Are portable met carts the new lactate testing?

[davidof]

There’s been some discussion about portable metabolic carts, but I haven’t seen a dedicated thread yet. I know some people here have used them or even own one such as the Calibrebio.

With recent advances in electronics, portable met carts are becoming more affordable and accessible to clubs, coaches, and even individuals.

A metabolic cart measures how much oxygen you inhale and how much carbon dioxide you exhale during exercise. Late last year I looked at a new device, the ZoneX, from French startup Pairfs (Grenoble).

A bit of background: as far back as the early 19th century, researchers noticed a link between exercise intensity and lactic acid. Today, blood lactate can be measured with portable analyzers, but testing is intrusive, error-prone, and not always easy to interpret. Portable met carts take a different approach by analysing respiration. They measure ventilation (VE), carbon dioxide output (V̇CO₂), and a derived metric called EqCO₂ (air volume divided by CO₂ volume).

During aerobic exercise, energy is produced using oxygen, with CO₂ as a by-product. As intensity increases, CO₂ rises faster than ventilation, causing EqCO₂ to fall. When anaerobic metabolism starts contributing, lactate begins to accumulate and EqCO₂ levels off. This marks the first ventilatory threshold (VT1 or LT1), with Zone 2 just below it. At higher intensities, buffering capacity is exceeded, ventilation increases sharply, and the second ventilatory threshold (VT2 or LT2) is reached.

So your metcart can help determine VT1 and VT2 as well as give you your fatmax.

Norwegian coach Olav Aleksander Bu has highlighted the value of frequent, non-intrusive testing. Portable met carts make this possible and can be used across different sports on a treadmill, rowing machine, ski erg, or even outside the lab.

In practice, the data isn’t always perfectly clean, just like lactate testing, trends become clearer with repeated measurements. Still, portable met carts are far less intrusive, much cheaper than lab systems (which can cost $30k+), and don’t require consumables for each test.

Here is a video I made about the ZoneX: https://www.youtube.com/watch?v=7AX9P2ncUms

and a link to the firm for anyone interested https://www.pairfs.com/

I have no relationship with this firm, I was just curious about the device and thought people here might be interested.

[GarageLab]

I own a consumer portable metabolic cart. Researched a bunch before buying in 2020 and bought a Pnoe at a nice discount. I wanted to try VO2master, but couldn’t get a loaner device to try. And with VO2master at $7k and Pnoe over $10k (with biannual O2 sensor replacement and calibration at $1.5-2k), they don’t offer good enough data for the price.

My Pnoe device is kind of ok, but the data is questionable at times. Even using calibration gas, you can do back-to-back resting tests and have RER at 0.75 one test, and then recalibrate and have RER showing 0.85 with no change in testing conditions. Or, I’m near VO2max, and RER still shows 0.95-0.99.

I haven’t seen any consumer cart that offers good data and there are some papers out there comparing multiple carts. I considered buying a refurb lab cart at $12-15k, but since I only really test myself, that’s a little steep for a hobby.

I learned a lot after collecting a couple hundred hours of data on myself, but at this point, I probably won’t buy a new cart when mine goes out of service.

[davidof]

Yes that is my feeling. I also have a calibrebio on test and have yet to get consistent results but that might be user error. I need to spend some more time with it. I found the same with lactate testing, messy, painful and not always easy to get clean readings - hence the thread title.

I’ve watched a few youtube videos with full lab tests and even there the results are not always clean with the “tester” kind of interpreting things based on his/her experience.

[RossH4]

I’ve worked in several research labs over the years and with a number of Lab grade carts. I wanted to add a smaller unit to my coaching business and leaned on some of the industry contacts and colleagues for additional input. The biggest concern is that without a CO2 sensor you are very limited in making accurate substrate measurements, and its also nice to have it involved in the detection of VT2. So the first overall problem is you are relying on a lot of estimations from algorithms. I think the value of estimates and the cost is a bit of a mismatch. As for specific units, while I haven’t used the Pnoe, colleagues who have, told me the software calculations and calculations from raw data seem to differ quite substantially. This was a major concern for those in more of a research setting. Calibre is really interesting but I cannot see how a startup which seems can come up with sensors which are so substantially cheaper than the bigger manufactures who would love to be able to offer a portable unit but have been unable to do so. In addition the issues with condensation build up is a bit of deal breaker for athletes who have greater tidal volumes and who will be working at some intensity. The VO2Master is interesting but without CO2 I just think its too limited for the cost. All in for something portable I’ll be sticking to lactate testing for now, the value in full metabolic testing for me comes from additional substrate assessments so without a reliable ability there I dont think its worth it. I do have some hope for calibre though and I have a colleague doing some independent testing there against a COSMED quark so itll be interesting to see how that goes.

[GarageLab]

@RossH4

I’ve worked in several research labs over the years and with a number of Lab grade carts.

I’d be curious to get your overall opinion on what devices give reasonable results. I should find a local lab to work part-time to get more experience.

@RossH4

The biggest concern is that without a CO2 sensor you are very limited in making accurate substrate measurements, and its also nice to have it involved in the detection of VT2.

@RossH4

The VO2Master is interesting but without CO2 I just think its too limited for the cost.

This was my #1 reason for not buying a VO2Master. Knowing VO2 is great, but not having VCO2 just doesn’t make sense. And VO2Master was saying they’re bringing CO2 measurement for years, but it hasn’t happened yet. Price went from $4,999 to $6,900 in just a couple years.

@RossH4

As for specific units, while I haven’t used the Pnoe, colleagues who have, told me the software calculations and calculations from raw data seem to differ quite substantially.

I would be interested in knowing more about this. I spent a year troubleshooting and getting a replacement device to get a working Pnoe unit. Spent several hours in meetings with support and the device inventor trying to understand how to ensure I get the best data. VO2 measurement seems ok, but CO2 is usually questionable even using calibration gas.

