ED attendance - 22 January 19. First we considered Mrs B’s ear complaint. When she attended the ED she reported her GP had prescribed her an ear spray and amoxicillin (an antibiotic) five days earlier, but this had not improved her pain. She reported she was also experiencing hearing loss, nausea and a fever.
20. GMC guidance says doctors assessing, diagnosing and treating patients must adequately assess the patient’s conditions, taking account of their history (including the symptoms), their views and values. It also says, where necessary, they should examine the patient.
21. The ED doctor reviewed Mrs B, took her history, and completed an examination. They noted Mrs B had an enlarged cervical lymph node (small structures in the back and sides of the neck and under the chin), which would indicate a local infection. Her ear drum was intact with no bulging which would indicate a middle ear infection and there was some erythema (redness) of the ear canal. No fever was noted on taking her temperature.
22. Our ED advisers told us the clinical examination of Mrs B was appropriate for the type of ear problem she presented with.
23. The ED doctor diagnosed her with otitis externa (inflammation of the external ear canal) and possibly media (inflammation of the middle ear). They prescribed a five day course of two different antibiotics and advised Mrs B to continue with the ear drops. They also prescribed painkillers and advised Mrs B to return to the ED if she was not improving or began to feel worse.
24. Considering the advice we have obtained and the account in Mrs B’s records, we can see Mrs B’s history was taken, her symptoms considered, and her ear was examined as we would normally expect to see. We are satisfied the care and advice for Mrs B’s ear symptoms was appropriate and in line with the GMC guidance.
25. Next, we considered Mrs B’s raised blood pressure. During triage the Trust checked Mrs B’s observations. These are measures of vital signs like heart rate and temperature. A score, called an early warning score (EWS) is calculated based on the results of the observations and this is used to assess the severity of a patient’s illness and their likelihood of deterioration, with higher scores being worse.
26. In Mrs B’s case her EWS was zero meaning there was a minimal risk of her deteriorating. However, her blood pressure was 198/83. BHF guidance explains a blood pressure reading of over 180/110 is an indicator of severe hypertension (abnormally high blood pressure).
27. NICE NG136 recommends taking an additional blood pressure reading at least an hour after an initial raised reading, when the patient is relaxed and has had pain relief. It explains this is because research shows just being in an emergency department can cause a significant raise in blood pressure. Additionally, it is widely recognised that pain can increase blood pressure.
28. BHF guidance lists symptoms of high blood pressure as blurred vision, nosebleeds, shortness of breath, chest pain, dizziness and headaches. Although the reading was raised, we have not seen any evidence Mrs B had any other symptoms of raised blood pressure. Our ED advisers confirmed there were no other concerning symptoms of very high blood pressure that would have suggested her raised blood pressure should have been treated any other way but routinely with further checks.
29. Therefore, considering Mrs B’s presentation and the guidance, the Trust should have taken a further routine blood pressure reading.
30. BHF guidance says readings between 140/90 and 180/110 indicate possible hypertension and require further checks. NICE NG136 explains that a systolic reading less than 180 on retesting (this is the first number), requires further monitoring and testing. Our ED advisers explained this would be completed by the GP, not the ED. In this scenario we would expect to see the ED refer Mrs B to the GP.
31. Had Mrs B’s systolic blood pressure remained over 180, NICE NG136 explains she should have had further tests at the hospital (we look at these tests in more detail later in our report).
32. In its complaint response the Trust said the doctor took a further reading and recalled the systolic reading as being 158. We were unable to identify any evidence to support the Trust’s account in the records. Additionally, we recognise Miss E’s concerns that this undocumented account is a recollection of events that happened months earlier.
33. It is possible the ED doctor’s recollection is correct, but when we weigh up the time that has passed, the number of other patients and blood pressure readings the doctor has likely seen and taken, and the lack of written evidence to support the reading, we cannot conclude it is an established fact that could be reasonably relied upon to form a balanced view.
34. Overall, we are not satisfied there is robust evidence a second reading was taken or that the Trust acted in line with BHF guidance and NICE NG136 when managing Mrs B’s raised blood pressure reading. This was a failing. We have looked at the impact of this later in our report.
