Urinary tract infection

Published on 09/03/2015 by admin

Filed under Obstetrics & Gynecology

Last modified 22/04/2025

Print this page

This article have been viewed 2118 times

CHAPTER 58 Urinary tract infection

Charlotte Chaliha, Michael R. Millar

Introduction

Urinary tract infections (UTIs) are a common cause of morbidity in women, affecting 50% of adult women at least once in their lives. In women aged 20–65 years, at least 20% will suffer an infection each year. Each year, approximately 5% of women will present to their general practitioners with dysuria and frequency (Hamilton-Miller 1994), and approximately half of these will have a UTI. It has been estimated that, on average, a case of uncomplicated acute cystitis results in 6.1 days of symptoms, 2.4 days of restriction of activities and 0.4 days in bed (Foxman and Frerichs 1985a). In women, the incidence of UTI increases with age and with the onset of sexual activity, and is highest amongst the elderly. In children under 1 year of age, the prevalence is higher in boys than girls, with a male:female ratio of 3 : 1–5 : 1. UTI in the neonate should be considered to be secondary to an underlying anatomical abnormality until proven otherwise (Kunin 1987).

The prevalence of UTI increases significantly with increasing age, and is higher in women than in men. In the elderly, the prevalence may be as high as 50%, especially if the woman is institutionalized (Boscia and Kaye 1987). This high prevalence in the elderly is thought to be secondary to cerebrovascular accidents, reduced mental and functional capacity, the use of bladder catheters and diabetes.

Approximately 25–30% of affected women will develop recurrent UTI that is not related to an underlying anatomical or functional abnormality of the urinary tract (Hooton 2001, Finer and Landau 2004).

This chapter will concentrate on UTI in the adult female and discuss strategies for treatment and prevention.

Terminology and Definitions

UTI is defined as inflammation of the urinary tract due to microbial invasion. There is considerable overlap in the clinical presentations of the various syndromes, including cystitis, pyelonephritis and urethritis. Complicated UTI is UTI associated with functional or anatomical abnormalities that increase the risk of serious complications or treatment failure, such as conditions which cause obstruction or relative stasis of urinary flow (Box 58.1).

Urine samples collected directly from the bladder, ureter or renal pelvis should be sterile. Urine passed through the urethra always contains some bacteria derived from the terminal urethra. Significant bacteriuria is defined by the culture of increased numbers of bacterial colony-forming units (CFUs). The absolute number needed to define significant bacteriuria depends on the sample type. The threshold of more than 10⁵ CFU/ml has been a standard for the definition of significant bacteriuria using carefully collected midstream urine (MSU) since the 1950s (Kass 1956). A significant proportion of patients with UTI will have less than 10⁵ CFU/ml (Johnson and Stamm 1987, 1989). Therefore, current recommendations (Warren et al 1999) suggest more than 10³ CFU/ml for a diagnosis of cystitis, and more than10⁴ CFU/ml for a diagnosis of pyelonephritis. An important consideration with these diagnostic criteria is that they rely on careful collection of the MSU. This requires that care is taken in the instruction and support that patients are given to collect these samples. Bacteriuria is common in association with any long-term catheter and is not in itself an indication for treatment of UTI.

Asymptomatic bacteriuria is defined as the presence of more than 10⁵ CFU/ml in two MSU samples in the absence of symptoms (Zhanel et al 1990). Cystitis is an inflammation of the bladder which may be due to infection or a variety of other causes. Urethritis is inflammation of the urethra and may be a consequence of a wide range of causes including UTI, sexually transmitted diseases such as chlamydia, vaginitis, trauma and allergy. Bacterial pyelonephritis is infection of the renal pelvices which may be acute or chronic.

Microbiology of Urinary Tract Infection

The majority of UTIs are caused by facultative bacteria and, occasionally, by fungi and viruses. Escherichia coli accounts for up to 70% of community-acquired infections (Grüneberg 1994), with the remainder predominantly caused by Staphylococcus saprophyticus and a variety of Gram-negative rods within the Enterobacteriaceae. In hospital-acquired infections, approximately 50% are caused by E. coli, 15% by Enterococcus spp. and the remainder by members of the Enterobacteriaceae, Pseudomonas spp., Staphylococcus spp. and yeasts (Bryan and Reynolds 1984). Hospital-acquired UTIs are frequently associated with iatrogenic risk factors such as instrumentation, and also with patient comorbidities. Antibiotic resistance is also much more likely to complicate hospital-acquired UTI.

