International Journal of Nursing Practice 2010; 16: 454–460

RESEARCH PAPER

Effect of foot massage to decrease physiological

lower leg oedema in late pregnancy:
A randomized controlled trial in Turkey ijn_1869 454..460

Ayden Çoban RN PhD

Assistant Professor, Department of Midwifery, Aydın School of Health, Adnan Menderes University, Aydın, Turkey

Ahsen Şirin PhD

Professor, Department of Obstetric and Gynecologic Nursing, Nursing School, Ege University, Ege, Turkey

Accepted for publication February 2010

Çoban A, Şirin A. International Journal of Nursing Practice 2010; 16: 454–460

Effect of foot massage to decrease physiological lower leg oedema in late pregnancy: A randomized
controlled trial in Turkey

This study aims to evaluate the effect of foot massage for decreasing physiological lower leg oedema in late pregnancy.
Eighty pregnant women were randomly divided into two groups; study group had a 20 min foot massage daily for 5 days
whereas the control group did not receive any intervention beyond standard prenatal care. The research was conducted
between March and August 2007 in Manisa Province Health Ministry Central Primary Health Care Clinic 1, in Manisa,
Western Turkey. Compared with the control group, women in the experimental group had a significantly smaller lower
leg circumference (right and left, ankle, instep and metatarsal–phalanges joint) after 5 days of massage. The results
obtained from our research show that foot massage was found to have a positive effect on decreasing normal physiological
lower leg oedema in late pregnancy.
Key words: foot massage, lower leg oedema, pregnancy, randomized controlled trial.

INTRODUCTION limbs) is a frequent and unpleasant accompaniment to

Physiological lower leg oedema is found in about 80% of pregnancy, causing discomfort, a feeling of heaviness,
all pregnancies, occurring in late pregnancy.1–5 It occurs as night cramps and painful paraesthesia.7
a result of the pressure of the gravid uterus, which In recent years, there has been an increased acceptance
impedes venous return; prostaglandin-induced vascular of the use of complementary therapies within the health-
relaxation; and reduced plasma colloid osmotic pres- care system.8–10 The use of non-pharmacological interven-
sure.3,4,6 Dependent physiological lower leg oedema tions to complement modern technological medicine is
(water retention in the interstitial space of the lower proving popular among nurses and midwives in clinical
practice.11–13 Foot massage is an example of an interven-
tion that can be used for specific conditions such as leg and
foot oedema as it moves extravascular fluid without dis-
Correspondence: Ayden Çoban, Adnan Menderes Üniversitesi, Aydın turbing intravascular fluid.14,15
Sağlık Yüksekokulu Gençlik Cad. No:7, 09100 Aydın/Turkey. Email: Very few studies have investigated alternative therapies
[email protected] that aim to reduce the effects of late pregnancy

