Summary of the evidence | Pyelonephritis (acute): antimicrobial prescribing | Guidance

Self-care

No systematic reviews or randomised controlled trials (RCTs) of any non‑antimicrobial treatments were identified that met the inclusion criteria.

Committee discussion on self-care

There was no evidence for the use of oral analgesia in acute pyelonephritis. However, paracetamol has a well-established efficacy and safety profile for managing pain. The committee agreed that it was reasonable to advise people about paracetamol for self-management of pain. A low-dose weak opioid, such as codeine, could be taken with paracetamol by adults and young people over 12 years for more severe pain.
Non-steroidal anti-inflammatory drugs, such as ibuprofen, are generally not recommended for people with acute pyelonephritis because of concerns about renal safety.
The committee discussed the need for an adequate intake of fluids to ensure a high urine output, which is believed to help resolve acute pyelonephritis through a mechanical flushing of bacteria from the kidney. No evidence was found for this and there was no evidence of what constitutes adequate hydration. However, based on committee experience that dehydration is often cited as a cause of UTIs, the committee agreed that people should be advised about drinking enough fluids to avoid dehydration.

Antibiotics

Acute pyelonephritis is a bacterial infection needing treatment with an antibiotic that reaches therapeutic concentrations in the kidney.
Gram-negative bacteria are the most common causative pathogens in acute pyelonephritis, with Escherichia coli (E. coli) causing 60 to 80% of uncomplicated infections. Other gram-negative pathogens include Proteus mirabilis (responsible for about 15% of infections) as well as Klebsiella (approximately 20%), Enterobacter and Pseudomonas species. Less commonly, gram-positive bacteria such as Enterococcus faecalis, Staphylococcus saprophyticus and Staphylococcus aureus may be seen.
Complications of acute pyelonephritis include impaired renal function or renal failure, sepsis and preterm labour in pregnancy.

Choice of antibiotic

Efficacy of antibiotics

An intravenous cephalosporin (ceftolozane/tazobactam or ceftazidime) was compared with an intravenous fluoroquinolone (levofloxacin or ciprofloxacin) in 2 RCTs (Wagenlehner et al. 2015 and Pasiechnikov et al. 2015) in adults with acute pyelonephritis, acute obstructive pyelonephritis or complicated UTI. Moderate quality evidence found that ceftolozane/tazobactam was significantly more effective than levofloxacin for improving composite cure (clinical cure and microbiological eradication and microbiological cure; 76.9% versus 68.4%, number needed to treat [NNT] 12 [range 7 to 43]) but there was no significant difference between antibiotics for clinical cure. Ceftazidime had a significantly higher rate of clinical cure compared with ciprofloxacin (88.9% versus 73.8%; NNT 7 [range 4 to 62]; very low quality evidence).
An intravenous cephalosporin (ceftriaxone or ceftazidime/avibactam) was compared with an intravenous carbapenem (ertapenem or imipenem/cilastatin) in 2 RCTs (Park et al. 2012 and Vazquez et al. 2012) in adults with acute pyelonephritis or complicated UTI. Very low to high quality evidence found that these cephalosporins and carbapenems were equally effective.
Very low quality evidence from a small single RCT (Moramezi et al. 2008) in pregnant women with acute pyelonephritis found no significant difference between intravenous cephalothin and intravenous ampicillin plus gentamicin in the duration of lower UTI symptoms or costovertebral angle pain. The mean time to end of fever was reduced with ampicillin plus gentamicin compared with cephalothin (mean 11 hours lower, p=0.01; very low quality evidence).
One RCT (Peterson et al. 2008) compared different fluoroquinolones (levofloxacin and ciprofloxacin: intravenous or oral) for acute pyelonephritis and complicated UTI in adults and found no significant differences in clinical or microbiological outcomes at follow‑up (high quality evidence).
One RCT (Talan et al. 2000) compared oral ciprofloxacin with oral co‑trimoxazole for acute pyelonephritis in adult women. Low to moderate quality evidence found that ciprofloxacin was significantly more effective for clinical cure (96.5% versus 82.9%; NNT 8 [range 5 to 18]) and microbiological cure (99.1% versus 89.1%; NNT 10 [range 7 to 28]) than co‑trimoxazole.
Low quality evidence from 2 RCTs (Wagenlehner et al. 2015 and Park et al. 2012) found no difference between antibiotics (ceftolozane/tazobactam versus levofloxacin, and ertapenem versus ceftriaxone) for treating bacteraemia secondary to complicated UTI or acute pyelonephritis in adults.
The evidence for children is based on 1 systematic review (Strohmeier et al. 2014) in acute pyelonephritis. No evidence from systematic reviews or RCTs was identified for children with complicated UTI. This systematic review did not find major differences in clinical effectiveness between different antibiotics compared in the studies (third- and fourth-generation cephalosporins, aminoglycosides, co‑amoxiclav and co‑trimoxazole; very low to moderate quality evidence).

