one on each of 2 con
secutive days. Measures of hand volume, finger cir
cumference, and finger stiffness were taken before
and after each treatment. Analyses comparing
mean percentage change scores for both treatments
showed large and significant tl'eatment effects for
aLL three dependent measures. The findings indicate
that, for this sample, CPM with limb elevation was a
more effective treatment for the reduction of hand
edema than Limb elevation alone. The results of
analyses performed on a subgroup of 11 subjects
with hemiplegia were similar, thus suggesting that
CPM with limb elevation may he an effective
method by which to reduce hand edema for this pa
tient population.

Marcia L. Giudice, MS, OTR/L, is Adjunct Instructor, Depart

ment of Occupational Therapy, Sargent College of Allied
Health Professions, Boston University, Boston, Massachu
setts (Mailing address: 42 Fuller Road, Wayland, Massa
chusetts 01778)

This article was accepted for publication Marcb 7, 1990

and edema is a problem frequently encoun

tered by medical professionals involved in
the rehabilitation of patients with paretic
upper extremities and in the rehabilitation of patients
following hand trauma or hand surgery (Vasudevan &
Melvin, 1979). The presence of edema in the hand
alters the viscous component of joint mObility, result
ing in joint stiffness that can hinder or restrict motion
(Brand, 1985). If the edema is prolonged, it can lead
to fibrosis, contracture, pain, inability to accomplish
tasks of daily liVing, and diminished quality of life
(Hunter & Mackin, 1984).
One of the most commonly recommended
methods of hand edema reduction for patients with
paretic hands and for posttraumatic or postsurgical
hand patients is limb elevation or elevation of the
extremity so that the hand is higher than the elbow,
and the elbow higher than the shoulder (Boyes, 1970;
Carter, 1983; Hunter & Mackin, 1984). Limb elevation
has been recommended for use alone (Carter, 1983)
as well as in combination with other treatment
methods for edema reduction, including active and
passive hand exercise, thermal modalities, massage,
and various forms of external compression (Brand,
1985; Hunter & Mackin, 1984; Vasudevan & Melvin,
1979)
Although causal mechanisms and quality of hand
edema may vary not only among patient groups but
also among patients with similar diagnoses (Exton
Smith & Crockett, 1957; Hurley, 1984), the recom
mended conservative treatment methods for edema
reduction share the common goal of increasing lym
phatic and venous drainage from the hand (Vasude
van & Melvin, 1979). Unfortunately, treatment results
for patients with paretic upper extremities and for
patients with postsurgical or posttraumatic edema that
persists beyond the normal healing time are often not
satisfactory, and edema persists (Exton-Smith &
Crockett, 1957; Witte, Witte, & Dumont, 1984). We
must therefore search for more effective methods
of hand edema management for these patient
populations.
Therapists are now using continuous passive mo
tion (CPM) machines, originally devised to aid in
tendon healing and early restoration of motion after
surgery (Salter et aI., 1984), as a way to reduce hand
edema (Petrone & Calvanio, 1989). The use of hand
CPM devices for edema reduction follows clinical ob
servations of reduced edema associated with the use
of CPM and is based on the rationale that CPM pro
vides a pumping action that increases lymphatic and
venous drainage (Coutts, Toth, & Kaita, 1984). Re
search is needed to investigate the efficacy of the use
of CPM as a treatment for hand edema reduction and
to proVide data on which to base treatment protocols.
The purpose of the present study was to investi-

