A randomized, controlled trial

comparing topical steroid application

to wet versus dry skin in children
with atopic dermatitis (AD)
Lucinda L. Kohn, MD,b Yanna Kang, PhD,b and Richard J. Antaya, MDa,b
New Haven, Connecticut

Background: Soak and smear (SS), a technique whereby a bath is followed by topical corticosteroid (TCS)
application to wet skin, is reported to be a beneficial adjunctive therapy for patients with recalcitrant atopic
dermatitis (AD).

Objective: We evaluated whether SS is of greater benefit than application of TCS to dry skin for the
treatment of childhood AD.

Methods: A randomized, investigator-blinded, controlled study was performed in children with AD.
Patients were randomized to apply TCS either via SS (n = 22) or to dry skin (n = 23) for 14 days.
The primary outcome was an improvement in the Eczema Area and Severity Index score. Secondary
outcomes included assessments of disease burden, pruritus, and sleep; morning cortisol levels; and adverse
effects.

Results: Patients with AD severity who applied TCS via SS or to dry skin improved 84.8% (95% confidence
interval 77.5-92.1) and 81.4% (95% confidence interval 70.3-92.4) by Eczema Area and Severity Index score,
respectively. There was no statistical difference between the 2 groups (P value = .85).

Limitations: Small sample size limited the power of our study.

Conclusions: We did not find that application of TCS to presoaked skin works better than application to
dry skin for the treatment of AD in children. ( J Am Acad Dermatol 2016;75:306-11.)

Key words: atopic dermatitis; bath; corticosteroids; eczema; hydration; soak and smear.

S oak and smear (SS) is a technique used to

increase the efficacy of topical corticosteroids
(TCS) for atopic dermatitis (AD) by the patient
soaking in lukewarm water before smearing
Abbreviations used:
AD:
EASI:
SS:
atopic dermatitis
Eczema Area and Severity Index
soak and smear
corticosteroid ointment on wet skin. Soaking TCS: topical corticosteroid
hydrates the stratum corneum, and smearing of TCS

From the Department of Dermatology and Pediatrics,a Yale ID. This study also comprises the MD thesis work of Dr Kohn,
University School of Medicine.b published under the name Lucinda S. Liu.
Supported by Yale University School of Medicine James G. Hirsch, Accepted for publication April 22, 2016.
MD, Endowed Medical Student Research Fellowship. Reprint requests: Richard J. Antaya, MD, Department of Dermatology,
Dr Antaya has been a consultant for the following: Promius, Anacor, Yale University School of Medicine, PO Box 208059, 333 Cedar St, LCI
Astellas, Ranbaxy, Pierre Fabre, and Hoffman-Laroche Pharmaceu- 501, New Haven, CT 06520-8059. E-mail: [email protected].
ticals. Drs Kohn and Kang have no conflicts of interest to declare. Published online May 31, 2016.
Presented as a poster at the Yale University School of Medicine 0190-9622/$36.00
Student Research Symposium on May 7, 2014 in New Haven, Ó 2016 by the American Academy of Dermatology, Inc. Published
CT, and as a poster at the Society for Pediatric Dermatology by Elsevier. All rights reserved.
40th Annual Meeting from July 9 to 12, 2014 in Coeur d’Alene, http://dx.doi.org/10.1016/j.jaad.2016.04.060

