Antibiotics for uncomplicated urinary tract infection in women

Drug review UTI Antibiotics for uncomplicated urinary tract infection in women Sarah Thompson BSc, MB ChB, Sarah Yacomeni BSc, MB ChB and Robert Townsend MSc, MRCPath, DTM&H Most well women with uncomplicated urinary tract infection should respond to three days treatment with a first-line antibiotic. Our Drug review describes the properties and side-effects of the recommended antibiotics, followed by sources of further information. Every year, around 5 per cent of women present to their GP with symptoms of dysuria and frequency, and approximately half of this number have confirmed urinary tract infection (UTI) upon laboratory investigation. This makes UTI one of the most common reasons for both presentation to primary care (especially in otherwise well women) and for anti biotic prescription. 1 Acute uncomplicated UTI is defined as UTI in a person with a structurally and functionally normal urinary tract. In contrast, a UTI can be considered complicated if there are functional or anatomical abnormalities within the urinary tract or underlying diseases that predispose to UTI, eg diabetes mellitus. 2 Acute uncomplicated UTI is predominantly observed in women, and recurrent cystitis is seen mostly in young, healthy women with normal urinary tracts (though the shorter female urethra is a predisposing factor to UTI). Complicated UTI may occur in anyone with predisposing illness or underlying abnormalities and www.prescriber.co.uk Prescriber 5 April 2010 15

treatment options may be more problematic, having to take into account the underlying condition. We will therefore restrict our discussion to the management of uncomplicated UTI. Bacterial spectrum The majority of UTIs are caused by Escherichia coli (over 70 per cent generally), with approximately 10-15 per cent caused by other enterobacteriaceae (such as Proteus and Klebsiella) and faecal streptococci (Enterococcus). Faecal flora account for most of the uropathogens encountered, probably utilising a mechanism of initial colonisation of the perineum, followed by passage up the female urethra (a process often aided by sexual intercourse). The skin commensal Staphylococcus saprophyticus accounts for around 5-10 per cent of UTI cases in young women. Treatment options There is evidence that untreated, uncomplicated cystitis in nonpregnant women appears not to be associated with any long-term sequelae, ie impaired renal function or increased mortality, and that untreated recurrent cystitis rarely progresses to an upper UTI. 3 The treatment of uncomplicated UTI is therefore aimed at the morbidity associated with potentially distressing (and disrupting) symptoms. Empirical therapeutic decisions should be guided by local knowledge of resistance patterns of the likely uropathogens. The trend away from sending uncomplicated UTI urine samples for culture and sensitivity means that these data may be lacking or incomplete. So, for example, if only urine from treatment failures is examined, then a skewed population of more resistant organisms may be seen (the more sensitive organisms would be less likely to fail treatment). One approach is to send urine for culture and at the same time prescribe a first-line urinary antibiotic: this not only provides good sensitivity data on which to base future resistance trends, but also a prepared result if the patient fails treatment. Short courses of antibiotic appear to be highly effective in the treatment of uncomplicated UTI and are desirable as they improve compliance, cost less and are associated with fewer adverse reactions. 4 Antibiotic Contraindications Common side-effects Rare but serious side-effects Trimethoprim trimethoprim hypersensitivity, nausea, vomiting, hypersensitivity, severe skin rash, blood dyscrasias, severe renal pruritus, rash blood dyscrasias impairment, pregnancy, porphyria Nitrofurantoin nitrofurantoin hypersensitivity, anorexia, nausea hypersensitivity, acute and chronic renal impairment, G6PD pulmonary reactions, hepatitis, deficiency, <3 months of age, peripheral neuropathy, severe skin porphyria rashes, pancreatitis, blood dyscrasias, neurological disturbance Amoxicillin and penicillin hypersensitivity, nausea, vomiting, hypersensitivity, severe skin rash, co-amoxiclav penicillin-associated hepatic diarrhoea, rash pseudomembranous colitis, dysfunction hepatitis, blood dyscrasias Cephalosporins cephalosporin hypersensitivity, nausea, vomiting, hypersensitivity, severe skin rash, porphyria diarrhoea, rash pseudomembranous colitis, blood dyscrasias Quinolones quinolone hypersensitivity, nausea, diarrhoea, neurological disturbance children, pregnant or vomiting, abdominal (including hallucinations and lactating women pain, headache, rest- psychosis), hypersensitivity, photolessness, rash, pruritus sensitivity, tendon inflammation and damage, hepatic necrosis, severe skin rash, blood dyscrasias Table 1. Contraindications and common and rare side-effects of antibiotics prescribed in UTIs 16 Prescriber 5 April 2010 www.prescriber.co.uk

