Hacker News Reader

Comment by klipt | Hacker News Reader

Scary story and I sympathize. I assume this is pretty rare though, which is probably why it's hard to treat in the first place (there haven't been that many patients to research better treatments on).

E.g. doesn't every mother getting an epidural while giving birth effectively have a lumbar puncture? And most of them turn out fine. Maybe the bed rest after labor helps.

The fact that CSF pressure is too high after patching makes sense, I guess the body ramps up CSF production to combat the leak. This probably also makes healing harder because there is constant increasing fluid pressure, which pushes fluid out, which makes any hole resist closing.

This suggests a better treatment might be actively tapering down the flow? E.g. instead of attempting to close the hole completely in one go, install some kind of controllable shunt and gradually reduce the flow over a longer time, then once reduced to nothing, close the hole?

View on HN

I'm a doctor training in anaesthesia. The idea with epidurals is to introduce local anaesthetic outside the compartment where the spinal cord is, literally epi-- --durally.

To sample the CSF, or to give a spinal anaesthetic, the needle needs to be introduced into the subarachnoid (one of the membranes that surrounds the central nervous system) space, but at a level below where the spinal cord has terminated. This is why they're done in the lower back typically.

Bending over is a totally legitimate way to perform them, and doesn't really alter the complication rate, while making the procedure easier. Also, there isn't really any realistic option to perform them in real time with imaging guidance. The author is right in that larger cutting needles are associated with increased rates of PDPH.

Hi there, thank you for the comment. Yes, bending over is a legitimate way to perform them, but some leak experts theorize that it can contribute to additional issues (e.g. an arachnoid bleb, which needs surgical intervention as EBPs won't suffice for treatment).

> instead of attempting to close the hole completely in one go, install some kind of controllable shunt and gradually reduce the flow over a longer time, then once reduced to nothing, close the hole?

That would only work if they can find the hole. According to the article, it often doesn't show up on imaging, and even if they know roughly where it is (as in the case of a lumbar puncture gone wrong), it's still incredibly difficult to pinpoint, and patching it usually involves trial and error.

It wasn't completely clear from the article, but it sounds like often they don't ever really pinpoint the hole, even when they are able to successfully patch it. The patching process seems more like "spray and pray": do some patching (she had 22 patches done, I think I read), and once it seems like the leaking has stopped, they assume they got it. Not sure if this is true, but that's the impression I got.

> That would only work if they can find the hole.

Well either that, or "spray and pray" to fix unknown holes, but deliberately make another small, known, controlled hole that can relieve pressure and gradually be shrunk until it's shut.

I actually knew someone who had a valve like that installed because her CSF pressure kept changing. Periods of high pressure were the most dangerous and led to rapidly deteriorating mental function—and eventually death.

The valve never really worked well, and she didn’t live long enough to determine why.

Retired neurosurgical anesthesiologist (38 years) here. I performed 500-1,000 spinal taps.

>... deliberately make another small, known, controlled hole that can relieve pressure and gradually be shrunk until it's shut.

This would make things worse. The pressure has already been relieved by the original hole with subsequent loss of cerebrospinal fluid.

Post-spinal/dural puncture headache results from NOT ENOUGH CSF surrounding the spinal cord and brain as a result of its leaking out: thus, the brain sags and positions itself on the base of the skull rather than floating, with subsequent headache.

For clarity, I had 4 rounds of targeted patching done, but I wanted to be clear that there were a total of 22 needles — it wasn't a high volume EBP, but smaller, directed patches with blood+fibrin.

Puncture leak sites rarely show on imaging, it's true. At Duke, Cedars, or other leak centers they have techniques for CTMs or DSMs that yield leak sites but that's usually for spontaneous CSF leak patients or post-surgical, not us puncture people. They're using a PCCT machine to find CSF venous fistulas at Mayo and at Duke, but again to my knowledge puncture leaks remain a needle in a haystack even on that quality of imaging.

Pretty rare, versus pretty hard to diagnose though? I'd be more surprised if everyone wasn't leaking a little and a lot of people are leaking enough that it would be noticeable if they knew to recognize it.

Re: your question I think the needle goes to different places.

> One large difference between a spinal tap and an epidural is the area that the needle is inserted. With an epidural the needle does not go into the spinal cord, but the epidural space just around the spinal cord. Spinal taps however, require the needle to be inserted all the way into and through your spinal cord. [1]

which at least matches my experience of needing a lumbar puncture for them to extract spinal fluid for testing. I've had them before, twice, never an epidural though. One leaked, and they didn't waste a moment doing a blood patch in urgent care (America). This entire post is a horror story, I would never have known to ask.

This person lost ten years of life and the rest of it irrevocably changed because some idiot didn't know something this basic. I'm just stunned, despite my intimate familiarity (lost the last three years frankly), it's just literally on the wiki for spinal tap headache treatments that you need to take action if they last more than 24 hours! [2]

If they hadn't thought to do a blood patch when I went to urgent care (which I had the privilege to do!) I'd literally be sitting here with probably the same leak now 13 years later. Malpractice, sorry, this is horrifying. It's one thing when it's hard to guess, I had a slipping rib that's apparently super hard to identify for sure and took those three years to be sure enough that just doing some PT wouldn't cut it. Three years. I'm not getting younger... I hope I at least get to be close to as in shape as I was three years ago.

This piece hit hard. I also spend days in bed just... tolerating... sometimes. Learning to be present and accept things that I cannot do anything about. I really hope they're able to go on walks in the park again.

[1] https://healthresearchfunding.org/spinal-tap-vs-epidural/

[2] https://www.mayoclinic.org/diseases-conditions/spinal-headac...

> This person lost ten years of life

Her lumbar puncture was in 2017, so not quite. But the leak came back in late 2018, and it seems like she's been living with it since then, which is incredibly awful. Based on her other issues (MCAS and EDS), diagnosed since the first leak, it does seem like she'll be living with this (and possibly worse) for the rest of her life.

At any rate, wherever you are on your healing journey, best wishes for further recovery!

Thanks, I missed that it was 10 years of being a nomad first and then this stuff. It said it so many times I lost track of time timeline!

I can very much relate to the meditation aspects. Sometimes I know that if I go to the Urgent Care the best they'll do is have me lay down for some hours until I'm okay to leave. I can do that at home, the hard part is letting myself believe everything is fine and that the best that can be done is to be present.

I am a retired neurosurgical anesthesiologist (38 years) and performed 500-1,000 spinal taps.

>Spinal taps however, require the needle to be inserted all the way into and through your spinal cord.

NO! Through the dura and into the subarachnoid space but NEVER into or through the spinal cord, which would likely result in permanent motor/sensory deficits and possibly paraplegia.

Thanks, the source seemed questionable.

I'm afraid I can't edit it, but thanks for correcting.

> It turns out that I have both a connective tissue disorder [Ehlers-Danlos Syndrome (EDS)], and a mast cell disorder, both of which make my situation more complicated.

...and predisposed her to getting a CSF leak in the first place. More details here: https://jodiettenberg.com/connecting-dots/

Reply on news.ycombinator.com