Transcript Post-operative visual loss, risk factors, mechanisms and prevention

Postoperative Visual
Loss
Marybeth A. Grazko MD
Postoperative Visual Loss
Impairment or total loss of sight following
an otherwise uncomplicated surgical
anesthetic requiring procedure (nonocular)
 Most commonly associated with spinal
surgeries.
 Most common pathophysiology is posterior
ischemic optic neuropathy (PION).

Postoperative Visual Loss
Uncommon,devastating and poorly understood.
 Uncommon

– Most often associated with spinal, cardiac, head
and neck surgeries. Incidence is spine surgery
0.2% (50 fold higher incidence )
– Posterior ischemic optic neuropathy is by far the most
common mechanism of permanent visual loss
– 83/93 cases reported in American Spine Association
registry were due to ischemic optic neuropathy
(ION),
– 53 P(posterior)ION, 19 A(anterior)ION, 10
unspecified ION, 10 central retinal artery occlusion
Case 1

65 year old male underwent
– Cystocopy, bilateral ureteral stent placement
– Laparascopic bilateral pelvic lymph node
dissection
– Laparscopic radical prostatectomy
– 16.5 hour procedure
– Trendelenburg dorso-lithotomy position for
most of this time
Pre-Op History

Past Ocular hx
– Retinal detachment repair
– Cataract
– Primary open angle glaucoma OS with glaucomatous
optic neuropathy (had afferent pupillaryt defect OS
pre-operatively)
– Humphrey 30-2 perimetry pre-operatively showed:
 Normal OD
 Arcuate defect OS
– pre-operative visual acuity of 20/20 OD, 20/50 OS
Intra-operative Course
After anesthetic induction blood pressure
dropped to 80/50 for 30 minutes but
otherwise maintained 110-130 mmHg
systolic and 70-80 mm Hg for the rest of
the case
 Blood loss 500cc
 Post op HCT was 38.3
 No known operative or anesthetic
complications

Post Op exam






Patient complained of visual loss OD upon
awakening
Visual acuity 20/40 OD, 20/60 OS; no afferent
pupil defect!
Mild punctate corneal erosions OD
Significant facial and periorbital edema
Dilated fundus exam was normal
Humphrey perimetry: OD peripheral depression
extending to fixation superiorly. OS unchanged
Case 2







22 year old female
Right laparoscopic pyeloplasty for right
ureteropelvic junction obstruction
6 hour surgery
Blood pressure 100/45 after induction, 100-110
over 50-60 mm Hg for duration of the case
Blood loss 50 cc
Post op HCT 34.2
Significant facial and periorbital edema
Post Operative status
Immediately noted bilateral loss of vision
but attributed it to the anesthesia
 Post-Op Day 7 Ophthalmology consult:
 Visual acuity 20/20 OU
 Relative afferent pupil defect OS
 Humphrey perimetry: peripheral
depression OD, diffuse depression OS
 Brain and orbit MRI were normal

Post Operative Visual Loss

Primary pathophysiology
– Anterior ischemic optic neuropathy
– Posterior ischemic optic neuropathy
– Central retinal artery occlusion

Associated surgical procedures
– Cardiac, thoracic, cerebrovascular, abdominal,
orthopedic, sinus and neck
– Majority of the cases are associated with
spinal surgeries
Majority of perioperative visual loss
has been due to PION (50-71%)

Posterior Ischemic Optic Neuropathy
– Sudden painless loss of vision
– Unilateral or bilateral (may or may not have
associated afferent pupillary defect)
– Normal fundus exam
– Nonspecific visual field abnormality
– In the absence of an afferent pupil defect may need
negative neuroimaging to confirm the diagnosis
Delay in Diagnosis
Often most of the bedside objective exam
will be normal.
 Patients often may attribute changes in
their vision to a recovery time following
anesthesia.
 May have other post-operative issues that
overshadow these symptoms.

Clinical Characteristics of Patients
Buono MB, Forrozan R, Surv Ophthalmol 50: 15-26, 2005
Lee et al, Anesthesiology 2006; 105: 652-9

Age
– 51.9 average
– 43.9 spinal surgery, registry average was 50

Degree of visual loss
– Count fingers or worse in 75.8%
– No light perception in 53.8% (56.6%)
– Bilateral involvement in 60.9% (66%)

Visual recovery
– Some visual recovery in 37.9% (42%)
– Patients with poorer vision initially tended to have
less improvement and poorer outcomes
– 54.9% had a final visual acuity of Hand motion or
worse
Clinical Characteristics of Patients
Buono MB, Forrozan R, Surv Ophthalmol 50: 15-26, 2005
No optic disc cupping noted
 Small congenital optic nerve head or
absent phyiologic cupping noted only in
4% in one series
 65.5% had one or more vascular risk
factors
 34.5% had no vascular risk factors

Clinical Characteristics
Lee et al, Anesthesiology 2006; 105: 652-9

Average anesthetic duration
– 9.8 h +/- 3.1 hours
– 6 hours or longer in 94%

Estimated blood loss
– 82% was 1000 cc or greater

Blood Pressure widely varied from less
than 20% of baseline to 50% of baseline
or greater, HCT (hematocrit) also had wide
variations between patients
Pathology
PION results from infarction of the
intraorbital portion of the optic nerve
 Infarction results due to decreased oxygen
delivery presumably to any number of
perioperative hemodynamic
derrangements

Pathophysiology
Decreased oxygenation leading to infarction

Decrease in perfusion pressure
– Hypotension (decreased Mean Arterial Pressure)
– Disturbed autoregulation (atherosclerosis,
hypertension)
– Increase in intraocular pressure ( prone and head
down positioning)
– Increase in Venous pressure
 Prolonged head down and prone position
 Internal jugular compression or ligation

