A Fridge too Far?

How apt that this morning I watched one of my favourite episodes of “The Simpsons,” “King-Size Homer” from series 7, in which Homer gains 60 Lbs in order to reach 300 Lbs, allowing him to work from home on a disability scheme.  As always, The Simpsons provide a very accurate commentary on modern society, in this case the obesity epidemic.  The episode contains one of my all time favourite lines from a Simpsons episode; when Homer is refused entry to the movie “Honk if you’re Horny” on account of the cinema seats being inadequate for his girth, a fellow movie-goer jibes “Hey, fatty. I’ve got a movie for ya – A fridge too far!”

Anyway, what does this have to do with intensive care.  Well an early release article appeared the other day in “Critical Care” that caught my eye.  Abhyankar et al retrospectively reviewed a large ICU database from a single hospital in Boston, MA.  Sounds dodgy, but the database had 16,812 patients covering a 7 year period from 2001-2008.  Some of the over 25,000 patients in the MIMIC database were excluded (children, no weight available) leaving 16,812 for analysis.

The authors compared 30 day and 1 year all cause mortality with BMI, using the standard WHO categories of underweight (<18.5), normal weight (18.6-25), Overweight (25-30) and Obese (>30).  They controlled for SAPS score, comorbidities, gender, insurance status, and ethnicity.  Height wasn’t available for 25% of patients, so they had a height allocated based on their other demographic factors.  The final analysis was conducted with and without these patients.

The results were pretty impressive.  OR for mortality in overweight patients was 0.81 at 30 days and 0.68 at one year.  Even more impressive were the results for obese patients, OR for death 0.74 at 30 days, 0.57 at one year.  So if you’re obese you’re almost half as likely to die following an ICU admission than someone who is normal weight.  Of note, the survival advantage disappears for the very obese (BMO >40).  Results were all significant.

I think that the authors did a pretty good job controlling for potential confounders like diabetes, obesity related cancers, age, SAPS score, etc.  The results were also similar when the 25% with guessed heights were removed.

So what’s going on?  The authors have a couple of hypotheses.  One is related to immunomodulatory effects of apidocytes.  The other is a bit more simple – overweight and obese people have more nutritional reserves, giving them a survival advantage.

So perhaps one of the most effective lifesaving tools that we have in the ICU is a bucket of fried chicken…?

The paper is open access, so have a look for yourself.  Here’s the abstract if you’re short of time.


Lower short- and long-term mortality associated with overweight and obesity in a large cohort study of adult intensive care unit patients

Swapna AbhyankarKira LeishearFiona M CallaghanDina Demner-Fushman and Clement J McDonald

Critical Care 2012, 16:R235

Published: 18 December 2012

Abstract (provisional)


Two-thirds of U.S. adults are overweight or obese, which puts them at higher risk of developing chronic diseases and of death compared to normal weight individuals. However, recent studies have found that overweight and obesity by themselves may be protective in some contexts, such as hospitalization in an intensive care unit (ICU). Our objective was to determine the relationship between body mass index (BMI) and mortality 30 days and one year after ICU admission.


We performed a cohort analysis of 16,812 adult patients from MIMIC-II, a large database of ICU patients at a tertiary care hospital in Boston, Massachusetts. The data were originally collected during the course of clinical care, and we subsequently extracted our dataset independently of the study outcome.


Compared to normal weight patients, obese patients had 26% and 43% lower mortality risk at 30 days and one year after ICU admission, respectively (OR 0.74 [95% CI, 0.64-0.86] and 0.57 [95% CI, 0.49-0.67]); overweight patients had nearly 20% and 30% lower mortality risk (OR 0.81 [95% CI, 0.70-0.93] and 0.68 [95% CI, 0.59-0.79]). Severely obese patients (BMI [greater than or equal to]40 kg/m2) did not have a significant survival advantage at 30 days (OR 0.94 [95% CI, 0.74-1.20]), but did have 30% lower mortality risk at one year (OR 0.70 [95% CI, 0.54-0.90]). There was no significant difference in admission acuity or ICU and hospital length of stay across BMI categories.


Our study supports the hypothesis that patients who are overweight or obese have improved survival both 30 days and one year after ICU admission.