© 2010 Blackwell Publishing Asia Pty Ltd doi:10.1111/j.1440-172X.2010.01869.x

Foot massage for lower leg oedema in pregnancy 455

oedema.2,3,16–18 Kent et al. compared the effect on preg- pregnancies and week of pregnancy and thus groups were
nancy oedema of standing on land, static immersion in formed with homogeneous distribution (P = 0.753,
water and water aerobics, each for 30 min, in 18 women P = 0.412, P = 0.093, respectively). Women in the
at 20–30 weeks of gestation. Although static immersion experimental group were given a 20 min foot massage
and water aerobics induced substantial diuresis (180 and daily for 5 days, whereas those in the control group
187 mL), only static immersion decreased leg volume.2 received standard prenatal care alone.
Hartmann and Huch found that a single exercise session of The data collection for the control group included
45 min in water significantly decreased severe bilateral filling out a form containing questions about their socio-
lower leg oedema in nine women with otherwise uncom- demographic and obstetric characteristics on the first day
plicated pregnancies.7 Mollart did not obtain significant of Stage I. Measurements were taken of the participants’
results on lower limb oedema from two reflexology tech- ankles, insteps and foot/toe junctions when they were in
niques when compared with bed rest.3 Katz et al. also a sitting position and made using a tape measure (stretch
found that immersion was a faster and safer way than bed property controlled). These measurements took on
rest to effect the mobilization of extravascular fluid.17 average 15 min at each stage. The ankle circumference
None of these studies have used foot massage. was measured medially and laterally above the malleoli,
Foot massage is manipulation of soft tissue of the foot where the diameter was the smallest. The instep circum-
and is more general and does not focus on specific areas ference was measured over the cuneiform and cuboid
that correlate with other body parts. Reflexology is bones distal to the heel, and the third circumference was
another therapy that applies pressure to specific areas of measured on the distal end of the foot, at the metatarsal–
the feet or hand.14,19 phalanges joint (the MP joint; where toe joins the foot) as
The current study was planned as preliminary random- shown in Figure 1.3,20 The same data were obtained from
ized controlled trial to assess the effect of foot massage in the control group in Stage II (Fig. 2).
decreasing physiological lower leg oedema in late preg- During the data collection for the experimental group,
nancy and this is the first study to evaluate this form of on the first day of Stage I, the same form that was used for
treatment. the control group was completed and the pre-foot massage
measurements were recorded. Then the pregnant women
METHODS were given a 10 min massage to each foot for a total of
After receiving Ege University Nursing School Scientific 20 min of massage. This was repeated every day in the
Ethics Committee approval, a randomized controlled trial same manner and at 10 o’clock time for 5 days. In Stage II,
took place in the Manisa Central Primary Health Care measurements were once again taken for the experimental
Clinic 1, Turkey. During a 6-month period between group. To determine whether the massage had a lasting
March 2007 and August 2007, women were invited to effect, in Stage III the measurements were again recorded 2
participate in the trial during antenatal visits. We tested days after the massage programme had finished (Fig. 2).
the one-sided hypothesis that a daily 20 min foot massage As mentioned above, the intervention consisted of a
for 5 days will decrease lower leg circumferences in lower 20 min massage daily for 5 days. These foot massages
leg oedema. were administered by the same investigator (A. Çoban),
The inclusion criteria for admission to the trial
included: (i) normal pregnancy of > 30 weeks of gesta-
tion; (ii) visible oedema of the ankles and feet; and (iii)
attendance at the Manisa Central Primary Health Care
Clinic 1 for pregnancy care. The exclusion criteria were
the presence of psychiatric problem, pre-eclampsia,
eclampsia and systemic coexisting disease.
After having their written consent, potential partici-
pants were pair-matched, then randomly assigned one of
two women from a matched pair to the experimental
group or the control group. The experimental and control Figure 1. Lower leg circumference measurement. MP joint,
groups were pair-matched statistically for age, number of metatarsal–phalanges joint.

© 2010 Blackwell Publishing Asia Pty Ltd

456 A Çoban and A Şirin

Figure 2. A diagrammatic representation of

the data collection procedure.

who was trained in foot massage by a professional wards and forwards.15 Each of these movements was
masseur. Johnson’s® baby oil (Johnson & Johnson, New carried out 10 times in succession. The right foot were
Brunswick, NJ, USA) was used during the foot massage to always massaged before the left and the same method was
prevent friction and possible resultant discomfort. The used for each foot in turn.
study used a standard massage technique without pressure The Statistical Package for Social Science 11.0 com-
on the points indicated on a reflexology foot as previously puter program (SPSS Inc., an IBM company, Chicago, IL,
described.19 The massage started with the foot being held USA) was used to analyse the data. The c2 test was used to
firmly, then stroked in its entirety from the toes to ankle ensure the groups’ homogeneity (age, gravidity and ges-
along the top of the foot using the whole hand, and return- tation). The quantitative data were analysed using paired-
ing under the foot to the toes using less pressure. The samples t-test to identify differences in foot measurement
second movement involved thumbs kneading the foot before and after the intervention. Student’s t-tests were
from the toes to the ankle while supporting the foot with also used to identify differences between the two groups.
the fingers underneath. The skin surface between each All statistical tests used a level of significance of 0.05.
tendon on the top of the foot was then stroked one after
another using thumbs. The foot was then grasped with RESULTS
both hands and gently manipulated from side to side. The Participants’ characteristics
toes were then held with one hand whereas the other hand Eighty women were invited to participate in the study.
supported the foot and the toes were gently bent back- All 80 participants completed the intervention and all

© 2010 Blackwell Publishing Asia Pty Ltd

Foot massage for lower leg oedema in pregnancy 457

Assessed for eligibility

(n = 84)

Excluded (n = 4)
Not meeting inclusion criteria
Enrollment (n = 2)
Refused to participate
Randomized (80) (n = 2)

Massage group (n = 40 ) Allocation Control group (n = 40)

Lost to follow up (n = 0) Follow up Lost to follow up (n = 0)

Figure 3. Summary: participants’ progress Analysed (n = 40) Analysis Analysed (n = 40)

throughout randomized controlled trial.

