Journal Pre-proof

Risk factors for recurrent Pelvic Inflammatory Disease

Myriam Safrai, Amihai Rottenstreich, Asher Shushan, Ronit Gilad,

Avi Benshushan, Gabriel Levin

PII: S0301-2115(19)30510-X
DOI: https://doi.org/10.1016/j.ejogrb.2019.11.004
Reference: EURO 11057

To appear in: European Journal of Obstetrics & Gynecology and Reproductive

Biology

Received Date: 12 September 2019

Revised Date: 4 November 2019
Accepted Date: 6 November 2019

Please cite this article as: Safrai M, Rottenstreich A, Shushan A, Gilad R, Benshushan A,
Levin G, Risk factors for recurrent Pelvic Inflammatory Disease, European Journal of
Obstetrics and amp; Gynecology and Reproductive Biology (2019),
doi: https://doi.org/10.1016/j.ejogrb.2019.11.004

This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.

© 2019 Published by Elsevier.

Title: Risk factors for recurrent Pelvic Inflammatory Disease

Authors: Myriam Safrai, MD¹±, Amihai Rottenstreich, MD¹±, Asher Shushan, Prof¹, Ronit
Gilad, MD¹, Avi Benshushan, MD¹, and Gabriel Levin, MD¹.

¹ Department of Obstetrics and Gynecology, Hadassah-Hebrew University Medical Center,

Jerusalem, Israel.
± Equal contribution

of
Corresponding author.

ro
Myriam Safrai M.D.
Department of Obstetrics and Gynecology,

-p
Hadassah-Hebrew University Medical Center, Ein-Kerem,
P.O. Box 12000,
re
Jerusalem, Israel
Zip code 9112001
lP

Tel: +972-50-4048813;
Fax: +972-2-6777541;
na

Email: myriamsafrai@gmail.com

Clinical article
ur

Word count: abstract: 312, Text: 2384

Jo

Abstract
Objective: Pelvic inflammatory disease (PID) is a common infection which can result in severe
long term morbidity, such as chronic pelvic pain and infertility. The morbidity increases in
correlation to the number of PID events. Our study aim to assess the risk factors for recurrence of
pelvic inflammatory disease.
Methods: A retrospective case control study was conducted using data for all women who were
admitted to a tertiary medical center for a recurrent PID over a duration of 15 years. Women who
had a recurrent PID were compared to women admitted for PID treatment without further
recurrence. Forward stepwise multivariate logistic regression analysis was subsequently carried
out.
Results: The study included 133 women of whom 33 had recurrent PID. Women in the recurrent

of
PID group had a higher rate of previous pelvic surgery (12 (36%) vs. 20 (20%), adjusted odds
ratio [OR] 2.2 (95% confidence interval CI 1.06-5.4, p=0.05) and more had intrauterine devices
(IUD) still in place if they had been previously present (5 (71.4%) vs. 9(25.7%), OR 7.2, (95%

ro
CI 1.18-43.9), p=0.02). The majority were treated with a combination of Ampicillin and
Gentamycin, fewer received Augmentin or a cephalosporin base regimen (28 (84.8%) vs 56

-p
(56.0%), OR 4.4, (95% CI 1.5-12.3, p=0.02), (1 (3.0%) vs 27 (27.0%), OR 0.08, (95% CI 0.01-
0.64), (4 (12.2%) vs 17 (17.0%)) respectively. In addition, invasive treatment had been required
re
in more patients who later had a recurrent PID (6 (18.1%) vs. 4(4.0%), OR 5.3 (95% CI 1.1.4-
20.2), p=0.007). Antibiotic regimens and invasive treatment were independently associated with
lP

recurrent PID (OR 2.69; 95% CI 1.13- 6.41, OR 2.10; 95% CI 1.19- 3.71, respectively).
Conclusion: Among women with PID, special awareness should be given to women with
previous pelvic surgery, who required an additional interventional treatment and have an IUD
na

inserted. Efforts should be made to achieve treatment success and optimal prevention to prevent
recurrent PID.
ur

