2021 Coagulation - oxygenator changeout

Permission to print: Yes

Category Coagulation
Incident type Good Catch No Harm Incident
Duration of incident: minutes
Description: Patient 39 YO male Ht 175cm and weight 58.5 KG with a history of RHD, S/P balloon MV
valvuloplasty. Procedure AVR and MVR with CPB. The circuit consisted of a Sorin S-3
roller pump, Sechrist gas blender, Terumo FX15 oxygenator and Terumo custom tubing
pack with X coated tubing set with 3/8 - 3/8” ID AV loop and a ½" ID boot. Monitoring
consisted of a Specturm M-4 Viper monitor. The patient was heparinized with 300u/KG
of unfractionated heparin. The post heparin ACT was 435sec. Since there was 10,000u of
heparin in the prime we proceeded to CPB. After initiating CPB the flow was increased to
4.2l/mon. The PaO2 was 136mmHg with 100% FiO2. ? This was lower than expected, Our
center is located 6300ft above sea level and our hospital oxygen system consists of an
oxygen concentrator system. Often our wall oxygen is between 82-95%. I had set my FIo2
at 100% but at that time the Oxygen Concentrator was put out 80-85%. The PO2
increased with cooling to 32C. As cooling proceeded the PaO2 increased to 180mmHg.
The initial post membrane line pressure was 195 mmHg. Fifteen minutes after the heart
was arrested with 1,000 ml of Del Nido Cardioplegia, it was noted that the post
membrane pressure dropped to about 150mmHg and there was a blood leak at a
connector before the oxygenator. The pre membrane pressure was measured and found
to be approximately 600mmHg. (Trans membrane pressure ~450, should be 100-140).
Additional 10,000 units of heparin was administered (subsequent ACT492). A change out
in this case was decided primarily due to high pressure excursion. I thought the circuit
would explode if we continued. Oxygenation was a marginal secondary problem The
membrane was changed without coming off bypass using the PRONTO Technique. Our
circuit has a 3/8 bridge around the oxygenator. The pre and post membrane pressure can
be measured from LL connection by simply moving a clamp. An inspire 6 oxygenator was
added to the circuit. The post membrane pressure increased to 220mmHg and the PaO2
increased to 400mmHg. After CPB the first oxygenator was rinsed and inspected. There
was high resistance across the oxygenator and visible clot seen in the fiber bundle.
GOOD CATCH - what went well The surgeon and anesthesia team was informed about the problem. The PRONTO
bridge in our circuit allowed us to check both pre and post membrane pressure.
The second oxygenator could be added to the circuit without coming off bypass.
There was no interruption of blood flow to the patient or signs of lactic acidosis.
We have used a PRONTO bridge in our circuit for every procedure.
What could we do Administered more heparin when the initial ACT was less than 480sec. This patient had
better double valve disease and a presumably higher blood volume. Checked the pre membrane
pressure and lower PO2 pre X clamp.
Preventive actions We might have considered loading with 400u/KG of heparin. A formal check that
perfusion is all ok including the pre membrane pressure before aortinc cross clampis
applied.
Hospital incident filed: No
Ext Authority Advised No
Rule issue Yes
Skill issue No
Discussed with team: Yes
Patient outcome variance Nil
Commentary This incident exemplifies a good catch with preventive action planning - using the
PRONTO bridge for seamless expeditious oxygenator changeout. This procedure was
published in 2002 and is attached. A youtube video shows the ease of use of this
technique. Https://www.youtube.com/watch?v=kehXUDGjZj8 . A very recent report to
 PIRS on an oxygenator change out fortunately able to be made before cross clamp
application elicited a correspondent to PIRS to question why the PRONTO procedure was
not in routine use. This report clearly demonstrates the safety advantage of the PRONTO
technique and centres should give serious consideration to its adoption. PIRS Ed