914
Downloaded From: http://ajot.aota.org/ on 10/24/2015 Terms of Use: http://AOTA.org/terms

October 1990, Volume 44, Number 10

gate the efficacy of the use of hand CPM in combina

tion with limb elevation to reduce hand edema. The
effects of 30 min of CPM of the digits with the limb
elevated were compared with the effects of 30 min of
limb elevation alone.
Limb elevation permits gravity to increase the
rate of venous and lymphatic flow from the extremity.
It also slows the rate of formation of edema fluid
through hydrostatic reduction of venous pressure and
a corresponding decrease in filtration pressure in the
capillary bed (Abramson, 1965). Despite the wide
spread use and acceptance of limb elevation as a
means of decreasing edema (Nicholas, 1977), there is
little data documenting its effectiveness for patients
with hand edema (Eccles, 1956; Salisbury et al.,
1973). Sims (1986) found that in 25 nondysfunctional
subjects, a significant (p < .001) mean decrease in
ankle volume (1522 m!) occurred after 20 min of
lower extremity elevation, thus supporting the effec
tiveness of limb elevation as a means of edema
reduction.
Active motion of the hand is commonly recom
mended in combination with elevation of the affected
extremity to reduce hand edema (Boyes, 1970; Brand,
1985; Hunter & Mackin, 1984). Active motion com
presses adjacent veins and lymphatic vessels, thereby
creating a pumping action that enhances venous and
I ymphatic flow (Guyton, 1977). Brand reported that
postoperative hand patiems showed greatly reduced
hand volume following 30 min of elevation combined
with active exercise.
Active motion, however, is not effective for re
ducing hand edema in patients whose hand function
is impaired by paralysis, weakness, immobilization,
stiffness, pain, fear, or poor compliance. The inability
or unwillingness to perform active hand motion and
effectively pump out the excess fluid has been pro
posed to be a major cause of persistent hand edema in
patients with paretic upper extremities (Exton-Smith
& Crockett, 1957) and in postsurgical hand patients
(Milford, 1985). For these patient populations, there
fore, passive motion may proVide an effective alterna
tive to active motion for the management of hand
edema.
Passive motion, like active motion, is proposed to
enhance venous and lymphatic flow from the hand
through the use of a pumping action, which results
from the compression of adjacent veins and lymphatic
vessels (Guyton, 1977) Evidence of enhanced lym
phatic drainage with passive motion has been ob
served in humans (McMaster, 1937) and in animals
(Calnan, Pflug, Reis, & Taylor, 1970; Elkins, Herrick,
Grindlay, Mann, & DeForest, 1953)
Although little documentation exists on the ef
fects of passive motion on edema, it has been pro
posed that continuous passive motion, as prOVided by

a CPM device, assists in edema reduction (Basso &

Knapp, 1987; Coutts et ai, 1984; Petrone & Calvanio,
1989). Coutts et al postulated that the observed "vir
tuallack of edema" (p. 127) in patients who received
2 weeks of CPM for a minimum of 20 hr per day
immediately after total knee replacement was due to
CPM assisting in venous and lymphatiC flow. To test
this postulate, Coutts et al. measured venous flow and
intramuscular pressure during CPM use in 19 lower
extremities after knee surgery. They documented a
consistent surge of venous flow in the femoral vein
and a simultaneous rise in the compartmental pres
sure of the posterior calf to a level above venous
pressure with each flexion cycle of the leg. They sug
gested that CPM produced a pumping effect that pre
vented the accumulation of edema fluid in the pa
tients who used CPM following knee replacement.
Further support for the existence of a pumping effect
produced by CPM was provided by O'Driscoll,
Kumar, and Salter (1983) and Aratow et al (1989)
Basso and Knapp (1987) compared the effects of
CPM administered for a minimum of 20 hr per day
with the effects of CPM administered for a maximum
of 5 hr per day in patients follOWing total knee re
placement. Reductions in midpatellar circumference
were found for both groups (- .65 em and - .80 cm,
respectively), with no significant difference between
the groups.
In 1989, Petrone and Calvanio reported the re
sults of an unpublished study comparing hand vol
ume reduction follOWing 2 hr of hand CPM to hand
volume reduction follOWing 2 hr of compression
wrapping for 10 hemiplegic subjects with hand
edema. They found that the CPM resulted in a mean
decrease of 19 ml in hand volume, which was signifi
cantly greater (p < .001) than the mean reduction of
5.6 ml found follOWing compression wrapping.
Given the lack of research documenting the ef
fects of CPM on hand edema and the encouraging
clinical reports of edema reduction associated with
the use of CPM, the need for research is evident. On
the basis of the assumption that both limb elevation
and passive motion enhance venous and lymphatiC
drainage from the hand, it seems reasonable that CPM
would be most effective for reducing hand edema
when combined with limb elevation. To control for
the effects of limb position as well as to investigate
the ability of CPM to augment the effects of limb ele
vation on edema reduction, I compared the effects of
limb elevation alone with the effects of CPM com
bined with limb elevation. On the basis of studies
reporting edema reduction follOWing treatment ses
sions lasting between 5 and 40 min (Brand, 1985;
Flowers, 1988; Masman & Conolly, 1976; McKnight &
Schomburg, 1982; Sims, 1986), I hypothesized that 30
min of CPM of the digits combined with limb eleva-

The Americall Journal o/Occupmional Therapy

Downloaded From: http://ajot.aota.org/ on 10/24/2015 Terms of Use: http://AOTA.org/terms

915

tion would result in a significantly greater reduction

of hand edema, as measured by hand volume, finger
circumference, and finger stiffness, than 30 min of
limb elevation alone.