Pathogenesis

There are host, iatrogenic and bacterial factors that contribute to the pathogenesis of UTI. Foreign bodies such as urinary catheters are major risk factors for infection through mechanisms that include trauma, compromise of local immunity, and by providing protected niche(s) for microbial proliferation and surfaces for biofilm production.

Bacterial virulence factors

The ability of bacteria to adhere to uroepithelial cells is a prerequisite for infection to occur, and reduces the chance of the bacteria being cleared from the urinary tract during voiding. There are various adherence factors, called ‘adhesins’; E. coli possess surface organelles called ‘pili’ that act as adhesins. These adhesins attach to complementary structures on the uroepithelial cell wall, and act not only to promote infection but also to help promote growth and toxin production (Zafriri et al 1987). There are many different types of adhesins, such as type 4 pili, outer membrane proteins, curli, filamentous haemagglutinins and adhesive pili. Other virulence factors that may facilitate infection are specific to each pathogen. These include the surface antigens on E. coli and haemolysins that are produced to help degrade cells, and aerobactins that enhance iron uptake which encourages E. coli growth.

Much of our understanding of UTI comes from the study of uropathogenic E. coli (UPEC). The type of pili of the different strains of UPEC may determine the site of disease in the urinary tract as they have specific cell affinity (Gunther et al 2002). The virulence of UPEC has been attributed mainly to the presence of type 1 fimbriae, a mannose-binding adhesion protein called ‘FimH’ (Abraham et al 1988). Another pathogenic mechanism is the development of intracellular UPEC pods which act as a reservoir for infection (Anderson et al 2003). These pods contain bacteria that are encased in a polysaccharide matrix and protected by a uroplakin coating which help to evade host defence mechanisms and antimicrobials. This initiates the invasion into cells to develop intracellular bacterial communities. It is this reservoir that can serve as a source of bacteria that may reinitiate infection (Figure 58.1)

Figure 58.1 Intracellular bacterial communities extend like pods into the bladder lumen. Infected bladders were bisected, fixed and visualized by either scanning electron microscopy (SEM) or light microscopy [haematoxylin and eosin (H&E) staining]. (A–C) SEM images of a pod on the surface of a C3H/HeJ mouse bladder infected with UTI89 for 24 h show large intracellular communities of bacteria inside pods. Scale bars: 50 µm (A), 5 µm (B), 0.5 µm (C). (D) SEM revealed no pods in C3H/HeJ bladders infected with MG1655. Scale bar: 50 µm. (E) H&E-stained sections of UTI89-infected C3H/HeJ mouse bladders show a bacterial factory 6 h after inoculation (top panel) and a pod 24 h after inoculation (bottom panel). Bacteria in the pod were densely packed, shorter and completely filled the host cell. Video microscopy has shown that bacterial factories mature into pods. Scale bars: 20 µm. (F) Pods were evident in wild-type C3H/HeN bladders infected with UTI89. Scale bar: 50 µm. (G) Confocal Z-section series from whole-mounted bladder infected with UTI89 expressing green fluorescent protein from the plasmid pcomGFP and stained with antibody to uroplakin (primary antibody) and tetramethyl rhodamine isothiocyanate-labelled secondary antibody, showing uroplakin coating the surface of pods. The series depicts the lumenal surface on the left and progresses through the epithelium toward the right. Optical section thickness: 1 µm. All images are representative of the entire sample surface and are from bladders harvested 24 h after inoculation, unless otherwise indicated.

Source: Anderson GG, Palermo JJ, Schilling JD, Roth R, Heuser J, Hultgren SJ 2003 Intracellular bacterial biofilm-like pods in urinary tract infections. Science 301: 105–107.

