MAJOR ARTICLE

Azithromycin Versus Doxycycline for the

Treatment of Genital Chlamydia Infection: A
Meta-analysis of Randomized Controlled Trials
F. Y. S. Kong,1 S. N. Tabrizi,2 M. Law,3 L. A. Vodstrcil,1,2 M. Chen,4,5 C. K. Fairley,5 R. Guy,3 C. Bradshaw,4,5 and J. S. Hocking1
1

Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, University of Melbourne; 2Murdoch Childrens Research
Institute, Parkville; 3Kirby Institute, University of New South Wales, Sydney; 4Sexual Health Unit, Melbourne School of Population and Global Health
Melbourne, University of Melbourne; and 5Melbourne Sexual Health Centre, Melbourne, Australia

Background. There has been recent debate questioning the efcacy of azithromycin for the treatment of urogenital chlamydia infection. We conducted a meta-analysis to compare the efcacy of 1 g azithromycin with 100 mg
doxycycline twice daily (7 days) for the treatment of urogenital chlamydia infection.
Methods. Medline, PubMed, Embase, Cochrane Controlled Trials Register, Cochrane reviews, and Cumulative
Index to Nursing and Allied Health Literature were searched until 31 December 2013. Randomized controlled trials
comparing azithromycin with doxycycline for the treatment of genital chlamydia with evaluation of microbiological
cure within 3 months of treatment were included. Sex, diagnostic test, follow-up time, attrition, patient symptomatic
status, and microbiological cure were extracted. The primary outcome was the difference in efcacy at nal follow-up.
Study bias was quantitatively and qualitatively summarized.
Results. Twenty-three studies were included evaluating 1147 and 912 patients for azithromycin and doxycycline,
respectively. We found a pooled efcacy difference in favor of doxycycline of 1.5% (95% condence interval [CI],
.1% to 3.1%; I 2 = 1.9%; P = .435; random effects) to 2.6% (95% CI, .5%4.7%; xed effects). Subgroup analyses
showed that the xed effects pooled efcacy difference for symptomatic men was 7.4% (95% CI, 2.0%12.9%),
and the random effects was 5.5% (95% CI, 1.4% to 12.4%).
Conclusions. There may be a small increased efcacy of up to 3% for doxycycline compared with azithromycin
for the treatment of urogenital chlamydia and about 7% increased efcacy for doxycycline for the treatment of symptomatic urethral infection in men. However, the quality of the evidence varies considerably, with few double-blind
placebo-controlled trials conducted. Given increasing concern about potential azithromycin failure, further welldesigned and statistically powered double-blind, placebo-controlled trials are needed.
Keywords.

genital chlamydia; meta-analysis; treatment efcacy; azithromycin; doxycycline.

Chlamydia is the most common bacterial sexually

transmitted infection (STI) worldwide [1]. For uncomplicated genital chlamydia, treatment with single-dose
azithromycin or 7 days of doxycycline is recommended
[2], with doxycycline less preferred because of compliance issues [3].
Received 12 December 2013; accepted 10 March 2014; electronically published
11 April 2014.
Correspondence: Fabian Yuh Shiong Kong, BPharm, MEpi, Melbourne School
of Population and Global Health, University of Melbourne, 3/207 Bouverie St,
Melbourne, Australia 3004 (kongf@unimelb.edu.au).
Clinical Infectious Diseases 2014;59(2):193205
The Author 2014. Published by Oxford University Press on behalf of the Infectious
Diseases Society of America. All rights reserved. For Permissions, please e-mail:
journals.permissions@oup.com.
DOI: 10.1093/cid/ciu220

A meta-analysis of randomized controlled trials

(RCTs) in 2002 reported a 98% cure rate for doxycycline and 97% for azithromycin (difference in efcacy,
1.0%; 95% condence interval [CI], 1.0% to 2.0%) [4].
However, 11 of the 12 studies used culture or immunoassay rather than sensitive nucleic acid amplication
tests (NAATs) to determine microbiological cure, and
one-third had attrition rates >25%. Furthermore, 3 of
the studies had cure rates of 81%92% for azithromycin
compared with 99%100% for doxycycline. Given the
use of culture rather than NAATs, it is possible that
the reported efcacies were overestimated [5, 6].
Repeat chlamydia infections are common after treatment; studies report high repeat infection rates ranging

Genital Chlamydia Treatment Meta-analysis

CID 2014:59 (15 July)

193

from 18% to 34% [79]. As most repeat infections are likely to

be reinfections, emerging evidence suggests that treatment failure following azithromycin may account for a substantial proportion [9, 10], and this has led to considerable debate in the
medical and scientic literature [5, 6, 1113]. A partner treatment study found that among women who reported having
no sex after treatment, 8% (95% CI, 5%11%) had persistent infection at follow-up [10]; another study of adolescent females
reported a treatment failure of 7.9% (4%10%) [9].
Given the recent concerns about the efcacy of azithromycin
and the fact that it is >10 years since the meta-analysis by Lau
and colleagues [4], we conducted an updated meta-analysis
to compare the efcacy of azithromycin 1 g vs doxycycline
100 mg twice daily for 7 days for the treatment of genital chlamydia infection in men and women.
METHODS
This meta-analysis was conducted according to the PRIMSA
(Preferred Reporting Items for Systematic Reviews and MetaAnalyses) statement [14].
Protocol and Registration

Analysis methods, inclusion criteria, and protocol were registered with Prospective Registration of Systematic Reviews
(PROSPERO) under registration number CRD42013003711
(available at: http://www.crd.york.ac.uk/PROSPERO/).
Search Strategy

Medline, PubMed, Embase, Cochrane Controlled Trials Register, Cochrane Reviews, and Cumulative Index to Nursing and
Allied Health Literature (CINHAL) were searched from the earliest date up to 31 December 2013. We hand-searched conference abstracts from the International Society for Sexually
Transmitted Diseases Research and the International Union
Against Sexually Transmitted Infections and the reference lists
of identied papers. Only English-language papers were sought.
Search terms included Chlamydia trachomatis OR chlamydia
AND azithromycin OR Zithromax. Medical subject headings
were used where possible.
Inclusion and Exclusion Criteria
We searched for published RCTs comparing the efcacy of azithromycin 1 g (single dose) with doxycycline 100 mg twice
daily for 7 days for treating genital (urethral or cervical) chlamydia infection in men and women. Eligible studies were English-language, included participants aged 15 years, and
measured microbiological cure (dened as a negative chlamydia
test result at the last follow-up) within 3 months of treatment.
Studies on treatment of prostatitis in men and pelvic

