Microscopic "Tuning Forks" Help Determine Effectiveness of Antibiotics

sciencehabit writes "A patient admitted to a hospital with a serious bacterial infection may have only a few hours to live. Figuring out which antibiotic to administer, however, can take days. Doctors must grow the microbes in the presence of the drugs and see whether they reproduce. Rush the process, and they risk prescribing ineffective antibiotics, exposing the patient to unnecessary side effects, and spreading antibiotic resistance. Now, researchers have developed a microscopic 'tuning fork' that detects tiny vibrations in bacteria. The device might one day allow physicians to tell the difference between live and dead microbes—and enable them to recognize effective and ineffective antibiotics within minutes."

Comments

Another rambling bullshit summary

[oldhack]

That's nice that a new technique is developed to measure/observe bacteria, but what's with all that bullshit about rushed bacterial infection?

PR idiots.

Re:Another rambling bullshit summary

[quantumghost]

That's nice that a new technique is developed to measure/observe bacteria, but what's with all that bullshit about rushed bacterial infection?

PR idiots.

As a clinical (critical care, if you care to know) physician, I too am a bit puzzled by the description.

Patients in septic shock are very sick and the prescription of antibiotics is a delicate subject....antibiotics need to be started within a few hours of diagnosis, and getting it wrong (prescribing an antibiotic to which the bacteria is resistant) and the patient has a 50% increase in mortality. To this end we use the broadest spectrum antibiotics available, and most hospitals develop an "Antibiogram " specific for their institution and their pt population. These antibiotics are so powerful, it is rare, but not unheard of, for organisms to be resistant to them.

The process goes like this:

Pt is admitted to an ICU

Cultures of all likely sources (urine, lung, blood, CSF, abscess fluid) are obtained

Antibiotics are started (sometimes before the cultures are drawn, but ideally after), as well as other therapies

Over the next few days the antibiotics are "De-escalated" as dictated by the cultures (see below)

Hopefully the pt recovers and their care is down-graded and ultimately discharged

The cultures are sent to the lab after being draw and in a process that (time-wise) parallels the above:

The sample is extracted from the specimen container and are plated on a growth medium or placed in a broth

They are allowed to grow for (around) 24 hours

The plates are examined to determine if anything actually grew (may take up to 3 days for blood)

If something grew, two processes happen:

The culture is sent through a variety of tests (gram-stain, etc) to determine the species of bacteria which will dictate the next step.

The specimen is then re-suspended in a culture medium and plated and allowed to grow in the presence of antibiotics thus yielding that particular organisms antibiogram

A you can see, there really isn't anywhere to rush the process. And I would be very interested to see how they can speed this up with their technology....the who purpose of the plating is to amplify the bacteria from the milieu of the body fluids and to find the dominant organism growing.

In addition, some cultures are already "contaminated" with body flora (e.g. upper respiratory and stool) and the purpose of the culture is to amplify pathological bacteria from the benign-normal flora.

Longer video that gives a better front to back description

Re:Another rambling bullshit summary

[pepty]

The process goes like this:

They are allowed to grow for (around) 24 hours

The plates are examined to determine if anything actually grew (may take up to 3 days for blood)

If something grew, two processes happen:

The culture is sent through a variety of tests (gram-stain, etc) to determine the species of bacteria which will dictate the next step.

The specimen is then re-suspended in a culture medium and plated and allowed to grow in the presence of antibiotics thus yielding that particular organisms antibiogram

A you can see, there really isn't anywhere to rush the process.

qPCR, RT-PCR, and/or ELISA tests to determine which bacteria are common and which antibiotic resistance genes are heavily expressed at the infection site or the blood stream. Would only take a few hours from taking the sample to having results, as opposed to 1-3 days to do a culture.

Re:Another rambling bullshit summary

[pchimp]

The real usefulness of a technique like this (as I understand from the sepsis researchers that I've interacted with), is improved antibiotic stewardship -- preventing overuse of antibiotics and reducing the time the patient need be treated. These rapid assays can provide a better and more timely means to monitor a patient's response to treatment.

