Confessions of a High-Flow Addict

Aaron B. Holley, MD


June 11, 2015

High-Flow Nasal Cannula: The Pros

Hello. My name is Aaron and I'm addicted to placing high-flow oxygen via nasal cannula (HFNC) on all of my ICU patients. They seem to like it. They can still interact with their families, they can still eat, and they don't get abrasions on their skin. The full face masks we synch onto patients when delivering noninvasive positive pressure ventilation (NIPPV) leave horrible marks. I'm not sure how well high-flow works, but I've read a few papers on the proposed physiology.[1] It's really cool.

Truth be told, I think the respiratory therapists are more addicted to HFNC than I am. Point being, I'm seeing it used increasingly more often. I really have no issue with this, mainly for the reasons I joked about above. However, we in the medical world love to adopt fancy new devices, especially when they advertise "superior humidification systems" and novel delivery methods. Unfortunately, it's usually not until we've used a new technology on patients that we get around to asking how clinically effective it really is.

I blogged about HFNC last fall after the American Journal of Respiratory and Critical Care Medicine published a paper on using it post-extubation.[2] A new trial published in the New England Journal of Medicine (NEJM) expands on what we think we know about HFNC.[3]

The authors randomly assigned three groups with hypoxemic (PaO2/FiO2 < 300 mm Hg), non-hypercapnic (PaCO2 < 45 mm Hg) respiratory failure to HFNC, NIPPV, or standard oxygen delivered through a face mask. Although they weren't able to show a statistically significant benefit to HFNC for their primary endpoint—proportion of patients intubated at day 28—HFNC was superior for ventilator-free days (at day 28) and mortality (at day 90) compared with the other two groups. A post-hoc subgroup analysis of patients with a PaO2/FiO2 < 200 mm Hg also showed benefit for HFNC.

Lingering Questions

The accompanying editorial makes the point that the investigators didn't find significance for their primary outcome, which is true.[4] The editorial author also speculates that the mortality benefit needs explanation. Could it have been due to the high tidal volumes (9.2 ± 3.0 cc) in the NIPPV group? This is an interesting question because we really don't know much about NIPPV causing lung injury. Is 9 cc/kg delivered noninvasively just as dangerous as high volumes delivered through an endotracheal tube? The short answer is that we don't know. Answering this question is complicated by several factors: (1) the estimates for tidal volume seen by the alveoli when using NIPPV vary by leak, mask interface, and device, and the authors give us very little information on these variables; (2) They do state that NIPPV was delivered via a Fisher-Paykel face mask hooked up to an ICU ventilator. We do know that leaks and asynchrony increase with a system that is not specifically designed for NIPPV.[5,6]

All of this matters because asynchrony can be a major issue with NIPPV. What if the improvement seen in VE/PaCO2 with HFNC was due to better synchrony and less anxiety and discomfort, and not a reduction in dead space as the authors suggest? Back to HFNC, though. Despite the issues with NIPPV delivery in the NEJM study, there was still an improvement with HFNC compared to oxygen delivered via face mask.

In conclusion, it's good to see clinical data on HFNC. The study in NEJM is an important addition to the literature and certainly increases my comfort level. It also matches up with my anecdotal clinical experience. Last, it highlights the fact that as NIPPV is used increasingly more often, trials that report on NIPPV outcomes need to be very specific on interfaces and devices if we're going to gain meaningful information about their efficacy. So, my name is Aaron, and I continue to be an HFNC addict.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.