Proximal and distal cannulation of the internal jugular vein for ECMO in a primate

Proximal

and Distal Cannulation of the Internal ECMO in a Primate

Jugular Vein for

By Hiroaki Kitagawa, Gordon Mccomb, Edward Ford, and James B. Atkinson

Los Angeles, California l Extracorporeal membrane oxygenation (ECMO) is lifesaving for infants with severe respiratory distress but is complicated by severe intracranial hemorrhage in 10% to 30% of patients. Intracranial venous hypertension, as a result of ligation of the internal jugular vein (IJV), has been hypothesized as a contributing factor to cerebral edema and subsequent hemorrhage. Accessory cephalad IJV cannulation may serve as a means of additional venous drainage to the pump as well as protection against intracranial venous hypertension. Proximal and distal cannulation of the IJV were studied in a primate model. The parameters monitored included sagittal sinus, right and left ventricular pressures as well as venous pressure in the ECMO circuit. The cephalad venous cannula was clamped and unclamped at 30.minute intervals. There was no significant difference in sagittal sinus or intracranial pressures during periods of cephalad cannula clamping or unclamping. Venous return was augmented when the cephalad cannula was unclamped. Cephalad cannulation has no demonstrable protective effect on intracranial, subarachnoid or venous pressures but does improve venous return to the ECMO circuit. It is concluded that cephalad venous cannulation is not necessary in all cases and should be reserved for those patients requiring additional venous drainage to support pump flow. Copyright a 1992 by W.B. Saunders Company

INDEX WORDS: (ECMO).

Extracorporeal

membrane

tored on a grass polygraph (Beckman Dynograph R-6-l 1, Fullerton, CA) and recorded at lo-minute intervals. Vessels in the right neck were used for extracorporeal bypass cannulation. The arterial cannula was placed in the carotid artery following distal ligation. Venous cannulation was accomplished by insertion of a catheter (Ellie Cath 10F) into the right atrium via the IJV. An additional SF venous cannula was placed in a cephalad direction in the IJV with the tip in the region of the mastoid process (Fig 1). Venoarterial ECMO was established. with systemic heparinization, and maintained at 400 mL/min throughout the experiment.’ The inspired 0: content (30%). mean blood pressure (90 mm Hg) and PC02 (40 mm Hg) were kept constant in order to dampen any effect of these parameters on autoregulation of cerebral blood flow. The animals were maintained under experimental conditions for 3 hours. The cephalad venous cannula was clamped and unclamped at 30minute intervals. Vital signs, intracranial venous (SSP) andventricular pressures (ICP), and circuit pressures were recorded at IO-minute intervals. Animals were killed at the conclusion of the experiment. All animals were treated in strict compliance with standards and guidelines established by the Animal Care Committee of the University of Southern California. RESULTS

During the experimental procedure the systolic blood pressures in both animals were maintained at a mean 90 mm Hg. The SSP, ICP, and circuit pressures were measured 10 minutes apart on nine occasions with the cephalad cannula open and on nine occasions with the catheter clamped. Results for ICP and SSP are summarized in Table 1 and graphically depicted in Figure 2. There was no significant change in SSP or ICP noted with or without cephalad drainage. Circuit pressures were available only from the second animal and did show significant augmentation of venous drainage at a constant flow of 400 mL/min (Fig 3). No temporal correlation was seen in blood pressure, ICP, or SSP with clamping or unclamping the cephalad cannula (Fig4).

oxygenation

E

XTRACORPOREAL membrane oxygenation (ECMO) in newborn infants is usually established by venoarterial bypass technique. Vascular access is achieved by right internal jugular vein (IJV) and common carotid artery ligation and cannu1ation.l Intracranial venous hypertension caused by ligation of the IJV has been hypothesized as contributing to cerebral edema and subsequent hemorrhage.’ An alternative to ligation of the TJV is ipsilateral placement of a cephalad catheter in addition to the standard atria1 directed cannula. The additional cannula would serve as a means of accessory venous drainage as well as to decompress presumed intracranial venous hypertension. A primate model was used to test the hypothesis of intracranial pressure changes being altered by cephalad cannulation of the IJV. MATERIALS AND

DISCUSSION

Infants supported with ECMO may show signs of venous congestion of the head and neck. This observaFrom the Department of Surgery, Childrens Hospital of Los Angeles. und the Divisions of Pediatric Suven, and Pediatric Neurosuqe:e,g Universiy of Southern California, L>osAngeles, CA. Date uccepted: April 3. 1991. Address reprint requests to James 8. Atkinson. MD, Division oj Pediatric SuTery, Box iO0, Childrens Hospital of Los Angeles, 4650 Sunset Blvd. Los Angeles, CA 90027. Copyright c 1992 by W.B. Saunders Cornpan! 0022.346819212709-0005$03.00l0