ED attendance – 29 January 35. On 27 January Mrs B’s GP referred her to the ENT team at the Trust’s ENT emergency clinic for further review. This appointment was scheduled for 29 January.
36. On 29 January Mrs B also attended the ED and was seen by the triage nurse at around 2pm. She reported ongoing pain in the left ear. Her observations showed she had a raised blood pressure of 154/87 but as before, it was within acceptable EWS range and when combined with all other normal observations her EWS score remained zero. These observations would have again suggested there was a minimal risk of Mrs B deteriorating.
37. Her discharge notes say she was reviewed by a GP in the ED department there are no notes explaining the doctor’s assessment and conclusions so we cannot say if the advice she was given was accurate.
38. We cannot determine exactly what happened, however we can see Mrs B was seen a couple of hours later by an ENT specialist who completed a full assessment of her ear pain (which we address below).
39. As mentioned previously NICE NG136 recommends taking an additional blood pressure reading at least an hour after an initial raised reading. This means Mrs B’s blood pressure reading should have prompted a further check and may have again triggered a referral to the GP for additional monitoring. As her systolic reading was under 180, and as explained above, research suggests blood pressure decreases following pain relief, it’s very unlikely her blood pressure would have triggered any immediate intervention by the ED.
ENT appointment – 29 January 40. That afternoon Mrs B attended her arranged ENT appointment. She was reviewed by an ENT doctor who documented her history and symptoms. Notably her diagnosis of diabetes, disproportionate pain in her external ear canal after ear syringing, swelling and an enlarged lymph node.
41. The ENT doctor discussed the patient with the ENT consultant on call, as per the Trust’s process. The ENT doctor suspected malignant otitis externa, (also referred to interchangeably as necrotising otitis externa, this is otitis externa that has spread to the surrounding tissue and bone). Our ENT adviser explained this is a rare condition. NICE CKS otitis externa explains older age and diabetes made her a higher risk of it.
42. The ENT consultant did not agree with the doctor’s suspected diagnosis and advised to manage the patient as simple otitis externa and not malignant otitis externa.
43. NICE CKS otitis externa explains both otitis externa and malignant otitis externa cause inflammation of the external ear canal. However, it says malignant otitis externa causes unremitting pain (as opposed to mild discomfort found with otitis externa) and malignant otitis externa may be present in elderly patients with diabetes who have had previous irrigation (cleaning of wax from the ear).
44. We asked our ENT adviser for their view on the diagnosis. They told us Mrs B clearly met the diagnostic criteria for malignant otitis externa on account of her risk factor and pain. Taking into consideration the guidance and advice of our specialist we think the advice provided by the ENT consultant is not supported by the evidence.
45. NICE CKS otitis externa explains malignant otitis externa is a potentially life-threatening progressive infection. ENT UK guidelines recommend inpatient admission for further diagnosis and management including IV antibiotics (a method of delivering medication directly into the blood stream), general medical care and monitoring, and pain management.
46. On the advice of their consultant the ENT doctor prescribed ear drops on a pope wick (a sponge in the ear canal to enable drops to be used in the presence of swelling) and discharged Mrs B home with a plan to review her in clinic two days later.
47. Whilst the ENT doctor’s clinical assessment was appropriate, the advice provided and follow up action was not in line with the relevant guidance, and Mrs B should have been admitted to hospital. This was a failing. We consider the impact of this below.
ED attendance – 31 January 48. Mrs B attended the ED again two days later, on 31 January. She raised concerns about her ongoing pain and explained she had seen no improvement with her recent treatment.
49. Mrs B’s observations recorded her blood pressure was still raised at 162/76, but as before it was within acceptable EWS range and when combined with all other normal observations her EWS score remained zero. These observations would have again suggested there was a minimal risk of Mrs B deteriorating.
50. Records show Mrs B reported she had been seen by the emergency ENT clinic, the clinic told her she would be referred for follow up ENT services and she was waiting on a further appointment. The Trust noted Mrs B needed further painkillers, and she was prescribed a dose of liquid morphine (a painkiller used to treat severe pain) and then discharged home.
51. We asked our ED advisers if this was an appropriate assessment. They explained given the specialist team involvement (ENT) the ED would not need to change her management plan. Pain appeared to be her presenting complaint.
