Hacker News new | past | comments | ask | show | jobs | submit login
Hospital’s iced water machine purifier stripped out chlorine, killing 3 patients (arstechnica.com)
82 points by rbanffy on March 8, 2023 | hide | past | favorite | 86 comments



> Whole genome sequencing ... revealed almost perfect matches of their M. abscessus isolates

How cool is it that in 2023 you can sequence samples from 4 patients as well as bacterial cultures found in separate ice machines and water sources to identify the source of an infection. Almost 30 years ago, when the first bacteria was sequenced, or 20 years ago, when the human genome project finished at a cost of $2.7B, or even 10 years ago, when sequencing cost more than $10,000 per run, this would have been unfathomable. But today it's a $600 routine expense, and it's perfectly reasonable to perform a dozen of them in the course of a large study.


Mycobacterium are no joke.

I got very severe sinus infection post sinus surgery from a incredibly rare Mycobacterium. (as far as I know I'm the only person in the world documented to have this bug in my sinuses)

It took almost 9 months to beat, 6 months with 3 types of antibiotics going at once.

The aggressive antibiotic caused severe neutropenia and put me back in the hospital for 16 days.

My infectious disease specialist basically equated treating mycobacterium infections similar to chemo.

Glad this episode is behind me, but never forget that tap water is not sterile.


>> (as far as I know I'm the only person in the world documented to have this bug in my sinuses)

Sort of like winning the lottery but in reverse. Sorry you had to go through that. Not just a reminder about tap water, but also not to have sinus surgery unless you really need it. I've heard from too many people that say even if it helps for awhile, the polyps usually grow back. I wonder if the success rates have improved recently.


Highly recommend skipping sinus surgery if possible... and definitely don't get surgery if you have an active infection regardless of what the experts say... until you know EXACTLY what that infection is with certainty.

After my first surgery my polyps were back worse than they started in just 3 weeks after the surgery (due to the infection... which hadn't been detected yet).

The doctors decided it was something unique with my immune response and said that the mycobacteria culture taken during the first surgery was a false positive. 2nd and 3rd opinions said the same thing!

Decided to have a 2nd surgery and mycobacteria came back in the culture again and they once again said it was a false positive/lab contamination... which is hard to prove until the culture has reached maturity.

Lab came back and it was indeed mycobacteria.


Stories like that kept me from getting surgery to correct a deviated septum. I have been able to avoid it up to now. The key that corrected most of my sinus problems was nasal rinsing with salt water. Always try surgical alternatives first.


If you're doing this. 100% boil your water and make sure your neti pot / nasal rinse tool is sterilized. Also if you are someone suffering read this study and try it: https://pubmed.ncbi.nlm.nih.gov/31560120/

Honestly it is my go-to when I have a really bad flare up.


I don't need to do it very often, but boiling the water seems like a real chore. (I know, call me lazy.) From my understanding, the risk of bacterial infection is extraodinarily rare, although I try to use distilled water.


Isn't a deviated septum a mechanical problem in the nasal passages? How does rinsing help?


True. Most people have a deviated septum. Just a question of degree. For me, salt water rinsing helps keep the tissue moist. When the tissue gets dried out, sinuses don't drain.


Its amazing how little they can know about the one thing theyre supposed to know.


> never forget that tap water is not sterile.

Which isn't a problem in most cases, but in this case it was also a case of an improperly cleaned machine; it apparently relied on some chlorine in the water to limit bacterial colonies. I think that would be fine overall, but only if the machine was emptied and thoroughly disinfected regularly. I can imagine a machine like that would also be a source of legionella.

Over here, after a major legionella incident [0] got 318 people sick of which at least 32 died, they're paying a ton more attention to it with things like public fountains, but they also seal firehoses nowadays - they would on occasion be used for utility like hosing things down, but then be unused for a long time, causing legionella and other nasties to build up in the room temperature water contained in the hoses.

[0] https://en.wikipedia.org/wiki/1999_Bovenkarspel_legionellosi...


>>The purification systems inadvertently stripped chlorine from the municipal tap water, allowing bacteria normally found at low levels to flourish and form biofilms inside the machines.

What is with all the hatred for Chlorine? Putting it in water is one of the great lifesavers, right up their with cooking and antibiotics. I hear people tell me about its health impacts, that it is "going to give cancer" and other crazy stuff. Well, as a former competitive swimmer, if chlorine does any harm whatsoever I am a write off. Having soaked in chlorinated pools for thousands of hours, and no doubt also drank in a not-small amount of that water, any hypothetical health impacts would have ended me decades ago. Stop filtering it.


