Lack of assessments and delayed investigations and the medical team did not recognise Mrs H had sepsis and delayed treatment into it
28. Mrs B complains there was a lack of assessments and delayed investigations into Mrs H’s symptoms when the Trust admitted her. In its complaint response, the Trust explained it suspected she had a UTI or pneumonia and provided treatment. It also said at that point, there was no indication for arranging an abdominal CT scan.
29. Mrs B also says the medical team did not recognise Mrs H had sepsis and delayed treatment into it. The Trust said it initially treated Mrs H for an infection, and it did not know what the source of sepsis was. It said it thought it was urinary in origin as the site of pain was Mrs H’s tummy and she experienced burning when passing urine.
30. We considered these issues together as they are linked, with help from our A&E and physician advisers.
31. GMC guidelines say that if doctors assess, diagnose or treat patients, they must adequately assess the patient’s condition and promptly provide or arrange suitable advice, investigations or treatment where necessary.
32. NICE sepsis guidance says doctors should assess people with any suspected infection to identify possible source of infection, factors that increase the risk of sepsis and any indications of clinical concern. It also says that an immediate review should be carried out to assess the patient and think of alternative diagnoses to sepsis, blood tests should be taken and a broad-spectrum antibiotic (medication effective against a wide range of bacteria) should also be given.
33. Mrs H attended A&E on 28 May. She had lower abdominal and right flank pain, a fever and she also felt short of breath. She had low blood pressure that responded to fluids which the ambulance crew administered prior to arrival at A&E.
34. The assessment in A&E recorded her medical history and the A&E team carried out a clinical examination. This noted that Mrs H’s abdomen was soft with mild tenderness. She also had blood tests taken which showed a white blood cell count (measures the number of white blood cells in the blood, which defend inflammation and infection) of 15 and C-reactive protein (a protein produced by the liver in response to inflammation) of 321. These blood tests showed raised inflammatory markers (indicates inflammation in the body that can be caused by infections, injuries or chronic disease).
35. The A&E team concluded that Mrs H had potential sepsis, and it treated her for this with intravenous fluids (IV – inserted directly into a vein) and antibiotics (we further consider the antibiotic choice in the next issue). The team thought the potential source of sepsis was either from the urinary tract or chest infection and they arranged a chest X-ray and urine specimen. The assessment and investigations meet the GMC and NICE guidance highlighted above.
36. NICE diverticular guidelines tell doctors to suspect diverticular disease (small bulging pouches in the lining of the large intestine) if a person presents with intermittent abdominal pain in the left lower quadrant with constipation, diarrhoea or occasional rectal bleeds, or with tenderness in the left lower quadrant on abdominal examination.
37. On her initial presentation to A&E there were no signs of suggestive diverticulitis, leading to perforation and peritonitis (redness and swelling of the lining of the belly or abdomen).
38. There was also no obvious indication to carry out a CT scan of Mrs H’s abdomen in A&E, this is because the A&E team recognised her symptoms as possible sepsis and carried out the correct investigations.
39. We next considered what happened when the Trust admitted Mrs H from A&E to the medical admissions unit for further investigations.
40. The clinical examination at the time noted that her abdomen was soft, not tender but with some right-angle tenderness (towards the upper back) and no guarding or rigidity.
41. At that point, the diagnosis was felt to be more likely a UTI. The doctor switched Mrs H’s antibiotics, and the plan was to wait for blood test results.
42. On the post-ward round on 29 May, the medical team examined Mrs H, and she still had some right-side tenderness in her abdomen. It still thought the sepsis was urinary or chest in origin. The plan was for cautious IV fluids as giving too much IV fluid, too quickly may cause fluid build-up on the lungs. The team also arranged an ultrasound to exclude abscess in the kidney or any obstruction to the urinary flow, leading to hydronephrosis (swelling of the urinary collecting parts of the kidney), which may need intervention. This did not identify any obvious abnormalities.
43. The laboratory analysed the urine sample the same day and this showed no nitrites or leucocytes. These indicate bacteria and inflammation and the absence of these, makes a UTI less likely. It also reported the urine sample as no evidence of infection, on 30 May.
44. The medical team reviewed Mrs H on 29 May (on several occasions) and discussed her case with a consultant, but there is no evidence the team reviewed the diagnosis at that point.