52. Whilst we know the ENT team should have admitted Mrs B to hospital two days earlier, when we consider the actions of the ED doctor we do this recognising they are not ENT specialists and they will not have the same insight into ENT conditions as an ENT doctor.
53. Had Mrs B not been actively under the ENT team we may have expected the ED doctor to involve an ENT specialist at this time. However, as Mrs B was receiving specialist input and appeared to be seeking pain management from the ED, we consider the ED doctor acted in line with GMC guidance in relation to Mrs B’s pain.
54. Mrs B’s systolic blood pressure reading was over 140 on arrival at the ED. As set out above, NICE NG136 says this should be retaken an hour later. We have again seen no evidence this was retaken. This was not in line with guidance, this was a failing. We look at the impact of this later.
ED attendance – 3 February 55. Mrs B next attended the ED two days after her last attendance, she had not been reviewed again by the ENT team as planned. The Trust told us there is no record it provided Mrs B with an ENT follow up appointment as documented.
56. Mrs B arrived in ED just after midnight reporting concerns with her ongoing ear pain. A nurse triaged her shortly after arrival. The blood pressure reading is difficult to read in full, but it clearly shows a systolic measurement of 174. We cannot determine the diastolic measurement. For the purposes of the EWS score the diastolic measurement is not a factor.
57. As before, the systolic reading and all other observations were within normal ranges. Mrs B’s EWS score was zero. Mrs B was given pain relief around three hours later. She decided not to wait for further care and self-discharged. Our ED advisers confirmed there was no reason she did not have capacity to make this decision.
58. Miss E says her mother’s blood pressure was checked twice in this admission and increased during that period from 159/74 to 174/58.
59. We have carefully reviewed the records and have found only one reading during this admission. We note there was a recording of 159/74 taken when Mrs B was initially brought in again by ambulance later that morning.
60. Whilst we cannot see two readings were taken, on the one reading we have seen the systolic reading indicated a further blood pressure reading should have been taken after at least an hour, in line with NICE NG136. We cannot see this happened. However, on this occasion Mrs B self-discharged, this means we cannot be critical the further reading was not taken, as her review and treatment was not completed.
61. Miss E also complains the Trust did not check how much pain her mother was in. A pain score is a way of understanding a person’s pain through a number or picture. In its complaint investigation the Trust acknowledged it should have asked Mrs B to score her pain, but it did not.
62. As all parties agree something went wrong here, we have not gone into detail about what happened. Below we have looked at the impact of the Trust’s failure to check Mrs B’s pain score.
Impact of the failings identified Raised blood pressure 63. The Trust should have taken a further blood pressure reading on 22 January, 31 January and 3 February.
64. Had a further reading shown Mrs B’s repeat blood pressure had been over 180/120 in ED, NICE NG136 says she should have had blood tests taken to look for organ damage, a urine sample to look for albumin (an indicator of kidney damage) and an electrocardiogram (ECG - a test that checks the heart’s function).
65. Considering Mrs B’s records from her admission in February, it is likely blood tests on 22 January, 31 January and 3 February would have been normal as her renal function and blood sugars were within normal limits even after she suffered an intracranial haemorrhage.
66. Had there been concerns on the ECG, NICE NG136 recommends repeating the blood pressure measurement in seven days. This would have been referred to the GP in the discharge notes. If there had been concerns with the repeat testing the GP would consider arranging 24-hour blood pressure monitoring.
67. If Mrs B’s blood pressure was still high on retesting, but below 180/120, as she had no other symptoms of raised blood pressure, she would have been referred back to her GP for a routine repeat blood pressure measurement, and if necessary, any relevant follow up (blood, urine and ECG testing and 24-hour blood pressure measurements).
68. In either scenario complete testing and diagnosis of any underlying blood pressure condition is likely to have taken several weeks. Mrs B sadly suffered the bleed 12 days after her initial visit to the ED. Additionally regular readings were taken in A&E during this period and did not prompt immediate intervention.
69. For these reasons we cannot conclude the further readings would have made a difference to Mrs B’s outcome, as associated investigations are unlikely to have led to any active treatment (if found to be required) before she sadly suffered the bleed.