Washed out oxygenator showing clot

 Perfusion 2002; 17: 447 ± 450

Parallel replacement of the oxygenator that is

not transferring oxygen: the PRONTO
procedure
RC Groom, RJ Forest, JE Cormack, KS Niimi and J Morton
Maine Medical Center, Portland, Maine, USA

We present here a technique to replace a failed oxygenator procedure may be easily applied to any standard CPB
by inserting a second oxygenator in parallel (PRONTO) circuit with a few minor alterations. The technique is
within the cardiopulmonary bypass (CPB) circuit. Oxygen- simple; it can be carried out rapidly. An important
ator failure is a potential hazard that may result in patient advantage of this technique is that it may be executed
injury or death. Although failures are rare, safety surveys without interrupting blood flow to the patient, which may
conducted over the last 25 years suggest that the inci- reduce the incidence of patient injury or death. Perfusion
dence of oxygenator failures is on the rise. This emergency (2002) 17, 447 ± 450.

Introduction

Failure of the oxygenator during cardiopulmonary Circuit modification and change-out

bypass (CPB) presents the surgical team with a seri- procedure
ous and potentially fatal problem requiring a swift
and well-coordinated response. Unfortunately, the
The CPB circuit is routinely set up with a shunt
incidence of oxygenator failure is on the rise. 1 ± 4 We
inserted between the inlet and outlet of the oxygen-
describe here a CPB circuit modification that allows ator using a 10-in. length of 3/8-in. tubing and two
rapid parallel replacement of an oxygenator not trans-
3/8 `y’ connectors. A luer-lock connector is placed in
ferring oxygen (PRONTO).
the center of the shunt. This luer-lock site may be
If poor oxygenation occurs during CPB, the surgi-
used to monitor the arterial line pressure, and by
cal team should be informed. The use of a checklist or
clamping the shunt distal or proximal to the luer-
oxygenator failure algorithm can be a useful tool to lock, pre- and postmembrane pressures may be
check for correctable malfunctions such as a leak in
measured as well.
the oxygenator gas supply, a malfunction of the oxy-
The equipment required for executing the change-
gen blender, or an incorrect blood flow path. These
out includes one additional tubing clamp, a pair of
components should be examined rapidly and thor-
sterile scissors, an alcohol swab, and a replacement
oughly. If the heart is beating and it is possible to oxygenator.
ventilate the patient and separate from CPB, this
In the event of an oxygenator failure, additional
should be carried out immediately.
volume, equivalent to the priming volume of the
If, however, the surgical procedure is at a point
membrane bundle and the heat exchanger, must be
where it is not possible to separate from CPB, the
added to the venous reservoir. A clamp is placed on
decision must be made expeditiously to replace
the oxygenator bypass line as close to the distal `y’ as
the oxygenator while on bypass. The PRONTO pro- possible (Figure 1). The tubing is swabbed and then
cedure described herein allows rapid transition to a
divided with sterile scissors equidistant between the
functioning oxygenator, without interrupting blood
luer-lock connector and the distal `y’ (Figure 2).
flow.
The new oxygenator is connected to the two exposed
ends of the tubing (Figure 3). The recirculation lines

Address for correspondence: Robert C. Groom, BS, CCP, Cardiac

Surgery, Maine Medical Center, 22 Bramhall Street, Portland, Presented at the Annual Seminar of the American Academy of
Maine 04102, USA. Cardiovascular Perfusion, New Orleans, LA, 27 ± 30 January
E-mail: [email protected] 2001.

© Arnold 2002 10.1191/0267659102pf606oa

 The PRONTO procedure
RC Groom et al.

