American Journal of Emergency Medicine (2012) 30, 1114–1117

www.elsevier.com/locate/ajem

Original Contribution

Laboratory-confirmed gonorrhea and/or chlamydia

rates in clinically diagnosed pelvic inflammatory disease
and cervicitis
Aaron M. Burnett MD a,⁎, Christopher P. Anderson MPH b , Michael D. Zwank MD a
a
Department of Emergency Medicine, Regions Hospital, St Paul, MN 55101, USA
b
HealthPartners Research Foundation, Minneapolis, MN 55440, USA

Received 9 June 2011; revised 13 July 2011; accepted 15 July 2011

Abstract
Objective: The aim of this study was to determine the rates of laboratory confirmed gonorrhea (GC) and
chlamydia (CT) in emergency department (ED) patients with pelvic inflammatory disease (PID) and
cervicitis who were diagnosed clinically and treated empirically. A secondary goal examines which
clinical criteria were present in patients with PID testing positive for GC/CT.
Methods: We conducted a retrospective chart review of all ED patents diagnosed with PID or cervicitis
during a 40-month period (January 2007–March 2010). Charts were reviewed for laboratory-confirmed
GC or CT. For patients with positive GC or CT studies, the presence of key clinical criteria used in the
diagnosis of PID was tallied.
Results: A total of 1469 patients were diagnosed with cervicitis and 343 with PID. Of these patients,
27 (1.8%) of 1469 and 15 (4.4%) of 343 were GC positive, and 136 (9.3%) of 1469 and 34 (10%) of
343 were CT positive. Twenty-six cervicitis (1.8%) and 9 PID (2.6%) patients were positive for both
infections. One hundred eighty-nine cervicitis (13%) and 58 PID (17%) patients were positive for at
least 1 sexually transmitted infection. Of the 58 patients with PID with laboratory-confirmed GC/CT,
the following clinical criteria were present: abdominal pain, 58 of 58; abdominal tenderness, 50 of 58;
cervical discharge, 47 of 58; cervical motion tenderness, 46 of 58; adnexal tenderness, 32 of 58;
vaginal bleeding, 8 of 58; and fever, 2 of 58. Ultrasound was preformed in 27 (47%) of 58 GC/CT-
positive patients with PID, with findings suggestive of PID in 12 (44%) of 27 ultrasounds. One
hundred percent of abnormal ultrasounds were associated with positive GC and/or CT results.
Conclusion: There is a generally low prevalence of GC and CT in this patient population diagnosed
with cervicitis or PID. There is a very low prevalence of coinfection.
© 2012 Elsevier Inc. All rights reserved.

1. Introduction tubes, and potentially into the peritoneal cavity [1]. It is a

Pelvic inflammatory disease (PID) is an infection that
starts in the vagina and ascends into the uterus, fallopian
⁎ Corresponding author. Tel.: +1 651 254 3666; fax: +1 651 254 5216.
E-mail address: aaron.m.burnett@healthpartners.com (A.M. Burnett).
polymicrobial process classically associated with a sexually
transmitted infection such as Neisseria gonorrhea (GC)
or Chlamydia trachomatis (CT). Prior studies have docu-
mented the rates of GC/CT recovered from tubo-ovarian
abscesses at laparoscopy as high as 77% [2]. This
association with GC/CT is reflected in the choice of