194

CID 2014:59 (15 July)

Kong et al

inammatory disease in women, as well as review and discussion papers, were excluded.
Data Extraction Process

We developed an Excel spreadsheet for extracting the following

data: study design, randomization method, participant numbers
by treatment arm and sex, presence of signs and/or symptoms at
diagnosis, diagnostic method used for assessing microbiological
cure, follow-up times, attrition, and microbiological cure. One
author (F. Y. S. K.) extracted data from included studies, and the
second author (J. S. H.) checked the extracted data. Disagreements were resolved by discussion between the 2 authors and
consultation with an additional author (C. K. F.) until a consensus was reached.
Outcome

The primary outcome was the difference in treatment efcacy

(efcacy for doxycycline minus efcacy for azithromycin) at
the last follow-up conrmed by a microbiological curethe
proportion with microbiological cure; the numerator was the
number of treated subjects with a microbiological cure and
the denominator was the number of subjects assigned to the
treatment and tested. Where available, the cumulative treatment
efcacy at the nal follow-up was used. For one study [15], efcacy data were extracted from the modied intention-to-treat
analysis. For another study, efcacy data for azithromycin and
doxycycline, with and without tinidazole, were used [16].
Analysis

Meta-analysis was used to calculate the pooled estimates of the

difference in treatment efcacy. We used the I 2 test to estimate
the proportion of total variability in point estimates that could
be attributed to heterogeneity other than by chance [17]. Both
xed and random effects pooled estimates were calculated.
Treatment efcacy difference was stratied by sex, signs/
symptoms, type of test (NAAT based vs culture/enzyme immunoassay [EIA], whether the study was double blind, follow-up
times (3 or >3 weeks), study attrition (calculated as the average
attrition across both treatment arms), whether the study was
drug company sponsored, whether doxycycline compliance
was measured, and when the study was published (before or
after 2002). Given the relatively small number of studies in
the subgroup analyses, we present both xed and random
effects.
Assessment of Bias and Quality
We assessed publication bias using a funnel plot. Asymmetry
was statistically evaluated using the Egger correlation test by regressing the difference in treatment efcacy by its standard
error. We used the Cochrane Collaboration Tool to assess bias
within studies [18]. Sensitivity analyses were undertaken to investigate the impact of removing studies appearing as outliers in

Figure 1. Flow of information in the systematic review. Abbreviations: CINHAL, Cumulative Index to Nursing and Allied Health Literature; RCT, randomized
controlled trial.

the funnel plot. Data were analyzed using Stata software version
12 (StataCorp, College Station, Texas).
RESULTS
Study Selection

The review process is shown in Figure 1 and the selected papers

summarized in Table 1. Of the 716 references identied, 58 papers were reviewed with 23 studies meeting the inclusion
criteria.
Study Characteristics

At the last follow-up visit, 1147 and 912 patients were evaluated
for azithromycin and doxycycline efcacy, respectively. Ten
studies (43%) included men only [15, 16, 20, 21, 23, 25, 26, 28,
29, 34], 8 studies (35%) included women only [24, 30, 31, 33, 35

37, 39], and 5 studies (22%) included both sexes [19, 22, 27, 32,
38]. Follow-up periods ranged from 1 to 6 weeks; 12 (52%) studies had follow-up times of 4 weeks [15, 16, 20, 22, 24, 25, 2730,
33, 38]. In 16 (70%) studies, the primary aim was to evaluate
treatments for nongonococcal urethritis (NGU) or cervicitis
with participants being randomized before the causative organisms were diagnosed and only a subgroup being diagnosed with
chlamydia [15, 16, 19, 2126, 28, 29, 31, 34, 35, 37, 39]. In the
other 7 studies, the primary aim was to evaluate treatment effect
among those diagnosed with chlamydia prior to randomization
[20, 27, 30, 32, 33, 36, 38]. Fourteen (61%) studies included only
patients with signs and/or symptoms [15, 16, 19, 21, 2326, 29,
30, 3437], with no studies providing results for asymptomatic
patients only. Four (17%) studies used EIA or DFA [31, 32, 34,
39], 6 (26%) used NAAT-based testing [15, 16, 24, 33, 37, 38],
and the remaining 13 (57%) studies used culture to assess

Genital Chlamydia Treatment Meta-analysis

CID 2014:59 (15 July)

195

196

Table 1.

Attributes of Randomized Controlled Trials Included in the Meta-analysis

CID 2014:59 (15 July)

Azithromycin
Study, First
Author, Year
Lassus, 1990 [19]a

Diagnostic
Method
Culture/EIA

Study
Blinding
Open label

Kong et al

Steingrimsson, 1990 [20]

Culture

Third-party
blinding

Whatley, 1991 [21]

Martin, 1992 [22]

Culture
Culture

Open label
Open label

Follow-up,
wk
1

Doxycycline

Male Female Attrition Rate (%) Male Female Attrition Rate (%)
22

16

3/20 (15.0%)

19

5/172 (2.9%)b

1
1

6
41

0
89

2/19 (10.5%)b
13/154 (8.4%)

6
37

0
77

2/19 (10.5%)b
8/133 (6.0%)

32

79

26

70

109/111 (98.2%)

96/96 (100%)

34

78

29

73

112/112 (100%)
Cumulative

101/102 (99.0%)
Cumulative

136/141 (96.5%)

122/125 (97.6%)

15/49 (30.6%)

44/44 (100%)
35/35 (100%)

42/42 (100%)
34/34 (100%)

13

1/14 (7.1%)

15/50 (30.0%)

42
34

Ossewaarde, 1992 [24]

Culturec

Open label

16

1/17 (5.9%)

4
1

16
16

Hammerschlag, 1993 [27] Culture

21

Open label

16
6

24

4
4
2
5

Culture

Double blind

Open label

13
6/30 (20.0%)

29
26

11/27 (40.7%)

22

5/46 (10.9%)

12
2

21

31

6
79

38
0

38
30

7/31 (22.6%)

24

29

Stamm, 1995 [29]

DNA Probe
test

3
1

Third-party
blinding

Horner, 1995 [30]

28
26

Culture

13
13

16

1
2

Open label

Steingrimsson, 1994 [28]

5/6 (83.3%)
112/114 98.2%)

Culture

6/6 (100%)
127/130 (97.7%)

49

44
35

Lister, 1993 [26]

20/20 (100%)
48/49 (98.0%)

5/172 (2.9%)b

1
2

Open label

19/20 (95.0%)