The talk about early diagnosis and saving lives is simply a lot more sexy and easier to sell than antibiotic stewardship. The grandstanding about this particular application aside, though, the technique itself is interesting.

Re:Another rambling bullshit summary

[the.original.drg]

qPCR, RT-PCR, and/or ELISA

Doesn't this assume you know what bacterium you are looking for, that you have the right primers, and that you know where the antibiotic resistance genes are? There seem to be too many variables to test for all of them. And why qPCR instead of regular old PCR? qPCR machines are pricey.

Re:Another rambling bullshit summary

[quantumghost]

Granted, but in my experience, common and uncommon are anything but....

Also PCR and ELISA are much more expensive and time consuming processes than plating or "brothing", and you also have to have a reasonable clue as to the organism that you expect to encounter (see below) - If I knew the bug before hand, I'd treat for it. Those in my position often get surprised by the organism that finally grows. I once had a case of endocarditis in a cardiac transplant patient, the culture came back with an unusual organism....googl-ing that organism yielded 4 hits (granted not the most scientific process) - sorry can't recall the bug, it was that unusual.

But also please reference my statement about amplification of the dominant organism above the ambient noise.

Sorry, just looked down and saw the reply by the.original.drg, basically the same argument, but with some personal experience thrown in

Re:Another rambling bullshit summary

[quantumghost]

The real usefulness of a technique like this (as I understand from the sepsis researchers that I've interacted with), is improved antibiotic stewardship -- preventing overuse of antibiotics and reducing the time the patient need be treated.

Please see my line about "De-escalated" - that is antibiotic stewardship in action.

These rapid assays can provide a better and more timely means to monitor a patient's response to treatment.

Again, where can this be applied in a "noisy" real world scenario? The researcher took an already cultured (i.e. purified) sample to prove the concept. I am trying illustrate the complexity of translating this into a real world application.

The talk about early diagnosis and saving lives is simply a lot more sexy and easier to sell than antibiotic stewardship.

They are two sides of the same coin....the bug I treat today with the tightest adequate spectrum of coverage is the bug I do not have to treat with the big guns later, or worse, have the big guns fail.....The situation for my patients really descends into the realm of "life sucks" when we're having the discussion "if we try this antibiotic I can treat his/her infection, but we're going to trash his/her {kidneys | lung | liver} in the process."

The grandstanding about this particular application aside, though, the technique itself is interesting.

No doubt that this is interesting, I again question where this is going to be useful....The best I can see is using it in place of the traditional sensitivity step - replacing the Kirby-Bauer test...but this is going to save hours. I guess that's going to be the "big saving"; still every hour I gain is something. Hopefully they can make this test cost-effective so we can actually persuade the bean-counters that it'll be worth it.

Re:Another rambling bullshit summary

[pepty]

Expense: Yes, those techniques are more labor and/or equipment intensive than culturing. I'm going on the assumption that there are cases where having the answer in 3 hrs instead of 3 days would drastically improve the outcome. How about this: If the only benefit to getting the result 20-70 hrs sooner was that patients ended up spending one less day in an ICU, the technique would save money if it cost less than ~$5k.

Knowing what to look for: You can differentiate A LOT of pathogens using a single set of primers, PCR, and RFLP analysis: combining speciation/strain ID and quantification is another matter, true, but that's why I mentioned qPCR. ELISAs can pick up which toxins and how much of each are actually present in a septic patient.

But I blanked on what would probably work best: array based techniques. Arrays on chips can identify ~100 different bacterial resistance genes, or simultaneously look for all of the most commonly feared biowarfare bactera and viruses. Bead arrays can simultaneously identify species/strain and quantify them. These techniques can also identify strains and viruses that are difficult to culture.

I read that as "Microsoft Tuning Forks"

[Anonymous Coward]

"It looks like you're trying to find a pitch!"

Re:I read that as "Microsoft Tuning Forks"

[oodaloop]

Funny, I read it as "Microsoft Turing Fords Kelp Detriment Affection of Antimatter" but I didn't want to bore anyone with my lack of reading ability.

