Update on the Management of Gonorrhea in Adults in the United States

Article (PDF Available)inClinical Infectious Diseases 44 Suppl 3(3):S84-101 · May 2007with49 Reads
DOI: 10.1086/511422 · Source: PubMed
Gonorrhea, the second most commonly reported notifiable disease, is an important cause of cervicitis, urethritis, and pelvic inflammatory disease. The selection of appropriate therapy for gonorrhea (i.e., safe, highly effective, single dose, and affordable) is complicated by the ability of Neisseria gonorrhoeae to develop resistance to antimicrobial therapies. This article reviews the key questions and data that informed the 2006 gonorrhea treatment recommendations of the Centers for Disease Control and Prevention. Key areas addressed include the criteria used to select effective treatment for gonorrhea, the level of antimicrobial resistance at which changing treatment regimens is recommended, the epidemiology of resistance, and the use of quinolones, cephalosporins, and other classes of antimicrobials for the treatment of uncomplicated gonorrhea.
S84 CID 2007:44 (Suppl 3) Newman et al.
Update on the Management of Gonorrhea in Adults
in the United States
Lori M. Newman,
John S. Moran,
and Kimberly A. Workowski
Division of STD Prevention, National Center for HIV/AIDS, Viral Hepatitis, STD and TB Prevention (proposed), and
National Center for
Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, and
Division of Infectious Diseases, Emory University,
Atlanta, Georgia
Gonorrhea, the second most commonly reported notifiable disease, is an important cause of cervicitis, ure-
thritis, and pelvic inflammatory disease. The selection of appropriate therapy for gonorrhea (i.e., safe, highly
effective, single dose, and affordable) is complicated by the ability of Neisseria gonorrhoeae to develop resistance
to antimicrobial therapies. This article reviews the key questions and data that informed the 2006 gonorrhea
treatment recommendations of the Centers for Disease Control and Prevention. Key areas addressed include
the criteria used to select effective treatment for gonorrhea, the level of antimicrobial resistance at which
changing treatment regimens is recommended, the epidemiology of resistance, and the use of quinolones,
cephalosporins, and other classes of antimicrobials for the treatment of uncomplicated gonorrhea.
Infection with Neisseria gonorrhoeae, a gram-negative
diplococcus, is an important cause of cervicitis, ure-
thritis, and pelvic inflammatory disease (PID). Infec-
tion is less frequently found in the pharynx, rectum,
conjunctivae, liver capsule, skin, heart valves, joints,
and meninges, or it can be disseminated in the blood-
stream. Untreated infection can result in infertility, ec-
topic pregnancy, and chronic pelvic pain [1]. Infection
with N. gonorrhoeae has also been associated with in-
creased HIV shedding [2].
Gonorrhea is the second most commonly reported
notifiable disease in the United States [3]. In 2004, a
total of 330,132 new cases of gonorrhea were reported
[4]. The estimated national prevalence of gonococcal
infection among persons 14–39 years of age was 0.25%
in the National Health and Nutrition Examination Sur-
vey (NHANES) and was 0.43% among persons 18–26
years of age in the National Longitudinal Study of Ad-
olescent Health (Add Health) [5, 6]. However, the prev-
Reprints or correspondence: Dr. Lori M. Newman, Centers for Disease Control
and Prevention, Mailstop E-02, 1600 Clifton Rd. NE, Atlanta, GA, 30333
Clinical Infectious Diseases 2007;44:S84–101
2007 by the Infectious Diseases Society of America. All rights reserved.
DOI: 10.1086/511422
alence is higher within specific populations (e.g., 2.1%
among black young adults in Add Health) and in some
areas (e.g., 5.5% among men 16–24 years of age en-
tering the National Job Training Program in Georgia)
[4, 6]. The direct cost of gonorrhea and PID attributable
to gonorrhea was estimated to be approximately $790
million in 1994 dollars [7].
Given the scale of the public health impact of gon-
orrhea, selection of appropriate therapy for gonorrhea
is essential. However, gonorrhea treatment is compli-
cated by the ability of N. gonorrhoeae to develop resis-
tance to antimicrobial therapies. To assist clinicians and
public health practitioners in the selection of gonorrhea
treatment, the Centers for Disease Control and Pre-
vention (CDC) reviews the scientific literature on the
treatment of gonorrhea and consults with experts to
produce national sexually transmitted disease (STD)
treatment guidelines. The present article reviews the key
questions and data behind the recommendations in the
2006 CDC STD treatment guidelines for the treatment
of N. gonorrhoeae infections in adults [8].