17/17 (100%)
43/44 (97.7%)

Double blind

Culture

Doxycycline

18/18 (100%)

44

Culture

Lauharanta, 1993 [25]

Azithromycin

3/23 (13.0%)

2
4

Nilsen, 1992 [23]

PCR

17

Microbiological Cure (%)

16

0/100 (0%)
10/40 (25.0%)

2/18 (11.1%)

13/13 (100%)
13/13 (100%)
12/13 (92.3%)

15/16 (93.8%)

12/13 (92.3%)

27/28 (96.4%)
24/26 (92.3%)

29/29 (100%)
26/26 (100%)

20/21 (95.2%)

22/24 (91.7%)

Cumulative cure
(5 wk) = 26/30
(86.7%)

Cumulative cure
(5 wk) = 28/30
(93.3%)

9/21 (42.9%)

27/29 (93.1%)

21/22 (95.5%)

3/27 (11.1%)

13/16 (81.3%)
28/30 (93.3%)

12/12 (100%)
22/23 (95.7%)

16

35/37 (94.6%)

18/18 (100%)

2
69

24
0

1/83 (1.2%)

42/44 (95.5%)
73/79 (92.4%)

23/26 (88.5%)
68/69 (98.6%)

28
23

8/29 (27.6%)

36/38 (94.7%)
27/30 (90.0%)

26/28 (92.9%)
23/23 (100%)

16/16 (100%)
16/16 (100%)
14/16 (87.5%)

16

2/18 (11.1%)

Cumulative

Cumulative

25/30 (83.3%)
16/16 (100%)

19/21 (90.5%)
16/16 (100%)

16

16

16/16 (100%)

15/16 (93.8%)

16

16

16/16 (100%)

14/16 (87.5%)

Table 1 continued.
Azithromycin
Study, First
Author, Year

Diagnostic
Method

Gruber, 1996 [31]

DFA

Study
Blinding
Not stated

Follow-up,
wk
1

Doxycycline

Microbiological Cure (%)

Male Female Attrition Rate (%) Male Female Attrition Rate (%)
0

22

7/80 (8.8%)

20

22

7/80 (8.8%)

20

Azithromycin

Doxycycline

20/22 (90.9%)

17/20 (85.0%)

21/22 (95.5%)

19/20 (95.0%)

323/351 (92.0%)
338/347 (97.4%)

146/164 (89.0%)
161/163 (98.8%)

Genital Chlamydia Treatment Meta-analysis

Thorpe, 1996 [32]

EIA

Open label

1
2

145

257

53/402 (13.2%)

77

118

30/195 (15.4%)

Hillis, 1998 [33]

PCR

Double blind

98

65/261 (24.9%)b

98

65/261 (24.9%)b

93/98 (94.9%)

94/98 (95.9%)

Tan, 1999 [34]

Sendag, 2000 [35]

Culture/EIA
Culture

Open label
Open label

2
2

13
0

0
4

0/13 (0.0%)
No data available

12
0

0
11

0/12 (0.0%)
No data available

13/13 (100%)
4/4 (100%)

12/12 (100%)
10/11 (90.9%)

Skerk, 2001 [36]

Culture/
hybridization
LCR

Open label

38

21/172 (12.2%)b

38

21/172 (12.2%)b

23/38 (60.5%)

28/38 (73.7%)

Open label

24

7/89 (7.9%)b

21

7/89 (7.9%)b

23/24 (95.8%)

19/21 (90.5%)

EIA, culture,
and LCRd

Not stated

11

1/17 (5.9%)

0/12 (0%)

16/17 (94.1%)

12/12 (100%)

4
6

6
6

11
11

4
4

8
8

15/17 (88.2%)
15/16 (93.8%)

12/12 (100%)
12/12 (100%)

EIA

Open label

TMA
NAAT

Double blind
Double blind

6
4

53
51

0
0

58
50

0
0

Rustomjee, 2002 [37]a

Jang, 2003 [38]

Guven, 2005 [39]

Schwebke, 2011 [16]
Manhart, 2013 [15]

2/18 (11.1%)b
12/65 (18.5%)
25/76 (32.9%)

2/18 (11.1%)b
8/66 (12.1%)
18/68 (26.5%)

7/7 (100%)

7/8 (87.5%)

41/53 (77.4%)
44/51 (86.3%)

55/58 (94.8%)
45/50 (90.0%)

Abbreviations: DFA, direct fluorescent antibody; EIA, enzyme immunoassay; LCR, ligase chain reaction; NAAT, nucleic acid amplification test; PCR, polymerase chain reaction; TMA, transmediated amplification.
a

Treatment efficacy for coinfection for chlamydia and gonorrhea.

CID 2014:59 (15 July)

Attrition across entire study.

Used both culture and PCR for evaluating microbiological cure. PCR results used in analysis.

Subjects confirmed positive by EIA or culture and positive by LCR at baseline were recruited. Microbiological cure by LCR and culture.

Treatment efficacy with cotreatment with tinidazole.

197

Figure 2. Difference in treatment efcacy between doxycycline and azithromycin. Abbreviations: CI, condence interval; D + L, DerSimonian and Laird
(random effects) method; ED, efcacy difference; I-squared, test for heterogeneity; MH, MantelHaenzel (xed effects method).

microbiological cure. Seven studies (30%) were based at specialist hospital clinics [19, 24, 25, 31, 35, 36, 39], with the remainder
based at sexual health clinics. Eight studies (35%) were drug
company sponsored [2023, 28, 29, 32, 37], and 5 studies
(22%) [15, 16, 3739] were published after the initial metaanalysis published in 2002 [4].
Treatment Efcacy

The xed effects pooled efcacy difference was 2.6% (95% CI,
.5%4.7%) showing a small but signicantly greater efcacy
for doxycycline, with negligible heterogeneity between studies
(I 2 = 1.9%; P = .435). The random effects estimate was 1.5%
(95% CI, .1% to 3.1%) (Figure 2). The xed effects pooled efcacy for azithromycin was 96.2% (95% CI, 94.9%97.5%), and
the random effects estimate was 94.3% (95% CI, 91.8%96.8%)
(Figure 3). The xed effects pooled efcacy for doxycycline was
97.4% (95% CI, 96.2%98.7%), and the random effects estimate
was 97.1% (95% CI, 95.6%98.6%) (Figure 3). Heterogeneity
between studies was considerably greater for azithromycin

198

CID 2014:59 (15 July)

