Acute renal failure associated with NSAIDSBMJ 1994; 308 doi: https://doi.org/10.1136/bmj.308.6932.857d (Published 26 March 1994) Cite this as: BMJ 1994;308:857
Editor,--G W Smith and S B Cohen report on a 79 year old woman who was taking non-steroidal anti-inflammatory drugs and developed renal impairment and hyperkalaemia when a urinary tract infection was treated with trimethoprim.1 They unreasonably dismiss the non-steroidal anti-inflammatory drug as a possible cause of the renal disease.
Non-steroidal anti-inflammatory drugs inhibit the normal tonic renal vasodilatation induced by prostaglandin. Loss of this tonic dilatation may be tolerated during health but not in unwell patients, particularly if there is volume depletion or a strong vasoconstrictive drive.2 Many illnesses, including urinary tract infection, can result in enhanced renal vasoconstriction mediated particularly by the renin-angiotensin axis. In the absence of drugs these renal effects of angiotensin II are opposed by prostaglandins.3 Acute salt or volume depletion is sufficient to render even young health volunteers at risk of renal impairment induced by non-steroidal anti-inflammatory drugs,2 and elderly people seem particularly susceptible to these effects.4
Thus an elderly patient who is well may have stable renal function while taking non-steroidal anti-inflammatory drugs; illness, however, may result in enhanced vasoconstriction which cannot be opposed by prostaglandins in the kidney and therefore results in a fall in the glomerular filtration rate. Hyperkalaemia is often severe and is a well recognised effect of non-steroidal anti-inflammatory drugs when renal function becomes impaired.2
The timing of the renal impairment in the case reported fits with that of the urinary tract infection (which was severe enough to merit admission to hospital) and its effect on a kidney deprived of vasodilatation induced by prostaglandin, just as readily as it fits with the use of trimethoprim. Once the patient was well again, non-steroidal anti-inflammatory drugs would not be expected to cause problems, and thus rechallenge with the drug would be of no discriminatory value in this case. It seems unnecessary to propose an unusual effect of trimethoprim when the case is readily explained by the known and commonly observed effects of non-steroidal anti-inflammatory drugs,