IV vs. PO steroids for asthma

Treating an asthma exacerbation with steroids–and the sooner the better–is undisputed. And it makes intuitive sense that higher doses of steroids work better, and that IV steroids work better than PO steroids, right?

Well, the answer’s not so simple.

First of all, I see patients >18 years of age (usually). A lot of the literature on asthma was studied in the pediatric population, so while the results might be generalizable, we don’t know for sure.

Is there a maximum dose for steroids?

No, although this Cochrane review suggests that upwards of 100 mg prednisone daily, you’re probably not getting much benefit. This separate Cochrane review concludes most of the evidence is pretty low-quality, so it’s hard to make a strong recommendation. Side effects such as hyperglycemia and agitation/delirium are dose dependent.

Are IV steroids better than PO steroids?

I’ve heard some people suggest that if someone has “failed” PO steroids, then they should get a trial of IV steroids for at least 48 hours. On the one hand, this seems reasonable, because you want to avoid intubating the patient for worsening respiratory distress. On the other hand, there’s no evidence to support the assertion that IV steroids are better. See this pediatric study or this randomized adult study. This review makes the argument that it can be hard to ascertain the true effect of IV steroids in retrospective studies because (1) there is often no true “control” or “placebo” group (2) more severe asthma attacks are more likely to get IV steroids upfront, so you have a “confounding by severity” problem.

IV steroids make sense as a one-time dose for initial therapy in the ED, or for patients who are too obtunded or can’t take PO for some other reason. Then again, I’ve had a couple of cases where we used IV steroids 4-5 days into someone’s asthma treatment, and they got better. There’s the literature, and then there’s practice.

To rule out or not rule out: the TB rule-out

The US is very, very lucky to only have approximately 9,000 cases of TB yearly. But TB can be a “great masquerader” with such serious public health consequences that we have to be constantly vigilant. After all, about ONE-THIRD OF THE WORLD POPULATION is infected with Mycobacterium tuberculosis.

Definitely tuberculosis.

Who needs to be ruled out for TB?

Someone who has clinical +/- epidemiological features of pulmonary TB. Only active pulmonary TB is infectious, so extra-pulmonary TB does not require quarantine. However, extra-pulmonary TB can be infectious if you’re directly biopsying or handling samples from infected lesions.

If the patient in question doesn’t really fit the above features, or you have a strong suspicion for an alternative diagnosis (plain old bacterial pneumonia, pulmonary infarct, DAH, vasculitis, pulmonary sarcoidosis, etc.) then you’re not required to do a rule out. But you better be DAMN SURE because if it turns out the patient does have TB, you could be putting a lot of people at risk of infection.

Who should be put in a negative pressure room “just in case?”

This is a 1999 multicenter, prospective study of a decision instrument to predict who would be at very low risk of TB and would not require isolation. Not surprisingly, people at low risk included those who were not homeless, did not have recent exposure to TB, were not incarcerated, and did not have suspicious x-ray findings. It stands to reason that you might want to be more cautious with people who do have the risk factors outlined above.

Does everyone getting ruled out for TB require hospitalization?

Theoretically, someone can be ruled out for TB at home if they stay away from other family members and don’t go out into the community. This EM:RAP article has a protocol for setting up outpatient TB rule-out. But because YOU CAN’T TRUST ANYONE (and perhaps because the ED doesn’t want to be held liable if something at home goes wrong), I’ve had even very low-public health risk cases admitted for rule-out. If a patient refuses to be admitted, you cannot legally hold them against their will just for TB rule-out, even if they are a public health hazard. Try to convince them of the merits of being admitted, and if they still refuse and leave AMA, call your local health department, who do have the authority to search for elopers and strong-arm them into getting admitted.

What are the possible outcomes for someone exposed to TB?

Can you rule out someone with a PPD or a quantiferon gold?

It would be so much easier than AFB x3 q8H +/- direct sampling! Alas. PPD and quant golds do not distinguish between latent and active TB. A negative PPD or quant gold could be reassuring, but a positive result doesn’t change management. Plus, up to 25% of people with active pulmonary TB will have a negative test result for these two tests.

What is an oxymizer?

What does this thing even look like?

Glad you asked:

Sometimes the oxymizer is referred to as a “nasal moustache” because the reservoir is positioned under the nose, as in this instructive 1-minute video from RT Clinic.

What’s the conversion between oxymizer and nasal cannula?

Roughly 1:2––For example, 3 L by oxymizer delivers approximately the same amount of FiO2 that 6 L NC does. Some oxymizer devices can provide oxygen at a 1:4 level (so 1 L oxymizer=4 L NC) and deliver up to 20 L NC! Pretty cool.