Own the Bronchial Blocker, Rent the Double Lumen Tube


You are the ICU registrar in a small teaching hospital.  Its 0300 and you get a panicked call from ED.  The nurse on the end of the phone tells you that they have a patient with a massive pulmonary haemorrhage who is agitated and hypoxic.  You quickly arrive to find the ED registrar struggling with a man in his 40s who has torrential haemoptysis.  He is agitated and combative, tachypnoeic and hypoxic with sats in the low 80s.  He keeps pulling off his oxygen mask.  A CXR taken just before he deteriorated shows a white-out on the left.  The anaesthetist on call has been contacted but is at least half-an-hour away.  It’s up to you to save this man’s life.

“The spitting of pus follows the spitting of blood; consumption follows the spitting of this, and death follows consumption…”


A long time ago

An apt quote borrowed from an article in the Annals of Thoracic Surgery from before I was born, but as true today as it was two thousand years ago.  There can be few scenarios in critical care medicine as terrifying as the patient with torrential pulmonary haemorrhage.  I can recall a couple from my brief career that both ended in the patient dying and me having to change my scrubs (because of blood, but they were pretty scary….)

After a discussion at a Sydney HEMS Clinical Governance Day recently, I thought I’d delve into the management of a patient with pulmonary haemorrhage (AKA massive haemoptysis) and in particular the techniques available to the non-thoracic anaesthetist.

The definition of massive haemoptysis varies, but most commonly quoted is somewhere between 200-600mL of blood loss per day.  Some have pointed out though, that blood loss can be difficult to quantify, and perhaps a more appropriate definition is any haemoptysis that significantly interferes with respiratory function.  These people are very sick, mortality as high as 80% has been reported.  They tend not to exanguinate.  Rather they tend to die of asphyxiation, so the main focus of management should be ensuring oxygenation by any means possible.

Lets briefly review the causes of massive haemoptysis.  The commonest cause depends on where you are.  In most parts of the world the leading cause of massive haemoptysis is tuberculosis, particularly when associated with bronchiectasis and cavitation.  In rich countries, cancer is the commonest cause.  Other causes include;

  • Trauma
  • Aspergilloma
  • Bronchiectasis
  • Aorto-pulmonary fistulae (scary)
  • AVMs
  • Lung abscess / necrotising pneumonia
  • Mitral stenosis
  • Autoimmune / vasculitides (e.g. Goodpasture’s, SLE, Wegener’s)

There are a couple of very context sensitive causes too;

  • Haemoptysis every month in a woman of childbearing age: pulmonary endometriosis
  • Massive haemoptysis in a patient in cardiac ICU: pulmonary artery rupture secondary to PAC.

Lets not forget contributing (and correctable) factors like coagulopathy and neutropenia.  And more importantly, lets not forget that the blood might not be coming from the lungs.  It could be coming from the nose, mouth, throat or gut.

Management is focussed on ensuring oxygenation.  The most experienced intubator should manage the airway.  The management plan depends on whether the bleeding is torrential and the patient dying before your eyes, or you think you have a bit of time.  If the bleeding isn’t immediately life threatening, the best plan is probably to intubate with the biggest tube you can get your hands on (a 9 or 10 if you can find one) and then get someone who knows what they’re doing to bronch the patient and hopefully find the source of bleeding.Obviously this is a perfect setting for delayed sequence intubation too.  The combative, hypoxic patient may respond well to a dose of ketamine in order to allow some form of pre-oxygenation.  Techniques such as local adrenaline or tranexamic acid could then be attempted.  The large tube also allows frequent suction of the clots that will block the tube.

If the bleeding is torrential, the bronchoscope is likely to be useless, and embolization or even surgery may be required.  In this case the best course of action is probably to isolate the bleeding lung with a double lumen tube or a bronchial blocker.  Hopefully you’ve been able to acquire a chest X-Ray and localise the source of bleeding to one lung or the other.

Traditionally the domain of the thoracic anaesthetist, double lumen tubes (DLTs) are big, scary and potentially lethal if you don’t know what you’re doing.  However they can also be potentially life saving, and may be worth a try in a dying patient.  Even if you’re not the doctor putting them in, occasionally a patient will end up in an ICU or a resus room with one in, so all docs and nurses working in critical care need some familiarity with them.  So lets have a look at them in a bit more detail.

Double lumen tubes have been around for a while now and are most commonly used in thoracic surgery when the surgeon needs one lung deflated in order to operate on it and the patient needs the other lung inflated in order to live.  They have a large external diameter, but obviously each lumen has a smaller internal diameter than a single lumen tube, which could be problematic if there are lots of clots coming up.  The left sided DLT is used most often.  Right sided tubes should only be used if an anatomic anomaly precludes left sided DLT placement, as the Right upper lobe bronchus can be easily occluded by a right sided DLT (which has an extra ventilation port in order to minimize this risk.)

