Brief Report

Neurologic manifestations of cerebral air
complication of central venous catheterization

embolism

as

a

Josef G. Heckmann, MD; Christoph J. G. Lang, MD; Klaus Kindler, MD; Walter Huk,
MD; Frank J. Erbguth, MD; Bernhard Neundörfer, MD
From the Departments of Neurology (Drs. Heckmann, Lang, Kindler, Erbguth, and
Neundörfer) and Neuroradiology (Dr. Huk), University of Erlangen-Nuremberg, Germany.
Address requests for reprints to: Josef G. Heckmann, MD, Department of Neurology,
University Erlangen-Nuremberg, Schwabachanlage 6, 91054 Erlangen, Germany.
Downloaded from http://journals.lww.com/ccmjournal by BhDMf5ePHKbH4TTImqenVPVWEjt0ik0+SG9c/P6Ve9I1c6F+mSUjQSpeqTXfjf6L on 11/07/2018

Objective, Patients, and Methods: A severe
case of cerebral air embolism after
unintentional central venous catheter
disconnection was the impetus for a
systematic literature review (1975-1998)
of the clinical features of 26 patients
(including our patient) with cerebral air
embolism resulting from central venous
catheter complications.
Results: The jugular vein had been
punctured in eight patients and the
subclavian vein, in 12 patients. Embolism
occurred in four patients during insertion,
in 14 patients during unintentional
disconnection, and in eight patients after
removal and other procedures. The total
mortality rate was 23%. Two types of
neurologic manifestations may be
distinguished: group A (n = 14) presented
with encephalopathic features leading to a
high mortality rate (36%); and group B
(n = 12) presented with focal cerebral

lesions resulting in hemiparesis or
hemianopia affecting mostly the right
hemisphere, with a mortality rate as high
as 8%. In 75% of patients, an early
computed tomography indicated air
bubbles, proving cerebral air embolism.
Hyperbaric
oxygen
therapy
was
performed in only three patients (12%). A
cardiac defect, such as a patent foramen
ovale was considered the route of right to
left shunting in 6 of 15 patients (40%).
More often, a pulmonary shunt was
assumed (9 of 15 patients; 60%). For the
remainder, data were not available.
Conclusion: When caring for critically ill
patients
needing
central
venous
catheterization,
nursing
staff
and
physicians should be aware of this
potentially lethal complication.
KEY
WORDS:
central
venous
catheterization; cerebral air embolism;
embolism; complication; adverse effect

Air embolism is a rare cause of
stroke, occurring in a number of situations
such as trauma, hemodialysis, oral-genital
sexual relations in women, childbirth, high
altitude accidents, diving accidents, open
heart
surgery,
thoracotomy,
pneumoradiologic
procedures,
arterial
angiography, and neurosurgery (1-6).
Previously, diagnosis was made by signs

and symptoms occurring during a high-risk
setting. In isolated cases, the introduction of
computed tomographic (CT) scan made it
possible to show air bubbles in the brain (1,
4).
Cerebral air embolism after central
venous catheterization procedures is a rare
but hazardous complication (1, 3, 4, 7). In
the literature are two reviews, one on 24

patients (8) and another on 79 patients (9).
In both reports, however, the authors did not
differentiate between cerebral air embolism
and embolism in other locations. In other
reviews, only general clinical and
experimental considerations of venous air
embolism are reported (10-12).
Few data exist on the neurologic
features of cerebral air embolism after
central venous catheterization. The aim of
this article is two-fold: first, a severe case of
cerebral air embolism in relation to
accidental disconnection of a central venous
catheter is reported; and second, all
published cases of cerebral air embolism
after central venous catheterization in the
literature between 1975 and 1998 are
reviewed in a systematic way. We hope to
provide information that will allow
clinicians to recognize the condition more
readily, to intervene more promptly, and to
prevent this hazardous complication,
whenever possible.
Case Report, Patients, and Methods
A 72-yr-old man was receiving cardiological
care after aortocoronary bypass surgery after
myocardial infarction and angiographydocumented coronary artery disease. There
was no intraoperative complication. A
postoperative pneumothorax was treated
immediately via a thoracic drain. Three days
later, the drain was successfully removed.
The patient was in stable cardiopulmonary
condition and without neurologic deficits
when transferred to the cardiological
department for further recovery.
In the evening, 14 days later, the patient
unintentionally disconnected the central
venous catheter and was found unresponsive
to verbal and noxious stimulation. All limbs
were paretic. Deep tendon reflexes were
symmetrical, and plantar reflexes were
extensor
bilaterally.
The
patient's
cardiopulmonary condition was stable.
Because continuous heparin had been given,
an intracerebral hemorrhage was initially