306
J AM ACAD DERMATOL Kohn, Kang, and Antaya 307
VOLUME 75, NUMBER 2

traps moisture in the stratum corneum, increases hydrocortisone 2.5% ointment twice daily. All pa-
permeability, and enhances drug absorption.1,2 tients were prescribed hydrocortisone 2.5% ointment
Previous studies have shown that SS results in for application to facial and intertriginous areas.
marked improvement or disease clearance in nearly Participants in the SS arm were instructed to use
all patients.3-5 There is limited evidence of its benefit the regimen once daily with the following instruc-
in the pediatric population; use of SS in children is tions: ‘‘Soak in a bath (not a shower) in plain,
based primarily on uncontrolled case series and lukewarm water for 10 minutes (use a timer) at night
expert opinion. This study then immediately, without
was undertaken to evaluate drying the skin, smear on
whether TCS application us- CAPSULE SUMMARY the corticosteroid ointment.’’
ing SS works better than TCS Participants in the control
d Previous uncontrolled studies suggest
applied to dry skin for the arm were instructed to
that soak and smear is a successful crisis
treatment of childhood AD. refrain from applying TCS to
intervention modality for patients with
wet skin. All participants
METHODS atopic dermatitis.
were educated regarding
Study design d Application of corticosteroid ointment maintenance and flare treat-
We conducted a random- using soak and smear was not superior ment for AD and were
ized, controlled, investigator- to application of corticosteroid ointment instructed to avoid soaps
blinded, single-center trial. to dry skin in children with atopic and cleansers.
Patients between 2 weeks dermatitis in this controlled study. Participants completed
and 18 years of age who met d In children, application of topical the treatments at home and
the clinical criteria for diag- corticosteroids for 2 weeks can result in documented progress, level
nosis of AD were eligible for greater than 80% clearance of atopic of pruritus, level of sleep
this study.6 Patients with less dermatitis if applied to wet or dry skin. deprivation, and compliance
than 5% total body surface in a daily log.
area involvement, with clini-
cally infected AD, allergy to Data collection and outcomes
triamcinolone 0.1% ointment or hydrocortisone 2.5% The main outcome was improvement in EASI on
ointment, or without access to a bathtub were day 14.7 The EASI was chosen as the objective
excluded. Participants were randomly assigned to measurement of disease, because it exhibits good
either the control or SS arm with a 1:1 allocation ratio interevaluator consistency, validity, and sensitivity to
using computer-generated permuted blocks of either change, and correlates well with other validated
4 or 6 in random order. Allocation was revealed only measures of AD severity.8 The EASI does not
after the patient had been recruited to the study. One subjectively measure AD. Therefore, we asked the
investigator was blinded and was responsible for patient or caregiver to record daily scores for itch,
evaluating patients. A separate investigator was not sleep, overall impact of disease, and compliance for
blinded and was responsible for demonstrating TCS the 14 days of treatment. Each patient or caregiver
application technique to the participants. Participants assigned a daily value to the level of pruritus (using a
initially randomized to the control arm were offered scale of 0-10 with 10 indicating severe itch); impact
the opportunity to cross over into the SS arm if their of disease defined as how much the AD affected
AD showed less than 75% clearance by Eczema Area overall quality of life (using a scale of 0-10 with 10
and Severity Index (EASI) score on day 14 (Fig 1). indicating worst quality of life); and quality of sleep
This study (protocol #1206010366) was approved (using a scale of 0-3 with 3 indicating that the patient
by the Yale IRB (Yale Human Research Protection slept poorly).
Program) on July 19, 2012 and the Yale Pediatric On day 14, participants returned for evaluation
Protocol Review Committee on June 1, 2012. Parents and were asked to report adverse effects (Fig 1).
gave written informed permission, and children older Participants had the option of participating in a
than 7 years gave written assent. The study design was 1-time blood draw to measure morning serum
posted on ClinicalTrials.gov and has the identifier cortisol level on day 14.
NCT01675232.
Statistical analysis
Study interventions Sample size calculation was based on the null
Patients 2 years of age or older were prescribed hypothesis that there would be no difference in
triamcinolone acetonide 0.1% ointment twice daily. treatment results between the 2 arms. Predicted
Patients younger than 2 years were prescribed outcomes were based on prior studies and showed
308 Kohn, Kang, and Antaya J AM ACAD DERMATOL
AUGUST 2016

Fig 1. Atopic dermatitis. Participant flow diagram showing randomization, study design, and
crossover design. EASI, Eczema Area and Severity Index.