Safety of antibiotics

Antibiotic-associated diarrhoea occurs in 2 to 25% of people taking antibiotics, depending on the antibiotic used (NICE clinical knowledge summary on diarrhoea – antibiotic associated).
About 10% of the general population claim to have a penicillin allergy; this is often because of a skin rash that occurred while taking a course of penicillin as a child. Fewer than 10% of people who think they are allergic to penicillin are truly allergic. See the NICE guideline on drug allergy for more information.
People with a history of immediate hypersensitivity to penicillins may also react to cephalosporins and other beta‑lactam antibiotics (BNF information on phenoxymethylpenicillin).
Trimethoprim has a teratogenic risk in the first trimester of pregnancy (folate antagonist; BNF information on trimethoprim). The manufacturers advise that it is contraindicated in pregnancy (trimethoprim summary of product characteristics).
Fluoroquinolones are generally not recommended in children or young people who are still growing (BNF information on ciprofloxacin). The manufacturers advise to avoid in pregnancy (ciprofloxacin summary of product characteristics). Tendon damage (including rupture) has been reported rarely in people receiving fluoroquinolones (BNF information on ciprofloxacin), and the European Medicines Agency's Pharmacovigilance Risk Assessment Committee in a press release (October 2018) has recommended restricting the use of these antibiotics following a review of disabling and potentially long‑lasting side effects mainly involving muscles, tendons and bones and the nervous system. Fluoroquinolones remain an option in acute pyelonephritis, which is a severe infection.
Aminoglycoside doses are based on weight and renal function, and whenever possible treatment should not exceed 7 days (BNF information on aminoglycosides).
Overall there did not appear to be major differences in adverse effects between antibiotics based on the included studies, although these were not well reported (very low to low quality evidence).
See the summaries of product characteristics for information on contraindications, cautions and adverse effects of individual medicines.

Committee discussion on choice of antibiotic

Based on evidence and experience, the committee agreed that acute pyelonephritis is a bacterial infection needing treatment with antibiotics that reach therapeutic concentrations in the kidney. Antibiotics that don't achieve adequate levels in renal tissue, such as nitrofurantoin, fosfomycin and pivmecillinam, are to be avoided.
A urine sample should be sent for culture to confirm susceptibility of the bacteria and inform treatment choice.
The committee reviewed the available evidence comparing different antibiotics in adults and children and agreed that it was limited by its setting (most studies in adults were undertaken in a hospital, and in children the setting of the studies was not reported). The studies included various different antibiotics, which may not reflect those chosen in UK practice. The committee discussed the evidence for a benefit of the intravenous third-generation cephalosporins, ceftolozane/tazobactam or ceftazidime, over an intravenous fluoroquinolone, but this was mainly limited to a benefit for composite cure (which included clinical cure, microbiological eradication and microbiological cure) and the absolute benefits were small.
The committee agreed, based on experience, that several oral and intravenous antibiotics should be available for people with acute pyelonephritis. This enables antibiotics to be selected based on the severity of illness, antibiotic susceptibilities from culture results when available, local resistance patterns, risk of resistant bacteria, the setting, and known patient factors (such as whether the person has a higher risk of developing complications). In line with antimicrobial stewardship, narrower-spectrum antibiotics should be used wherever possible.
Nationally for England, resistance of E. coli (the main causative organism of acute pyelonephritis) in laboratory-processed urine specimens to the following antibiotics is:
- cefalexin: 9.9% (varies by area from 8.1 to 11.4%)
- ciprofloxacin: 10.6% (varies by area from 7.8 to 13.7%)
- co-amoxiclav: 19.8% (varies by area from 10.8 to 30.7%)
- trimethoprim: 30.3% (varies by area from 27.1 to 33.4%).
  
