Nonspecific Vaginitis Following Sexual Abuse in Children

(1)

Nonspecific

Vaginitis

Following

Sexual

Abuse

in Children

Margaret

R. Hammerschlag,

MD, Marinella

Cummings,

BS,

Brinda

Doraiswamy,

MD,

Phyllis

Cox,

RN, and

William

M. McCormack,

MD

From the Division of Immunology and Infectious Diseases, Department of Pediatrics, and Division of Infectious Diseases, Department of Medicine, State University of New York, Downstate Medical Center, Brooklyn

ABSTRACT. Nonspecific vaginitis, one of the most com-mon causes of vaginitis in adults, is a polymicrobial infection in which vaginal anaerobes act synergistically with Gardnerella vaginalis. The diagnosis is made by examination of the vaginal secretions for clue cells, the development of a fishy odor after the addition of 10%

KOH to vaginal secretions, and a vaginal pH greater than 4.5. To determine whether nonspecific vaginitis occurs in sexually abused children, we obtained vaginal washes from 31 abused and 23 nonabused children, 2#{189}to 13 years of age. A child was considered to have definite nonspecific vaginitis if her wash contained both clue cells and odor; she was considered to have possible nonspecific vaginitis if her wash contained either clue cells or odor. We did not use vaginal pH as a diagnostic criterion

because the normal range has not been standardized in prepubertal girls. Possible nonspecific vaginitis (odor only) was found in only 1/23 (4%) of nonabused children. This girl was asymptomatic and findings from her ex-amination were normal. Only one of the 31 abused chil-dren had possible nonspecific vaginitis (odor) detected at the initial examination, less than 48 hours after the

episode of abuse, whereas 4/31 (13%) developed definite, and 4/31 (13%) possible nonspecific vaginitis at the fol-low-up visit more than seven days after the episode of abuse or rape. Five of these eight girls developed either a new vaginal discharge or dysuria; three were treated with

metronidazole with resolution of their symptoms and

reversion of the vaginal wash to normal. These findings suggest that nonspecific vaginitis is uncommon in normal children and that it can be acquired after sexual abuse. Nonspecific vaginitis was the most frequent cause of vaginitis in the abused girls in this study. Examination of a vaginal wash for clue cells and odor should be part of the evaluation of sexually abused girls who develop vulvovaginal symptoms. Pediatrics 1985;75:1028-1031;

vaginitis, sexual abuse, vaginal dLscharge.

Received for publication Feb 21, 1984; accepted June 27, 1984.

Reprint requests to (M.R.H.) Department of Pediatrics, Box 49,

Downstate Medical Center, 450 Clarkson Aye, Brooklyn, NY 11203.

The evaluation of a child suspected of being

sexually abused often includes only cultures for

Neisseria gonorrhoeae and a serologic test for

syph-ilis.’ However, one would expect these children to

be at risk for acquiring other infections prevalent

in the adult population. One of these infections is

nonspecific vaginitis. The role of sexual

transmis-sion in the acquisition of nonspecific vaginitis is

not well established.2’3 Nonspecific vaginitis is a

synergistic infection involving Gardnerella vaginalis

and various anaerobic bacteria including

Bacte-roides and Peptococcus species.4 The diagnosis is

made by examination of the vaginal secretions for

clue cells, development of a characteristic odor after the addition of 10% KOH to the vaginal secretions, and a vaginal pH greater than 4#{149}5#{149}3The prevalence of nonspecific vaginitis in children, as determined

by these criteria, is unknown. As part of a study of

infections in sexually abused children, we examined

vaginal washes from a group of abused and

nona-bused girls for evidence of nonspecific vaginitis.

METHODS

Patients

Children admitted to Kings County Hospital,

Brooklyn, NY, for the evaluation of suspected

sex-ual abuse were enrolled in the study. All children

with suspected sexual abuse are routinely admitted

for medical and social evaluation. A group of

healthy girls attending a well-child clinic were also

studied. After an explanation of the procedures to

be performed, written informed consent was

ob-tamed from the parents or legal guardian.

The abused children were examined as soon as

possible after admission, and two or more weeks

(2)

ARTICLES

1029

and an examination completed by one of the

inves-tigators. The nonabused children were only studied

once. The results were recorded on a standard form.

Specimen Collection and Microbiologic Methods

Vaginal washes were obtained by injecting 2 mL

of sterile saline into the vagina with a transfer

pipette. The specimen was aspirated from the

va-gina and transferred to a 1-dram, screw-top vial

and immediately taken to the laboratory. One drop

of fluid was placed on a slide, a drop of 10% KOH

was added, and the fluid was examined for the

presence of a fishy odor. Another drop of the wash

was examined under the microscope for the

pres-ence of clue cells (epithelial cells heavily studded with bacteria) and trichomonads. The investigator

examining the washes did not know the clinical

status of the patient.

