Pediatric Dermatology Vol. 32 No.

3 353–357, 2015

Molluscum Contagiosum: To Treat or Not to

Treat? Experience with 170 Children in an
Outpatient Clinic Setting in the Northeastern
United States
Hatice Basdag, M.S., Barbara M. Rainer, M.D., and Bernard A. Cohen, M.D.

Division of Pediatric Dermatology, Departments of Pediatrics and Dermatology, Johns Hopkins University School of
Medicine, Baltimore, Maryland

Abstract: Despite the high prevalence of molluscum contagiosum

(MC) in children, epidemiologic data on this common self-limited viral
infection is limited. In this report we review our experience with the
demographic characteristics, clinical characteristics, management, and
time to resolution of MC in 170 children. A retrospective medical chart
review and telephone survey were conducted on children younger than 16
years of age evaluated for MC in the Division of Pediatric Dermatology at
the Johns Hopkins Children’s Center, Baltimore, Maryland, from January
1, 2008, to December 31, 2011. Of 170 children with MC, 51.8% were female
and 77.1% were Caucasian. The median age at diagnosis was 5 years and
46.5% had a history of atopic dermatitis (AD). Children with AD had
significantly more MC lesions than those without (p < 0.05); 72.9% of
children did not receive any treatment. MC lesions completely cleared
within 12 months in 45.6% of treated and 48.4% of untreated children and
within 18 months in 69.5% of treated and 72.6% of untreated children.
Treatment (if any), sex, race, diagnosing physician, number of lesions at
diagnosis, number of anatomic locations, or history of AD did not predict
time to resolution of MC lesions. MC lesions completely resolved in
approximately 50% of children within 12 months and in 70% within
18 months. Treatment did not shorten the time to resolution.

Molluscum contagiosum (MC) is a common, self- poxvirus that is subdivided into four genetically
limiting cutaneous viral infection with a prevalence of distinct but clinically indistinguishable MC virus
5.1% to 11.5% in children from 0 to 16 years old (MCV) types (3,4). MCV genotype 1 is most prevalent
(1,2). MC is caused by a double-stranded DNA and the cause of 98% of cases in the United States (5).

Address correspondence to Barbara M. Rainer, M.D., Depart-

ment of Dermatology, Johns Hopkins University School of
Medicine, 601 N. Caroline Street, #6068, Baltimore, MD 21287,
or e-mail: [email protected].