Decreased oxygen carrying ability
– Anemia
– Hemodilution
Hemodynamic derangements
– Hypotension and blood loss are common in cardiac
bypass surgery. Low incidence of PVL seen in cardiac
bypass patients suggests that another mechanism is
playing an additional role in the mechanism for
permanent visual loss.
– Anatomic variation in blood supply of intraorbital
nerve (absent anastomoses)
– Positioning creating either increased intraocular
pressure and/or increased orbital venous pressure
From: Ozcan : Anesth Analg, Volume 99 (4), Oct 2004. 1152-1158
Blood Supply of the Nerve
Facial Edema

Facial edema
– Often data was not provided but in 19 patients where
this was commented upon, 17/19 had post operative
facial edema and in 11 was considered to be severe.
(both of our cases had severe facial edema)
Evidence that positioning can increase
intraocular pressure (maximal in Trendelenberg)
 Increased intraocular pressure can lead to
decreased perfusion.

Pathogenesis

Hemodynamic derangement
– Evidence to date points to prolonged surgery, blood
loss or combination of both

In a patient with a specific susceptibility
which at this point remains unclear
– Incomplete anastomoses, altered anatomy

Risks factors of positioning and periorbital/facial
edema may contribute but at this point are less
clearly factors
Recent Practice Advisory
Am Soc of Anesthesiologists
Spine procedure
 Positioned prone
 Receiving general anesthesia
 Uncommon occurrence

– Less than 0.2% of spine surgeries
– Makes it a difficult thing to study, mostly case
control studies and case reports in terms of
what is in the literature.
ASA Practice Advisory

Task force of 12 members
–
–
–
–
–
–


4
3
1
1
2
3
Anesthesiologists
Neuro-Ophthalmologists
Orthopedic Spine Surgeon
Neurosurgeon
Methodologists from ASA Practice Parameters committee
physicians served as liasons from national organizations
 NANOS
 Am Academy of Orthopedic Surgeons
 Am Association of Neurologic Surgeons
Opinions were also solicited from Society for Neurosurgical
Anesthesia and Critical Care, NANOS, and North American Spine
Society
Summary of the available literature, expert opinion, open forum
commentary and consensus surveys
Predictive Risk Factors
Vascular risk factors (hypertension,
diabetes, smoking,obesity,glaucoma,
carotid artery disease)
 Pre-operative presence of anemia
 Prolonged procedures (range was 2-12
with average of 6.5 hours)
 Substantial blood loss (ranges was 10200% with average of 44.7% of total
blood volume)

Pre-Operative Assessment

Task Force was in consensus that
– There is no evidence that an ophthalmic or
neuro-ophthalmic exam pre-operatively is
useful for identifying patients at risk
– Recommendation was to consider informing
patients in whom prolonged procedures,
substantial blood loss or both are anticipated
that there is a small unpredictable risk of
perioperative visual loss.
BP,fluid,blood loss managment
Recommend continual monitoring in high
risk patients (CVP monitoring)
 Deliberate hypotensive techniques have
not been shown to be associated with
peri-operative visual loss.
 There is no absolute hematocrit number to
use a transfusion threshold, individually
dependent

Patient Positioning
Several case reports of direct pressure to the
eyes from use of a sheet roll or head rest
resulting in central retinal artery occlusion
(CRAO) or ION but also in patients with no
headrest (head held by pins)
 All task force members agreed that direct
pressure on the eye should be avoided to
reduce risk of CRAO and other ocular damage.

Head positioning?






No pathophysiologic mechanism by which facial edema
can cause perioperative ION
Eyes of prone patients should be regularly checked
Peri-operative facial edema is common in high risk
patients
Postion the head level or higher than the heart for highrisk patients
Keep the head in a neutral forward position when
possible
Direct pressure on the eye should be avoided to avoid
CRAO
Treatment
Correction of hemodynamic
derangements, systemic steroids,
antiplatelet therapy, measures to lower
intraocular pressure or intracranial
pressure.
 None effective though a few case reports
of improvement with transfusion and
keeping blood pressure over 140/80

Peri-operative Visual loss registry
http://depts.washington.edu/asaccp/eye/i
ndex.shtml
 www.asaclosedclaims.org

– Established in 1999
– Recently published on 93 spine cases (93/131
cases 72%)
– Lee et al Anesthesiology 2006, 205:652-659.
– Anonymous reporting
Advisory
PION can be seen with a variety of
surgical procedures but is more likely to
be seen with spinal procedures
 There is no evidence that compression is a
factor in the pathophysiology of this entity
although compression is clearly related to
CRAO
 PION does correlate with prolonged
surgery (> 6 hours)

Advisory
While PION does correlate with blood loss,
the amount of blood loss required is
unclear and can still occur in the absence
of significant blood loss as our cases
demonstrate.
 PION is somewhat more likely with
prolonged hypotension but again the
range where this is significant is likely to
be very patient dependant.

Advisory

Patients in a prolonged supine position
with significant peri-orbital swelling (our
two cases) while theoretically this could
predispose them to PION has yet to be
recognized as an independent risk factor.
Prognosis

Visual recovery
– Some visual recovery in 37.9% (42%)
– Patients with poorer vision initially tended to have
less improvement and poorer outcomes
– 54.9% had a final visual acuity of HM or worse

Both of our patients had some improvement
both in their acuity and visual fields but again
were more mildly affected than many of the
other reported cases
Summary
Postoperative Visual Loss is a fortunately
rare but devastating condition that is still
poorly understood.
 Recent reports by the ASA Registry and
reviews of the literature would suggest
counseling patients who are undergoing
procedures that are prolonged and/or with
expected substantial blood loss about the
potential risk for this condition.