Host factors

Regular voiding flushes the urinary tract of pathogens and the acidity of urine inhibits bacterial growth. A healthy vaginal flora is thought to be essential in reducing infection. This flora is predominantly lactobacilli and this maintains an acidic pH in the vagina. Lactobacilli and uromucoid in the urine are thought to interfere with bacterial adherence and colonization. It is also thought that the composition of the flora is important as it provides a continuous microbial stimulus to the host immune system, such that it is primed to respond to pathogens. In women with recurrent UTI, the flora has reduced lactobacillus composition (Kirjavainen et al 2009). The glycosaminoglycan layer of the bladder also serves as a protective layer preventing bacterial adherence.

Factors that increase the risk of infection include:

• impaired bladder emptying which can occur with neurogenic disorders such as diabetes, multiple sclerosis, cerebrovascular events and anatomical abnormalities. Anticholinergic drugs may also impair bladder emptying;

• instrumentation of the urinary tract which may traumatize the urethra;

• foreign bodies such as catheters and stones increase the risk of infection as they are a focus for infection;

• pelvic tumours and inflammatory bowel disorders may invade the bladder directly and also affect bladder emptying;

• glycosuria that occurs in diabetes mellitus is a potent culture medium for bacterial growth;

• genetic factors have been postulated to increase the risk of recurrent infection. Women with recurrent infection are more likely to be non-secretors of histo-blood group antigens, and E. coli is found to adhere better to uroepithelial cells of non-secretors than secretors (Lomberg et al 1986, Sheinfeld et al 1989). Further evidence for a genetic susceptibility is that female relatives of women with recurrent UTI are more likely to suffer from UTI as an adult (Hopkins et al 1999);

• in menopausal women, the lack of oestrogen reduces lactobacillus growth and, with the rise in vaginal pH, leads to a predisposition to growth of enterobacteria (Cardozo 1996); and

• sexual intercourse and contraception are strongly associated with the onset of UTI (Hooton et al 1996, Foxman et al 1997). Sexual intercourse not only results in trauma and disruption to the uroepithelial cells, but may also introduce rectal and vaginal bacteria into the urethra. In a woman who has had sex in the previous 48 h, the odds ratio for a UTI is increased 60 times over a woman who has not (Nicolle et al 1982). The use of spermicides with diaphragms alters vaginal flora, increases vaginal pH and decreases lactobacilli concentration, promoting colonization with E. coli.

In addition to all these factors, there is a well-developed and effective innate and adaptive host response to bacterial invasion. The mucosal lining of the urinary tract has a number of immune surveillance molecules that function to recognize invading pathogens. The best characterized of these is the Toll-like receptor (TLR) family (Samuelsson et al 2004, Zhang et al 2004, Anderson-Nissen et al 2007). These receptors function to initiate appropriate host immune defences when triggered by a pathogen. There are 11 TLRs; TLR4 is the best characterized and is present on the epithelial cells of the bladder and kidney, and promotes cytokine and chemokine responses to Gram-negative pathogens. This TLR4 response can still occur even once intracellular bacterial invasion has occurred. The importance of host-mediated immunity can be seen in women with recurrent UTI. A study comparing vaginal, urine and blood samples from 22 women wirh recurrent UTI and 17 controls showed an aberrant immune response. Women with recurrent UTI had defective T-cell activation and a lower concentration of tissue-repair-associated vascular endothelial growth factor (Kirjavainen et al 2009).

Natural History

In children up to 1 year of age, 1.1% of girls and 1.2% of boys may suffer a symptomatic UTI. In school-age children, a UTI has been reported in 8% of girls and 2% of boys (Hansson et al 1997). The sequelae of UTI in neonates and young children include pyelonephritis and renal scarring, especially if infection occurred before 5 years of age. Vesicoureteric reflux is a significant aetiological factor in the occurrence of UTI in the young.

In young women, UTI can result in asymptomatic bacteriuria, cystitis and pyelonephritis. These do not usually affect renal function, except in the elderly where bacteriuria has been associated with a decrease in glomerular filtration rate. In pregnancy, UTIs have been associated with an increased risk of prematurity, perinatal mortality and perinatal complications (Maclean 2000). The incidence of asymptomatic bacteriuria is similar to non-pregnant women (4–7%) (Patterson and Andriole 1997); however, in pregnancy the risk of developing a symptomatic UTI is much higher, and 10–30% of women with asymptomatic bacteriuria develop pyelonephritis.

Presentation

A UTI may present either as an asymptomatic bacteriuria, acute cystitis or, more seriously, as acute pyelonephritis, bacteraemia and renal failure. The classic symptoms of acute cystitis include dysuria, frequency, urgency and suprapubic pain. If the upper urinary tract is involved, haematuria, loin pain, renal angle tenderness and fever may also occur. In children and the elderly, the classic clinical features of a UTI may not be present. In young children, clinical features of a UTI may be failure to thrive or a non-specific abdominal pain. In the elderly, UTIs often present as confusion and general malaise.

Physical Signs

In cases of acute cystitis or urethritis, there is often suprapubic tenderness and, occasionally, fever. The clinical presentation of acute pyelonephritis is often much more florid, with the patient often looking unwell with pyrexia and tachycardia. There is usually loin tenderness and, if severe, features of septicaemia may be present. In young children and the elderly, as with symptoms, clinical signs may be non-specific and atypical in nature.

Other associated physical signs which may point to the aetiology are a sensory neuropathy or evidence of a spinal cord lesion in cases of a UTI secondary to a neuropathic bladder.

Management

History and examination

In many women, infection can be diagnosed on the basis of history and clinical examination alone. A history should be taken to identify any predisposing features, such as recent UTI, recent urinary tract operations, recent sexual intercourse and the use of the contraceptive diaphragm and condom. Poor bladder emptying secondary to neurological disorders or the use of anticholinergic therapy, pregnancy, the presence of pelvic tumours and diabetes mellitus may also predispose to infection.

Clinical examination should include a general systemic examination, especially if the patient is pyrexial. Examination of the renal angles is required to elicit signs of pyelonephritis. If a neuropathy is suspected, a neurological examination of the S2–4 nerve roots should be performed, assessing for sensation around the buttocks. A gynaecological examination should be performed to exclude residual urine, a pelvic mass or pregnancy. Differential diagnoses that should be considered include detrusor overactivity, cystitis, bladder stones or tumours, ovarian torsion or cysts, ectopic pregnancy and miscarriage.

Investigations

Investigations should be aimed to help select appropriate treatment and to exclude any underlying cause which may predispose to recurrence (Arnold 2000).

Treatment

There are three principles in the management of infection. General supportive measures relieve symptoms and may help to eradicate infection. Antimicrobial therapy should be instituted appropriately, and if an underlying cause is found, such as obstruction, this should be treated. Finally, prevention of further infection will help to reduce recurrence.

General measures

Generally, patients are advised to maintain a high fluid intake of at least 2 l/day and to void regularly to ensure adequate bladder emptying. However, there is not much evidence that this practice improves outcomes over and above appropriate antibiotic therapy. If the patient is septicaemic, more intensive supportive measures and monitoring are required. Intravenous fluids, vasoactive drugs and treatment of the septicaemia should be considered.

Antimicrobial therapy

The aim of antimicrobial therapy is to eradicate pathogenic organisms with minimal local and systemic side-effects. A suitable antimicrobial agent should reach a suprainhibitory concentration in urine and have activity against the range of common causative agents of UTI. An ideal antibiotic would have a low potential to select for bacterial resistance and to give rise to side-effects, be inexpensive and easily administered. Selection of antibiotics for empirical treatment should take account of local resistance patterns, which may vary geographically and are also dependent on whether the infection is hospital or community acquired. Side-effects include anaphylaxis, skin rashes, gastrointestinal disturbances, fungal infection and Clostridium difficile colitis (particularly in the elderly). Multiresistant E. coli is becoming more commonly recognized as a cause of community-acquired as well as hospital-acquired infection, making the requirement to take account of local antibiotic resistance information increasingly important when designing treatment algorithms.

Many antibiotics administered systemically reach much higher concentrations in urine than in serum. These include β-lactams, aminoglycosides, fluoroquinolones and trimethoprim, so large doses of these agents are rarely required. Amoxycillin resistance is now so common in laboratory isolates in the UK that it is best avoided in the empirical treatment of UTI, and in some areas, trimethoprim resistance is reaching similar levels. The true level of antibiotic resistance amongst agents of community infection is not known. The levels of resistance in laboratory isolates from patients in the community may be an overestimate because of biases in the way in which laboratories are used. For example, samples may only be sent to the laboratory when patients return to a doctor after failure of empirical treatment. Estimates of the levels of antibiotic resistance for hospital infections may be more accurate because of the relative ease of use of the laboratory. Alternatives to amoxicillin or trimethoprim for oral use include nalidixic acid, penicillin/enzyme inhibitor combinations (such as amoxicillin with clavulanate), nitrofurantoin, oral cephalosporins and quinolones. The British National Formulary gives good advice on antibiotic selection and treatment durations for specific clinical scenarios. The duration of therapy has come under some debate with a move to shorter regimes to increase compliance, as these will have less effect on the faecal and vaginal flora and reduce the risk of resistant strains. Three-day regimes are as effective as 5- and 7-day regimes for those with uncomplicated UTI (Norrby 1990). Ideally, protocols should be developed with local microbiologists and/or infectious disease specialists that take account of local resistance information. Additional information on the use of antibiotics is provided in the section dealing with specific clinical presentations.

Prevention

For many women with recurrent infection, suggested preventive measures include maintaining a high fluid intake, instructions on perineal hygiene such that the perineum is wiped from front to back after defaecation and micturition (thus reducing the risk of faecal contamination of the urethra), and the avoidance of bubble baths, vaginal deodorants and specific underwear. The benefits of these practices are unclear and they have not been shown to reduce the frequency of infections in case–control studies (Foxman and Frerichs 1985b, Remis et al 1987).

There is, however, a strong association between sexual behaviour and contraceptive use (Foxman and Chi 1990). If sexual intercourse is a precipitating factor, postcoital treatment and voiding are recommended. In women using spermicides and diaphragms for contraception, alternative methods may be recommended.

The beneficial effects of cranberry juice are receiving increasing attention as a simple remedy that reduces the incidence of recurrent infection; however, most studies have been relatively small and inconclusive. There are two postulated mechanisms of action: competitive inhibition of the E. coli fimbrial subunit to the uroepithelial cells, or prevention of the expression of the normal fimbrial subunits (Patel and Daniels 2000). In a randomized double-blind trial to determine the effect of cranberry juice on bacteriuria and pyuria in 153 elderly women, there was reduced frequency of bacteriuria (15% vs 28%) with daily ingestion of 300 ml of cranberry juice (Avorn et al 1994). There was a decrease in the incidence of symptomatic UTI but this did not reach statistical significance; however, antibiotic use decreased by approximately 50% in the group who drank cranberry juice. The Cochrane analysis of cranberries for prevention of UTI found that the studies assessed were flawed as the amount and type of cranberries given differed, and there was only weak evidence to support the use of cranberry juice for prevention (Jepson et al 2004).

In postmenopausal women, there is increased susceptibility to infection secondary to the changes in the vaginal flora and the uroepithelium secondary to oestrogen deficiency. There are few randomized trials of hormone replacement therapy in the prevention of UTI. In a double-blind, placebo-controlled trial of oestriol cream for the treatment of recurrent infection in 93 women, Raz and Stamm (1993) found that those on topical oestriol had a lower incidence of UTI (0.5 vs 5.9 episodes/patient-year). The decrease in UTI was seen together with a decrease in vaginal pH and recolonization with lactobacilli. In a later study by Cardozo et al (1998), daily use of oestriol cream was found to be superior to placebo in the treatment of recurrent UTI.

The recent Cochrane review (Perrotta et al 2008) of oestrogen use in postmenopausal women with recurrent UTI reviewed nine studies which included 3345 women. The data were difficult to summarize as studies were heterogeneous and used different application methods and doses. Pooled data from four studies showed that oral oestrogens did not reduce UTI compared with placebo. There were two small relevant studies comparing vaginal oestrogens which showed a reduction in UTI.

Low-dose prophylactic antibiotics can be considered if the frequency of attacks is two or more per 6 months or three or more over 12 months (Nicolle and Ronald 1987, Stamm and Hooton 1993). The aim of treatment is to eradicate urinary bacteria without affecting the healthy flora of the bowel and vagina, or causing the development of resistant strains. The usual suppressive dose is one-quarter to one-third of the antimicrobial dose required to treat an acute infection. This is usually prescribed at night to maintain a high antimicrobial concentration for as long as possible. The antibiotics of choice are trimethoprim, trimethoprim and sulphamethoxazole, nitrofurantoin, nalidixic acid and cephalexin (Harding et al 1982). In patients with neuropathic bladders, long-term indwelling catheters or ileal conduits who are at increased risk of recurrent infection, it is not advisable to begin long-term prophylaxis as this increases the risk of resistance and antimicrobial side-effects. In these cases, the patient should be advised to seek early treatment when a UTI is suspected.

Future approaches to treatment

The increasing problem of antimicrobial resistance and the lack of efficacy of conventional therapy in recurrent infection has led to new strategies to reduce bacterial penetration and alter the host innate or adaptive immunity.

There is much interest in the use of lactobacillus-containing probiotics for preventing recurrent infection. Probiotics are ‘live micro-organisms’ and the rationale for their use is that they restore the commensal flora of the genitourinary tract which inhibits colonization with uropathogens. Four randomized trials (Reid et al 1992, 1995, Baerheim et al 1994, Kontiokari et al 2001) have studied lactobacilli for the prevention of recurrent UTI; however, only one reported a reduction in episodes of recurrent UTI compared with the year prior to the study (Reid et al 1995). There are also concerns regarding the stability of probiotics and strain-specific effects.

The identification of various E. coli adhesins, including the type I and P fimbriae, has stimulated the search for vaccines against the development of UTI. Animal studies have shown that vaccination with antibodies to the FimH adhesin, which mediates binding to the bladder mucosa, reduces bacterial binding, bacteriuria and pyuria (Langermann et al 2000).

As the host immune response may determine susceptibility to infection, modulators of the immune system have been proposed as adjuvants to help boost the normal immunological mechanisms to vaccines (Ishii and Akira 2007, Parker et al 2007). Data from animal studies have shown a greater antigen-specific immune reponse and a reduced bacterial load when immunomodulators have been used (Bishop et al 2007, Huleatt et al 2007).

The clinical application and efficacy of these new strategies for treatment are still unclear; however, if found to be useful, they may act in combination or instead of antimicrobials to decrease recurrent infection and increase treatment efficacy.

Specific Clinical Situations

Asymptomatic bacteriuria

Asymptomatic bacteriuria is defined as the presence of increased numbers of bacterial CFUs in a urine sample in the absence of clinical signs or symptoms of UTI. In a carefully collected MSU sample, this number is more than 10⁵ CFU/ml. If left untreated, 60–80% of patients will spontaneously clear the infection without long-term sequelae. However, there are several situations where treatment is advisable. In obstetric and gynaecological practice, these situations include pregnancy and urogenital surgery.

There has been much debate on the benefits and cost-effectiveness of screening pregnant women for asymptomatic bacteriuria. The prevalence of asymptomatic bacteriuria has been reported to range from 2% to 13% (Norden and Kass 1968). There is an association with premature birth and low birth weight (Andriole and Patterson 1991), as well as an increased risk of developing pyelonephritis. In a study of 5000 antenatal patients, it was reported that in women with asymptomatic bacteriuria in pregnancy, 36% progressed to acute pyelonephritis if untreated compared with 5% if treated (Little 1965). Current recommendations in the UK (National Institute for Health and Clinical Excellence 2008) support the screening of pregnant women for bacteriuria at the time of booking by urine culture, because treatment with antibiotics reduces the risk of pyelonephritis. Antibiotic choices should be based on the culture result and should take account of the safety profile of the selected agent in pregnancy. Tests for urinary nitrite and leukocyte esterase are unreliable for the diagnosis of asymptomatic bacteriuria (Tincello and Richmond 1998).

Recurrent infection

Up to 20% of women with acute cystitis develop recurrent UTI which is defined as three or more laboratory-confirmed infections per year. These occur due to either reinfection or a relapse of persistent infection. Risk factors such as atrophic vaginitis should be looked for and treated. Maintaining a good urine output is advisable to promote voiding and avoid urinary stasis. Prophylactic use of trimethoprim, nitrofurantoin or norfloxacin therapy may reduce the risk of recurrent attacks but may also select for antibiotic resistance. Cranberry juice may reduce the risk of recurrent UTI without the risk of antibiotic resistance (Kontiokari et al 2001).

Postcoital prophylaxis with a single dose of an antibiotic can be used if infection is related to intercourse. Alternatively, patients can be given antibiotics to use when symptoms occur, or daily prophylaxis with nitrofurantoin, trimethoprim, norfloxacin or cephalexin. A 6-month trial of prophylaxis is usually recommended, with the patient observed to see if infection recurs once the regime is discontinued. However, recurrence seems to occur and a longer period of prophylaxis is now advocated for at least 2 years. Antimicrobials such as trimethoprim have been effective and well tolerated for up to 5 years (Nicolle and Ronald 1987, Nicolle et al 1988).

Acute pyelonephritis

Prompt antimicrobial treatment reduces the risk of serious adverse outcomes, so treatment should be commenced as soon as a diagnosis of urosepsis is considered (clinical signs and symptoms of UTI such as dysuria or loin pain in association with systemic signs of infection such as fever, rigors, hypotension, tachypnoea and tachycardia) and prior to the results of urine culture. Potential complications of pyelonephritis include Gram-negative bacteraemia, endotoxic shock and disseminated intravascular coagulation. In general, acute pyelonephritis resolves without long-term renal damage in the majority of women. However, in the presence of obstruction, such as with stones, infection may result in papillary necrosis, renal or perinephric abscess, or xanthogranulomatous pyelonephritis (Cattell 1998). Treatment consists of aggressive supportive treatment including rehydration and intravascular volume expansion.

Drugs of choice for parenteral therapy include cephalosporins, fluoroquinolones or an aminoglycoside. After the culture results are available, treatment may be changed to the appropriate antibiotic if necessary in consultation with a microbiologist or infectious disease specialist. The duration of treatment is usually 10–14 days (Bailey 1998, Cattell 1998).

Imaging of the renal tract is not usually necessary unless there has been no response to antibiotics or if there is a strong clinical suspicion of renal tract obstruction. Intravenous urography will be normal in 75% of cases of uncomplicated acute pyelonephritis (Fraser et al 1995), as well as renal ultrasound. CT imaging is usually normal unless infection is severe, when changes include renal enlargement, focal swelling and parenchymal attenuation.

Catheter-associated infection

Urethral catheterization is a major risk factor for UTI and local trauma, so should not be undertaken lightly. Those who carry out urinary tract catheterization should have received appropriate training in technique and catheter type (Carter et al 1990). A UTI is reported to occur in approximately one-third of patients catheterized in hospital (Hayley et al 1985). The risk of a UTI after an in–out catheter is 1–2% (Turck et al 1962), but the risk is higher in pregnancy, with a high bladder residual and in the immunocompromised. Basic measures such as the use of a closed-drainage system and gravity-dependent drainage of urine decrease the risk of UTI, and a policy of expediting catheter removal as soon as clinically appropriate should be practised.

Long-term urinary catheters become colonized with bacteria, and UTI is a frequent complication. The underlying cause is the development of a pathogenic biofilm on the surface of an indwelling catheter. The rate of bacteriuria is 3–10%/day and approaches 100% in those with long-term catheters (Warren et al 1982, Saint et al 2002). Inhibiting biofilm formation is one mechanism to reduce UTI, and urinary catheters have been modified to promote this. Impregnation of catheters with antimicrobial agents such as silver have been shown to delay or reduce the onset of bacteriuria; however, there is the possibility of future resistance to silver (Brosnahan et al 2004). Treatment may be required when the patient develops systemic signs or symptoms of infection. Antibiotic treatment choice is best based on culture results.