Key words: pelvic inflammatory disease; PID; recurrent; pelvic pain; prevention; infertility,
antibiotics, IUD, intrauterine device
Jo

Article:
Introduction
Pelvic inflammatory disease (PID) is a common infection affecting 4% of women during their
reproductive years. It is caused by the ascension of microorganisms from the lower genital tract
to the endometrium, fallopian tubes and adjacent structures [1].
PID is a major health burden leading to such complications as chronic pelvic pain, ectopic
pregnancy and infertility. [1, 2, 3] The hallmark of PID diagnosis is pelvic tenderness.
Unfortunately the diagnosis is lacking both sensitivity and specificity and clinicians should refer
to the Centers for Disease Control and Prevention (CDC) recommended diagnosis criteria for

of
PID. [2]
Previous studies have identified predisposing risk factors for PID, such as multiple sexual
partners, sexually transmitted infection, young age and African American ethnicity. Whereas

ro
barrier contraception methods seem to be protective [4,5] and the first line of treatment consists
of antibiotics with a high success rate [6], newer studies are trying to define markers associated

-p
with the failure of conservative treatment in order to adapt therapy in the acutest way [7].
It is well established that repeated episodes of PID increase the morbidity and worsen the
re
reproductive outcomes [8,9]. Despite the importance of detecting patients at risk for subsequent
PID infection there are currently no effective means of predicting or preventing recurrent PID. In
lP

this study, we aimed to investigate the predisposing factors for recurrent PID.
na

Material and methods

Study population
This was a retrospective case control study of women who had been hospitalized for treatment
ur

after being diagnosed with a PID and subsequently had a recurrence of PID requiring inpatient
treatment. The data were obtained from a tertiary medical center for patients who had been
Jo

admitted between 2003 and 2018. These patients were compared to 100 patients hospitalized
with a primary episode of PID without future recurrence.
In order to achieve a random selection of control patients during the years of the study: all the
PID hospitalizations were chronologically listed, for every recurrent case 3-4 patients listed
above were chosen as a control. Diagnosis was made in accordance with diagnostic criteria for
PID published in recent guidelines by the CDC [2]. In short, PID diagnosis was established in
sexually active women experiencing pelvic or lower abdominal pain, excluding other etiologies
for their complaints and including one of the following: cervical motion tenderness, uterine
tenderness or adnexal tenderness. In our center, patients with PID are hospitalized in cases of:
presence of a tubo-ovarian abscess (TOA), severe illness (as most patients in emergency
department encounters suffering from a PID are experiencing mild-moderate disease), nausea
and vomiting or fever >38°C, and patients who are judged not to be compliant with outpatient
treatment regimen. In order to evaluate risk factors of recurrence of primary infections, previous
PID reported in the hospital admission records, was an exclusion criteria, (as our medical center

of
is a tertiary center, many patients are admitted with an history of PID treated at other centers. We
have excluded those, as analysis of risk factors is not possible as data regarding their admission
is not available).

ro
A PID was considered a new event, aka recurrent, if 30 days or more had elapsed since the first

-p
episode. The 30 day limit was set as consistent with the common definition for readmission after
hospitalization [10] and the cure rates for PID evaluated with the cutoff of 30 days in the
"PEACH" trial [3].
re
Data collection
lP

For the purpose of this study, we extracted patient’s hospital admission records, gynecological
ward follow-up charts, laboratory and imaging scan reports, operation reports and discharge
letters from the electronic medical record databases of the gynecological unit in our medical
na

center. Records were reviewed by a single reviewer (M.S).

