Predicting recovery in patients thought to be dying


A woman in her mid 80s had a perforation and was not fit for surgery. She had not responded to five days of antibiotics. I was asked to see her because of SOB.

Her airway was clear. I noted her sats were 88% on 2L O2/min, and she had a diffuse expiratory wheeze. I prescribed a salbumatamol nebuliser, put the O2 up to 5L/min and took an ABG.

During my ABC assessment, Outreach asked me to think about her ceiling of care. 10 minutes later, after discussion with the consultant, she was on the LCP.

A few days later, the same patient made some slight improvements in her cognition and seemed to be getting slightly better. She requested specific foods, having not spoken for days. She had a decent urine output, although she was not catheterised. The consultant’s impression was that she may have recovered, as the gallbladder was draining into the small bowel, and not spreading infection into the peritoneum. The LCP was stopped.

The issue

She had been on the LCP for about a week, and was taken off it and kept in our hospital. However, later that day she was put back on the LCP. She died a few hours later.

Could this have been predicted?

Yes and no. I don’t think anyone is to blame here at all. Nothing in the LCP is ever 100% certain. However, there must be markers that suggest irreversibility and others which suggest improvement. Over the next few weeks, I intend to research and discuss this and write up what I find.

What have you found so far?

The question is, if we look back is there anything we could have done differently? The decision to be on the LCP must be regularly reviewed, as it is not a death sentence but a constantly renewed decision to do what is best for the patient. That said, after 7 days on the LCP, is a recovery possible? Her mouth was bone dry and had frank necrosis in it. She was massively edematous. She had not been eating or drinking for such a long time. Even if the underlying pathology had reversed, had the combined toll of the underlying problem plus the LCP with its lack of attendance to the vital signs and end organ perfusion left her in an unrecoverable state? And does this mean that if the LCP is maintained on any patient for longer than say 5 or 6 days, that is it probably now irreversible in any case?

In other words, is the LCP self-fulfilling? This is an accusation often made about the LCP, and is the subject of a DoH review due in summer 2013.

There was a case of a patient who was on the LCP for about 5 days during her admission, but today was discharged from hospital in back to her joking self and mobilising between bed and chair. How does one predict the different destinies of patients on the LCP when there are no observations, clinical assessments or investigations?

Things became clearer when I thought about shock. This progresses through a few stages, and the early ones are reversible. However, more and more aggressive therapy is needed the further the shock progresses. Eventually you reach refractory shock, which is when adenosine leaks out of cells greater than the maximal rate at which it can be synthesised, meaning that the cell will never generate enough ATP and death is inevitable. At this point, the focus must be on symptom relief, and anything else is completely pointless.  Even though it seems like just minute earlier we were doing all sorts of high-end resuscitation therapy for the patient who has just tipped into refractory shock, it does make sense to switch the treatment goals so suddenly. The decision to be on the LCP likewise should be when the physiology of the patient is irreversible, which justifies focusing on pure symptom relief.

Unfortunately, we do not have real time measures of adenosine leaking or any other specific markers of death. We can however consider what the patient’s baseline physiology was like, and how much of an insult it is likely to be able to take. We can look at response to treatment, and if this is not working after a certain length of time then perhaps it will never work. This is because if it did not work when the patient was functioning at 40% of his/her baseline and the patient is deteriorating, it is even less likely to work at 30% of his/her baseline. We can consider whether the disease process itself is terminal, such as end stage renal failure not for any renal replacement therapy.

What I think palliative care is about

I have been looking up general guidance on the LCP, mostly to understand where it came from and appreciate the nuances. I found this guide to the LCP medications, and noticed a few comments towards the end of the document about what good palliative care really is.

‘You matter because you are you.
You matter to the last moment of your life,
and we will do all you can,
not only to help you die peacefully
but to live until you die.’

Sometimes, that’s all there is to do. And we can either do it well or lose ourselves trying to do the impossible and reverse irreversibly deranged physiology, causing massive distress to the patient and family. Finally, the LCP has started to make sense to me.


