The Society of Clinical Perfusion Scientists of Great Britain and Ireland and The College of Clinical Perfusion Scientists of Great Britain and Ireland
Perfusion Reports

Welcome to the Perfusion Reports Page.

Perfusion Reports are taken from the Safety Committee web based reporting system and after appropriate anonymising disseminated to the membership. The content should be informative and hopefully encourage similar sharing from the perfusion community.

Details of incident/Nature of defect:

60year old 80kg male patient. Blood group B+ve. Previous PCI During coronary artery bypass grafting surgery, once Cardiopulmonary bypass (CPB) commenced, there were very high pre-oxygenator P1 line pressures in the bypass circuit (450mmHg). The oxygenator was a Medtronic Fusion. P2 post-oxygenator pressures were within 'normal' range. All potential causes were eliminated but it persised.

The high P1 pressures were not resolving and the p02 oxygen measurement started to fall gradually, although not compromising patient oxygenation. Clotting levels were within range and no clots were seen in the oxygenator. The surgeon was kept informed and the situation monitored before progressing with operation. The decision was made to change out the oxygenator. The anaesthetist was informed and the patient was ventilated conventionally.

CPB was stopped and the oxygenator changed. CPB recommenced and the same high P1 pressures seen again but with no compromise to oxygenation so bypass was continued and monitored very closely. After 30 minutes, the pressures fell to 'normal' levels and remained that way for remainder of the bypass. It was also noted that patient haematocrit on CPB was high (38%) and following discussion with the anaesthetist the patient was diluted with the addition of crystalloid.

Post CPB and protamine, a distal vein graft needed repair and heparin was administered and CPB recommenced. P1 and P2 were both within 'normal levels'. As high P1 pressures seen with both devices on initial use, this was suggestive of a patient specific incident. No untoward effects were noted in the post-op period.

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

First oxygenator returned to Medtronic for evaluation. No fault was found with the oxygenator.

In-house discussion:

Following the incident there was a feeling that perhaps the oxygenator was changed out too soon and high pre oxygenator pressures would have stabilised.

Details of incident/Nature of defect:

The brand of heparin used in Cardiac Surgery in the hospital was changed. Immediately difficulties were noted opening the new branded ampoules. Two perfusionists received minor cuts and a number of ampoules shattered on opening and others left jagged edges. Glass splinters were observed in a roller pump.

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

In excess of 20 ampoules were returned to pharmacy whose initial response was to buy ampoule breakers. Ampoules continued to break unevenly with ampoule breakers and the pharmacy were contacted again.

Reply from Hospital Pharmacy

"I have highlighted this issue with the All Wales contracting group, and I have also shared this with other pharmacy procurement colleagues across Wales. Cardiff & Vale have also highlighted similar issue with the 20ml ampoules. The suppliers (Kent Pharmaceuticals) have also been contacted, and they are in the process of sourcing new manufacturers in view of the issue with Fannin brand. Once I hear of any update, I will let you know."

Details of incident/Nature of defect:

A very unusual emergency for a Thoracic procedure occurred where, the perfusion and cardiopulmonary bypass support was requested. The dry build pump which was set up earlier has been wheeled in to the emergency theatre (Non cardiac). It was necessary to institute femoral-femoral CPB to support the patient in an uncontrolled and extremely quick situation. The emergency was such and the instability of patient made the sucker test not performed before it was put in the pleura (patient was heparinised and ACT was greater than 250 secs)

It was made obvious from the scrub team that the sucker from the was set up in wrong direction and was blowing rather than sucking. Due to the experience of the perfusionist the tubing was quickly changed and no consequence to the patient was seen.

Details of injury (to patient/healthcare professional)

No injury

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

The manager has been informed and local DATIX has been filled in.

Details of incident/Nature of defect:

During a 3 day period this unit experienced 2 incidents of abnormally high trans-membrane pressure gradients (TMP) during the early stages of cardiopulmonary bypass. The department routinely utilises roller pumps for bypass and measures pre- and post-membrane pressures for all cases. Abnormal gradients are defined within the unit as P >300mmHg, with attention to related factors e.g. flow etc. Consideration for oxygenator change out is only made when perfusion becomes limited and potentially inadequate. For the cases there were no observed common factors e.g. surgeon or perfusion scientist; however the same oxygenator model (Sorin Inspire 6F) was utilised but each had a different batch number.
 

