12. Mrs A told us the Trust did not provide the care and treatment it should for sepsis. We looked at each admission to see if what happened was in line with relevant guidance.
First admission: mid July
13. A doctor assessed whether R might have been suffering from sepsis on her admission on admission. This was because she had pyrexia (elevation of core body temperature). The assessment was done to establish if she had sepsis, and to determine its cause. This was in line with the NICE guidance ‘Sepsis: recognition, diagnosis and early management’ that stresses the importance of recognition and early assessment of possible sources of infection, and the value of blood tests and early treatment with antibiotics,
14. According to the NICE guidance, doctors should consider ‘could this be sepsis?’ when a patient exhibits signs or symptoms suggestive of a potential infection.
15. Our adviser told us that when carrying out the assessment for sepsis, the Trust followed the established guidelines in the UK Sepsis Trust Sepsis Screening Tool. This involved considering the possibility of sepsis due to R’s elevated temperature, and then administering intravenous antibiotics and intravenous fluids, both of which are advised in the NICE guidance.
16. A resident doctor reviewed R a few hours later and gathered a comprehensive medical history. The doctor performed a physical examination, checked observations and ordered tests including X-ray, blood tests, and ECG (electrocardiogram, an electrical tracing of the heart). The doctor discussed R’s symptoms with a medical registrar (a more senior specialist doctor).
17. This was in line with the GMC guidance Good Medical Practice, which says:
‘You must provide a good standard of practice and care. If you assess, diagnose or treat patients, you must: • adequately assess the patient’s conditions, taking account of their history […]; where necessary, examine the patient • promptly provide or arrange suitable advice, investigations or treatment where necessary • refer a patient to another practitioner when this serves the patient’s needs.’
18. Following the conversation with the medical registrar, the treating doctor considered whether an alternative diagnosis to sepsis could be made. This was because blood tests showed R had normal CRP levels (C-reactive protein, a marker in the blood where high levels indicate inflammation, often due to infection).
19. The doctor also noted the presence of tachycardia (rapid heart rate) and tachypnoea (rapid breathing rate) that were significant enough to affect R’s mobility. This raised concerns that she may have a pulmonary embolism (a life-threatening blockage of a lung artery).
20. This process of reassessing and exploring other possible diagnoses, along with continuous evaluation and treatment, was consistent with the GMC guidance in paragraph 17 and also the following:
‘In providing clinical care you 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 • provide effective treatments based on the best available evidence […] • consult colleagues where appropriate.’
21. From this point forward, the Trust determined that R’s symptoms were unlikely to be due to sepsis. Our adviser told us this was a reasonable assessment, considering the absence of any source of infection and the lack of elevated CRP levels. Our adviser told us R’s white cell count was elevated, which could indicate infection but is also consistent with various other conditions.
22. Throughout R’s admission, the Trust continued to look for an underlying cause of her symptoms and arrived at a reasonable conclusion that her symptoms were the result of a viral infection. Additionally, the Trust organised the following follow-up appointments, as documented on R’s discharge:
‘1. Outpatient (OP) endocrine follow up requested 2. OP infectious disease follow up requested 3. OP cardiology follow up for incidental finding of pericardial cyst.
4. OP Bone health clinic requested.’
23. This was all in line with the guidance in paragraph 20.
24. R was discharged with safety netting advice (information given to a patient or carer about actions to take if their condition fails to improve). This is in line with the Department of Health and Social Care guidance ‘Hospital discharge and community support guidance’, which says:
‘Health and social care professionals working in NHS bodies and local authorities should ensure that ‘safety netting’ is provided whereby the individual is provided with advice on discharge. The person should be given the contact details of their discharge team at the point of discharge and advised to make contact if they are concerned about anything.’
25. To summarise this admission, the Trust did not establish a conclusive diagnosis of sepsis, as there were other more likely causes for R’s symptoms. Sepsis was among several medical conditions initially considered.
