Acute pulmonary edema following
venous air embolism during a
neurosurgical procedure
Case report
WILLIAM F. CHANDLER,M.D., DONALD G. DIMCHEFF, M.D.,
AND JAMES A. TAREN, M.D.
Section of Neurosurgery and Pulmonary Division, Department o] Internal Medicine,
University of Michigan Medical Center, Ann Arbor, Michigan

A case of venous air embolism during a procedure in the sitting position is
reported in which the patient developed fulminant pulmonary edema within 2
hours of the embolus. Possible cause and effect relationships are discussed.
KEY WORDS
venous air embolism
wet lung syndrome

CUTE pulmonary edema following
neurosurgical procedures is uncommon, but well documented. ~,~2
Venous air embolism during neurosurgical
procedures in the sitting position has a
reported incidence of from 1.6% to
15%, ~2,~3 and a mortality rate, generally
from hypotension and cardiac arrest, ~ of up
to 93% in untreated cases, 4 and from 0 to
28% in treated cases. 9,~ In surveying the
literature we were unable to find any cases
where acute pulmonary edema, or the "wet
lung" syndrome, was associated with venous
air embolism. We are reporting such a case
in which we verified the diagnosis of venous
air embolism by fortuitously noting air in
the dural sinuses during routine fluroscopy
and video taping.

A

Case Report
At the age of 10 years this 27-year-old
dO0

pulmonary edema

left-handed man noted increased tone and
athetoid movements in his right arm and leg
which progressed slowly over a 10-year
period, and stabilized so that his right arm
was virtually useless. The abnormal movements were accentuated by stress, and
caused him to avoid social situations. The
patient was diagnosed as having dystonia
musculorum deformans and admitted for a
left stereotaxic thalamotomy to decrease the
tone and abnormal movements.
Examination. The positive neurological
findings were confined to the right arm and
leg and included marked increase in tone,
minimal hypesthesia to pinprick, and athetoid movements, more marked distally; all
of these findings were greater in the right
arm. CerebeUar testing was normal; there
was no specific weakness and the deep
tendon reflexes were 2 + bilaterally. The
general physical examination was normal.
J. Neurosurg. / Volume 40 / March, 1974

Acute pulmonary edema following venous air embolism
On the day before surgery the patient
underwent a pre-thalamotomy pneumoencephalogram during which 110 cc of
oxygen were instilled via a 20-gauge spinal
needle at L3-4, and 55 cc of CSF were
removed. The patient tolerated the procedure well, and the study demonstrated
slightly enlarged ventricles with no cortical
atrophy.
Operation. The patient was placed in the
sitting position and given 4 cc of Innovar*
intravenously. As it was necessary for the
patient to be conscious and cooperative for
neurological examination, neither a "G suit"
nor an esophageal stethoscope was used.
Under local lidocaine anesthesia a left
occipital trephine opening was made and
bone wax was applied; the dura appeared
sunken and pulsatile. A Guiot-Gillingham
stereotaxic apparatus was then applied; the
patient was talkative and able to move all
extremities. As a small cruciate incision was
made in the dura he complained of neck
pain, and his pulse rate fell from 80 to 40
with no change in blood pressure. Atropine,
0.4 mg, abolished the bradycardia. At this
point, two attempts were made to enter the
left lateral ventricle with the use of the
stereotaxic ventricular needle, but the ventricles were not encountered and, therefore,
no air was injected. By using the "C-arm"
fluoroscopy unit and video taping, the
trajectory of the needle was checked and
appeared appropriate. It was noted on the
TV monitor that air was streaming over the
convexities from the vertex region, flowing
posteriorly and inferiorly toward the base of
the skull. This was thought initially to be
subarachnoid air. However, at the same
time, the patient became increasingly
drowsy, and subsequently unarousable; the
procedure was terminated immediately. During the closing procedure, the blood pressure
dropped from 140 to 90 systolic, but
responded to 30 ~ semi-Fowler's position.
Several premature ventricular~ontractions
were noted on the EKG monitor.
Postoperative Course. In the r~overy
room the patient was unresponsiv~ and
intermittently decerebrate in response to