Case 1. The patient was an average sized 18 year old male scheduled for redo sternotomy for an aortic root replacement having previously undergone a valvotomy in childhood. After sternotomy and heparinisation, a small bleed occurred requiring quicker than expected bypass initiation however a longer than adequate ACT had already been achieved. Initially full flow was reached, with a normal TMP 150mmHg. Cooling to 30°C was commenced. More dissection of the heart was required for mobilisation and therefore the aortic cross clamp was not applied. After about 20mins on bypass the TMP gradually increased to >300mmHg. The circuit was checked for physical obstructions, a second perfusion scientist was consulted, the surgeon was informed and the case perfusion scientist requested to rewarm the patient hoping this may resolve the situation.  The pressure however continued to rise, the pre-membrane pressure reaching in excess of 600mmHg (TMP>450mmHg). After discussion, and as the bleeding issue had been resolved but knowing a prolonged bypass was expected the decision was made to come off bypass and change the oxygenator. The patient was ventilated, filled and normal ejection was established for change out. This occurred without any untoward event, bypass was re-established, and normal pressures were observed and maintained for the duration of the case. The patient was weaned successfully and underwent an uneventful post-op course. There was no observed effect to the patient from this event.



Case 2. A 66year old unremarkable male was scheduled for a primary mitral valve repair. Bypass was initiated after an ACT=541 and the patient cooled to 34°C according to specific surgeon request, TMP 200mmHg. After about 30mins on bypass and the application of the aortic cross clamp and opening of the heart chamber increasing TMP was noted, again the circuit was checked for obstruction or error and another Perfusion Scientists opinion requested. The pre-membrane pressure continued to rise to >600mmHg , TMP>400mmHg. The situation was discussed with the surgeon and as the heart was open cooling of the patient commenced. The TMP increased to >500mmHg despite reducing arterial flow (and therefore ABP) and the decision was made to replace the oxygenator. Change out occurred at 31°C, CPB was interrupted for 1.5 minutes and re-established without incident. Normal pressures returned and the bypass, weaning and post-op period occurred uneventfully. No negative effects were observed in the patient, neurological or otherwise. 


Details of injury (to patient/healthcare professional)

No injury


Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

In both cases the oxygenators were preserved and returned to the manufacturer for analysis.

The findings are outlined below:

“DHR verification did not reveal any relevant information possibly linked with the claimed event….
 Significant haematic deposits were found in both oxygenator and heat exchanger fibre mats.
…this... is classified as a rare event, with a 0.03% global complaint rate in 2015.
Most likely the root cause of the issue is multi-factorial, where a combination of triggering factors might be originating the phenomena including interaction with clinical procedure and patient. 
A number of activities…are on-going to identify the possible root causes of the issue…
Internal R&D, QA and marketing resources together with Consultants from national and international Universities are working on the matter”



The department has reviewed its protocols for dealing with these cases after an extensive review of the relevant literature; however each case needs to be assessed relative to its immediate situation. The decision whether to change out an oxygenator lies with the case perfusion scientist following discussion with the Consultant Surgeon and Anaesthetist.

Details of incident/Nature of defect:

The units custom-made ECMO circuit contains a soft shell reservoir bag to assist with priming and de-airing. During the set-up of an ECMO circuit crystalloid fluid was added to the reservoir bag and the centrifugal head, oxygenator and associated tubing circuit primed passively. After correct positioning of the head in the system driver and applying power to the circuit no flow could be produced. The circuit was checked for obstruction and on close inspection it was found that the outlet of the bag was almost completely adhered to the bag structure. Cavitation was observed in the centrifugal head inlet line. The application of negative pressure to the outlet effectively sealed the bag preventing flow. The seal was pulled apart sufficiently enough to allow flow however slight damage to the bag occurred. The bag is always removed from the circuit prior to the initiation of ECMO and after adequate de-airing this was performed in the usual manner. ECMO was then established in the patient uneventfully.