  (Public Health England. Antimicrobial resistance quarterly surveillance: March 2018)
The committee also discussed that prescribers should be aware of their local antimicrobial prescribing data, because resistance rates do vary by area.
The committee agreed that any recent previous urine culture and susceptibility results, and antibiotic prescribing, should be reviewed before choosing an antibiotic.
Based on experience, the committee agreed that if the results of urine culture suggest the bacteria are resistant to the antibiotic given, people with acute pyelonephritis should be contacted and the antibiotic changed regardless of whether symptoms are improving or not. The committee agreed that acute pyelonephritis is a serious infection and antibiotics should be changed to ensure cure.

Non-pregnant women and men with acute pyelonephritis

Based on evidence, their experience and resistance data, the committee agreed to recommend a choice of first-line oral antibiotics, at usual doses for acute pyelonephritis. These are:
- cefalexin (a first-generation cephalosporin); based on its broad spectrum of activity and acceptable levels of resistance
- co-amoxiclav (a penicillin with a beta-lactamase inhibitor); which is only suitable if culture results are available and bacteria are susceptible, because resistance rates are high
- trimethoprim; which is only suitable if culture results are available and bacteria are susceptible, because resistance rates are high
- ciprofloxacin (a fluoroquinolone); based on its broad spectrum of activity and acceptable levels of resistance (particularly for people who have had previous treatment with penicillins, or cannot tolerate or are allergic to penicillins).
The committee noted that use of broad-spectrum antibiotics, such as later-generation cephalosporins, fluoroquinolones or co‑amoxiclav, can create a selective advantage for bacteria resistant to these second-line broad-spectrum agents, allowing such strains to proliferate and spread. And, by disrupting normal flora, broad-spectrum antibiotics can leave people susceptible to harmful bacteria such as Clostridium difficile in community settings. However, these antibiotics are appropriate for the empirical treatment of acute pyelonephritis, where coverage of more resistant strains of common bacterial pathogens is required.
The committee was aware of the European Medicines Agency's Pharmacovigilance Risk Assessment Committee recommendation to restrict the use of fluoroquinolone antibiotics following a review of disabling and potentially long-lasting side effects mainly involving muscles, tendons and bones and the nervous system. However, they discussed that fluoroquinolone antibiotics are a valuable option for the treatment of acute pyelonephritis, which is a severe infection, and it is appropriate to reserve fluoroquinolone use for such conditions. Resistant gram-negative organisms are a particular concern in acute pyelonephritis, and the committee agreed that ciprofloxacin should remain a first-choice option to cover what can be a complex infection. The committee was keen to point out, however, that cefalexin, co‑amoxiclav and trimethoprim are also first-choice options, and antibiotics should be chosen on an individual patient basis, taking fluoroquinolone safety concerns, as well as susceptibility and resistance, into account.
Based on evidence, experience and resistance data, the committee agreed to recommend a choice of first-line intravenous antibiotics, at usual doses, for people with acute pyelonephritis who are unable to take oral antibiotics due to vomiting, or are more severely unwell. These are:
- co‑amoxiclav (only in combination or if culture results are available and bacteria are susceptible)
- cefuroxime (a second-generation cephalosporin) or ceftriaxone (a third-generation cephalosporin)
- ciprofloxacin (taking safety concerns into account)
- gentamicin or amikacin (aminoglycosides); which may be appropriate for some people with acute pyelonephritis, particularly those with severe infection or sepsis, but that efforts should be made to identify the causal bacteria and use reviewed at 48 hours. Gentamicin is the preferred aminoglycoside in the UK, but shortages of certain antibiotics may result in the use of alternatives; for example, amikacin in place of gentamicin.
The committee agreed, based on experience, that it may be necessary to combine antibiotics in the care of people with suspected sepsis. This should be done according to local policy or on the advice of a microbiologist, taking into account local antimicrobial resistance data.