assumed, and the patient was transferred at
midnight to a radiologic institution where
CT scan was available. An intracerebral
hemorrhage was excluded, but unfortunately,
the CT results were misinterpreted. Small air
bubbles in the right hemisphere were not
detected (Fig. 1, arrows). The chest
radiograph
performed
later
was
unremarkable with regard to a hemato- or
pneumothorax.

Figure 1. Cerebral computed tomography 3
hrs after onset of signs and symptoms.
Multiple air bubbles are seen more
prominently in the right hemisphere
subarachnoid vessels.
The patient remained in a severe
tetraparesis and akinetic mutism, leading to
a first neurologic consultation. The control
CT indicated prominent hypodense lesions
of the right hemisphere, minor lesions of the
left frontotemporal region, and very mild
lesions in both hypothalamic areas (Fig. 2).
Consequently, the first CT was reviewed
and the air bubbles were detected. Air
embolism was documented by using
measurement in (-16) Hounsfield units (1, 4).
At this time, a specific treatment with
hyperbaric oxygen was not considered to be
of value and, therefore, was not performed.

embolism,
neuroradiologic
findings,
cardiological findings, and the course of the
disease. In some cases, conflicting or
missing data could not be resolved; in other
cases, data had to be simplified for
tabulation.
Table 1. Relevant clinical findings, 1975 to
1998

Figure 2. Follow-up computed tomography
3 days later showing hypodense lesion of the
right hemisphere.
A presurgical coronary catheter
investigation demonstrated severe proximal
coronary artery stenosis of the left anterior
interventricular branch and the left
intermediate branch. Ventriculography
demonstrated
hypokinesia
of
the
anterolateral wall. There were no hints of a
patent foramen ovale or similar right-to-left
shunts.
Three months later, the patient was
still in akinetic mutism, without signs of
neurologic improvement. Focusing on this
problem, we performed an extensive
literature search (MEDLINE, Index Medicus,
Excerpta Medica, Psychological Abstracts,
Chemical Abstracts, and Current Contents)
under the headings "central venous
catheterization,"
"central
venous,"
"catheterization," "cerebral air embolism,"
"cerebral air," "air," "embolism" for the
years 1975-1998. All available data
including follow-up examinations were
evaluated. Central venous catheterization
procedures,
clinical
symptomatology,
treatment, course, CT findings, and cardiac
findings
were
transformed
into
homonymous terms and broken down into
tables containing the most relevant clinical
features (Table 1). Stress was placed on the
type of catheterization procedures, the
clinical manifestation of cerebral air

Table 1. Continues.

RESULTS
We strongly believe that cerebral air
embolism in our patient occurred during the
unintentional central venous catheter
disconnection and that the air bubbles
probably traversed from the veins to the
arteries via pulmonary right-to-left shunts. A
cardiac defect was not probable because
ventriculography showed neither a patent

foramen ovale nor an atrial septum defect. In
addition, any relationship to the previous
coronary bypass
surgery and
the
postoperative minor thoracotomy after
pneumothorax was considered unlikely
because the cerebral accident occurred ≤14
days after thoracic drain removal and 17
days after coronary artery surgery.
The results of the literature search are
reported in Table 1. Including our patient,
the search encompassed 17 men and nine
women ranging in age from 19 to 79 yrs
(average, 54.7 yrs). In 12 patients (38%) the
unintended injury occurred during a
subclavian vein catheterization procedure, in
eight patients during jugular vein
catheterization (31%), in one patient (4%)
resulting from an inadvertent air injection
into an antecubital vein, and in five patients
(19%) no detailed information was provided.
Most events happened during disconnection
(14 patients; 54%), less frequently after
removal and other procedures (eight patients;
31%), or insertion (four patients; 15%). The
neurologic manifestations can be divided
into two groups. Group A (n = 14) had
predominantly encephalopathic features,
typically with an acute confusional state,
akinetic mutism (as with our patient),
various stages of coma, or generalized
seizures. Group B (n = 12) demonstrated
focal neurologic signs, of which hemiparesis
(9 of 12, mostly left-sided) was most
frequent, followed by homonymous
hemianopia (5 of 12), hemihypesthesia (1 of
12), aphasia (1 of 12), and stupor (1 of 12).
A CT scan available in 16 patients
demonstrated air bubbles in 12 patients
(75%) and signs of infarction in two patients
(13%); CT was unremarkable in two patients
(13%).
Data on cardiac defects were available
in 15 patients. Cardiac defects occurred in
six patients (40%). In nine patients,
including our patient, a physiologic
arteriovenous
pulmonary shunt
was

