Bag Valve Mask (BVM)

The Idea

The COVID-19 global epidemic has brought to light the issue of lack of ventilators and the resulting dire situations that can result when hospitals are overrun. In response to this issue, groups around the world have joined the effort of designing low-cost, easily manufacturable, basic ventilators. One of the most popular concepts in the crowdsourcing community is the bag valve mask (BVM) squeezing ventilator. A bag valve mask is a hand held device used to provide positive pressure ventilation, often in emergency situations. Since these devices are already designed to provide air volume and pressure to patients, adapting them to be squeezed automatically seems like an obvious way to create a basic ventilator.

The COSMIC team decided to design and build a BVM ventilator in order to determine the viability of such a device for use in a worst-case COVID-19 scenario. We wanted to establish the know-how and figure out supply chain issues should the need to build 50+ of these simple ventilators arise. We would like to acknowledge that there are a myriad of other groups working on BVM ventilator designs, including the MIT eVent and RICE ApolloBVM. These teams, and many others, have made incredible advancements in the understanding of the feasibility and viability of BVM ventilators.

Bag Valve Masks - How Much Can They Take?

We determined early on that a key consideration in building a BVM ventilator was the reliability of the BVM itself. Could a BVM be squeezed continuously for hours, days, or even weeks on end without failing and leaving a patient with no breath? To answer this question, we designed a test jig to compress an Ambu SPUR II BVM continuously. The bag was compressed using a wooden block attached to a piston and each compression resulted in ~450mL of air escaping the bag.

The BVM was able withstand >570,000 compressions without failure. At the end of the test, the BVM showed minor scuffing and was slower to reinflate (taking ~4 seconds to fully reinflate). This is a very promising result, given that even at the relatively high breathing rate of 30 bpm, a BVM ventilator could in theory sustain a patient for over a week without the BVM itself failing.

Calculation from our test - BVM compressed 36 times per minute for 11 consecutive days:

Calculation for a patient breathing at 30 bpm for 7 consecutive days:

The Design

Design Process

Our team went through many design iterations. Because there were so many other open source designs for BVM ventilators, we made sure to regularly check up on the designs of other teams. In the spirit of creating an extremely low-cost, simple ventilator, our first prototype consisted of a 3D printed hinge attached to a stepper motor by a string to compress the BVM.

The beauty of this design was that the total cost of components was less than $50 and it was extremely easy to build. Initial tests of the design showed that it could compress the BVM adequately, but there were concerns about the BVM’s ability to reinflate itself against the force of the hinge.

It was around this time that we came across RICE University’s ApolloBVM design. The design was fully open source, with all the files necessary to produce it available on their website. Furthermore, at the time it was ranked 3rd on the list of more than 100 open source ventilator designs. We arrived at the conclusion that due to the RICE team being so much further ahead than us, and in the spirit of open source and adopting the best solution at any cost, we abandoned our own design and focused our efforts on building the ApolloBVM.

Building the ApolloBVM

Building the ApolloBVM was not as simple as following the instructions provided on RICE’s website. We had to make their design work within the context of our local supply chain in Vancouver, BC, as opposed to what they had access to in Houston, Texas. Some of the components in their design were only available online through international distributors that would take months to arrive. This was not suitable for our needs, especially given that we wanted to be able to rapidly scale up and build 50+ devices in a short time span. Therefore we made several key adaptations to the RICE design so that we could source all components locally:

After making these adaptations and ensuring that all the components still fit together, we completed our prototype of the RICE ApolloBVM.

Where we think the RICE ApolloBVM design could be improved is in the addition of patient triggered respiration. If the controls system were upgraded, it could allow for detection of when the patient is trying to inspire or expire, and then trigger the squeezing/release of the BVM accordingly. This improvement would make the ventilator more suitable for patients who aren’t unconscious or who don’t have access to heavy sedatives.

Archiving the Project

Now that we have successfully built a BVM ventilator in Vancouver, we have decided to archive the project. We no longer see the need for BVM ventilators in British Columbia as public health officials are not currently forecasting a shortage of ventilators. However, we can now say that we have established the know-how and figured out the local supply chain to build these devices should the need ever arise in the future.

Thank you to all of our sponsors, donors, and of course, volunteers who have contributed to the success of our project.