Referral recommendations are based on expert opinion in the National guideline on the diagnosis and treatment of gonorrhoea in adults, published by the British Association for Sexual Health and HIV (BASHH), which states that referral to a genito-urinary medicine clinic is 'strongly encouraged' [BASHH, 2005b].

Referral

• Genito-urinary medicine clinics and general practices providing enhanced sexual health services have the resources to ensure effective diagnosis and treatment of gonorrhoea, as well as screening for other sexually transmitted infections (including HIV), counselling, follow up, and contact tracing (partner notification).
• Men who are suspected of having complications caused by gonorrhoea require specialist management (for example extended courses of antibiotics) [Handsfield and Sparling, 2005; Bignell, 2009].
• Gonorrhoea in pregnancy is associated with spontaneous abortion, premature labour, early rupture of fetal membranes, perinatal mortality, and gonococcal conjunctivitis in the newborn [Handsfield and Sparling, 2005].
• Pelvic inflammatory disease requires treatment with intramuscular ceftriaxone, which is not always available in primary care [BASHH, 2005a; RCOG, 2009].

Admission

• Admission is required for disseminated gonorrhoea because it can develop into life-threatening infection (for example gonococcal meningitis). Treatment in secondary care will typically involve higher doses of an intramuscular or intravenous cephalosporin for up to a week [Bignell, 2009].
• Pelvic inflammatory disease that is severe or has complications requires admission for specialist management, such as intravenous antibiotics [BASHH, 2005a; RCOG, 2008].

Sexually transmitted infection and HIV screening

• Screening for other sexually transmitted infections is in line with guidelines from BASHH and the Health Protection Agency [BASHH, 2006; RCGP and BASHH, 2006; HPA, 2008].

Treatment recommendations are based on the National guideline on the diagnosis and treatment of gonorrhoea in adults, published by the British Association for Sexual Health and HIV (BASHH) [BASHH, 2005b], a primary care guideline [RCGP and BASHH, 2006], and a European guideline [Bignell, 2009].

Treatment with antibiotics

• For ethical reasons, the efficacy of antibiotics for gonorrhoea has not been established compared with placebo. However, historical evidence suggests that left untreated, Neisseria gonorrhoeae will cause prolonged infectivity and symptoms over several weeks, with a risk of chronic complications [Handsfield and Sparling, 2005].
• Good evidence from several comparative randomized controlled trials (RCTs) supports the use of antibiotics in the treatment of gonorrhoea. Historically, cephalosporins, fluoroquinolones, and macrolides have produced microbiological cure rates in excess of 95% in these trials [Moran and Levine, 1995].

Choice of antibiotics

• Since the introduction of sulphonamides in the 1930s, there has been growing evidence of gonococcal resistance to antibiotics [Workowski et al, 2008]. Therefore, antibiotic recommendations are mainly based on current gonococcal sensitivity to antibiotics, rather than historical evidence of their efficacy from RCTs [Tyson, 2005].
• Cephalosporins are recommended first line because N. gonorrhoeae is currently sensitive to this drug class [HPA, 2010a]. Cefixime and ceftriaxone are third-generation cephalosporins that have shown evidence of a cure rate greater than 95% in RCTs (the accepted level of effectiveness) [Moran and Levine, 1995].
• The use of cefixime for the treatment of gonorrhoea is off-label [ABPI Medicines Compendium, 2008a], but one RCT showed it to be effective at a single oral dose of 400 mg (cure rate 96%, 95% CI 94 to 98) [Handsfield, 1991].
• Ceftriaxone is less convenient than cefixime but was shown to give a 100% cure rate in one trial of people with uncomplicated infection [Zajdowicz et al, 1983], and in one trial which included people with pharyngeal infection [Christophersen et al, 1989].
• Confirmed therapeutic failure to cefixime is rare, and is undocumented for ceftriaxone in England and Wales [HPA, 2010a]. However, in 2009, 1.2% of gonococcal isolates demonstrated reduced susceptibility to cefixime, and 0.3% to ceftriaxone.
• For treatment of pharyngeal gonorrhoea, BASHH recommends ceftriaxone first line [BASHH, 2005b]. However, because ceftriaxone is not usually readily available in primary care, some experts recommend an alternative 3-day regimen of cefixime on the basis of pharmacokinetic principles [Horner, Personal Communication, 2009]. This is reasonable, as a test of cure is always recommended for pharyngeal gonorrhoea (see Follow-up).
• Ciprofloxacin and ofloxacin are fluoroquinolones that are recommended if cephalosporins are contraindicated, most commonly because of documented beta-lactam allergy (allergy to penicillins, cephalosporins, and associated antibiotics). Between 1% and 10% of people with penicillin allergy are thought to have cross-sensitivity to cephalosporins [Workowski et al, 2008].
• Good evidence from RCTs indicates that fluoroquinolones have been effective in the treatment of gonorrhoea [Bignell, 1996]. However, evidence from surveillance data indicates that there is endemic resistance of gonorrhoeal isolates to ciprofloxacin, with a prevalence of around 35% [HPA, 2010a]. Therefore, it is important that the sensitivity of the infection is known before treatment with a fluoroquinolone is prescribed.
• The manufacturers of ciprofloxacin (Ciproxin^®) and ofloxacin (Taravid^®) state that fluoroquinolones should be avoided in people with tendonitis or epilepsy [ABPI Medicines Compendium, 2008b; ABPI Medicines Compendium, 2009]. However, the adverse effects of fluoroquinolones are not likely to be significant, as they are given as a single dose for the treatment of gonorrhoea.