Kong et al

efcacy (I 2 = 52.4%; P = .002) than doxycycline efcacy

(I 2 = 9.1%; P = .336).
Subgroup analyses (Table 2) found that moderate heterogeneity was associated with men, symptomatic men, doubleblind allocation of treatment, participants with signs or
symptoms of chlamydia, NAAT-based testing, follow-up
times >3 weeks, attrition between 10% and 19%, measurement
of drug compliance, and studies published since 2002. Fixed effect estimates showed signicantly higher efcacy for doxycycline in studies of men, in studies of symptomatic men, in
studies where NAAT-based testing was used, in studies that
were double blind, and in studies published since 2002. No differences in efcacy were observed when stratied by duration of
follow-up, attrition, whether drug compliance was measured, in
trials based on culture/EIA/DFA, or among women.
Between-Study Bias

The funnel plot (Figure 4) showed some asymmetry, with an

absence of small studies reporting higher efcacy with

Figure 3. A, Azithromycin efcacy. B, Doxycycline efcacy. Abbreviations: CI, condence interval; D + L, DerSimonian and Laird (random effects) method;
I-squared, test for heterogeneity; IV, inverse-variance (xed effects) method.
Genital Chlamydia Treatment Meta-analysis

CID 2014:59 (15 July)

199

Table 2. Fixed and Random Effects Pooled Estimates for Difference in Treatment Efcacy, by Subgroup
Fixed
Variable

No.
(Doxycycline)

No.
(Azithromycin)

ED

95% CI

912
341

1147
357

2.6
6.2

.54.7
2.310.1

Overall pooled estimate

Men

Random
ED
1.5
4.3

95% CI

Test for Heterogeneity

I 2 (P Value) and 95% CI

.1 to 3.1
.2 to 8.9

1.9% (.435)
42.2% (.076)

Women

225

225

0.5

4.9 to 5.8

0.6

5.1 to 3.9

0.0% (.657)

Symptomatic onlya
Both symptomatic + asymptomaticb

342
570

349
798

5.2
1.2

.69.9
.8 to 3.2

2.6
1.4

2.2 to 7.3
.4 to 3.1

31.4% (.125)
0.0% (.763)

Symptomatic men

223

234

7.4

2.012.9

Symptomatic women
Culture/EIA/DFA

99
660

98
889

1.1
1.9

9.1 to 11.3
.4 to 4.1

Culture

453

497

2.9

.2 to 6.1

1.6

.8 to 3.9

0.0% (.521)

NAAT/PCR/LCR
Double blinding

252
163

258
169

4.7
8.1

.09.5
1.614.5

3.7
6.4

2.6 to 9.9
4.4 to 17.3

34.9% (.175)
72.5% (.012)

Follow-up 3 wk

554

790

1.9

.4 to 4.2

1.4

.2 to 3.0

0.0% (.873)

Follow-up >3 wk
10% attrition

540
321

580
355

2.6
1.6

.1 to 5.3
1.3 to 4.5

1.9
1.6

1.7 to 5.5
1.1 to 4.3

48.3% (.031)
0.0% (.834)

5.5
3.1
1.3

1.4 to 12.4
12.3 to 6.1
.4 to 2.9

48.6% (.058)
11.0% (.343)
0.0% (.645)

>10% to <20% attrition

335

528

2.9

.5 to 6.4

0.8

5.9 to 7.5

52.6% (.039)

20% attrition
Drug company sponsored

245
488

260
706

3.7
1.6

.8 to 8.2
.5 to 3.6

1.8
1.4

1.7 to 5.4
.4 to 3.1

0.0% (.464)
0.0% (.701)

Not drug company sponsored

424

441

4.0

.07.9

2.2

1.8 to 6.2

17.5% (.258)

Drug compliance measured

Drug compliance not measured

499
413

709
438

2.7
2.4

.45.1
1.3 to 6.1

1.9
1.4

1.2 to 5.1
1.4 to 4.1

33.2% (.175)
0.0% (.560)

Studies 2001 or earlier

763

996

1.8

.2 to 3.9

1.3

.3 to 2.9

0.0% (.761)

Studies 2002 and after

149

151

6.9

.2 to 14.0

4.4

5.1 to 13.9

44.4% (.126)

Abbreviations: CI, confidence interval; DFA, direct fluorescent antibody; ED, efficacy difference; EIA, enzyme immunoassay; LCR, ligase chain reaction; NAAT,
nucleic acid amplification test; PCR, polymerase chain reaction.
a

Randomized controlled trials (RCTs) where all trial participants were reported to have genital symptoms such as urethritis or cervicitis.

RCTs included both symptomatic and asymptomatic participants. It was not possible to separate asymptomatic participants from the studies.

doxycycline than azithromycin. It also showed 4 outlier

studies1 large study favoring doxycycline [16] and 3 small
studies favoring azithromycin [21, 35, 39]. However, Egger test
showed no statistical evidence of publication bias, with a coefcient of .05 (.60 to .70; P = .874).
Within-Study Bias

All studies reported random sequence generation, but 16 (70%)

studies did not provide any information about allocation concealment (Table 3). There was a moderate risk of performance
bias, with the majority of studies (n = 19 [83%]) being open
label. Attrition by treatment arm was available for 15 (65%)
studies [15, 16, 19, 2230, 32, 34, 38], with 6 studies (26%) [15,
23, 25, 26, 29, 33] reporting attrition rates of 20% in each treatment arm. Attrition was comparable between the azithromycin
and doxycycline arms (16.7% vs 17.2%, respectively; P = .75) in
these 15 studies. Sample size calculations were not provided in
20 (87%) studies, comparison of baseline characteristics between trial arms was provided in 16 (70%) studies [15, 16, 19,
22, 23, 25, 28, 29, 3137, 39], and drug compliance was measured

200

CID 2014:59 (15 July)

Kong et al

in 7 (30.4%) studies [16, 22, 24, 29, 32, 33, 37]. Chlamydia was a
secondary outcome in 15 (65%) studies [15, 16, 19, 2123, 25, 26,
28, 29, 31, 34, 35, 37, 39].
Sensitivity Analysis