Failure to admit on 29 January 70. ENT UK guideline says the ENT clinic should have admitted Mrs B for IV antibiotics, general medical care and monitoring, and pain management. We looked at what difference this would have made.
71. The postmortem report does not specifically comment on the ear canal or temporal bone (where malignant otitis externa would have been seen) so we do not have any evidence of the impact of Mrs B’s infection in this area. However, the postmortem clearly excludes infection as the cause of the bleed or death.
72. Based on the information that is available to us we cannot make a link between the failing to provide IV antibiotics to treat infection and Mrs B’s death.
73. As an inpatient Mrs B would have received regular observations, including blood pressure monitoring through regular EWS checks. As previously mentioned, research shows pain management would likely have reduced her blood pressure. However, we have also seen repeat readings where her blood pressure remained raised. Whilst the readings that were taken were raised, Mrs B’s blood pressure never scored outside of the EWS parameters, and she did not present with any other symptoms of chronic hypertension.
74. The postmortem also excludes blood vessel damage as a cause of the bleed or death. Blood vessel damage is caused by chronic uncontrolled periods of blood pressure, that build up over time.
75. Taking account of the likely monitoring Mrs B would have received and considering the evidence within the postmortem we cannot conclude additional monitoring as an inpatient would have made a significant difference. We recognise Miss E’s view that her mother’s admission would have prevented her death. Whilst there is not sufficient evidence for us to say this was the case, we recognise her mother should have been admitted and not having certainty about what difference this would have made has a lasting impact for Miss E.
76. Admission would also have provided better pain management. Disproportionate unremitting pain is a primary symptom of malignant otitis externa. Mrs B reported to the ENT doctor that co-codamol (a pain killer used for moderate to severe pain) made no difference to her pain and she attended the ED to seek relief. Additionally, research into pain explains that the severe, disproportionate pain caused by the condition has a severe impact on patients’ mental health and quality of life.
77. The records show liquid morphine offered her some pain relief, but it could not be easily obtained, resulting in frequent ED trips with lengthy, uncomfortable waits. Had Mrs B been admitted, she would have likely received consistent reviews of her pain level which would have led to overall better pain management.
78. From this we can determine Mrs B likely experienced significantly more pain than she would have if not for the failing. We were very sorry to see this was her experience, we appreciate this must be difficult for Miss E and the family to know. We go on to make recommendations to address this below.
3 February 79. The Trust recognised it did not check Mrs B’s pain score. A high pain score would have meant Mrs B would have been seen in the ‘majors’ area of the ED, and she would likely have received pain relief sooner. It is possible this could have led to her seeing a doctor sooner, however we do not know how long the wait would have been and it may still have been that she decided to go home, which was around three hours after arriving.
80. We have also seen there was a further opportunity to better manage Mrs B’s pain in this visit. As mentioned above she had extreme pain that was clearly very distressing for her. We do not underestimate how uncomfortable she was.
81. We have already explained it is widely recognised that pain can increase blood pressure. We looked at how likely it was pain caused Mrs B’s bleed.
82. NICE CKS hypertension looks at interactions between hypertension and intercranial bleeds. It explains short periods of high blood pressure, such as those caused by pain alone, are called malignant hypertension. This is where rapid onset of high blood pressure causes damage to the organs. This is a very rare condition effecting only 1 in 50,000 people a year. The postmortem findings and tests completed on admission did not suggest this happened to Mrs B.
83. When we look at the available evidence, the low chances of this type of bleed happening and the lack of organ damage seen in testing and found by the pathologist, on the balance of probabilities we cannot say the Trust’s failure to manage Mrs B’s pain led to the fatal bleed.
Conclusion 84. In summary Mrs B experienced avoidable pain that could have been more successfully managed had she been admitted sooner at any point from 29 January.
85. On the balance of probabilities, we cannot see the failings caused her death. However we recognise things went wrong, Mrs B became unwell very soon after, and we cannot give a certain answer. This means Miss E and the family will be left with some ongoing uncertainty. We understand how hard this will be for them.
86. We have set out what the Trust should do to remedy the impact of the failings we have found below.