For those of you needing a bit of a refresher on Bronchial anatomy, check out the amazing bronchoscopy simulator at thoracic-anesthesia.com (and stay tuned for “own the bronchoscope” in the near future)

Here a great Youtube video by PainH8R (who have a series of excellent procedural videos) that gives a pretty good rundown on the basics.

It is important to note that bronchoscopic guided insertion is unlikely to be successful in torrential haemorrhage.

DLTs are sized using the French scale (outer diameter in mm x 3) rather than the traditional scale of inner diameter in mm for single lumen ETTs.  Sizing is typically based on patient’s height (you can measure the left main bronchus on the CXR and use that, but that seems like a lot of work in an emergency) with most adult females being a 37 and most adult males being a 41.

Insertion is tricky, even for experienced intubators.  It is demonstrated pretty well on this Youtube video though.

Once it’s in you inflate both cuffs and then isolate the bleeding lung by clamping the circuit just proximal to the suction port of the appropriate lumen.  As mentioned in the videos, by convention the bronchial lumen is blue (All the Bs), which will be the left side >90% of the time.

Hopefully the anaesthetist has arrived and spared you the trauma of inserting a tube that looks like it’s as big as your fist.  But once the DLT is in, you probably can’t leave the patient’s side, as there’s a lot that can still go wrong.  The average ED or ICU nurse has probably never even seen a DLT before and they can easily become malpositioned leading to catastrophe, so everyone involved in the patient’s needs to know that there’s something special going on with the airway and breathing.

As exciting as DLTs are though, the device that tends to appear from nowhere when someone is dying of haemoptysis is the bronchial blocker. This device looks like a small foley catheter, and is inserted into the bleeding lung with bronchoscopic guidance if possible (blind insertion is possible though by rotating the tip clockwise or anticlockwise in an effort to direct it into the appropriate lung).  Once in position the balloon is inflated, thereby isolating the bleeding lung and preventing soiling of the good lung.  You definitely need a CXR first if you’re going to use a bronchial blocker.  A special 4 way adaptor that comes with the bronchial blocker allows you to insert the blocker, use the bronchoscope and ventilate all at the same time.


The benefit of the bronchial blocker is that it can go down a normal ETT (however special tubes with a separate port for the blocker are available), which is a good thing is staff are unfamiliar with DLTs.

Of course our friend PainH8R has prepared an excellent video on the bronchial blocker too too…

Once the patient is able to be oxygenated effectively its off to either angio or OR (or the hybrid suite if you’re lucky and rich) for definitive management.  If the patient is stable enough after lung isolation you can have a go with the bronch, but if the bleeding was so severe as to require emergent lung isolation, bronchoscopy is unlikely to be helpful.

Something worth dwelling on is what issues might arise once you have isolated a lung, either with a DLT or a bronchial blocker.  One lung ventilation (OLV) is an unusual physiologic state and it’s worth reviewing.  The take home message is that hopefully not much will change.  The patient will effectively already be on one lung ventilation due to soiling of the affected lung.  Poor oxygenation of the affected lung creates a shunt, and (hopefully) leads to hypoxic pulmonary vasoconstriction, which decreases perfusion and strives to minimise ventilation/perfusion (VQ) mismatch.  This means that almost the entire cardiac output goes through the ventilated lung, which may worsen VQ mismatch by increasing physiologic dead space if ventilation to the ventilated lung is poor.  Hypoxaemia in one lung ventilation in the ICU should initially be treated by increasing PEEP and FiO2.  It is also important to consider that hypoxaemia may be due to obstruction or malposition of the tube, especially if a DLT is in place.

Ananthem et al wrote a good review of OLV in the ICU which is worth a read (it’s open access)

So there you have it, an idiots guide to a potentially lifesaving procedure.  All you need to do now is visualise the procedure a few times in a variety of different settings (or even better, try to get into cardiac theatre and have a go!)

Are we drowning our patients?

I’ve been a bit busy across at sydneyhems.com for a while, but while I’m writing my next little bit, here’s a link to the good people at jems.com who have published something I wrote a while back on fluid resuscitation.  JEMS (Journal of Emergency Medical Services) is an American EMS journal that mainly publishes reviews of topics aimed at paramedics.

It’s a bit of a rant (imagine that) and fairly lighthearted.  More thought provoking and hypothesis generating than practice altering.  But I think that the way we resuscitate people with fluid will change significantly over the next few years. Anyway, have a read and see what you think?