Removing the 4 outlier studies had negligible impact on the

xed effects pooled efcacy difference (2.0%; 95% CI, .0% to
4.1%), but if only the study by Schwebke et al was excluded [16],
the pooled efcacy difference dropped to 1.7% and was no longer statistically signicant (95% CI, .3 to 3.8%) (Table 4). Similar results were found in other subgroup analyses where the
study by Schwebke et al was excluded.
DISCUSSION
The results of our meta-analysis show a small difference of between 1.5% and 2.6% in favor of doxycycline for the treatment
of urogenital chlamydia infection. We also found that doxycycline may be about 7% more effective for the treatment of
symptomatic urethral infection in men. The World Health

a change to the treatment guidelines. Readers should consider

that >80% of chlamydia infections are asymptomatic; the
xed effects estimate of azithromycin efcacy is >95% (96.2%;
95% CI, 94.9%97.5%); the condence interval of the more conservative random effects estimate is >95% (94.3%; 95% CI,
91.8%96.8%), even though the point estimate is <95%; and
there are considerable limitations with the quality of available
evidence.
We found that doxycycline may be more efcacious for urethral infection in men, particularly for those with symptoms.
Past studies have found that patients with signs and/or symptoms have higher organism loads [4144], with some suggesting
that higher organism load may be related to azithromycin treatment failure. Although the measure of bacterial load in an Australian study was awed, the study suggested that women with
higher organism loads where reinfection had been excluded
were more likely to test positive again within 3 months, raising
the possibility of treatment failure [7]. A mass azithromycin
treatment trial for trachoma observed that 2 months after

Figure 4. Funnel plot. *Four outlier studies in circles.

Organization sexually transmitted infection treatment guidelines recommend using treatments with an efcacy >95%
[40], so readers will need to decide whether our results warrant

Table 3. Summary of Risk of Bias for Included Studiesa,b

Study, First Author, Year

Random
Sequence
Generation

Allocation
Concealment

Blinding of
Participants/
Researchers

Blinding of
Outcome
Assessment

Incomplete
Outcome
Data

Selective
Reporting

Other Bias

Steingrimsson, 1990 [20]

Lassus, 1990 [19]

+
+

+
?

++
++

?
?

+
+

+
+

++
++

Whatley, 1991 [21]

++

++

Martin, 1992 [22]

Nilsen, 1992 [23]

+
+

+
?

++
+

?
?

+
+

+
+

++
+

Ossewaarde, 1992 [24]

++

++

Lauharanta, 1993 [25]

Lister, 1993 [26]

+
+

?
?

++
++

?
?

+
+

+
+

++
++

Hammerschlag, 1993 [27]

++

++

Steingrimsson, 1994 [28]

Stamm, 1995 [29]

+
+

+
+

++
+

?
?

+
+

+
+

+
++

Horner, 1995 [30]

++

++

Gruber, 1996 [31]

Thorpe, 1996 [32]

+
+

?
?

++
++

?
?

+
+

+
+

++
++

Hillis, 1998 [33]

++

Tan, 1999 [34]

Sendag, 2000 [35]

+
+

?
?

++
++

?
?

+
?

+
+

++
++

Skerk, 2001 [36]

++

++

Rustomjee, 2002 [37]

Jang, 2003 [38]

+
+

?
?

++
++

?
?

+
+

+
+

+
++

Guven, 2005 [39]

++

Schwebke, 2011 [16]

Manhart, 2013 [15]

+
+

+
+

+
+

?
?

+
+

+
+

+
+

Abbreviations: +, low risk of bias; ++, medium risk of bias; ?, unclear risk of bias.
a

Assessed using the Cochrane Collaboration Tool for assessing risk of bias.

See Supplementary Table.

Genital Chlamydia Treatment Meta-analysis

CID 2014:59 (15 July)

201

Table 4. Sensitivity Analysis

Fixed

Random

Variable

ED

95% CI

ED

95% CI

Test for Heterogeneity

I 2 (P Value)

Overall pooled estimate

Overall pooled estimate excluding Schwebke et al [16] only

2.6
1.7

.54.7
.3 to 3.8

1.5
1.3

.13.1
.3 to 2.8

1.9% (.435)
0.0% (.844)

Overall pooled estimate excluding 3 small studies [21, 35, 39]

2.9

.85.0

1.7

.2 to 3.6

9.2% (.340)

Overall pooled estimate excluding all 4 studies [16, 21, 35, 39]
Men, excluding Schwebke et al [16] only

2.0
4.1

.04.1
.18.1

1.4
2.4

.2 to 2.9
.7 to 5.4

0.0% (.807)
0.0% (.468)

Men, excluding 3 small studies [21, 35, 39]

6.6

2.710.5

4.7

.09.4

47.0% (.058)

Men, excluding all 4 studies [16, 21, 35, 39]

Symptomatic, excluding Schwebke et al [16] only

4.5
2.9

.58.5
2.1 to 7.9

2.5
0.7

.6 to 5.6
2.8 to 4.3

0.1% (.428)
0.0% (.562)

Symptomatic, excluding 3 small studies [21, 35, 39]

5.9

1.210.6

3.2

1.9 to 8.3

39.3% (.079)

Symptomatic, excluding all 4 studies [16, 21, 35, 39]

Symptomatic men, excluding Schwebke et al [16] only

3.6
4.2

1.5 to 8.6
1.8 to 10.2

1.0
1.8

2.6 to 4.7
2.4 to 6.1

0.0% (.456)
0.0% (.447)

Symptomatic men, excluding 3 small studies [21, 35, 39]

8.1

2.613.6

6.3

Symptomatic men, excluding 4 studies [16, 21, 35, 39]

NAAT/PCR/LCRa, excluding Schwebke et al [16] only

5.0
1.2

1.1 to 11.0
3.8 to 6.1

2.4
1.0

.9 to 13.5
2.3 to 7.1
3.6 to 5.7

54.3% (.041)
5.0% (.384)
0.0% (.859)

Studies 2002 and after, excluding Schwebke et al [16] only

0.6

7.9 to 9.2

0.4

7.6 to 8.5

0.0% (.566)

Double blinding excluding Schwebke et al [16]

Nondrug companysponsored trials excluding Schwebke et al [16] only

3.3
1.9

4.0 to 10.7
2.2 to 6.1

1.1
0.9

3.8 to 5.9
2.6 to 4.4

0.0% (.453)
0.0% (.699)

Abbreviations: CI, confidence interval; DFA, direct fluorescent antibody; ED, efficacy difference; LCR, ligase chain reaction; NAAT, nucleic acid amplification test;
PCR, polymerase chain reaction.
a

All small studies used culture/enyzme immunoassay/direct fluorescent antibody.