FIG. 1. Left carotid arteriogram, brow-up
venous phase, demonstrating retraction of frontal
lobes (small arrowheads) and stretching of a bridging vein (large arrowhead).

noxious stimuli. The blood pressure was
stable at 150/80, and an EKG showed sinus
tachycardia of 160. Physical examination
demonstrated marked tachypnea and bilateral rales. The patient had small minimally reactive pupils but maintained the
oculocephalic reflex. A nasotracheal tube
was placed, and emergency left carotid and
vertebral arteriograms showed no intra- or
extracerebral hematoma but did show: considerable air subduraUy; the frontal lobes
appeared retracted from the skull, and the
bridging veins were stretched (Fig. 1) ;
minimal transtentorial herniation was also
demonstrated. During the arteriogram the
patient was dusky, his arterial blood dark,
and he showed signs of peripheral vasoconstriction. Arterial blood gases drawn
while the patient was receiving 100% oxygen
revealed a pO2 of 51 and pCO2 of 28. A
chest film showed bilateral pulmonary
edema (Fig. 2). A central venous catheter
was placed, and the central venous pressure
was found to be 3 cm of water. A review of
the video tape at this point showed that the
streaming air appeared to be in the dural
sinuses
(Fig. 3). The chest film was taken
*Innovar (fentanyl 0.05 mg and droperidol 2.5
mg per cc) manufacturedby McNeil Laboratories, 2 89hours after the air had been seen on
fluoroscopy.
Inc., Fort Washington, Pennsylvania 19034.
J. Neurosurg. / Volume 40:4/ March, 1974

40]

William Chandler,

FIG. 2. Chest film 2 hours postoperatively
showing bilateral pulmonary edema.

The "wet lung" syndrome 11 was treated
by using a volume cycle respirator with large
title volumes (18 cc/kg) and positive end
expiratory pressure (PEEP) of 10 cm H20.
Pancuronium was used for neuromuscular
paralysis. Within 6 hours, the percentage of
inspired oxygen (FIO2) was progressively
decreased from 100% to 40%, so that
adequate peripheral oxygen transport was

Donald

Dimcheff and James Taren

assured; i.e., an arterial pO2 of 75 to 85, 02
saturation of 90% to 95%, normal pH, and
normal hemoglobin concentration. Fluid
management was monitored with hourly
urinary output, together with central venous
pressure, pulse, blood pressure, and body
weight. A daily weight loss of at least 0.5 kg
was maintained so that any increase in lung
water from fluid management was negligible. 21 Furosamide was used intermittently to
assure this weight loss. Physiological and
roentgenographic evidence of pulmonary
improvement appeared slowly. The PEEP
and paralysis were discontinued on the 5th
day of mechanical ventilation, and he was
weaned from the ventilator by the 8th
day.
It was difficult to assess the patient
neurologically while paralyzed with pancuronium, but he was able to protrude his
tongue on command on the first postoperative day as the drug effect subsided,
although he maintained decerebrate posture.
Pupils were equal and reactive, and the
EEG showed moderate diffuse slowing,
consistent with hypoxic changes. After the
pancuronium was stopped on the 5th
postoperative day he could follow with his
eyes, protrude his tongue, but had no
spontaneous movements of his extremities.
Gradually over the next 4 weeks, his mental
status progressed from stupor to confusion
to complete normalcy. The cranial nerve
and sensory examination was normal within
2 weeks. He progressed from decerebrate
rigidity to generalized spasticity, with motor
function appearing first in the left arm at 3
weeks, and in his legs by 4 weeks.
Three months postoperatively, his left
arm is normal, he walks unassisted with a
slightly spastic gait, and his right arm
remains spastic and flexed at the elbow, with
only minimal function. He is able to feed
and dress himself.
Discussion

Fro. 3. Medical illustrator's drawing of the
television videotape of fluoroscopy during the
operation. The posterior portion of the skull is to
the left; and the white dots represent air seen
traveling in the dural sinuses and internal jugular
vein. Air is also seen in the lateral ventricles.
402

Precise interpretation of the sequence of
events in this patient is difficult and not
entirely conclusive, but we believe the
relationship between the events is as
follows: the preoperative pneumoencephalogram, with a large exchange of air
and CSF, plus a postulated continuous leak
1. Neurosurg. / Volume 40 / March, 1974

