Lumbar puncture is a fascinating procedure. It is cheap, it can be done in relatively remote places, and it can be learnt easily given access to enough people who need it.
LP has an incredible role in the management of many complications related to HIV. I've heard stories about how there used to be 5 LPs done every day in VMH during the time when HIV was causing rampant destruction in Karnataka and India. When I was there, we would do about 5 in two weeks. Nevertheless, when a colleague asked on Twitter about CSF analysis, I thought I should write down some of the things I believe to know about Lumbar Puncture itself, especially in relation to its use in management of complications of HIV.
The first many LPs I saw were all done for spinal anesthesia in KR Hospital. Till then all I knew about spinal anaesthesia was a friend's description of the back ache he had post a "cool" hernia surgery because they had "poked many times for anaesthesia". I think I hadn't really thought about it till I was doing my anaesthesia rotation during internship. The first LP I did was also done during the same time - the "pop" and being in the space that you can learn only by doing. (If anyone thinks that all knowledge is codifiable like I do, here is what it feels like. Imagine there is a thick plastic layer laid around a piece of rusk. Imagine your needle piercing through the rusk and then splitting open the plastic layer. Now you are in the space.)
The first time I saw an LP done for diagnostic reasons was in the medical emergency ward of KR Hospital where a young patient with some sort of neurological condition was being pinned down to the bed by 4 people and the postgraduate resident was dancing with the needle along with the squirming patient. Despite the grotesqueness of the picture, I found it incredible that 20-40 drops of a particular fluid can be so valuable in diagnosis.
I learned the reasons when I was in VMH. There were many "spot" diagnoses we made using LP:
1) Perceived high opening pressure in an HIV infected patient with neurologic symptoms - we send for cryptococcal antigen and it is almost certainly positive. (Always use Cryptococcal antigen test. Indian ink looks fancy under the microscope when it is positive, but is not as sensitive)
2) High lymphocytes and proteins - you can keep your various tuberculosis diagnoses active. But even otherwise, you can't rule out TB ever.
3) RBCs and you can suspect sub-arachno... Just admit that you did a traumatic tap.
But LP was mainly used for ruling out the infections. It is very simple to miss CNS infections in HIV infected patients. For example they will come with vomiting and you will examine their mouth and see oral (and possibly oesophageal) candidiasis written all over it. But rather unknown to you, they might also be having cryptococcal meningitis.
It might be difficult to treat cryptococcal meningitis because Flucytosine is not something you find easily in India and therefore you are stuck with Fluconazole and Amphotericin B and good luck to you if you plan to give the latter in peripheral venous lines. (I'm not sure if the liposomal variety of Amphotericin B doesn't cause as much phlebitis). But cryptococcal meningitis is a diagnosis you do not have to miss, if you are doing LP.
It is a messy thing, but it is a life saving diagnosis. I've seen one patient die during the treatment, even though we were doing regular therapeutic lumbar punctures to reduce the intracranial pressure. But I've seen almost everyone else survive (including the case where I had to take PEP). I've also heard a very inspiring story from Dr Ramakrishna Prasad about a patient whom everyone else had given up on, coming back to life after switching over to the liposomal variety.
A (thankfully) much rarer thing is HIV CSF escape syndrome. Hearing about this for the first time is when I realized which peak of the Dunning-Kruger effect I was on. You see, the blood brain barrier is a real thing. And not all of the HIV drugs cross this barrier the same way (paradoxic?). And therefore there are patients who can have no virus in their plasma, but if you do a CSF viral load test you will have a real surprise waiting.
A not so uncommon thing which can be diagnosed through CSF is neurosyphilis. I always have to read the guidelines three times about when to use a VDRL test and how much to rely on it, but this is a test that we used to do as a protocol while doing an LP in HIV infected.
Things like gadolinium enhanced MRI are becoming more useful than CSF analysis in diagnosis of things like tubercular meningitis. But from what Dr Rahul Abraham once told a group of us about his experience with MSF in Bihar, lumbar puncture will remain with us till the end of the HIV pandemic.
No comments:
Post a Comment