Listen to the patient: He is teaching the students

“Sure thing. Teaching at 2:45pm, I’ll find an interesting case.”

It was 2:10pm by the time I finished clerking a new arrival and I still needed to have lunch. How was I going to find a case and prepare for it?

I then remembered a patient I had seen earlier on a ward round. He was an elective admission for an angio of the right leg because of 3 weeks of worsening claudication. He had no accompanying medical notes at the time. I had to clerk him fully, which turned out to be a blessing for the medical students.

The patient turned out to have some incredible, rare conditions. Each condition deserves a separate blog post, so for now I’d like to focus on his Parkinson’s disease for 5 years, managed on co-beneldopa and entacapone.

I initially planned to go through a Parkinson’s examination with the students in an OSCE format. Most of medical education on Parkinson’s makes it seem like this exotic disease with unique signs that you must exhibit, nod sagely as you demonstrate them before moving straight into a discussion with the examiner about the complications of long term levodopa. Unless we are working in neurology or care of the elderly, maybe its hard to appreciate what a day in the life of a patient with significant parkinsonism entails. I wanted to find out.

“How does the Parkinson’s affect you?”

“Well I can’t sign cheques for the business anymore, that’s for the missus now.”

Micrographia must be horrible. Hearing it from people unable to do their normal business as a result made me appreciate the significance of this. Imagine having a great idea and being unable to jot it down. I could really sense some of the frustration when I read about it in these forum posts.

“And I had this episode where I was walking along, and got completely stuck. I was there for a few minutes. It was weird. In my head I kept telling myself to move but I just wouldn’t. You feel like something’s very wrong and you wonder what you’ve done to cause it.”

The patient was describing the freezing phenomenon. Classically, the feet are rooted to the ground but everything above the waist works just as normal. You could even make a cup of tea if the table were close enough. The Parkinson’s UK guide for patients on freezing is so helpful. I remember being taught that patients with Parkinson’s may freeze when they get to an obstacle e.g. a narrow doorway but could follow a tape across the same obstacle. The principle is that in Parkinson’s, there seems to be a problem synchronising all the complex movements we have subconsciously mastered since we learnt to walk. Anything that stresses the mind or adds to the challenge of walking can leave a patient with Parkinson’s unable to just walk. However, if you can break up the task of walking into a sequence of smaller goals, this seems to get around the problem.

I had always thought that the reason people fall in Parkinson’s disease was because of a loss of postural reflexes and the gait disturbances. It seems it’s a multifactorial and sadly synergistic cocktail of problems, of which the freezing phenomenon is quite important. Again, the patient information leaflet on why people fall in Parkinson’s is a great read, and complements the freezing leaflet.

“The doctor told me to use the smallest dose of levodopa to make it last longer. I wish I could take more, but I know it’s for the best. It sorts everything out in about 30 minutes.”

It must be a daily struggle requiring a little bit of Zen to choose to not give yourself a medication that you know can take away the symptoms so effectively.

“What was the first symptom you noticed?”

“I had drooling right from the beginning.”

A recent paper suggests that the non-motor symptoms of Parkinson’s disease may predate the motor symptoms by years. In particular, “excess saliva, forgetfulness, urinary urgency, hyposmia, and constipation” are potential early symptoms. Drooling occurs in about half of all patients with PD.

I sense there may be a shift from seeing Parkinson’s disease as an isolated lesion of the dopamingeric neurons in the substantia nigra affecting extrapyramidal movements to a more holistic view of a neurological disease with a wide range of clinical manifestations. The non-motor questionnaire helps draw attention to these symptoms.

Medical students and MRCP candidates will probably have to churn out a Parkinson’s examination at some stage. I’ve made 3 things for this:

  1. A Parkinson’s OSCE/PACES examination crib sheet.
  2. A video of the patient’s tremor (with consent of course)
  3. A podcast on the management of Parkinson’s disease, designed to nullify any attempted grilling from your examiner

Medical students can make doctors perform better

Shifted around

The twilight shift agrees with my circadian rhythm like no other. The hours are 1600-0000, which is 8 hours just like your regular 9-5. But OMG, it just works so much better.

