Tuesday, March 10, 2020

Specifications for a Pandemic Ventilator for Coronaviris COVID-19

Someone mentioned the BARDA specifications in one of the comments.  I was not familiar with that so I checked it out and they are specifications for a solicitation for a ventilator to be stockpiled that could be used for a mass casualty or all-hazards event, but specifically for a pandemic influenza outbreak.
The link to the HHS solicitation is here:
I took the liberty of adapting these specifications somewhat, considering that the COVID-19 outbreak is not dangerous for children and I also eliminated references to cost and replaced them with parts availability and ability of the design to be replicated and modified as needed.

Here is my list:

  • Be able to be constructed by readily available components
  • A filtration device to reduce or prevent the spread of virus into the air
  • Only needs to accommodate adults.  Critical illness in children in this pandemic is rare and any cases that do arise can be accommodated by reallocation of existing commercial devices.
  • Intuitive user interface for ease of use so that inexperienced health care providers with limited or no respiratory support training, can safely and easily use the device.
  • Weight and size is a minor issue at this point.
  • Closed-loop inspiratory pressure control system that can operate with minimal time and attention from staff.
  • Accommodate breathing circuits and with universal connectors appropriate to patient population.
  • Adequate shielding for radio frequency interference (RFI), electromagnetic interference (EMI), and conductive interference, such as a power surge, and must be demonstrated and tested in appropriate environments.
  • Open source, well documented software that can be modified to easily use alternative processors, blowers and sensors
  • Real time integrated O2 and CO2 monitoring may be optional as these could possibly be also accomplished with stand-alone devices, but respiratory rate is not
  • Parts list and repair and troubleshooting manual
  • Easily recognizable and operational audible and visible alarm systems
  • Accommodate a heat and moisture exchanger (HME) or humidifier.

Meet an evaluation testing program(s) and other testing:
·         Operational, usability, and functional performance tests for adult patient populations.
·         Electromagnetic (EMI), radio-frequency (RFI) and conductive interference testing
·         Review of executed software validation protocols.
·         Usability performance for ease of use and set up type testing for inexperienced health care providers with limited or no respiratory support training and outlines all recommended characteristics and specifications incorporated into the design of the ventilator.
·         Full functional testing of the ventilator.
·         Functional testing in a clinical environment in all modes of operation under simulated use conditions in the simulated environment in which the device is expected to be used by inexperienced health care providers with limited or no respiratory support training.
·         Testing in a non-clinical environment for ease of set up and performance,
·         Differentiation of audible and visible alarm failure modes .

Other Resources for Specifications and testing:
·         ASTM F 1100-90 (Standard Specifications for Ventilators Intended for Use in Critical Care),
·         ASTM F 1246-91 (Standard Specification for Electrically Powered Home Care Ventilators, Part 1-Positive-Pressure Ventilators and Ventilator Circuits)

·         IEC 6061-2-12(Standard for Critical Care Ventilators).

Is there anything I should add or leave out?
.......................................................................................

In my last post I assumed that most developed countries would have similar ratios of ventilators to population.  Then I wondered if any of these  BARDA HHS ventilators have actually been deployed.  I found a recent source of ventilator numbers in the
from May 3, 2018, and here are their numbers:

865,000 US residents would be hospitalized during a moderate pandemic (as in 1957 and 1968) 
64,875 would need mechanical ventilation

9.9 million during a severe pandemic (as in 1918).       
742,500 would need mechanical ventilation

US acute care hospitals own approximately 62,000 full-feature mechanical ventilators.   
28,883 of these ventilators (46.4%) can be used to ventilate pediatric and neonatal patients. 
an additional 98,000 ventilators that are not full-featured but can still provide basic function.
Based on these numbers, the maximum number that can be potentially ventilated is around 160,000.

The US has 20.5 ICU beds with mechanical ventilation capability per 100,000 population
Canada has 8.7 ICU beds with mechanical ventilation capability per 100,000 population
Australia & New Zealand have only 5.4 ICU beds with mechanical ventilation capability per 100,000 population

In addition, the CDC Strategic National Stockpile has an estimated 8,900 ventilators as of 2010. 
Ventilators are stored and kept as managed inventory.