Details of injury (to patient/healthcare professional)

No injury


Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

The bag was kept and returned to the manufacturer.


The following response was received.


"This is an isolated incident, unfortunately caused by an operator error during manufacture. The connector was over bonded when inserted into the outlet at the bottom of the reservoir bag where the bonding together occurred. Operatives have been made fully aware of the issue, an extra quality check along with details of this complaint have been added to the relevant stage of manufacture, which should prevent reoccurrence"



The other members of the department were made aware of the incident so that the appropriate checks could be made to subsequent ECMO circuits when used.


Details of incident/Nature of defect:

The oxygenator (a Sorin Inspire 8F) was setup as per our normal and pressure tested for 10 minutes prior to priming. All routine pressure / occlusion checks were performed and passed. The pump was left to recirculate during mammary harvest at 140mmHg. Upon return, we noticed several ‘micro’ bubbles in the oxygenator near the white recirculation port but no visible leak which was de-aired with flow and pressure being applied. A very small leak was noted at this point but the location of the leak was difficult to visualize. Therefore, all luer lock connectors for the sampling manifold and both red and white recirculation ports were tightened, at which point the leak seemed to stop.



We continued with an arterial R.A.P which was uneventful. The venous tubing was then R.A.P.’ed and it was this at this point where a fast and almost continuous blood leak was noted out the bottom of the oxygenator seemingly originating from inside or around the temperature port.


Details of injury (to patient/healthcare professional)

No injury


Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

Lines were clamped off to patient and oxygenator was changed out.


Details of incident/Nature of defect:

For a period of time Sorin Inspire 8F oxygenators were arriving detached from the plastic holding arm. As a consequence the oxygenator could not be mounted either in a holder or onto the reservoir.

This connection should be a permanent connection.

We were advised by the company that as long as they did not arrive detached the bond would be secure. Two incidences then occurred where the device was checked and appeared intact but subsequently became detached - once during building and once during priming.

Details of injury (to patient/healthcare professional)

None

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

Supplier contacted on each occasion. This covered numerous lot numbers. After the incidences of detachment during building and priming we contacted the supplier again. They were unable to assure us that the problem had been fully identified and resolved and could not guarantee that there was no risk of an oxygenator falling off the holder during use. We have discontinued the use of the device and all packs have been returned to Sorin.

Details of incident/Nature of defect:

After 2 days on ECMO, the paragon oxygenator started to leak gas from around the oxygenator gas inlet top plate.

Details of injury (to patient/healthcare professional)

This showed as reduced patient SaO2 and poor post oxygenator blood gases.

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

This is the second such leak in 2 weeks, same batch number. All other products with this batch number has been removed from stock and quarantined for collection by manufacturer.

Details of incident/Nature of defect:

IABP Console (Maquet) was active on a patient. The nurse was taking obs and re-zeroed the pressure trace. As she did so, the screen went blank. The pump turned off, stopped ballooning and emitted a constant alarm tone. The power to the pump was switched off and on again to see if this would rectify the problem and it did not.

The service representative was unable to retrieve any fault logs. Due to the age and the cost of a potential Main PCB replacement the IABP was condemned.

Details of injury (to patient/healthcare professional)

None.

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

On call perfusionist was called in, console was swapped out for a newer model and therapy resumed.

Details of incident/Nature of defect:

Once the pump was primed and left circulating in theatre, it was noticed that there were bubbles in the prime. Gas was leaking from the gas phase into the blood phase from the bottom of the oxygenator module (Sorin EOS). The pump and oxygenator were removed from theatre and swapped out for a new one.

Details of injury (to patient/healthcare professional)

None

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

Pump/oxy swapped out, device returned to manufacturer for investigation.

Details of incident/Nature of defect:

Device (Paragon Midi oxy) found to be leaking from heat exchanger whilst priming.