Two major reasons IMO:

1. Chlorine tastes nasty in water. (The correct solution IMO here is to use chloramine instead.)

2. Chlorine removal may be a side effect. Carbon filters are used to remove all kinds of nasties and/or unpleasant tastes and odors. And they remove chlorine. Carbon-filtering without removing chlorine may be difficult or impossible.

> Well, as a former competitive swimmer, if chlorine does any harm whatsoever I am a write off.

Research is mixed, but the actual danger to swimmers might not be the chlorine per se. There is some (conflicting) evidence that nitrogen trichloride and/or dichloramine exposure increases asthma risk in swimmers and in people who work near indoor swimming pools. They also smell terrible — the nasty “chlorine” smell, especially around indoor pools, is dichloramine and nitrogen trichloride.

Fortunately your tap water is unlikely to meaningfully expose you to these. You need a nitrogen source and excess chlorine, which should not exist at the same time in any substantial amount in tap water. A chlorinated swimming pool plus sweat and/or urine is a different story.


In other countries they add fluoride to tap water, which I'm pretty sure is worse; it causes fluoridosis (minerals being stripped from teeth) and may cause behavioural differences (the science is not conclusive if you google it, one of the more batshit headlines says "A chemical used to fluoridate the drinking water of 150 million Americans may foster violent behavior and cocaine use in some of those who drink the water")


They are not specifically filtering chlorine, they are just filtering the water for taste.


>> What is with all the hatred for Chlorine?

Ignorant people.

Less dickishly - people incapable of assessing risks as a spectrum alongside other risks - combined with a giant echo chamber in everyone's pocket.


Chlorine tastes terrible. And people want to drink water, not chlorine water.


It's the same people who think they need to filter water in places like America or Europe. I drink chilled tap water. Tastes great and there isn't a problem. You know what tastes gross to me? Filtered water.


You need to travel more. Many places have really bad tasting water. It might be safe but it's basically undrinkable without filtering.


I've lived extensively in America (12 different states), Canada, the UK, Ireland, Spain, Germany, France, Italy, Greece, Japan, Korea, Australia, and New Zealand.

The tap water tasted fine in everyone one of them.

¯ \ _ ( ツ ) _ / ¯

The most disgusting water I taste consistently is from a plastic bottle.

In your defense, there's a big difference between room temp, tap cold, and chilled water. At room temp there is a flavor of water. Chilled with ice or a fridge mutes flavors.

If you did a blind taste test of chilled tap and "filtered" chilled water you wouldn't be able to tell a difference. Consequently, if you did the same test at room temp, you'd find both to have bad flavors.


I hate cold water (cool water is nice though). So that doesn't work for me.

Also, when you fill a pitcher of water and let it stand, the chlorine evaporates, so you don't actually like chlorine - you always make sure to remove it :)


We've got a water filter for one of our cats, the de-mineralization really does a number on the flavor. Some of the best water I've had came from a tap at a campsite close to a water source. Said water source was drinkable too, straight from the river.


"in places like <contintent> or <other contintent>"


Impressive root cause analysis. I wish all hospitals were this thorough and open.


Harvard better have their RCA’s in order. (They are affiliated with Brigham and MGH).

Completely unrelated other harvard medical news, their resident union is getting some major concessions from the administrators right now. That’s pretty cool. Pay at some of these progressive institutions is completely inadequate.


Sounds to me like a dirty ice machine was to blame and not the water.


Agreed. Ice machines aren't supposed to depend on chlorine in the water to keep them clean. Commercial ice machines need to be cleaned and sanitized every six months in normal settings, and you'd probably expect a hospital to do it even more frequently.

The article says:

> Although hospital records indicated that the machines had been maintained and cleaned per the manufacturer's instructions, the researchers noted visible bacterial biofilms on the inside components of the machines.

But it seems like a hospital should have its own standards for cleaning and disinfecting, not relying on manufacturer cleaning instructions for a non-medical device being used for patients with highly elevated risk of infection.


One of the food safety rules of travelling in rougher places, where only boiled or bottled drinks are safe enough, is to never take ice in drinks. I suppose we can now generalize to wherever maintenance looks shoddy.


> Commercial ice machines need to be cleaned and sanitized every six months in normal settings

It makes me wonder how many (or few) hotels keep on top of this kind of maintenance.


I wonder if they made assumptions based on the water filter that also did UV disinfection.