0735-6757/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.ajem.2011.07.014
Laboratory confirmed GC/CT in PID
antibiotics recommended by the Centers for Disease
Control and Prevention (CDC) for empiric treatment of
PID. Often, PID is preceded by an isolated cervicitis that
can ascend into the upper genital tract during menstruation
or douching [3,4]. Currently, it is estimated that 750 000
American women experience acute PID each year, and
75 000 become infertile because of it [5]. A large portion
of ectopic pregnancies are caused by fallopian tube
scarring from prior PID. The literature describes primary
epidemics of GC and CT, followed by secondary
epidemics of PID. The term tertiary epidemic has been
used to refer to the expected surge in ectopic pregnancies
after the PID outbreak [6].
The CDC sets guidelines for the diagnosis of PID
(Table 1) [7]. The classic doctrine in medical education
stresses the need for a low threshold for diagnosing and
treating PID. In general practice, a female with lower
abdominal pain and tenderness on bimanual pelvic
examination without an alternate explanation (appendicitis,
diverticulitis, cystitis, etc) is often given a clinical
diagnosis of PID. However, cervical motion, uterine, and
adnexal tenderness are physical examination findings with
a significant subjective component. The criterion standard
for PID diagnosis is identification of purulent salpingitis on
laparoscopy; however, this is a highly invasive and
resource-intense procedure that is very infrequently
performed in the modern era of medicine [8]. The
challenge for the emergency physician lies in the fact
that most of the diagnostic criteria are clinically based,
subjective, or nonspecific. It is known that the clinical
diagnosis of PID is imprecise [9].
Currently, polymerase chain reaction tests of cervical
or vaginal samples are most commonly performed to
evaluate GC or CT infection. Polymerase chain reaction–
based tests are a form of nuclear acid amplification
technology with sensitivities and negative predictive
values of more than 95% for both GC and CT; however,
the results are not immediately available (turnaround time
Table 1 CDC diagnostic criteria for PID
Inclusion criteria: sexually active with lower abdominal or
pelvic pain with no alternate etiology
Minimum criteria: cervical motion tenderness or uterine
tenderness or adnexal tenderness
Additional criteria to increase specificity
Oral temperature N101°F (38.3°C)
Abnormal cervical or vaginal mucopurulent discharge
Presence of abundant numbers of WBC on saline microscopy
of vaginal secretions
Elevated ESR
Elevated CRP
Laboratory documented cervical GC/CT infection
WBC, white blood cells; CRP, C-reactive protein; ESR, erythrocyte
sedimentation rate.
1115
is 2-3 days at many institutions) [10,11]. This limits the
utility of this diagnostic test for the emergency physician
who is deciding on an empiric antibiotic treatment for an
individual patient.
Determining if an individual patient has subjective
physical examination findings that meet the CDC case
definition of PID can be a challenge. Even when the CDC
minimum criteria are met, the specificity for PID is only
22% [9]. Undertreatment of PID may yield a significant
morbidity to the individual patient; however, false-positive
diagnoses may contribute to antibiotic resistant organisms. In
2006, the CDC discontinued the recommendation for using
quinolones in PID treatment, given the unacceptably high
rates of quinolone resistant GC.
There have been no studies that have looked at the
correlation between clinically diagnosed PID or cervicitis
and positive laboratory studies for GC/CT in the
emergency department (ED) population. Prior studies
have examined the rate of positive culture results from
laparoscopically drained tubo-ovarian abscesses; however,
this is not very useful to the emergency physician at the
bedside [12]. Our study attempts to add to the
knowledge base by examining the rates of laboratory-
confirmed GC/CT in patients clinically diagnosed with
PID or cervicitis in an urban ED. Secondary goals are to
determine which of the clinical criteria were present in
the individuals diagnosed with PID who tested positive
for GC/CT.
2. Methods
This study is a retrospective chart review. It was approved
by our institutional review board. We identified as key
variables the results of laboratory studies for GC/CT that
were performed on patients clinically diagnosed with PID or
cervicitis. Secondary variables that we studied included
physical examination findings in these patients and results of
radiographic and laboratory investigations.
This study was conducted in the ED of an academic,
urban trauma center that hosts an emergency medicine
(EM) residency. Midlevel providers including EM and off-
service residents as well as physician assistants perform
direct patient care under the supervision of board-certified
