8 ms·

Apple Watch detecting coronary ischaemia during chest pain episodes

132 points by 1cvmask 7y ago

aladoc99 7y ago
Impressive utility of the single-lead ECG recording, but I have to say that if I'm in charge, that patient's headed for bypass surgery, not stenting.
- unsrsly 7y ago
  I'm not sure the HN community participates in the CABG PCI wars, but maybe your comment will lead to a fiery debate.
  - ashtonkem 7y ago
    Never underestimate the ability for HN to participate in an extremely niche flame war.
    - Austin_Conlon 7y ago
      Please don't make comments like this. We’re also well-equipped to participate in flame wars about the popular culture.
- epmaybe 7y ago
  Eh, I think the data is controversial and case-dependent in two-vessel disease. If this were three vessel disease, the data is far more in favor of CABG. The fluoroscopy doesn't look convincing to me of major stenosis, so a PCI/stent could certainly be placed for symptomatic relief without a CABG. Additionally, the patient definitely did not need emergency CABG. My understanding is that CABG could be completed after stenting, and may very well have happened for this patient without our knowledge.
- Herodotus38 7y ago
  I’ll take you up on the bypass statement. :) This is an 80 yo woman, that alone would probably make me not want to put her under for bypass. Also, she had a history of paroxysmal afib and prior pulmonary embolism (they don’t state if it wad provoked or not) but is not on anticoagulation. By CHADS2-Vasc criteria she should have been on AC alone, but isn’t. She’s clearly intelligent and has no issues with access to care (engineer, has an Apple Watch) which further makes me think that her underlying health warrants a more conservative approach.
  - op03 7y ago
    I agree. Redoing plumbing of an 80 year old house, makes sense only under certain conditions eg- part of a living history museum.
    - Herodotus38 7y ago
      Plus, I don’t want to risk giving the house a stroke or cognitive injury from the risks of the operation and general anesthesia when a comparable or better outcome could be done with PCI. It’s all about risks and benefits for her.
- azalemeth 7y ago
  In three vessel disease, PCI is decidedly worse than CABG [1], but the latest evidence is, AFAIK, equivocal for CABG over PCI in one and two vessel disease [2]. I've personally often thought that this is due to the degree of collatoralisation that goes on in the ischaemic myocardium, even when we are convinced that the reduced FFR is all the story: stenosis does not happen overnight and metabolic adaption to, e.g., decreased O2 starts to begin. In three vessel disease you've burnt your reserves and are desperate for O2. In two vessel disease you're (mal)adapted. Incidentally, this is why my scientific research is focused on using a slightly crazy CMR technique to directly image glycolysis and beta-oxidation in the ischaemic myocardium. We can make PCI kill fewer people if the interventionalists can be told "This region is ischaemic and has shifted towards anaerobic glycolysis due to this branch of (LCx/whatever) artery. Go revascularise". [1] https://academic.oup.com/eurheartj/article/35/40/2821/2293222 https://academic.oup.com/eurheartj/article/35/40/2821/229322... [2] https://openheart.bmj.com/content/3/2/e000489?utm_term=usage-012019&utm_content=consumer&utm_campaign=oh&utm_medium=cpc&utm_source=trendmd https://openheart.bmj.com/content/3/2/e000489?utm_term=usage...
  - unsrsly 7y ago
    Hyperpolarized 13C MR?
    - azalemeth 7y ago
      Yep :-)
    - azalemeth 7y ago
      Are you in the field -- and if so, where do you work? UCSF? (I'm in Oxford :-)
atian 7y ago
Treat the Apple Watch as a single lead and you can piecewise together a full ECG. Each of the single-lead ECG recording be off by 30 seconds but is good enough for most purposes.
- johannes_ne 7y ago
  The apple watch ECG corresponds to lead I. Lead II and III is right and left arm respectively to left foot, so they will be possible with a bit of flexibility. The 6 precordial leads are not possible, since they are measuring the voltage between an electrode on the chest and the mean of the 3 extremity electrodes.
- yosito 7y ago
  I have very little understanding of how an ECG works. Can you ELI5 why a single-lead ECG recording can be off by 30 seconds, and what that means?
  - LeoPanthera 7y ago
    He means that you would have to do many single-lead recordings separately, rather than a multi-lead recording once. The standard length of the Apple Watch ECG recording is 30 seconds.
- sradman 7y ago
  The poor man’s single lead ECG is complemented by continuous PPG heart rate data. Being able to combine the data of multiple sensors that are synchronized with a single clock provides new opportunities for monitoring and analysis. HRV data is also useful and I’m surprised that it hasn’t been used to estimate the respiration rate.