The following data were extracted: patient characteristics [age, origin, marital status, gravidity,
ur

parity, history of PID, pelvic surgery and mode of contraception – no distinction was made
between different IUD’s types], current admission characteristics [body temperature, C reactive
protein (CRP) serum level, white blood cell (WBC) count, neutrophils percent, presence of a
Jo

TOA and abnormal cervical discharge]. Medical treatment and management [antibiotics regimen,
IUD removal and invasive treatment], clinical outcomes including length of stay, as well as
culture results of vaginal, cervical, urine, IUD, blood and aspirated TOA samples were also
reviewed. Marital status was defined as accepted, with a widow status being regarded as a single
status. Oral temperature measured in Celsius centigrade was recorded on admission. TOA was
diagnosed when evident by an imaging modality (vaginal ultrasonography, computed
tomography). Inpatient treatment regimens were in accordance with the CDC guidelines [2]. In
our department the physician can decide which antibiotic regimen to prescribe mostly based on
the CDC recommendations. The selection of a treatment regimen should consider availability,
cost, patient acceptance, antimicrobial acceptability and the clinical findings at presentation. All
patients were treated with one of the following regimen: Clindamycin/Gentamicin regimen:
Clindamycin 900 mg IV every 8 hours plus Gentamicin single daily dosing (5mg/kg),
Augmentine: Ampicillin/clavulanic acid (our local substitute to ampicillin/sulbactam,
administered at a daily dose of 1g, every 8 hours) plus Doxycycline 100 mg orally every 12

of
hours or a Cephalosporin base regimen plus Doxycycline 100 mg orally every 12 hours.
Treatment continued until 48 hours after clinical improvement, followed by oral therapy with
clindamycin (450 mg orally four times daily) or doxycycline (100 mg twice daily) on an

ro
outpatient basis. IUD removal was performed at the discretion of the attending gynecologists at
the emergency department or during the hospitalization. Invasive procedures were defined as

-p
percutaneous aspiration/colpotomy or laparoscopy, we did not differentiate between the invasive
procedures.
re
Statistical analysis
lP

Patient characteristics are described as proportions for categorical variables and medians,
interquartile ranges and means for continuous variables without a normal distribution.
Significance between groups was assessed using the Chi squared test and -Fisher's exact test for
na

categorical variables. The Student t test was used for analysis of continuous variables with
normal distribution and the Mann–Whitney U test for analysis of continuous variables with
skewed distribution. A 2-sided P-value < 0.05 indicated statistical significance. The data were
ur

analyzed using Software Package for Statistics and Simulation (IBM SPSS version 22, IBM
Corp, Armonk, NY).
Jo

Ethical approval: Approval was obtained from the institutional review board (IRB) of the
Hadassah-Hebrew University Medical Center before data extraction was performed. The
requirement for written informed consent was waived by the IRB.

Results
During the study period, 711 cases of PID were recorded and admitted for inpatient treatment. Of
those, 33 (4.8%) women had a recurrent event of PID. The characteristics of both groups are
presented in table 1.
Women with a recurrent PID had a higher rate of previous pelvic surgery (12 (36%) vs. 20
(20%), OR 2.2 (95% CI 1.06-5.4, p=0.05). Most of them were treated with a combination of
Gentamicin and Clindamycin (28 (84.8%) vs 56 (56.0%), OR 4.4, (95% CI 1.5-12.3, p=0.02),
versus patients treated with Augmentine who had a lower risk of recurrence (1 (3.0%) vs 27
(27.0%), OR 0.08, (95% CI 0.01-0.64). Women with a recurrent PID had a higher chance of

of
requiring invasive treatment during their first admission (6 (18.1%) vs. 4(4.0%), OR 5.3 (95% CI
1.1.4-20.2), p=0.007), or to have their IUD retained after the first episode of PID (5 (71.4%) vs.
9(25.7%), OR 7.2, (95% CI 1.18-43.9), p=0.02).

ro
The results of the multivariate logistic regression model are presented in table 2. Treatment with
a combination of Gentamicin/Clindamycin and invasive treatment were independently associated

-p
with PID recurrence (adjusted OR 2.69; 95% confidence interval [CI] 1.13- 6.41, adjusted OR
2.10; 95% [CI] 1.19 - 3.71, respectively).
re
lP

