Hacker News Reader

Hacker News

Comment by kuprel | Hacker News Reader

Parent Full thread

kuprel

I read that a CT scan outperforms the lab test at detecting it

View on HN

A CT scan is not an appropriate screening test for COVID-19.

Also, that paper will have no clinical impact. Thread: https://twitter.com/drlukeor/status/1233547995192279040?s=21

Follow-up from the American College of Radiology: https://www.acr.org/Advocacy-and-Economics/ACR-Position-Stat...

"CT should not be used to screen for or as a first-line test to diagnose COVID-19"

The latest guidelines published by the Chinese CDC claimed so too, a CT scan is required, and RT-PCR tests have to be applied multiple times before claiming a patient is cured.

Any idea why a chest X-ray cannot be used, as is for TB?

CT scans involve a lot of ionizing radiation. Definitely something to be avoided unless necessary.

According to [1], a chest CT scan rad dose ranges from 4 to 18 milliSieverts. Annual background dose is ~3 milliSv. [2] states that a 100 mSv dose would increase risk of cancer by 0.5% (5/1000).

[1] https://www.health.harvard.edu/cancer/radiation-risk-from-me...

[2] https://www.scientificamerican.com/article/air-travel-expose...

You cannot compare an annual background dose to the dose received in a CT scan. Similarly, the 100 mSv dose of your reference number 2 is an inappropriate comparison because it is spread out over time compared to a CT scan. Your premise is like saying, walking puts an impact of .8 J on my foot. I walk 10,000 steps a day. 10,000*.8 is 8,000 J. A bullet produces an impact of 168 J when fired from a .22 rifle into a human body. Therefore walking is much more dangerous than a bullet from a rifle.

What you suggest seems plausible, but I can't immediately calculate, for example, DNA repair rate vs. X-ray photon flux and cross-sections. Can you point me to something that confirms incidence or severity of radiation-induced disease depends on flux rather than dose? Tnx.

Sun burns are known to significantly increase cancer risks vs similar exposure over longer periods. Sun burns are rather similar to accuse radiation poisoning suggesting similar issues.

Both dose and flux matter.

Above, there is a link to a report from Harvard Medical School, that claims otherwise.

Maybe they are wrong, but a post on HN is not exactly a credible refutation.

And yet, this Harvard Medical School link (not a report, not peer reviewed, but whatevs) clearly states that one CT scan of 4 to 18 mSV increases cancer risk by 0.7%, which is higher than the increase of 0.5% from 100 mSV exposure over a greater period of time. So, no. The Harvard Medical School link does not refute in any way my point. In fact, it supports my point.

The Sievert unit is based upon the Gray unit of radiation dose, but it also includes a weighting factor for the type of biological tissue that is exposed to radiation. You can also read up on the linear no-threshold model of radiation exposure for more information. In terms of cancer risk, radiation doses on a range from 1 mSv to 100 mSv can definitely be compared, regardless of the duration of the exposure. There is a range of higher doses e.g. above 1 Gy that would cause acute radiation syndrome if received all at once, where different biological effects would predominate.

0.5% absolute points or by 0.5%? Chance of cancer is already really low. Multiplying it by 1.005 doesn't make much of a difference. Adding 0.5% does.

Wouldn't tracking a virus like this be such a necessary situation?

Reply on news.ycombinator.com