Various other factors constrain the capacity of the US healthcare system from providing ventilation therapy. Using mathematical models, one study found that the limiting factor during a pandemic-level crisis would be the number of respiratory therapists—maxing our ventilator therapy capacity at around 135,000—significantly lower than the estimated 742,500 needed
........................................................................................
Here is some information for Canada
Ontario has released a document Ontario’s Critical Care Ventilator Stockpile September 2019
It states that Ontario has 209 additional ventilators stockpiled, 50 of which are for neonates
This would be somewhere around 15% surge capacity

........................................................................................
If you want to know which ventilator designs to emulate when designing a pandemic ventilator in order to make it easy for staff to learn and use the device, this paper published by the CDC
gives some very useful info for stockpiling of ventilators for a pandemic.
........................................................................................

Here is a really interesting article from National Geographic
U.S.has only a fraction of the medical supplies it needs to combat coronavirus

Here are some excerpts:

“In a severe pandemic, we certainly could run out of ventilators, but a hospital could just as soon run out of respiratory therapists who normally operate these devices,” says Toner. The Bureau of Labor Statistics estimates that the U.S. employs 134,000 respiratory specialists, or approximately 20 of these technicians for every hospital in America.

The actual demand and supply for health care equipment during this outbreak will depend on myriad variables, one of which is an outbreak’s attack rate. As of this moment, that is a mystery for COVID-19.

The attack rate is what percentage of a population catches an infectious disease overall. If a hundred people live in a city, and a virus’ attack rate is 20 percent, then 20 citizens would be expected to get sick. Both the CDC papers and Toner’s models rely on attack rates ranging from 20 to 30 percent, a standard estimate for severe pandemics. 

27 comments:

  1. Very interesting, thank you for sharing this work. This may be a life saver in case the virus spread even more aggressively.

    ReplyDelete
  2. Is there a parts list? directions for assembky?
    Thank you... so scared for my dad..

    ReplyDelete
    Replies
    1. Yes, I am working on it at this page
      https://panvent.blogspot.com/2020/03/design-for-positive-pressure-pandemic.html

      Delete
    2. Hi,
      I am Software/embedded engineer and I am from Pakistan. I would love to help you out in any way possible because my people (pakistanis) here need it direlu as we are running short of ventilators in our best hospitals. Anxiously, waiting for your response.

      Delete
  3. I was wondering if we can use eletric pumps like we use in Aquarium, large ones. Or even air compressor (eletric ones)

    ReplyDelete
    Replies
    1. Any compressor used mus have no oil. Oil in the air will damage the lung. Aquarium typ systems usually do not have sfficient flow and pressure

      Delete
  4. I looked at a lot of blowers today and worked out the specs required. I want something readily available. Small vacuum cleaners with high power like the Dyson meet both the flow and pressure requirements and are readily available. have alook at the "Design for a Positive Pressure Pandemic Ventilator" for more info.

    ReplyDelete
    Replies
    1. "high power like the Dyson" ... if you are looking at high blower like Dyson, and adding all the other components, and build time. May as well get a proper portable ventilator for $5k

      Delete
    2. Sure, If there were hundreds of thousands of ventilators in stock at suppliers and ready to go, they could just be ordered and used. Problem solved. The thing is, there is no stock, every manufacturer is using the "just in time" system now. The supply chains are mostly shut down. Many of the vital components need to come from China. I tried to order some components from China. They are 3 to 4 month lead times with no guarantee.

      Delete
    3. Should say, there is a small stock that is being deployed. Th manufacturers are doing everything they can to make as many ventilators as they possibly can right now. It well help a lot but will still not be enough for some regions.

      Delete
    4. What about a pump to inflate beds like this one? https://www.amazon.com/dp/B000FE5PE2?creativeASIN=B000FE5PE2&imprToken=3CEBrpFCUeyOpqso4G15-g&slotNum=9&tag=supercomfysleep-20&linkCode=ogi&th=1&psc=1

      https://supercomfysleep.com/best-mattress-pumps/

      Delete
  5. What are the flow and pressure requirements? In my garage I have a compressor (small enough to pick up but still 7cfm into a 20litre tank at 110lb psi), a leaf blower/vacuum or go crazy on the car's turbo or the common rail diesel pump. what range are we looking at?

    ReplyDelete
    Replies
    1. Estimated requirements (one patient) is 0.8 cu ft /min at 0.85 psi positive press.

      Delete
  6. Why not use a negative pressure ventilator which does not require tubing? That way, you need to build a regulator to regulate the pressure, a microcontroller to control the regulator, a UI to configure settings and a shell to place on your belly.