But, any source of food and water should be subject to frequent testing, it should have flagged up much sooner than this. This also applies outside of hospitals, the dodgy looking ice machines in e.g. hotels probably don't get tested at all.


The ice machine would not have gotten dirty(bacteria growing) if the water had chloramine in it. The article says the service and maintenance were performed correctly according to manufacturers specs.


I worked in restaurants all through college, you absolutely must clean the ice machines frequently. Sounds like they assumed they were self cleaning or just were not cleaning them often or thoroughly enough.


Makes you wonder if the "cleaning according to manufacturers specifications" simply wasn't correct and would have prevented this as well.


Seriously doubt that machine was being scrubbed down weekly/monthly. Hah


The breadth of cases with a DNA match suggest it wasn't cleaned thoroughly enough to kill the colony over at least 2 years


Also the fact that biofilms had developed is evidence to further support that. If you wipe a surface that’s covered in a biofilm it’s going to be very visible that it’s filthy. This is just pure negligence.


You might not be able to see the surface in question.

Here’s a manual for an ice machine:

https://www.webstaurantstore.com/documents/pdf/hid_manual.pd...

See page 19. I don’t know whether 10 minutes of exposure to 100ppm sanitizer will kill a biofilm adequately. Even “ppm” is vague. 100ppm of (active) hypochlorous acid is quite high and ought (IMO) to sterilize things effectively. But the recommended sanitizer is Stera Sheen Green Label, which, like most powdered sanitizers, is based on dichlor. 100ppm free chlorine worth of dichlor also adds about 90ppm of cyanuric acid, and cyanuric acid buffers hypochlorous acid, such that 100ppm FC + 90ppm CYA is a much worse disinfectant than 100ppm FC by itself.

Also, what disinfects the tubes between the filter and the reservoir?

(Why isn’t chlorine dioxide used for this purpose more often? It’s supposedly much better than chlorine for killing biofilms. Possibly less damaging to plastics as well.)


To be fair, it was likely on the inside. But still, it should be flushed out (and tested) on the regular.


Semi-tangentially related story. I was asked to be responsible for a water purifier in a science lab. They needed very pure water for sensitive NMR experiments. Well, I forgot to put the chlorine tabs in for a few months after they finally traced the contamination back to me (all the NMR spectra looked messed up).

It's one of the reasons I returned to being a computational biologist.


How does this apply to our water filters inside our refrigerators?


It's a risk if you don't swap the filter regularly as recommended by the manufacturer, and everything downstream of the filter (tubing etc) is at elevated risk for biofilm buildup when the filter is stripping out chlorine/chloramine.

It's an even bigger risk, generally, with pitcher/reservoir type systems (e.g. Brita or Pur) that require manual fills, given that they get a lot of environmental exposure. If you use a Brita pitcher for a while and then leave the filter media and pitcher somewhere at room temp for a couple weeks, you can often see evidence of fungal/microbial growth popping out of the bottom of the filter cartridge. It's pretty gross. I've also had a relative end up in the hospital with a pretty severe amoebic infection, which investigators traced back to her Brita filter cartridge she had neglected to change for some time.


Sure, but as long as you change the filter every 2-3 months and keep it in the fridge there's nothing to worry about.

And if you leave almost anything out at room temperature for weeks kept moist you're going to see gross growth. It's nothing specific to a filter.


Part of the reason I purposefully bypassed the filter on my fridge. The tap water is fine and doesn't need filtering but an old water filter can cause all sorts of nastiness.


Well the main point of a Brita filter is to remove the chlorine so you tap water doesn't taste like it was mixed with pool water.

But this is why you replace your Brita filter every 2-3 months and wash the container.

Refrigerator filters generally need to be replaced every 6 months.

It's both because they lose effectiveness and because bacteria builds up inside them.


Ice machines are open to the air, water lines in a fridge are not.

And your ice maker is normally always cold.

But clean any parts that do get warm.


Cheap filters don't do much and the more expensive ones don't remove chlorine.


It's really shocking how ineffective things like pitcher filters are when compared to a quality cartridge filter. The pitcher filters really do practically nothing.


Hard disagree. My tap water has a strong chlorine taste, I hate it. In my Brita pitcher, it's totally gone. Just tastes like water again.

It's the total opposite of "practically nothing", it does exactly what I bought it for -- effectiveness 100%.


There's no question that Brita filters improve water taste by reducing chlorine/chloramine levels, and they will remove relatively large particulates, but they pale in comparison to proper pressure-driven cartridge filter systems if you start looking at comparisons for other contaminants.


