The Pediatric Cardiac Surgery Inquest Report



The operation - December 20

Starting at 1240 hours on Tuesday, December 20, Erin underwent what was expected to be the insertion of a Blalock-Taussig shunt. It would become necessary during the procedure for Odim to ligate the ductus arteriosus and to insert a central shunt. As a result, the operation became very lengthy. Erin died on the morning after surgery in the NICU.

Erin Petkau- pre-operative heart
Erin Petkau- pre-operative heart

Diagram 8.10 Erin Petkau - post-operative heart
Diagram 8.10 Erin Petkau - post-operative heart
1 - Modified Blalock-Taussig Shunt, clipped
2 - Ligation and division of ductus arteriosus
3 - Central shunt

Compare the pre- and post-operative diagrams side by side


To conform with the written instruction from Blanchard as set out in the memorandum of December 6, 1994, Odim was required to call Hamilton to have him assist with this procedure. Odim testified that he did not because "that wasn't my understanding of my conversation with Dr. Blanchard." (Evidence, page 26,256)

Odim stated that he called Hancock because she had been assisting him throughout the year. "I could have called anyone I wanted who might be available to assist me." (Evidence, page 26,257) He further testified that he understood that he was to have Hamilton's assistance at only open-heart neonatal procedures, not all neonatal procedures. The evidence suggests that on that point Odim was not correct. In failing to have Hamilton present for this procedure, Odim was acting outside the written directive that he had received.

The operating team is set out in the accompanying chart.

TABLE 8.9: Persons involved in the operation on Erin Petkau, December 20, 1994
OR team member   Persons involved
Surgeon J. Odim
Surgical assistant B.J. Hancock
Anaesthetists A. McNeill, N. Svorkdal (resident)
Scrub nurses H. Skomorowski, C. McGilton, A. Glenday
Circulating nurses C. Youngson, K. Rodgers, K. Cox, and H. Skomorowski
Perfusionists T. Koga, C. McCudden

TABLE: 8.10 Length of phases of the operation on Erin Petkau, December 20, 1994
Phase of the operation   Time taken
Induction 1 hour 15 minutes
Bypass 1 hour 39 minutes
Total surgical time 10 hours
Total operating-room time 11 hours 50 minutes

During induction, Odim performed a cut-down for the insertion of an arterial line. This procedure involved surgically exposing an artery in the wrist and then inserting a monitoring line.

According to Cornel and Hudson, Erin had a metabolic acidosis at the start of her operation that continued throughout surgery.

The initial chest incision was made at 1355 hours. In the operation, Odim used a 3.5-millimetre Gortex tube graft to connect the subclavian artery to the right pulmonary artery. When the shunt was opened at approximately 1510 hours, the systemic blood pressure fell. Odim testified that he became concerned that this was a sign that too much blood was being shunted to the lungs, as opposed to the body. (The systemic pressure falls when there is not sufficient blood flow to the body.) Erin became increasingly acidotic as a result. Odim believed that his only options were to make the shunt smaller or close the ductus arteriosus, thereby reducing blood flow to the lungs. As a first step to doing this, he stopped the flow of prostaglandin, in hopes that the ductus arteriosus would close on its own. He then manually pinched the ductus arteriosus and the pressures went up. After speaking with Giddins by phone, Odim decided that he would ligate the ductus arteriosus. He hoped that this would reduce the blood flow to the lungs and improve systemic blood pressure. At the same time, McNeill was transfusing additional fluids and treating Erin with dopamine in an effort to increase her blood pressure.

Throughout this portion of the operation, McNeill ventilated Erin with 100 per cent oxygen. In his testimony, Odim said that it might have been appropriate to reduce the amount of oxygen with which Erin was being ventilated after the shunt was inserted. He testified that oxygen would have dilated the vessels leading to the lungs, thus increasing pulmonary blood flow at a time when he wanted it decreased. He testified that he was not aware that Erin was being given 100 per cent oxygen at that time and he did not inquire about a change in oxygenation.

The PDA was ligated, with some improvement. However, each time Odim attempted to close Erin's chest, her blood pressure dropped. Odim testified that at one point the pressure was so low that there was no pulmonary blood flow through the shunt. The decline in blood pressure and blood flow created the possibility of clotting in the shunt. There was also less blood going to the coronary arteries, which perfuse the heart muscle itself, threatening to weaken the heart. McNeill began treating Erin with epinephrine, a stronger inotrope.

Odim reopened Erin's chest at 1610 hours and examined the Blalock-Taussig shunt. He found the shunt blocked with clotted blood. He was able to momentarily open the shunt by passing a catheter though it, but the shunt soon closed again. This took place at approximately 1700 hours.

Blockage of the shunt created a new problem, since at this point, with the ductus arteriosus closed, the shunt was the only source of blood flow into Erin's lungs. To reverse this, Odim released the ligature on the ductus arteriosus. An infusion of prostaglandin was restarted. However Erin's pressures remained low and the ductus arteriosus failed to reopen.

Odim testified that, at that point, he was not able to determine the cause of the problem.

