Coma in a spinal patient

There are many ways to break this down into a logical sieve. This is a modified version of the approach described by Bala Venkatesh in Oh’s Intensive Care Manual:

Differential Diagnosis of Coma

This, of course, is a very generic approach. Immediately post op, you can also think of causes as being:

• Related to the anaesthetic
• Related to the surgery
• Related to post-operative events

As discussed in Ophthalmological Befuddler 001 there are two possible causes of anisocoria (unequal pupillary size)

• one pupil is abnormally small or constricted (miotic)
  OR
• one pupil is abnormally large or dilated (mydriatic)

Again, an approach modified from that of Bala Venkatesh in Oh’s Intensive Care Manual is shown:

For further review read Neurological Mind-boggler 002 on the causes of coma with small pupils, and Ophthalmological Befuddler 001 for the causes of a dilated pupil.

In addition to a close review of what happened in the operating theatre and recovery, the following investigations may help:

• ABG including glucose and lactate
• Blood tests (see Q2 above)
• CT brain +/- MRI brain +/- spine
• EEG

The currently held belief is that as the epidural abscess was drained, canal compression was released and CSF flow was altered. A bolus of intrathecal baclofen (from the pump, which had always been running) then reached the brain in high concentration, and caused coma.

The anisocoria was also attributed to the baclofen overdose, although this usually causes bilateral pupillary dilation. It is difficult to explain why only one pupil was affected  — perhaps there was a degree of  post-operative Horner’s syndrome affecting the side of the ‘small’ pupil?

With supportive treatment complete neurological recovery occurred over the next 48 hours.

Read Wall et al (2006) for an open access case report of a baclofen overdose in a different context, with a literature review.

Is antimicrobial chemotherapy inadequate?

• check that current antibiotic dosing is adequate for the MIC of cultured organism
• re-culture everything, empirically broaden cover and narrow when you get a new organism

Is there an ongoing source of sepsis?

• Was all the abscess drained? Re-drain & washout.
• Is the metal work infected? Remove it if suspected and its possible.
• Is the baclofen pump infected? Remove it!
• Look for ventilator associated pneumonia (VAP): CXR, blind tracheal aspirate & non-quantitative culture — see Canadian Clinical Trials Group (2006).
• Look for endocarditis (TTE +/- TOE depending on the TTE result)
• Look for collections elsewhere

Antibiotic cover was increased to vancomycin & meropenem. An echocardiogram showed no vegetations. A further MRI showed evidence of remaining collection. In OT, the internal fixation was removed, the intrathecal baclofen pump was removed and the abscess drained. The only organism isolated was the MSSA and long term (at least 6 weeks) IV benzyl penicillin was all that was needed. The MSSA presumably came from her scratched skin and seeded to her metal rods/epidural space.

Of course, other complications occurred but I won’t bore you with them!

• Canadian Critical Care Trials Group. A randomized trial of diagnostic techniques for ventilator-associated pneumonia. N Engl J Med. 2006 Dec 21;355(25):2619-30. PMID: 17182987.
• Wall GC, Wasiak A, Hicklin GA. An initially unsuspected case of baclofen overdose. Am J Crit Care. 2006 Nov;15(6):611-3. PMID: 17053268. [free fulltext]