Method
A repeated-measures design was used with each of 16
subjects receiving both treatments. Eight of the sub
jects received elevation alone on one day and CPM
with elevation on the next day (at the same time of
day). The other 8 subjects received the treatments in
the reverse order. During the subjects' participation in
the study, their usual activities, including treatment
for edema, were unaltered.

Figure 1. Treatment setup for continuous passive motion with

limb elevation.

Subjects

Patients from four rehabilitation facilities who had vis

ible hand edema more than 4 weeks after upper ex
tremity injury or surgery or after the onset of upper
extremity paresis were referred by their primary occu
pational therapist and asked to volunteer for the
study. Eleven men and 5 women, ranging in age from
38 to 78 years, gave informed consent and completed
the study.
No subjects had pain or limitation of motion in
the fingers that precluded passive flexion of the
fingers to within 1V2 in. of the palm. No subjects had
any known upper extremity lymphatic blockage or a
secondary condition (e.g., rheumatoid arthritis) that
resulted in upper extremity edema. No subjects had
unhealed wounds, infections, or unstable fractures in
their edematous upper extremity.

intrarater reliability for this measurement protocol

was r= 1.0.
A measure of finger stiffness was determined
through the measurement of the degrees of passive
range of motion of metacarpophalangeal flexion
when a constant force of 200 g was applied to the
proximal phalanx of the most visibly edematous
finger for 5 sec (Brand, 1985). A finger goniometer
was used to measure passive range of motion, and a
force gauge was used to monitor the amount of force
applied. The intrarater reliability for this measure
ment protocol was r = .92.
Procedure

To ensure reliable measurement of hand volume,

finger circumference, and finger stiffness, a precise
protocol was developed for each measurement. 1 In
trarater reliability was determined for each protocol
with Pearson product-moment correlations for two
successive measures of 10 normal hands (i.e., of 5
subjects).
Hand volume was assessed through the averaging
of two successive volumetric measures (in milliliters)
of the affected hand. The protocol was based on the
research and recommendations of Devore and Ham
ilton (1968), Eccles (1956), McKnight and Schom
burg (1982), and Waylett and Seibly (1981). Intrarater
reliability for the measurement protocol on normal
hands was r = 1.0.
A circumferential measure (in millimeters) was
taken of the proximal phalanx of the most visibly
edematous finger with the use of a tape measure. The

Measurements were taken immediately before and

after treatment in the following order: (a) finger cir
cumference , (b) finger stiffness, and (c) hand vol
ume. The amount of time reqUired to take the mea
surements was similar for each subject across treat
ments, but varied among subjects from 12 to 30 min.
At all times unaccounted for by measurement or
treatment, the affected extremity was supported so
that it was neither elevated nor dependent.
For the condition of elevation alone, the subjects
were supine on a flat surface with the extremity posi
tioned on a stand adapted to maintain the limb in
approXimately 30 of shoulcler abduction, 30 of
shoulder flexion, and 70 of elbow flexion. For the
condition of CPM combined with elevation, the sub
jects were positioned in the same manner as for ele
vation alone while a CPM hand unit proVided continu
ous passive flexion and extension of the index, mid
dle, ring, and little fingers 2 (see Figure 1).
During both treatments, the subjects' wrists were
supported with the universal wrist splint proVided
with the CPM unit. Additionally, for both treatments,

1 Measurement protocols are available on request by sending a

self-addressed stamped envelope LO the author

2 Treatment proLOcols are available on request by sending a self

addressed stamped envelope to the author.

Instrumentation

916
Downloaded From: http://ajot.aota.org/ on 10/24/2015 Terms of Use: http://AOTA.org/terms

October 1990. Volume 44. Number 10

the subjects were instructed to relax and not to move

the elevated extremity.

Table 2
Results of Two-Way Mixed Analyses of Variance
Calculated on Mean Percentage Change Scores
for the Total Group (N = 16)

Data Analysis

Source

Scores for all 16 subjects and for the subgroup of 11

subjects with cerebrovascular accident (CVA) were
analyzed separately. The mean change scores (pre
score - postscore) and mean percentage change
scores [(mean change score/prescore) X 100J for
hand volume, finger circumference, and finger stiff
ness for elevation alone and for CPM with elevation
were compared with the use of three 2 X 2 (i.e., one
between factor and one repeated factOr) mixed analy
ses of variance (ANOVAs). The between factor was
Treatment Sequence and the repeated factor was Day
(1 and 2). The treatment effect was tested in the in
teraction effect of sequence with day, which is essen
tially a test of the difference in the mean change
scores for CPM with elevation versus elevation alone
This test is similar to a simple t test comparing the
mean change scores, but it uses a more precise error
term, that is, the variance attributed to sequence and
day are removed.
The probability level of .02 (.05/3) was selected
as the criterion for Significance, a level adjusted for
the use of three planned ANOVAs (Rosenthal & Ros
now, 1984). Because this adjusted level is conserva
tive and may lead to failure to reject the null hypoth
esis when it is, in fact, false (i.e., <I Type II error), the
effect size r was also calcu lated for each treatment
effect. The effect size r is a measure that shows the
magnitude of the difference between the two treat
ments: An r of .10 is a small effect, an r of .30 is a