I was wondering whether the pulmonary arteries were too small in this child, and we were having problems with the outflow bed. Because I was perplexed, I had never been in a situation where I was just so over shunted with a 3.5 millimetre shunt.

Q: And what happened then?

A: When the shunt occluded, we were still faced with a tremendous acidemia and requirements for inotropes. And Erin needed, you know, another shunt, needed pulmonary blood flow, because now we had no pulmonary blood flow. (Evidence, page 26,292)

From 1700 hours onwards, there were periods when Erin's blood pressure and oxygen saturation were acceptable, but these periods did not last for more than fifteen minutes. By 1840 hours, her blood pressure and oxygen saturation had fallen again and were not responding to treatment. During this period, she suffered a reduction in oxygen supply to her tissues (hypoxemia) that was, at times, life-threatening.

Odim concluded that creation of a second shunt was necessary. To do this, he would have to use bypass, so that he could gain access to the small blood vessels that were to be connected by a central shunt.

Barbara Petkau testified that, at approximately 1830 hours, Savani informed the family that the initial shunt procedure had not gone well. As a result, Erin would be put on bypass and a second shunt inserted.

The anaesthetists packed Erin's head in ice at approximately 1900 hours, when the perfusionists were called. The team went on bypass at 1927 hours. The move to go on bypass was delayed when it was discovered that the single venous cannula that was being used was too small to supply an appropriate level of blood flow from Erin to the bypass machine. Koga had recommended a smaller-size cannula, since he was under the impression that Odim would be using two cannulas for the venous return, whereas Odim intended to use only one cannula. Although the problem lasted only a few minutes, it was further evidence of ongoing communication problems between team members.

Following the operation, team members also had differing accounts of how the original cannulation problem arose. McNeill felt that Odim should have recognized that the cannula was too small before he inserted it. Odim said that Koga had simply failed to provide him with the appropriate cannula. However it is obvious that Odim failed to notice that he had been given the wrong-size cannula for what he intended.

By the time bypass was initiated, two and a half hours had passed from the time that the shunt had clotted. Once bypass was initiated, Odim constructed a four millimetre central shunt. He also clipped the original B-T shunt to ensure that it did not reopen and create excessive pulmonary blood flow once again. The team then attempted to wean Erica from bypass, which had lasted for 109 minutes. High doses of inotropes were required to allow separation from bypass, which took place at 2106 hours.

When Erin first came off bypass, her blood pressure was low and was treated with dopamine and epinephrine. She required this treatment to maintain her blood pressure at acceptable levels while in the OR.

Erin's condition was so serious while she was coming off pump that the OR team called Giddins and asked him to come down to the HSC. The team also requested that a chaplain be called to the hospital. NICU nurse Armitage called the chaplain and took him to see the Petkaus. The parents were told that Erin's situation looked serious. Shortly afterwards, Giddins arrived at the HSC. By that point Erin had been successfully weaned from bypass. Giddins communicated this news to Erin's parents and told them that things were looking well for Erin. Unfortunately, this was very shortly after Armitage and the chaplain had left the Petkaus with the impression that Erin's condition was bleak. The unfortunate timing of these two events was distressing for Erin's parents. This was particularly so since Giddins, quite unintentionally, had left them with what proved to be an overly optimistic picture of Erin's condition.

After Erin came off bypass, she suffered extensive bleeding. This was due to a number of causes, including the fact that she had been treated with heparin to prevent clotting. She also had some surgical bleeding from her chest wall at the site of the insertion of the right atrial catheter (a small pressure monitoring line) and at the site of the Blalock-Taussig shunt. In her testimony, McGilton said that it took approximately three hours to control the post-bypass bleeding. She said that Odim made several attempts to close Erin's chest. However, he had to abandon each one because the closure caused the blood pressure and oxygen saturation to fall and was accompanied by bleeding from the chest.

Giddins commented that the bleeding did not surprise him:

The fact that it hadn't been bleeding during the case and the fact that it perhaps starts to leak a little bit, if there is a coagulopathy or tendency to bleed, doesn't particularly surprise me. You are not making it water proof. You are making a near water proof seal that depends on blood clotting to completely finish the job, but a little area that needed one extra suture doesn't strike me as being unusual. (Evidence, pages 4,547-4,548)

McNeill gave this description of the cause of the coagulopathy:

I would attribute it to the effect of bypass on her coagulation status; and because of her size, she would have had a significant dilution of clotting factors and platelets because of the bypass situation, and whatever contribution there would be during bypass to platelet function deterioration, and then the massive transfusion post-operatively, which would further exacerbate the problem. (Evidence, page 13,614)

When the drapes were removed before Erin was taken to the NICU, it was discovered that she had been bleeding from the cut-down site in her wrist. In their evidence, McNeill and Odim differed in their accounts of the amount of bleeding that took place. McNeill viewed it as an extensive hemorrhage, while Odim said that it was insignificant. McNeill was asked about the significance of this bleeding.

