Thoughts on Specifications
MIT has published a helpful list of specifications for emergency ventilators.
We agree with most of these specifications but wanted to provide a little elaboration on the rationale behind them, for engineers who may be struggling to meet one spec or another. And also to help interpret whether a ventilator that misses on one might still have a chance of being valuable in an emergency, and for roughly how large a subset of patients.
First, it is worth noting that in in the polio epidemic in 1952, lives were saved by people manually squeezing bags like the Ambu bag for hours on end. These folks did not have lots of sensors, did not have closed loop electronic pressure or volume control, and did not have many other features found in current clinical ventilators. It is helpful to remember this when considering which features are “nice to have” and which are truly
“must have”.
Next, a big question our team had as we went through this process is whether the patient must be paralyzed during ventilation via chemicals similar to those used in surgery. The answer from discussions with pulmonologists is that the patient must at least be heavily sedated, but not necessarily chemically paralyzed. Likely, physicians would try heavy sedation first, and see how well the patient tolerates ventilation. Some will tolerate it well. Some will fight the ventilator. The latter group would require chemical paralysis. This has implications for the required sensing systems needed in a minimum useful ventilator, as discussed below.
A few other specifications from the E-Vent team and our thoughts on them are as follows:
- Patients should be under the management of a trained physician
We 100% agree!
- BPM (Breaths per Minute): Should be adjustable between 8-40 BPM
We agree with this range, and the need for adjustability. However, if you are struggling to get it with your design, 10-30 BPM or 10-35 BPM would also cover a large percentage of patients. Our system provides a knob (connected to a potentiometer) that the physician can turn to manually adjust BPM. The ability to adjust BPM is a “must have”. The exact range over which it can be adjusted is slightly negotiable.
- Tidal Volume (TV) (air volume pushed into lung each breath): 200-800 mL based on patient weight.
This is a good range, in general. If you are having trouble getting this, a clinical rule of thumb is 6ml/kg body weight (a very rough estimate – it depends on lots of factors – but helpful here for engineers). So, for example, 300-600 ml would (nominally) cover people weighing 110lbs-220lbs, which is still a sizable fraction of the population. In our system, the tidal volume is adjusted by moving the set point of the windshield wiper motor mechanism with respect to the fixed location of the Ambu bag. This adjusts the magnitude of compression of the bag, per stroke.
- I/E Ratio (inspiratory/expiration time ratio): Start around 1:2; best if adjustable between 1:1 and 1:4.
We agree with the nominal being 1:2. If you can’t quite get to 1:4, that is ok; 1:3 is reasonable. Note that if you can do it, it also is useful to be able to do up to 2:1 as well (called “inverse ratio ventilation”). This is a “nice to have” not a “must have”. Some adjustability around the nominal of 1:2 is somewhere in between, i.e. close to an a “must have”, but not quite. If your vent could only do a fixed ratio of 1:2 it would still be potentially helpful to patients. But it is highly useful to build in some adjustability around this nominal value.
- Assist Detection pressure. When a patient tries to inspire, they can cause a dip on the order of 1-5 cm H2O, with respect to PEEP pressure (not necessarily = atmospheric).
This depends on whether the patient is chemically paralyzed via anesthesia or not. If yes, this is not needed. If no, then assist detection may help. In a true emergency, in US hospitals, the patients could be paralyzed via anesthesia and likely helped with a machine that does not include assist detection.
- Airway pressure must be monitored. Maximum pressure should be limited to 40 cm H2O at any time; Plateau pressure should be limited to max 30 cm H2O
We agree with both these specifications. Note that measuring plateau pressure requires the machine to briefly hold at the point of full inhale. The plateau pressure is an important predictor of ventilator-induced lung injury and clinicians will want to monitor it closely. Our design provides the clinician with a button that when pressed and held causes the system to pause at the next full inhale, until the clinician releases the button, enabling him/her to read out plateau pressure manually (i.e. plateau pressure is the pressure at full inhale after dynamics have died down). This kind of button press-and-hold is a feature of clinical ventilators, making it familiar for clinicians.
Note that pressure sensing does not have to be electronic. We provide both a “high tech” approach with a digital pressure sensor, and a “low tech” approach with a water manometer. Either can be used alone, so don’t stress if you can’t source a good pressure sensor in the midst of the COVID-19 crisis. You can make a water manometer out of a piece of plastic tubing.
- Use of a passive mechanical blow-off valve fixed at 40 cm H2O is strongly recommended
We agree with this recommendation. The passive mechanical blow-off valve is a *really* nice to have, but possibly not strictly necessary, as long as you have a good high-pressure alarm. Note: Check whether your Ambu bag’s PEEP valve has this function. Some have the passive mechanical blow-off valve built in. Some don’t.
- Clinicians require readings of plateau pressure and PEEP
We generally agree (see earlier discussion of our plateau pressure button hold solution). Plateau pressure is an important measurement for physicians that is a predictor of mortality. Note however that in earlier times when patients were treated with hand-squeezed bags, there was no way to measure plateau pressure and some patients were still saved. Another “nice to have” is to be able to hold at the point of full exhale to make a measurement similar to plateau pressure there as well. This measurement is the actual PEEP which may be high if the patient does not have time to exhale fully during each cycle. It is helpful data for clinicians.
- PEEP of 5-15 cm H2O required; many patients need 10-15 cm H2O
PEEP (set PEEP, as opposed to actual PEEP) is absolutely required. A non-negotiable “must have”. Our
Ambu bags have a plastic PEEP valve built in that can be set by the physician to the desired pressure range. Patients with COVID-19 requiring mechanical ventilation will likely not be helped without PEEP.
- Failure conditions must permit conversion to manual clinician override, i.e. if automatic ventilation fails, the conversion to ventilation must be immediate.
We completely agree. In our system the Ambu bag can easily be picked up out of our device and hand squeezed, in this scenario.
- Ventilation on room air is better than no ventilation at all. Blending of oxygen and air gas mixture to adjust FiO2 is not important in an emergency scenario. It is certainly nice to have that ability and can easily be implemented with an oxygen/air gas blender that some hospitals already have.
We completely agree.
- Covid-19 can get aerosolized (airborne), so HEPA filtration on the patient’s exhalation is required or between the ventilator unit and the patient (at the end of the endotracheal tube) to protect clinical staff from certain infection. In-line HEPA filters can usually be purchased alongside manual resuscitator bags.
We agree – it is likely that your local hospital will have some of these that attach to the Ambu bag’s outlet. It is unlikely you will have to build your own custom filter if you are in the USA working with a local hospital.
- Heat and moisture exchanger should be used in line with the breathing circuit.
Note that this is not standard on Ambu bags. If the patient is going to be on this vent for days, dry air may cause thick secretions to build up, and cold air may cause hypothermia and bronchospasm (as in asthma). Heat and humidity tend to alleviate these. For short term emergency use, just like “ventilation on room air is better than no ventilation at all”, we would say that ventilation without heat and humidity is better than no ventilation at all. So, heat and moisture exchange is a “very nice” to have, rather than a “must have” if you are building one of these in your garage and trying to help your local doctor.
- Failure conditions must result in an alarm.
We agree with this in general. We would also add that a low pressure alarm is very close to a must have. The high
pressure alarm mentioned earlier is a must have. Low pressure can also indicate a variety of failures (leak in the system, ruptured cuff on the endotracheal tube or active attempts at breathing from the patient) that need attention by the physician, so we include both in our design.