A bad combination - medical case study

Case study on a prescribing system failure

Day one

A GP is called to attend a diabetic patient at home. Mr Q has for the past 24 hours been suffering from frequency of micturition and in the previous night has developed nausea and vomiting. His blood sugar is high.

The GP examines the patient’s abdomen and asks for a urine sample. He examines the specimen visually and suspects Mr Q has a urinary infection. He prescribes a short course of trimethoprim and returns to the surgery where he arranges to have the sample sent off to the laboratory.

In Mr Q’s records is a note of pre-existing conditions including ankylosing spondylitis and ongoing treatment for that condition with the drug methotrexate. Treatment with trimethoprim is a known contraindication in patients on methotrexate as it can lead to acute bone marrow suppression. The practice computer system generates an automated hazard alert but the GP does not routinely pay attention to these, finding the low specificity (lots of false positive alerts) “annoying”.

Day three

The surgery receives a phone call from Mrs Q to say her husband has been taking the prescription for two days but his condition has grown worse with further nausea. Later that day the GP phones to say that the lab results indicate that trimethoprim was not the correct antibiotic for the type of infection and he issues a new prescription for cephalexin.

Day five

Mr Q’s condition does not improve and he is eventually taken by ambulance to hospital. He is catheterised and passes bloodstained urine. He requires a central line and eventually, because of deteriorating renal function, dialysis. His condition continues to deteriorate and in time Mr Q ends up in the ITU sedated and on a ventilator. He develops septicaemia and it is thought that this may be due to immunosuppression caused by an adverse interaction between methotrexate and trimethoprim.

ANALYSIS/OUTCOME: MR Q eventually recovers but continues to suffer subsequent health problems. Later solicitors acting on his behalf begin legal proceedings alleging clinical negligence.

An expert opinion on the case is solicited by MDDUS. The expert advises that methotrexate is an immunosuppressive drug that is eliminated largely by the kidneys. In renal impairment the drug can build up to toxic levels causing neutropenia. In examining the patient notes on admission to hospital the expert finds that Mr Q was suffering from marked neutropenia not present in earlier blood tests. But he advises that it was very unlikely that much trimethoprim would have been taken and absorbed in the 48-hour period it was administered, especially if the patient had been vomiting. Nor does he believe it likely that such a limited dose of trimethoprim would play a significant role in the development of neutropenia.

In the end, MDDUS acknowledges that there has been a breach of duty in prescribing trimethoprim in the presence of methotrexate but denies that the error contributed in any significant degree to Mr Q’s illness.

A few months later the case against the GP is abandoned. However, the incident does prompt the practice to undertake a significant event analysis which concludes that the practice must be more diligent in attending to hazard alerts generated by the practice IT system. A protocol is developed to ensure this is done.

KEY POINTS

  • Consider potential contraindications in all prescribed drugs.
  • A medical error in itself does not amount to negligence; there must also be a causative link, i.e. it resulted in harm.
  • Check for contraindications to prescribing in all patients when returning to the surgery after home visits.

Alan Frame is risk adviser with MDDUS Training and Consultancy