Details of injury (to patient/healthcare professional)

None

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

Device changed out

Details of incident/Nature of defect:

During cardiopulmonary bypass (AVR) the perfusionist noticed blood in the water lines that connect the heater cooler unit to the oxygenator's (paragon) heat exchanger 37 minutes in to bypass. This suggests a leak in the heat exchanger and the patients' blood and water could be mixing.

The water lines were immediately switched off and disconnected, and the consultant surgeon informed. The patient remained on bypass while a new bypass circuit was prepared. Bypass was temporarily stopped and re-commenced with the new circuit using a new heater-cooler and the patient fully rewarmed. Bypass was discontinued and the patient was stable. It should be noted that a full heat exchanger leak test was performed before the bypass circuit was primed and no fault was detected (water circulated at 37oC for 5mins prior to adding prime solution, no pressure test). Blood and water samples were sent for testing. Antibiotics were given by the consultant anaesthetist.

Details of injury (to patient/healthcare professional)

Patient stable and extubated following day. Blood samples sterile to date.

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

Microbiology was informed and antibiotics were given. A water sample from the heater cooler unit was sent to the lab for investigation, and the oxygenator was sent to Chalice. A Datix report has been completed and MHRA informed by Chalice.

Details of incident/Nature of defect:

Vent Pump Boot tubing inserted in pump the wrong way round. Wet test failed to show up.

New bypass machines were in use.

The tubing pack was still under modification to fit new machines. Vent sucker required extending to reach new reservoir position.

Wet test performed by scrub person and confirmed as being okay. PA vent inserted, surgeon immediately reports that there was no suction. Suction was stopped and the occlusion was checked, but still no suction. Wet test repeated and the vent observed blowing bubbles. The perfusionist spots a pump boot error and reverses the boot into the correct orientation. Wet test repeated, vent observed sucking. Case proceeds as normal. Anaesthetist was viewing TOE throughout and reported no air. Patient recovered as normal.

Details of injury (to patient/healthcare professional)

None.

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

Discussed and raised awareness of potential errors with the introduction of new equipment. Modification of tubing sets to eliminate the need to extend the vent sucker and also increase the visibility of the connector for the vent at the inlet side of the pump. Review of all sucker tubing indicators in the operative field.

Increased robustness of wet test.

It was felt that the failure of the wet test may have been due to the use of a connecting silicone 'U' bend that connects 3/8th inch arterial to 1/2 inch venous in the tubing sash. This is cut out of the system by the scrub person and retains some priming fluid which was being dipped into for the wet test. It may have indicated a positive result as a sucker tip pushed into the 3/8th inch side of the 'U' could displace fluid up the 1/2 inch side which would then naturally swing back up the sucker tip. Based on this the wet test is now performed from an open jug and visualised and confirmed by the perfusionist and the scrub person.

One way valves in all suction lines was highlighted as well as a final dip test by the operator prior to insertion of intra cardiac suction. Neither of these measures are currently in place.

Better indicators on the pump for direction off flow are being explored.

Details of incident/Nature of defect:

Within minutes of initiating bypass (paragon oxy), blood was observed to be entering the water lines of the heater cooler unit. Bypass was immediately suspended and a new circuit connected to the patient. Bypass was re-commenced and continued uneventfully, the failed unit was removed for investigation.

Details of injury (to patient/healthcare professional)

No immediate injury observed. Blood gas parameters were within normal ranges and microbiology screens negative at reporting date.

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

Local investigation, MHRA informed, supplier informed. The use of this device has been suspended pending outcome of the investigation.

Details of incident/Nature of defect:

S3 pump control knob unresponsive and unable to increase or decrease blood flow on bypass. E62 error noted on control panel. RPM remained fixed but flashed visibly to 0 RPM whilst blood flow readings remained constant.

If control knob turned to reduce or increase RPM, there was no response however, LED dot marked the current RPM value.

Details of injury (to patient/healthcare professional) -

No injury to the patient.