448

Figure 1 Isolating the bridge in preparation for parallel replace- Figure 3 Replacement oxygenator placed in parallel with failed
ment of the oxygenator. oxygenator.

from the new oxygenator are then connected to the with this procedure, which should require less than
cardiotomy reservoir and are left open. The clamp at 90 seconds to complete.
the inlet of the oxygenator is removed, allowing the
unit to fill and causing air to be expelled through
the recirculation lines (Figure 4). The oxygenator is Discussion
tipped and tapped gently to allow all of the air from
the connections to be dispelled. After visual inspec- A 1980 survey of open heart centers in the USA
tion to confirm that all of the air has been purged from determined the incidence of oxygenator failure to
the device, the gas line is moved to the replacement be 1:56 000.1 Similar USA surveys conducted in
oxygenator, the recirculation lines are closed, and the 1986 and 1999 showed oxygenator failure rates of
clamp distal to the replacement oxygenator is 1:13 600 and 1:2700, respectively.2,3 Surveys of the
removed and placed proximal to the inlet of the failed UK and Ireland in 1993 and 1997 showed oxygenator
oxygenator (Figure 5). The patient is then monitored failure rates requiring change-out of 1:4034 and
to confirm adequate oxygenation. After adequate oxy- 1:4631, respectively.4 This increasing rate of oxygen-
genation has been established, the heat exchanger ator failures is alarming. However, the incidence
may be connected to the water supply as necessary. remains so infrequent that it has been difficult to
During this procedure, safety devices such as the air definitively determine the causes of these oxygenator
bubble detector and low-level alarm remain func- failures. A number of proposed mechanisms have
tional. The arterial line pressure will be reading been suggested as potential causes.
premembrane pressure and, if desired, a purge line Palanzo et al. reported a rapid rise in oxygenator
from the arterial filter can be connected to the transmembrane pressure shortly after the initiation
pressure monitor to determine transmembrane pres- of CPB.5 It was believed that this pressure excursion
sure. There are a minimal number of steps involved may have been triggered by cooling the patient and

Figure 2 Bridge divided equidistant between luer lock and dis- Figure 4 Removal of proximal clamp, allowing replacement oxy-
tal `y’. genator to prime.
 The PRONTO procedure
RC Groom et al.

449
8
starches in the CPB priming solution. Palanzo
et al. have shown that the addition of albumin in
the prime solution prior to CPB can be protective for
platelets. Furthermore, this protective effect occurs
with as little as 2 g of albumin in the prime
solution.9
It has also been suggested that sodium nitroprus-
side or nitric oxide may ameliorate this phenomenon
by decreasing platelet activation. 10,11 In a study by
Rauch et al., nitric oxide was introduced into the
sweep gas of an in vitro circuit. The scanning electron
microscopy study of the oxygenator membranes
showed a decrease in fibrin and cellular deposits in
Figure 5 Clamp moved from outlet of the replacement oxygenator the membranes that had nitric oxide added to the
to the inlet of the failed oxygenator.
sweep gas compared to membranes that were venti-
lated without nitric oxide. 11
Wendel et al. examined the incidence of high trans-
was alleviated upon warming. This same phenom- membrane pressure gradients and the use of coated
enon was reported by other centers in the USA, circuits. They found a higher rate of abnormally high
Canada, and Europe, occurring with various hollow transmembrane pressures in the noncoated circuits
fiber oxygenators.6 Although the exact mechanism when compared to coated circuits. The group
remains unknown, there have been numerous hypothesized that the coated circuits caused less
hypotheses offered. Bearss et al. attribute these blood activation, which may play a role in the build-
increases in transmembrane pressures to the deposi- up of fibrin and platelets in membrane oxygenators.12
tion of cold reactive proteins upon the surface of the Parallel replacement of a hollow-fiber oxygenator
heat exchanger, causing a narrowing of the blood was reported as early as 1988 by Hart et al.13 The
path.7 Schaadt believes that the increased incidence authors’ technique required the placement of sections
of these high transmembrane pressure gradients is of tubing and connectors at the oxygenator inlet and
due to oxygenator designs that focus on increasing outlet. They also recommended the flushing of the
efficiency while decreasing prime volume, and device with carbon dioxide. The technique required
describes that, in an effort to design high-efficiency, dividing the tubing in two places. They also reported
low-prime oxygenators, manufacturers must de- that the advantage of parallel replacement is that
crease the blood channel dimensions. Thus, any it eliminates the period of circulatory arrest. They
further reduction in these channels’ dimensions, further stated that it minimizes the amount of hemo-
caused by depositions, would cause a marked in- dilution that occurs when a failed oxygenator is
crease in resistance to blood flow and an increase in removed from the circuit and replaced.
transmembrane pressure.6 It has also been noted that An oxygenator failure is a potentially catastrophic
these pressure gradients are not limited to low-prime event that necessitates an immediate and precise
oxygenators, but that this phenomenon is more response. Whenever possible, ventilation of the lungs
pronounced and occurs sooner in these than in and controlled separation from CPB are the best
higher-prime oxygenators.5,6 option, provided the heart has not been arrested,
Others have suggested that these increases in and the heart and aorta are intact. The PRONTO
transmembrane pressures are caused by the activa- technique allows rapid replacement of a failed oxy-
tion of platelets and their subsequent deposition genator, there are a minimal number of steps required
with fibrinogen on the membrane surface. Aniuk to perform, and, more importantly, it is safe as there is
et al. have noted that many of these occurrences no interruption in blood flow and all major safety
have been associated with the use of hydroxyethyl devices remain active and functional.