Antibiotic Trimethoprim Nitrofurantoin Amoxicillin and co-amoxiclav Quinolones Cefalexin Interactions warfarin, phenytoin, digoxin, antimalarials, ciclosporin, cytotoxics antacids warfarin, oral contraceptives NSAIDs, antacids, warfarin, phenytoin, ciclosporin, iron, theophylline warfarin, oral contraceptives Table 2. Drug interactions with antibiotics used to treat UTIs A study of 75 clinical trials by the Infectious Diseases Society of America produced guidelines that were later reviewed by the American Urological Association and the European Society of Clinical Microbiology and Infectious Diseases (ESCMID). 5 In summary, they concluded: trimethoprim, co-trimoxazole and quinolones given for three days was as effective as treatment for a longer duration single-dose regimens appear to be less effective than the same antibiotic given for longer and should not be recommended longer duration of treatment usually causes higher rates of adverse events although co-trimoxazole was the most commonly studied drug, it was found that trimethoprim alone was equivalent to co-trimoxazole in eradication and, given the side-effect profile of sulphonamides, trimethoprim alone may be considered the preferred drug fluoroquinolones showed a similar three-day efficacy to co-trimoxazole/trimethoprim; however, given the greater expense and possibility of higher incidence of adverse events, they are not recommended as first-line agents for empirical therapy. 2 Resistance There is a trend towards increasing rates of resistance in uropathogens. Extended-spectrum beta-lactamase (ESBL)-producing Gram-negatives are a particular problem, usually conferring resistance to all penicillins and cephalosporins. In the community setting, prior antibiotic exposure is the single largest risk factor for acquisition of an ESBL-producing organism. Such resistance can be spread between different Gram-negative bacteria on a mobile section of RNA (a plasmid). Plasmids can also carry mutations giving rise to resistance to other classes of antibiotics, eg fluroquinolones. UTIs untreatable with currently available oral antibiotics are rare but do occur. Such infections clearly cause significant management difficulties. The increasing problems with resistant organisms highlights the importance of vigilance regarding antimicrobial prescribing, and also of knowing your local resistance patterns. Supportive treatments General measures shown to be of benefit include increased fluid intake (to increase urinary output) and an analgesic (to control pain and/or fever). Oral agents that alkalinise the urine may also be of benefit and may help to alleviate symptoms. Antibiotics Antibiotics commonly used in uncomplicated UTI include trimethoprim, nitrofurantoin, beta-lactam antibiotics and quinolones. Their contraindications and side-effects are shown in Table 1, and their drug interactions in Table 2. Trimethoprim Trimethoprim is a dihydrofolate reductase inhibitor first introduced to augment the activity of sulphon - amides by sequential inhibition of folic acid synthesis, and is the commonest first-line therapy. Sensitivity of enterobacteriaceae to this antibiotic varies according to location and whether primary- or secondary-care populations are examined. Trimethoprim has no activity against Pseudomonas and variable activity against enterococci. Clinically, the most important method of resistance is by the use of plasmid-mediated dihydrofolate reductases that are resistant to the effects of this antibiotic. Antibiotic resistance to trimethoprim is increasing among uropathogens, with one recent study showing that approximately 39 per cent of isolates of E. coli are now resistant. 6 Trimethoprim is absorbed readily from the GI tract and reaches peak serum levels one to four hours after ingestion. About 60-80 per cent of the dose is excreted in the urine, achieving an excess of the minimum inhibitory concentration (MIC) for most of the urinary pathogens. Trimethoprim monotherapy at a dose of 100-200mg twice daily is effective first-line therapy for uncomplicated UTI in women, but should be avoided 18 Prescriber 5 April 2010 www.prescriber.co.uk