  - 2dvisio 7y ago
    Quite unlikely they did not try. Reflectance pulse oximetry is unfortunately affected a lot by ambient light and so, even for the Apple Watch, that might have caused unreliable measures. Using the ECG for RR estimation though should be more than feasible...
londons_explore 7y ago
If I am reading the abstract correctly, it seems a medical procedure was carried out here based on evidence solely from an apple watch. Other medical tests either came out fine or weren't carried out. Doesn't that break a million rules? What if that trace was just the result of Apple testing out a new data postprocessing algorithm?
- mritun 7y ago
  Naah, they merely looked at Apple Watch graphs and skipped more noninvasive diagnostics directly to invasive diagnostic procedure - aka catheterization aka angiography. Angiography is the real test that can find the blockage but is invasive, so not done unless there is solid suspicion. Summary of the paper: Apple Watch was right. They found the blockage and fixed it.
- tinus_hn 7y ago
  I don’t think Apple can get FCC approval if they just test new algorithms on end user devices.
option_greek 7y ago
Is this specific to apple watch ? Are the other fitness trackers not suited for this ?
- sterlind 7y ago
  I think the Apple watch is (nearly) alone in being FDA-approved as a medical device, and I wouldn't trust anything that isn't. If I didn't need an iPhone to power it I'd buy one in a heartbeat just for health reasons.
  - yosito 7y ago
    > I'd buy one in a heartbeat I see what you did there.
  - BadOakOx 7y ago
    "FDA-cleared" not approved. There is a huge difference: https://www.cnet.com/news/fda-approved-vs-fda-cleared-whats-the-difference/ https://www.cnet.com/news/fda-approved-vs-fda-cleared-whats-...
  - mosselman 7y ago
    What is wrong with an iPhone?
    - roschdal 7y ago
      Very expensive telephone.
      - sudhirj 7y ago
        Can get the SE, which is pretty midrange unless you’re comparing it to feature phones.
        hexmiles 7y ago
        where i am the SE is 500€. Is a LOT more than the 180€ that i payed my xiaomi. I wouldn't really call it midrange... but i guess it depends where you live? Considering another 460€ for the watch it become very expensive. Also not long ago 500€ (here) was considered a notch under top of the line. edit: as a reference, feature phone extremely rarely surpass the 100€ on amazon and are generically around 50-60€. edit: fix grammar
mbubb 7y ago
pretty cool - anecdotally I've had a funny call or two involving Apple watches (as an EMT). My favorite was during a local celebration of the "Holi" festival (aka Indian festival of colors). It was setup in a waterfront park and we had an EMS station and got a series of calls through the day. One case a young woman came up to our station to say a guy was maybe having a heart attack... We followed her to a tall, lanky college age kid seated on a stone bench freaking out - labored rapid breathing, etc Talked to him a bit to get some history and he kept showing us his watch - his skin and watch (and just about everything else) was caked in the fine dust which folks were throwing into the air and at each other. It was reading some absurdly high (or low I don't remember) pulse rate. But the guy was talking in full sentences - had good perfusion and seemed to know who he was and where he was, etc. I don't remember if we gave him oxygen - we took his vitals manually and eventually convinced him that the watch was not reading correctly. He calmed down and refused further attention and was back hurling packets of colored dust at his buddies in a few minutes... Its awesome people are using these devices (pulse ox devices are excellent for detectign silent hypoxia in Covid cases even before a person has obvious signs - see recent NYTimes opinion piece) Even if they are nto accurate (and they generally arent) they give the person a baseline. "Usually my pulse is x and now it is y" is useful information for an EMT anyway (we don't diagnose as we are not doctors - we look for obvious life threats and rate of change for various external signs). Just don't freak out if your device is malfuntioning. Have a second or third way to verify a vital sign...
- vulcan01 7y ago
  > pulse ox devices are excellent for detectign silent hypoxia in Covid cases even before a person has obvious signs - see recent NYTimes opinion piece I'm having trouble finding the opinion piece... If someone bookmarked it please post it here :)
  - mbubb 7y ago
    https://www.nytimes.com/2020/04/20/opinion/sunday/coronavirus-testing-pneumonia.html https://www.nytimes.com/2020/04/20/opinion/sunday/coronaviru...