45. RCP guidance recommends the nursing team aggregate observations as a National Early Warning Score (NEWS), which is an NHS system used to detect, monitor and respond to clinical deterioration in patients. A score of 7 or higher should prompt emergency assessment by a senior team or clinician.
46. NHS ‘Seven Day Services Clinical Standards’, explains that a consultant should review and see all patients with high dependency needs, at least once a day.
47. On 30 May, the nursing team noted that Mrs H’s NEWS was 8 and the hospital out of hours team were aware.
48. Between 30 and 31 May (a bank holiday weekend), there is no evidence in the medical records that the senior medical team reviewed Mrs H. This is not appropriate given Mrs H’s high NEWS. We also think that the senior medical team should have reviewed Mrs H’s diagnosis at this point, given there was no evidence of a urine infection, including carrying out a CT scan of her abdomen to further evaluate other possibilities. This means a failing in her care occurred.
49. On 1 June at 5am, the senior medical team reviewed Mrs H. During this examination, the doctor noted that her heart rate was high, her abdominal pain was coming in waves and there was blood in her stool. Mrs H had lower abdominal tenderness and localised guarding (a sign of irritation of the lining of the abdomen called peritonitis). The doctor prescribed IV metoprolol to control her fast heart rate and some IV fluids.
50. The doctor noted she had urinary sepsis, with diverticular disease and a bleed. They switched her antibiotics to Tazocin and requested a surgical review to consider a CT scan. The surgical doctor reviewed Mrs H promptly at 7.20am and agreed a CT scan was needed. The assessments and investigations requested, were appropriate and in line with GMC guidelines.
51. The critical care outreach nurse saw Mrs H at 11.30am and again at 4.10pm, along with the surgical doctor. The surgical doctor noted that Mrs H had a sudden onset of abdominal pain on 31 May, and her CT scan showed a perforation (leak) from the bowel.
52. The ‘Report on the Peri-operative Care of the Higher Risk General Surgical Patient’, indicates a CT scan for high-risk patients (which Mrs H was), should be within four hours.
53. The surgical doctor identified that Mrs H required a CT scan at 7.20am, it was carried out at 12.41pm and reported at 2.07pm. This is not in line with the information given above.
54. The findings were suggestive of a perforated diverticulum (occurs when an inflamed diverticular pouch bursts, causing a hole in the bowel wall) and the surgical team decided she needed an emergency laparotomy (high-risk surgical procedure involving a large incision into the abdominal cavity) which they carried out on 1 June.
55. In summary, the evidence shows that the assessment and investigations carried out in A&E, were appropriate and the medical team correctly identified that Mrs H had suspected sepsis. Although the medical team did not identify the source of sepsis straight away, it provided treatment.
56. When the Trust admitted Mrs H onto the ward, we have identified that the senior medical team did not review Mrs H for two days when she had a high NEWS and delayed requesting a CT scan. Furthermore, when the surgical doctor requested the CT scan, further delays occurred. These are failings in Mrs H’s care. The impact from this interacts with other issues, so we have considered this below from paragraph 73.
Monitoring 57. Mrs B says the nursing team did not monitor Mrs H consistently and when her NEWS was high, did not escalate her care to the senior medical team. In its complaint responses, the Trust acknowledged that while the nursing team carried out Mrs H’s observations, and NEWS scores recorded, it did not do this as frequently as it should have. We understand this was a worrying time for Mrs B and her family.
58. RCP guidance explains the clinical response that should be given in relation to NEWS and how often a nursing team should monitor a patient. If a patient has a NEWS of 1 to 4, they are at low risk, and the nursing team should monitor them every four to six hours. A NEWS above 5 indicates a review should take place, usually with a doctor or acute nurse team and the nursing team should monitor the patient every hour. A NEWS of 7 or more triggers an emergency assessment by a senior medical team.
59. We reviewed this issue with help from our nursing adviser.
60. Our nursing adviser explained the nursing team did not always monitor Mrs H in line with the above guidance.
61. There were some delays in the team monitoring Mrs H and where this should have been hourly due to her NEWS being 5 or 6, it went over an hour but not by a significant amount of time. This shows a failing in her care occurred.
62. We also considered whether the nursing team escalated Mrs H’s care appropriately to the senior team when required. The Trust acknowledged that the escalation of the NEWS scores was done so in the context that Mrs H had already received a senior medical review and there was an agreed plan of care in place.