Acute pulmonary

edema following venous air embolism

from the lumbar subaraclmoid space, removed the normal fluid cushion of the brain,
and allowed caudal displacement in the
sitting position. This was evidenced by the
markedly sunken dura seen on completing
the craniectomy. We believe that when the
dura was opened this served as a vent,
allowing air to enter the subdural space, and
in turn produced a further caudal shift of the
brain. Since air was visualized in the
superior sagittal sinus over the vertex at this
point, we conclude that the caudal movement of the brain caused tearing of a
bridging vein, and allowed the ready access
of subdural air into the venous sinus.
Although an esophageal stethoscope was not
available for evaluation of cardiac murmurs,
the bradycardia, 5 hypotension,6 and premature ventricular contractions 15 are all
consistent with the venous air embolism
documented on the TV monitor.
Shortly after the air embolism, the patient
developed severe pulmonary edema, due
either to the air embolism or a neurogenic
stimulus. The pathophysiology of these two
entities involve distinctly different mechanisms. Clinically, neurogenic pulmonary
edema has been associated with trauma,
intracranial hematoma or neoplasm, and
seizures? It has been seen experimentally
following sudden increase in intracranial
pressure," bilateral lesions in the preoptic
areas, 2" hypothalamic lesions, 17 and intracisternal injection of fibrin ~~ or endotoxin.2~
Whatever the initial lesion, most investigators have recorded intense sympathetic
overactivity leading to redistribution of
blood volume to the pulmonary circuit, with
systemic and pulmonary arterial hypertension, left atrial and pulmonary venous
hypertension, and subsequent pulmonary
edema.~.3-1e,22
On the other hand, pulmonary edema
following air embolism has not been
documented in humans, but has been well
studied in dogs. Is,z~ Following large air
emboli, a foamy mixture of air, blood,
fibrin, and fat are seen in the right heart and
pulmonary arterial tree. s Initial cardiopulmonary responses include bradycardia,
systemic hypotension, and pulmonary arterial hypertension, with no left atrial or
pulmonary venous hypertension,s The pathophysiology of the subsequent pulmonary
1. Neurosurg. / Volume 40 / March, 1974

edema observed is not completely understood, but may involve mechanisms similar
to blood clot emboli. TM Theoretically, then,
measurement of the pulmonary wedge
pressure as an estimation of the left atrial
and pulmonary venous pressure should
differentiate the pulmonary edema due to air
emboli from that due to neurogenic stimuli.
Because of the lack of increased intracranial pressure, the initial bradycardia and
hypotension, and the documentation of large
air embolism on video tape, we favor air
embolism as the cause of pulmonary edema
in this patient. Since left atrial pulmonary
capillary wedge pressures were not measured, exclusion of brain stem injury as a
cause of "neurogenic" pulmonary edema is
impossible.
The exact etiology of the brain stem
injury is not apparent, but may also have
been due to the sudden caudal displacement
of the brain. As the patient appeared in the
immediate postoperative period to have a
"Count of Monte Cristo" syndrome,TM being
able to move only his eyes and tongue, we
believe the brain stem injury was in the
ventrorostral pons. A lesion here, be it
traumatic, hemorrhagic, or hypoxic, may
involve the corticospinal tracts, but spare
the aberrant pyramidal tracts to any or all
cranial nerve nuclei, x~
We believe that the rapid onset of
pulmonary edema following a neurosurgical
procedure in the sitting position should
suggest the diagnosis of air embolism as well
as a "neurogenic" etiology, and in either
case, should be treated by mechanical
ventilation and positive end expiratory
pressure. This case suggests that procedures
in the sitting position the day following
pneumoencephalography should be carefully
considered.