A Medline search was conducted in February 2005,
using PubMed, for articles published since 2000 under
the major headings of “gonorrhea/drug therapy,” “gon-
Subscription Information for:
Management of Gonorrhea in Adults CID 2007:44 (Suppl 3) S85
orrhea/therapy,” or Neisseria gonorrhoeae/drug effects.” Addi-
tional searches were performed for related reports posted on the
Internet by internationally recognized public health agencies.
The articles identified through the methods described above
were compiled and presented to a panel of expert consultants
in April 2005. The panel used evidence-based methods to review
the quality of the data that addressed a series of key questions,
and it drafted and approved conclusions and recommendations
for these key areas. The CDC incorporated these expert con-
sultant recommendations into the revised national STD treat-
ment guidelines published in August 2006 [8]. Although the
present article largely reflects material prepared for the April
2005 meeting, Medline searches were updated before guidelines
publication, and preliminary 2006 data (January through June
2006) from the Gonococcal Isolate Surveillance Project (GISP)
were reviewed. Given the importance of the GISP data, a de-
cision was made to further update the August 2006 published
guidelines and include those changes in this article.
It is widely accepted that, whenever possible, the treatment for
gonorrhea should be a safe, highly effective, single-dose, and
affordable regimen [9]. The safety and efficacy of gonorrhea
treatment are important because of the large number of patients
who are treated for gonorrhea and the high proportion of them
who become reinfected and require repeated courses of therapy
[10]. Single-dose oral therapies are recommended to improve
patient compliance and to reduce risk of iatrogenic infection
associated with the use of injections. The large number of gon-
orrhea infections diagnosed and treated each year requires that
gonorrhea therapy be affordable in a wide variety of primary
and specialty care settings [4].
The definition of efficacy for the treatment of gonorrhea,
however, has not been easily delineated. The World Health
Organization has supported the concept that an efficacious
treatment is one that results in 95% of infections being cured,
but it has done so without strongly documented evidence as
to the rationale behind the recommendation [9]. The United
States has also struggled to define acceptable efficacy for the
treatment of gonorrhea. In 1992, Handsfield et al. [11] rec-
ommended that an acceptable clinical efficacy be defined as a
cure rate of 95% with a lower 95% CI of at least 90%. A
more stringent criterion for clinical efficacy, defined as a 95%
cure rate with a sufficient number of subjects studied to es-
tablish a lower 95% CI of 95% in summed clinical trials, was
articulated in a 1995 publication by Moran et al. [12]. At the
time this criterion was proposed, there was a wide variety of
antimicrobial regimens with summed overall cure rates
Therefore, Moran et al. [12] suggested that limiting recom-
mendations to only those regimens with the optimal perfor-
mance (i.e., those with a lower 95% CI of 95%) would per-
haps reduce the risk of therapeutic failure and limit the
development of antimicrobial resistance.
The pharmacokinetic profile of a regimen is also important
in predicting the efficacy of a regimen. Jaffe et al. [13] found
that curing gonococcal urethritis in males by use of penicillin
was best predicted when the serum penicillin concentration
was 3–4 times the MIC for 7–10 h. Subsequent work has eval-
uated the applicability of this standard for other gonorrhea
treatment regimens and for populations rather than individuals
[11, 12]. It is generally accepted that efficacy of a regimen in
a population is likely if the serum concentration of a regimen
is at least 4 times the MIC
for at least 10 h after the peak
concentration is reached [12].
The CDC adopted the more stringent clinical efficacy cri-
terion of 95% efficacy with a lower 95% CI of 95% in the
preparation of the 1993, 1998, and 2002 STD treatment guide-
lines. On the basis of this criterion, ceftriaxone (125 mg in-
tramuscularly [im]), cefixime (400 mg orally [po]), and fluo-
roquinolones (ciprofloxacin [500 mg po], ofloxacin [400 mg
po], and levofloxacin [250 mg po]) were the recommended
regimens in the 2002 STD treatment guidelines for the treat-
ment of uncomplicated gonococcal infections of the cervix,
urethra, and rectum. All patients treated for gonorrhea should
also be treated with a regimen effective against uncomplicated
genital infection with Chlamydia trachomatis unless chlamydial
infection has been ruled out. Alternative regimens in the 2002
guidelines (spectinomycin [2 g im], ceftizoxime [500 mg im],
cefoxitin [2 g im] plus probenecid [1 g po], cefotaxime [500
mg im], gatifloxacin [400 mg po], norfloxacin [800 mg po],
and lomefloxacin [400 mg po]) were regimens that may have
met the stringent criterion but were more expensive, injectable,
or had limited clinical data, compared with the primary rec-
ommended regimens. Because of the emergence of quinolone
resistance, quinolones were not recommended in 2002 for in-
fections acquired in Asia or the Pacific, including Hawaii, as
well as in California and other areas with increased prevalence
of quinolone-resistant N. gonorrhoeae (QRNG) infection [14].