Re:I read that as "Microsoft Tuning Forks"

[techno-vampire]

It's so good to see Ugol's Law in action! Once I'd realized that I'd misread the title, I wondered if anybody else would be willing to admit that they'd done the same thing. I feel so much better!

This is still in the research stadium

[vikingpower]

From TFA:

"It's a brilliant method," provided subsequent investigations confirm the researchers' interpretation of their data

I can, however, already hear the feet of the major pharmaceutical multinationals stampeding to get to Dublin....

Re:This is still in the research stadium

[Anonymous Coward]

My seismometers are picking it up too, this really could be quite a killer way to screen for new antibiotics.

The only problems I foresee are finding chemicals that cause bacteria to merely hibernate instead of just dying and breeding super-bacteria.

We could become accustomed to quickly finding new antibiotics to replace ones that have stopped working, that we end up with an Über-bacterium that's that's resistant to everything because we used up everything.

Re:This is still in the research stadium

[pepty]

This sounds like it may well be an interesting technique, but probably more for niche research uses than screening antibiotics. You could tell those e. coli cultures were full of live or dead bacteria just by swirling the test tubes and eyeballing them - ampicillin lyses the cells and makes the cultures look different. There are plenty of good, fast ways to screen for antibiotics that can be done with hundreds or thousands of samples at a time, as opposed to dedicating an atomic force microscope to each sample.

Tuning Fork Detectors?

[Anonymous Coward]

Do tuning forks detect vibrations? I thought the whole point was to produce vibrations.

Re:Tuning Fork Detectors?

[Anonymous Coward]

Do tuning forks detect vibrations? I thought the whole point was to produce vibrations.

Works both ways, years ago there was a tool that was a row of tuning forks or vibrating reeds. The operator watched which ones vibrated and that gave a rough idea of the spectral content of the vibrating mass that the tool was in contact with.

Entering the hospital may be first mistake

[Bearhouse]

Since you're most likely to contract a hard to cure infection....in hospital..

Re:Entering the hospital may be first mistake

[Okian Warrior]

Medical error ranks third among causes of death in the US.

Estimates of risk vary depending on which complications are counted, but it's always in the top 10. Any trip to the hospital results in a 1 in 300 chance of dying from medical mistake. For comparison, your chance of dying in an airplane accident is 1 in 10,000,000 per flight.

A rational plan would spend time and effort where it will do the most good. Instead of inventing new cures and treatments, perhaps we should be looking into ways to make our existing process safer?

For comparison, the risk of death by medical error is higher than the risk of death from diabetes. I'm not saying that diabetes research should be halted, but shouldn't higher risk factors be addressed as well?

Re:Entering the hospital may be first mistake

[slick7]

Since you're most likely to contract a hard to cure infection....in hospital..

After 5 return visits to an un-named hospital, notorious for infections, I used MMS - Master Mineral Solution - and have yet to see a recurrence of the infection two years later. The hospital insisted on a 5 day, intravenous, in hospital, course of treatment followed by an at home one week oral course. Within weeks I was re-hospitalized worse than before. To boot, they used IRS to get their money taken from my tax return, how nice. Talk about a monopoly, infect, treat, re-infect, re-treat, etc, etc, etc, yet never admitting the infection issue n the first place.

Re:Entering the hospital may be first mistake

[Anonymous Coward]

Quite true.

After an episode of acute aspergillosis resulting in a ruptured abscess in my lung and landing me in the hospital for almost 3 weeks, I hope to never spend time in a hospital bed again. Major lung surgery and the associated pain was nothing compared to the horrible food, uncomfortable bed, and lack of sleep due to constant nightly checking of vitals and blood draws. Given the inept handling of caring for the ~12" incision on my side (running from the back to the chest) I'm quite surprised I never caught a secondary infection. Maybe I'm one of the lucky people who didn't develop problems with IV of broadspectrum antibiotics twice daily.