EM physicians.
We enrolled a convenience sample of consecutive patients
diagnosed with PID or cervicitis in our ED during a 40-
month period. Inclusion criteria were patients of any age who
were diagnosed with either cervicitis or PID in the ED and
had a laboratory study for GC/CT.
Individual charts were reviewed for laboratory-confirmed
GC or CT. Samples were obtained from the endocervical or
vaginal canal and were analyzed by polymerase chain
reaction. Occasional urine samples were used for analysis,
which is consistent with the Food and Drug Administration–
1116
approved sampling method for the tests. For patients' with a
clinical diagnosis of PID and positive laboratory studies,
further chart review to identify additional clinical criteria of
PID was performed. For any patient with laboratory-positive
PID who had a radiographic procedure performed at the
discretion of the treating physician, the results were reviewed
for abnormalities suggestive of PID.
Descriptive statistics were used, with results reported as
absolute numbers and percentages. This will allow us to
report the prevalence of laboratory-confirmed GC/CT in our
study population. A power analysis was not performed, as
this was a purely descriptive study.
3. Results
A total of 1469 patients were diagnosed with cervi-
citis, and 343 patients were diagnosed with PID. Of
the patients with cervicitis, 1.8% (27/1469) were GC
positive and 9.3% (136/1469) were CT positive. Patients
diagnosed with PID had a 4.4% (15/343) rate of positive
GC and 10% (34/343) rate for CT. Coinfection with both
organisms was documented in 1.8% (26/1469) of the
patients with cervicitis and 2.6% (9/343) of the patients
with PID (see Fig. 1). For the 58 patients diagnosed with
PID who also had positive laboratory results for either
organism, the following physical examination findings
were noted (see Fig. 2). White blood cell counts were
checked in 25 patients with laboratory-confirmed PID and
ranged from 2 to 22, with an average of 12.1.
Ultrasonography was obtained in 47% (27/58) of the
GC/CT-positive patients with PID and demonstrated
findings concerning for PID in 44% (12/27), with 1
frank abscess. Of the 12 suspicious ultrasounds, 5 patients
were positive for CT, 3 for GC and 4 had coinfection with
both organisms.
Fig. 1 Infection rates among patients diagnosed with PID.
A.M. Burnett et al.
Fig. 2 Signs and symptoms among patients with laboratory-
confirmed PID.
4. Discussion
Pelvic inflammatory disease is a serious infection that
can lead to future infertility and ectopic pregnancy. Due to
the complications that can arise from untreated infections,
guidelines that encourage aggressive treatment for patients
who present with signs and symptoms of PID have been
established. Unfortunately, there is currently no physical
examination finding or radiographic study with sufficiently
high sensitivity to exclude this diagnosis. Although GC and
CT are the most common organisms identified, PID is a
polymicrobial process. The consensus in the literature has
been to recommend treating for PID in patients with even a
moderate pretest probability. Inherent in this argument is an
acceptance of overtreatment of true-negative cases; how-
ever, the widespread use of antibiotics also has complica-
tions. Microbial resistance to antibiotics has led to the
emergence of methicillin-resistant Staphylococcus aureus
and vancomycin-resistant Escherichia coli. In 2007, the
CDC recommended that the entire quinolone class of
antimicrobials be removed from the armamentarium
targeted at GC due to the development of resistant
gonococci. The spectrum of agents available for PID
treatment was, thus, narrowed.
This study demonstrates that there is a low prevalence of
laboratory-confirmed GC and CT in patients diagnosed with
PID at our institution. In patients with PID, we observed a
higher rate of confirmed GC and CT as compared with the
patients diagnosed with cervicitis. Cervicitis, however, can
be treated with a single dose of antibiotics, whereas PID
requires up to 14 days of treatment. Examination findings
including abdominal tenderness, cervical discharge, and
cervical motion tenderness were closely associated with
laboratory-confirmed GC/CT. Leukocytosis was not a
reliable predictor of positive laboratory results in our cohort
Laboratory confirmed GC/CT in PID
of patients with PID. An ultrasound deemed by the treating
emergency physician as suspicious for PID was useful in
predicting positive GC/CT laboratory results 100% of the
time, suggesting that an abnormality on ultrasound is of
diagnostic utility.
Given the emergence of drug-resistant gonococci com-
bined with the low rates of GC in our patients clinically
diagnosed with PID, it gives us pause to reflect on our
prescribing practices. The risk to individual patients who are
undertreated is clearly high; however, the risk to the general