treatment, 91% of individuals with a low chlamydia organism

load at baseline had no infection, whereas only 74% with higher
organism loads were infection free at follow-up (P < .01) [45].
Although it is unclear whether antibiotic resistance is a cause
for treatment failure, chlamydia isolates demonstrating heterotypic resistance to macrolides (including azithromycin) in vitro
have been reported [46, 47], but only at high, not low, levels of
organism load [5, 13, 48].
We found no statistical evidence of publication bias using the
Egger test, but the funnel plot showed an absence of published
small studies reporting a higher efcacy for doxycycline; it is
unclear whether this represents any bias. Our subgroup analyses
found some evidence that treatment efcacy favored doxycycline in trials that were not drug company sponsored, raising
questions of whether the drug company funding was a source
of publication bias [49, 50].
We found 4 outlying studies in the funnel plot: 3 small studies (<10 subjects in each treatment arm) [21, 35, 39] and 1 larger
study by Schwebke et al [16]. Schwebke et al contributed considerable heterogeneity to the pooled efcacy difference and when
excluded, the pooled efcacy difference was reduced and nonsignicant (1.7%; 95% CI, .3% to 3.8%). This study of NGU
among men found chlamydia was eradicated in 41 of 53 (77%)
men given azithromycin, compared with 55 of 58 (95%) given
doxycycline (efcacy difference, 17%; 95% CI, 5%30%). These
results were in sharp contrast to a similar study conducted by

202

CID 2014:59 (15 July)

Kong et al

Manhart et al in 2013 that found an efcacy difference of 4%

(95% CI, 9% to 16%) among men with NGU [15]. Both studies were well designed, included a similar population of men,
and were reported according to the Consolidated Standards of
Reporting Trials (CONSORT) statement. The main difference
was that microbiological cure was reported 3 weeks following
treatment in Manhart et al and after 6 weeks in Schwebke
et al, although it was measured at both 3 and 6 weeks in this
latter study.
We obtained further data from Schwebke and colleagues that
showed that men taking doxycycline were more likely to have
microbiological cure measured at 3 weeks rather than at 6
weeks as in the azithromycin arm (38% vs 30%; P = .40). Although this was nonsignicant, it does raise the possibility of
differential follow-up bias and whether there was a greater
risk of reinfection in the azithromycin arm compared with
the doxycycline group. However, it could also be argued that
measuring cure at 3 weeks rather than 6 weeks might lead to
an increased risk of false-positive results due to persistent
DNA without viable chlamydia [51]. This does highlight the
importance of ensuring that trial arms are comparable particularly with respect to follow-up and outcome measurement and
ascertainment.
We would expect to nd that the efcacy difference in favor
of doxycycline has increased since the last meta-analysis in 2002
[4], if increased antibiotic resistance is contributing to

azithromycin treatment failure for chlamydia, as has been observed for Mycoplasma genitalium [52]. Our data suggest that
efcacy may have increased since 2002 to be between 4.4%
and 6.9% higher for doxycycline, but these results were not signicant, there was moderate heterogeneity in the studies and
the results were very sensitive to the inclusion of the study by
Schwebke and colleagues. However, there have been only 5
studies published since the last meta-analysis [15, 16, 3739],
and this smaller sample size (151 in the azithromycin group
and 149 in the doxycycline group) may have reduced the statistical power to nd a signicant efcacy difference.
There are a number of limitations in our meta-analysis. First,
it was limited to published, English-language studies, which
could limit the generalizability of our ndings. We were unsuccessful in obtaining results from 2 drug companysponsored
studies that found no difference in efcacy between azithromycin and doxycycline [53]. Some of our subgroup analyses were
based on a small number of studies, and although the I 2 test for
heterogeneity only reached statistical signicance for 2 subgroups (>3 weeks of follow-up and attrition of 10%20%), moderate, but nonstatistically signicant heterogeneity was
observed for several subgroups (men, symptomatic individuals,
studies using NAATs, studies where compliance was measured,
studies published since 2002). The strengths of our metaanalysis are that we looked for bias within and between studies, and
we conducted a comprehensive subgroup and sensitivity analyses
to investigate whether our pooled estimates were robust.
The quality of the studies varied considerably. Only 4 studies
were double blind, placebo controlled [15, 16, 23, 29], and when
treatment regimens differ substantially as they do here (single
dose vs dosing for 7 days), this can dramatically bias the results.
It is possible that individuals taking 7 days of doxycycline may
delay resuming sex while they are on treatment, whereas individuals taking single-dose azithromycin may resume sex earlier,
making them more susceptible to reinfection. If this happened,
we would expect to see a higher efcacy for doxycycline. Most
studies did not provide sample size estimates, and in nearly twothirds of studies, chlamydia was only a secondary outcome and
may not have been adequately powered to nd a difference if it
existed. Further, the majority of studies were among high-risk
populations, usually sexual health clinic patients; several studies
included symptomatic patients only; and there were few data
available for women, reducing the generalizability of the results.
The relative efcacy of doxycycline vs azithromycin in a more
general, largely asymptomatic population remains unclear.
Only 7 studies measured pill compliance [16, 22, 24, 29, 32,
33, 37], even when compliance with doxycycline has previously
shown to be poor. If doxycycline compliance was poor [54] in
these studies, we would have expected that the difference in efcacy to favor azithromycin, although partial compliance with
doxycycline has been reported to produce high cure rates [3].

Importantly we were unable to nd any RCTs comparing

doxycycline and azithromycin for the treatment of rectal chlamydia infection, despite reports of treatment failure of up to
13% following treatment with 1 g azithromycin [55, 56].
Given the potential risk of human immunodeciency virus
transmission associated with rectal chlamydia infection, it is
vital that efcacious treatment be made available [57]. Given
these concerns, the European guidelines currently recommend
that chlamydia rectal infection be treated with 7 days of doxycycline [58].
CONCLUSIONS
Our meta-analysis showed that there may be a small increased
treatment efcacy of up to 3% in favor of doxycycline for the
treatment of urogenital chlamydia and about 7% increased efcacy for doxycycline for the treatment of symptomatic urethral
infection in men. In considering whether a change to the treatment guidelines is warranted, readers should consider that
>80% of chlamydia infections are asymptomatic and that
there are considerable limitations with the quality of available
evidence. Given increasing concern about possible azithromycin treatment failure, further well-designed double-blind,
placebo-controlled RCTs are warranted. Similarly, RCTs
addressing rectal chlamydia infections are also urgently needed.
Supplementary Data
Supplementary materials are available at Clinical Infectious Diseases online
(http://cid.oxfordjournals.org). Supplementary materials consist of data
provided by the author that are published to benet the reader. The posted
materials are not copyedited. The contents of all supplementary data are the
sole responsibility of the authors. Questions or messages regarding errors
should be addressed to the author.