What are the advantages of using an oxymizer?

For patients at home, oxymizers make portable oxygen tanks last longer (because the flow rate is lower). How much longer? It depends on the individual patient’s O2 needs and activity level. The lower flow rate also decreases nasal dryness and the discomfort of oxygen blasting into the nares all the time. From a pragmatic standpoint, most clinicians would be nervous about keeping patients at 7-8 L NC at all times (especially on a general medical floor), but an oxymizer can be used to keep patients with higher O2 requirements who are otherwise stable on the floor. For most patients with long-term, high-flow oxygen needs, oxymizers just make more sense.

If it’s “better” than nasal cannula, why don’t we use it all the time? Are there any downsides to the oxymizer?

Cost-wise, oxymizers can be expensive. Not all insurance companies cover oxymizers. The retail price of a single device is (as of this post) about $400 and the manufacturers recommend replacing the device every 3-4 weeks to ensure the membrane continues to work properly. In addition, some patients find the heavier tubing uncomfortable to wear. As far as I know, that’s pretty much it.

Disclaimer: I don’t get paid by Big Respiratory to write this. I have noticed more and more patients wearing oxymizers in the past couple of years who have been able to stay out of stepdown units and ICUs just because they “need more than 6 L NC, which is too much for the floor.”

Interpreting hypoxia on an ABG: PaO2 and SaO2

Let’s say you have a 57 year old patient breathing comfortably on room air, and when you walk in the next morning, he’s suddenly on 6 L O2 by nasal cannula. He doesn’t look like he’s in respiratory distress, but you decide to investigate by getting an ABG.

Result: 7.27/47/68

He is satting 93% on 6 L NC. Is that good? Is that bad? How does his O2 sat compare to the PaO2 on his ABG?

Normal PaO2=80-100 mm Hg. PaO2 is affected by age (tends to be lower) and altitude (tends to be lower).

PaO2 and O2 sat can be related through the oxygen-hemoglobin dissocation curve! See this table for PaO2 to O2 sat conversion. Remember that from first year of med school?

Straight from Wikipedia

As you can see, under normal conditions, an O2 sat of 90% correlates with a PaO2 of 60 mm Hg  (bonus points if this makes you realize an O2 sat  of 90% is not totally normal, although for sick, hospitalized patients,  it is acceptable). This curve is useful because it shows that giving supplemental O2 is most useful when someone has an O2 sat <90%. The curve also shows that O2 sat falls slower than the PaO2–a change in PaO2 from 96 to 70 may only show up as a change in O2 sat from 97% to 92%.

FiO2 can also affect an ABG reading. The PaO2 on your ABG should equal FiO2 x 500. If it doesn’t, there’s probably an A-a gradient. The PaO2/FiO2 ratio (or P/F ratio) is useful for categorizing hypoxia as potentially severe (when applied to ARDS).

So what about the patient above? His PaO2 of 68 mm Hg correlated perfectly with an O2 sat of 93%. However, he was also on 6 L NC, and the FiO2 was 40%. This implied that there was a significant A-a gradient

Random notes below:

Why are air bubbles bad? The PO2 of room air is 150 mm Hg, which means any air bubbles trapped in the ABG sample will shift the oxygen value towards 150 mm Hg.

When does the ABG have to be put on ice? If it can’t be processed in 15 minutes. (Residual blood cells will continue to use oxygen and make the PaO2 seem lower than it really is.) An ABG on ice can still be analyzed for up to an hour after collection.

If I get a value like PaO2=213, what does that mean?! At least you know the patient’s not hypoxic? PO2 is measured directly via electrode. The electrode is calibrated for values between 0-140. Therefore values >150 are of unclear accuracy. Remember that FiO2 affects the value as well.


UC Denver handout

Clinical Methods (E.P. Trulock III)

American Nurse Today

What is pulmonary toilet?

AKA “pulmonary hygiene.” Pulmonary toilet is advocated by many people, and does sound like a good idea in theory. However, the literature on whether pulmonary toilet actually improves outcomes for various patient populations is very mixed. In general, the patients who seem to benefit most are the cystic fibrosis and critically ill/intubated populations.

The purpose is to THIN and LOOSEN secretions. Having an awake, alert patient who can cough on their own is the best kind of pulmonary toilet.

What are the specific components of pulmonary toilet?