a predicted improvement of 95.2% with SS and 60.5% Table I. Demographics

with TCS application to dry skin yielding a difference Control Soak and smear
in proportions of 35.3,4,9 Given that our predicted
Patients 23 24
effect size was calculated using vastly different
Age, y, mean 6 SD 3.1 6 4.0 3.2 6 3.4
studies, we chose to set it at a value that was more (range) (0.3-16) (0.3-11)
conservative than calculated. For this study to detect Gender female:male 8:15 10:12
a difference of at least 15% (by EASI) between the 2 Race
treatment arms, with an estimated SD of 20%, and White 14 12
using a 2-sided t test with a = 0.05, we required a Black 2 5
sample size of 26 participants in each arm to reach Asian 5 2
80% power. Anticipating a dropout rate of 5%, we Hispanic 0 2
calculated a need for 55 participants to reach Other 2 1
significant power. This was an intention-to-treat Recruited in summer 3 3
(June, July, August)
analysis. Two-sided Wilcoxon rank sum tests were
Initial EASI score, 15.1 6 6.9 15.8 6 9.1
performed to compare values in the 2 study arms.
mean 6 SD (range) (2.8-29.7) (4.6-34.95)

EASI, Eczema Area and Severity Index.

RESULTS
In all, 46 eligible patients with AD presented to the
Yale Pediatric Dermatology Clinic between July 2012 Three patients were asked to cross over from the
and July 2013, of whom 45 were enrolled in this control arm into the SS arm, and of those, 2 returned
study (Table I). Every patient was referred from a for a second day-14 follow-up visit. The family who
general pediatrician or community dermatologist. In did not return reported difficulty in keeping the
all, 22 patients were enrolled in the SS arm and 23 in follow-up appointment because of work schedules
the control arm. All 45 patients returned on day 14 for and frustration with lack of AD clearance. The data
a follow-up visit and were included in data analysis. were analyzed excluding the crossover group, and
J AM ACAD DERMATOL Kohn, Kang, and Antaya 309
VOLUME 75, NUMBER 2

Fig 2. Atopic dermatitis. Eczema Area and Severity Index (EASI ) score, sleep quality, pruritus,
and overall impact of disease scores in soak and smear (SS ) and control groups on days 1, 7,
and 14.

again including the crossover group. There was Serum cortisol levels
no change in statistical significance between the Ten of 45 participants volunteered to investigate
2 arms with the inclusion of the crossover group. serum morning cortisol level. Blood was drawn
The data presented below excludes the crossover between 8:15 AM and 8:45 AM. Nine patients had
data. serum morning cortisol levels within the normal
Both SS and control arms showed improvement in range, and 1 patient in the control arm had a serum
disease severity measured by EASI. The SS arm morning cortisol level of 4.75 g/dL, which was
showed an 84.8% reduction by EASI and the control below the lower limit of normal of 5 g/dL (Fig 3 and
arm showed an 81.4% reduction (P = .85), demon- Table II).
strating no statistically significant difference between
the 2 arms. Both groups reported improvement in Adverse events
pruritus, overall impact of disease, and sleep on days Three participants in the SS arm and 5 participants
7 and 14. There was no statistically significant in the control arm developed folliculitis. No patient
difference detected in caregiver-reported subjective developed clinical evidence of skin atrophy or
scores between the 2 treatment arms on days 7 or 14 suppression of the hypothalamic-pituitary-adrenal
(Fig 2 and Table II). axis. There were no other reported or observed
adverse events.
Compliance
Patients in the SS arm missed on average Early termination of the study
0.67 6 1.09 days, whereas patients in the control Because of investigator-related constraints on
arm missed on average 0.48 6 0.59 days of TCS time, we were not able to reach our goal sample
application. There was no statistically significant size of 55 participants. To assess if ending the study
difference between compliance rates in the 2 arms before reaching our goal recruitment was a reason-
of the study (P = .8). able option, we performed a futility assessment with
310 Kohn, Kang, and Antaya J AM ACAD DERMATOL
AUGUST 2016

