The Practice
27. Mr A complains that the Practice prescribed his wife an antibiotic in June 2019 that she previously had an adverse reaction to in 2017.
28. The urinalysis result on 6 June 2019 confirmed Mrs A had a UTI. The Practice’s decision to prescribe antibiotics was in line with NICE NG109, which recommends the use of antibiotic treatment for UTI.
29. Mrs A had experienced nausea and vomiting after taking nitrofurantoin on two previous occasions. The first was in October 2017 and the second in May 2019, just weeks before the appointment in question. In May 2019, Mrs A’s nausea and vomiting was so significant that she went to hospital.
30. Both occasions are clearly documented in the Practice’s records, along with a letter from the hospital when Mrs A was discharged on 19 May 2019 noting: ‘presumed reaction to nitrofurantoin’. Records show these side effects were known at the appointment on 6 June 2019, because the Practice prescribed Mrs A with anti-sickness medication along with nitrofurantoin.
31. We do not consider the Practice’s decision to prescribe nitrofurantoin on 6 June 2019 appropriate, as it was known Mrs A experienced side effects twice previously, and most recently to the extent she needed hospital treatment. This was not in line with GMC Guidance which says clinicians must: ‘prescribe drugs or treatment, including repeat prescriptions, only when you have adequate knowledge of the patient’s health and are satisfied that the drugs or treatment serve the patient’s needs’. We consider this a failing.
32. The Practice explains there were no other antibiotic options, due to Mrs A’s allergies to some, and side effects she experienced from others. Records make clear Mrs A had a known allergy to penicillin. The Practice says it did not prescribed cephalexin because 10% of people allergic to penicillin are also life-threateningly allergic to cephalexin, which is considered a ‘cousin of penicillin’.
33. The Practice records show it had prescribed cephalexin to Mrs A on seven previous occasions, without her reporting any notable side effects. These were 23 October 2013, 29 July and 7 August 2014, 9 October 2017, 4 September and 20 December 2018, and 26 March 2019.
34. While the entry in August 2014 notes Mrs A ‘blames cephalexin’ for her nausea at that time, she also reported her smoking cessation medication was making her ill. We do not see her reporting concerns at any other time cephalexin was prescribed. In October 2017 and September 2018, records note Mrs A had tolerated it well, with no allergies or sensitivities previously. Notably, she took cephalexin just ten weeks before the appointment in question, with no reported side effects.
35. We consider it would have been appropriate and in line with GMC Guidance for the Practice to have prescribed Mrs A cephalexin for her UTI on 6 June 2019. Had the Practice felt uncertain, we consider it could have reasonably sought microbiology advice on which other antibiotics they could have prescribed.
36. While antibiotics were appropriate to prescribe, records do not suggest Mrs A was particularly unwell with UTI symptoms at the appointment. Our GP adviser suggests any short delay in prescribing, caused by a wait for microbiology advice, would not have unreasonably denied Mrs A treatment considering she was not particularly unwell at that time.
Impact
37. We have thought carefully about the impact of the failing we have identified. We first considered Mr A’s view that his wife had such a severe reaction to the nitrofurantoin that she became unresponsive and was hospitalised, needing critical care and supportive measures, including oxygen.
38. Records clearly show Mrs A had a sudden health deterioration on the afternoon of 6 June, and we see she was taken to hospital by ambulance and then suffered a PEA arrest, before being resuscitated and cared for in the CCU. Yet, we cannot say with any certainty that this was caused by an anaphylactic reaction to nitrofurantoin.
39. Trust records note that on 7 June, clinicians felt Mrs A’s presentation was more likely due to sepsis than anaphylaxis. This means clinicians felt it more likely Mrs A’s deterioration was due to an extreme response to her UTI infection.
40. Blood tests were taken at the time of the PEA arrest and again the next day. A marker in the blood that identifies anaphylaxis was not raised and was within normal limits, both at the time of the PEA arrest and the following day.
41. The Trust’s discharge summary, completed on 27 June 2019, notes: ‘?anaphylactic reaction to nitrofurantoin’. The inclusion of a question mark in medical records means it is only a possibility, or that there is doubt about the observation. There is no evidence to suggest anaphylaxis was ever a definitive diagnosis.
42. Having considered all evidence available to us, we think it is possible Mrs A’s deterioration on 6 June could have been due to an anaphylactic reaction to nitrofurantoin. We think it is also possible her deterioration could have been due to sepsis. It could have been both. The normal serum tryptase levels do not provide clear clinical evidence of anaphylaxis. We cannot say for certain that nitrofurantoin was the direct cause of the events that took place that day.
43. Mr A says his wife was never the same, that her health deteriorated, and he considers her death in December 2019 the direct result of the Practice’s decision to prescribe nitrofurantoin six months previously. It is clear this was an incredibly difficult time for both Mr and Mrs A, and we are very sorry to hear about her declining health and her death.