Well, the major significance of it was that this was of a covert bleeding site that we wouldn't have actually seen during the period of time when we were trying to volume resuscitate her. And whenever somebody is being transfused, you use their hemodynamic profile to determine whether you are adequately transfusing them. So, you know, it wasn't, we weren't actually aware that she was bleeding from her wrist, but we would have, we don't only use the bleeding that we see to replace blood loss. (Evidence, page 13,629)

While this blood loss would have been replaced by transfusion, it increased the volume of blood that needed to be transfused, thus raising the likelihood that Erin would develop a coagulopathy. There was no apparent explanation for the bleeding at the cut-down site. There was no evidence of bleeding when Erin was draped; nor does it appear that the line became dislodged during the operation.

After the bleeding sites had been identified and sutured, and resuscitative drugs and large amounts of blood products had been given, Erin was taken to the NICU with her chest incision covered with a silastic membrane.

In his report for this Inquest, Hudson wrote:

There was massive intraoperative hemorrhage. The surgeon's and anesthetist's comments indicate that inadequate surgical hemostasis [this refers to the surgeon's control of bleeding] was an important cause of the bleeding. The patient received a total 800-900 ml of blood products during surgery, including albumin, PRBC [packed red blood cells], platelets, FFP [fresh frozen plasma] and cryoprecipate. This is over 3 times the blood volume of this patient (~90-100 ml/kg). (Exhibit 307, page 12.14)

In his testimony, Odim said he judged that the need to keep the shunt open was of greater significance than the need to bring the coagulopathy under full control.

By the time Erin left the OR, her heart had been subjected to considerable extremes in blood pressure and oxygenation. She had undergone an extremely lengthy operative procedure and suffered extensive bleeding. The evidence suggests that her system would have been very weakened as a result of this operation.

Erin arrived in the NICU at 0028 hours on December 21. Initially Odim considered her vital signs satisfactory. However, by 0200 hours, Erin had suffered a series of episodes in which her oxygen saturation fell significantly, and she was treated with massive doses of inotropes. She continued to have the same problems with bleeding, low oxygen (hypoxemia), metabolic acidosis and low calcium (hypocalcemia) that she had in the OR.

In response to the fall in her oxygen saturation, Erin's chest was reopened at 0225 hours. This was done in the NICU with the assistance of a nurse who was not scrubbed in and without the assistance of an anaesthetist. After discovering a clot blocking the central shunt, Odim flushed the shunt and passed a catheter through it.

Hudson questioned the propriety of opening the chest in the NICU without the presence of an operating-room nurse and an anaesthetist. He said that optimal management of the case would be "difficult or impossible" in their absence. (Exhibit 307, page 12.15)

The reopening of Erin's chest in the NICU was followed by a period of significant bleeding. In his testimony, Odim said that the underlying post-operative problem was Erin's declining cardiac output, caused by her weakened heart. This decline led to lower blood pressures and the blocking of the shunt. The problems were complicated by her small pulmonary arteries. The weakness of her heart muscle arose from the fact that, for the previous 18 to 20 hours, her heart been subjected to variations in blood pressure and oxygen saturation.

At 0500 hours, Erin's heart rate and blood pressure fell. These were signs of primary cardiac failure. Decreases in blood pressure led to a slowing of blood flow through the shunt and increased the chances of clotting. Giddins testified that as he understood it, Erin's problems were primarily with her heart rate and blood pressure and not with the shunt itself.

From 0500 hours onwards, Erin deteriorated rapidly, suffering a cardiac arrest on several occasions. For two hours, continuous efforts (including cardiac massage and flushing of the shunt) were made to resuscitate her. According to Hudson, Erin once again was given transfusions, which amounted to three times her blood volume, in response to what he termed a massive hemorrhage. (In his report, Hudson stated that Erin did not receive clotting factors post-operatively, but the post-operative record contradicted this statement.)

Erin's parents were summoned to her bedside. Barbara Petkau testified that she felt that it was only at this point that she was made aware that her daughter was near death. Erin died in her mother's arms at 0750 hours.

Armitage testified that she believed that some of Odim's comments about Erin's death were too technical and possibly insensitive to the parents. However, when giving her testimony, Barbara Petkau could not recall the comments.



Current Home - Table of Contents - Chapter 8 - The operation - December 20
Next Autopsy findings
Previous Pre-operative status
Section 1 Chapter 1 - Introduction to the Issues
  Chapter 2 - Pediatric Cardiac Issues
  Chapter 3 - The Diagnosis of Pediatric Heart Defects and their Surgical Treatment
  Chapter 4 - The Health Sciences Centre
Section 2 Chapter 5 - Pediatric Cardiac Surgery in Winnipeg 1950-1993
  Chapter 6 - The Restart of Pediatric Cardiac Surgery in 1994
January 1, 1994 to May 17, 1994
  Chapter 7 - The Slowdown; May 17 to September 1994
  Chapter 8 - Events Leading to the Suspension of the Program
September 7, 1994 to December 23, 1994
  Chapter 9 - 1995 - The Aftermath of the Shutdown
January to March, 1995
Section 3 Chapter 10 - Findings and Recommendations
Appendix 1 - Glossary of terms used in this report
Appendix 2 - Parties to the Proceedings and counsel
Appendix 3 - List of witnesses and dates of testimony
Search the Report
Table of Contents