26. Our adviser explained it is important to recognise that in addition to adhering to guidelines, physicians employ clinical judgement when treating patients. Clinical signs and tests frequently do not lead definitively to a single diagnosis, and recommendations for one diagnosis may conflict with those for another.
27. Physicians must evaluate the evidence comprehensively and apply clinical judgement to investigate what they consider to be the likely diagnosis. This was the case with the care and treatment the Trust gave to R, and we consider it was appropriate and in line with the guidance quoted.
Second admission: end of July
28. During her second admission, R presented with symptoms similar to those observed during her previous admission. She again had a raised temperature, and the doctor carried out a comprehensive assessment. This included a detailed medical history, a physical examination, ongoing observation, and a review of earlier imaging. The Trust carried out blood tests, which indicated that R’s inflammatory markers (including white cell count and CRP) were not elevated.
29. This was in line with the guidance outlined in paragraph 17.
30. The doctor made diagnoses of pyrexia of unknown source, and anxiety. Our adviser told us these diagnoses were justified, given the clinical evidence and the details R provided. The records show she expressed her desire to return home and said her anxiety was her primary concern and felt ‘unbearable’.
31. The doctor drew up a reasonable plan. This included taking imaging, a mental health review, screening for a viral infection, an infectious disease specialist review and further blood tests. This was in line with the guidance outlined in paragraph 20.
32. Our adviser told us the Trust did not make a definite diagnosis of sepsis, and there is no indication that R had, or was treated for, sepsis on this admission. They said this was reasonable given the history and her presentation, which did not suggest sepsis was likely.
33. R told the medical team she wanted to be discharged on the last day. The Trust agreed to her discharge, with arrangements for follow-up care as an outpatient.
34. Our adviser told us it is widely acknowledged that extended hospital stays can lead to increased risk factors, such as hospital-acquired infections, blood clots, and elevated levels of stress or anxiety, all of which can hinder recovery. Consequently, discharging R and arranging for outpatient follow-up was a sensible course of action.
35. We know Mrs A was concerned that the Trust did not tell the family about sepsis on discharge. We are not critical that the Trust did not provide specific safety netting advice regarding sepsis, as this was not considered a probable diagnosis.
36. Our adviser explained that in cases where no specific safety netting advice is given, it is generally assumed that patients will seek medical attention if their condition deteriorates or if they feel unwell. Nevertheless, it is always prudent to clearly communicate this. The Trust appeared to have done this with the advice given to Mrs A, as documented: ‘I reiterated at the end of the call if [Mrs A] was worried for R’s health or wellbeing she should present to A&E or call 999’.
37. To summarise, on this admission the care and treatment in relation to sepsis was in line with guidance, as there was no indication this was a likely diagnosis. Investigations were ongoing for R as an outpatient, and this was in line with the guidance quoted.
Third admission: mid to end of August
38. R was admitted on this occasion with a tramadol overdose (a strong painkiller). She was admitted to the intensive treatment unit (ITU) where she continued to have a raised temperature. The Trust initially suspected the raised temperature and other symptoms were related to the tramadol overdose.
39. This was a reasonable conclusion to reach as the NICE clinical knowledge summary ‘What are the signs and symptoms of drugs commonly involved in poisoning or overdose?’ explains, fever can be a symptom of overdose. The Trust noted, and continued to investigate, the ongoing history of pyrexia of unknown origin. There is a note on the records that the Trust also considered aspiration pneumonia (a lung infection caused by inhaling food or stomach contents into the airways).
40. Our adviser told us that throughout this period of care and treatment the Trust thoroughly assessed R, carried out relevant blood tests and investigations and gave antibiotics appropriately for the suspected aspiration pneumonia.
41. The Trust involved specialist microbiologists (infection experts who advise on antibiotics) when deciding to stop antibiotics, as there were no signs of infection at that stage. The Trust restarted antibiotics when there was an increase in CRP, suggesting an infection. This was all in line with the guidance in paragraphs 17 and 20.