63. It also acknowledged that over the bank holiday weekend, the nursing team did not always escalate Mrs H’s care to more senior physicians. It said if a senior clinician had reviewed her over the weekend, this may have prompted earlier decisions to arrange a CT scan to try to establish if there were additional sources of infection. It said this did not affect the overall outcome.
64. Prior to 29 May, the medical records document that the nursing team escalated Mrs H’s care appropriately when her NEWS scores triggered a deterioration.
65. On 29 May up to 2.42pm, Mrs H’s NEWS was between 5 and 8 and the nursing team made the medical team aware.
66. Between 3.56pm and 5.02pm on the same day, her NEWS was between 9 and 10 and there is no indication the nursing team escalated this to the senior medical team at the time.
67. After this time, her NEWS was 7 and there was a further delay, before the nursing team escalated this to the senior medical team late on 29 May.
68. On 30 May, Mrs H’s NEWS was between 7 and 8 and the nursing team escalated this appropriately to the medical team.
69. From 31 May onwards, the nursing team carried out Mrs H’s NEWS in a timely manner, and escalated her care accordingly to the medical team, when required.
70. On 1 June, Mrs H’s NEWS score was 4 at 2am, indicating the nursing team did not need to repeat it for four to six hours, and this was at the nurse’s discretion.
71. At 5am, Mrs H complained of abdominal pain and her NEWS was 7. The nursing team escalated this and made an urgent bleep to the medical team. After this the senior medical team reviewed Mrs H from 5 to 8am, which was appropriate.
72. In summary, there were some delays in the nursing team monitoring Mrs H and on 29 May, it appears the nursing team did not always escalate her care when the NEWS score was high. This is not in line with RCP guidance and is a failing in care. We have considered the impact from this, below.
Impact
73. We have identified there were some delays in the nursing team monitoring Mrs H, and on 29 May, the nursing team did always not escalate her care to the senior medical team when her NEWS score was high.
74. We also identified the senior medical team did not review Mrs H for two days over the bank holiday weekend when she had a high NEWS. Furthermore, the medical team delayed requesting a CT scan and when the team did request it, further delays occurred.
75. Our physician adviser explained that although there were some delays in the nursing team monitoring Mrs H, this did not contribute to Mrs H’s deterioration.
76. This is because Mrs H’s NEWS scores were high throughout the admission. Her NEWS was also 4 on 1 June, which shows that the lack of monitoring prior to this, did not affect her clinical state and her deterioration appeared to settle.
77. We think the lack of escalation and senior medical review, contributed to a delay in the medical team considering other possible diagnoses, including requesting an earlier CT scan. It is likely that an earlier CT scan would have demonstrated changes associated with the abdominal infection.
78. Unfortunately, even if the senior medical team had reached an earlier diagnosis, we do not know if this would have affected Mrs H’s chances of survival. This is because she had a severe condition and a significant risk of death due to this. We acknowledge the considerable uncertainty this will undoubtedly cause Mrs B.
79. Our complaint standards say organisations should look for continuous improvement and use the lessons learnt from complaints to make sure they do not repeat maladministration or poor service and apologise when things go wrong.
80. Within its complaint responses, the Trust acknowledged these failings and apologised for them, including that an earlier senior review, and CT scan may have established additional sources of infection. It has also explained a range of improvements it has made to its service.
81. This included a multidisciplinary daily safety huddle to take place on each ward to discuss patients with a high NEWS, an improvement group developing an electronic deteriorating patient alert which will visibly flag up patients with deteriorating NEWS and new mandatory training on sepsis and an improved patient screening deteriorating tool.
82. The Trust also raised a business case to increase the size of the critical care outreach team and staff and discussions took place to maximise Trust-wide learning in a forum with medical and nursing staff, in terms of achieving timely senior review, particularly out of hours. The Trust also said it discussed Mrs H’s care in relevant departments to promote future learning.
83. We think the steps taken by the Trust show reflection on the events, learning from it and improvements made to its service so that nursing staff appropriately monitor patients with high NEWS scores, and senior reviews are carried out where necessary.
84. These actions are in line with our complaint standards. Mrs B is also seeking a financial remedy, and we have made a recommendation in relation to this at the end of our report.