Table 1
Mean Change and Percentage Scores for the Total
Group (N = 16)
Elevation Alone
Measure

CPM With Elevation

Change
Score'

Percentage
Change b

Change
Score'

Percentage
Chani\e b

61
95

11
1.8

145
84

27
16

0.6
06

0.8
08

1.4
09

1.9
12

05
4.9

1.5
8.8

58
40

115
8.1

Hand volume (ml)

M
SD

Finger circumference
(mm)

M
SD
Finger stiffness
(degrees)
M
SD

Nole. CPM = continuous passive motion.

" Change score for hand volume and finger circumference = pre

score - postscore; for finger stiffness, postscore - prescore. Thus,

the higher the score, the greater the reduction in hand volume,

finger circumference, and finger stiffness. hPercentage of change

score = (mean change score/prescore) X 100

df

Ii'

062

<.22

2.20
723

<08
<.009

0.28

<30

023
521

<32
<.02

397

<03

003
1036

<.43
<.003

HAND VOLUME
Between sequence (5)
Error
Within day (D)
S X 0 treatment C
Error

I
14
1
1
14

2.2
3.4
57
18.8
2.6

58

FINGER CIRCUMFERENCE
Between sequence (S)
Error
Within day (D)
S X 0 treatment C
Error

1
14
1
1
14

03
09
0.4
85
1.6

52

FINGER STIFFNESS
Between sequence (S)
Error
Within day (D)
S X 0 treatment C
Error

1
14
1
1
14

2709
68.2
25
791.0
76.3

65

, One-tailed. b r = treatment effect size, calculated by the square

root of F/(F + dferror). Clnteraction (S X D) = treatment effect,
indicating the degree to whicll continuous passive motion with
elevation resulted in a greater improvement than elevation alone.

moderate effect, and an r of .50 is a large effect

(Cohen, 1977)
Pearson product-moment correlations were cal
culated separately for each dependent measure and
for each treatment to determine if a relationship ex
isted between treatmeD[ outcomes (ie, mean per
centage change scores for hand volume, finger cir
cumference, and finger stiffness) and the factors of
(a) time in weeks <lfter onset of diagnosis, which re
sulted in hand edema, and (b) amount of pretreat
ment edema [(prevolume - unaffected hand vol
ume)/unaffecred hand volume] X 100.
Results

Findings[or the Total Group

The mean change scores <lnd mean percentage
change scores for rhe dependeD[ measures of the total
group (N = 16) for both treatments are displayed in
Table 1. The results of rhe ANOVAs and the effect
sizes calculated on the me<ln percentage change
scores are displayed in Table 2. Significant and large
treatmeD[ effects were found for <Ill three dependent
me<lsures. For this sample, the treatment of CPM with
elevation resulted in a significantly greater reduction
of hand edema than the rreatment of elevation alone.
Sequence effects were not significant for the
measures of hand volume or finger circumference.
The sequence effect for the measure of finger stiffness
[F(l,14) = 3.97, P < .03], however, approached the

Tbe Americanjournal of Occupational Tberapy

Downloaded From: http://ajot.aota.org/ on 10/24/2015 Terms of Use: http://AOTA.org/terms

917

significance level of ,02 set for the ANOVA, This finding indicates that subjects who received CPM with
elevation first tended to have a greater reduction in
finger stiffness than did the subjects who received
elevation alone first, which suggests a carryover of
reduced stiffness following CPM with elevation, That
the pretreatment measures of stiffness for Days 1 and
2 are similar for both sequences does not support the
existence of a carryover effect, but does suggest that
the effect was due to chance, The effect of day was not
significant for any of the dependent measures, given
the level of significance set for this study,
The results of the correlational analyses investigating the relationship between treatment outcomes
and (a) time after onset and (b) amount of pretreatment edema are shown in Table 3, The coefficients
relating outcome and time after onset showed a small
to moderate positive relationship (the greater the
time after onset, the greater the treatment effect) for
reduction in hand volume and stiffness following elevation alone and for reduction in stiffness following
CPM with elevation Almost no relationship was
found for hand volume or finger circumference following CPM with elevation or for finger circumference following elevation alone,
Correlation coefficients relating outcomes to
amount of pretreatment edema show a moderate to
large positive relationship (the greater the amount of
pretreatment edema, the greater the treatment effect)
for hand volume and finger circumference following
CPM with elevation, but no relationship for these
same measures following elevation alone, A moderate
positive relationship was found between the amount
of pretreatment edema and decreased stiffness following elevation alone, and a small positive relation-