considered the reason for right-left shunting
(60%).
Detailed information about the
treatment of cerebral air embolism was
available for 22 patients. Specific hyperbaric
oxygen therapy, which is recommended by
some authors (1-4), was performed in only
three patients (12%). The other patients
received symptomatic treatment, including
cardiopulmonary resuscitation and intensive
care unit therapy.
Cerebral air embolism was lethal in six
patients (23%); however, nine patients (35%)
survived with severe residual neurologic
deficits and ten patients (38%) survived with
good or complete recovery. In one patient
(4%), no information on the course of the
disease was given. With regard to the two
types of neurologic manifestations, group A
with an encephalopathic pattern had a higher
mortality rate (36%) and a lower rate of
good or complete recovery (29%) than
group B, in which the mortality rate was 8%
and the rate of good or complete recovery
was 50%. Group A patients had a higher age
(mean, 63.9 yrs) than group B patients
(mean, 44 yrs).

DISCUSSION
The
initial
approach
to
our
anticoagulated patient with acute neurologic
dysfunction was to look for intracerebral
hemorrhage and to exclude toxic, metabolic,
or infectious disorders. Cerebral air
embolism was not suspected initially and,
therefore, diagnosed late. Approximately 72
hrs after the onset of signs and symptoms,
the correct diagnosis was made and the
mechanism of stroke was reconstructed.
Ideally, in patients with central venous
catheter, cerebral air embolism should be
suspected early during the course of
neurologic
dysfunction.
Appropriate
diagnostic measures useful to identify this
entity include arterial blood gas analysis,

chest radiography, echocardiography, and
CT (11-13). In addition, if air embolism
during monitoring procedures occurs, a
decrease of end-tidal PCO2 is one of the
earliest signs, reflecting the disordered
ventilation-perfusion
matching
and
increased physiologic deadspace breathing
(12).
Venous air embolism may occur
during insertion, use, or maintenance of the
catheter and after catheter disconnection and
removal (14-17). In the patients reviewed,
the most hazardous situation was
unintentional disconnection.
After venous air embolism, severe
cardiovascular and pulmonary changes
typically occur. The sequelae depend on the
amount of air, the speed of entry, and the
location within the body (12). An "air lock"
in the right heart obstructs the outflow,
resulting in right heart failure with
consequent
hypo-oxygenation
and
cardiovascular depression (12). It is worth
mentioning that through a 14-gauge needle,
if the pressure gradient is 5 cm H2O, it is
possible that 100 mL of air per second will
enter the vascular system (18). Paradoxic air
embolism occurs under cardiac defect
conditions and via physiologic pulmonary
arteriovenous shunts (19-21).
Two
types
of
neurologic
manifestations can be observed. In group A,
patients presented with clinical signs and
symptoms of encephalopathy, including
decreased consciousness, deep coma,
seizures, and sometimes overlapping focal
signs. These patients have a poor outcome
and a high mortality rate. In addition, the
mean age of this subgroup is high. Whether
in these patients a depression of the
cardiocirculatory system is leading to a
general cerebral hypoxia and ischemia
cannot be determined based on this review.
It may be speculated that the higher age may
be a risk factor because in elderly
individuals, the reserve capacity of oxygen

