Day #92 - "I don't know"

Wow. What a great case. It isn't often that I truly throw my hands up and say that "I don't know" but this was one of them.

But I hope the exercise in reasoning was useful for you. I think we demonstrated an approach that was safe, cost-effective, and broad in going through the history and physical exam.

Dr. Gold previously spoke about meningitis (bacterial) and that talk is available here.

In approaching a patient with multiple infarcts we need to consider the following differential:

• Emboli - Arising from the heart or great vessels. One possibility is subacute bacterial endocarditis, particularly given the history of night sweats. I think this patient needs blood cultures drawn with blind subculturing (endocarditis blood cultures) to look for unusual pathogens like brucella. I also think he needs a trans-esophogeal echo looking for a cardiac source including a bubble study to exclude PFO. This can also look at the aortic arch.
  
• Ischemia -- Either bad luck from uncontrolled risk factors, or genetic stroke syndrome like CADASIL, Fabry's or MELAS, or thrombophilia like antiphospholipid antibody syndrome, ATIII deficiency or hyperhomocysteinemia. I would exclude thrombophilia. As a last resort I would look into genetic testing. Night sweats don't fit with ischemia.
  
• Vasculitis (Night Sweats would fit with this too)
  
  • Primary:
    
    • Primary CNS angiitis
    • GCA/Takayatsu, polyarteritis nodosa (patient has Hep B!)
      
  • Secondary due to:
    
    • Infection:
      
      • Syphilis always needs to be considered in young patients who present with stroke like syndromes.
      • HIV can do this -- though he is suppressed.
        
      • CNS TB can cause this also (usually have CNS pleocytosis).
        
      • Fungal infections like cryptococcus can do this too.
        
    • Malignancy -- Angioinvasive lymphoma or lymphovascular lymphamatosis
      
• Other

If the echo is clean, I would consequently perform angiography (MRA possibly conventional) in this patient and look for evidence of vasculitis. I think given the night sweats and epidemiology that TB needs to be excluded prior to high dose steroids. I would repeat the lumbar puncture to ensure that it is still acellular (early TB can be acellular) and again send TB studies.

I would confirm whether or not he has had a mantoux. He should have one prior to immunosuppression anyways and should be treated for latent TB if this does not represent CNS TB.

I would probably scan his chest/abdomen/pelvis to look for evidence of malignancy, lymphoma or TB disease -- or something safer than the brain to biopsy and make a diagnosis.

Ultimately he may require a stereotactic brain biopsy to exclude TB and make a diagnosis. In a young man, with progressive infarcts and no other cause, I would discuss this with the patient.

TB is treatable, thrombophilia and emboli preventable, and lymphoma may be cured but CADASIL can not so we had better be sure!