Vaginal specimens were obtained with sterile

cot-ton-tipped swabs which were broken into a sterile

glass vial containing 2.0 mL of Mycoplasma

trans-port media and examined for Mycoplo,sma hominis

and Ureaplasma urealyticum by means of

previ-ously described methods.5 A second vaginal swab

culture was broken into a sterile glass vial

contain-ing 1 mL of Feinberg-Whittington media6 which

was incubated at 36.5#{176}C.The culture was examined

daily, excluding Saturday and Sunday, for seven

days for motile trichomonads. Rectal and vaginal

swab cultures were examined for isolation of

Chkz-mydia trachomatis. Cultures were obtained with calcium alginate swabs and immersed in a collection

medium consisting of 0.2 M sucrose in a 0.02 M

phosphate buffer with 10 g/mL of gentamicin and

1 tg/mL of amphotericin B. The specimens were

kept at 4#{176}Cfor up to 24 hours, then cultured or

frozen at -70#{176}Cuntil examined. Isolation of C

trachomatis was performed using

cycloheximide-treated McCoy cells grown in 96-well microtiter

plates.7 Negative specimens were inoculated a

sec-ond time.

RESULTS

Paired vaginal wash specimens were obtained

from 31 abused children. A single specimen was

obtained from 23 children without a history of

recent sexual abuse. The mean ages of the abused

children and control subjects were 8.1 years (range

3.5 to 13 years) and 7.8 years (range 2.5 to 12 years),

respectively. All the children were black. Nine of

the abused girls (29%) and three (13%) of the

control subjects were postmenarcheal. One abused

girl had a history of voluntary sexual activity.

Vag-ma!

penetration as determined by history or

phys-ical examination occurred in 18 (58%) ofthe abused

girls. Physical evidence included signs of trauma

and the presence of semen. Two girls enrolled as

control subjects had a history of sexual abuse 3

years previously. None ofthe other control children

had a history of known sexual abuse and none gave

a history of voluntary sexual activity.

Diagnosis

of Nonspecific

Vaginitis

The diagnosis of nonspecific vaginitis was

con-sidered definite if the vaginal wash contained both

clue cells and odor. If the vaginal wash contained either clue cells or odor, the diagnosis of nonspecific

vaginitis was considered to be possible. Definite

nonspecific vaginitis was not detected in any of the control children. Possible nonspecific vaginitis was

detected in only 1/23 (4%) control children. This

child was clinically asymptomatic but had only odor

present in her vaginal wash specimen. There was

no history of known sexual abuse.

One abused girl had possible nonspecific vaginitis

(odor) at the initial visit. Her specimen was

ob-tamed less than 48 hours after a single episode of

abuse by a friend of her father. She was

asympto-matic and a follow-up vaginal wash specimen

ob-tamed four weeks later was normal.

Four girls who had normal washes initially had

definite nonspecific vaginitis when examined seven

or more days after the episode of sexual abuse. In

addition, four girls had possible nonspecific

vagi-nitis at the follow-up examination. The girls ranged

in age from 5 to 12 years (mean 9 years). Three

were postmenarcheal (all were 12 years of age).

Three of these girls were victims of rape by a

stranger, and five had been raped by a male relative. Vaginal penetration occurred in all eight girls.

As shown in the Table, four of these eight (50%)

girls developed a new vaginal discharge and one

developed dysuria. The discharges were thin and

white; two had an obvious foul odor. An abnormal

vaginal discharge was present in 5/22 (23%) abused girls without evidence of nonspecific vaginitis. In four of these girls, the discharge was present at the

initial examination. Thus, 4/8 (50%) girls with

definite or possible nonspecific vaginitis developed

a new vaginal discharge or dysuria, compared with

1/22 (4.5%) abused girls without nonspecific

vagi-nitis (P = .01, Fisher exact test). Two girls with

definite nonspecific vaginitis and one with possible

nonspecific vaginitis who developed a new vaginal

discharge were treated with metronidazole, which

resulted in resolution of their symptoms and rever-sion of the vaginal wash to normal.

Other Infections

Trichomonas vaginalis was not isolated from the

at Viet Nam:AAP Sponsored on September 7, 2020 www.aappublications.org/news

(3)

TABLE. Vaginal Wash Specimen Findings and Clinical Symptoms in Eight Abused Children

Patient No.

Age (yr)

Vaginal Wash S pecimen Findings Symptoms

#{149}. #{149}.