Discussion
Our study is the first to evaluate multiple predictor of recurrent PID (a search in the Cochrane
database, PubMed, and Google was performed). In our study population, the PID recurrence rate
na

was 4.8% which is lower than the 20-25% rate usually reported in the literature [11,12].
However, these data must be viewed cautiously, since PID is associated with an increased risk of
subsequent chronic pelvic pain in general [12]. In order to collect accurate data, our study
ur

focused on PID cases requiring hospitalization. This may limit the misinterpretation of chronic
pelvic pain, and the recorded recurrences to only more severe cases.
Jo

The CDC estimates that more than 1 million women are affected with PID each year in the
United States [13]. This, combined with a high recurrence rate, results in a large population of
women who suffer from recurrent PID and long term complications. [1,3,5]. For secondary
prevention to occur it is of paramount importance to identify the sub group at risk for recurrent
PID in order to provide the adequate treatment to minimize recurrence.
The risk factors for a primary PID infection are well known and include sex, especially with
multiple partner [5], age between 15-25 years [14], previous PID [12, 15] and sexually
transmitted diseases. [16, 17] It has been proved that consistent use of condoms offers a
significant risk reduction [9, 18]. Ness et al., demonstrate that persistent use of condoms after a
primary episode of PID reduced the risk of recurrent PID [9], and Haggerty et al., note in their
study that PID may recur among women who engage in risky sexual behavior during treatment
[19]. Yet no clear risk factor for recurrent PID has been established.

of
The CDC have developed guidelines for the treatment of PID [2]. There are empirical treatments
using broad-spectrum combination regimens of antimicrobial agents to cover likely pathogens
[20,21]. Cochrane analysis didn’t find any conclusive evidence that one regimen of antibiotics

ro
was safer or more effective than any other [22].
In our study we found that patients who had been treated with a combination of Gentamicin and

-p
Clindamycin had a higher risk of recurrence versus patients treated with Augmentin which seems
to lower the recurrence rate. Without being formally recommended the combination of
re
Clindamycin and an aminoglycoside may be particularly appropriate for patients with a TOA
[22, 23] and was prescribed more frequently in more severe cases. We can assume that patients
lP

treated with this combination of antibiotics had a more severe disease resulting in higher
recurrence later on. Moreover, we demonstrate that patients who required any invasive treatment
were more prone to experience a subsequent episode of PID. Furthermore, the preferable
na

outcome of Augmentin regiment require further evaluation with prospective trial to assess and
improve PID treatment and long term outcomes.
ur

Regarding the patient’s history, previous pelvic surgeries were associated with recurrent PID.
Surprisingly, this is not a risk factor known for predisposing women to a primary PID. Pelvic
Jo

surgery might cause tubal damage and adhesion [24]; it is possible that tubal damage may impair
full tubal recovery and predispose to further recurrent infection.

Previous studies have suggested that IUD insertion increases the risk of PID [25]. To our
knowledge there are no studies looking at the relationship between IUD and recurrence of further
PID, but only into the outcome of the current event of PID. Our study demonstrated that if a PID
occurred with an IUD in place, even after optimal treatment and cure there is still a greater
chance for recurrence if the IUD remains in situ. We must point out that due to the retrospective
analysis of the data, causability can’t be proved and the IUD itself may not be the direct cause of
recurrence, in example the lack of use of barrier contraception with IUDs known as protective
[8] may contribute.
Prevention of long-term sequelae is determined to a large extent by early administration of
appropriate antimicrobial therapy [2]. Therefore, our study can lead to an easy and cost effective
clinical implication. A precise analysis of the data at the time of a first PID episode might be

of
used to define patients prone to recurrent event. We believe that women prone to a secondary
infection may be informed and recommended preventive behavioral changes which can reduce
the risk for further PID infection. In the case of a high risk patient with an IUD, IUD removal

ro
should be easily recommended as this is a benign procedure that can lower recurrence. Finally, in
especially high risk cases, when reproduction is no longer being considered, a curative and

-p
prophylactic salpingectomy may be consider in order to minimize recurrent PID and long term
sequelae.
re
Our study has several limitations. The prominent one is the retrospective nature of the study that
lP

carries an inherent selection bias and information bias derived from medical record coding.