METHODS

Two adult cynomongolous monkeys (weighing 7 and 10 kg) were anesthetized using balanced technique with oxygen, narcotic. and muscle relaxant. A primate model was used because of cerebral vascular anatomy close to humans. Burr holes were placed in the skull over the sagittal sinus and both lateral ventricles. Nylon transducing catheters, 0.5 mm inner diameter. were placed into the ventricles and sagittal sinus. Pressures were continuously moniJournalofPed!atrnz Surgery,

Vol 27, No 9

(September),

1992:

pp 1189-l

191

1189

1190

KITAGAWA ET AL

Arterial Cannula

25

0

2 0

Clamped Unclamped

2

TIME

(Hours)

Fig 3. Bladder pressure was measured in the second animal studied and there was consistent augmentation of venous pressure in the circuit when the cephalad cannula was open.

Atrial Venous Fig 1. lae.

Can&la

The positions of the two venous and single arterial cannu-

tion has raised the question of whether intracranial bleeding during ECMO therapy may occur secondary to intracranial venous hypertension.2 An alternative to ligation of the IJV is simultaneous cephalad cannulation. A second venous cannula, inserted into the cephalad jugular vein, would theoretically serve Table 1. Results Case 1

Sagittal sinus pressure intracranial pressure Bladder pressure

Case 2

10.75 + 4.83

9.22 + 2.30*

8.11 + 2.28

8.60 r 1.90t

7.25 t 2.42

10.73 r 2.67*

8.31 + 4.79

10.22 + 2.64

-

-5.2

k 4.71*

6.18 + 9.75t ‘Cephalad cannula clamped. tcephalad

2.

cannula unclamped.

Sagittal Sinus Pressure

Intra-cranial

as an accessory to venous drainage as well as protect against intracranial venous hypertension. The specific goals of this experiment were: (1) to document intracranial venous and cerebrospinal fluid pressure changes generated by unilateral jugular venous ligation; (2) to determine the effect of accessory venous cephalad cannulation on intracranial pressures; and (3) to determine the ability of an accessory cephalad venous catheter to augment venous return to the circuit. The results of the study failed to demonstrate any correlation between intracranial pressures, either subarachnoid or sagittal venous sinus, and the status of drainage of blood from the cephalad directed accessory venous catheter. The lack of observed change in the pressures would suggest that the brain has sufficient alternative venous pathways to replace the ligated vessel. This observation lead to the conclusion that venous hypertension does not exist in the cerebral circulation of the primate cannulated for ECMO by unilateral cervical vessel ligation. These results confirm the findings of Stolar et al that changes in intracranial pressure in sheep associated

Pressure 7 Clamped

I

I

I

I

I

I

,

Case 1 0 Case 2

I5

l

-

o--o

7 Unclamped

n

SSP ICP

BP Cephalad Cannula

0

Clamped

[?

Unclamped 10 ECMO Flow 400nll,rn,”

P=NS

P=NS

Fig 2. The mean intracranial pressure values determined in each case, with the standard deviation. Comparison is made between values obtained with and without the cephalad cannula open, with no significant differences noted.

oiLL!0

2

1

3O

TIME (Hours)

Fig 4. The intracranial pressures varied over time without correlation to the blood pressure or status of the cephalad venous cannula.

CANNULATION

OF THE IJV FOR ECMO

1191

with ECMO are due to hypoxic disruption of cerebral blood flow autoregulation rather than a unilateral venous ligation4 The routine placement of an accessory cephalad venous cannula would not be predicted based on these results to alter the rate of intracranial bleeding complications. Adequate venous return to the ECMO pump is

critical to successful support of patients. In this experiment the cephalad cannula was demonstrated to uniformly augment venous return. It is recommended that the insertion of a cephalad cannula be considered in patients with inadequate venous return but that it is not beneficial in routine cases to prevent intracranial bleeding complications.

REFERENCES 1. Bartlett RH, Fong SW, Burns NE, et al: Prolonged partial venoarterial bypass: Physiologic. biochemical and hematologic responses. Ann Surg 180:850-856.1974 2. Vogt JF. Gangitano ES, Liberman RL, et al: Cannulation of cephalad segment of the internal jugular vein during extracorporeal membrane oxygenation in the neonate. Presented at The 5th Annual Children’s Hospital National Medical Center ECMO Symposium, Snowmass, CO. February 1989

3. Atkinson JB. Emerson P. Wheaton R, et al: A simplified method for autoregulation of blood flow in the extracorporeal membrane

oxygenation

circuit. J Pediatr

4. Stolar CJH, Reyes C: Extracorporeal causes artery 1988

significant

changes

blood flow in newborn

in intracranial

Surg 24:251-252, membrane pressure

lambs. J Pediatr

1989

oxygenation and

carotid

Surg 23:1163-1168,