Table 1 Demographic characteristics by groups

Characteristics Massage group (N = 40) Control group (N = 40)

Mean Range SD Mean Range SD

Age (years) 27.4 20–35 4.87 26.4 20–35 4.13

Gravidity 1.82 1–4 1.03 1.62 1–4 0.77
Gestation (weeks) 33.45 30–39 2.99 34.62 30–40 2.71

outcome measures (Fig. 3). Descriptive statistics for the According to the survey in the experimental group,
two groups’ means, standard deviations and ranges are 85.0% of the informants have three main meals a day,
expressed in Table 1. whereas 51.4% have two snacks between meals per day
In the experimental group, average length of time and 47.5% have consumption normal dietary salt; on
participants had been married was 5.0 (SD 4.9; range average, they drink 9.3 (SD 4.3) glasses of water per day.
1–18), and they had given birth an average of 1.6 (SD In the control group, 72.5% of women have three main
0.3) times; on average, they had 1.1 (SD 0.3) surviving meals a day, 42.4% have two snacks between meals per
children, and these averages in the control group were day and 55.0% have consumption normal dietary salt; on
4.6 (SD 4.0; range 1–15), 1.2 (SD 0.4), 1.2 (SD 0.4), average, they drink 6.5 (SD 3.1) glasses of water per day.
respectively. In terms of weight gain according to the duration of

© 2010 Blackwell Publishing Asia Pty Ltd

458 A Çoban and A Şirin

pregnancy, 35.0% women in the experimental group and The experimental group’s average measurements on
36.0% women in the control group of informants the fifth day after massage were lower than on the
reported they had gained more weight than expected even seventh day (except for the left ankle). This difference
allowing for the pregnancy. was found to be statistically significant for the right
The informants also reported that participants in the ankle (t = -3.102, P = 0.004), right MP joint
experimental group spent on average 5.8 (SD 2.6) h (t = -2.159, P = 0.037) and left MP joint (t = -3.058,
standing up and 5.6 (SD 2.1) h sitting down during the P = 0.004).
day, and these average in the control group were found The women in the control group were found to have
6.0 (SD 2.7), 5.2 (SD 2.5), respectively. higher average measurements for all measurements on
both legs between the first day and the fifth day. A statis-
Effect of foot massage tically significant difference was found between the first-
Measurements taken of the right ankle (paired t = 3.873, and fifth-day measurements for the right MP joint
P = 0.000), right instep (paired t = 5.024, P = 0.000), (t = -2.813, P = 0.008), left ankle (t = -4.762,
right MP joint (paired t = 2.592, P = 0.013) and left MP P = 0.000), left instep (t = -3.863, P = 0.000), left MP
joints (paired t = 4.482, P = 0.000) of participants in the joint (t = -3.172, P = 0.003), but not for the right ankle
experimental group after their fifth-day programme of (t = -2.016, P = 0.051) or right instep (t = -1.980,
daily 20 min foot massages showed a statistically signifi- P = 0.055).
cant difference between these measurements and those Average lower leg circumference measurements by
taken before the 5-day massage programme began, group are shown in Table 2. A number of participants in
although they did not indicate a statistically significant the control group displayed increases in lower leg circum-
difference in average measurements taken of the left ankle ference measurements; however, in the experimental
(t = 0.029, P = 0.984) and left instep (t = 1.553, group all mean circumference measurements decreased,
P = 0.129). as shown in Table 3. There was a statistically significant

Table 2 Average score on lower leg circumferences by groups

Variables Control group (N = 40) Massage group (N = 40)

Stage I Stage II Stage I Stage II Stage III

Before After 5 days of Second day after
intervention intervention the last intervention
Mean ⫾ SD Mean ⫾ SD Mean ⫾ SD Mean ⫾ SD Mean ⫾ SD
(range) (range) (range) (range) (range)

