Complications of BNTI:

Still lacking ED RSI at Charity, we frequently use BNTI, acknowledging
its inferiority to RSI where complications are concerned. Despite meticulous
preparation, we still encounter frequent nasal trauma (2,3) and occasional
mediastinal intubation with morbid results (1). 

What we usually proceed
to do in the ICU, once a patient's immediate urgent airway and other
lifethreatening concerns are stabilized, is to change out most nasal
intubations to oral ones; this to prevent sinusitus, to allow improved
pulmonary toilet, to decrease airway resistance, and to allow bronchoscopy.

(1) Seaman M. Mediastinitis following nasal intubation in the emergency
department. Am J Emerg Med 1991 Jan;9(1):37-9 

ABSTRACT:
A patient who developed a retropharyngeal abscess and fatal mediastinitis
following emergent nasal intubation is described. Despite aggressive
surgical therapy the patient died of mediastinitis. Although mediastinitis
as a complication of oral intubation has been described, mediastinitis
following nasal intubation has not previously been reported. 

(2) O'Hanlon J. Epistaxis and nasotracheal intubation--prevention with
vasoconstrictor spray. Ir J Med Sci 1994 Feb;163(2):58-60 

ABSTRACT:
Eighty patients having anaesthesia for oral surgery requiring nasal
intubation were randomly allocated to be intubated with either a plain
Magill red rubber or cuffed polyethylene endotracheal tube and in a double
blind manner, to receive xylometazoline 0.1% vasoconstrictor nasal spray. 
The extent of any epistaxis occurring was assessed by an independent
observer. With the Magill tube there was bleeding in one out of twenty
patients in both the vasoconstrictor group and non vasoconstrictor group at
intubation and no bleeding in either of the two groups at extubation. With
the polyethylene tube sixteen out of twenty patients had bleeding in the
non vasoconstrictor group. This improved to seven out of twenty with the
administration of vasoconstrictor drops at intubation (chi square 10.2; p <
0.01) in the polyethylene tube group. At extubation ten out of twenty
patients had bleeding in the non vasoconstrictor group improving to two out
of twenty with the administration of the vasoconstrictor (chi square 9.6;
p,0.01). The use of the vasoconstrictor xylometazoline helped to reduce
epistaxis that occurred during nasal intubation and further study into the
type of endotracheal tube is recommended. 

(3) Cameron D. Inadvertent brain penetration during neonatal nasotracheal
intubation. Arch Dis Child 1993 Jul;69(1 Spec No):79-80 

ABSTRACT:
During routine nasal intubation of a premature infant, the endotracheal
tube penetrated the brain. Bloodstained cerebrospinal fluid and neural
tissue was apparent. Initial cranial sonography was normal, but the infant
later developed extensive intracranial haemorrhage. Rotation of an
endotracheal tube to facilitate insertion angles the bevel at the tip
upwards, increasing risk of brain penetration. Great care is required
during nasal intubation; use of a small feeding tube over which to slide an
endotracheal tube may be helpful. 

(James Li, M.D.)

 What am I supposed to conclude from the observation of ref 1?: 
"Although mediastinitis as a complication of oral intubation has been
described, mediastinitis following nasal intubation has not previously been
reported."  This 1994 paper was the first case report of this complication
with BNTI while mediatinitis as a result of OTI had been previoulsy
observed.  Seems to me that this, if anything, makes a good case for BNTI
over OTI and not the other way around!

 Refernce 2 makes the earth shattering observation that topical 
vasocontrictors may decrease nasal bleeding with BNTI.  Well, let me tell
you a dirty little secret:  I rarely even bother to take the time to use
them.  Generally you will see a little nasal bleeding, rarely of
troublesome proportions.  Common sense should be used.  Avoid the procedure
in patients with a coagulapathy.  The main drawback of nasal bleeding is if
BNTI fails then the blood will somewhat obscure your view when the patient
is intubated orally.  Make sure suction is available and know when to stop
BNTI before the bleeding becomes voluminous. If the procedure is done in
the sitting position (another advantage of BNTI) then aspiration should not
be a problem.  Blood, in any case, is rather inert and non-toxic when
aspirated unlike gastric juice unless the patient exanguates and
asphyxiates.  The mortality as a result of gastric aspiration is a function
of pH.

 My prefered technique (generic) is (baring contraindications to 
BNTI such as head trauma) to premedicate with a benzodiazepine or ketamine
depending on the circumstance (use fentanyl or combinations if you like).
I'll have the nurse draw up sux and prepare my laryngoscope etc.
_just_in_case_. Meanwhile,  with the patient breathing spontaneously with
supplemental oxygen and in the upright position (thus avoiding the hazards
of gastric distension from positive pressure ventilation) BNTI is attempted
first in one nostril and then in the other.  With failure to pass the tube
the patient is placed horizontally, sux is administered and the patient
intubated orally.  Two days ago I did a BNTI in 5 seconds.  Yesterday going
through the same procedure I failed to pass the tube nasally and proceded
to paralyze and OTI.  The little extra time that the benzo has to work
makes the administration of of an NMB a little more humane. If you don't
have RSI as a backup procedure then you're screwed.

 Reference 3 describes the intracranial penetration of the tube in 
a neonate.  That patient had no business being intubated nasally.  I 
generally don't even consider it under the age of 12 or so.  The anatomy of
a child is completely different with a high anterior larynx and profuse
lymphoid tissue and does not lend itself readily to BNTI.  So if you see a
complication in a situation where a procedure should not have been utilized
in the first place don't express any surprise.	

(Harvey Louzon, M.D.)