Also, the device has a serious design flaw, and needs an additional inline moisture separator from the flow sensor to the unit because the helapet filter will clog with breath moisture and give false readings. You have to swap filters to maintain air flow, which isn’t a problem, but limits the lifespan of the filters and adds yet another thing to babysit as you try to focus on testing.

@RossH4

I do have some hope for calibre though and I have a colleague doing some independent testing there against a COSMED quark so itll be interesting to see how that goes.

I just can’t believe Calibre, yet. Like you mentioned, how can a startup offer the claimed accuracy when Cosmed, VacuMED, etc. can’t offer a portable device for less than $20K+?

@RossH4

All in for something portable I’ll be sticking to lactate testing for now, the value in full metabolic testing for me comes from additional substrate assessments so without a reliable ability there I dont think its worth it.

I’m with you here. A lactate meter and a power meter are more than enough for me at this point. I can’t justify spending that much money for the lackluster performance of the portable metabolic devices.

[GarageLab]

@davidof

Yes that is my feeling. I also have a calibrebio on test and have yet to get consistent results but that might be user error. I need to spend some more time with it. I found the same with lactate testing, messy, painful and not always easy to get clean readings - hence the thread title.

It took me a while to get really comfortable with my equipment. But you have to practice, or else it’s hard to know when you’re not getting good data or something is wrong. I’ve done a lot of testing and retesting to see what changes, trying different best practices, following the manufacturer’s instructions to the letter, etc. These devices require a lot of attention and the data needs processing before analysis, like removing spikes, dropouts, etc.

[RossH4]

So the clinical grade equipment is all going to be deemed within accurate limits and variance will be much the same (COSMED, Vyntus, Metamax and Metalyzer are the ones I know best), they all have a little quirk here and there so its probably less about accuracy and more about reliability and user friendliness. Right now I’m using a new COSMED Quark, and its got its quirks but generally works well so it probably would be no.1. The other units I used were pretty old and heavily used, so I suspect I didn’t see them at their best. I don’t like the automatic detection of thresholds on any of them. I accept auto-detection of VO2max purely for standardisation and defence of methodology (for write-ups). I also wouldn’t use respiratory data for training zones and thresholds. I think the shape of a lactate curve gives more texture to things and just gives better insight. Wouldn’t use automated detection tools for lactate either but thats more a preference and how I was taught way back during my MSc.

I think when Vo2master has CO2 the cost will fall in line with other more portable analysers like Metamax. Anything that doesn’t need calibration for me is questionable, even with one athlete, the air composition of the room changes a decent amount, so I struggle to see how you can correct for that with out some calibration gas.

As for PNOE all I have there is secondhand info but its software was described to me as a random number generator. So for me I just stayed clear.

Cardiocoach is another I’ve seen being used by local testers but any data reports I’ve seen have been a bit suspicious. Don’t know if that’s operator or device related.

So yea I’m still intrigued by calibre, the only validation study I saw was quite promising and it has raised a lot of interest amongst my colleagues. We are all sort of wating to see because it did somehow work it would open a lot of doors. The old saying, if it seems to good to be true. But I’ll be watching that one.

So Metamax is probably my vote for a portable device. But you still need calibration gas and all the licenses and kit that go with it.

[GarageLab]

Thanks so much for the input. I spent several days re-investigating metabolic carts. The Metamax is the device I’ve been looking for, but even that device has been shown to have “offset” readings from true value. For me, I just need a device to give me reliable data and put the data in a CSV. That’s it.

I starting putting together an app to look at my data and plot the device uncertainty along with the readings. Based on Pnoe’s stated sensor specs for O2, CO2, and flow, the readings really can appear random. Performing Haldane computatons and comparing to the device reported values shows this. Of course, I could be calculating things incorrectly.

Independent studies show high error in Ventilation (VE) (~44%) and unreliable fat oxidation data (~76% error).

But on a different test, I see these results:

But, hey, I can produce “pretty” Wasserman-style Plots

And when doing treadmill graded-tests, it’s nearly impossible to compare:

Still fun to practice, though.

[RossH4]

I think thats sort of the trade off with the portable devices. Keep in mind that the O2 sensors on the COSMED are so sensitive, that even vibrations around the device through the floor can impact data. So a moving analyser sensitive enough to be accurate will pick up some noise. I guess you don’t need to be 100% accurate because what is that really. You just need it to be consistent and reliable. Accuracy is in some ways subjective because you are comparing to a gold standard which might in itself have quirks in how it handles data. I think the metamax is as good as anyone would need. It’s certainly used in published data sets.

[davidof]

that’s really cool, I went down the same route with calibrebio writing my own scripts to analyse the data, I posted the code on github if anyone is interested

GitHub - davidzof/calibrebio-analyzer: Some python code to analyse the data from the...

that’s the easy bit, getting good, reliable data is harder

Fatmax looks plausible, around 132 bmp.

This is VO2/VCO2 with VT1 and VT2 automagically calculated by the program. However you see that VCO2 and VO2 go a bit haywire at 950S - did the sensors get saturated. Calibre thought it was a mask fit issue. A VT1 at 138 and VT2 at 152bpm are in the ballpark but is this just wishful thinking?

I watched quite a few videos done in the lab and they seem to have issues with VO2/VCO2 at the upper and of the test and just do a lot of armwaving with clients. VO2 of 2.75l/m seems a bit low to me though.

Finally VO2/VCO2, well it looks pretty.

I also did my own Haldane calculations but my garage was really cold that day and the Calibrebio readings were not really usable. I’m going to come back to that when I retest, don’t read too much into the above graphs, it is not a definitive review of Calibrebio.