in pregnancy especially during the first trimester due to its antifolate mode of action (see Table 1). Nitrofurantoin Another commonly used first-line agent, nitrofurantoin has a poorly understood mode of action but appears to require intracellular reduction via nitrofuran reductase. The active components are capable of blocking inducible enzymes and also damaging bacterial DNA. Susceptible organisms vary somewhat, with E. coli, Citrobacter, enterococci and Staph. saprophyticus being generally sensitive, but members of the Proteus, Providencia, Morganella and Pseudomonas groups being resistant. GI absorption is enhanced when the drug is taken with food. Although serum concentrations are low with a serum half-life estimated at only about 30 minutes, the drug concentration in the urine is substantial and easily able to exceed the MIC of susceptible organisms. Alkalinising agents adversely affect the activity of nitrofurantoin and concurrent use should be avoided. Nitrofurantoin should not be used in patients with renal insufficiency as they may achieve subtherapeutic levels in the urine and raised serum levels. Nitrofurantoin may cause neonatal haemolysis and should therefore be used with caution in pregnancies at term. This antibiotic should only be used for its anti - bacterial action in the bladder and lower urinary tract, and not if there is any evidence or suggestion of upper urinary symptoms. Beta-lactam antibiotics Beta-lactams are a large group of broad-spectrum antibiotics that target cell wall synthesis in a range of Gram-positive and Gram-negative organisms. They act by attaching to penicillin-binding proteins that, aside from binding to beta-lactam antibiotics, are responsible for the final cross-linking of the bacterial cell wall, thus conferring strength, rigidity and the ability to withstand osmotic pressures. This large group of antibiotics is comprised of penicillins, aminopenicillins, cephalosporins, monobactams and carbapenems. Although many betalactams are used in the treatment of UTI, we shall restrict ourselves to the oral agents commonly used in primary care for simple uncomplicated UTI. The three main ones are therefore amoxicillin, co-amoxiclav and cefalexin. As always, before commencing on a course of betalactam antibiotics, penicillin allergy should be enquired about and borne in mind. 20 Prescriber 5 April 2010 www.prescriber.co.uk

Amoxicillin Initially introduced as a development of penicillin that demonstrated a broader spectrum of activity both for Gram positives and negatives, its usefulness is, however, now limited by bacterial resistance. Amoxicillin is thus no longer recommended as empirical therapy for uncomplicated UTI given the level of beta-lactamase-mediated resistance, which is around 50 per cent. 6 It may be used as a second-line agent when the results of culture are known. Amoxicillin is safe to use in pregnancy but only when antibiotic sensitivities are known (as discussed above). It demonstrates good bioavailability from oral dosing and is visibly excreted in the urine. Co-amoxiclav Co-amoxiclav is a combination drug comprising amoxicillin and clavulanic acid. Clavulanic acid is an inhibitor of a range of bacterial beta-lactamase enzymes and thus restores some of the lost spectrum of activity to amoxicillin. It is thus suitable for empirical therapy. Co-amoxiclav may cause clinical resolution of symptoms caused by ESBL-producing organisms, but should not be relied upon to treat systemic or complicated infection. This combination drug is often expressed as a ratio of amoxicillin:clavulanate and demonstrates absorption and bioavailability similar to that of amoxicillin. However, the addition of clavulanate means caution should be exercised in pregnancy and lactating mothers, and where possible an alternative should be chosen. Cefalexin This is a first-generation cephalosporin antibiotic that is stable to some of the beta-lactamase enzymes that have so limited the use of amoxicillin. While retaining much of its Gram-negative activity, it does not have the drawback of co-amoxiclav in that it is not combined with a beta-lactamase inhibitor (clavulanate), which means that it has the advantage of being safe to use in pregnancy. ESBL-producing bacteria may limit the use of cefalexin in the future. Levels of resistance are currently under 10 per cent. 6 Caution must be exercised in patients with known penicillin allergy, especially since the reported crosshypersensitivity between penicillins and cephalo - sporins is around the 10 per cent mark. Cefalexin is the most commonly used oral cephalosporin for the management of UTI, showing reasonable bioavailability and excretion into the urinary tract. Cephalosporins have no activity against the uro - pathogen Enterococcus and often poor activity against Staph. saprophyticus. Quinolones These are a large family of synthetic bactericidal