Initial Visit Follow-up V.isit. Initial. . Visit. Follow-up

Clue Cells Odor Clue Cells Odor

1 7 0 0 0 + None None

2 12 0 0 + 0 None Discharge

3 12 0 0 + + None Discharge*

4 12 0 0 + + None Discharge*

5 7 0 0 + + None None

6 9 0 0 0 + None Discharge5

7 8 0 0 + + None None

8 5 0 0 + 0 None Dysuria

S Treated with metronidazole.

vaginal cultures of any of the control subjects, but it was present in one abused child with definite nonspecific vaginitis at the follow-up visit and an-other abused child with no evidence of nonspecific

vaginitis. Similarly, gonococcal infection was not

present in any of the control subjects. N

gonor-rhoeae was isolated from 3/22 (13.6%) abused

chil-dren without nonspecific vaginitis. All three

chil-dren were first seen with complaints of a vaginal

discharge. C trachomatis was isolated from the

vag-inal cultures of 3/23 (13%) control subjects. Two of

these girls were sisters who had a history of previous

sexual abuse 3 years previously. C trachomatis was

also isolated from the rectal culture obtained at the

follow-up visit of one abused girl with possible

nonspecific vaginitis. The data concerning C

tra-chomatis and sexual abuse have been presented in detail elsewhere.8

M hominis and U urealyticum were each isolated

from the vaginal cultures of 2/23 (9%) control

subjects. Although both M hominis and U

urealy-ticum were isolated more frequently from the

abused children compared with control subjects (for

M hominis,

x2

= 3.37, P < .1 > .05, for U

urealyti-cum, 2 = 6.5, P < .025), neither organism was

isolated more frequently from the girls with

non-specific vaginitis in comparison with the other

abused children. M hominis was isolated from 1/4

(25%) girls with definite nonspecific vaginitis and

2/5 (40%) with possible nonspecific vaginitis, and

6/22 (27%) abused girls without nonspecific

vagi-nitis

(x2

0.4, 2 df, P > .75). U urealyticum was isolated from the vaginal cultures of 2/4 (50%) girls

with definite nonspecific vaginitis and 3/5 (60%)

with possible nonspecific vaginitis, and 8/22 (36%)

abused girls without nonspecific vaginitis

(x2

= 0.9,

2df,P>.5).

DISCUSSION

The term nonspecific vaginitis is actually a

mis-nomer since it is a specific clinical entity, not a

nonspecific genital infection. Efforts to ascribe

nonspecific vaginitis solely to infection with G vag-inalis may have contributed to the confusion. Since 1955, when Gardner and Dukes9 described nonspe-cific vaginitis and identified G vaginalis (then called

Haemophilus vaginalis), there have been many

studies that have and have not found a relationship

between isolation of the organism and nonspecific

vaginitis.’#{176}’3 Subsequent studies4 have suggested that nonspecific vaginitis is the result of the syn-ergistic interaction of G vaginalis and various

an-aerobic bacteria. In 1978, Pheifer et al”

demon-strated that nonspecific vaginitis responded to ther-apy with metronidazole.

The following criteria have been proposed for the

diagnosis of nonspecific vaginitis: (1) the presence

of a thin homogeneous vaginal discharge, (2) a

vaginal pH greater than 4.5, (3) the release of a

fishy odor after the addition of 10% KOH to one

drop of vaginal fluid, and (4) the presence of clue

cells on a wet mount of the vaginal fluid.2’3”4 These

are all simple procedures that can be performed in

the office, in the clinic, or on the ward.

Nonspecific vaginitis is one of the most common

causes of vaginitis in adult women. The prevalence

has been reported to range from 25% to 36%.23,14

More than half of adult women with nonspecific

vaginitis are asymptomatic.2 Some data suggest that acquisition of nonspecific vaginitis is related

to sexual activity.2’3 G vaginalis may be isolated

from the urethra of approximately 80% of male

partners of women with nonspecific vaginitis.”2

There are few data on the prevalence of G

vagi-nalis or of nonspecific vaginitis among

preadoles-cent children. Hammerschlag et al,’5 in a study of

unselected girls attending a walk-in clinic, isolated

G vaginalis from vaginal cultures in 14%. The

pres-ence of the organism was not associated with

vagi-nitis. Paradise et al’6 failed to isolate G vaginal is

from 54 premenarcheal girls with vulvovaginitis

and 40 healthy control subjects. Neither of these

two studies used selective media, or examined

vag-inal secretions for evidence of nonspecific vaginitis.

(4)

ARTICLES

1031

secretions of children have been examined for clue

cells or odor.

By examining vaginal washes of 31 abused

chil-dren, we detected definite nonspecific vaginitis in

four and possible nonspecific vaginitis in four. Only

the presence of clue cells and the presence of odor

were used as diagnostic criteria because of the lack

of standards for vaginal pH in children. Clue cells

were not found in the vaginal washes from any of

the control subjects and were not present in the

initial washes from any of the abused children. The

presence of odor was less specific because it was

detected in the vaginal wash specimen of one

con-trol subject and in the initial specimen of one

abused child.

