Ventilator Waveform Anatomy - The Basics You Gotta Know

If your "ventilator resume" looks like mine, you started with something like an AutoVent, and progressively upgraded to the critical care ventilator you are using now. Having not been through formal respiratory therapist training, understanding ventilator waveforms was not something I was taught initially. Thankfully, they are not nearly as difficult as I imagined, and prove to be incredibly helpful in tailoring your vent settings to the patient.
There are two core waveforms you have to understand, the pressure waveform and flow waveform.  In the illustration below, notice how the inspiratory segments of the pressure and flow waveforms line up, as do the expiratory segments. This is a critical piece if information that we will use at the bottom of the page to learn how to optimize your I:E ratio.
The pressure waveform is made up of basic settings and monitored parameters you already know: PIP or Ppeak, PEEP, pRamp, and I:E ratio. Look at the yellow pressure waveform below, and how it matches up with the monitored and set parameters.
     - PEEP plus Pcontrol = PIP (or Ppeak)
     - In pressure control modes, notice how Driving Pressure is the same thing as Pcontrol
     - A higher pRamp (in this case 200ms, vs 50 ms in the first illustration) causes more of a slope in the inspiratory phase
Next we turn our attention to the red and blue lines above the pressure waveform. The red line is set by the High Pressure alarm. The blue line is always 10 cmH2O below the red line. The ventilator allows the yellow pressure waveform to climb as high as the blue line, but no further as a safety measure. When your settings call for the peak airway pressures to exceed the blue line, this triggers the "Pressure limitation" alarm you see in the image below. This can be solved in one of 3 main ways:
     - Optimize your patient (loosen seatbelts, sit upright, suction, analgese/sedate)
     - Change your settings to stay within the blue line (decrease Vt or Pcontrol, increase RR to meet Vte goals)
     - Adjust the high pressure alarm to raise the red and blue lines. This makes the high pressure alarm different from the rest of the alarms: it functions more like a setting than an indicator.
The final concept to explore in the pressure waveform is the concept of Mean Airway Pressure, or pMean. If you fill in the area below the pressure waveform, you can see a great representation of the airway pressures during inhalation and exhalation. Averaging this gives you the pMean, represented by the dashed red line overlaid on the diagram below. You can find this value on page 2 of the monitoring tab.  
Mean airway pressure and FiO2 are the two main ways we optimize oxygenation. The illustration below shows how increasing the inspiratory time significantly effects the pMean. Using a 1:1 ratio or even slight I:E inversion should not cause hypercapnia as long as the flow waveform is returning to baseline (see next section). Slowly increasing the inspiratory time is a great way to optimize oxygenation when PEEP and FiO2 has already been dialed in to ideal (or max) settings. 
As we start looking closer at the flow waveform, there is one principle that stands out as most important. The portion of the waveform above the white baseline shows inspiratory flow, and the portion below the line represents exhalation. AutoPEEP or air trapping is shown in the diagram below, where the expiratory segment doesn't return to baseline before the next breath, indicating all the air inhaled wasn't able to be exhaled before the next breath was triggered. This should be corrected immediately by lowering the RR or TI. 
There are several other characteristics of the flow waveform that are helpful to understand.  Ventilators in traditional volume modes have a square inhalation box, due to a fixed flow rate. Pressure modes (and volume-target modes like PRVC or (s)CMV+ ) use a decelerating flow waveform like what is shown below. This starts with a fairly high initial flow rate and then tapers off, much more like we naturally take breaths. For that reason the decelerating waveform has been described as more "physiologic", meaning more comfortable for patients. 
When you compare the two flow waveforms below, you see some notable differences. First off, the obstructive example has much lower initial flow rates (15 liters per minute, vs 75 in the "normal" example). This is caused by a restriction like bronchospasm, and shows less volume being put into the patient. This is where patience pays off: airway pressures are a product of flow and resistance. Decreasing the flow rate and allowing the turbine to run longer (by lengthening TI) allows you to get more volume past the obstruction without increasing pressures.
Another difference is how the obstructive flow waveform does not decelerate back to baseline, again indicative of less volume being able to be put in. The makes the inspiratory portion of the flow waveform look more like a box instead of a triangle. It's important to recognize this, and realize that increasing the TI will directly increase the amount of tidal volume able to be pushed in.
At the very top of this post, we discussed how the inspiratory time (TI) and expiratory time (TE) match up on the pressure and flow waveforms. You can see on the normal example above how the flow stops before the TI and TE has completed, leaving the purple line on the baseline before the next phase is triggered. In the obstructive example above, the exhalation returns to baseline and stays at the baseline for a short time before the next breath begins. In normal patients this is not a problem, but in a patient with hypercapnia due to obstructive pulmonary disease, this space can be used to optimize minute volume. 
I like to think of that time at baseline right before inhalation as "time on the table" just waiting to be used. You can increase the RR or TI to take advantage of this, or even increase the PIP if you have room to to that safely (which will cause more air to be inhaled, thus more to exhale, thus taking longer and extending the exhalation phase). 
Looking at the obstructive example above, I'd recommend starting with increasing TI. The squarish inhalation phase that is NOT returning to baseline is a clue that increasing TI will directly increase tidal volume. Increasing Vt as opposed to RR gives you a little more "bang for your buck" since it directly increases alveolar minute volumes, and increasing RR increases both alveolar and dead space minute ventilation. 
You can also "use this baseline" in ARDS patients, as it allows you to increase the I:E ratio to maximize pMean and oxygenation, without worsening hypercapnia. As long as the patient is still able to exhale their entire breath, then it's safe to slowly being inverting the I:E without worry of causing hypercapnia. 
After spending the past four years relying on ventilator waveforms in transport, taking them away feels as egregious as taking away my ETCO2 waveform. Vent waveforms increase patient safety, give us clues in optimizing settings, and constantly provide us with the information we need to troubleshoot alarms. Lets summarize a few of the high points:
     1) Use the pressure waveform to visualize how to improve pMean (using PEEP, TI, RR, and/or PIP).
     2) Always watch to make sure the expiratory flow is returning to baseline to avoid AutoPEEP/air trapping.
     3) When ventilating 'sick lungs' (obstructive or ARDS patients), look to see if you are "leaving time on the table" and use up the baseline with RR or TI to increase pMean or Vte without causing hypercapnia or air trapping.