Wake up whenever you feel ready, go to the gym, prepare some OSCE teaching for medical students, come in around 2pm, find a patient and go through the examination properly with no risk of being bleeped. Meet all the other teams between 1600 and 1700 to check for sick patients and that they have done all the jobs which you should not be doing e.g. deciding if this patient is for the LCP this evening or whether or not to re-dress the wound this evening.

Such a straightforward shift compared to the weekend shift, where we do a ward round of the entire surgical inpatient population. This ward round goes from 8am till 2pm. Then do the jobs from that, plus all the weekend bloods for the entire surgical inpatient population, whilst getting bleeped for the usual cannulas/TTOs/fluids etc. And then you have the sick patients.

Teach others to teach yourself

A few medical students hung around to do the twilight shift with me last week. It was definitely a different experience having to explain my actions as I did them, and it probably made me more thorough than normal. I try not to offload cannulas and bloods onto them, but when it is busy then I justify it on the grounds that we can go through more teaching the quicker the mundane jobs are done.

Last Thursday seemed like it was going to be a disappointing collection of piddly jobs, without much for the student to learn from. You never want a patient to be seriously sick, but you do want the opportunity to go through the ABC assessment of a real patient with the student. At around 10pm, when we were thinking of letting the student go home, we got called to see a patient who had pulled out his NG tube and cannula.

This patient was a man in his early 60s who was day 4 of acute severe pancreatitis, scoring 6 on the Glasgow criteria. He had been declared not suitable for HDU/ITU, and was not for resuscitation. I had been warned by the outgoing day team at 5pm that he was ‘impossible’ to cannulate. I have recently discovered this savior vein between the thumb and first finger, about 3cm distal to the anatomical snuffbox. This vein is remarkably well tethered, and thankfully ignored by other cannulators.

Gray's anatomy - veins on the back of the hand

At 6pm, I decided to give it a go. Careful not to burst the vein by pausing as soon as I sight blood in a cannula to retreat the needle, I was treated to extensive flashback and a satisfying flush. I taped the cannula down with extra MicroPore tape, and walked away with the satisfaction of a job well done. I couldn’t wait to let the day team know I got it in first time. During this time, I conversed with the patient and he was well orientated.

ABC is always the right approach

For the same patient to now pull out everything a few hours later was significant. His sats were 92% on 2L of oxygen according to the saturation probe. This was where he had been for the past few days. However, his respiratory rate was 28, and his work of breathing was substantial, with use of accessory muscles and nasal flaring. He was barely able to talk in sentences, which was different. For the patient to need to breathe like that to maintain his sats was clearly abnormal.

I hunted where I had success before. The area was a bit swollen, but I could just about feel the ruins of the vein. The medical student had a different approach. She went straight for the right antecubital fossa, and found something there. I asked her to go for a pink cannula if she’s confident. She got it in. Boom.

During this time, we first fully appreciated his confusion and incredible work of breathing. I asked the student to go through an ABC approach.

His airway was patent. Now, although his O2 sats were where they were normally, he was clearly in respiratory distress. This meant O2 was necessary to reduce the work of breathing before he fatigues out. On examination of the chest, there were bilateral crackles.

Acute severe pancreatitis is a known risk factor for ARDS. ARDS occurs in about 10-20% of episodes of acute pancreatitis, and has a mortality of about 30-40%. It is strongly associated with pancreatic necrosis. This was my clinical impression at the time. I realised the patient was not a candidate for HDU/ITU, and ward based NIV does not happen in our hospital. This meant that the best ward based care would be oxygen. The fluid regime for this patient would also become tricky, as ARDS is treated with a fluid conservative approach, whereas pancreatitis is a state of ongoing third space losses.

The patient did not have any risk factors for hypercapnic respiratory failure, so we really should have gone full whack with the O2 at that stage.

Circulation wise, I decided to give a fluid challenge despite the bibasal crackles. He had worsening mental status and this was pancreatitis after all. His cap refill and urine output were fine, and his pulse was 112 from 100 at 5pm and his BP had dropped to 120/80 from 150/90 at 5pm. He had no history of cardiac problems, and I suspected a non cardiac etiology for the bibasal crackles (ARDS). I wanted to see if this made him better or worse, and it was only 250ml.