Re:Entering the hospital may be first mistake

[Anonymous Coward]

After 5 return visits to an un-named hospital, notorious for infections, I used MMS - Master Mineral Solution - and have yet to see a recurrence of the infection two years later.

Hospitals aren't perfect and your solution is snake oil? Paying someone money to let you drink bleach isn't exactly smart.

Re:Entering the hospital may be first mistake

[Anonymous Coward]

The "EHR" is working on it, though whether it will succeed or not is anyone's guess. Maybe after everything is barcoded out the wazoo with 50 things to beep for each step of the hospital stay, they'll be able to get your kosher bacon cheeseburger mostly correct. One of these bracelets has the barcode that I need to beep to get the right baggie to put on the IV stand (that needs to also be beeped) before I can beep the needle and finally beep the valve to open it to the correct drip rate.

Fighting against the forces of change are the doctors who are fighting tooth and nail to keep their illegibly scrawled instructions and verbal orders shouted at the nurses who aren't in earshot. Meanwhile, half the doctors don't agree on which side of your body is "left" and are too entrenched to consider standardizing. Then there's the nurses, who don't want to have to fiddle stuff into the right place in the computer when they used to just write your penicillin allergy on the corner of the chart and give it a green sticky so everyone knew. Why can't the computer just have a green sticky to keep people from prescribing penicillin? Finally the patients who throw a shitfit about impersonal care if the doctor so much as glances at a computer screen looking up resources about a disease rather than memorizing the entire body of medical knowledge for every obscure problem.

Re:Entering the hospital may be first mistake

[sjames]

The snake oil may not be a great idea, but it's inevitable. Consider, where else are you expected to pay the costs to correct the service provider's mistake? If the mechanic screws up your engine, he's liable for the repairs. if the plumber floods your basement, he's on the hook to make it right.

So why is it that when the hospital staff infect you with unwashed hands (REALLY! There's a problem getting doctors and nursed to wash their hands!!!) why do they think it's on the patient to pay the huge bill? They make used car salesmen look like paragons of virtue.

Re:Entering the hospital may be first mistake

[neonsignal]

While I don't dispute that there can be improvement in hospital processes, many of these people who died were already at risk. If I have a heart attack, a 0.3% risk of dying through medical error at the hospital is nothing compared to my risk if I don't go to the hospital. What was the average risk of death had a patient not undergone a hospital procedure? Maybe there are classes of patients who should not undergo procedures! It isn't terribly informative to quote such a statistic without any context, and without giving details about the original illnesses and the common errors involved.

It's a bit like saying that many people die in their bed. A true fact, but not on its own enough to warrant an inquiry into bed safety.

Re:Entering the hospital may be first mistake

[slick7]

If it's "snake oil", then why does it work? Read the literature. You drink chlorine, flourine and whatever else is in water and yet, you chastize me? I suppose you believe that hemp oil is "snake medicine" because the petrochemical / pharmaceutical industries say so, how sad. Ayervedic medicine and Traditional Chinese Medicine have approximately 5000 years each in written records and each of these countries has over 1 bbbbbbillion people. Granted, Western medicine excels in trauma and emergency procedures, yet there is lacking in chronic issues. I do not believe in taking unnecessary chances, yet too many surgeries, too many antibiotics, too many doctors saying "It's not my fault" and having their signatures on orders, hands on scapels makes me want different protocols that are not so invasive. Somewhere I remember a saying that goes "Physician, heal thyself."

Re:Entering the hospital may be first mistake

[sjames]

The AC clearly believed it to be that and seemed unlikely to be convinced otherwise. For such a person, the best hope is to remind them that unless western medicine can truly make a better offer, people will turn away from it and to the things he/she clearly disapproves of.

I am not familiar with that particular product, but I am familiar with a variety of traditional and alternative medicine. I have found some that is ineffective and some that is stunningly and even inexplicably effective.

Certainly I tend to take any pharmaceutical claims with a few grains of salt these days.