Right Ankle 25.60 ⫾ 1.70 26.19 ⫾ 2.36 26.28 ⫾ 1.63 26.08 ⫾ 1.73 26.16 ⫾ 1.73
(23.00–30.00) (23.00–36.00) (23.00–31.00) (22.50–31.00) (22.50–31.00)
Instep 25.68 ⫾ 1.63 25.93 ⫾ 1.70 26.40 ⫾ 2.18 26.15 ⫾ 2.07 26.21 ⫾ 2.14
(21.00–29.00) (21.00–30.00) (23.00–33.00) (22.70–32.50) (23.00–33.00)
MP joint 23.81 ⫾ 1.60 24.40 ⫾ 1.98 23.71 ⫾ 1.29 23.50 ⫾ 1.25 23.58 ⫾ 1.24
(20.00–27.00) (20.00–32.00) (22.00–27.00) (21.00–26.00) (22.00–26.30)
Left Ankle 25.33 ⫾ 1.72 25.69 ⫾ 1.89 26.28 ⫾ 1.95 26.27 ⫾ 2.85 26.09 ⫾ 2.04
(22.00–30.00) (22.00–31.00) (22.00–31.50) (22.00–39.50) (22.00–31.50)
Instep 25.74 ⫾ 1.75 26.14 ⫾ 1.86 25.93 ⫾ 1.88 25.74 ⫾ 1.84 25.81 ⫾ 1.84
(21.00–30.00) (21.00–30.00) (23.00–32.00) (23.00–32.00) (23.00–32.00)
MP joint 23.86 ⫾ 1.81 24.09 ⫾ 1.85 23.41 ⫾ 1.25 23.26 ⫾ 1.28 23.31 ⫾ 1.26
(21.00–29.00) (21.00–30.00) (22.00–27.00) (21.00–27.00) (21.20–27.00)

MP joint, metatarsal–phalanges joint; SD, standard deviation.

© 2010 Blackwell Publishing Asia Pty Ltd

Foot massage for lower leg oedema in pregnancy 459

Table 3 Differences in average lower leg circumferences measurements by groups (between Day 1 and Day 5)

Mean measurement changes Control group (N = 40) Massage group (N = 40) (95% CI) P-value

Right ankle 0.60 -0.20 0.19–1.38 0.010

Left ankle 0.36 -0.50 -0.15–0.88 0.167
Right instep 0.25 -0.25 0.23–0.77 0.000
Left instep 0.39 -0.19 0.37–0.81 0.000
Right MP joint 0.60 -0.21 0.35–1.26 0.001
Left MP joint 0.23 -0.14 0.13–0.61 0.002

A positive number denotes an increase and a negative number denotes a decrease. CI, confidence interval; MP joint, metatarsal–phalanges
joint.

difference between the two groups in circumference mea- Conclusion

surements (except for the left ankle). Physiological lower leg oedema is a common problem in
late pregnancy, and foot massage might provide effective
relief for this condition. Our study results suggest that
DISCUSSION regular foot massage is beneficial in terms of decreasing
We sought to evaluate the effects of foot massage on
physiological lower leg oedema in healthy women without
physiological oedema in the lower leg. Because massage
obstetric complications.
has been shown to be an effective treatment for oedema,
stimulating circulation, we were hopeful that foot
massage might show a similar effect on physiological ACKNOWLEDGEMENTS
lower leg oedema in late pregnancy.2,14,21,22 The authors wish to thank Mr M. Yalın (the masseur in
In the current study, the right foot was always mas- Hospital of Celal Bayar University) and Dr U. İnceboz
saged before the left. This might explain unusual finding (Professor, PhD, School of Medicine, Balıkkesir Univer-
that the massage seemed to be more effective for the right sity), the staff and patients for their cooperation and par-
foot than the left, because the investigator who adminis- ticipation in the study.
tered foot massage might start to fatigue and the foot
massage of second foot tended to be less effective than the
first foot. REFERENCES
The findings from the study indicated that a 20 min 1 Cunningham FG, Gant NF, Leveno KJ et al. Williams
foot massage daily for 5 days significantly reduced physi- Obstetrics, 21st edn. New York, USA: McGraw, 2001.
2 Kent T, Gregor J, Deardorff L, Katz V. Edema of preg-
ological lower leg oedema in late pregnancy. These find-
nancy: A comparison of water aerobics and static immer-
ings support the effectiveness of foot massage as an sion. Obstetrics and Gynecology 1999; 94: 726–729.
intervention to reduce physiological lower leg oedema. 3 Mollart L. Single-blind trial addressing the differential
The experimental group showed a consistent decrease in effects of two reflexology techniques versus rest, on ankle
all circumferences compared with the control group in the and foot oedema in late pregnancy. Complementary Therapies
study. The current study indicates that a daily 20 min foot in Nursing and Midwifery 2003; 9: 203–208.
massage can produce significant decreases in physiological 4 Reynolds D. Severe gestational edema. Journal of Midwifery
lower leg oedema. Regular intervention might play an and Women’s Health 2003; 48: 146–148.
5 Bamigboye AA, Hofmeyr GJ. Intervention for leg edema
important role in its effect on physiological lower leg
and varicosities in pregnancy: What evidence? European
oedema in the study. Kent et al.2 and Hartmann and Journal of Obstetrics and Reproductive Biology 2006; 129: 3–8.
Huch7 published similar findings, using different methods. 6 Schroth BE. Evaluation and management of peripheral
Mollart3 and Katz et al.18 did not obtain significant results edema. JAAPA: Official Journal of the American Academy of
on physiological lower leg oedema in late pregnancy. Physician Assistants 2005; 18: 29–34.

