Is “nitrofurantoin lung”on the increase?BMJ 2013; 346 doi: https://doi.org/10.1136/bmj.f3897 (Published 18 June 2013) Cite this as: BMJ 2013;346:f3897
Gupta and Trautner suggest using nitrofurantoin as prophylaxis for recurrent urinary tract infections (UTIs) in non-pregnant women.1 They mention the risk of pulmonary toxicity (“nitrofurantoin lung”) associated with prolonged treatment but suggest that this occurs after years of treatment.
In Grampian we have identified 13 patients in the past four years with nitrofurantoin lung, 12 of whom were non-pregnant women receiving prophylactic treatment for recurrent UTI. Ten of them had been treated with nitrofurantoin for 18 months or less before diagnosis (mean treatment duration 14 months). None had been aware of potential lung toxicity, and the prescribing clinicians were surprised by the diagnosis. All patients were advised to stop taking nitrofurantoin, and six also received empirical oral steroids. Serial spirometry (available in nine patients) showed dramatically improved lung function after nitrofurantoin was stopped (mean predicted forced vital capacity 78% at diagnosis, 110% after cessation); serial chest radiographs also showed improvement.
Data from Prescribing Information System for Scotland show that community prescribing of nitrofurantoin more than tripled from 2008 to 2012—from 3.4 to 11 prescribed items per 1000 patients. We believe this change in prescribing, advocated in current guidelines,2 3 is reflected in our local experience and elsewhere in the UK.4 5 We anticipate a continued increased in the incidence of nitrofurantoin lung and worry that many clinicians have forgotten the potential for lung toxicity.
Nitrofurantoin lung was initially misdiagnosed as cardiac failure, pneumonia, and, in one case, metastatic cancer. Patients should be advised to report any respiratory symptoms—such as worsening cough or breathlessness—that develop. Current guidelines and primary care prescribing systems should emphasise the potential for toxicity, which is reversible if the association is recognised early.
Cite this as: BMJ 2013;346:f3897
Competing interests: None declared.