Table 3
Correlation Coefficients Relating Treatment Outcomes
With Number of Weeks After Onset and Amount
of Pretreatment Edema
CPM With
Elevation

Measure
No,

OF WEEKS AFTER ONSET

Hand volume
Finger circumference
Finger stiffness

(N= 16)

,06
,04
,23

AMOUNT OF PRETREATMENT EDEMA

Hand volume
Finger circumference
Finger stiffness

Elevation
Alone

.41

.52
,18

23
06
,27
h

(Il

= 15) c

02
,02
.41

= continuous passive motion

Mean percentage change [(mean change score/prescore) X 1001,
h[(Prevolume - unalfected hand volume)/unaffecled hand volume]
X 100, 'Unalfected hand volume could not be calculated for one of
the 16 subjects with bilateral hand edema,
Note, CPM

Table 4
Change Scores for Hand Volume, Finger Circumference,
and Finger Stiffness for the CVA Subgroup (n = 11)
Elevation Alone
Measure

CPM With Elevation

Change
Score"

Percentage
Change b

Change
Score"

Percentage
Change b

5,9
10.7

1.1

2.1

13,7
9.4

2,6
1.8

0.7
0.7

09
0.9

1.6
09

2.2
1.3

0.4
5.4

1.8
9.0

45
4,2

8.7
7.7

Hand volume (001)

M

SD
Finger circumference
(moo)
M

SD
Finger stiffness
(degrees)
M

SD

Note. CVA = cerebrovascular accident. CPM = continuous passive

motion.
Change score for hand volume and finger circumference = prescore - postscore; for finger stilfness, postscore - prescore, Thus,
the higher the score, the greater the reduction in hand volume,
finger circumference, and finger stiffness. bpercenrage change
score = (mean change score/prescore) X 100.

ship was found between these variables following

CPM with elevation.

Findings for the eVA Group

Mean change scores and mean percentage change
scores for the dependent measures of the CVA subgroup (n = 11) are shown in Table 4. Results of the
ANOVAs and the effect sizes calculated on the mean
percentage change scores for this subgroup are shown
in Table 5.
The ANOVAs showed no significant sequence effects or day effects. The treatment effects approached
the conservative level of significance set for this study
(.02) for the measures of hand volume (p < .03),
finger circu mference (p < .03), and finger stiffness (p
< .06). In addition, the effect sizes for all three dependent measures were large (r> .50). The large
effect sizes indicate a high degree of difference between the two treatments and suggest that the differences between the mean percentage change scores
for the two treatments would have been significant at
the .02 level if a larger sample had been used. For the
11 subjects with hemiplegia secondary to CVA, 30 min
of CPM with elevation was more effective in the reduction of hand edema than was 30 min of elevation
alone.
Discussion
The use of CPM as a method of hand edema reduction
is a relatively new concept, and no data have been
published documenting its efficacy. The report of a
mean reduction of 19 ml in the hand volumes of 10

918
Downloaded From: http://ajot.aota.org/ on 10/24/2015 Terms of Use: http://AOTA.org/terms

October 1990, Volume 44, Number 10

Table 5
Results of Two-Way Mixed Analyses of Variance
Calculated on Mean Percentage Change Scores
for the CVA Subgroup (n = 11)
d/

Source

[f

043

<.26

193
4.78

<.09
<.03

003

<43

0.14
473

<35
<03

1.41

<13

085
313

<.19
<06

r"

HAND VOLUME
Between sequence (S)
Error
Within day (D)
S X 0 treatment<
Error

9
1
1

2.0
4.8
67
166
35

059

FINGER CIRCUMFERENCE
Between sequence (S)
Error
Within day (D)
S X 0 treatment'
Error

1
9
1
1
9

003
10
0.3
98
21

059

FINGER STIfFNESS
Between sequence (S)
Error
Within day (D)
S X 0 treatment'
Error

1
9
1
1
9

1308
928
505
1863
595

0.51

Note. CVA = cerebrovascular accident.