In 2004, the CDC further recommended limiting quinolone
use to heterosexual men and women, on the basis of identi-
fication through GISP and other data sources of a high prev-
alence of QRNG infection among men who have sex with men
(MSM) [15].
The recent widespread emergence of QRNG (see further dis-
cussion below), combined with the recent unavailability of ce-
fixime [16] and a lack of new therapeutic options for the treat-
ment of gonorrhea, means that the more stringent US criteria
may not allow for the use of oral agents in many settings. In
response to these challenges, experts at the 2005 STD treatment
guidelines meeting decided that the primary recommended reg-
imens should meet the stringent criteria but that alternative
Table 1. Selected reports of quinolone-resistant Neisseria gonorrhoeae (QRNG) published between January 2000 and June 2006, sorted by region, population, and year.
Region and population
Year(s) of
data collection
No. of
isolates tested Methodology
Prevalence of
resistant isolates
North America
Canada 2003 NR NR 2.0%; “as high as 12.5% in some
PHAC [20]
Canada 2001 3000–5000 NR 2.1% (Atlantic Canada, 4.4%; central
Canada, 2.1%; western Canada,
Sarwal et al. [21]
Hawaii 2001 267 Agar dilution, disk
diffusion, CLSI
20% Newman et al. [22]
Michigan 2004 540 Disk diffusion, Etest, CLSI 1.5% Macomber [23]
Michigan 2003 582 Disk diffusion, Etest, CLSI 2.9% Macomber [23]
United States (GISP) 2004 6322 Agar dilution, CLSI Overall, 6.8%; MSM, 23.8%; hetero-
sexual men, 2.9%; when data from
CA, HI, and MSM were excluded,
CDC [24]
South America/Caribbean
Argentina 2000 1 Agar dilution, CLSI 1 case of QRNG (MIC, 16 mg/mL) Fiorito et al. [25]
Argentina 1995–1996 81 Agar dilution, CLSI 0% Famiglietti et al. [26]
Brazil 1998 81 Agar dilution, CLSI 0% Dillon et al. [27]
Cuba 1995–1999 120 Agar dilution, CLSI 0% Llanes et al. [28]
Trinidad and Tobago 1999 128 Agar dilution, CLSI 0% Castor et al. [29]
Trinidad, Guyana, St. Vincent 1994–1995 282 Agar dilution, CLSI 0% Dillon et al. [30]
Argentina, Chile, Colombia, Peru,
Uruguay, and Venezuela
1990–1999 2806 Agar dilution, CLSI 2 resistant isolates in Venezuala in
1995; 2 resistant isolates in Uruguay
in 1997
Dillon et al. [31]
Austria 2002 202 Disk diffusion, CLSI 59.4% Uthman et al. [32]
Denmark 1997 177 Agar dilution, CLSI 5.6% Su and Lind [33]
Denmark 1998 197 Agar dilution, CLSI 6.1% “due to importation of resistant
Su and Lind [33]
England and Wales (GRASP) 2004 1744 Agar dilution Overall, 14.1%; women, 5.0%; hetero-
sexual men, 10.7%; MSM, 27.1%
GRASP [34]
England, London 2003 952 Agar dilution 7.9%; QRNG is “spreading endemically
in high risk groups”
Martin et al. [35]
Finland (FiRe) 1998 118 Disk diffusion, CLSI 5.1% Nissinen and Huovinen [36]
France (RENAGO) 2001–2003 473 Etest, CLSI 9.7% Herida [37]
Germany 2004–2005 65 Etest, CLSI 47.7% Enders [38]
Greenland 1998–1999 61 Agar dilution, Etest, disk
0% Dragsted et al. [39]
Ireland 2003–2004 158 Disk diffusion, CLSI 7% Hopkins [40]
Netherlands 2003 772 Disk diffusion 7.3%; MSM, 10.5%; heterosexual men,
3.4%; women, 0%
Kolader et al. [41]
Netherlands 2004 1529 Disk diffusion, Etest, CLSI,
and Dutch guidelines
14.9% Borgen [42]
Scotland 1999 540 Agar dilution, Etest 2.2% Forsyth et al. [43]
Sweden 2003 NR NR “Increased ciprofloxacin resistance
among heterosexual men and
Berglund et al. [44]
Sweden 1998–1999 348 Etest Resistant and intermediate suscepti-
bility: overall, 18%; if exposed in
Sweden, 8%; if exposed in Asia,
Berglund et al. [45]
Spain 2001 81 Agar dilution, CLSI 9.9% Arreaza et al. [46]
Turkey 1998–2002 78 Agar dilution, CLSI 1.3% Aydin et al. [47]