Sure, but for people who are getting perfectly safe water from their municipal water supply, and just want to get rid of the chlorine taste -- that's where pitcher filters work perfectly. You don't need anything else.


It’s still probably not doing anything, chlorine readily evaporates out of water left in a pitcher. It’s a common technique used by indoor growers of certain plants - leave a bucket out for a day.


I can taste the water 5 minutes after it goes through the filter. The chlorine is gone. (Which is not the case if I pour it into a glass.) It has nothing to do with how long it sits around for.

I find it strange you're insisting filters don't work when one can tell from taste that they clearly do. And scientifically, activated carbon absorbs chlorine -- that's not a myth. So everything checks out.

Pitcher filters work for chlorine.


Sure, they work for chlorine, but there are plenty of other issues (arsenic, PFCs, other VOCs, glyphosates, TTHMs, lead for non 'Elite' Brita filters) that Brita filters are woefully inadequate to address.


Are any of those issues for something like the NYC municipal water supply?

I think most people buy filter pitches to fix the taste of water. Not for health or safety reasons.


Actually, NYC is a major concern in terms of water quality because it's very common for buildings to have old plumbing that releases trace heavy metals, and many buildings use rooftop water reservoirs that are a common source of organic contaminants (and occasionally cause building-wide outbreaks of things like legionella).

Similar issues exist in most 'older' US cities. I started becoming worried about my water filter efficacy after moving to downtown DC and learning that a significant portion of the municipal water supply's last mile piping contains lead, and that while the municipal water is safe at the source, water in DC housing is often heavily contaminated by lead and other compounds resulting from the use of extremely aged pipe infrastructure.


Yes it will. Any carbon filter will remove chlorine, it's impossible for it not to.

Cheap filters improve taste, not water quality. And in most places that's all you need.


Thanks for clarifying.


They are dirty and risk for your health. Unless you clean it carefully every week.


“ Our cluster demonstrates the risk for unintended consequences associated with systems designed to improve hospital water”

Sounds more like incompetence especially you are in a hospital setting


Hospitals are one of the dirtiest place to be. Somewhat better for critical care patients, but once you start getting better you want out of there as fast as you can.


> Both carbon filters and ultraviolet light are known to decrease chlorine concentrations.

Shouldn't the UV light also kill bacteria, though?


A friend of mine is technically on a different industry (water for semiconductor use) but he said that UV light is not really that effective, especially if a biofilm has grown. Even if everything works correctly and the setup killed stuff, it doesn't remove their corpses (only filtration can do that).

Edit: however it seems that the machines were maintained, so it might not be about this specifically.


great, so a vaccine then? /s


Sure, but in an ice machine the water is exposed to the air which also carries the bacteria.

Ice machines are notoriously filthy because they have tons of tiny spaces that never dry. I remember looking inside an ice machine at work and the inside was covered in a thin film of slime.


This is the kind of stuff that us laymen (relatively speaking, in the context of microbiology) find puzzling.

Another thing I never quite understood was how bacteria thrive in purified water anyway, to the point where they can form biofilm. Where do they even get the necessary nutrients?


I'm a registered nurse, and while not an expert in microbio, I have more training in it than the average bear.

It doesn't take much for bacteria or fungi to start growing on previously filtered water once it's open to air. Just take a glass of filtered water and leave it out for a few weeks. Some glasses won't grow anything. Others will have all sorts of things floating on the surface.

Pasteur is credited for formalizing this knowledge when he figured out that if you used a crooked necked flask to store media, it reduces the chance of airborne contamination.

Once bacteria form a biofilm, it could make it more difficult for any chlorine or UV light to disinfect or sanitize the surface. The bacteria have formed a sort of protective bubble.

Mycobacteria are particularly good at resisting various antibiotics because they have a thicker than average cell envelope. Mycobacterium tuberculosis is another bug in this family.


Air maybe? Deposition of dust? Mostly dead particles of skin and so on? Or maybe clothing?


I've been looking into these home water purification systems. A good one will strip pretty much anything out of the water, so as a window cleaning company that used their own water to clean the glass put it, its dead water, in theory nothing can survive in it. Now to touch and feel that "dead water" it was like handling liquid silk without the thickness or resistance. It would be like having the most luxurious bath or shower you'd ever have without any cleaning products like shower gels, bath soaks etc. It also means the amount of cleaning product used, like washing powder for clothes, shower gels, dishwasher product, goes down and becomes more effective as there is no mineral to interfere with it. Technically pure water is an acid.