Additionally, the physician choice of antibiotics protocol was not random, which may lead to a
na

selection bias regarding the regimen administrated. Another bias might be a confounding bias for

unknown variables not accounted for in our study. Third, the lack of sensitivity and specificity in
ur

the diagnosis of PID may affect the accuracy of the diagnosis [26], an issue which is

unfortunately is well known regarding PID disease and PID studies. Moreover there is no
Jo

standardized definition of recurrent PID, choosing 30 days was based on the standard

readmission [10] rate and the PEACH trial [3] , but as it was arbitrary it might be a source of

information bias. In addition, we cannot exclude the possibility that patients who did not

experience recurrent PID may still have a recurrence in the future, although we tried to minimize
this limitation by having long a follow-up of 15 years and the knowledge that PID is most

common among sexually active young and adolescent women [23, 27,]. Another potential

caveat, is the relatively small sample size of the recurrent PID group. Finally, we cannot rule out

the possibility that women with a first PID experienced a recurrent PID but were treated

elsewhere. This point might be a source of information bias, although it is probably minimal, as

our hospital serves the greater Jerusalem area covering a heterogeneous, multicultural and

multinational population from a heavily populated urban center as well as women from rural,

of
scarcely populated areas.

ro
Conclusion

-p
In conclusion, there is a subgroup of women who are prone to subsequent PID. Women with risk
factors may especially benefit from preventive life style changes and being informed of those
re
recommendations in order to minimize the risk of further event.
Moreover, in order to reduce the long term sequelae of PID, empirical and early treatment is
lP

warranted specifically in these patients. Now that a high risk group has been defined, further
evaluation in prospective studies is needed in order to establish the best preventive way to reduce
recurrent PID.
na

Declaration of interests
ur

The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
The authors declare the following financial interests/personal relationships which may be considered as
potential competing interests:
Jo
References

1. Ford GW, Decker CF. Pelvic inflammatory disease. Disease-a-month: DM. 2016
Aug;62(8):301.
2. Workowski KA, Bolan GA. Sexually transmitted diseases treatment guidelines, 2015.
MMWR. Recommendations and reports: Morbidity and mortality weekly report.
Recommendations and reports. 2015 Jun 5;64(RR-03):1-137.
3. Trent M, Bass D, Ness RB, Haggerty C. Recurrent PID, subsequent STI, and reproductive
health outcomes: findings from the PID evaluation and clinical health (PEACH) study.

of
Sexually transmitted diseases. 2011 Sep;38(9):879.
4. Lee, N. C., et al. (1991). "Measures of sexual behavior and the risk of pelvic inflammatory

ro
disease." Obstetrics and gynecology 77(3): 425-430.
5. Kreisel K, Torrone E, Bernstein K, Hong J, Gorwitz R. Prevalence of pelvic inflammatory

-p
disease in sexually experienced women of reproductive age—United States, 2013–2014.
MMWR. Morbidity and mortality weekly report. 2017 Jan 27;66(3):80.
re
6. Hemsell DL, Little BB, Faro S, Sweet RL, Ledger WJ, Berkeley AS, Eschenbach DA,
Wolner-Hanssen P, Pastorek JG. Comparison of three regimens recommended by the Centers
lP

for Disease Control and Prevention for the treatment of women hospitalized with acute pelvic
inflammatory disease. Clinical infectious diseases. 1994 Oct 1;19(4):720-7.
7. Levin G, Herzberg S, Dior UP, Shushan A, Gilad R, Benshushan A, Rottenstreich A. The
na

predictive role of CA-125 in the management of tubo-ovarian abscess. A retrospective study.

European Journal of Obstetrics & Gynecology and Reproductive Biology. 2019 Jul 1;238:20-
4.
ur

8. Trent M, Haggerty CL, Jennings JM, Lee S, Bass DC, Ness R. Adverse adolescent
reproductive health outcomes after pelvic inflammatory disease. Archives of pediatrics &
Jo

adolescent medicine. 2011 Jan 3;165(1):49-54.