Africa/Middle East
Benin 1998–1999 143 Agar dilution 0% Van Dyck et al. [48]
Morocco 2001 154 Agar dilution, CLSI 2.6% Alami et al. [49]
Israel 2000 100 Etest, CLSI 61% Dan et al. [50]
Israel 2000 22 Disk diffusion, Etest, CLSI 54.5% Yagupsky et al. [51]
Kuwait 2003 44 Disk diffusion, CLSI 36.4% Sharma [52]
Rwanda 1999–2000 139 Agar dilution, CLSI 0% Van Dyck et al. [53]
Saudi Arabia 1996–2000 93 Disk diffusion, CLSI
!1% Balkhy et al. [54]
South Africa, Durban 2005 248 NR 42.0% Moodley [55]
Asia/Central Asia/South Pacific
Australia 2005 3886 Agar dilution 28.6% AGSP [56]
Australia, Victoria 2001 645 Agar dilution, CLSI 11% Veitch et al. [57]
Bangladesh 2003 63 NR 90.5% ICDDR,B [58]
China 2002 100 Agar dilution 98% Fei et al. [59]
India 2000–2001 241 Etest, WHO 67.3% Bala et al. [60]
Indonesia 1996 267 Agar dilution, CLSI 0% Ieven et al. [61]
Indonesia, Bali 2004 147 Agar dilution, CLSI 40.1% Donegan et al. [62]
Japan 2002 221 Agar dilution 78.3% Ito et al. [63]
Kyrgyzstan 1999–2000 120 Agar dilution, CLSI Cpfx resistance, 10.0%; Ofx resis-
tance, 15.0%
Dorlencourt et al. [64]
Mongolia Before 2001 56 NR 25.0% Lkhamsuren et al. [65]
Nepal 2003 16 Agar dilution, CLSI 87.5% Chaudhary [66]
New Zealand 2002 413 Agar dilution 6.8% Hefferman et al. [67]
Pakistan, Karachi 2002 26 Etest, CLSI 42% Jabeen [68]
Philippines 1996–1997 115 Agar dilution, Etest 63% Aplasca De Los Reyes et al.
Taiwan 2003 29 Agar dilution, CLSI 93.1% Hsueh et al. [70]
Thailand 1999 110 Agar dilution, CLSI 25.4% Trees et al. [71]
Australia 2004 3542 Various 21.4% WHO WPGASP [72]
Brunei 2004 113 Various 40.7% WHO WPGASP [72]
China 2004 1203 Various 94.3% WHO WPGASP [72]
Hong Kong 2004 2811 Various 94.2% WHO WPGASP [72]
Japan 2004 261 Various 81.6% WHO WPGASP [72]
Korea 2004 93 Various 70.0% WHO WPGASP [72]
Laos 2004 48 Various 88.0% WHO WPGASP [72]
New Zealand 2004 773 Various 19.1% WHO WPGASP [72]
Papua New Guinea 2004 92 Various 1.0% WHO WPGASP [72]
Philippines 2004 175 Various 47.4% WHO WPGASP [72]
Singapore 2004 160 Various 50.0% WHO WPGASP [72]
Vietnam 2004 156 Various 52.5% WHO WPGASP [72]
Bangladesh 1999–2000 110 Disk diffusion, CLSI 76% Ray [73]
India, Chennai 2001 80 Disk diffusion, CDS 11.2% Ray [73]
India, New Delhi, SJH 2001 108 Disk diffusion, CDS 88.4% Ray [73]
India, New Delhi, MAMC 2001 10 Disk diffusion, CLSI 100% Ray [73]
India, Hyderabad 2001 46 Disk diffusion, CDS 57% Ray [73]
India, Kolkata 2001 58 Disk diffusion, CLSI 22.4% Ray [73]
India, Nagpur 2001 74 Disk diffusion, CDS 35.1% Ray [73]
India, Pune 2001 37 Disk diffusion, CLSI 10.6% Ray [73]
Nepal 2001 9 Disk diffusion, CLSI 11.1% Ray [73]
Sri Lanka 2000 235 Disk diffusion, CLSI 8.2% Ray [73]
NOTE. Italics indicate a QRNG prevalence 2%; bold italics indicate a QRNG prevalence 5%. AGSP, Australian Gonococcal Surveillance Programme; CDS, calibrated dichotomous sensitivity; CLSI, Clinical and
Laboratory Standards Institute; Cpfx, ciprofloxacin; FiRe, Finnish Study Group for Antimicrobial Resistance; GISP, Gonococcal Isolate Surveillance Project; GRASP, Gonococcal Resistance to Antimicrobials Surveillance
Programme; ICDDR,B, International Centre for Diarrhoeal Disease Research, Bangladesh; MAMC, Maulana Azad Medical College; MSM, men who have sex with men; NR, not reported; Ofx, ofloxacin; PHAC, Public
Health Agency of Canada; RENAGO, National Reference Centre for Neisseria gonorrhoeae; SJH, Safdarjang Hospital; WHO, World Health Organization; WPGASP, Western Pacific Gonococcal Antimicrobial Surveillance
Data are the percentage of resistant isolates, unless otherwise indicated.