42. The Trust suspected sepsis towards the end of August, and escalated R’s care because of her deterioration. This was because of the combination of her ongoing fever, spike in her temperature, and a reduction in her level of awareness.
43. The Trust consulted with the microbiologist and the critical care consultant in drawing up a treatment plan. This included blood tests, antibiotics, fluids and discussion with a consultant. Our adviser told us the Trust treated R in line with the guidance in paragraph 17, and with the NICE guidance on sepsis which says:
‘arrange for a clinician with core competencies in the care of acutely ill patients to urgently assess the person’s condition and think about alternative diagnoses to sepsis
carry out a venous blood test, including for: • blood gas, including glucose and lactate measurement • blood culture • full blood count • C-reactive protein • urea and electrolytes • creatinine • liver function tests • a clotting screen
• give antibiotics in line with recommendation 1.8.3 and the recommendations on choice of antibiotic therapy • refer to the senior clinical decision maker as soon as possible • use clinical judgement to decide whether to discuss with a consultant.’
44. R’s condition continued to deteriorate, and despite escalation to ITU she sadly died. All care and treatment for sepsis on this admission was balanced and in line with the guidance quoted.
Antibiotics from mid to end of August
45. Mrs A told us about delays and missed doses of antibiotics. She said she was concerned there were failings in relation to this that had led to R’s death. She was not able to pinpoint the specific dates and times within the period, so we asked our adviser to consider this.
46. Our adviser told us that, as previously outlined in the overview of the care and treatment in paragraphs 38 to 44, it is their view that the decision-making and overall administration of antibiotics was in line with GMC guidance. We concluded there were no indications of failings in general in relation to the overall administration of antibiotics during this period.
47. We looked at the specific issues Mrs A raised when discussing these concerns with us. She outlined an occasion where the family say the Trust failed to give R the antibiotic she needed for 14 hours. A review of the medical records shows this relates to an occasion on 20 August when the Trust was unable to give the antibiotic because R’s cannula was dislodged. A cannula is a small tube usually inserted in the back of a patient’s hand to deliver fluids and medication.
48. We can see R had difficulties with cannulas and had either removed them, or the site had become swollen and they had to be removed. The records show that for this reason the doctor took the decision on that night, to wait until the next morning to reinsert the cannula. Our adviser told us this was reasonable, given the fact R had had four cannulas displaced in the previous 24 hours.
49. While there is no specific guidance on the insertion of a cannula, it must be done by trained personnel and there is an expectation that these trained personnel will use clinical judgement to minimise unnecessary attempts, and follow sterile procedures to prevent infection.
50. In these circumstances, and given R’s recent history of repeated insertions, it was reasonable for the Trust to wait to the next morning so the cannula could be inserted safely by trained personnel, even if this did mean one dose of antibiotic was given later than prescribed.
51. This was in line with the GMC guidance which says:
‘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.’
52. There were also two occasions of delays the family raised with the Trust when they made the complaint.
53. On the first occasion there was a delay of three and a half hours in giving the antibiotic dose. The complaint response from the Trust explains it had changed the antibiotic type and gave the first dose of this at 9.25pm. The next dose was due at 3am, but was not given until 6.30am. This was due to a human error in managing the move from three to four doses a day. This is a reasonable explanation for how this mistake was made.
54. We cannot say that this slight delay is what led to R’s death. This is because our adviser told us there is no indication R experienced any adverse effects from it. We would have expected to see a worsening in R’s symptoms soon after the delayed dose, if it had had a serious impact that was connected to her later deterioration and sad death. There is no evidence of this from the records. The Trust has provided reassurance that learning has taken place, and we consider this and the apology are a suitable remedy for what happened.
55. On the second occasion the records show it was not possible to give one dose of antibiotics in the early hours because there was no venous access. As outlined previously, R had difficulties with cannulas and had removed them. The records show on that night she was unsettled. Our adviser said in that circumstance, and with this history, it was reasonable for the Trust to wait until the next morning to reinsert the cannula. This would allow the cannula to be inserted safely by trained personnel, when she was less agitated, even if this did mean one dose of antibiotic was given later than prescribed.
