Haemophagocytic lymphohistiocytosis

Haemophagocytic lymphohistiocytosis is listed as an L2 condition in Section 2.1.11 ("Haematological and Oncological Intensive Care") in the CICM Syllabus for the Second Part Examination (1st ed).  For the time poor candidate, the assessment can be covered well enough by the paper on the diagnostic criteria by Henter et al (2004), and for management, La Rosée et al (2019) present some recommendations (being reluctant to call them "guidelines"). One might rightly point out that no intensivist would ever manage this condition in a vacuum without a lot of input from haematologists and immunologists, and so it would make no sense to hang a high-stakes exit exam on the ability of trainees to recall exotic monoclonal antibodies used to treat it. On the other hand, one may point out that the ICU is the final common pathway for these patients, and though haematologists might occasionally turn up and sprinkle some etoposide into the broth, the coordination of their complex multisystem care will fall to the CICM graduate. As such, some SAQs on this have appeared in the exam papers, most notably:

• Question 18 from the second paper of 2021
• Question 24.1 from the first paper of 2020

Haemophagocytosis and states of macrophage activation

HLH is only the charismatic frontman of the haemophagocytic syndromes, a little-known boyband that contains a number of obscure conditions that no CICM trainee would ever be expected to name (see Janka & Lehmberg, 2014, for details). It will suffice to say that they all follow a common pathophysiological choreography:

• A trigger of some sort (usually a virus, malignancy,  intracellular bacterial pathogens like Rickettsia, etc) has the effect of activating CD8 T-cells by the activity of antigen-presenting cells, which present the lymphocytes with antigens.
• This should usually produce a population of cytotoxic effector T-cells. These effector cells then would normally express perforins and lyse the antigen-presenting cells, or induce apoptosis in them, to suppress further antigen presentation and control the magnitude of the T-cell response.
• If something happens to be wrong with this suppression step , the negative feedback loop does not have the desired downregulatory effect, and the antigen presenting cells remain free to continue activating lymphocytes, which means a large number of CD8 T-cells undergo a maladaptive hyperactivation (that can be detected by overabundance the soluble sCD25 surface marker).
• This enraged clone of T cells leads to the production of a whole range of cytokines, of which the worst is probably IFN-γ
• IFN-γ is a pleotropic molecule, but its most relevant action here is to stimulate macrophages, whose Janus kinases (JAKs) then mediate the synthesis of numerous cytokines, from whence all the clinical manifestations are seen.
• Specifically, the macrophages produce:
  • TNF-α, which causes:
    • fever by direct effects on the hypothalamus
    • hypertriglyceridemia by inhibition of lipoprotein lipase (Grunfeld et al, 1989)
    • ferritin synthesis by macrophages
  • IL-6, L-1 and IL-18, which cause fever and a raised CRP 
  • Plasminogen activators, which  cause coagulopathy and deplete fibrin
• For some reason not very clearly defined, this specific sort of macrophage activation makes them specifically hungry for red cells, which leads to the haemophagocytosis.

Epidemiology of HLH in ICU

On one hand, this is an exceptionally rare occurrence, with the entirely of the US recording only 16,136 admissions between 2006 and 2019, or 6-11 per 100,000. On the other hand, one can be fairly certain that the majority of these would have ended up in the ICU, making the intensivist a logical target for HLH exam content. This also opens us to criticism that we fail to recognise it when we see it: Okabe et al (2012) describe three cases that each presented with a different diagnosis and suggest that lots of things that look like sepsis might actually be HLP under the surface. Of course to distinguish the two is difficult, considering both present with fever and shock among patients who are susceptible to infection. Which brings us to:

Assessment of HLH in ICU

Acknowledging the difficulty in making the diagnosis of HLH in the presence of a strong suspicion for sepsis (or, frank actual sepsis), the Histocyte Society has produced a scoring system (Henter et al, 2004), where 

Meena et al (2020) suggested that having 5 criteria had a specificity of 97% and a sensitivity of 70%, which improved when the ferritin cutoff was increased to 1000 (which still seems somewhat low).

Now, readers might point out that this is largely a laboratory diagnosis, but the "assessment" syllabus vocabulary item in the CICM exams usually means one needs to think in terms of the history, examination and investigations. For something like this, a prepared set of thoughts is probably essential, as compared to something like (say) acute myocardial dysfunction, which lends itself much better to a physicianly outpatient-clinic vibe.

History

• Fever
• Fatigue
• Malaise
• Abdominal pain
• Oedema
• Ataxia
• Seizures
• Skin manifestations (rash, purpura)

Clinical examination

• Delirium
• Hepatosplenomegaly
• Features of coagulopathy (eg. ecchymoses)
• Bleeding
• Specific identifying features of some kind of genetic syndrome

Relevant investigations

• Obviously, those that are listed in the diagnostic criteria:
  • FBC
  • Triglycerides
  • Fibrinogen
  • Ferritin
• Useful tests that look for complications:
  • Blood film (fragmented red cells)
  • EUCs, CMOs, LFTs, LDH (organ system failure)
  • CRP and procalcitonin (because also possibly sepsis?)
  • Coags (because always coagulpathic)
• Looking for causes, which is admittedly academic:
  • Viral serology or PCR for: EBV, HSV, CMV, HHV6, HIV, VZV and parvovirus
  • Imaging to look for malignancy and splenomegaly (if you do not wish to palpate the abdomen)
• Gold standard
  • Bone marrow biopsy
  • Soluble IL-2 receptor alpha (sCD25) 

Management of HLH in ICU

With resuscitation of this state resembling the resuscitation of any distributive shock state (fluids and squeeze), the money in any exam answer is likely to be in the more juicy immunological therapies. This is a rare weird disorder and exam candidates who write extensive notes about their crystalloid choices and the gauge of the two cannulas they will insert are unlikely to score any marks with their efforts.

Specific management

Treat the underlying cause:

• Not all of the underlying causes have a treatment
• When something is clearly responsible, eg. an EBV infection or a CAR-T cell therapy infusion, the success of mananagement is dependent on your ability to defeat this primary trigger

And for the HLH:

• High dose dexamethasone: start with 10mg/m² BSA, so about 17-20mg/d (the choice is based on the better CNS penetration of dexamethasone; as HLH has nontrivial CNS effects, such as seizures and PRES)
• Etoposide 150mg/m² BSA (so, 250-300mg)- this is a podophylootoxin derivative which tends to target T cells and macrophages; the aim is to abort the cytokine storm
• Two weeks later, start cyclosporin A and methotrexate; mainly to prevent relapse
• Etoposide and pulsed dexamethasone will need to continue until the precipitating cause of the HLH has subsided, or until the patient gets a bone marrow transplant.
•  

And while waiting for these to have effect,

• Infliximab (a TNF-a monoclonal antibody) or anakinra (an IL-1 monoclonal antibody) can be used to control the cytokine storm.
• Dasatanib, a tyrosine kinase inhibitor, can be effective (it interferes with the JAK signalling in macrophages, preventing the cytokine synthesis and release)
• ATG (antithymocyte globulin) can be used to rapidly remove T cells from the circulation
• Emapalumab (anti-IFN-γ monoclonal antibody, used to silence the IFN-associated macrophage activation)