© 2010 Blackwell Publishing Asia Pty Ltd

460 A Çoban and A Şirin

7 Hartmann S, Huch R. Response of pregnancy leg edema to 15 Hayes J, Cox C. Immediate effects of a five-minute foot
a single immersion exercise session. Acta Obstetricia et Gyne- massage on patients in critical care. Complementary Therapies
cologica Scandinavica 2005; 84: 1150–1153. in Nursing and Midwifery 1999; 6: 9–13.
8 Pinn G, Pallet L. Herbal medicine in pregnancy. Complemen- 16 Jacobs M, McCance K, Steward M. Leg volume changes
tary Therapies in Nursing and Midwifery 2002; 8: 77–80. with EPIC and posturing in dependent pregnancy oedema.
9 Anderson FWJ, Johnson CT. Complementary and alterna- Journal of Nursing Research 1986; 35: 86–89.
tive medicine in obstetrics. International Journal of Gynaecol- 17 Katz V, Ryder R, Cefalo R, Carmichael S, Goolsby R. A
ogy and Obstetrics 2005; 91: 116–124. comparison of bedrest and immersion for treating the
10 Gaffney L, Smith CA. Use of complementary therapies in edema of pregnancy. Journal of Obstetrics and Gynaecology
pregnancy: The perceptions of obstetricians and midwives 1990; 75: 147–151.
in South Australia. Australian and New Zealand Journal of 18 Katz V, Rozas L, Ryder R, Cefalo R. Effect of daily immer-
Obstetrics and Gynaecology 2004; 44: 24–29. sion on the oedema of pregnancy. American Journal of Peri-
11 Wilkinson JM, Simpson MD. Personal and professional use natology 1992; 9: 225–227.
of complementary therapies by nurses in NSW, Australia. 19 Wang MY, Tsai PS, Lee PH, Chang WY, Yang CM. The
Complementary Therapies in Nursing and Midwifery 2002; 5: efficacy of reflexology: Systematic review. Journal of
142–147. Advanced Nursing 2008; 62: 512–520.
12 Leach M. An examination of factors influencing natural 20 Brijker F, Heijdra YF, Elshout FJJ et al. Volumetric mea-
therapy use in the Royal District Nursing Service. Australian surements of peripheral oedema in clinical conditions. Clini-
Journal of Holistic Nursing 2002; 9: 41–49. cal Physiology 2000; 20: 56–61.
13 Brolinson PG, Price JH, Ditmyer M, Reis D. Nurses’ 21 Field T, Hernandez RM, Hart S, Treakston H, Schanberg S,
perceptions of complementary and alternative medical Kuhn C. Pregnant women benefit from massage therapy.
therapies. Journal of Community Health 2001; 26: 175– Journal of Psychosomatic Obstetrics and Gynecology 1999; 20:
189. 31–38.
14 Tuna N. A’dan Z’ye Masaj, 6th edn. İstanbul, Turkey: Nobel 22 O’Brien JG, Chennubhotla SA. Treatment of edema. Ameri-
Tıp Kitabevleri, 2004. can Family Physician 2005; 71: 2111–2117.

© 2010 Blackwell Publishing Asia Pty Ltd