Action taken (includes any action by patient, carer or healthcare professional, or by the manufacturer or supplier)

The fault occurred 10 mins before the end of bypass during de-airing of the heart after mitral surgery.
The surgeon informed of fault and second perfusionist brought into theatre for support.
Unable to reach calculated flow, blood gases were monitored closely to ensure adequate perfusion.
Patient tolerated enforced low flow levels.
De-airing of the heart performed using clamps to control blood outflow from the pump unit.
Patient weaned from bypass safely again using clamps.
Pump rebooted once safely off bypass which cleared the fault.
Critical incident reported at local level.
Heart lung machine removed from service and colleagues informed of incident.
Sorin engineer called to investigate and replace the faulty unit. Thought to be a fault with the drive unit not the control panel.

Details of incident/Nature of defect:

Before bypass a small amount of fluid was noticed on the floor beneath the oxygenator, although this was believed to be condensation. On bypass a small amount of blood was noticed to be leaking from the oxygenator fiber block. The volume was minimal, approximately one drop every 5mins, the surgeon was informed and it was agreed to continue the case without further action unless the leak volume increased.

Details of injury (to patient/healthcare professional):

None.


Action taken (includes any action by patient, carer or healthcare professional or by the manufacturer or supplier):

The case proceeded without any intervention. Following the procedure the oxygenator was flushed and retained for analysis by the supplier. The subsequent report from Medtronic stated the following:

'At Medtronic’s Quality Laboratory, inspection showed a blood path through the mandrel-to-heat exchanger bond. No fiber leak was observed. The leak in this case was coming from a mechanical joint and clearly not a ‘water to blood’ leak across the heat exchanger. This was a leak path to the outside of the device due to a bonding issue that has since been corrected in production, you can see from the diagram where the leak path to the outside of the device is located and from where the Affinity in question was leaking.

Medtronic has identified that mandrel-heat exchanger bond leaks are most likely related to a UV adhesive void in the bond joint. In response to these leaks, an internal corrective action was initiated and resulted in improved manufacturing process. The mandrel is actually a mechanical joint between two pieces of plastic bonded together. It is the mating joint where the heat exchanger is attached to the oxygenator bundle, last step in the assembly process. The diagram shows where the joint (i.e. mandrel) is located. The joint is designed so that it is always under positive pressure and leaks from the inside out, thereby being a fluid leak from the blood side of the oxygenator to atmosphere.'

Fault pic

 

Details of incident/Nature of defect:

During cardiopulmonary bypass (MVR+CABG) blood was seen in the water lines that connect the heater cooler unit to the oxygenator’s heat exchanger. This obviously suggests a leak in the heat exchanger and the patients’ blood and water could be potentially mixing. The water lines were immediately switched off and disconnected, and the consultant surgeon informed.
The recommended procedure is to change out the oxygenator following heat exchanger leak. However, the consultant cardiac surgeon felt that this was not a safe option at the time as the oxygenator was still working, so we remained on bypass. The cross clamp was removed and the patient was rewarmed from 28°C to 34°C using the separate heat exchanger on the cardioplegia circuit and two fluid warmers donated by the anaesthetic team. Once the patient reached 34°C, the patient was ventilated and bypass was terminated and the oxygenator replaced with a new one. Bypass was re-commenced and the patient fully rewarmed. Bypass was discontinued and the patient was stable. It should be noted that a full heat exchanger leak test was performed before the bypass circuit was primed and no fault was detected (water circulated at 37oC for 5mins prior to adding prime solution, no pressure test).

Details of injury (to patient/healthcare professional):

Unknown. Patient stable and extubated the following morning.


Action taken (includes any action by patient, carer or healthcare professional or by the manufacturer or supplier):

Microbiology was informed and antibiotics were given. A water sample from the heater cooler unit was sent to the lab for investigation, and the oxygenator was sent to Medos. A Datix report has been completed and MHRA informed by Medos.

Details of incident/Nature of defect:

Avant D903 has a dual chamber reservoir, i.e., The cardiotomy reservoir can be separated from the venous reservoir. There was a leak of the priming fluid observed before going on Cardiopulmonary Bypass. There might have been a crack or not sealed enough (Inadequate Glue) where both the cardiotomy and venous chambers join.

