Last updated: Lesson of the Month - January 2019…
on 01 Jan 2019

October 2012

Matthew Graham-Brown

An 85 year old male was admitted with a general deterioration in his physical state. He had been brought in by ambulance, when carers had found him to be more confused and less responsive than normal. The admitting SHO found it difficult to get any coherent history overnight, but did find some recent clinic letters that gave information about co-morbidities and drug history.


  1. Type 2 Diabetes
  2. Previous NSTEMI
  3. CKD stage 3 - presumed secondary to renovascular disease
  4. Gout
  5. Hypertension
  6. Heart Failure


  1. Ramipril 10mg OD
  2. Spironolactone 25mg OD
  3. Simvastatin 40mg ON
  4. Metformin 500mg BD
  5. Allopurinol 100mg OD
  6. Ibuprofen 400mg PRN


On examination

The SHO noted the patient was confused and disorientated, with an AMT (abbreviated mental test) score of 0/4. The SHO also noted that the patient looked dehydrated.

Bedside Observations:

  1. Temp - 34.2
  2. RR - 25
  3. Sats - 99% on air
  4. BP - 94/46
  5. Pulse - 110

General examination was unremarkable, and the JVP was noted to not be visible, with dry mucous membranes. Heart sounds were normal, but tachycardic and regular, and the chest was clear to auscultation. The abdomen was soft and non-tender, with normal bowel sounds, but the patient was noted to be incontinent of urine.

The initial diagnosis was recorded as:

  1. ?UTI
  2. ?Underlying dementia

The initial plan was recorded as:

  1. IV access
  2. Bloods - FBC, U&E, LFT, CRP
  3. Urine Dip + MSU if indicated
  4. CxR
  5. ECG
  6. IV fluids 1in 8
  7. IV antibiotics

Tests were done, fluids were started and the patient received IV antibiotics. He was reviewed on the post take ward round he following day, with the results:

  1. ECG - which showed sinus tachcardia
  2. CxR - slightly increased cardio-thoracic ratio, but no evidence of infection/pulmonary oedema
  3. Urine Dip - + ketones, but otherwise negative
  4. Bloods - Hb - 11.2, WCC - 9.8, Plt - 132, CRP - 18, Na - 152, K - 3.7, Ur - 26.2, Cr - 383, eGFR - 20

The patient was moved off the medical admissions unit overnight, and sent up to the geriatrics ward. He was reviewed on a post-take ward round on Friday morning by the geriatric team on the ward. His case and results were reviewed by the junior staff and the elderly care SpR, and the diagnosis continued to be:

  1. UTI
  2. ?Dementis
  3. 'Renal Failure'

The plan read as follows:

  1. Stop IV fluids (avoid volume overload - patient known heart failure)
  2. Encourage oral fluid
  3. Antibiotics to oral
  4. Repeat bloods
  5. Dementia screen when confusion resolved
  6. OT/PT and social review

The patient was given a predicted discharge date of Friday the following week, and he was not put on the weekend review list. Bloods were done, and blood forms were also put out for the Saturday morning phlebotomy round. Bloods results were not reviewed, and the patient was not reviewed on the Saturday.

The on-call weekend SHO was called to see the patient on Sunday, because the nurses were concerned that he was passing little urine and he had become more tachycardic, and hypotensive. When the SHO arrived she found the patient to be somnolent, and clearly dehydrated, with the following bedside observations:

  1. Temp - 37.2
  2. RR - 30
  3. Sats - 99% on 2L nasal cannulae
  4. BP - 78/41
  5. Pulse - 118

The patient had only passed 120ml of concentrated urine in the preceding 10 hours. The SHO also reviewed the bloods from the proceeding 2 days:

  1. U&E's Friday day time: Na - 153, K - 4.1, Ur - 31.1, Cr - 481, eGFR - 12
  2. U&E's Saturday morning: Na - 155, K - 5.4, Ur - 35.0, Cr - 526, eGFR - 9

