|
There is
insufficient evidence to support existing diagnostic criteria, based
on clinical presentation, for the diagnosis of pelvic inflammatory
disease.
Diagnosis of pelvic inflammatory
disease: time for a rethink.
Simms I, Warburton F, Westrom L.
Sexually
Transmitted Infections 2003;79:491-494.
Summary:
Question
What is the accuracy with which signs and symptoms predict the presence of
pelvic inflammatory disease (PID) as determined using laparoscopy as the
diagnostic gold standard?
Design
This study describes the use of three analytical techniques to compare the
accuracy with which clinical presentation predicted the presence of
laparoscopically diagnosed PID.
Participants
Data from 623 women who attended the department of Obstetrics and
Gynecology, Lund University Hospital, between 1960 and 1969, with first
episodes of suspected PID were analyzed.
Description of Tests and Diagnostic
Standard
The minimum criteria for a diagnosis of PID based on signs and symptoms
were lower quadrant bilateral abdominal or pelvic pain of less than 3
weeks' duration, together with two or more of the following: abnormal
vaginal discharge, fever >38oC, vomiting, menstrual
irregularity, ongoing bleeding, symptoms of urethritis, rectal temperature
>38oC, marked tenderness of pelvic organs on bimanual
examination, adnexal mass, and ESR >15 mm in the first hour.
Laparoscopy was used to confirm a diagnosis of PID.
Main Outcome Measures
The sensitivity and specificity of each sign or symptom for the diagnosis
of PID as determined by laparoscopy were calculated. The likelihood ratio
and post-test probability of each sign or symptom were compared with the
pretest probability of PID. The combination of signs and symptoms that
most effectively predicted the presence of PID were determined.
Main Results
Laparoscopy confirmed PID in 494 (79%) of 623 women. The sensitivity of
each sign or symptom for the diagnosis of PID compared to laparoscopy
ranged from 10% to 74% except for tenderness of pelvic organs on bimanual
examination (99%) and ESR (81%). The specificity of each sign or symptom
for the diagnosis of PID compared to laparoscopy ranged from 0.01% to 77%
except for proctitis symptoms (92%) and vomiting (88%). The post-test
probability of PID was not significantly different from the pretest
probability for any sign or symptom. The likelihood ratios ranged from
0.97 to 1.73. Three variables combined significantly influenced the
prediction of the presence of PID: ESR, fever, and adnexal tenderness. The
combination of these three signs and symptoms compared with laparoscopy
for the diagnosis of PID is shown in the table. These variables correctly
classified 65% of patients with laparoscopically diagnosed PID.
Authors' Conclusions
The Lund and other available datasets
provide insufficient evidence to support existing PID diagnostic
guidelines and have limitations for the formulation of new ones. A new
evidence base is urgently needed, which will require either a new
investigation of the association between clinical presentation and PID
based on a laparoscopic gold standard, or the development of new
diagnostic tests.
Source of funding:
Not given
For correspondence: Ian Simms, HPA
Communicable Disease Surveillance Centre, 61 Colindale Avenue, London NW9
5EQ, UK. E-mail address: ian.simms@hpa.org.uk.
|
.gif){kind=small}