Pregnant women with acute pyelonephritis

Based on experience and resistance data, the committee agreed to recommend cefalexin (a first-generation cephalosporin) as the first-choice oral antibiotic for pregnant women who don't require intravenous antibiotics, and cefuroxime (a second-generation cephalosporin) as the first-choice intravenous antibiotic.
Ciprofloxacin and trimethoprim are not recommended because they should be avoided in pregnancy. Co‑amoxiclav was not recommended because of high resistance levels nationally and the risks of treatment failure in pregnancy.
The committee agreed, based on experience, that local microbiologists should be consulted for advice on second-choice antibiotics, or combining antibiotics if susceptibility or sepsis is a concern.

Children and young people with acute pyelonephritis

The committee was aware that the NICE guideline on urinary tract infection in under 16s makes recommendations on diagnosing acute pyelonephritis and considering referral to a paediatric specialist.
Based on evidence, their experience and resistance data, the committee agreed to recommend cefalexin or co‑amoxiclav (only if culture results are available and bacteria are susceptible) at usual doses for acute pyelonephritis, as first-choice oral antibiotics.
Based on evidence, experience and resistance data, the committee agreed to recommend a choice of first-line intravenous antibiotics, at usual doses, for children and young people who are unable to take oral antibiotics due to vomiting, or are more severely unwell. These are:
- co-amoxiclav (only in combination or if culture results are available and bacteria are susceptible); which can be given intravenously
- cefuroxime (a second-generation cephalosporin) or ceftriaxone (a third-generation cephalosporin); which would be suitable alternatives to co‑amoxiclav
- gentamicin or amikacin (aminoglycosides); which may be appropriate for some children and young people with acute pyelonephritis, particularly those with severe infection or sepsis, but that efforts should be made to identify the causal bacteria and use reviewed at 48 hours.
The committee agreed, based on experience, that it may be necessary to combine antibiotics in the care of children and young people with suspected sepsis. This should be done according to local policy or on the advice of a microbiologist, taking into account local antimicrobial resistance data.

Antibiotic course length

The evidence for antibiotic course length in the treatment of acute pyelonephritis in adults comes from 2 systematic reviews (Eliakim-Raz et al. 2013 and Kyriakidou et al. 2008) and 1 RCT (Ren et al. 2017). No significant differences were found for clinical, microbiological or safety and tolerability outcomes between short courses and longer courses of antibiotics (7 days or fewer compared with 10 days to 6 weeks in 1 systematic review, and 7 to 14 days compared with 14 to 42 days in the other systematic review [very low to moderate quality evidence]). There were no significant differences between a short course (5 days) of intravenous levofloxacin (750 mg once daily) and a longer course (7 to 14 days) of intravenous and then oral levofloxacin (500 mg once daily; moderate quality evidence).
There were some significant differences in clinical effectiveness between different antibiotic course lengths in 1 systematic review in children with acute pyelonephritis (Strohmeier et al. 2014). However, this was limited to 1 RCT comparing 10 days of oral sulfafurazole with 42 days (moderate quality evidence), with other studies in the review finding no differences in outcomes (very low quality evidence). Safety and tolerability outcomes were not reported.

Committee discussions on antibiotic course length

The committee agreed that the shortest course that is likely to be effective should be prescribed to reduce the risk of antimicrobial resistance and minimise the risk of adverse effects.
Based on evidence, the committee agreed that a short course of antibiotics was generally as effective as a long course of antibiotics for acute pyelonephritis, but the definition of short and long course differed depending on the clinical trial definition and the antibiotic used.
In line with the NICE guideline on antimicrobial stewardship and Public Health England's 'Start smart – then focus' toolkit, the committee agreed that the use of intravenous antibiotics should be reviewed by 48 hours (taking into account the person's response to treatment and susceptibility results from urine culture) and switched to oral treatment where possible.

