Membership information 0800 561 9000
Medicolegal advice 0800 561 9090

Concealed sepsis

01 May 2014

Mr D, 53, had suffered with osteoarthritis in his right knee since turning 50. This had been confirmed with arthroscopy. It rarely bothered him and he continued to work as a PE teacher. He had experienced a flare-up of knee pain at the start of the autumn term but this settled quickly with analgesia.

He contacted the GP out-of-hours service on the first weekend of the Christmas holidays, complaining of two days of bilateral knee pain, which was unrelieved by his usual codydramol. A home visit was arranged. He was seen by Dr C, who documented a normal right knee on examination, but limited movement in the left knee, with positive meniscal signs and no effusion.

Dr C also noticed that Mr D had a painful swollen left little finger, which he had jammed in the door two weeks earlier. Since he was afebrile, Dr C attributed the symptoms to OA and advised Mr D should also arrange to get an x-ray of his finger to exclude a fracture. She provided him with naproxen analgesia.

The pain continued after the weekend and Mr D had been unable to leave the house to arrange the x-ray. He spoke to Dr V at his own surgery and an appointment was arranged for the next morning. The following day, Mr D was still unable to get to his car and called the surgery again, this time speaking to Dr A, who agreed to a home visit.

Dr A recorded an effusion and worsening right knee pain now radiating to the calf and hip. He also mentioned that Mr D now had swelling over the dorsum of his injured hand, and he also spotted two erythematous patches on the right elbow and left foot. Mr D had not reported feeling feverish and so vital signs were not recorded.

Dr A prescribed a course of antibiotics to cover for possible infection in the right hand, and documented that the knee pain was likely to be a strain. She queried gout as a possible cause and recorded that she was uncertain what the satellite lesions represented. She advised Mr D contact the surgery again the next day.

The next day was Christmas Eve and Dr B was on duty for the day. He visited Mr D at home as requested by Dr A. By now he was feeling better, and the swelling in his hand had reduced, but he was feeling “spaced out” on the codeine analgesia he was now taking. Dr B asked the patient to get out of bed for a full examination, which he was able to do.

Mr D’s wife recalled the doctor taking her husband’s blood pressure and advising he omit his antihypertensive medication. Dr B made no record of this examination. He later recalled that he examined the patient fully, including his temperature, and as he found nothing of concern he did not make a note of this. His advice was to complete the course of antibiotics and increase his fluid intake.

Mrs D recalled that her husband became worse towards the end of the day, with slurred speech and generalised weakness. He made an attempt to go to the toilet with the assistance of his son and it took him 40 minutes. Mrs D awoke the next morning to find her husband was dead.

The pathologist who carried out the postmortem concluded that Mr D had died from complications of septicaemia, but the focus of the infection remained uncertain. He noted splenomegaly but no lymphadenopathy. Experts agreed that the cause of death was perplexing but that the knee was the least likely site, with either the hand or an upper respiratory tract infection being the most likely causes.

Crucially, expert opinion agreed that if intravenous antibiotics and volume replacement had been commenced on 23 or 24 December, then arguably the fatal episode of sepsis could have been avoided.

Expert opinion also found that neither Dr A nor Dr B had recorded anything like enough to suggest that their assessments were adequate. In Dr B’s case, with no clinical details recorded and no plausible diagnosis, there would be no possible chance that a court would accept that his assessment was reasonable. Similarly, Dr A had not recorded enough to show that her assessment was reasonable on 23 December.

The case was settled for a substantial sum.

Learning points

  • Good note-keeping is essential. In this case, recording the vital signs and patient’s mobility would have demonstrated that an adequate assessment had been carried out and made the actions of the doctors involved easier to defend.
  • Clinical presentation can change quickly. Expert opinion was critical of a lack of a plausible diagnosis. It is not clear from the note-keeping how unwell Mr D was when assessed by Dr A. It may have been the case that Mr D appeared so well that Dr A felt it unnecessary to document normality. However, without adequate information or a clear diagnosis to prove that a reasonable assessment was carried out, it is difficult to defend her action given the symptoms of polyarthritis with patches of erythema suggestive of infection.
  • Patients should be advised on the signs to look out for and when to seek further help if they continue to feel unwell.
  • Identifying sepsis early can save lives. The diagnosis may not always be immediately obvious and a high index of suspicion is required to make the diagnosis and prevent fatalities. The surviving sepsis campaign, http://survivesepsis.org is an educational resource to train healthcare professionals in the recognition and immediate management of sepsis.