Notes
Author contributions. J. S. H. conceived the research question, wrote
the protocol, checked extracted data, supervised the analysis and contributed to the manuscript, F. Y. S. K. extracted the data, conducted the analysis
and wrote the manuscript; M. L. supervised the analysis and contributed to
the manuscript; S. N. T., M. C,. R. G., and C. B. contributed to the interpretation of the results and drafting of the manuscript; L. A. V. contributed to
the protocol and drafting of the manuscript; C. K. F. advised on the data
extraction and contributed to drafting the manuscript.
Financial support. This work was supported by an Australian Postgraduate Award for F. Y. S. K. and the National Health and Medical Research
Council (APP1042907 for J. S. H.).
Potential conicts of interest. All authors: No reported conicts.
All authors have submitted the ICMJE Form for Disclosure of Potential
Conicts of Interest. Conicts that the editors consider relevant to the
content of the manuscript have been disclosed

References
1. World Health Organization. Global prevalence and incidence of selected curable sexually transmitted infections. Geneva, Switzerland: WHO,
2012.

Genital Chlamydia Treatment Meta-analysis

CID 2014:59 (15 July)

203

2. Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines, 2010. MMWR Morb Mortal Wkly Rep
2010; 59(RR-12):1116.
3. Bachmann L, Stephens J, Richey C, et al. Measured versus self-reported
compliance with doxycycline therapy for chlamydia-associated syndromeshigh therapeutic success rates despite poor compliance. Sex
Transm Dis 1999; 26:2728.
4. Lau C-Y, Qureshi AK. Azithromycin versus doxycycline for genital
chlamydial infections: a meta-analysis of randomized clinical trials.
Sex Transm Dis 2002; 29:497502.
5. Horner P. The case for further treatment studies of uncomplicated genital Chlamydia trachomatis infection. Sex Transm Infect 2006; 82:
3403.
6. Handseld HH. Questioning azithromycin for chlamydial infection. Sex
Transm Dis 2011; 38:10289.
7. Walker J, Tabrizi SN, Fairley CK, et al. Chlamydia trachomatis incidence
and re-infection among young womenbehavioural and microbiological characteristics. PLoS One 2012; 7:e37778.
8. LaMontagne D, Baster K, Emmett L, et al. Incidence and reinfection
rates of genital chlamydia infection among women aged 16 to 24
years attending general practice, family planning and genitourinary
medicine clinics in England: a prospective cohort study by the chlamydia recall study advisory group. Sex Transm Infect 2007; 83:292303.
9. Batteiger BE, Tu W, Ofner S, et al. Repeated Chlamydia trachomatis
genital infections in adolescent women. J Infect Dis 2010; 201:4251.
10. Golden MR, Whittington WLH, Handseld HH, et al. Effect of expedited treatment of sex partners on recurrent or persistent gonorrhea or
chlamydial infection. N Engl J Med 2005; 352:67685.
11. Horner PJ. Azithromycin antimicrobial resistance and genital Chlamydia trachomatis infection: duration of therapy may be the key to improving efcacy. Sex Transm Infect 2012; 88:1546.
12. Sandoz KM, Rockey DD. Antibiotic resistance in Chlamydiae. Future
Microbiol 2010; 5:142742.
13. Wang S, Papp J, Stamm W, et al. Evaluation of antimicrobial resistance
and treatment failures for Chlamydia trachomatis: a meeting report.
J Infect Dis 2005; 191:91723.
14. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med
2009; 6:e1000097.
15. Manhart LE, Gillespie CW, Lowens MS, et al. Standard treatment regimens for nongonococcal urethritis have similar but declining cure
rates: a randomized controlled trial. Clin Infect Dis 2013; 56:93442.
16. Schwebke JR, Rompalo A, Taylor S, et al. Re-evaluating the treatment of
nongonococcal urethritis: emphasizing emerging pathogensa randomized clinical trial. Clin Infect Dis 2011; 52:16370.
17. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in
meta-analyses. BMJ 2003; 327:55760.
18. Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaborations tool for assessing risk of bias in randomised trials. BMJ 2011; 343:
d5928.
19. Lassus A. Comparative studies of azithromycin in skin and soft-tissue
infections and sexually transmitted infections by Neisseria and Chlamydia species. J Antimicrob Chemother 1990; 25(suppl A):11521.
20. Steingrimsson O, Olafsson JH, Thorarinsson H, et al. Azithromycin in
the treatment of sexually transmitted disease. J Antimicrob Chemother
1990; 25(suppl A): 10914.
21. Whatley JD, Thin RN, Mumtaz G, et al. Azithromycin vs doxycycline in
the treatment of non-gonococcal urethritis. Int J STD AIDS; 1991;
2:24851.
22. Martin DH, Mroczkowski TF, Dalu ZA, et al. A controlled trial of a single dose of azithromycin for the treatment of chlamydial urethritis and
cervicitis. The Azithromycin for Chlamydial Infections Study Group. N
Engl J Med 1992; 327:9215.
23. Nilsen A, Halsos A, Johansen A, et al. A double blind study of single
dose azithromycin and doxycycline in the treatment of chlamydial urethritis in males. Genitourin Med 1992; 68:3257.

204

CID 2014:59 (15 July)