Three of the eight children who developed

defi-nite or possible nonspecific vaginitis were

asymp-tomatic. The other five girls complained of either a

new vaginal discharge (four girls) or of dysuria (one girl). One of these children also had trichomoniasis,

which could also have accounted for her symptoms.

Trichomoniasis may be associated with an

abnor-mal odor,2’3 however, it is not associated with the

presence of clue cells, which were seen in this child’s specimen.

Some studies have suggested that infection with

M hominis is also associated with nonspecific

vag-initis.’7 In one study of adult women,’4 M hominis

was not isolated more frequently in women with

nonspecific vaginitis compared with control

sub-jects. Although isolation of M hominis occurred

more frequently in the abused children than in the

control subjects, M hominis was not isolated sig-nificantly more frequently in the girls with nonspe-cific vaginitis.

The development of nonspecific vaginitis was

temporally related to the episode of abuse in our

patients. In eight girls with definite or possible

nonspecific vaginitis, the abnormal findings were

present only in the specimens obtained at the

fol-low-up examination. This observation suggests that

nonspecific vaginitis was sexually acquired in these

children. Nonspecific vaginitis was also the most

common cause of a new vaginal discharge in our

population.

In conclusion, the development of a new vaginal

discharge in a child who has been raped should

suggest nonspecific vaginitis. Examination of a

vag-inal wash specimen for clue cells and odor should

be part of the routine medical evaluation of sexually

abused children who develop vulvovaginal

symp-toms.

REFERENCES

1. Jones JG: Sexual abuse of children: Current concepts. Am J Dis Child 1982;136:142-146

2. Blackwell A, Barlow D: Clinical diagnosis of anaerobic va-ginosis (nonspecific vaginitis). Br J Vener Dis 1982;58:387-393

3. Amsel R, Totten PA, Spiegal CA, et al: Nonspecific vaginitis. Am J Med 1982;74:14-22

4. Spiegal CA, Amsel R, Eschenbach, D, et a!: Anaerobic bac-teria in nonspecific vaginitis. N Engl J Med 1980;303:601-607

5. McCormack WM, Almeida PC, Bailey PE, et al: Sexual activity and vaginal colonization with genital mycoplasmas.

JAMA 1982;221:1375-1377

6. Feinberg JG, Whittington M: A culture medium for

Trich-omonas vaginalis Donn#{233}and species of Candida. J Clin Pat/wI 1957;10:327-329

7. Yoder BL, Stamm WEK, Koester CM, et al: Microtest

procedure for isolation of Chkzmydia trachomati.s. J Clin Microbiol 1981;13:1036-1039

8. Hammerschlag MR, Doraiswamy B, Alexander ER, et al: Are rectogenital chlamydial infections a marker of sexual abuse in children? Pediatr Infect Die 1984;3:100-104

9. Gardner HL, Dukes CD: Haemophilus vaginalis vaginitis.

Am J Obstet Gynecol 1955;69:962-976

10. McCormack WM, Hayes CH, Rosner B, et al: Vaginal colonization with Corynebacterium vaginale (Haemophilus vaginalis). J Infect Dis 1977;136:740-745

11. Pheifer TA, Forsyth PS, Durfee MA, et al: Nonspecific vaginitis: Role of Haemophilus vaginalis and treatment with metronidazole. N EngI J Med 1978;298:1429-1434

12. Ratnam S, Fitzgerald BL: Semiquantitative culture of non-specific vaginitis. J Clin Microbiol 1983;18:344-346

13. Dattani IM, Gerken A, Evans BA: Aetiology and manage-ment of nonspecific vaginitis. Br J Vener Dis 1982;58:32-35

14. Hill LH, Ruparella H, Embil JA: Nonspecific vaginitis and other genital infections in three clinic populations. Sex Transm Dis 1983;10:114-118

15. Hammerschlag MR, Alpert S, Rosner I, et al: Microbiology ofthe vagina in children: Normal and potentially pathogenic organisms. Pediatrics 1978;62:57-62

16. Paradise JE, Campos JM, Friedman HM, et al: Vulvovagi-nitis in premenarcheal girls: Clinical features and diagnostic evaluation. Pediatrics 1982;70:193-198

17. McCormack WM, Moller BR, Mardh P-A: Mycoplasma

hominis-A human pathogen. Sex Transm Dis 1983;10:160-165

(5)

1985;75;1028

Pediatrics

William M. McCormack

Margaret R. Hammerschlag, Marinella Cummings, Brinda Doraiswamy, Phyllis Cox and