44. Despite the PEA arrest in June 2019, records show Mrs A did recover, and she was well enough to be discharged home just weeks later. We do not see any evidence to suggest this would have resulted in any long term impact, considering Mrs A’s recovery before discharge. We also do not see any evidence Mrs A’s death, six months later, was linked to the nitrofurantoin prescription. The cause of death is documented as bronchopneumonia, caused by her underlying COPD.
45. We take this opportunity to clarify that we do not see any evidence to suggest Mrs A was allergic to nitrofurantoin. While she was known to experience nausea and vomiting when taking nitrofurantoin previously, these side effects were not due to an allergy to the drug. This is supported by NICE CG183.
46. Also importantly, while the side effects of nausea and vomiting were known, it could not have been foreseen that Mrs A would have a different, or more severe, reaction to nitrofurantoin when it was prescribed in June 2019. As we have explained, we cannot say with certainty that Mrs A did experience any different or severe reaction to the drug on this occasion.
47. We do not see evidence that the nitrofurantoin prescription directly caused the events on 6 June, or that it caused any long term impact, including with regards to Mrs A’s death. However, it remains a failing. We consider it caused Mr A considerable distress, particularly as he has been left, up to now, without clarification and thinking that this led to a more significant impact on his wife’s health.
The Trust
48. Mr A says his wife was transferred far too quickly after her admission to the Trust on 3 December 2019. Records show the Trust transferred Mrs A from hospital A to hospital B on 6 December, three days after her admission.
49. There are no specific national guidelines that set out what constitutes an appropriate inter-hospital transfer in this situation. Our respiratory adviser explains the only guidelines available are for the transfer of critically ill patients from one CCU to another.
50. While Mr A says his wife required critical care and supportive treatment, records do not show Mrs A had any clinical need for intensive care support. She was admitted onto the acute medical unit at hospital A, and transferred onto the respiratory ward at hospital B. The available guidelines do not apply to Mrs A, as she was not on the CCU at hospital A and records show she was not critically ill at the time of transfer.
51. While Mr A is concerned about the transfer occurring too soon after admission, it is not the timing following admission, but the clinical need of the patient that determines whether transfer is appropriate.
52. Records show Mrs A was reviewed by a doctor at hospital A on 6 December, just before her transfer. Evidence suggests that while Mrs A was not critically ill, she was not clinically stable. Our respiratory adviser says, in general, transferring patients between hospitals is not considered appropriate if they are clinically unstable, unless they are being transferred to specialist centre for urgent treatment. This was not the reason for Mrs A’s transfer.
53. GMC Guidance says clinicians must: ‘refer a patient to another practitioner when this serves the patient’s needs’. There is no record of any rationale for the transfer, and it did not occur because Mrs A needed referral to receive specialist or urgent treatment. We do not see evidence to support the decision to transfer Mrs A, and do not consider it followed GMC Guidance. We consider this a failing.
Impact
54. We have carefully considered the impact of the failing we have identified. Mr A questions whether his wife received the care and treatment she needed at hospital B after her transfer. We can assure Mr A we see evidence to show that she did.
55. Records show all the treatment given to Mrs A at hospital A was also given at hospital B. She continued to receive antibiotic, steroid, nebuliser and oxygen treatment for her pneumonia, influenza, and COPD at hospital B. This treatment was in line with NICE CG138, BNF guidance and NICE NG115 respectively.
56. Mr A considers his wife’s death directly resulted from the Trust’s decision to transfer her away from hospital A. We do not see any evidence to suggest the transfer to hospital B contributed to her death. Records tell us Mrs A did not deteriorate because of the transfer. Her death is documented as due to bronchopneumonia, caused by her underlying COPD. This was despite the Trust providing the appropriate treatment for these conditions, in line with the guidance referenced above, at both hospitals.
57. A decision had already been made at hospital A that escalating Mrs A’s care to the CCU would not have been appropriate. Sadly, there was no additional treatment that could have reasonably been given at hospital B.
58. Mr A says that staff at hospital B voiced concern both to, and in front of, him about the decision to transfer his wife. In response to his complaint, the Trust acknowledged staff at hospital B were initially concerned Mrs A was transferred too quickly. While we can assure Mr A the transfer had no negative impact on his wife’s condition or care, it clearly caused him significant distress.
59. In response to the complaint, the Trust explained it had produced an action plan in which it stated lessons would be learned to ensure all patients are identified as medically safe before transfer. It said this event and the learning taken from it would also be discussed at monthly governance meetings.
60. We are satisfied this Trust action is in line with Our Principles for Remedy, in assuring us lessons have been learned and measures are now in place to ensure it will not happen again. We do not see the Trust has acknowledged the failing or directly apologised to Mr A for the distress caused by it, and so we set this recommendation to resolve the matter.