Common household appliance points to better path for ventilating multiple patients with single machine
With further testing, gas appliance pressure regulators could provide each patient with their own personalized input pressures
July 7, 2020
COVID-19 is once again filling up intensive care units with patients in respiratory distress. While many countries now have better availability of ventilators relative to just two months prior, the prospect of ventilating more than one patient with a single ventilator remains a possibility in resource-constrained settings.
Now, a team from the Bay Area has demonstrated a safer way to ventilate two patients from a single ventilator, using gas appliance pressure regulators, which are inexpensive and widely available. The method applies commonly available pressure regulators connected to each of two inspiratory limbs, supplying different pressures to each patient from a common pressure source delivered by a ventilator. Separate gas flow analyzers for each patient carefully monitor tidal volume in realtime. Using synthetic test lungs, the team demonstrated the potential for this technique to deliver separate inspiratory pressures and tidal volumes to each of two patients from a single ventilator. The full open-access preprint is uploaded here, and a video demonstration by Dr. Rishel is now available on YouTube.
The team of engineers and physicians draws its strength from disparate fields including radiology, anesthesiology, electrical engineering, and mechanical engineering. Lead author Ram Srinivasan MD PhD is founder of Orbit CME, an educational technology that simplifies CME compliance for physicians across the country, including anesthesiologists, internal medicine doctors, radiologists, and other specialties. Chris A. Rishel MD PhD and Barret J. Larson MD are anesthesiology faculty at Stanford and co-founders of WikiAnesthesia, with backgrounds in systems neuroscience, outcomes research, and medical device development. Larson also runs the Stanford Anesthesia Innovation Laboratory, which integrates entrepreneurship into the Anesthesiology curriculum at Stanford, and previously founded Leaf Healthcare. Juhwan Yoo PhD is an electrical engineer with expertise in mixed-circuit design and biomedical research. Ned M. Shelton MS is a mechanical engineer and roboticist. The mathematical connection between respiratory mechanics and resistor-capacitor circuits, established in the mid-20th century, allowed the team to reason quickly about various solutions to split ventilation before converging on their proposed solution.
While FDA-approved ventilator tubing adaptors have been manufactured for this “split ventilator” scenario, there is no good way to supply each patient with different inspiratory pressures from a common ventilator with existing configurations. Some papers in the prominent anesthesiology journal Anesthesiology and elsewhere have suggested flow regulators, essentially pinching tubing to variable amounts to deliver different pressures to each patient, but the authors acknowledge this is a brittle and challenging approach. The best-in-class protocol has chosen to avoid flow regulators, instead requiring carefully matching patients on the same ventilator to avoid lung damage in a disease that already requires carefully adjusted inspiratory pressures.
Following weeks of experimenting with various approaches including circuit analysis explored by Srinivasan and Yoo, roboticist Ned Shelton suggested pressure regulators as a robust approach to ventilator splitting that would allow providers to dial in different inspiratory pressures for each patient, sourced from a single ventilator operating at a single inspiratory pressure. Unfortunately, most commercially available pressure regulators require a very high input pressure, typically several orders of magnitude higher than the ~50 cmH2O that ventilators comfortably deliver.
The team got its break when Srinivasan’s natural gas went out during a late night cooking session at his home in Berkeley. The following morning, the PG&E service person came to inspect the gas line, and the two got to talking about gas pressure regulators. The PG&E service person noted that gas to homes is supplied at approximately 50 cmH2O, and the gas appliance pressure regulator converts that pressure into about 10–15 cmH2O for gas stoves. Serendipitously, this performance range is similar to what the team was looking for in pressure regulators that could solve the split ventilator application. The service person suggested that Srinivasan contact Maxitrol, a major gas appliance pressure regulator manufacturer.
Within a week, engineers at Maxitrol had sourced the gas appliance pressure regulator with an internal spring calibrated appropriately to provide output pressures in the range of 10–40 cmH2O, with input pressures of about 50 cmH2O, exactly what the team needed to support split ventilators. Using his skills as a mechanical engineer, Shelton rapidly constructed and 3D printed the 3/4" NTP to 20 mm adaptor needed to connect the Maxitrol regulator to ventilator tubing. Files required to 3D print this adaptor are included in the article preprint.
With the Maxitrol pressure regulators and adaptors in hand, Srinivasan and his anesthesiologist-engineer collaborators Chris Rishel and Barrett Larson at Stanford Hospital set out to show that two test lungs with identical compliance could receive different controlled tidal volumes in a modified split-ventilator configuration. In what amounted to over 24 hours of late-night experimentation over multiple days, the team refined their setup and experimental design, collecting the measurements included in this new preprint that suggests that gas appliance pressure regulators may be able to achieve the steady, independent control of tidal volume needed for safe split ventilation in COVID-19 patients.
These results are promising, but the research team is quick to emphasize the need for extensive safety and usability testing before employing this method on patients. Although gas appliance pressure regulators are used in the critical application of handling natural gas, the method could benefit from human factors engineering, such as measures to enforce proper regulator orientation. Stress testing with prolonged duty cycles would further ensure safety. A better understanding of training requirements for providers involved with the devices would be needed. Custom mechanical controls would be helpful; the team adjusted output pressures during this demonstration with a Philips-head screwdriver. Finally, gas flow analyzers combined with an easy-to-interpret display and carefully designed alarm system would be necessary to monitor tidal volumes and other critical values on an ongoing basis for each patient separately. The cost of these analyzers would be a major consideration for international use.
For a link to the full open-access preprint, click here. For the video demonstration, click here. You can also learn more about the team’s adjacent efforts, including streamlining CME for anesthesiologists and other doctors through Orbit CME, aggregating open-access anesthesiology knowledge through WikiAnesthesia, or fostering entrepreneurship in anesthesiology at the Stanford Anesthesia Innovation Laboratory.