Kong et al

24. Ossewaarde JM, Plantema FH, Rieffe M, et al. Efcacy of single-dose

azithromycin versus doxycycline in the treatment of cervical infections
caused by Chlamydia trachomatis. Eur J Clin Microbiol Infect Dis 1992;
11:6937.
25. Lauharanta J, Saarinen K, Mustonen MT, et al. Single-dose oral
azithromycin versus seven-day doxycycline in the treatment of nongonococcal urethritis in males. J Antimicrob Chemother 1993;
31(suppl E): 17783.
26. Lister PJ, Balechandran T, Ridgway GL, et al. Comparison of azithromycin and doxycycline in the treatment of non-gonococcal urethritis in
men. J Antimicrob Chemother 1993; 31(suppl E): 18592.
27. Hammerschlag MR, Golden NH, Oh MK, et al. Single dose of azithromycin for the treatment of genital chlamydial infections in adolescents.
J Pediatr 1993; 122:9615.
28. Steingrimsson O, Olafsson JH, Thorarinsson H, et al. Single dose
azithromycin treatment of gonorrhea and infections caused by
C. trachomatis and U. urealyticum in men. Sex Transm Dis 1994;
21:436.
29. Stamm WE, Hicks CB, Martin DH, et al. Azithromycin for empirical
treatment of the nongonococcal urethritis syndrome in men. A randomized double-blind study. JAMA 1995; 274:5459.
30. Horner M, Heller VC, Ziegler C, et al. Azithromycin in the treatment of
chlamydial cervicitis and urethritis. Acta Dermatovenerol Alp Panonica
Adriat 1995; 4:1215.
31. Gruber F, Tomic D, Brajac I. Comparative trial with azithromycin and
doxycycline in gonococcal and chlamydial infections in females. G Ital
Dermatol Venereol 1996; 131:4036.
32. Thorpe EM Jr, Stamm WE, Hook EW 3rd, et al. Chlamydial cervicitis
and urethritis: single dose treatment compared with doxycycline for
seven days in community based practises. Genitourin Med 1996;
72:937.
33. Hillis SD, Coles FB, Litcheld B, et al. Doxycycline and azithromycin
for prevention of chlamydial persistence or recurrence one month
after treatment in women. A use-effectiveness study in public health settings. Sex Transm Dis 1998; 25:511.
34. Tan HH, Chan RK. An open label comparative study of azithromycin
and doxycycline in the treatment of non-gonococcal urethritis in males
and Chlamydia trachomatis cervicitis in female sex workers in an STD
clinic in Singapore. Singapore Med J 1999; 40:51923.
35. Sendag F, Terek C, Tuncay G, et al. Single dose oral azithromycin
versus seven day doxycycline in the treatment of non-gonococcal
mucopurulent endocervicitis. Aust N Z J Obstet Gynaecol 2000;
40:447.
36. Skerk V, Schnwald S, Strapa Z, et al. Duration of clinical symptoms in
female patients with acute urethral syndrome caused by Chlamydia trachomatis treated with azithromycin or doxycycline. J Chemother 2001;
13:17681.
37. Rustomjee R, Kharsany ABM, Connolly CA, et al. A randomized controlled trial of azithromycin versus doxycycline/ciprooxacin for the
syndromic management of sexually transmitted infections in a resource-poor setting. J Antimicrob Chemother 2002; 49:8758.
38. Jang D, Sellors J, Howard M, et al. Correlation between culture testing
of swabs and ligase chain reaction of rst void urine from patients recently treated for Chlamydia trachomatis. Sex Transm Infect 2003;
79:2379.
39. Guven MA, Gunyeli I, Dogan M, et al. The demographic and behavioural prole of women with cervicitis infected with Chlamydia trachomatis, Mycoplasma hominis and Ureaplasma urealyticum and the
comparison of two medical regimens. Arch Gynecol Obstet 2005;
272:197200.
40. World Health Organization. Guidelines for the management of sexually
transmitted infections. Geneva, Switzerland: WHO, 2003.
41. Geisler WM, Suchland RJ, Whittington WLH, et al. Quantitative culture
of Chlamydia trachomatis: relationship of inclusion-forming units produced in culture to clinical manifestations and acute inammation in
urogenital disease. J Infect Dis 2001; 184:13504.

42. Eckert LO, Suchland RJ, Hawes SE, et al. Quantitative Chlamydia trachomatis cultures: correlation of chlamydial inclusion-forming units
with serovar, age, sex, and race. J Infect Dis 2000; 182:5404.
43. Michel C-EC, Sonnex C, Carne CA, et al. Chlamydia trachomatis load at
matched anatomic sites: implications for screening strategies. J Clin Microbiol 2007; 45:1395402.
44. Wiggins R, Graf S, Low N, et al. Chlamydia Screening Studies Study Group.
Real-time quantitative PCR to determine chlamydial load in men and
women in a community setting. J Clin Microbiol 2009; 47:18249.
45. West SK, Munoz B, Mkocha H, et al. Infection with Chlamydia trachomatis after mass treatment of a trachoma hyperendemic community in
Tanzania: a longitudinal study. Lancet 2005; 366:1296300.
46. Somani J, Bhullar VB, Workowski KA, et al. Multiple drug-resistant
Chlamydia trachomatis associated with clinical treatment failure. J Infect Dis 2000; 181:14217.
47. Bhengraj AR, Vardhan H, Srivastava P, et al. Decreased susceptibility to
azithromycin and doxycycline in clinical isolates of Chlamydia trachomatis obtained from recurrently infected female patients in India. Chemotherapy 2010; 56:3717.
48. Jones RB, Van Der Pol B, Martin DH, et al. Partial characterization of
Chlamydia trachomatis isolates resistant to multiple antibiotics. J Infect
Dis 1990; 162:130915.
49. Lexchin J, Bero LA, Djulbegovic B, et al. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ 2003;
326:116770.

50. Bourgeois FT, Murthy S, Mandl KD. Outcome reporting among drug
trials registered in ClinicalTrials.gov. Ann Intern Med 2010; 153:
15866.
51. Dukers-Muijrers NHTM, Morr SA, Speksnijder A, et al. Chlamydia
trachomatis test-of-cure cannot be based on a single highly sensitive laboratory test taken at least 3 weeks after treatment. PLoS One 2012; 7:
e34108.
52. Anagrius C, Lor B, Jensen J. Treatment of Mycoplasma genitalium. Observations from a Swedish STD clinic. PLoS One 2013; 8:e61481.
53. Johnson RB. The role of azalide antibiotics in the treatment of chlamydia. Am J Obstet Gynecol 1991;164(6 pt 2):17946.
54. Cerin A, Grillner L, Persson E. Chlamydia test monitoring during therapy. Int J STD AIDS 1991; 2:1769.
55. Drummond F, Ryder N, Wand H, et al. Is azithromycin adequate treatment for asymptomatic rectal chlamydia? Int J STD AIDS 2011;
22:47880.
56. Steedman NM, McMillan A. Treatment of asymptomatic rectal Chlamydia trachomatis: is single-dose azithromycin effective? Int J STD
AIDS 2009; 20:168.
57. Bernstein KT, Marcus JL, Nieri G, et al. Rectal gonorrhea and chlamydia
reinfection is associated with increased risk of HIV seroconversion.
J Acquir Immune Dec Syndr 2010; 53:53743.
58. Lanjouw E, Ossewaarde JM, Stary A, et al. 2010 European guideline for
the management of Chlamydia trachomatis infections. Int J STD AIDS
2010; 21:72937.

Genital Chlamydia Treatment Meta-analysis

CID 2014:59 (15 July)

205

Copyright of Clinical Infectious Diseases is the property of Oxford University Press / USA
and its content may not be copied or emailed to multiple sites or posted to a listserv without
the copyright holder's express written permission. However, users may print, download, or
email articles for individual use.