Antibiotics not recommended for confirmed gonorrhoea

• Ampicillin combined with probenecid is recommended as an alternative regimen by BASHH, provided that bacterial sensitivity is known [BASHH, 2005b], but this is an impractical option in primary care because of the limited availability of probenecid [BNF 57, 2009]. Likewise, spectinomycin is recommended for empirical treatment but is not widely available in primary care in the UK.
• Other cephalosporins are options but have not been as extensively studied as ceftriaxone, and they do not offer any additional benefits [BASHH, 2005b].
• Azithromycin is not recommended by BASHH, despite having shown adequate efficacy in RCTs. The guideline development group state 'the emergence of azithromycin resistant N. gonorrhoeae has been reported and clinical efficacy does not always correlate with in vitro sensitivity testing'. In addition, there are concerns cited that high-dose azithromycin is associated with gastrointestinal intolerance [BASHH, 2005b].

Empirical treatment

• As it is not possible to accurately differentiate between infection with N. gonorrhoeae and Chlamydia trachomatis by clinical features alone, and the infections often coexist, empirical treatment aims to cover both organisms [RCGP and BASHH, 2006].
• Azithromycin and doxycycline have both been shown to be effective in the treatment of chlamydia by RCTs [Lau and Qureshi, 2002], and they are recommended by national guidelines for chlamydia [BASHH, 2006] and non-gonococcal urethritis [BASHH, 2007].

Treatment recommendations for women who are pregnant or breastfeeding are based on the National guideline on the diagnosis and treatment of gonorrhoea in adults, published by the British Association for Sexual Health and HIV (BASHH) [BASHH, 2005b], and a European guideline [Bignell, 2009].

Treatment of confirmed gonorrhoea

• There is evidence from a meta-analysis of two comparative randomized controlled trials that cefixime and ceftriaxone are effective in the treatment of confirmed gonorrhoea in pregnant women [Brocklehurst, 2002].
• Amoxicillin (combined with probenecid) has also been shown to be effective, but it is probably not as effective as cephalosporin antibiotics [Brocklehurst, 2002], and is evidence that Neisseria gonorrhoeae is becoming increasingly resistant to penicillin-based antibiotics [GRASP Steering Group, 2007].

Empirical treatment of gonorrhoea

• Empirical treatment aims to provide coverage for both Neisseria gonorrhoeae and Chlamydia trachomatis, as these infections often coexist and cannot be accurately distinguished by clinical features alone [RCGP and BASHH, 2006].
• Azithromycin, erythromycin, or amoxicillin is recommended by BASHH for the treatment of chlamydia in pregnant women [BASHH, 2006]. These antibiotics have been shown to be effective in controlled trials [Brocklehurst and Rooney, 1998] and are not contraindicated in pregnancy or breastfeeding.
• Penicillin has been shown to induce latency in vitro, and some experts are concerned there is a theoretical risk of re-emergence of infection, However, amoxicillin has been found to be as effective as erythromycin in a meta-analysis of trials, and has considerably less adverse effects [BASHH, 2006].