agents that do not act on the cell wall as beta-lactams Figure 1. Dipstick urinalysis showing positive for leukocytes and nitrites is suggestive of infection; a short (three-day) course of empirical treatment is recommended in uncomplicated UTI do, but instead at the level of the bacterial chromosome. They inhibit the activity of DNA gyrase and topoisomerase, enzymes essential for the correct packaging and managing of cellular DNA. These enzymes have no effect on their mammalian counterparts. Quinolones can be divided into generations like cephalosporins. The use of nalidixic acid, the firstgeneration quinolone, has largely been superseded by second-generation fluroquinolones such as ciprofloxacin. This antibiotic class has good Gramnegative action but very limited Gram-positive action, especially against Gram-positive uropathogens. Quinolone antibiotics should be avoided in children, pregnancy and breastfeeding due to concerns regarding arthropathy in animal studies. Caution should also be exercised when considering quinolone use in patients with G6PD deficiency or epilepsy. Resistance of uropathogens in the community to ciprofloxacin is currently less than 10 per cent, 6 which is a valuable reason for this important antibiotic not to be a first-line choice. 7 Quinolone antibiotics are known to select for and encourage MRSA (meticillin-resistant Staph. aureus) SPL www.prescriber.co.uk Prescriber 5 April 2010 23

and should be used cautiously in patients known (or suspected) to be MRSA positive. 8 Likewise, in numerous recent publications quinolone usage has also been implicated in Clostridium difficile infection, particularly with reference to the newly described O27 hyper - virulent strain. 9-11 Quinolone antibiotics are not recommended for first-line management of uncomplicated UTI partly for the reasons already expressed: to attempt to limit resistance occurring to this potent group of anti - biotics, because they are more expensive than some of the alternatives already mentioned, and because of some of the issues concerning MRSA and C. difficile. Taking MSUs While nearly all primary care physicians would request an MSU be sent in more complex cases such as cases of UTI in children, pregnancy and in diabetic men, it has been shown that in otherwise well women there can a wide variation in sample requesting. 12 The optimum management of uncomplicated UTIs should ideally include the taking of an MSU just prior to starting the course of antibiotics for several reasons: this would reassure the physician that the empirical choice was correct when the report is duly available history and examination suggest a urinary tract infection positive result (presence of leukocytes and nitrites) 3-day course of empirical treatment dipstick urinalysis negative result reassure, consider alternative diagnosis, eg STIs, candidosis, pelvic pathology (treat and investigate as appropriate) failure of treatment or recurrence of symptoms send urine to laboratory for culture and sensitivity no resolution or recurrence of symptoms positive result negative result treatment course as per sensitivity report failure of treatment or recurrence of symptoms, if more than 3 episodes of recurrence, refer for investigation and consider prophylaxis consider alternative diagnosis, eg STIs, candidosis, pelvic pathology (treat and investigate as appropriate) Figure 2. Recommended management of women with an uncomplicated UTI; the management of a patient with a complicated UTI should routinely include an MSU sample - see Resources, European Association of Urology guidelines 24 Prescriber 5 April 2010 www.prescriber.co.uk

it would provide accurate epidemiological and antibiotic resistance data essential for guiding future strategy (rather than just sending in complex, relapse or recurrent cases) if the empirical therapy fails due to resistance, then a second antibiotic with known sensitivity can be selected without having to wait for an MSU to be sent in for someone who has now been symptomatic for at least three days. Sending an MSU will also be useful if it demonstrates pyuria, ie white cells but no growth, in which case it may point to another pathology/diagnosis. However in the majority of otherwise well women the cost of routinely sending MSUs to the lab for culture and sensitivity may be the most expensive part of the treatment. A recent study demonstrated that sending an MSU in a patient with an uncomplicated UTI did not significantly improve symptom duration and outcome. 13 Separate study data suggest that 23 urine cultures would need to be performed to predict one infection in which resistance would cause clinical failure. 