Further Thoughts: Septic Afib w/ Precision Rate Control


You walk into the ED to find your septic patient has a MAP of 50 and heart rate bouncing between 110 and 140. As you walk closer to the hospital monitor, your suspicion of atrial fibrillation with rapid ventricular (RVR) response is confirmed. At some point in your career you may have asked yourself: "is the rate causing the hypotension, or is the rate compensating for the hypotension?"

I believe most of us would chalk this tachycardia up to compensation, evaluate how much fluid the patient has received, and have a low threshold to pull the trigger on levophed. However, if this patient is coming out of an ICU (or extended resuscitation in the ED), already received an adequate amount of fluid, and currently on pressor therapy - one must assume that the tachycardia may have made a shift from a compensatory mechanism to now a response from circulating exogenous catecholamines. 

We know that preload is manipulated by multiple factors, but heart rate is an independent variable. I think of it like the gigantic buckets at waterparks, they fill up over several minutes and then dump water on all the kids standing underneath. If someone were to manually dump the bucket every few seconds, the amount of water released would be underwhelming to say the least.

A little less than two years ago I did a podcast with Dr. Josh Farkas from Pulmcrit on atrial fibrillation with rapid venticular response (RVR) in the septic patient. As a primer to the podcast, Farkas did an amazing blog which can essentially be summarized with this quote.

The average stroke volume of the heart is 50-100 ml's. If we multiply this by our heart rate we will get our cardiac output. Typically if stroke volume goes down, heart rate will increase to maintain cardiac output. However, just because we decrease rate, does not automatically mean we will increase stroke volume. This can be seen in the two calculations below for a patient whose heart rate is decreased from 120 to 80.

Example A : the heart rate decreased to only 80 and without an increase in stroke volume, the cardiac output dropped by two liters.

Example B: the heart rate decreased to 80 and a generous 25 ml's of stroke volume was added. This is an ideal situation, but not assured.

The ideal agent to reduce heart rate after subjective adequate resuscitation would be one in which the heart rate could be tightly controlled and would have a short half-life if stroke volume did not respond to increased diastolic filling.

A study in 2013 evaluated the use of esmolol for this exact scenario. This was a single center, phase II open label RCT, and outcomes were focused on heart rate control (between 80-94), pressor requirements, and fluid requirements. 

The results were promising and showed excellent precision in regard to rate control, decreased pressor requirements, and decreased fluid requirements within the treatment arm. When 28 day, ICU, and hospital mortality was evaluated as a secondary outcome - there was some signal of a mortality benefit as well. 

Graphics have been re-created and simplified for the blog. It is important to download this article from JAMA and evaluate the evidence for yourself. The link is included in the references.

While larger studies are needed with emphasized patient centered primary outcomes, I anecdotally love how tightly I can dial in heart rate on aortic anuerysm patients who I transport on esmolol. This study was an excellent example of the precision and safety margin that can be obtained with esmolol. The graph below shows the average heart rate for both the treatment and control arms. 

It makes sense to me to use an agent that is titrates as an infusion and has a short half-life. I have seen diltiazem reduce heart rate by barely 10 bpm, and other times reduce by 50 bpm. When diastolic filling is a dependent variable, reducing an independent variable such as heart rate should be carefully calculated. 

In this study they did not begin to treat rate untill 24 hours after hemodynamic resuscitation. This probably is wise and for the acute setting, you will be safer to assume the tachycardia is due to compensation in the presense (or suspicion) of sepsis. 

I do believe there are a subset of patients that have increased their heart rate above compensatory therapy. This is especially true in atrial fibrillation, where cardiac output is already diminished due to loss of atrial kick. In this setting, a shared decision-making model with the referring physician is wise prior to transfer.


Morelli A, Ertmer C, Westphal M, et al. Effect of Heart Rate Control With Esmolol on Hemodynamic and Clinical Outcomes in Patients With Septic ShockA Randomized Clinical TrialJAMA. 2013;310(16):1683–1691. doi:10.1001/jama.2013.278477

Flipping The Podcast: Approach to shocky patient in AF w/ RVR - Dr. Josh Farkas

FOAMfrat Podcast 54 - Approaching The Shocky Patient In Afib w/ RVR.

© 2020 FOAMfrat LLC. All Rights Reserved.