'Onetailed. b r = treatment effect size, calculated by the square
roOt of F/(F + d/error). 'Interaction (S X D) = treatment effect,
indicating the degree to which CPM with elevation resulted in a
greater improvement than elevation alone.

hemiplegic subjects follOWing 2 hr of hand CPM (Pe

trone & Calvanio, 1989) is encouraging. The authors,
however, did not control for the effects of limb posi
tion, which makes it difficult to separate the effects of
the CPM from the welldocumented (Sims, 1986) ef
fects of limb position.
In the present study, 30 min of CPM with limb
elevation resulted in a significantly greater reduction
in hand edema than did 30 min of limb elevation
alone. One can speculate that CPM augments the ef
fects of elevation by creating a pumping action or a
rhythmiC compression of adjacent veins and Iym
phatic vessels (Coutts et aI., 1984; O'Driscoll et aI.,
1983), thereby enhancing lymphatic and venous
drainage from the hand.
That the findings for the total group (N == 16)
were similar to the findings for the eVA subgroup (n
== 11) suggests that CPM combined with limb elevation may be an effective treatment methocl to reduce
hand edema for patients with hemiplegia after CVA.
Further generalization of the results to specific pa
tient populations is not possible, given the small sam
pie size and the lack of homogeneity among the subjects' conditions. A common Factor across all subjects,
however, was an impaired ability to use the eclematous hand for functional activity (15 subjects had hand
paresis or paralysis, and 1 subject had limited hand
use due to stiffness and pain accompanying

shoulder-hand syndrome). ePM with limb elevation,

therefore, may proVide an effective alternative for
those patients who, due to impaired hand function,
are unable to effectively use active motion for edema
reduction.
It is important to note that in this study and in the
study by Petrone and Calvanio (1989), the measures
of edema that were reduced follOWing ePM treatment
generally returned to pretreatment levels within 24
hr. Investigation of the long-term effects of ePM as
well as of methods and protocols to improve the carryover from treatment to treatment are needed to establish whether CPM is effective for attaining more
than a temporary reduction in hand edema. Research
is also needed to investigate whether even temporary
reduction of edema through the use of CPM is beneficial, particularly, whether treatment could significantly delay or prevent the process of tissue fibrosis
and loss of range of motion that can occur with prolonged edema.
Two correlational analyses were performed to
investigate Factors that could potentially affect a person's response to treatment. One factor was the
amount of time after onset of the condition that resulted in the edema Brand (1985) suggested that simple edema of low viscosity (e.g., edema caused by
limb dependency or lack of muscle action) is easier to
move and redistribute than the more viscous edema
found with inflammation He also suggested that
edema becomes more viscous over time, particularly
when tissues remain stagnant, thereby allowing invasion by protein cells and fibroblasts. This suggestion
was supported by Exton-Smith and Crockett (957),
who collected data on the concentration of protein in
fluids removed from 21 edematous hands of hemiplegic subjects anel proposed that there was a correlation between protein content and time after onset of
hemiplegia.
Assuming that there is a direct relationship between protein content, viscosity, and response to
treatment for edema, it is reasonable to propose that
the longer edema persists, the less responsive it is to
treatment (3 negative correlation between time after
onset and treatment outcome). This postulate was not
supported by the results of the regression analyses
performed on the el3t3 in the present study. Either no
relationship or small to moderate positive relationships were found between the weeks 3fter onset of
the condition that resulted in the hand edema and
treatment ou tcomes.
The second potential factor affecting treatment
outcome that was investigated was the amount of pretreatment edema. The results of the correlational
analyses show a moderate to large positive relationship between the amount of pretreatment edema and

The American./oumal a/Occupational fherapy

Downloaded From: http://ajot.aota.org/ on 10/24/2015 Terms of Use: http://AOTA.org/terms