References
1 Stoney WS, Alford WC, Burrus GR et al. Air embolism 3 Mejak BL, Stammers A, Rauch E et al. A retrospective
and other accidents using pump oxygenators. Ann Thor- study on perfusion incidents and safety devices. Perfu-
ac Surg 1980; 29: 336 ± 40. sion 2000; 15: 51 ± 61.
2 Kurusz M, Conti VR, Arens JF et al. Perfusion accident 4 Fisher AR. The incidence and cause of emergency oxy-
survey. Proc Am Acad Cardiovasc Perfus 1986; 7: 57 ± 65. genator changeovers. Perfusion 1999; 14: 207 ± 12.
 The PRONTO procedure
RC Groom et al.

450
5 Palanzo DA, Manley NJ, Montesano RM et al. Potential 10 Mellgren K, Mellgren G, Lundin S et al. Effect of nitric
problems when using the new lower prime hollow-fiber oxide gas on platelets during open heart operations. Ann
membrane oxygenator with uncoated stainless steel heat Thorac Surg 1998; 65: 1335 ± 41.
exchangers. Perfusion 1996; 11: 481 ± 85. 11 Rauch E, Stammers A, Mejak B et al. The effect of nitric
6 Schaadt J. Oxygenator thrombosis: an international phe- oxide on coagulation during simulated extracorporeal
nomenon. Perfusion 1999; 14: 425 ± 35. membrane oxygenation. J Extra-Corpor Technol 2000;
7 Bearss M, Ericson D, Parks P et al. The puzzling case of 32: 214 ± 19.
the cryoprecipitates. AMSECT Today 1996; June: 21 ± 12 Wendel H, Philipp A, Weber N et al. Oxygenator throm-
25. bosis: worst case after development of an abnormal
8 Aniuk L, Miller M, McAdoo S et al. The phenomenon of pressure gradient-incidence and pathway. Perfusion
pressure excursion (`oxygenator vasoconstriction’). Proc 2001; 16: 271 ± 78.
Am Acad Cardiovasc Perfus 1999; 20: 73 ±76. 13 Hart MA, Abshier DA, Pacheco SL et al. A technique for
9 Palanzo DA, Zarro DL, Montesano RM et al. Albumin in the change-out of a malfunctioning membrane oxygen-
the cardiopulmonary bypass prime: how little is enough? ator without terminating cardiopulmonary bypass. Proc
Perfusion 1999; 14: 167 ± 72. Am Acad Cardiovasc Perfus 1988; 9: 105 ± 18.