Tazocin 85. Mrs B says the medical team delayed administering Tazocin to treat Mrs H’s infection. The Trust said Tazocin was given on 29 May and on the post-ward round, the antibiotics were rationalised back to co-amoxiclav as this was an effective treatment for urine and chest infections. It said the medical team discussed the change of antibiotics with microbiology and the surgeon, and neither felt it would have made a significant difference (it did not find a bacteria resistant to co-amoxiclav).
86. It is, however, usual Trust practice to change antibiotics to Tazocin if there is further deterioration (in case of resistant bacteria). It acknowledged this was done later than recommended.
87. NICE sepsis guidance explains that a broad-spectrum antibiotic should be given to patients with suspected sepsis. The Trust suspected Mrs H had an infection, and this was urinary or chest in origin.
88. The medical team administered co-amoxiclav on 28 May when it suspected that Mrs H had sepsis. This is a broad-spectrum antibiotic used to treat a wide range of bacterial infections.
89. The medical team changed this and administered Tazocin on 29 May at 6.15am. Tazocin is also a broad-spectrum antibiotic and is typically used for more severe or broader infections and the BNF also recommends it for the treatment of sepsis.
90. The medical team changed this back to co-amoxiclav, and Mrs H received this antibiotic until the medical team changed it back to Tazocin on 1 June.
91. In line with NICE sepsis guidance, the Trust correctly prescribed Mrs H a broad-spectrum antibiotic when it suspected she had sepsis. We acknowledge that in line with Trust standard practice and the BNF, it should have changed to Tazocin sooner, when Mrs H continued to deteriorate. We have considered any impact from this failing, below.
92. Our physician adviser explained that prescribing co-amoxiclav rather than Tazocin first, is unlikely to have had any significant effect. This is because Mrs H had an infection in her abdomen, and unfortunately antibiotics alone could not have treated this. We therefore do not think the delay in administering Tazocin had an impact on Mrs H’s condition, although we recognise this was a worrying time.
Nursing staff moving Mrs H 93. On the morning of 1 June, Mrs B says that two nurses roughly turned Mrs H which caused her pain and may have caused her pelvic abscess to combust. We considered what happened, with help from our nursing adviser.
94. The NMC code says nurses should treat people as individuals and uphold their dignity by treating them with kindness, respect and compassion.
95. During the admission, and up to Mrs H’s deterioration on 1 June, the medical records note she was independent, mobile going to the toilet and turning herself in bed.
96. The nursing team carried out a patient handling risk assessment on 29 May. This stated that nursing staff would not be required to assist with manual handling. It said if nursing staff were required to assist, this would only require one member of staff.
97. Therefore, when nursing staff did move her on 1 June, it appears they took the risk assessment into account, and more than one staff member turned her. This is more than the risk assessment recommended. There is no evidence in the medical records that the nursing staff did this in a rough manner. The care provided was in line with the NMC code above.
98. Our nursing adviser also explained that turning someone roughly, would not cause a bowel perforation. We hope this provides some added reassurance to Mrs B.
Surgery 99. Mrs B has raised several issues that occurred when Mrs H had surgery on 1 June. We have considered these issues together as they are linked. We recognise this was a difficult and challenging time for Mrs B and her family and we do not underestimate the worry they felt.
Rectus sheath inserted incorrectly, and the surgical team administered the local anaesthetic incorrectly
100. Mrs B says that during the operation, the surgical team inserted the rectus sheath catheters incorrectly and did not aspirate them properly. She says the surgical team also administered the local anaesthetic incorrectly.
101. The rectus muscle is the ‘six pack’ muscle which covers the front part of the abdominal wall. An emergency abdominal operation usually involves a midline incision in the abdominal wall. At the end of an operation, placement of rectus sheath catheters (plastic tubes inserted into the muscle) allows the instillation of a low volume of anaesthetic directly into the muscle on either side.
102. The Trust said the surgeon inserted the rectus sheath catheters in a standard manner and bilaterally. It said they attached the two separate catheters to the Y-connector (a Y-shaped junction on the catheter) and filtered and aspirated via this, prior to the injection. It said the surgeon saw no blood, so injected the local anaesthetic down the catheters via the filter and Y-connector. Almost immediately after this, Mrs H suffered from a rapid fall in blood pressure and fast heart rate. The surgical team made a diagnosis of LAT and provided treatment.