9. Ness, R. B., et al. (2004). "Condom use and the risk of recurrent pelvic inflammatory disease,
chronic pelvic pain, or infertility following an episode of pelvic inflammatory disease."
American journal of public health 94(8): 1327-1329.
10. Kansagara D, Englander H, Salanitro A, Kagen D, Theobald C, Freeman M, Kripalani S.
Risk prediction models for hospital readmission: a systematic review. Jama. 2011 Oct
 19;306(15):1688-98.
11. Goyal M, Hersh A, Luan X, Localio R, Trent M, Zaoutis T. National trends in pelvic
inflammatory disease among adolescents in the emergency department. Journal of Adolescent
Health. 2013 Aug 1;53(2):249-52.
12. Weström L. Effect of acute pelvic inflammatory disease on fertility. American Journal of
Obstetrics & Gynecology. 1975 Mar 1;121(5):707-13.
13. Centers for Disease Control and Prevention, Division of STD Prevention, National
Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention, Pelvic Inflammatory Disease

of
(PID): CDC Fact Sheet. http://www.cdc.gov/std/PID/STDFact-PID.htm#common. Accessed
April 21, 2009

ro
14. Trent M. Status of adolescent pelvic inflammatory disease management in the United
States. Current opinion in obstetrics & gynecology. 2013 Oct;25(5):350.

-p
15. Flesh G, Weiner JM, Corlett Jr RC, Boice C, Mishell Jr DR, Wolf RM. The intrauterine
contraceptive device and acute salpingitis: a multifactor analysis. American journal of
re
obstetrics and gynecology. 1979 Oct 1;135(3):402-8.
16. Kamwendo F, Forslin L, Bodin L, Danielsson D. Decreasing incidences of gonorrhea-and
chlamydia-associated acute pelvic inflammatory disease: a 25-year study from an urban area
lP

of central Sweden. Sexually transmitted diseases. 1996 Sep 1;23(5):384-91.

17. Eschenbach DA, Harnisch JP, Holmes KK. Pathogenesis of acute pelvic inflammatory
na

disease: role of contraception and other risk factors. American journal of obstetrics and
gynecology. 1977 Aug 15;128(8):838-50.
18. Haggerty, C.L., Totten, P.A., Astete, S.G., Lee, S., Hoferka, S.L., Kelsey, S.F. and Ness,
ur

R.B., 2008. Failure of cefoxitin and doxycycline to eradicate endometrial Mycoplasma

genitalium and the consequence for clinical cure of pelvic inflammatory disease. Sexually
Jo

transmitted infections, 84(5), pp.338-342.

19. Sweet RI, Draper DL, Schachter J, James J, Hadley WK, Brooks GF. Microbiology and
pathogenesis of acute salpingitis as determined by laparoscopy: what is the appropriate site to
sample?. American journal of obstetrics and gynecology. 1980 Dec 1;138(7):985-9.
20. Soper DE, Brockwell NJ, Dalton HP. Microbial etiology of urban emergency department
acute salpingitis: treatment with ofloxacin. American journal of obstetrics and gynecology.
 1992 Sep 1;167(3):653-60.
21. Sweet RL. Treatment of acute pelvic inflammatory disease. Infectious diseases in
obstetrics and gynecology. 2011;2011:561909.
22. Brunham RC, Gottlieb SL, Paavonen J. Pelvic inflammatory disease. New England
Journal of Medicine. 2015 May 21;372(21):2039-48.
23. McNeeley SG, Hendrix SL, Mazzoni MM, Kmak DC, Ransom SB. Medically sound,
cost-effective treatment for pelvic inflammatory disease and tuboovarian abscess. American
journal of obstetrics and gynecology. 1998 Jun 1;178(6):1272-8.

of
24. Ten Broek RP, Issa Y, van Santbrink EJ, Bouvy ND, Kruitwagen RF, Jeekel J, Bakkum
EA, Rovers MM, van Goor H. Burden of adhesions in abdominal and pelvic surgery:

ro
systematic review and met-analysis. Bmj. 2013 Oct 3;347:f5588.
25. Hubacher D. Intrauterine devices & infection: Review of the literature. The Indian