  • Bronchodilator (albuterol)
  • Mucolytic (such as NAC or Mucomyst): NAC in particular has become less popular because of lack of demonstrated utility (a good example is    this meta-analysis on cystic fibrosis). I still see it used in the ICU, though.
  • Hypertonic (7%) saline: may cause bronchospasm, coughing
  • Chest PT
    • Vibrating chest vest
    • Percussion (clapping patient on the chest)
    • Postural drainage
  • PEP device and flutter valve (such as Accapella)  
    •       NB: the Accapella =/= incentive spirometer
  • Suctioning
  • Bronchoscopy: using a bronchoscope to “wash out” secretions for atelectasis. This is only helpful in patients who are critically ill/intubated who cannot do any other kind of secretion clearance on their own. My personal observation is that when BAL is used for pulmonary toilet, it turns into a problematic cycle of provoking even more secretions and the need for repeated bronchs. A bronch is also not a completely benign procedure and may be associated with barotrauma, temporarily worsened oxygenation, etc.

When is pulmonary toilet useful? This clinical review is pretty negative. It states that chlorhexidine oral rinses are the only intervention that reduces rates of nosocomial pneumonia, and that other common practices like mucolytics and chest PT are not associated with improved outcomes and may in fact cause harm like bronchospasm and mechanical trauma, respectively. (A lot of evidence comes from pre-2000’s papers/guidelines, though). Let’s look at some common situations:

  • COPD: This meta-analysis reports that daily oral NAC was associated with fewer COPD exacerbations in patients who had a spirometric diagnosis of COPD, but also in patients who did not have a spirometric diagnosis of COPD. However, this joint paper from ACP-ACCP states that multiple trials have not found benefit for mucolytics or chest PT in COPD exacerbations.
  • Cystic fibrosis: Yes to chest PT. There is also good evidence that dornase alfa (DNase) and hypertonic saline have small effects.
  • Atelectasis: for acute atelectasis, if patients cannot cough on their own, chest PT is helpful. (The same review talks about how “bronch for airway clearance” is only indicated in selected cases of atelectasis that are multilobar or severe.)
  • Post-surgical patients :There is very limited objective evidence that pulmonary toilet decreases pulmonary complications (pneumonia, atelectasis, edema, etc.) in the postoperative period. This Cochrane review shows that in patients who underwent  upper abdominal surgery, incentive spirometry didn’t make a difference. Maybe this will make us feel better about all those unused incentive spirometers sitting by patients’ bedsides.        

Does supplemental oxygen cause loss of respiratory drive in COPD patients?

It is common to set the O2 saturation goal for hospitalized patients with COPD exacerbations at 88-92%, and patients without COPD to 94-98%. This is in accordance with British Thoracic Society guidelines. The O2 sat goals are lower for patients with COPD because of the risk of hypercapnenic respiratory failure.

I’m not questioning the O2 sat goals. What I do want to discuss is one of the oft-cited mechanisms for this respiratory failure: that a higher O2 sat will depress the respiratory drive in these patients. Is this true?

Not really. The explanation is very well stated in LITFL.  Patients with COPD suffer from parenchymal damage that increases V/Q mismatch. To compensate for this, those smart pulmonary arterioles vasoconstrict to deliver O2 preferentially to the parts of the lungs with the least damage. Giving someone supplemental O2 in this scenario causes the pulmonary arterioles to dilate, causing increased blood flow to the damaged parts of the lungs, too, which increases V/Q mismatch again. See this article for helpful diagrams and more detailed explanation. In addition, due to the Haldane effect (=introducing more O2 will cause CO2 to dissociate more readily from hemoglobin), adding supplemental O2 theoretically causes COPDers’ CO2 levels to increase. I’m not sure that anyone has actually demonstrated this experimentally, but it makes sense.

This review from Respiratory Care on the use of supplemental O2 cites a 2010 study of about 400 patients with COPD exacerbations. They were randomized into a 88-92% O2 sat group and a non-titrated group (so presumed >94%); each group could get as much supplemental O2 as needed to reach those goals. The titrated group had a 58% reduction in hypercapnea and respiratory failure compared to the non-titrated group.

In conclusion: patients with COPD exacerbations should have a lower target O2 sat, but the justification for this is not that it will affect their respiratory drive–instead, think about V/Q mismatch and CO2 dissociation. 

What if a patient has COPD but does not have an exacerbation while they’re in the hospital? I couldn’t find a clear answer but would assume that since the same physiologic properties above are in play, it makes sense to continue the lower O2 sat goal.

Tangent: oxygen has traditionally been recommended as standard management of MI. However, this 2017 Swedish RCT showed that putting patients who had an O2 sat >90% on supplemental O2 did not change outcomes. The British Medical Journal shows that maintaining very high O2 sats does not improve mortality and has a great infographic on why maintaining an O2 sat between 90-94% is appropriate for most patients.