References
1. Ducker TB: Increased intracranial pressure
and pulmonary edema. Part 1: Clinical study
of 11 patients. J Neurosurg 28:112-117,
1968
2. Ducker TB, Simmons RL: Increased intracranial pressure and pulmonary edema. Part
2: The hemodynamic response of dogs and
monkeys to increased intracranial pressure. J
Neurosurg 28:118-123, 1968
3. Ducker TB, Simmons RL, Martin AM Jr:
403

William Chandler, Donald Dimcheff and James Taren
Pulmonary edema as a complication of
intracranial disease. Am J Dis Child 1111:
638-641, 1969
4. Ericsson JA, Gottlieb JD, Sweet RB: Closedchest cardiac massage in the treatment of
venous air embolism. New Engl J Med
270:1353, 1964
5. Gilbert RGB, Brindle F, Galindo A: Anesthesia for Neurosurgery. Boston, Little,
Brown, 1966, p 39
6. Gottlieb JD, Ericsson JA, Sweet RB: Venous
air embolism: a review. Anesth Anaig
44:773-779, 1965
7. Gurewich V, Cohen ML, Thomas DP:
Humoral factors in massive pulmonary embolism: an experimental study. Am Heart J
76:784-794, 1968
8. Hartveit F, Lystad H, Minken A: The
pathology of venous air embolism. Brit J Exp
Path 49:81-86, 1968
9. Hunter AR: Air embolism in the sitting
position. Anaesthesia 17:467-472, 1962
10. Kahn EA, Crosby EC, DeIonge BR: Tumors
of the posterior fossa, in Kahn EA, Crosby
EC, Schneider RC, et al (cds): Correlative
Neurosurgery. Springfield, Ill, Charles C
Thomas, 1969, ed 2, pp 211-213
11. Kumar A, Konrad BS, Falke KJ, et al:
Continuous positive-pressure ventilation in
acute respiratory failure. Effects on hemodynamics and lung function. New Engl J
Med 283:1430-1436, 1970
12. Marshall BM: Air embolus in neurosurgical
anaesthesia: its diagnosis and treatment.
Canad Anaesth Soc J 12:255-261, 1965
13. Michenfelder JD, Martin IT, Altenburg BM,
et al: Air embolism during neurosurgery: an
evaluation of right-atrial catheters for diagnosis and treatment. JAMA 208:1353-1358,
1969
14. Michenfelder JD, Terry HR Jr, Daw EF, et
al: Air embolism during neurosurgery: a new
method of treatment. Anesth Analg 45:390395, 1966
15. O'Higgins JW: Air embolism during neuro-

40,4

16.

17.
18.

19.

20.

21.

22.

23.
24.

surgery. A case report. Brit J Anesth
42:459-462, 1970
Paine R, Smith JR, Butcher HR, et al: Heart
failure and pulmonary edema produced by
certain neurologic stimuli. Circulation 5:759765, 1952
Reynolds RW: Pulmonary edema as a
consequence of hypothalamic lesions in rats.
Science 141:930-932, 1963
Richardson HF, Coles BC, Hall GE: Experimental gas embolism. I. Intravenous air
embolism. Current Med Ass J 36:584-588,
1937
Sarnoff SJ, Sarnoff LC: Neurohemodynamics
of pulmonary edema, n . The role of
sympathetic pathways in the elevation of
pulmonary and systemic vascular pressures
following the intracisternal injection of fibrin.
Circulation 6:51-62, 1952
Simmons RL, Ducker TB, Martin AM, et ah
Intrathecal endotoxin: the role of the central
nervous system in septic shock. Surg Forum
28:12-14, 1967
Sladen A, Laver MB, Pontoppidan H:
Pulmonary complications and water retention
in prolonged mechanical ventilation. New
Engl J Med 279:448-453, 1968
Urabe M, Segawa Y, Tsubokawa T, et al:
Pathogenesis of the acute pulmonary edema
occurring after brain operation and brain
trauma. Jap Heart J 2:147-169, 1961
Wycoff CC, Cann JE: Experimental pulmonary air embolism in dogs. Calif Med
105:361-367, 1966
Yuceoglu YZ, Rubler S, Eshwar KP, et al:
Pulmonary edema associated with pulmonary
embolism: a clinicopathological study. Angiology 22:501-510, 1971

Address reprint requests to: William F. Chandler, M.D., Department of Surgery, Section of
Neurosurgery, University of Michigan Medical
Center, Ann Arbor, Michigan 48104.

1. Neurosurg. / Volume 40 / March, 1974