Safety of antibiotics in pregnancy

• The benefits of antibiotics for gonorrhoea outweigh the risks in pregnancy.
• Left untreated, gonorrhoea in pregnancy is associated with spontaneous abortion, premature labour, early rupture of fetal membranes, perinatal mortality, and gonococcal conjunctivitis in the newborn [Handsfield and Sparling, 2005].
• Chlamydia is associated with premature birth and spontaneous abortion in pregnant women [Stamm et al, 2005] and may develop into pelvic inflammatory disease, which increases the risk of ectopic pregnancy, infertility, and chronic pelvic pain in later life [BASHH, 2006].
• Cefixime and ceftriaxone are not specifically licensed in pregnancy but 'are not known to be harmful' [BNF 57, 2009]. Although cephalosporins cross the placenta, no evidence from animal studies indicates that they are toxic to the embryo or the fetus, and no harms have been shown in observational studies of pregnant women who have been exposed to cephalosporins [Schaefer et al, 2007; NTIS, 2008d].
• Amoxicillin is considered to have a good safety profile in pregnancy, and its use is licensed in this setting [ABPI Medicines Compendium, 2008c]. There is extensive experience of using penicillins in pregnancy, with no evidence to suggest that they are associated with an increased risk of malformations or other forms of fetal toxicity [NTIS, 2008c].
• Macrolides are considered to be suitable treatment options in pregnancy where indicated [NTIS, 2008c].
• Erythromycin is generally preferred because there is more experience with it than with other macrolides [NTIS, 2008b]. However, it is probably not as effective as azithromycin in the treatment of chlamydia [Brocklehurst and Rooney, 1998].
• Azithromycin is usually recommended as second-line treatment in pregnancy; animal studies and limited observational data have not indicated an increased risk of congenital malformations [NTIS, 2008a].

Drugs not recommended

• Fluoroquinolones are not recommended in pregnancy or breastfeeding except 'for the treatment of serious or life-threatening conditions unresponsive to standard antibiotic therapy', because there is a theoretical risk of arthropathy in the child [Schaefer et al, 2007; NTIS, 2008c].
• Tetracyclines have been associated with discolouration of teeth and bones, staining of the lenses, and the development of cataracts in newborn babies when used in the second and third trimesters. They are therefore contraindicated in pregnancy and breastfeeding [Schaefer et al, 2007; NTIS, 2008c].

Recommendations for providing information and advice are based on expert opinion from the National guideline on the diagnosis and treatment of gonorrhoea in adults, published by the British Association for Sexual Health and HIV [BASHH, 2005b].

Recommendations for partner notification are based on expert opinion from the National guideline on the diagnosis and treatment of gonorrhoea in adults, published by the British Association for Sexual Health and HIV [BASHH, 2005b].

Partner notification in primary care

• There is expert consensus that if referral to a GUM clinic (or to a general practice providing an enhanced sexual health service) is not possible, contact tracing should be undertaken in primary care, and this should be documented [Fitzgerald et al, 1996].

Period of partner notification

• The recommended periods for notification are consistent with what is known about the natural course of symptomatic infection with gonorrhoea [Handsfield and Sparling, 2005]. However, there is less certainty about the natural course of asymptomatic infection or infection at sites other than the genitals and rectum, and a period of 3 months is therefore advised as a precautionary measure [BASHH, 2005b].

Method of partner notification

• A Cochrane systematic review (search date: around 2001) found evidence that both provider and contract referral methods result in more partners presenting for medical evaluation compared with patient referral [Mathews et al, 2001]. However, in practice, provider referral can be difficult to organize, and most GUM clinics in the UK use patient referral as the contact method of first choice [Stokes and Schober, 1999].

Recommendations on follow up are based on the National guideline on the diagnosis and treatment of gonorrhoea in adults, published by the British Association for Sexual Health and HIV [BASHH, 2005b], a primary care guideline [RCGP and BASHH, 2006], and a European guideline [Bignell, 2009].

• Anogenital gonorrhoea usually responds well to antibiotics to which it is known to be sensitive, and test of cure is not necessary unless symptoms persist.
• CKS recommends a routine test of cure for pregnant women, as the infection can have serious consequences for the mother and baby if it persists (including spontaneous abortion, premature labour, early rupture of fetal membranes, perinatal mortality, and gonococcal conjunctivitis in the newborn) [Handsfield and Sparling, 2005].
• Pharyngeal gonorrhoea responds less well than anogenital gonorrhoea to treatment with antibiotics [Moran and Levine, 1995] and may cause no symptoms. Therefore, a test of cure is reasonable.
• Swabbing for culture is recommended, as this allows sensitivity testing (which should be done at least 3 days after antibiotic treatment to avoid false-positive results). Nucleic acid amplification tests do not provide information on sensitivity and require a longer washout period [BASHH, 2005b].