and energy substrates is diminished when
compared with those values in younger
individuals. Group B presents mostly
predominant focal signs pointing to the right
hemisphere and the right posterior
circulation, probably because air bubbles
follow the bloodstream directly from the
aorta to the right innominate artery or the
right vertebral artery (22).
Computed tomography is an valuable
tool in detecting cerebral air bubbles,
especially when performed soon after the
incident (13). However, it is essential that
the examiner be familiar with bubble
appearance. Late CT changes are not
specific and are difficult to distinguish from
ischemic
infarction
or
diffuse
leukoencephalopathy, as our CT follow-up
demonstrated.
Successful treatment of air embolism
consists of the following: a) stopping the air
entry; b) aspiration of the air from the right
ventricle if a central catheter is being used; c)
placing the patient in a head-down left
lateral position (Durant's maneuver); and d)
initiating general intensive care unit
measures as soon as possible, including
oxygen supply and cardiopulmonary
resuscitation. Hyperbaric oxygen treatment
in special centers is recommended (1-4);
however, its real efficacy has recently been
questioned (23). In the literature we
reviewed, hyperbaric oxygenation could
only be performed in a minority of patients,
probably because of logistic problems
involved in the long distance transport of a
critically ill patient to a specialized
treatment center. However, these three
patients all survived, two of them recovering
completely and one of them recovering well.
Despite this, the data reviewed do not
warrant the assumption that hyperbaric
oxygenation is superior to general
symptomatic treatment. In our patient, the
diagnosis of cerebral air embolism was
unfortunately delayed. At the time of correct

diagnosis, the CT scan no longer indicated
air bubbles and hyperbaric oxygenation was,
therefore, not initiated.
Paradoxic embolism is typically
considered to be related to cardiac defects,
such as a patent foramen ovale or atrial
septal defect. The results of this review,
however, show that this condition is less
frequent than hemodynamically relevant
pulmonary shunts. This is important, for
example, in neurosurgical procedures in
which cerebral air embolism cannot only
occur via cardiac defects, but also via
transpulmonary passage (24). The new
method of contrast-enhanced transcranial
Doppler during Valsalva maneuvering will
be able to detect these pulmonary shunts
(25).
In conclusion, when caring for patients
with central venous catheters, strong
emphasis has to be placed on preventing air
embolism.
General
guidelines
(Trendelenburg position, occlusion of the
needle, adequate central venous catheter
care, and occlusive dressing after removal)
should be strictly observed (11, 12).
Neurologic deterioration during central
venous catheterization procedures should
always be taken as a potential sign of
cerebral air embolism and lead to prompt
diagnosis, thus limiting cerebral damage.

REFERENCES
1.

2.

3.

Hwang TL, Fremaux R, Sears ES, et al:
Confirmation of cerebral air embolism
with computerized tomography. Ann
Neurol 1983; 13:214-215
Mader JT, Hulet WH: Delayed
hyperbaric treatment of cerebral air
embolism: Report of a case. Arch
Neurol 1979; 36:504-505
Menkin M, Schwartzman RJ: Cerebral
air embolism: Report of five cases and
review of the literature. Arch Neurol
1977; 34:168-170

4.

5.
6.

7.

8.

9.

10.

11.

12.

13.

14.

Voorhies RM, Fraser RAR: Cerebral
air embolism occurring at angiography
and diagnosed by computerized
tomography. J Neurosurg 1984;
60:177-178
Vourc'h G: Paradoxical air embolism.
N Engl J Med 1971; 285:56
Wijman CAC, Kase CS, Jacobs AK, et
al: Cerebral air embolism as a cause of
stroke during cardiac catheterization.
Neurology 1998; 51:318-319
Ponsky JL, Pories WJ: Paradoxical
cerebral air embolism. N Engl J Med
1971; 284:985
Kashuk JL, Penn I: Air embolism after
central venous catheterization. Surg
Gynecol Obstet 1984; 159:249-252
Seidelin PH, Stolarek H, Thompson
AM: Central venous catheterization
and fatal air embolism. Br J Hosp Med
1987; 38:438-439
Defalque
RJ,
Fletcher
MV:
Neurological complications of central
venous cannulation. JPEN J Parenter
Enteral Nutr 1988; 12:406-409
Feliciano DV, Mattox KL, Graham JM,
et al: Major complications of
percutaneous
subclavian
venous
catheters. Am J Surg 1979; 138:869873
Orebaugh SL: Venous air embolism:
Clinical
and
experimental
considerations. Crit Care Med 1992;
20:1169-1177
Jensen ME, Lipper MH: CT in
iatrogenic cerebral air embolism.
AJNR Am J Neuroradiol 1986; 7:823827
Dilkes MG, Dunwoody G, Bull TM, et
al: A case of intracerebral air
embolism secondary to the insertion of
a Hickman line. JPEN J Parenter
Enteral Nutr 1991; 15:488-490