132 points by 1cvmask 7y ago

Impressive utility of the single-lead ECG recording, but I have to say that if I'm in charge, that patient's headed for bypass surgery, not stenting.

unsrsly 7y ago
I'm not sure the HN community participates in the CABG PCI wars, but maybe your comment will lead to a fiery debate.
- ashtonkem 7y ago
  Never underestimate the ability for HN to participate in an extremely niche flame war.
  - Austin_Conlon 7y ago
    Please don't make comments like this. We’re also well-equipped to participate in flame wars about the popular culture.
epmaybe 7y ago
Eh, I think the data is controversial and case-dependent in two-vessel disease. If this were three vessel disease, the data is far more in favor of CABG. The fluoroscopy doesn't look convincing to me of major stenosis, so a PCI/stent could certainly be placed for symptomatic relief without a CABG. Additionally, the patient definitely did not need emergency CABG. My understanding is that CABG could be completed after stenting, and may very well have happened for this patient without our knowledge.
Herodotus38 7y ago
I’ll take you up on the bypass statement. :) This is an 80 yo woman, that alone would probably make me not want to put her under for bypass. Also, she had a history of paroxysmal afib and prior pulmonary embolism (they don’t state if it wad provoked or not) but is not on anticoagulation. By CHADS2-Vasc criteria she should have been on AC alone, but isn’t. She’s clearly intelligent and has no issues with access to care (engineer, has an Apple Watch) which further makes me think that her underlying health warrants a more conservative approach.
- op03 7y ago
  I agree. Redoing plumbing of an 80 year old house, makes sense only under certain conditions eg- part of a living history museum.
  - Herodotus38 7y ago
    Plus, I don’t want to risk giving the house a stroke or cognitive injury from the risks of the operation and general anesthesia when a comparable or better outcome could be done with PCI. It’s all about risks and benefits for her.
azalemeth 7y ago
In three vessel disease, PCI is decidedly worse than CABG [1], but the latest evidence is, AFAIK, equivocal for CABG over PCI in one and two vessel disease [2]. I've personally often thought that this is due to the degree of collatoralisation that goes on in the ischaemic myocardium, even when we are convinced that the reduced FFR is all the story: stenosis does not happen overnight and metabolic adaption to, e.g., decreased O2 starts to begin. In three vessel disease you've burnt your reserves and are desperate for O2. In two vessel disease you're (mal)adapted. Incidentally, this is why my scientific research is focused on using a slightly crazy CMR technique to directly image glycolysis and beta-oxidation in the ischaemic myocardium. We can make PCI kill fewer people if the interventionalists can be told "This region is ischaemic and has shifted towards anaerobic glycolysis due to this branch of (LCx/whatever) artery. Go revascularise". [1] https://academic.oup.com/eurheartj/article/35/40/2821/2293222 https://academic.oup.com/eurheartj/article/35/40/2821/229322... [2] https://openheart.bmj.com/content/3/2/e000489?utm_term=usage-012019&utm_content=consumer&utm_campaign=oh&utm_medium=cpc&utm_source=trendmd https://openheart.bmj.com/content/3/2/e000489?utm_term=usage...
- unsrsly 7y ago
  Hyperpolarized 13C MR?
  - azalemeth 7y ago
    Yep :-)
  - azalemeth 7y ago
    Are you in the field -- and if so, where do you work? UCSF? (I'm in Oxford :-)

atian 7y ago

Treat the Apple Watch as a single lead and you can piecewise together a full ECG. Each of the single-lead ECG recording be off by 30 seconds but is good enough for most purposes.

johannes_ne 7y ago
The apple watch ECG corresponds to lead I. Lead II and III is right and left arm respectively to left foot, so they will be possible with a bit of flexibility. The 6 precordial leads are not possible, since they are measuring the voltage between an electrode on the chest and the mean of the 3 extremity electrodes.
yosito 7y ago
I have very little understanding of how an ECG works. Can you ELI5 why a single-lead ECG recording can be off by 30 seconds, and what that means?
- LeoPanthera 7y ago
  He means that you would have to do many single-lead recordings separately, rather than a multi-lead recording once. The standard length of the Apple Watch ECG recording is 30 seconds.
sradman 7y ago
The poor man’s single lead ECG is complemented by continuous PPG heart rate data. Being able to combine the data of multiple sensors that are synchronized with a single clock provides new opportunities for monitoring and analysis. HRV data is also useful and I’m surprised that it hasn’t been used to estimate the respiration rate.
- 2dvisio 7y ago
  Quite unlikely they did not try. Reflectance pulse oximetry is unfortunately affected a lot by ambient light and so, even for the Apple Watch, that might have caused unreliable measures. Using the ECG for RR estimation though should be more than feasible...