14 Therefore requesting an MSU in all straight - forward cases may be impractical and expensive, in which case restricting sample sending for only complex cases, treatment failures and those at risk of resistant isolates (previous antibiotics, hospitalisation or known resistant organisms) may represent the most optimal use of resources, while acknowledging that this may skew local resistance data. Structured surveillance of local resistance patterns will provide essential data for drawing up local guidelines. For example, knowing the resistance to agents such as amoxicillin and cefalexin may change the first- and second-line agents. As such, microbiological input should be sought for drawing up and altering such guidelines, and will be essential in the forward monitoring of antibiotic resistance patterns. Conclusions Uncomplicated UTI in an otherwise well woman is a common problem for primary care and accounts for www.prescriber.co.uk Prescriber 5 April 2010 27

Key points uncomplicated UTI is a very common complaint and one of the most common reasons for GP referral one of the commonest reasons for the prescribing of antibiotics the majority (>70 per cent) are caused by E. coli, with coliforms, enterococci and staphylococci making up most of the remainder three days of a first-line empirical antibiotic is usually sufficient treatment ideally, urine samples should be sent for analysis to provide local epidemiology and antibiotic resistance data and to guide subsequent management should first-line treatment fail; this may, however, be impractical and expensive a considerable amount of morbidity, antibiotic prescribing, laboratory time and NHS cost. It has been demonstrated that, generally speaking, the majority of well women should need no more than three days antibiotic therapy with a first-line agent. Short courses, where possible, to minimise antibiotic exposure are essential if we are to maintain the effectiveness of the currently available anti biotic classes in Resources Further reading BMJ Collected Resources. All articles published in the BMJ since the first issue. http://bmj.bmjjournals. com/collections. British national formulary. September, 2009. Clinical Knowledge Summaries. www.cks.nhs.uk/ urinary_tract_ infection_lower_women. Coping with cystitis. Clayton C. London: Sheldon, 1995. Guidelines on urological infections. European Association of Urology, 2009. www.uroweb.org/professionalresources/guidelines. Medical microbiology. 3rd ed (updated). Mims C, et al. Edinburgh: Elsevier Mosby, 2005. Medical microbiology: a guide to microbial infections: pathogenesis, immunity, laboratory diagnosis and control. 17th an era of increasing resistance (including ESBL-producing E. coli and Klebsiella species) while still providing effective treatment. References 1. Hamilton-Miller JM. J Antimicrob Chemother 1994;33:63-73. 2. Naber KG. J Antimicrob Chemother 2000;46(S1):23-7. 3. Hooton TM, et al. Infect Dis Clin North Am 1997;11(3): 551-81. 4. Naber KG. Current Opinion in Urology 1999;9:57-64. 5. Warren JW, et al. Clin Infect Dis 1999;29(4):745-58. 6. Bean DC. Ann Clin Microbiol Antimicrob 2008;7:13. 7. Drug & Therapeutics Bulletin 1998;36(4). 8. Weber SG, et al. Emerg Infect Dis 2003;9(11):1415-22. 9. Kazakova SV, et al. Arch Intern Med 2006;166(22):2518-24. 10. Cookson B. Postgrad Med J 2007;83(979):291-5. 11. Delaney JA. Emerg Infect Dis 2007;13(5):761-3. 12. Hillier S, et al. J Antimicrob Chemother 2006;58:1303-6. 13. Little P, et al. BMJ 2010;340:c199 doi:10.1136/bmj.c199. 14. McNulty CA, et al. J Antimicrob Chemother 2006;58:1000-8. Dr Thompson is a microbiology registrar and Dr Yacomeni is an SHO on rotation in microbiology at Sheffield Teaching Hospitals, and Dr Townsend is a consultant microbiologist at Sheffield Teaching Hospitals, an honorary clinical lecturer at the University of Sheffield and honorary clinical fellow at Sheffield Hallam University ed. Greenwood D, et al. Edinburgh, New York: Churchill Livingstone/Elsevier, 2007. SIGN guidance no. 88. Management of suspected bacterial urinary tract infection in adults. www.sign.ac.uk/ pdf/sign88.pdf. Websites EdRenINFO. Information provided by the Royal Infirmary of Edinburgh Renal Unit for patients and for doctors who are not renal specialists. This site includes information about many kidney problems including urinary tract infections. http://renux. dmed.ed.ac.uk/edren/edreninfohome.html. www.patient.co.uk. Information leaflets for patients: Cystitis in women, Cystitis recurrent infections in women, Urine infection in men, Urine infection in pregnancy, and Urine infection in children. Information and support for people with any kind of cystitis: www.cobfoundation.org/. 28 Prescriber 5 April 2010 www.prescriber.co.uk