population from polydrug-resistant microbes must also be
considered. The rates of coinfection were noted to be 2.6%
for the group with PID, yet standard practice is to cover
patients with PID for both GC and CT. As new multidrug-
resistant bacteria emerge, our therapeutic options are already
becoming narrowed.
To assist the emergency clinician in the diagnosis of
PID, our specialty should focus our efforts on quantifying
the sensitivity and specificity of tests that are used in the
workup of patients with suspected PID. Novel radio-
graphic approaches should be considered. The primacy of
transvaginal ultrasound in the diagnosis of tubo-ovarian
abscess has recently been called into question, with
reported sensitivities and specificities of 56% to 93% and
86% to 98%, respectively [13]. A recent study evaluating
the utility of magnetic resonance imaging in the diagnosis
of PID found that the radiographic diagnosis agreed with
the laparoscopic diagnosis in 20 of 21 patients with lapa-
roscopically confirmed PID [14]. A reevaluation of which
combinations of laboratory, physical examination and
radiographic studies may be useful for raising the posttest
probability of PID is in order. We should explore the
response to antibiotics of patients with culture-confirmed
infections to determine if there is a predictable course
over the first several days. Perhaps, starting a 14-day
course of empiric antibiotics may not require finishing
the entire course if we could identify factors that would
allow a physician at a follow-up visit to safely say that
PID is unlikely.
The findings in this study should cause us to reexplore
our current approach to PID and to stimulate new research
in this area. Ways in which we can more narrowly
focus our treatment regimens will be needed, and the role
of empiric antibiotics will continue to evolve. Consensus
guidelines that are developed to assist clinicians will
need to reflect the best available science. Risks to the
individual patient will need to be considered in the context
of the public health threats of multidrug-resistant bacteria.
We hope that the data presented here contribute to the
dialogue and ongoing reassessment of what is the best
medical practice.
1117
5. Conclusion
There is a generally low prevalence of GC and CT in this
ED population diagnosed with cervicitis or PID. There is a
very low prevalence of coinfection.
Acknowledgments
Health Partners Institute for Medical Education provided
an internal grant of $2000 for this project.
References
[1] Paavonen J, Valtonen V. Chlamydia trachomatis as a possible cause
of peritonitis and perihepatitis in a young woman. Br J Vener Dis
1980;56:341.
[2] Soper DE, Brockwell NJ, Dalton HP, et al. Observations concerning
the microbial etiology of acute salpingitis. Obstet Gynecol Surv
1994;49:689.
[3] Ness RB, Soper DE, Holley RL, et al. Douching and endometritis:
results from the PID Evaluation and Clinical Health (PEACH) study.
Sex Transm Dis 2001;28:240.
[4] Sweet RL, Blankfort-Doyle M, Robbie MO, et al. The occurrence of
chlamydial and gonococcal salpingitis during the menstrual cycle.
JAMA 1986;255:2062.
[5] Center for Disease Control and Prevention. Pelvic inflammatory disease
fact sheet. Available at: http://www.cdc.gov/std/PID/STDFact-PID.htm.
Accessed November 10, 2010.
[6] Simms I, Stephenson J. Pelvic inflammatory disease epidemiology: what
do we know and what do we need to know? Sex Transm Infect 2000;76:80.
[7] Centers for Disease Control and Prevention. Morbidity and mortality
weekly report: update to CDC's sexually transmitted diseases
treatment guidelines, 2006: Fluoroquinolones no longer recommended
for treatment of gonococcal infections. Morbidity and Mortality
Weekly Report 2007;56(14):332-6.
[8] Lareau SM, Beigi RH. Pelvic inflammatory disease and tubo-ovarian
abscess. Infect Dis Clin North Am 2008;22:693-708.
[9] Peipert JT, Ness RB, Blume J, et al. Clinical predictors of endometritis
in women with symptoms and signs of pelvic inflammatory disease.
Obstet Gynecol Surv 2001;56:468.
[10] Thejls H, Gnarpe J, Gnarpe H, et al. Expanded gold standard in the
diagnosis of Chlamydia trachomatis in a low prevalence population:
diagnostic efficacy of tissue culture, direct immunofluorescence,
enzyme immunoassay, PCR and serology. Br Med J 1994;70:300.
[11] Van Dyck E, Ieven M, Pattyn S, et al. Detection of Chlamydia
trachomatis and Neisseria gonorrhoeae by enzyme immunoassay,
culture, and three nucleic acid amplification tests. J Clin Microbiol
2001;39:1751.
[12] Landers DV, Sweet RL. Tubo-ovarian abscess: contemporary
approach to management. Rev Infect Dis 1983;5:876-84.
[13] Lee DC, Swaminathan AK. Sensitivity of ultrasound for the diagnosis
of tubo-ovarian abscess: a case report and literature review. J Emerg
Med 2010.
[14] Tukeva TA, Aronen HJ, Karjalainen PT, et al. MR imaging in
pelvic inflammatory disease: comparison with laparoscopy and US.
Radiology 1999;210:209.