The response was a slight improvement in BP to 130/84 and pulse of 106. This suggested he was fluid deplete and could do with extra fluids.

We took an ABG on 2L of oxygen, which returned:

pH: 7.4
PO2: 7.7kPa
PCO2: 4.5kPa
BE: -4
Glu: 20.9

We put the O2 up to 5L / min via a facemask, and saw the saturation probe return values of around 94-96% which we were happy with.

I asked for capillary ketones, which came back as 1.4.

With our only IV line, I asked for an insulin sliding scale to get glucose under control, as the high glucose would dehydrate the patient further. I then informed the registrar, who would review the patient.

Bloods were taken, which showed LDH 1560, lactate 2.5, WCC 20.8 (similar to earlier), CRP 250 (similar to earlier) and unchanged eGFR of about 60. Corrected calcium was normal. LFTs were remarkably normal.

There was no baseline LDH level to compare to. The reason I took this was not just to reassess the severity according to the Glasgow criteria but to look for a rise, which may go with pancreas necrosis.  This is an area of controversy, but it makes sense that the release of a ubiquitous intracellular enzyme is a decent marker for cell death somewhere.

The registar reviewed the patient later, and started antibiotics. The evidence for antibiotics in pancreatitis is not definitive, but imipenam (good pancreas penetration) may be used when there is strong clinical suspicion of infection and/or more than 30% pancreatic necrosis.

Until then, the British Society of Gastroenterology has recent guidance on pancreatitis. And I must be more vigilant to go back to ABC when there is the slightest suspicion, and not let anything come in the way of a full assessment.

Eidt 10/4/13 – I have recently learnt that most of the time hypoxia in pancreatitis is not caused by ARDS. Have a look at an SBA on this.

Why CPAP works in acute heart failure and other bedtime stories

The potassium issues from the previous post resolved with IV magnesium, and the patient was discharged on oral magnesium indefinitely. It is ironic that oral magnesium supplements cause diarrhoea, which is the most common source of magnesium loss.

I have been wandering around the hospital looking for cases to learn from. It’s almost like being an safari, knowing that there’s probably something fascinating out there if you look long and hard enough. Last Friday there was an ITU admission of dermatomyositis with respiratory failure, which brought up plenty of discussion about NIV and invasive ventilation from the medical students. Let’s go through what we found out.

A 65 year old man presented with worsening fatigue and weakness for the previous 10 days, with a rash occurring at the same time.

The weakness seemed to affect “every movement”, and he was brought in by ambulance when his partner felt he looked awful and was poorly responsive.

By the time I saw him, he had already been on high dose steroids, and the most spectacular aspects to his rash were apparently gone. Some scaling from resolving Gottron papules was still visible around the knuckles, but there was no rash around the eyelids and his upper chest and face were normal.

This was not the classical presentation. The textbooks will speak of a proximal weakness that spares the distal muscles. Our patient seemed to have a particularly rapid progressing variant.

Anti Jo-1, which is present in only 20-30% of cases of myositis, signifies a poor prognosis. Our patient was negative for this. A CT abdo/thorax also looked for any suspicious lesions causing dermatomyositis as a paraneoplastic syndrome and was clear.

His CK was elevated around 5000, and was coming down. In myositis, the CK levels can be used to assess the response to therapy, as the muscle inflammation is of a background, continuous sort, leaking to a steady leak of CK that will reduce as the myositis is treated. In contrast, in rhabdomylosis following say a fall, the CK rise occurs as a response to the initial injury, which is usually one off. In rhabdomyolysis, CK rises within 12 hours of the muscle injury, peaks at 1–3 days and then declines 3–5 days after the original muscle insult. The peak CK level in rhabdomyolysis can be used to guide treatment and assess prognosis, but daily CK levels are less helpful for assessing the response to treatment once it has been shown to be falling.

I had never known myositis to be so serious as to cause respiratory muscle failure, but apparently respiratory compromise is not uncommon. Patient UK does not mention respiratory complications from myositis. I found a short paper on three case reports where patients with myositis needed ventilation support, which really helped me appreciate high-end spectrum of this disease.

This got me thinking more broadly about the indications for invasive ventilation.  I was wondering why this patient was not a candidate for NIV instead, given he was making efforts and there was no airway obstruction and there was no vomiting/secretions into the airway. Perhaps it was because of his depressed consciousness, leading to a potentially poor gag and risk of aspiration? I’ll have to find out from the anaesthetists on Monday.

I found this handy summary of the indications for intubation, which basically splits it into:

  1. Obtaining and maintaining an airway
  2. Correcting gas exchange
  3. Protecting from aspiration.

NIV can correct gas exchange, but cannot maintain an airway nor protect from aspiration.

The medical students also asked why CPAP is used for acute heart failure leading to pulmonary edema. In all honesty, I realised that I did not understand this well enough to teach it with confidence. I had a simple understanding that CPAP literally blows the fluid back into the pulmonary vasculature from the alveoli/interstitium.  I decided to read it over the weekend and explain it on Monday. I’d like to write down it while I understand it, and would love to hear if anyone else has heard any other explanations.

There are two problems:

  1. Fluid is in the alveoli and in the interstitium of the lungs. This is impairing gas exchange.
  2. The left ventricle has a limited amount of contracting power, so the cardiac output is not enough to get fluid out of the lungs. Ultimately, to fix the pulmonary edema we need to get the left ventricle’s cardiac output back up.

So what determines the cardiac output in a failing heart?

Well, we know that cardiac output is stroke volume x heart rate. This means we need to increase the stroke volume, the heart rate, or a combination.

The problem with increasing the heart rate is that this will impair myocyte perfusion, as increases in the heart rate decrease the proportion of the time the heart spends in diastole (when the coronary arteries perfuse the myocardium) whilst increasing the oxygen demand. The rate can only go so far.

On an aside, it’s worth pointing out that the diastolic pressure, which seems to be treated as the curious sidekick to the heroic systolic blood pressure, actually determines cardiac perfusion. In a tachycardic septic patient, you will never get systolic blood pressure up if the coronary perfusion pressure (diastolic pressure minus right atrial pressure) is 15mmHg.

So what else determines the stroke volume, and hence cardiac output?

When the preload is less than what the left ventricle is good enough to deal with, preload determines cardiac output.

This is the case in healthy hearts at most preloads. The left ventricle in Mo Farah can deal with pretty crazy amounts of preload thrown at it, and will just pump it straight out. No pulmonary edema for him.

When the preload is greater than what the left ventricle can deal with, afterload determines the cardiac output.

This is what has happened to our heart failure patient. The preload determines the amount of myocyte stretch you have just before the myocytes contract. In a heart failure patient, there simply isn’t the intrinsic contractility to deal with stretch. The myocytes will follow the normal Starling laws upto a point (say  15mmHg preload). After that, any further increase in the preload (say to 18mmHg) will only produce the contractility that their maximum contractility (at 15mmHg) will allow. You therefore have a limited cardiac output as preload increases, which means fluid congests in the inflow to the left atrium.

So, we need to find a way of increasing cardiac output. We can’t increase the preload. What we can do instead is reduce the afterload. When the afterload is reduced, the left ventricle can eject blood more rapidly, and empty a greater proportion of its volume into the aorta. This increases its stroke volume. It’s like squeezing a full water balloon connected to a hose full of high pressure water versus squeezing a full water balloon connected to an empty hose. For the same effort, you will empty much more of the water balloon into the empty hose.

This means our only ward-based option is to reduce afterload. We could go to ITU and use ionotropes to increase the contractility, but this comes with risks (especially arrhythmias), increases the myocardial oxygen demand and may worsen mortality.

CPAP has three key effects in acute left ventricular failure:

1. The extra intra-alveolar pressure counteracts the extra hydrostatic pressure in the pulmonary vasculature that was causing fluid to translocate into the alveoli and interstitium.

2. The increase in thoracic pressure gives a boost to the contraction of the LV ventricle. You could argue that it also impairs filling, but the effect on boosting contraction in a failing heart with no shortage of filling pressure is more significant.

3. The thoracic pressure is increased but there is no pressure increase outside the thorax. This is the equivalent from the LV’s point of view of pressure around the thorax staying the same, but the pressure outside the thorax being lowered. This will feel like a reduced afterload outside the thorax to the LV.

For these three reasons, CPAP is the king when it comes to pulmonary edema from heart failure.

Impossible potassium

The prize for the most extreme hypokalemia ever is not awarded solely for the magnitude of the hypokalemia. It is much more impressive when the potassium refuses to rise above 2.5 despite more than 160mmol of potassium being given per day for 5 days, as in the case of a lady in her mid 80s who had come in for an EVAR.

Her operation was uneventful. Day 1 post operatively, her potassium sank to 2.6 and never rose above this value till the day this post was written some 6 days later.


The causes of any serum electrolyte depletion can be split into inadequate intake, intracellular shift and excess losses.

Our patient was getting about 160mmol of potassium a day. Her intake was clearly more than sufficient.

Intracellular shift of potassium is usually caused by one of three things:

  • insulin (as in refeeding syndrome, the best condition ever)
  • a beta-adrenoreceptor agonist like salbutamol
  • alkalosis

Our patient was having some occasional salbutamol for recently feeling SOB, but nowhere near enough to cause K of 2.2 as it was yesterday morning. A venous bicarbonate was 23.

So she must have been she losing it from somewhere…

GI fluid is potassium rich, but post operatively there had been no diarrhoea or vomiting. A CT Abdomen showed no anastomic leak. She was on no laxitives.

What about urine?

She was on no diuretics. That said, her magnesium was 0.4-0.5 since the operation. Her electrolytes seemed to worsen when she had her fortisips resumed. I was thinking of refeeding syndrome, which as you may be starting to realise, is fast becoming my favourite thing ever (her phosphate was low-normal).

The team initially wanted to treat her magnesium cautiously, with oral supplements. However, the magnesium barely budged on this for 5 days. In fact, reading the BNF will reveal that oral magnesium is only for preventing recurrence of the deficit, not for correction of a deficit. I decided to switch to IV magnesium after 5 days of this, on the grounds that hypomagnesemia can make hypokalemia impossible to correct. I also discussed the case with the endocrine registrar, who suggested adding urine U&Es to determine if the kidneys are to blame.

Why exactly does a lack of magnesium lead to persistently low potassium?

See this review. The gist is that there are magnesium-inhibited ROMK channels in the late distal tubule/collecting duct that let potassium into the urine. If you lose the magnesium, those channels go into freeflow mode.

What were the results of the urine U&Es, and how did you interperate them?

The fractional Na excretion was about 1.2%, which is normal for a euvolemic patient. <1% would go with pre-renal failure, or a normal response to hypovolemia as the kidneys try to preserve sodium. Anthing above 2% is suggestive of a problem with sodium reabsorption, and as most reabsorption happens in the tubules this suggests a tubular problem, like acute tubular necrosis.

The normal bicarbonate pushes us away from renal tubular acidosis as a cause for the hypokalemia, although does not rule it out.

So what’s your conclusion?

In our patient’s case, the amount of potassium lost was inappropriate for the serum level of potassium, so this suggested the cause of the hypokalemia was renal. In light of the generally OK function of the tubules as suggested by the fractional Na excretion, and the lack of acidosis as suggested by the bicarbonate (though I do appreciate a normal bicarbonate is not proof of normal pH), this was not a tubular problem, and in particular not renal tubular acidosis. Something was selectively letting potassium leak through the kidneys.

My money is on low magnesium helping K to leak via the ROMK channel. I corrected the magnesium intravenously yesterday, and the magnesium level was back to normal this morning, with a K of 2.8. I’ll post what happens to the K tomorrow. If it’s still abnormal, we’ll have to call the endocrine gods.