Table II. Study outcomes

Control Soak and smear Wilcoxon
EASI score
Day 1 15.1 (12.1-18.1) 15.8 (11.7-19.8)
Mean (95% CI)
Day 14 2.6 (0.8-4.5) 2.5 (0.9-4.1)
Mean (95% CI)
% Difference 81.4 (70.3-92.4) 84.8 (77.5-92.1) P = .85
Mean (95% CI)
Pruritus
Day 1 5.4 (4.0-6.7) 6.3 (4.9-7.7)
Mean (95% CI)
Day 7 2.7 (1.5-4.0) 2.9 (1.7-4.0) P = .83
Mean (95% CI) 48.8 (28.0-69.7) 39.0 (e8.2 to 86.1)
% Difference (95% CI)
Day 14 1.9 (0.7-3.1) 1.6 (0.6-2.6) P = .85
Mean (95% CI) 75.8 (62.6-88.9) 77.8 (66.4-89.2)
% Difference (95% CI)
Sleep
Day 1 1.3 (0.8-1.8) 1.3 (0.8-1.9)
Mean (95% CI)
Day 7 0.7 (0.3-1.0) 0.6 (0.2-1.0) P = .9
Mean (95% CI) 54.7 (21.8-87.6) 59.0 (34.8-83.2)
% Difference (95% CI)
Day 14 0.7 (0.3-1.0) 0.5 (0.1-0.9) P = .7
Mean (95% CI) 54.4 (27.8-81.1) 62.5 (34.6-90.4)
% Difference (95% CI)
Overall impact of disease
Day 1 5.7 (4.4-7.1) 7.2 (6.1-8.2)
Mean (95% CI)
Day 7 2.9 (1.7-4.0) 3.1 (2.2-4.1) P = .8
Mean (95% CI) 54.1 (37.4-70.8) 52.3 (35.2-69.3)
% Difference (95% CI)
Day 14 2.0 (0.8-3.1) 2.2 (0.8-3.7) P = .6
Mean (95% CI) 74.0 (60.5-87.4) 64.8 (41.4-88.2)
% Difference (95% CI)
Morning cortisol, g/dL
Day 14 12.3 (6.4-18.3) 15.0 (4.5-25.5) P = .6
Mean (95% CI)

CI, Confidence interval; EASI, Eczema Area and Severity Index.

that, with the data we had accrued, there was only

5% chance of rejecting the null hypothesis.

DISCUSSION
This study shows that twice daily application of
TCS over 2 weeks results in significant improvement
in AD severity regardless of whether the TCS is
applied to dry skin or wet skin using the SS
Fig 3. Atopic dermatitis. Morning (AM ) serum cortisol technique. Both groups showed remarkably
levels (g/dL) measured in volunteer participants on day high disease clearance rates of over 80% by EASI.
14. SS, Soak and smear. There was no difference in AD improvement by
EASI, pruritus, sleep quality, or overall impact of
conditional power.10,11 The calculated conditional disease. We failed to demonstrate that SS is more
power was based on data gathered after accruing 45 beneficial than TCS application to dry skin for AD in
patients and was calculated to be 0.05, which means children.
J AM ACAD DERMATOL Kohn, Kang, and Antaya 311
VOLUME 75, NUMBER 2