919

treatment outcome for hand volume and finger circumference following ePM with elevation, but no relationship for these same measures following elevation alone. These findings suggest that ePM with elevation may be more effective for patients with large
amounts of edema. More data are needed, however, to
explore this concept
Although all three measures of hand edema
showed a greater reduction following ePM with elevation than following elevation alone, the measure
that reflected the greatest difference in the subjects'
responses to the two treatments was finger stiffness. I
hypothesized that ePM would result in a greater reduction in hand edema, as measured by a decrease in
finger stiffness. The magnitude of the result can only
be explained, in part, by a reduction in edema. The
decreased stiffness following ePM with elevation
probably was also due to the well-known effect of
passive motion on tissue lengthening (Sapega, Quedenfeld, Moyer, & Butler, 1981). Although it is
beyond the scope of this paper to address the effects
of ePM on joint stiffness and range of motion, it is
worth noting that increased range of motion and decreased stiffness from tissue lengthening are other
potential benefits of ePM (Ketchum, Hibbard, & Hassanein,1979).
Conclusion
The results of this study support the hypothesis that
30 min of ePM of the digits in combination with limb
elevation results in a significantly greater reduction of
hand edema than 30 min of limb elevation alone. The
findings were similar for both the total group (N = 16)
and the subgroup of subjects with hemiplegia secondary to eVA (n = 11), which suggests that ePM with
limb elevation is an effective treatment for the reduction of hand edema in the hemiplegic patient. This
study supports further investigation of the use of ePM
as an adjunctive treatment modality for the reduction
of hand edema, particularly for those patients who do
not respond favorably to traditional treatments for
edema reduction due to an impaired ability to move
or use the hand.
Research is needed to (a) further document the
efficacy of the use of ePM with limb elevation as a
method of edema reduction; (b) collect data on the
immediate, cumulative, and long-term effects of
treatment; (c) develop optimal treatment protocols
for specific patient populations as well as for specific
characteristics of hand edema; and (d) investigate the
reliability and validity of the measures of hand edema
for patients with limb paralysis.
I recommend to those interested in replicating
the study that they (a) use a larger sample size, (b)

use a more homogeneous sample, (c) include only

those subjects with moderate to severe edema, and
(d) increase controls for subjects' activity and limb
position before treatment and between treatment
days . .&
Acknowledgments
I gratefully acknowledge the assistance of Carol McGough,
Director of Occupational Therapy at Braintree
Hospital, and my thesis advisors Catherine Trombly, MA,
OTR/L, I'AOTA, Linda Tickle-Degnen, PhD, OTR/L, and Cynthia
Philips, MA, OTR/L. I would like to thank Sutter Biomedical,
Inc., and LMB Hand Rehab Products, Inc., for proViding
eqUipment and the American Occupational Therapy Foundation for their grant.
This paper is based on a study submitted in partial
fulfillment of the reqUirements for the master of science
degree from the Department of Occupational Therapy at
Boston University's Sargent College of Allied Health Professions, Boston, Massachusetts.
MS, OTR/L,

References
Abramson, D. (1965). Physiologic basis for the use of
physical agents in peripheral vascular disorders. Archives of
Physical Medicine and Rehabilitation, 46, 216-244.
Aratow, M., Coutts, R. D., Mai, M. T., Blevins, F. T.,
Crenshaw, A., & Hargens, A. R. (1989, February). The effect
of continuous passive motion on muscle compartment
pressure and femoral vein flow in the lower extremity.
Paper presented at the 35th annual meeting of the Orthopedic Research Society, Las Vegas, NV.
Basso, D. M., & Knapp, L. (1987) Comparison of two
continuous passive motion protocols for patients with total
knee implants. Physical Therapy, 67, 360-363.
Boyes, J H. (1970). Joints. In J. H Boyes (Ed), Bunnell's surgery of the hand (5th ed., pp. 287-290). Philadelphia: Lippincott
Brand, P. W. (1985). Clinical mechanics of the hand.
St. Louis: Mosby.
Calnan, J S., Pflug, J J., Reis, N. D., & Taylor, L. M.
(1970) Lymphatic pressures and the flow of lymph. British
journal ofPlastic Surgery, 23, 305-317.
Carter, P. R. (1983). Edema controL In P R. Carter
(Ed.), Common hand injuries (pp. 61-65). Philadelphia:
Saunders.
Cohen, J. (1977). Statistical power analysis for the behavioral sciences. New York: Academic Press.
Coutts, R. D., Toth, C, & Kaita,J. H. (984). The role of
continuous passive motion in the rehabilitation of the tOtal
knee patient. In D. L. Hungerford, K. A. Krackow, & R. V.
Kenna (Eds.), Total knee arthroplasty: A comprehensive
approach (pp. 126-132). Baltimore: Williams & Wilkins.
Devore, G. L., & Hamilton, G. F (968) Volume measuring of the severely injured hand. American journal of
Occupational Therapy, 22, 16-18.
Eccles, M. V. (956). Hand volumetrics. Britishjournal
of Physical Medicine, 19, 5-8.
Elkins, E. C, Herrick,). F., Grindlay,). H., Mann, F. C,
& DeForest, R. E. (953). Effect of various procedures on
the flow of lymph. Archives ofPhysical Medicine and Rehabilitation, 34, 31-39.