103. It also said during the emergency treatment, the surgeon took the Y-connector and filter off and aspirated each catheter individually. Blood was clearly seen coming up the right side of the catheter adding to the likelihood of LAT, secondary to intravascular placement of the catheter. It said the surgeon put the catheters in using the right technique, but sometimes they end up in the blood vessel within the rectus sheath and if not detected, can lead to intravascular injection which would cause toxicity, even from a smaller amount of local anaesthetic.
104. The Trust said that bench testing following the incident, confirmed that aspirating each catheter separately (without a filter) prior to the injection may have increased the likelihood of recognising the intravascular placement.
105. Our surgical adviser said there are no national guidelines or standard operating procedures for insertion of rectus sheath catheters. However, there is accepted standard practice for inserting local anaesthetic into any body structure and this involves aspirating the catheters to check that there is not an inadvertent injury to a blood vessel prior to injection of local anaesthetic. There is also no accepted guidance on how to aspirate catheters that are connected by a Y-piece, and a surgical team can carry this out jointly or separately.
106. Before injecting local anaesthetic in any area of the body (using a needle or a catheter), it is correct practice for the surgeon to aspirate the needle or catheter, to check that they have not inadvertently injured or entered a blood vessel. Part of this technique is ‘blind’, even if the surgeon has placed local anaesthetic in the skin with a needle and syringe, it can cut into a small blood vessel. If the surgical team see blood aspirated into the needle, then it indicates that a vessel has been injured or entered and, in this circumstance, the surgeon should not inject the local anaesthetic. Various research papers include information about standard practices, including the BJA and McDermott’s papers.
107. At the end of the operation, the surgeon inserted rectus sheath catheters into Mrs H’s abdominal wall, and the right sided rectus sheath catheter was inadvertently placed in a small blood vessel. When the surgeon inserted the anaesthetic down the catheter, the anaesthetic solution went into her general circulation, causing an immediate detrimental effect on her. We understand how upsetting this time was for Mrs B.
108. There is no documented evidence in the medical records which describes the rectus sheath catheter insertion. Given the result of LAT directly related in time to the rectus sheath catheter insertion, we think it is likely the surgical team did not aspirate the catheter correctly before injecting the local anaesthetic. The catheters were likely to have been in the correct overall position, but the surgical team injured a blood vessel in this layer and did not identify this prior to injecting the anaesthetic.
109. This means a failing occurred and the surgical team did not aspirate the rectus sheath catheters correctly and administered the local anaesthetic incorrectly as this went directly into Mrs H’s blood stream. We consider the impact from this below.
The surgical team injected the wrong dosage and strength of local anaesthetic to Mrs H during the operation
110. Mrs B also complains the surgical team injected the wrong dosage and strength of local anaesthetic to Mrs H during the operation, and this was based on her actual body weight, instead of her ideal body weight.
111. The anaesthetist during Mrs H’s operation recommended 35ml of 0.5% of the local anaesthetic, levobupivacaine (175mg) as it is the maximum dose stated in the BNF, suitable for regional blocks. It said Mrs H’s weight was estimated at 110 to 120kg and if her actual body weight was used, a dose of 2mg/kg would have been at least 200mg (40ml x 0.5%).
112. The Trust said the surgical team confirmed that the anaesthetist should give 40ml of local anaesthetic, but had not realised that it was 0.5%, rather than the usual 0.25% of 40ml. The Trust said it is not known who checked the ampoules. It said an anaesthetist trainee agreed the dose of local anaesthetic in error and was too high based on ideal body weight. It said patients needing to be treated on an emergency basis, frequently have to have their weight estimated.
113. It said its review concluded that a range of human factors caused the error, including multiple distractions, handovers of responsibility, another urgent case needing to be dealt with rapidly and the use of a higher strength of local anaesthetic, all contributed to the high dose being given.
114. Our surgical adviser explained it is standard practice for two people to check the anaesthetic dose and volume together, prior to instillation.
115. Manufacturers of the local anaesthetic levobupivacaine, recommend a maximum single dose of 150mg (30ml of a 0.5% solution), while guidance provided by the BNF suggests giving the drug on a lean body weight basis in obese patients. At a dose of 2mg/kg and a lean weight of 70kg, this would have been 140mg (28ml of a 0.5% solution) for Mrs H.