londons_explore 7y ago

If I am reading the abstract correctly, it seems a medical procedure was carried out here based on evidence solely from an apple watch. Other medical tests either came out fine or weren't carried out. Doesn't that break a million rules? What if that trace was just the result of Apple testing out a new data postprocessing algorithm?

mritun 7y ago
Naah, they merely looked at Apple Watch graphs and skipped more noninvasive diagnostics directly to invasive diagnostic procedure - aka catheterization aka angiography. Angiography is the real test that can find the blockage but is invasive, so not done unless there is solid suspicion. Summary of the paper: Apple Watch was right. They found the blockage and fixed it.
tinus_hn 7y ago
I don’t think Apple can get FCC approval if they just test new algorithms on end user devices.

option_greek 7y ago

Is this specific to apple watch ? Are the other fitness trackers not suited for this ?

sterlind 7y ago
I think the Apple watch is (nearly) alone in being FDA-approved as a medical device, and I wouldn't trust anything that isn't. If I didn't need an iPhone to power it I'd buy one in a heartbeat just for health reasons.
- yosito 7y ago
  > I'd buy one in a heartbeat I see what you did there.
- BadOakOx 7y ago
  "FDA-cleared" not approved. There is a huge difference: https://www.cnet.com/news/fda-approved-vs-fda-cleared-whats-the-difference/ https://www.cnet.com/news/fda-approved-vs-fda-cleared-whats-...
- mosselman 7y ago
  What is wrong with an iPhone?
  - roschdal 7y ago
    Very expensive telephone.
    - sudhirj 7y ago
      Can get the SE, which is pretty midrange unless you’re comparing it to feature phones.
      - hexmiles 7y ago
        where i am the SE is 500€. Is a LOT more than the 180€ that i payed my xiaomi. I wouldn't really call it midrange... but i guess it depends where you live? Considering another 460€ for the watch it become very expensive. Also not long ago 500€ (here) was considered a notch under top of the line. edit: as a reference, feature phone extremely rarely surpass the 100€ on amazon and are generically around 50-60€. edit: fix grammar

mbubb 7y ago

pretty cool - anecdotally I've had a funny call or two involving Apple watches (as an EMT). My favorite was during a local celebration of the "Holi" festival (aka Indian festival of colors). It was setup in a waterfront park and we had an EMS station and got a series of calls through the day. One case a young woman came up to our station to say a guy was maybe having a heart attack... We followed her to a tall, lanky college age kid seated on a stone bench freaking out - labored rapid breathing, etc Talked to him a bit to get some history and he kept showing us his watch - his skin and watch (and just about everything else) was caked in the fine dust which folks were throwing into the air and at each other. It was reading some absurdly high (or low I don't remember) pulse rate. But the guy was talking in full sentences - had good perfusion and seemed to know who he was and where he was, etc. I don't remember if we gave him oxygen - we took his vitals manually and eventually convinced him that the watch was not reading correctly. He calmed down and refused further attention and was back hurling packets of colored dust at his buddies in a few minutes... Its awesome people are using these devices (pulse ox devices are excellent for detectign silent hypoxia in Covid cases even before a person has obvious signs - see recent NYTimes opinion piece) Even if they are nto accurate (and they generally arent) they give the person a baseline. "Usually my pulse is x and now it is y" is useful information for an EMT anyway (we don't diagnose as we are not doctors - we look for obvious life threats and rate of change for various external signs). Just don't freak out if your device is malfuntioning. Have a second or third way to verify a vital sign...

vulcan01 7y ago
> pulse ox devices are excellent for detectign silent hypoxia in Covid cases even before a person has obvious signs - see recent NYTimes opinion piece I'm having trouble finding the opinion piece... If someone bookmarked it please post it here :)
- mbubb 7y ago
  https://www.nytimes.com/2020/04/20/opinion/sunday/coronavirus-testing-pneumonia.html https://www.nytimes.com/2020/04/20/opinion/sunday/coronaviru...