DOI: 10.1111/pde.12504

© 2015 Wiley Periodicals, Inc. 353

354 Pediatric Dermatology Vol. 32 No. 3 May/June 2015

It is also the most common cause of MC in children. respiratory), and frequent infections (ear infections,
MCV genotype 2 occurs more commonly in the throat infections, skin infections, pneumonia or any
anogenital area of sexually active adolescents and serious internal infections such as meningitis). MC
adults (6). MCV is contagious and can be spread was “treated” or “untreated.” Specific treatment
through several modes of transmission, most com- options for MC included topical therapies (tretinoin
monly direct skin-to-skin contact and less commonly cream, imiquimod, cidofovir, cimetidine, cantharidin)
fomites on bath sponges and towels (7). Investigators and locally destructive therapies (curettage, cautery,
have hypothesized that activities resulting in close cryotherapy). Time to resolution of MC lesions was
physical contact, such as contact sports, swimming defined as time (in months) from when the patient or
pools, day care, and shared baths among siblings, are family member first noticed the lesions until their
risk factors (8,9). Although MC rarely causes serious complete resolution. The collected data was verified
medical complications, it is often associated with with the patients’ medical records on the Johns
anxiety and social stigma (10). Although MC is Hopkins electronic patient records database.
commonly encountered in clinical practice, there is Patients were excluded from the analysis if they
little epidemiologic data on MC infection in children, were unable to recall specific events, they could not
with most case series focused on small specific popu- complete the survey, or their recollection of events did
lation subgroups (11–13). not match their medical records. SPSS Statistics 21
We conducted a retrospective medical chart review (SPSS, Chicago, IL) was used for data analysis.
and telephone survey of 170 pediatric patients to Categorical variables were compared using chi-square
investigate demographic characteristics, clinical char- and Fisher exact tests as appropriate. The significance
acteristics, management, and time to resolution of level was p < 0.05.
MC infection in children and to assess whether
treatment shortens the course of the infection.
RESULTS
MATERIALS AND METHODS Patient characteristics
The Johns Hopkins Institutional Review Board Of 308 identified children, 170 (55.2%) participated in
approved the study protocol. Retrospective medical this study; 138 refused to participate, were ineligible,
chart reviews and telephone interviews were con- or were excluded from the analysis (unable to schedule
ducted at the Johns Hopkins Pediatric Dermatology interview, language barrier, unable to recall specific
Clinic in Baltimore, Maryland, between April and events). The median age at diagnosis was 5 years
August 2013. Eligible patients were 15 years of age (range 1–15 years), 88 (51.8%) were female, and 131
or younger at the time of clinical diagnosis of MC as (77.1%) were Caucasian; 79 (46.5%) had a personal
assessed by a pediatric dermatologist at the Johns history of AD. There was a history of asthma in 33
Hopkins Pediatric Dermatology Clinic between Jan- (19.4%), drug allergies in 20 (11.8%), respiratory
uary 2008 and December 2011. allergies in 15 (8.8%), and food allergies in 15 (8.8%).
Selected patients and their families were first Three children had a history of immunodeficiency.
notified by mail that they had been enrolled in the Demographic characteristics and personal medical
study and could opt out by returning the letter or history are presented in Table 1.
contacting our clinic over the telephone. To optimize
contact with parents and legal guardians and to
Clinical characteristics of MC lesions
minimize selection bias, telephone interviews were
conducted at different times of the day (between 8 The median length of time to diagnosis was 2 months
a.m. and 10 p.m.) and on all days of the week. If the (ranging from a few days to 9 months). The majority
patient was old enough to understand the questions of children (59.4%) had 10 to 50 lesions. The most
asked and recall the sequence of events, their response commonly affected location was the trunk (62.9%); 71
was included in the telephone interview. Telephone (41.8%) children had lesions in three or more
interviews were conducted to assess demographic anatomic locations. Fifteen (8.8%) children had a
characteristics (age, sex, race), MC characteristics family member previously evaluated for MC. The
(number of lesions, distribution, symptoms, side clinical characteristics of the MC lesions are shown in
effects, management, time to resolution), diagnosing Table 2. MC lesions were significantly more numer-
physician, and personal medical history, including ous in children with a personal history of AD than in
atopic dermatitis (AD), asthma, allergies (food, drug, those without (p < 0.05) (Table 3).
 Basdag et al: Pediatric Molluscum Contagiosum 355

TABLE 1. Patient Characteristics and Personal Medical TABLE 3. Number of Molluscum Contagiosum Lesions at
History (N = 170) Diagnosis in Patients With and Without a History of Atopic
Dermatitis (AD)
Characteristic Value
<10 10–49 50–99 ≥100
Sex, n (%)
Male 88 (51.8) History of AD n (%)
Female 82 (48.2)
Race, n (%) No 37 (40.7) 53 (58.2) 1 (1.1) 0
Caucasian 131 (77.1) Yes 23 (29.1) 48 (60.8) 5 (6.3) 3 (1.8)
African American 24 (14.1)
Hispanic 9 (5.3)
Asian 4 (2.4)
Other 2 (1.2)
Personal medical history, n (%) TABLE 4. Management of Molluscum Contagiosum
Atopic dermatitis 79 (46.5) Lesions
Asthma 33 (19.4)
History of infections 27 (15.9) Treatment n (%)
Drug allergy 20 (11.8)
Food allergy 15 (8.8) No 124 (72.9)
Respiratory allergy 15 (8.8) Yes 46 (27.1)
Immunodeficiency 3 (1.8) Tretinoin 14 (8.2)
Age, median (range) 5 (0–15) Imiquimod 12 (7.1)
Cantharidin 10 (5.9)
Cryotherapy 8 (4.7)
TABLE 2. Clinical Characteristics of Molluscum Curettage 7 (4.1)
Contagiosum Infection (N = 170)