These results are unexpected given the previous difference between the SS and control arms. With
literature published on SS.3-5 Three studies have more patients, we may have been able to capture a
reported great benefit using SS: a retrospective study subset for which SS could be a better treatment.
of adults with severe, refractory dermatitis; a Another limitation is that we were unable to directly
prospective study of combat soldiers with severe observe the SS regimen. Lastly, our quality-of-life
AD flares; and a retrospective review of children and assessments are not validated.
adults with severe or moderate AD. In both In conclusion, we did not find that application of
retrospective studies, the methods of assessing AD TCS via SS works better than applications to dry skin
improvement are not well validated.12,13 Many of the for the treatment of AD in children.
patients in the retrospective study in adults were
older individuals with associated xerosis. One could REFERENCES
postulate that the disease clearance observed in 1. Schaefer H, Zesch A, St€ uttgen GN. Skin permeability. Berlin:
these studies may be attributable to a beneficial Springer-Verlag; 1982.
2. Rothman S. Physiology and biochemistry of the skin. Chicago:
effect of SS in specific populations, such as older
University of Chicago Press; 1954.
adults with xerosis or patients with more severe 3. Rustad K, Henning S. Direct observed ‘‘soak and smear’’
disease. A major confounder of the study therapy for severe eczematous dermatitis in the combat
conducted by the US Army was that patients were setting. Dermatitis. 2009;20(5):275-277.
relocated from a combat area to a hospital. 4. Gutman AB, Kligman AM, Sciacca J, James WD. Soak and
smear: a standard technique revisited. Arch Dermatol. 2005;
Hospitalization alone may improve the severity of
141(12):1556-1559.
AD.14 Because none of these studies had a control 5. Hajar T, Hanifin JM, Tofte SJ, Simpson EL. Prehydration is
arm, they do not show that SS is superior to standard effective for rapid control of recalcitrant atopic dermatitis.
treatment. Dermatitis. 2014;25(2):56-59.
The improvement in AD clearance in our patients 6. Eichenfield LF, Tom WL, Chamlin SL, et al. Guidelines of care
for the management of atopic dermatitis: section 1. Diagnosis
was most likely attributable to application of
and assessment of atopic dermatitis. J Am Acad Dermatol.
adequate amounts of TCS. All participants were 2014;70(2):338-351.
prescribed a 454-g jar or equivalent of TCS and 7. Hanifin JM, Thurston M, Omoto M, Cherill R, Tofte SJ,
were provided with an estimate of how much TCS Graeber M. The Eczema Area and Severity Index (EASI):
should have been applied by the end of the 2-week assessment of reliability in atopic dermatitis. EASI Evaluator
Group. Exp Dermatol. 2001;10(1):11-18.
study intervention based on age and affected body
8. Schmitt J, Langan S, Williams HC, European
surface area. Dermato-Epidemiology Network. What are the best outcome
Our patients’ families initially presented with measurements for atopic eczema? A systematic review. J
varying degrees of reluctance toward applying Allergy Clin Immunol. 2007;120(6):1389-1398.
adequate amounts of TCS, which we addressed by 9. Schuttelaar ML, Coenraads PJ. A randomized, double-blind
study to assess the efficacy of addition of tetracycline to
education on the benefits and adverse effects. In the
triamcinolone acetonide in the treatment of moderate to
study, they were highly adherent and missed less severe atopic dermatitis. J Eur Acad Dermatol Venereol. 2008;
than 1 day of treatment on average. This may be a 22(9):1076-1082.
result of multiple factors, including the Hawthorne 10. Chow SC, Chang M. Adaptive design methods in clinical trials -
effect, use of a daily score sheet, a reminder a review. Orphanet J Rare Dis. 2008;3:11.
11. Lachin JM. A review of methods for futility stopping based on
telephone call at 1 week, and a short follow-up
conditional power. Stat Med. 2005;24(18):2747-2764.
time of 2 weeks.15 12. Rehal B, Armstrong AW. Health outcome measures in atopic
The only adverse event reported was folliculitis, dermatitis: a systematic review of trends in disease severity and
although no participant reported symptoms. quality-of-life instruments 1985-2010. PLoS One. 2011;6(4):e17520.
Systemic absorption of TCS and effects on the 13. Futamura M, Leshem YA, Thomas KS, Nankervis H,
Williams HC, Simpson EL. A systematic review of Investigator
hypothalamic-pituitary-adrenal axis were negligible.
Global Assessment (IGA) in atopic dermatitis (AD) trials: many
Only 1 patient in the control arm exhibited a options, no standards. J Am Acad Dermatol. 2016;74(2):
morning serum cortisol value below the lower limit 288-294.
of normal, and he was asymptomatic. 14. Lee JH, Kim J, Lee SW, et al. The clinical effects of hospital-
The biggest limitation of our study was the small ization in a low pollutant room on atopic dermatitis. Asia
Pacific Allergy. 2011;1(2):87-92.
number of participants. An effect may have been
15. Feldman SR, Camacho FT, Krejci-Manwaring J, Carroll CL,
seen if we had recruited more patients. However, our Balkrishnan R. Adherence to topical therapy increases
futility calculations indicate that even if we reached around the time of office visits. J Am Acad Dermatol. 2007;
our intended sample size, we would not have seen a 57(1):81-83.