    ReplyDelete
    Replies
    1. Thanks, Yes That will work. Not useful for people with COVID symptoms but can be used for other people in the hospital that will be triaged and competing for remaining ventilators. This is what you mean. https://www.youtube.com/watch?v=pvrUQCMa3a8
      I am talking to this guy

      Delete
  7. Following this with much interest. Trying to figure out how to organize a mass production out of our isolated home workshops. Maybe Maker Spaces?

    ReplyDelete
  8. why will negative pressure ventilators not work with covid 19?

    ReplyDelete
  9. Not useful for people with COVID symptoms because lungs are too stiff fom fluid and infection to inflate properly. A positive pressure ventilator can force the alveoli open and keep them open. A negative pressure unit could be used for other people in the hospital that will be triaged and competing for remaining ventilators. This is what you mean. https://www.youtube.com/watch?v=pvrUQCMa3a8 It is the same principle, Called a Cuirass ventilator. Uses much less space. Issues with CO2 retention in these patients though.

    Cuirass Ventilator the DIY way
    I made a proof of concept of a negative pressure Curiass ventilator. The principle is to lower the pressure over the chest to make you breath in, then you breath out naturally, with a bit of help from the wetsuit squeezing you. I used a 12 litre box and a 12v airbed fan - using the deflate port to create negative pressure. The fan is powered by ...
    www.youtube.com

    I am talking to this guy

    ReplyDelete
  10. You should add "Ability to provide 100% oxygen" to your list of requirements. Hypoxemia seems to be a major problem with COVID-19.

    ReplyDelete
    Replies
    1. Excellent idea, backed up by other emails I have received. Is implemented in this design. https://panvent.blogspot.com/2020/03/updated-bellows-style-pandemic.html

      Delete
  11. This document from Richardson products gives a good overview of design features. Alarms are one example: with a cobbled together design does someone have to sit with an unconscious patient to make sure it keeps working?
    http://www.richardsonproducts.com/fdavent.pdf

    ReplyDelete
  12. This document by Richardson Products has some good explanation of design features such as alarms. For a bare bones machine, would someone have to sit with an unconscious patient to make sure it keeps working? Lots to think about.
    http://www.richardsonproducts.com/fdavent.pdf

    ReplyDelete
  13. We found some of the standards.

    https://www.reddit.com/r/DataHoarder/comments/fplzki/human_standards_help_share_the_international/

    ReplyDelete
  14. Is it been approved for Coronavirus Pandemic! use? Could you share more details? Origin, Specifications, Approvals & QC/QA reports. Thanks.

    ReplyDelete
  15. Hello my name is Andres, I'm not a tech guy, I'm more of a number cruncher and I visualize machines and then create them. The issue I see with this is the numbers and time. Here is my visualization..... Not exactly accurate just a zeroed out estimate.... Ok
    40000 respirators needed in NY(2weeks),
    20000000 people in NY,
    ???? Respirators needed for USA,
    350000000 people in USA(2weeks)
    The question is they need air!!correct!
    Ok,
    Take one 50 hp rotary screw compressor
    Tankless made in America at 200-235 CFM
    With constant 115 psi... Ok...
    Now put it in a corner of a NFL stadium
    Put one air feed line running up and down the field with space in between for patients to lay with there head at the feed line. Now all you need is a little control stations at each person's head to control the air for each person with oxygen tanks next to them as needed. With the numbers flying through my head and visualizing it. The power supply will never run out due to back up generators in the stadium. One rotary screw will act a one extremely big respirator that will save 26600 people at one time. I dout one stadium could hold that many. Not I see the air supply running through the stands as well. Ok. The stadium turns into one big respirator. Just need the little pieces to fill in..such as valves, air lines, controls for each individual, etc. etc... . My numbers maybe over kill. But we need to think bigger scale now because this virus can hit puberty any min. It is still a baby. And at a 6 percent death rate can easily turn into 60 percent. If I had the funds and the manpower I could create it in two weeks. There is some great minds in this blog, my first blog, if you all can fill in the individual patient aspect it can work, and work quickly. The less moving parts the less chance of failure. WE ALL SHOULDN'T BE AIMING FOR THE STARS, WE SHOULD MAKE THE STARS OUR WALKWAY!!! Call me at 4049884371 and tell me if it worked.

    ReplyDelete
    Replies
    1. Generally these patients require individualized settings and so each need their own machine. Scaling up has been done up to a factor of 4 on 1 ventilator in testing, and 2 per ventilator in practice, but it completely eliminates individualized settings and compromises safety systems.

      Delete

Commenting has been limited for this post

Note: Only a member of this blog may post a comment.