920
Downloaded From: http://ajot.aota.org/ on 10/24/2015 Terms of Use: http://AOTA.org/terms

October 1990, Volume 44, Number 10

Exton-Smith, A. N., & Crockett, D.]. (1957). Nature of

oedema in paralysed limbs of hemiplegic patients. British
Medical journal, 2, 1280-1283.
Flowers, K. R. (1988). String wrapping versus massage
for reducing digital volume. Physical Therapy, 68,
1072-1076
Guyton, A. C. (1977). Basic human pbysiology: Normal
function and mechanisms of disease (2nd ed.). Philadelphia: Saunders.
Hunter,]' M., & Mackin, E.]. (1984). Edema and bandaging. In]. M. Hunter, L. H. Schneider, E.]. Mackin, & A.
D. Callahan (Eds.), Rehabilitation of the hand (pp.
146-153). St. Louis: Mosby
Hurley, ]. V. (984). Inflammation. In N. C. Staub &
A. E. Taylor (Eds), Edema (pp. 463-488). New York:
Raven.
Ketchum, L. D., Hibbard, A., & Hassanein, K. M.
(1979). Follow-up report on the electrically driven hand
splint. journal ofHand Surgery, 4, 474-481.
Masman, A. R., & Conolly, W. B. (976). Intermittent
pressure in the management of post-traumatic oedema and
lymphoedema. Medicaljournal of Australia, 1, 87-89.
McKnight, P T, & Schomburg, F. L. (1982) Air pressure splint effects on hand symptoms of patients with rheumatoid arthritis. Archives ofPhysical Medicine and Rehabilitation, 63, 560-564.
McMaster, P. D. (1937). Changes in the cutaneous lymphatics of human beings and in the lymph flow under normal and pathological conditions. journal of Experimental
Medicine, 65, 347-371.
Milford, L. (1985). Complications of flexor tendon
grafts. In S. C. Sandzen (Ed), The hand and wrist (p 283)
Baltimore: Williams & Wilkins.
Nicholas, ]. S. (1977) The swollen hamL Pbysiotherapy, 63, 285-286

O'Driscoll, S. W, Kumar, A., & Salter, R B (1983) The

effect of continuous passive motion on the clearance of a
hemarthrosis from a synOVial joint. Clinical Orthopaedics
and Related Research, 176, 305-311.
Petrone, P., & Calvanio, R. (1989, April). Use of tbe
Stryker hand exerciser to control upper extremity edema in
the hemiplegic patient. Paper presented at the National
Stroke Rehabilitation Conference, Boston, MA.
Rosenthal, R., & Rosnow, R L. (984). Essentials of
behavioral research. New York: McGraw-Hill.
Salisbury, R. E., Loveless, S., Silverstein, P., Wilmore,
D. W., Moylan,]. A., & Pruitt, B. A. (1973). Postburn edema
of the upper extremity: Evaluation of present treatment.
journal of Trauma, 13, 857-862.
Salter, R. B., Hamilton, H. W., Wedge, ]. H., Tile, M.,
Torode, I. P., O'Driscoll, S. W., Murnaghan,].]., & Saringer,
J. H. (984). Clinical application of basic research on continuous passive motion for disorders and injuries of synOVial
joints: A preliminary report of a feasibility study. journal of
Orthopaedic Research, I, 325-342.
Sapega, A., Quedenfeld, T, Moyer, R., & Butler, R.
(1981). Biophysical factors in range of motion exercise.
Pbysician and Sportsmedicine, 9, 57-65.
Sims, D. (986). Effects of positioning on ankle
edema. journal of Orthopaedic and Sports Physical Tberapy, 8, 30-33
Vasudevan, S. V, & Melvin,]. L. (1979) Upper extremity edema control: Rationale of the techniques. American
journal of Occupational Therapy, 33, 520-523
Waylett, J., & Seibly, D. (1981). A study to determine
the average deviation accuracy of a commercially available
volumeter. journal of Hand Surgery, 6, 300.
Witte, C. L., Witte, M. H, & Dumont, A E (1984)
Pathophysiology of chronic edema, lymphedema, and fibrosis In N. C. Staub & A. E. Taylor (Eds), Edema (pp.
521-542) New York Raven.

Coming in November:
Special Issue on Multiple Personality Disorder
An overview of multiple personality disorder
Child abuse as an antecedent of multiple personality disorder
Understancling and cooperation among alter and host personalities

UPDATE

Occupational therapy with patients with multiple personality disorder

.
Turn to A]OTfor the latest information on occupational
therapy treatment modalities, aids and equipment, legal and
social issues, education, and research.

The American.lournal of Occupational Tbl.'rap.r

Downloaded From: http://ajot.aota.org/ on 10/24/2015 Terms of Use: http://AOTA.org/terms

921