-p
journal of medical research. 2014 Nov;140(Suppl 1):S53.
26. Yang SF, Wu TF, Tsai HT, Lin LY, Wang PH. New markers in pelvic inflammatory
re
disease. Clinica Chimica Acta. 2014 Apr 20;431:118-24.
27. Rekart ML, Gilbert M, Meza R, Kim PH, Chang M, Money DM, Brunham RC.
lP

Chlamydia public health programs and the epidemiology of pelvic inflammatory disease and
ectopic pregnancy. The Journal of Infectious Diseases. 2012 Oct 24;207(1):30-8.
na
ur
Jo
 Single PID Future Odds ratio p value
n=100 recurrence (95% CI)
n=33

Age, years 34 [28-43] (35) 32 [25-37] (32) 0.09

Origin 0.46
 Jewish 79 (79.0%) 28 (84.8%)
 Arab 21 (21.0%) 5 (15.2%)

of
Marital status 0.43
 Married 68 (68.0%) 20 (60.6%)

ro
 Unmarried 32 (32.0%) 13 (29.4%)

Gravidity 2 [0-4] (3) 2 [0-5](3) 0.94

Parity 2 [0-3] (2)

-p 2 [0-3](2) 0.36
re
EP 3 (3.0%) 3 (9.0%) 0.15

CS 23 (23.0%) 7 (21.2%) 0.99

lP

AB 28 (28.0%) 13 (39.3%) 0.23

Previous pelvic surgeries 20 (20.0%) 12 (36.3%) 2.2 (1.06-5.4) 0.05

na

After uterine instrumentation 10 (10.0%) 1 (3.0%) 0.20

IUD carrier 35 (35.0%) 7 (21.2%) 0.14

ur

Oral contraception 10 (10.0%) 2 (6.0%) 0.233

Jo

CRP at admission, mg/L 2.5[0.3-6.8] (5.7) 2.3[0.1-15.8] 0.41

(7.8)

WBC at admission, ×1,000/mm3 10.8 [8.0-14.3] 12.3 [7.8-15.3] 0.72

(11.5) (11.9)
Neutrophils, % 72.9 [59.9-83.0] 75.0 [59.9-83.4] 0.97
(69.2) (69.1)

Length of hospitalization, days 5[4-7] (6) 6 [5-8] (7) 0.12

Tubo-ovarian abscess at presentation 31 (31.0%) 10 (30.3%) 0.94

Fever >38°C 22 (22.0%) 10 (30.3%) 0.33

of
Treatment 0.02
 Clindamycin/Gentamicin 56 (56.0%) 28 (84.8) 4.4 (1.5-12.3)

ro
 Augmentin 27 (27.0%) 1 (3.0%) 0.08 (0.01-0.64)
 Cephalosporin based 17 (17.0%) 4 (12.2%)

-p
Length of treatment IV, days 5[3-6] (5) 5[5-6] (6) 0.33

Need for invasive treatment 4 (4.0%) 6 (18.1%) 5.3 (1.4-20.2) 0.007

re
Positive culture 33 (33.0%) 10 (30.3%) 0.77
lP

IUD retained 9 (25.7%) 5 (71.4%) 7.2 (1.18-43.9) 0.02

Abnormal cervical discharge 16 (16.0%) 10 (30.3%) 0.07

na

Table 1. Patient, disease characteristics and outcomes according to PID recurrence association
All continuous variables are expressed as median [interquartile range] (mean).
ur
Jo
 Variable Prevalence of risk Recurrence in OR (95% CI)
factor (%) future

Treatment

Conservative 92.5% 22.1% Reference

Invasive 7.5% 60.0% 2.10(1.19-3.71)

Antibiotics

of
Non Clindamycin/Gentamicin 50.3% 16.4% Reference

ro
Clindamycin/Gentamicin 49.7% 33.3% 2.69(1.13-6.41)

-p
re
Table 2 Multivariate analysis of factors associated with recurrent PID
lP
na
ur
Jo