She noted that 2 months ago the patients eGFR had been 65. She reviewed the drug kardex, and immediately stopped all medications (ramipril, spironolactone, metformin, simvastatin, ibuprofen and allopurinol) and reduced the IV antibiotic to 'renal' dose. She found the patient was confused and dehydrated, and took a venous blood gas along with a repeat set of U&E's. Whilst she went to check the gas, she asked the nursing staff to give the patient a fluid challenge of 250ml normal saline. Blood gas showed the patient had a metabolic acidosis:

pH 7.198, Lactate 4.7, Base Excess -14, Bicarb 12

On returning to the patient, the fluid challenge had improved the patients blood pressure to 86/53, and a further 250ml improved BP to 93/55. The SHO proceeded to fluid resuscitate the patient with emphasis to be paid to meticulous fluid balance. The SHO contacted the medical registrar to make them aware of her findings, and treatment, and to see if the registrar felt the patient would benefit from HDU care. The medical registrar came and reviewed the case, and agreed with the course of action taken. He also contacted the intensive care team for review. It was agreed with ITU that the patient should continue to be managed on the ward, as he was not considered a suitable candidate for renal replacement therapy. The renal team were contacted for review, who agreed that in the initial stages they needed to see how the patient responded to fluid resuscitation on the ward before decisions were made about higher level care.

The patient actually started to make some improvement over the coming days. He started to pass urine and his confusion improved, and his renal function also had started to slowly improve. Unfortunately, however, he contracted a hospital acquired pneumonia, and despite further antibiotic therapy he did not respond to treatment and passed away.


The intial management of this case was extremely poor. The admitting SHO did not recognise that the patient was even at risk of AKI, and presumed (as happens quite often) that the patient had a UTI, despite a negative urine dip, and normal inflammatory markers. The confusion was taken on the post-take ward round to be a potential sign of dementia, and at no point were significant efforts made to contact family or friends to find out about the patients normal functional status, or the circumstances that had lead to his admission. Nobody recognised (in the initial stages) that the patient was acutely unwell. Blood test were done, and not reviewed - this is unacceptable - and no formal handover of the patient was made to the weekend team to review blood tests done on the Saturday.

The SHO who reviewed the patient on the Sunday recognised the acutely ill patient, and took appropriate steps to rectify the situation. She involved seniors in her decision making, and ensured the patient was seen by appropriate teams. Sad to say that if initial resuscitation had been better, he may have improved and been discharged before he developed hospital acquired pneumonia. Recognising and responding to acutely abnormal physiology can be life saving, and as the weekend SHO demonstrated, it is simple measures (done properly) that make all the difference.


  1. Patients at risk of AKI need to be identified early. There was a new KDIGO (Kidney Disease Improving Global Outcomes) guideline published in MARCH 2012, (which builds on the excellent work done by AKIN (acute kidney injury network) and the group who developed the RIFLE classification of AKI). This guideline includes simple ways of staging AKI, and steps that should be taken/considered at each stage.
  2. Nephrotoxic medicines should be stopped in patients at risk of, or with established AKI, and good fluid balance is the corner stone of management.
  3. Fluid management can be difficult in patients with 'heart failure', but a diagnosis of 'heart failure' written in the notes is not a particularly helpful or meaningful statement. Heart failure (like AKI and CKD) is a syndrome, with many causes and many degrees of severity. In patients who have a diagnosis of 'heart failure' it is essential to get objective information to go with this - is it systolic or diastolic, right sided, left sided or congestive? - is there a recent ECHO, and if so, what is the ejection fraction? Are there any valvular abnormalities (murmurs/ECHO findings) - Have they seen a cardiologist and what did the most recent letter say? Once you have this information, you will have a better idea of how their heart pumps, and that can help you to judge fluid replacement. In terms of how much fluid to give, there is no universal answer to this, but by measuring simple parameters (BP, urine output, pulse) as you give fluid you can assess response. Do not just write up fluid and walk away, give fluid, assess response, ensure they are not getting fluid overloaded and proceed.
  4. Involving seniors and intensive care to help with decisions on more invasive management (such as in patients requiring renal replacement therapies) should be done sooner rather than later.
  5. A venous gas gives you essential information quickly and can also give you objective feedback of how well you are resuscitating/treating your patient.
  6. The geriatric team seemed quite keen to make a diagnosis of dementia, and whilst this may have been very important for the patient had they been well and nearing discharge, it is pointless trying to assess cognitive function when a patient has acute confusion (delirium). Serial abbreviated mental test (AMT) scores, over the course of admission can be very useful to see if the patients acute confusion is resolving, but formal tests for dementia should only be done in patients who have no acute confusion.
  7. As we have said in previous 'Lessons of the Month', communication between team members, follwoing up jobs, and handing over jobs and sick patients are essential to patient safety and are the cornerstone of good clinical care. If you do a test, look at the result and act appropriately - if you cannot look at it, it is your responsibility to make sure someone else does.

Back To Index