NGS sequencing of ribosomal sequences is faster

[gringer]

An RNASeq run, either targeted to the ribosome or total (given that rRNA takes the lion's share) is a little bit quicker than culture, as long as the bioinformatics side of it is appropriately set up (e.g. massively parallel mapping, and automated count summarisation).

Sample preparation will take a few hours, and there are sequencers that will get results out in a few hours -- the mythical Oxford nanopore sequencers will speed both of these things up as well.

Actual Article Summary

[TheSwift]

"We made a tiny bar that vibrates when it's surrounded by bacteria! It stopped vibrating when the bacteria were given antibiotics and we think this means the bacteria were dead. We don't know why it vibrates and currently we have no way of telling the difference between different kinds of bacteria."

Cool technology, but keep your pants on. This has very little application for a very long time.

Royal Rife

[transporter_ii]

They should proceed with caution. They could end up quacks at any time. The famous Royal Rife machine used vibrations to kil bacteria. And here it is, all these years later, and it turns out bacteria *does* vibrate:

Rife also reported that a 'beam ray' device of his invention could weaken or destroy the pathogens by energetically exciting destructive resonances in their constituent chemicals.[4]

MLS

[Anonymous Coward]

Microbial susceptibility testing is done by Medical Laborotory Scientists. In the US we are credentialed by the American Soceity of Clinical Pathology. Contrary to what "House" ( and for you old folks "Quincy") has led to you believe, most doctors never set foot in a lab.

Omg I'm not awake yet...

[YoungManKlaus]

read "Microsoft" the first time when I saw the news in reader and was "Wtf? Microsoft Tuning Fork? Parsing error!" and now after opening the story did the same thing again ;)

Live/dead staining

[the.original.drg]

There is already a fast way to tell if bacteria are dead or alive; it's called live/dead staining. Basically, it stains living cells one fluorescent colour and dead cells another. You can then look at the sample under a fluorescent microscope or with a flow cytometer to quantify the amount of killing caused by the antibiotic.

More direct measurement of bacterial metabolism

[Anonymous Coward]

"Rushing" (which, as you say, is sort of impossible) would be concluding "no growth" (antibiotic effective) too early.

However, growth to visible cultures is composed of hundreds of generations, and if you had a more sensitive detector of bacterial reproduction, that didn't have to wait so many generations, you could reach colclusions a lot faster; limited primarily by the drug uptake rate.

They claim they can detect bacterial metabolism directly. So for bactericides, at least, they don't even have to wait one generation to detect results.

I suspect that bacteria could be classified crudely using some variant of flow cytometry, and then you could test antibiotics against each group.

Re:More direct measurement of bacterial metabolism

[quantumghost]

However, growth to visible cultures is composed of hundreds of generations, and if you had a more sensitive detector of bacterial reproduction, that didn't have to wait so many generations, you could reach colclusions[sic] a lot faster; limited primarily by the drug uptake rate.

Hmmm....me thinks you should look into your math...:-) (I'm being purely humorous at this point, not meant to be mean, but with real math)

doubling time is about 20 min in ideal circumstances so 100*20 is 2000 min or ~ 33.3 hrs

100 generations == 2^100 bacteria or 1,267,650,600,228,229,401,496,703,205,376

each bacteria weights about 9.5^-13g

so total bio mass is about 13,343,690,528,718,204g or 13,343,690,528,718kg or 1.3e13kg

for reference, the earth weighs...5.9e24 kg, (moon is merely 7.3e22, the USS Iowa battleship is about 5.2e7kg, a supertanker is 1.1e10 kg)

For a gram of bacteria (that's a lot!) only takes about 40 generations under ideal circumstances....now I'm ignoring lag and standing phases and focusing on the log phase just to give you an idea of the order of magnitude.

They claim they can detect bacterial metabolism directly. So for bactericides, at least, they don't even have to wait one generation to detect results.

Well, there's still the lag for them to grow to culture, again its the in vivo milieu that needs to be discarded.

I suspect that bacteria could be classified crudely using some variant of flow cytometry, and then you could test antibiotics against each group.

Looks like its been done. Didn't read too far in, I suspect it is not ost effective tho.