Anyway I have not found any suitable answer for cleaning the water pipes after these strip everything out of the water filter systems, and whilst it shouldnt be a problem if you have copper pipes throughout the house because copper kills most bacteria and some viruses, hence copper water vessels from the past, if you have the new plastic flexible piping in the house, cleaning will almost certainly be required.

Penicillin based antibiotics when they are broken down in the liver become penicillamine and this causes the body to release lots of copper into the blood stream, so how much is that having an antibacterial effect on the body? Plus copper is also needed for interleukin-6 which kills alot of cancer, so if doing lots of penicillin antibiotics, your risk of cancer could go up if your dietary copper intake is not high! Something the medical professionals dont tell you!

So something like what the pubs use to flush their pipes leading up from the barrels and kegs should do it for cleaning the water system pipes after the water filter, but like I say, I havent found anything satisfactory from these household water supply filter people. Certainly copper pipes are best though, and if doing some DIY, dont leave any flux on the pipe because in damp cold conditions, condensation reacts with the flux and it causes tiny pinholes to form in the pipe. A good plumber will clean their pipes down to stop this from happening otherwise you get minute water leaks that can get worse very quickly.

UV-C is the main disinfecting UV light source, but you dont want to go near it because UV-C destroys pretty much anything which is why most of it is blocked in the upper atmosphere.

I know hospitals use it to clean wards and rooms, but it will destroy materials much more quickly, ideally hospitals would clean a ward and room like normal to remove ground on dirt and then use the robots that move around shining UV-C all over the place.


The standard procedure for fixing similar issues with well water is standard household bleach + time to sit + a lot of running water to move everything through, FWIW. Look up instructions for "chlorine shocking well water". You need to calculate the volume/concentration correctly, but there are lots of guidelines to make reasonable estimates of pipe + well volumes.

FWIW, the main sanitizer used for kegs is the same thing used in restaurants for the standard "three sink" cleaning system (last dip is in sanitizer). Star San is the main brand name. It's a mild acid and it's both cheap and readily available. It works basically by denaturing proteins, which any acid will do, but the specific one is chosen to be harmless and relatively tasteless as well as easily clinging to surfaces to give extra contact time.

I'm not sure it would be a good idea to use it, though, as it might eat away at some plumbing fixtures (e.g. galvanized pipe).

If you did, you'd want the non-foaming variety (which is not called star san... Saniclean, I think? Same brand, either way).


>Technically pure water is an acid.

https://finance.yahoo.com/news/insane-golden-chamber-contain...

scroll down to the "Terrifyingly pure water" section for a fun read on how fun really pure water can be.


Just need one to survive for it to multiply somewhere later down the pipeline. It can also be introduced other way I guess


The UV bulbs need to be replaced periodically. This was most likely negligence.


Article says the machines were maintained.

It doesn't matter if you purify water then expose it to room air. It's going to get contaminated again.


The 'machines' may have been maintained, it doesn't mean the filters installed on the waterline before the machines were. Properly functioning UV filters kill bacteria, there would be no way to introduce bacteria into the system from the waterline if the UV filters were working adequately (assuming the upstream carbon filter wasn't acting like a reservoir exceeding the specs of the UV filter also due to lack of maintenance).

Since the other machines elsewhere didn't have these filters, it's safe to assume nobody had them on the maintenance list and serviced them.


It's an ice machine. Have you seen one? The trays the ice are made in are open to room air.

No conspiracy theory needed.


There was a maintenance record. Not that the machines were maintained. Having visible biofilm I don’t think there was done anything on them.


The machines were operating normally when the investigation was done. Why assume the UV light didn't work and maintenance reports were faked when the easier explanation is the machine was just dirty?


This seems a bit of a conspiracy thought. What if the machine was designed poorly or something extraordinary happened causing the film to build.


Would bypassing the water filtration system for a few days a month help with this?


How does UV reduce chlorine? It's a freaking element. It's especially weird inside an ice machine, where the chlorine can't selectively evaporate or something, which is my first guess at a mechanism in open air.


The UV doesn't reduce chlorine, it kills bacteria. There's also a filter, and that's what removes the chlorine


That's what I would have thought. But the article said:

> Both carbon filters and ultraviolet light are known to decrease chlorine concentrations.

Seems crazy, right? That's why I'm asking.


You expect scientific accuracy from journalists?


UV purification systems can also remove chlorine.


Title is incorrect, this wasn't the hospital's overall water purification system, but 2 devices with purification systems.


Thanks - we put 'iced' in the title above to clarify that it was more specific.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: