Clinical Case 113: Feeding Me Softly

Today’s case comes from Dr Trent Little – chronic Broome JMO and now my right-hand man in the Education Department (  i.e.. we share a desk ;-]  ).  This is a great Kimberely case – all the usual mix of third and first world medicine that just doesn’t happen much in the city.  Lets jump right in:

Solomon is a 44 y.o. man, he presents to the ED with a painful finger.

On examination he has a gangrenous looking right index finger. He is febrile, but not particularly looking that unwell. He also incidentally complains of having an increasing productive cough over the last week.

He is however well known to our department here and has a long medical history.
Previous TB
Chronic pancreatitis
Malnourished and ETOH abuse
Chronic normocytic anaemia
CKD stage 2

He reluctantly agrees to admission and is brought into hospital with presumed osteomyelitis and commenced on IV piperacillin / tazobactam tds to cover his chest and finger.

Some clever doctor thought we best do a full blood work up given his history which showed:
Hb =98
Na = 124 (134-146)
K = 2.4 (3.4-5.0)
Cl = 86 (98-108)

Urea = 8.4 (3.0  – 8.0)

Creat = 183 (about double upper range of normal for US readers.)

Phosp = 0.92 (0.8-1.50)
Mag = 0.38 (0.7-1.10)
Cal = 2.20 (2.15-2.60)
Alb = 37 (35-50)
CRP = 220  [Ed: we do not encourage the use of CRP around here, but if you must….]
Lipase = 220 (ever so slightly elevated)

His electrolyte abnormalities were thought to be largely due to malnutrition and chronic disease.

In addition to his anti-microbial cover he his given initially IV K + Mag and then high oral doses. 2g oral Mag tds and 1200mg BD slow K.
His K rises slowly as planned, but despite the high doses oral Mag his serum Mag concentration does not reach >0.4.

On the third day of his admission he absconds from the ward for a few hours to have a couple of cigarettes. On return he his tachycardic with a rate of 110. He is completely asymptomatic and he gets and ECG.

ECG shows:

ECG 1 113

new ST elevation in V3 V4 and perhaps V5. What do you think may be going on here? Would you treat for STEMI/ACS?

Advice from the remote cardiologist is sought.  Solomon is given


– enoxaparin  1mg/kg bid

– bisoprolol (bisoprolol was thought to be a better B-blocker option given his respiratory history).

The next morning his ECG had returned to normal and his high sensitivity Troponins were all negative at 6 & 12 hours

Now we are confused. What other investigations?

His bloods were revisited that morning.
Na = 126
K  = 4.2
Cl = 90
Cr  = 186
Urea= 8.4
Cal = 2.12
Phosp = 0.44
Mag = 0.38
Lipase 979
CRP is now 10  [Ed: wooohooo! Cured, discharge – hurrah for us…  sorry I just really don’t like CRP as a test. CP]

So, what was all that about then?  Whats the diagnosis ??

He was diagnosed with refeeding syndrome. Our ICU friends thought that the ECG changes were presumed myocardial dysfunction secondary to refeeding syndrome.

How do we manage refeeding sydnrome?  We would be interested to hear anyones opinion who has experience in this area.

What patients are at risk of Refeeding syndrome?

The best way of preventing re-feeding syndrome is identifying those who are at high risk. This BMJ article gives a nice summary.

Anyone not feeding for more than >5 days is at theoretical risk of refeeding syndrome. Our common high risk groups patients groups include:

Anorexia nervosa
Chronic Alcoholism
Postoperative (particularly GI surgery)
Oncology patients
Patients with uncontrolled diabetes
Malabsorptive syndromes such IBD, CF and chronic pancreatitis
Some patients who we would be less likely to consider, but who are also at risk include long term antacid users (Mag and aluminium salt binding phosphate) and long term diuretic users (loss of electrolytes).

It is largely a clinical diagnosis with a few characteristic biochemical abnormalities. Here is a nifty flow diagram from the BMJ review of the NICE guidelines for managing referring syndrome

refeed mx



  1. Katrin Hruska says:

    Great case! The best treatment is, as always, prevention!
    Sounds very unlikely that CRP would go from 220 to 10 in a day? We get CRPs on almost every patient who sets foot in the ED. It’s easy and fast, venous or just capillary POC. I don’t think I’ve ever seen anyone go from a three figure CRP to 10 in a day. A few questions:

    1. Did the patient get thiamine on admission?

    2. Why did they get an ECGs on an asymptomatic patient who had a mild tachycardia after going out for a smoke?

    3. What was the diagnosis that prompted the cardiologist to suggest full ACS treatment? STEMI? NSTEMI?

    • Thanks for your comments Katrin,
      This was about 3-4/7 later after antibiotic treatment that his CRP had reduced. I would also be interested to see what anyone thought about the rising lipase? This was also asymptomatic.

      1. Yes. This is almost routine for the majority of our patients given the high rates of alcoholism and malnutrition
      2. I believe this was nurse initiated. His PR was about 30-40 above his normal baseline. I presume the nurse was concerned about a possible arrhythmia. But yes, his PR could have just increased after his nicotine hit.
      3. I wasn’t involved in managing this case on the ward or the discussion with the cardiology consultant, but I believe they were concerned about a silent ischemic cardiac event, and were being cautious.

      • I think that an asymptomatic lipase rise deserves a CT scan; it’s like MIs, where a trop rise + ECG changes would make you quite worried about a silent MI; a lipase rise + CT changes would make you strongly consider treating the patient for pancreatitis.

  2. Hey Trent, I’m a fresh medical student, and I’m curious, what’s wrong with the CRP test? Not defending it, just interested in hearing an educated point of view on it, I hear this come up often enough, but people aren’t really clearing it up for me.

  3. The major issue with CRP and why you’ve probably heard the criticism is that it is not a particular specific test. It can be raised by a variety of things; trauma, surgery, bacterial/fungal infections and autoimmune disease. Where it is probably most useful is in monitoring trends and response to treatment. Fortunately we already have some useful ways of doing that, i.e. the patients clinical picture, temperature etc.
    Like any test you should have a particular clinical question in mind before you order it. In some situations it therefore can be useful. I know Casey loves to bad mouth CRP so he might have something more to add or point you in the direction of some useful resources.
    tw. @trentrural

    • It has it’s uses but the genuinely useful moments aren’t as numerous as the on-the-ground frequency the test gets used. I’ve seen it used for trend monitoring in the tertiary setting for serious infections, establishing the diagnosis of ARF etc where guidelines demand it etc. As always intelligent use of the test along with the history and clinical picture will give you a reasonable idea of how useful the test is going to be.

Speak Your Mind