Tasty Morsels of Critical Care 028 | HIV in the ICU

18 Feb

Welcome back to the tasty morsels of critical care podcast. A meandering monologue through critical care fellowship exam preparation.

HIV in the ICU is becoming a bit of a rare beast as the ID docs seem to have it so well controlled that it becomes a chronic co-morbidity rather than the cause of their presence in the ICU. There are of course exceptions – typically in the poor soul in difficult circumstances, trapped in poverty and with no means to maintain stability in their lives. These patients often have fulminant and uncontrolled disease.

Primary HIV will not typically land you in the ICU but it is an important syndrome with a clinical appearance a bit similar to infectious mononucleosis. The key is to consider it and be liberal with testing. In particular patients are often most viraemic at this stage and can spread HIV significantly when they are very much unaware of their status.

The CD4 receptor is the primary binding site for HIV and over time these CD4+ T cells become very deplete and indeed this is one of the means of immunosuppresion in HIV. CD4 counts are rarely available out of hours but we do have the lymphocyte count as a reasonably well established surrogate. (Do you remember back in early COVID times the low lymph counts were considered somewhat predictive of SARS CoV2?)

CD4 levels <200 are generally considered to put people at risk of opportunistic infections. Though the risks and weirdness of the infections gets greater as the CD4 falls. In general at the higher levels (say 100-200) expect respiratory illnesses, at the super low (<100) expect to see funky things like crypto and toxo more commonly.

In terms of diagnosis, as far as I can work out (and I’m no virologist…), most tests to diagnose HIV are some form of antigen/antibody testing. The viral load that we see is used primarily to measure response to treatment.

These patients will often be on a lot of unpronounceable medications. The general rule is to continue them all if possible. Breaks in treatment can lead to resistance. However there are often problems with keeping this stuff going. For example:

  • no parenteral preparations currently available and of course, working in the ICU we find it hard to believe that drugs given by a non IV route actually do anything.
  • worse than this many are not crushable so even with an NG we mightn’t be able to deliver them.
  • absorption likely impaired by critical illness with poor cardiac output
  • ileus is common and further reduces absorption
  • feeds need stopped for administration – impairs nutrition
  • clearance may be altered by either hepatic (reduced flow, cirrhosis, acute liver injury) or renal injuries (reducing renal clearance)
  • interactions with other meds used in ICU changing effect of meds
  • All ART are hepatotoxic
  • the NRTI are famous for causing lactic acidosis (at least in exam stems…)

Finally it’s worth being aware of an entity called IRIS (Immune Reconstitution Inflammatory Syndrome). This occurs in patients who have relatively advanced HIV at presentation with very depressed CD4 counts. These patients have immune systems that have been culled of their senior and competent staff, somewhat like the hospital at 3am on the Sunday after Christmas. Lots of opportunistic infections have sneaked in and are not causing much of a fuss and are entirely unnoticed. Along comes effective antiretroviral therapy and all of a sudden there are legions of CD4s marching the corridors and uncovering CMV and crypto in every crevice. The newly reconstituted immune system gets excited and does what it does best and causes inflammation somewhat to the detriment of the body overall.

All that is to say that occasionally in the critically ill, newly diagnosed HIV patient, you may find your ID team making strong recommendations not to commence ART.


Oh’s manual Chapter 69

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.