Non-pregnant women and men with acute pyelonephritis

Based on evidence, experience and resistance data, the committee agreed that, for oral treatment, a 7‑day course of ciprofloxacin was sufficient to treat acute pyelonephritis in non‑pregnant women and men. However, because there was no evidence for 7‑day courses of cefalexin or co‑amoxiclav, a range of 7 to 10 days was recommended for these antibiotics. For trimethoprim, a 14‑day course was recommended because there was no evidence for course lengths shorter than 14 days.
For intravenous treatment, antibiotics should be reviewed by 48 hours and stepped down to oral antibiotics where possible, for a total of 7 days.

Pregnant women with acute pyelonephritis

Based on evidence, experience and resistance data, the committee agreed that, for oral treatment, a 7- to 10‑day course of cefalexin was required to treat acute pyelonephritis in pregnant women. For intravenous treatment, antibiotics should be reviewed by 48 hours and stepped down to oral antibiotics where possible, for a total of 7 days.

Children and young people with acute pyelonephritis

Based on evidence, experience and resistance data, the committee agreed that a 7- to 10‑day course of oral antibiotics was required to treat acute pyelonephritis in children and young people. For intravenous treatment, antibiotics should be reviewed by 48 hours and stepped down to oral antibiotics where possible, for a total of 10 days.

Antibiotic dose frequency

No systematic reviews or RCTs that compared the frequency of antibiotic dosing in adults were identified that met the inclusion criteria.
Evidence from 1 systematic review in children with acute pyelonephritis (Strohmeier et al. 2014) found no significant difference in the clinical effectiveness of aminoglycosides with once-daily administration compared with 8‑hourly administration (moderate quality evidence). There were no significant differences in the number of children with hearing impairment or kidney dysfunction (very low quality evidence).

Antibiotic route of administration

The evidence for route of antibiotic administration in acute pyelonephritis is based on 1 systematic review of 15 RCTs in adults and children (Pohl 2007). This review addressed different modes of administration of antibiotics, which cover:
- sequential intravenous then oral treatment compared with intravenous or intramuscular treatment
- sequential intravenous then oral treatment compared with oral treatment
- oral treatment compared with intravenous or intramuscular treatment
- single-dose intravenous or intramuscular treatment then oral treatment compared with sequential intravenous then oral treatment.
Overall, this review found that oral antibiotics were as effective as other routes of administration in treating symptomatic severe UTI (including pyelonephritis) in both adults and children. Intravenous or intramuscular antibiotics were significantly better for bacteriological cure than oral antibiotics at the end of treatment, but this is based on 1 small RCT (NNT 4 [range 3 to 15]; low quality evidence).
There were no significant differences in adverse effects between different routes of administration of antibiotics (very low quality evidence).
Further evidence is available from 1 systematic review in children with acute pyelonephritis (Strohmeier et al. 2014), which compared different routes of administration, which cover:
- oral treatment compared with sequential intravenous then oral treatment
- sequential intravenous then oral treatment (short course of 3 to 4 days) compared with intravenous treatment (longer course of 7 to 14 days)
- single-dose intramuscular then oral treatment compared with oral treatment
- oral treatment compared with rectal treatment.
Overall, this review found no significant differences in the clinical effectiveness of oral antibiotics (cephalosporins or co‑amoxiclav) in children with acute pyelonephritis compared with other routes of administration (very low to moderate quality evidence).
Safety and tolerability outcomes were poorly reported in the RCTs included in Strohmeier et al. 2014, but there did not appear to be any significant differences between different routes of administration of antibiotics (very low quality evidence).

Committee discussions on antibiotic route of administration

Based on evidence, the committee agreed that, overall, oral antibiotics were as effective as other routes of administration for treating acute pyelonephritis in adults and children.
The committee agreed, based on evidence and experience, that oral antibiotics should be given first line when people can take oral medicines and the severity of their condition does not require intravenous antibiotics.
The committee agreed, based on evidence and experience, that intravenous antibiotics can be used for people who are unable to take oral antibiotics due to vomiting, or are more severely unwell, in line with Public Health England's 'Start smart – then focus' toolkit.

See the full evidence review for more information.