Characteristic n (%)
TABLE 5. Time to Resolution (Months) of Molluscum
Number of lesions Contagiosum Lesions in Treated and Untreated Patients
0–9 60 (35.3)
10–49 101 (59.4) 0–5 6–11 12–17 18–23 ≥24
50–99 6 (3.5)
≥100 3 (1.8) Treated n (%)
Anatomic location
Trunk 107 (62.9) No 14 (11.3) 46 (37.1) 30 (24.2) 10 (8.1) 24 (19.4)
Legs 96 (56.5) Yes 6 (13) 15 (32.6) 11 (23.9) 6 (13) 8 (17.4)
Arms 74 (43.5)
Head and neck 53 (31.2)
Buttocks 29 (17.1)
Genitalia 28 (16.5)
Feet 9 (5.3) Overall, 92.9% experienced inflammation during
Hands 8 (4.7)
Number of anatomic locations
the course of the infection and 83.5% were prescribed
1 40 (23.5) a topical steroid as a result. MC lesions completely
2 59 (34.7) cleared within 12 months in 21 (45.6%) treated and 60
≥3 71 (41.8)
Time to resolution, months
(48.4%) untreated children and within 18 months in
0–5 20 (11.8) 32 (69.5%) treated and 90 (72.6%) untreated children
6–11 61 (35.9) (Table 5). Treatment (if any), sex, race, diagnosing
12–17 41 (24.1)
18–23 16 (9.4)
physician, number of lesions at diagnosis, number of
≥24 32 (18.8) anatomic locations, and history of AD did not predict
Diagnosing physician time to resolution.
Pediatrician 86 (50.6)
Dermatologist 84 (49.4)
DISCUSSION
We reviewed the demographic characteristics, clini-
Management, side effects, and time to resolution
cal characteristics, management, and time to resolu-
One hundred twenty-four (72.9%) children did not tion of MC in 170 children who attended our clinic.
receive any treatment. Of the treated children, 36 were In almost half of the children, the MC lesions
treated with topical therapies (tretinoin cream, imiq- resolved within 12 months; the time to resolution
uimod, cantharidin) and 15 with local destructive was similar for treated and untreated patients.
measures (cryotherapy, curettage). Management of Children with a history of AD had significantly
the MC lesions is presented in Table 4. more MC lesions than those without. Sex, race, age,
356 Pediatric Dermatology Vol. 32 No. 3 May/June 2015

number of MC lesions, diagnosing physician, history different treatment modalities for MC lesions in
of AD, asthma, and allergies did not affect time to children.
resolution.
The peak prevalence of MC was at 5 years of age,
CONCLUSION
and girls and boys were equally affected. These
findings are comparable with results from previous Approximately half of MC infections resolved within
surveys conducted in Japan (14), the Netherlands (15), 12 months and 70% within 18 months regardless of
the United States (16), and France (2). In the current whether lesions were treated. Children with a history
study, 77% of subjects were Caucasian, 14% African of AD had more MC lesions, but time to resolution
American, 5% Hispanic, and 2% Asian. Compared was similar to that of children without AD. Age, sex,
with the racial composition of the general population race, diagnosing physician, number of lesions at
in the United States (approximately 75% Caucasian, diagnosis, and number of anatomic locations involved
14% African American, 16% Hispanic, 6% Asian) did not change the time to resolution of MC lesions.
(17), our patient population was composed of signif-
icantly fewer Hispanic and Asian children. The
REFERENCES
referral pattern of our patients, the ethnic composi-
tion of our community, and a language barrier, 1. Olsen JR, Gallacher J, Piguet V et al. Epidemiology of
because English is not always the primary language molluscum contagiosum in children: a systematic
review. Fam Pract 2014;31:130–136.
spoken in some Hispanic and Asian households, may 2. Osio A, Deslandes E, Saada V et al. Clinical character-
explain this selection bias. istics of molluscum contagiosum in children in a private
Forty-seven percent of children with MC had a dermatology practice in the greater Paris area, France: a
personal history of AD. This rate is higher than the prospective study in 661 patients. Dermatology
estimated AD prevalence of approximately 10% to 2011;222:314–320.
3. Scholz J, R€ osen-Wolff A, Bugert J et al. Epidemiology
20% in children (18) . Individuals with AD are of molluscum contagiosum using genetic analysis of the
probably more susceptible to MC because of defects viral DNA. J Med Virol 1989;27:87–90.
in immunologic and skin barrier function. Similar to 4. Bugert JJ, Darai G. Recent advances in molluscum
other studies (18,19), children with AD were more contagiosum virus research. Arch Virol Suppl
likely to have more MC lesions than those without 1997;13:35–47.
5. Smith KJ, Yeager J, Skelton H. Molluscum contagio-
AD (p < 0.05), but there was no difference in time to sum: its clinical, histopathologic, and immunohisto-
resolution between children with or without a history chemical spectrum. Int J Dermatol 1999;38:664–672.
of AD. 6. Becker TM, Blount JH, Douglas J et al. Trends in
Although there are various treatment modalities molluscum contagiosum in the United States, 1966–
for the treatment of MC, the decision to treat or not to 1983. Sex Transm Dis 1986;13:88–92.
7. Choong KY, Roberts LJ. Molluscum contagiosum,
treat MC in children should take into account the self- swimming and bathing: a clinical analysis. Australas J
limited nature of MC and the lack of evidence that Dermatol 1999;40:89–92.
supports therapeutic intervention. Our data do not 8. Castilla MT, Sanzo JM, Fuentes S. Molluscum conta-
support the use of destructive, painful methods in the giosum in children and its relationship to attendance at
management of MC infection in immunocompetent swimming-pools: an epidemiological study. Dermatol-
ogy 1995;191:165.
children. 9. Brown J, Janniger CK, Schwartz RA et al. Childhood
There are several limitations to the findings in our molluscum contagiosum. Int J Dermatol 2006;45:93–99.
review. These limitations are selection and recall bias 10. Bayerl C, Feller G, Goerdt S. Experience in treating
and limited sample size, particularly for the subsets of molluscum contagiosum in children with imiquimod
specific treatment modalities. Selection bias includes 5% cream. Br J Dermatol 2003;149(Suppl 66):25–29.
11. Kakourou T, Zachariades A, Anastasiou T et al.
the sociodemographic (urban and in a first-world Molluscum contagiosum in Greek children: a case
country) and socioeconomic (recruitment through a series. Int J Dermatol 2005;44:221–223.
dermatologist, language barrier, telephone in the 12. Arican O. Topical treatment of molluscum contagiosum
household) characteristics of the population studied. with imiquimod 5% cream in Turkish children. Pediatr
Larger controlled trials are needed to validate our Int 2006;48:403–405.
13. Hawley TG. The natural history of molluscum contagio-
experience and to more precisely assess factors influ- sum in Fijian children. J Hyg (Lond) 1970;68:631–632.
encing the resolution of MC, identify subgroups of 14. Niizeki K, Kano O, Kondo Y. An epidemic study of
children with delayed resolution or low resolution molluscum contagiosum: relationship to swimming.
rates, and compare the effectiveness and safety of Dermatologica 1984;169:197–198.
 Basdag et al: Pediatric Molluscum Contagiosum 357

15. Koning S, Bruijnzeels MA, van Suijlekom-Smit LW et 18. Berger EM, Orlow SJ, Patel RR et al. Experience with
al. Molluscum contagiosum in Dutch general practice. molluscum contagiosum and associated inflammatory
Br J Gen Pract 1994;44:417–419. reactions in a pediatric dermatology practice: the
16. Dohil MA, Lin P, Lee J et al. The epidemiology of bump that rashes. Arch Dermatol 2012;148:
molluscum contagiosum in children. J Am Acad 1257–1264.
Dermatol 2006;54:47–54. 19. Simonart T, De Maertelaer V. Curettage treatment for
17. U.S. Census Bureau. American Community Survey molluscum contagiosum: a follow-up survey study. Br J
[online]. 2010. Available at http://factfinder2.census. Dermatol 2008;159:1144–11477.
gov/ Accessed November 17, 2013.