56. As outlined in paragraph 51, this was in line with guidance.
Nutrition, hydration and diabetes management
57. We were sorry to hear from Mrs A that she thought the care and treatment given by the Trust in these areas led to R becoming dehydrated and so contributed to her deterioration and death.
58. The medical records did not give a complete account of the nutrition and hydration the Trust gave to R. We asked our adviser to give an overview of whether we could reach the conclusion Mrs A had, that indications of failings in these areas contributed to R’s deterioration and sad death.
59. We were not able to reach this conclusion. Our adviser gave us an explanation of how these factors affected R.
60. They told us R was receiving nasogastric feeding (a way of giving nutrition through a thin tube that goes from the nose down into the stomach). There are numerous mentions of this within the records. While our adviser was not able to say from the evidence available whether the nutrition and hydration were sufficient, they were able to say that it is likely they did not play a major role in R’s decline and sad death.
61. Our adviser said the evidence to support this assertion is that the blood tests showed no compelling evidence of dehydration, such as raised creatine and urea levels. They would also have expected to see R’s kidney function become impaired if she had been dehydrated. Throughout the course of her treatment, R’s urea and creatinine levels remained normal and there was nothing to suggest any kidney function impairment. For this reason we cannot reach the conclusion that these factors were linked to R’s deterioration and death.
62. In relation to R’s diabetes, the Trust gave her an insulin infusion. Our adviser said this suggests her blood sugar levels were closely monitored, and the insulin infusion was adjusted based on her blood glucose readings.
63. Our adviser explained managing blood sugar levels in critically ill patients is very challenging, and the priority is often to prevent low blood sugar (hypoglycaemia), even if this results in higher than normal blood sugar levels (hyperglycaemia). This may have been the reason R’s blood sugar levels tended to be elevated.
64. Another measure of effective diabetes management is the levels of blood ketones and acid, which can pose serious risks when significantly abnormal. In R’s case, they were within acceptable ranges, with only minimal increases.
65. Considering the above, we cannot say the actions of the Trust in relation to nutrition, hydration and diabetes management led to R’s deterioration and death.
Monitoring in the last week of August
66. Mrs A says the Trust did not monitor R in line with National Early Warning Score guidance (NEWS, a score-based system of monitoring to warn if someone is becoming more unwell). She says this led to a missed opportunity to give earlier treatment that may have prevented R’s death.
67. Mrs A did not tell us exactly when she thinks the Trust did not monitor R, in line with NEWS. The complaint she made to the Trust outlined her concerns that R was not moved to the ICU, despite her NEWS being frequently high.
68. Our adviser carried out a general review of the records and did not see anything in the charts to show R was not monitored as she should have been.
69. The RCP National Early Warning Score guidance says that when NEWS is above 7 it should trigger an urgent senior review and consideration of higher level care. There is no guidance to say a high NEWS should routinely prompt admission to the ICU.
70. At the end of August the Trust escalated R’s care because of her deterioration when her NEWS of 8 required this. This was due to the combination of her ongoing fever and spike in temperature, and a reduction in her level of awareness. The details of this are outlined in paragraphs 42 and 43.
71. The records show that at all points where R’s NEWS was elevated the Trust arranged reviews in line with the guidance. The critical care team reviewed her on many occasions between during this last week.
In conclusion
72. We are deeply sorry for the sad death of R and the ongoing impact on Mrs A. We understand Mrs A remains concerned that R was not given the care and treatment she should have been, and that this was related to R’s medical history of ME. We did not see any evidence to support the view that there were indications of failings in R’s care that had any relation to ME.
73. We understand how important this complaint is to Mrs A and we thank her for sharing her concerns with us. We hope she will be reassured that we have not found anything to make us think we need to ask the Trust to take further action in relation to the issues we considered.