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

Jacobsen WK, Briggs BA, Mason LJ:
Paradoxical air embolism associated
with a central total parenteral nutrition
catheter. Crit Care Med 1983; 11:388389
Moorthy SS, Tisinai KA, Speiser BS,
et al: Cerebral air embolism during
removal of a pulmonary artery catheter.
Crit Care Med 1991; 19:981-983
Schlotterbeck K, Tanzer H, Alber G, et
al: Zerebrale luftembolie nach
zentralem venenkatheter. Anasthesiol
Intensivmed Notfallmed Schmerzther
1997; 32:458-462
Ordway CB: Air embolus via CVP
catheter without positive pressure. Ann
Surg 1974; 179:479-481
Black M, Calvin J, Chan KL, et al:
Paradoxic air embolism in the absence
of an intracardiac defect. Chest 1991;
99:754-755
Butler BD, Hills BA: Transpulmonary
passage of venous air emboli. J Appl
Physiol 1985; 59:543-547
Tommasino C, Rizzardi R, Beretta L,
et al: Cerebral ischemia after venous
air embolism in the absence of
intracardiac defects. J Neurosurg
Anesthesiol 1998; 8:30-34
Caplan LR: Posterior circulation
disease: Clinical findings, diagnosis,
and treatment. Cambridge, MA,
Blackwell Scientific Publications,
1996
Layon AJ: Hyperbaric oxygen
treatment for cerebral air embolism:
Where are the data? Mayo Clin Proc
1991; 66:641-646
Marquez J, Sladen A, Gendell H, et al:
Paradoxical cerebral air embolism
without an intracardiac septal defect. J
Neurosurg 1981; 55:997-1000
Heckmann JG, Hilz MJ, Hecht M, et al:
Transesophageal
echocardiography

26.

27.

28.

29.

30.

31.

32.

33.

34.

and contrast-transcranial Doppler
sonography yield a higher rate for the
detection of patent foramen ovale than
one method alone. Eur J Neurol 1998;
5:S78
Grace DM: Air embolism with
neurologic complication: A potential
hazard of central venous catheter. Can
J Surg 1977; 20:51-53
Michel L, Poskanzer DC, McKusick
KA, et al: Fatal paradoxical air
embolism to the brain: Complication
of central venous catheterization.
JPEN J Parenter Enteral Nutr 1982;
6:68-70
Neubauer N, Umek H, Kristoferitsch:
A cerebral air embolism due to a
central venous catheter in the
computed tomogram. Rontgenblatter
1988; 41:462-464
Phifer TJ, Bridges M, Conrad SA: The
residual central venous catheter track:
An occult source of lethal air
embolism-Case report. J Trauma 1991;
31:1558-1560
Turnage S, Harper JV: Venous air
embolism occurring after removal of a
central venous catheter. Anesth Analg
1991; 72:559-560
Ploner F, Saltuari L, Marosi MJ, et al:
Cerebral air emboli with use of central
venous catheter in mobile patient.
Lancet 1991; 338:1331
Mennim P, Coyle CF, Taylor JD:
Venous air embolism associated with
removal of central venous catheter.
BMJ 1992; 305:171-172
Aikawa H, Mori H, Miyake H, et al: A
case of cerebral air embolism
predominant in the left hemisphere
following subclavian catheterization.
Radiat Med 1995; 13:73-76
Kearns PJ, Haulk AA, McDonald TW:
Homonymous hemianopia due to
cerebral air embolism from central

35.

36.

37.

venous catheters. West J Med 1984;
140:615-617
Halliday P, Anderson DN, Davidson
AI, et al: Management of cerebral air
embolism secondary to a disconnected
central venous catheter. Br J Surg
1994; 81:71
Hartung EJ, Sgouropoulou S, Bierl R,
et al: Severe air embolism caused by a
pulmonary artery introducer sheath.
Anesthesiology 1994; 80:1402
Yu ASL, Levy E: Paradoxical cerebral
air embolism from a hemodialysis
catheter. Am J Kidney Dis 1997;
29:453-455