116. Most publications, including the BJA paper, relating to the rectus sheath catheter technique suggest the use of 0.25% levobupivacaine. If 20ml of this concentration had been given down each catheter (40ml total volume), a dose of 100mg would have been given, less than the calculated maximum of 140mg and 150mg noted above.
117. The evidence shows us the surgical team therefore administered the wrong strength and dosage of local anaesthetic, and this should have been based on Mrs H’s ideal body weight, instead of her actual body weight. It also appears the surgical team did not correctly check the strength and dosage of the drug, prior to injecting it.
Impact
118. We have explained above that the evidence indicates the surgical team did not aspirate the rectus sheath catheters correctly and administered the local anaesthetic incorrectly. The surgical team also gave the wrong strength and dosage of local anaesthetic to Mrs H. This led to Mrs H experiencing LAT. We recognise how worried Mrs B and her family were at this time.
119. The clinical management of the LAT was prompt to counter act the side effects of the toxicity and in line with local anaesthetic toxicity guidelines. The surgical team then transferred Mrs H to the intensive care unit.
120. Our surgical and anaesthetic advisers explained that it was the inadvertent, IV injection of the local anaesthetic that caused the toxicity, that in turn caused a significant physiological change to Mrs H’s condition (immediate low blood pressure). It is likely it would have happened if the correct, lower levobupivacaine dose of 140 to 150mg had been given instead. It is therefore not the dose that caused the toxicity but the route of inadvertent administration, as much of the drug was given directly into the blood stream rather than between layers of tissue in the abdominal wall, and this resulted in toxicity.
121. This physiological insult coupled with the septic insult from a perforated diverticulum, rapidly worsened Mrs H’s condition and increased her mortality risk (likelihood of death within a specific timeframe and any mortality score above five is deemed as high risk).
122. Unfortunately, even without the local anaesthetic toxicity, Mrs H had a very high predicted mortality. This was 22.5% based on pre-operative scoring system which means about 1 in 4 people are expected to die from the same surgery. Furthermore, Mrs H could only walk 50 yards before getting short of breath. This indicates poor functional capacity which means she was a very high-risk patient and may not have been strong enough to cope with major surgery, with a higher chance of serious complications.
123. We think it is likely that although LAT may have contributed to Mrs H’s death, even on balance of probabilities, we cannot say how much. This is due to Mrs H being a high-risk patient, with a background of cardiac failure and systemic sepsis, having a severe illness and high-risk surgery which came with its own risks and possible complications, all of which means she may have sadly died anyway. We recognise this will lead to further uncertainty and distress for Mrs B.
124. The Trust has acknowledged what went wrong and offered its apologies to Mrs B, as well as completing a duty of candour letter. This apologised for the complications Mrs H suffered and any part that may have played in her death. The Trust also made several improvements following this complaint and completed an action plan. This included the following:
• wrote a standard operating procedure about the use of rectus sheath catheters with surgeons and anaesthetists to ensure best practice is followed, including recommendations regarding insertion, dosage and aspiration • posters introduced on all local anaesthetic cupboards stating that ideal body weight must be checked to calculate maximum body weight • amended the theatre etiquette policy to make it clear the person injecting anything is responsible for checking the drug and has the required knowledge of doses, actions and side effects and how to treat this • the Trust wrote a regional local anaesthetic policy and shared this to ensure standardisation of safe practice across all areas • human factors training and clinical simulation training was rolled out • the case was discussed at governance meetings and the case presented at the anaesthetic and surgical directorate for wider awareness and learning across teams • staff involved in the incident also reflected on the care given to Mrs H, including discussing it with educational supervisors and during regular appraisals.
125. We reviewed the changes made, along with our surgical and anaesthetic advisers.
126. They concluded that the Trust has learnt from the events and put in place learning and processes that will minimise a recurrence of the events. This is also specific to the errors the Trust made in Mrs H’s care.
127. We think the Trust set out what happened in Mrs H’s care, acknowledged what went wrong and took action to identify learning and used this to improve its services, as set out in our complaint standards. We hope this provides some reassurance to Mrs B and her family.
128. Mrs B is also seeking a financial remedy that the Trust has not provided. We have made a recommendation for this below.