Details of injury (to patient/healthcare professional):

None


Action taken (includes any action by patient, carer or healthcare professional or by the manufacturer or supplier):

Have informed the Surgeon, Anaesthetist about the leak and changed the entire reservoir for another one. Since it had been noticed while priming and before going on CPB no harm had been done to anyone.

Details of incident/Nature of defect:

Trainee Perfusionist added an extra sucker to a pre built circuit & failed to complete the 3 perfusion sucker checks. The 4th and final sucker check, a visual wet test of the suckers with the scrub nurse was only completed for 2 of the 3 suckers i.e. not the additional sucker.

The additional sucker was used as a vent and after x clamping the surgeon asked for the vent to be turned on and noted it was blowing not sucking, this was switched off and corrected. The surgeon was adamant that the Aorta was X clamped before the vent was switched on and that no air entered the patient.

Details of injury (to patient/healthcare professional):

Post operatively the patient showed Lt sided stroke and the family was informed of the incident involving the reversed suction tubing.

The patient and family have been kept fully informed as to the possible reasons for the stroke - and that stroke is a known risk from cardiac surgery and cardiopulmonary bypass.

The family has also been kept fully informed of the mistakes made during setup and checking of the suckers.


Action taken (includes any action by patient, carer or healthcare professional or by the manufacturer or supplier):

A Hospital Incident Report was filed on the day of the incident. A panel met and discussed the incident and decided to conduct a full SUI investigation with outcome of investigation sent to family and the SHA.

A full review of practices has been undertaken, this has resulted in a number of changes to procedure and/or protocols.

Details of incident/Nature of defect:

Patient commenced on cardiopulmonary bypass, flow/line pressure normal, on bypass just over 1 minute venous blood appeared very blue as did arterial blood, had just cross clamped and cardioplegia just running, informed surgeon and anaesthetist problem with blood gas blender. FiO2 at 60%, and gas flow at 2.5L, on commencing bypass. Adjusted FiO2 to 100%, and gas flow to 10L, palpated end of air/O2 line with finger, very little air/O2 going into oxygenator. Traced line back to vaporiser, large tear observed on tubing at back of he vaporiser, line seemed perished. Anaesthetist put tape/sleek on tubing and a tie wrap, which solved problem. Poor oxygenation for over 2minutes, blood gases were pO2=6.7kPa, PCO2=6.8kPa. Repeat blood gases after the repair were PO2=53.2kPa, PCO2=7.2kPa. Adjusted FiO2 and gas flow then repeated blood gases were PO2=29.6kPa, PCO2=4.5kPa. Case continued without incident, exited CPB without problem.

Details of injury (to patient/healthcare professional):

None.


Action taken (includes any action by patient, carer or healthcare professional or by the manufacturer or supplier):

Replaced defective tubing, must replace/check tubing more frequently.

Details of incident/Nature of defect:

Gas blender flow control knob jammed no flow of gas possible. Perfusionist dis-assembled the knob and could manipulate the flow with a tubing clamp. Having been serviced this subsequently occurred a second time on the device which was only 6 months old. Device returned to manufacturer for a full overhaul and is back in service.


Details of injury (to patient/healthcare professional):

None.


Action taken (includes any action by patient, carer or healthcare professional or by the manufacturer or supplier):

Device returned to manufacturer for overhaul. All perfusionists alerted how to avoid and dis-assemble this device if event re-occurs. Other methods of gas supply are currently available in operating room.

Details of incident/Nature of defect:

An Isoflurane vaporiser was full and on a hanger suspended from a vertical pole. The vaporiser was allowed to rest on the base of the pump console. This caused the vaporiser seals to leak causing a reduction of flow. An unusually high flow rate was necessary to maintain normal blood gases to the patient. Once the vaporiser was repositioned without touching the base no further problems were encountered.


Details of injury (to patient/healthcare professional):

None.


Action taken (includes any action by patient, carer or healthcare professional or by the manufacturer or supplier):

Vaporisers should not be allowed to touch the base of the console if suspended from a hanger pole due to the possibilities of gas leakages from the seals. This has been reported in the anaesthetic journals.