Management of Gonorrhea in Adults CID 2007:44 (Suppl 3) S89
regimens could include those with a clinical efficacy of 95%
and a lower 95% CI of 90%.
In addition to the intrinsic efficacy of a regimen against fully
susceptible strains, selection of a recommended regimen must
also take into account the epidemiology of antimicrobial re-
sistance, because infection with a resistant strain can result in
treatment failure. The World Health Organization has suggested
that an antimicrobial should not be used when
15% of strains
demonstrate resistance [9]. Although, ideally, the assessment of
“resistance” would be based on the clinical failure rate, in reality
the most readily available data are generally in vitro data, which
have a close but not perfect correlation with clinical failure
rates. In 1987, in response to an increasing prevalence of pen-
icillinase-producing N. gonorrhoeae (PPNG), the CDC pro-
posed lower thresholds for modifying treatment recommen-
dations; such recommendations were based on input from
subject-matter experts, because there was a paucity of available
evidence [17]. A “nonendemic area” was defined as an area
!1% of gonorrhea reported in a 2-month period was
caused by PPNG. In nonendemic areas, a basic control program
(e.g., antimicrobial susceptibility testing of treatment failures
and gonococcal isolates from public laboratories and routine
treatment with a regimen effective against penicillin-susceptible
strains) was sufficient. In “endemic areas,” those areas where
1%–3% of gonorrhea was caused by PPNG, it was recom-
mended that antimicrobial susceptibility testing of specimens
in both private and public laboratories be conducted and that
only providers whose patients with gonorrhea were from areas
with known increased prevalence of PPNG should be encour-
aged to switch regimens. A “hyperendemic area” was defined
as one in which PPNG accounted for
13% of all gonococcal
strains. In hyperendemic areas, all providers should be en-
couraged to switch regimens in addition to performing anti-
microbial susceptibility testing for all patients. Even in the
1980s, when gonococcal culture was widely used to diagnose
gonorrhea, these guidelines were impractical for many areas
because they required more knowledge of local susceptibility
patterns than was generally available. Although GISP now pro-
vides routine susceptibility data from 30 sites around the
United States, the dramatic shift in recent years to the use of
nonculture techniques for the diagnosis of gonorrhea means
that guidelines based on knowledge of prevalence of resistance
are still impractical for most areas in the United States [18].
A more comprehensive approach to identifying when to
switch treatment strategies requires knowledge of not only the
prevalence of resistance but also of other factors, such as gon-
orrhea prevalence and the cost of drugs and diagnostic testing.
For example, Roy [19] modeled cost-minimizing strategies
from the health care system perspective for both diagnosis (cul-
ture followed by antimicrobial susceptibility testing vs. non–
culture-based testing) and treatment (ciprofloxacin vs. ceftriax-
one) of gonorrhea in women. Her analysis suggested that
switching from ciprofloxacin to ceftriaxone was optimal when
the prevalence of gonorrhea was
13% and the prevalence of
ciprofloxacin resistance was
15%. However, models such as that
used by Roy [19] cannot take into account such factors as the
benefits of keeping in reserve an already existing antimicrobial
agent, such as ceftriaxone, for as long as economically feasible
in a setting in which alternative regimens are limited.
On the basis of the evidence and opinions presented above,
the CDC STD treatment guidelines expert consultants felt that
the CDC should continue to use 5% as the resistance threshold
at which an antimicrobial therapy should no longer be routinely
recommended. The decision of whether or not to abandon quin-
olones when QRNG levels are 3%–5% should take into consid-
eration other