As noted at the outset, this case gave rise to the following questions:

• Were Erin's parents provided with sufficient information to allow them to give informed consent to the procedure?
• Should the ventilation intra-operatively have been different?
• Was the Blalock-Taussig shunt too small?
• What led to the shunt failures?
• Should Dr. Andrew Hamilton have assisted in this operation?
• What was the cause of death and was it preventable?

Were Erin's parents provided with sufficient information
to allow them to give informed consent to the procedure?

Finding

The family seems to have been fully informed as to the risks associated with the particular procedure that Erin was to undergo. However, they were not informed of the program's recent history, including the summer shutdown and the intention to have the program reviewed by an external examiner. They were entitled to know this information, before giving their consent. This evidence tends to suggest that Erin's parents were not provided with sufficient information to allow them to give informed consent to the procedure.

Should the ventilation intra-operatively have been different?

Odim testified that he thought that, in providing 100 per cent oxygen after the B-T shunt was opened, McNeill had violated a basic tenet of the ventilation of neonates. He testified that, if the level of oxygen given to Erin had been reduced, her oxygen saturation would have decreased, with a resultant lessening of the excessive blood flow to her lungs. As a further result, Erin's systemic blood pressure would have been increased when the shunt was opened and the problems with low blood pressure and acidosis avoided.

Odim testified that he relied on the anaesthetists for management of oxygenation and ventilation. He said that this had not been a problem in previous cases. He said he told McNeill that there was too much pulmonary blood flow and asked for her help, but he did not inquire about the amount of oxygen being delivered.

This is a serious question. Hudson and Soder, both of whom are anaesthetists, were questioned on this point. Their testimony indicated that there was nothing inappropriate about the amount of oxygen delivered by McNeill during the stage of the operation when the pulmonary blood flow was excessive. It was suggested that lowering the oxygen might have had a positive effect in terms of blood pressure, but this action carried with it other risks.

Finding

It is not possible to state on the basis of the evidence what might have happened. However, it can be stated that McNeill's actions were considered reasonable practice. The underlying issue here, as it is in many of the cases under review, is of the extremely low level of communication and heavy reliance by Odim on what he believed to be 'standard practice' in any given situation.

Was the Blalock-Taussig shunt too small?

Cornel questioned Odim's use of a 3.5-millimetre graft, which he said was difficult to work with and could easily occlude. Cornel said that he preferred to use five-millimetre grafts. He said that the artery providing the blood, not the size of the shunt, governed the amount of blood that was shunted. In his report he wrote:

Stark and DeLaval recommend use of larger size grafts and I have found that a 5mm graft can be used in virtually every patient. I have used a 4mm graft on only 2 or 3 occasions. Flow is controlled by the subclavian artery rather than by the graft itself. There is a much higher failure rate in small grafts. (Exhibit 353, page 71)

In his testimony, Duncan said that when he wrote his report, he had not heard of anyone using a 3.5-millimetre shunt. He said that in Vancouver, however, where he had subsequently taken up a position at the British Columbia Children's Hospital, 3.5-millimetre shunts were used in that program, with good results. He testified that the issue was the length, as well as the diameter, of the shunt. A long, thin shunt, he said, was more likely to clog. He said that one reason why there was a good chance of the B-T shunt clogging was the fact that the pulmonary vessels were so small. Soder testified that it was common to use 3.5-millimetre shunts at the Izaak Walton Killam Hospital for Children in Halifax, where he worked.

Giddins was asked if he thought that the shunt was too small.

No, I don't think so. This was a relatively small child, weight of 2.6 kilos. Normally, 3.5 millimetre shunts, in my experience, are standard for normal newborn weights, which is three, 3.5, even 4 kilos. Also she had small lung blood vessels, and the standard size normal newborn size shunt would, I would think, be perfectly adequate for her situation of size and lung blood vessel size. (Evidence, page 4,534)

Giddins was also asked why the central shunt was larger:

Also, going up a size seemed to me, although I don't remember being part of that decision, it seemed to me a very reasonable decision, in view of the fact they had been having difficulties up until that point in the-not at the very beginning, but up until that point with low oxygen saturation levels. In other words, this shunt had potentially more to do than the first one. This shunt would provide more flow to both sides than potentially just one side. (Evidence, page 4,537)

Finding

From the testimony that was presented it would appear that the use of a 3.5-millimetre shunt was a reasonable choice, although it led to a higher degree of difficulty for the surgeon.

What led to the shunt failures?

Finding

There is no conclusive evidence as to why the B-T shunt and the central shunt failed. The evidence indicates that a B-T shunt was an appropriate choice for this child and that switching to a central shunt was also a reasonable decision. The evidence also indicates that at the outset, both shunts were patent-indeed, following the opening of the B-T shunt, Erin had difficulties with too much blood flow.

Witnesses to this Inquest suggested, without reaching firm conclusions, that Erin's problems were related to underdeveloped (hypoplastic) pulmonary blood vessels, low cardiac output, hypoxia and acidosis. These problems were all inter-related and fed upon one another.

As described above, problems with inadequate blood pressure and cardiac output led to both shunts eventually clotting. It should be noted that Odim experienced difficulties with the central shunt when he tried to close Erin's chest. This speaks to the technical difficulties of this procedure, particularly in ensuring that the shunt does not bend or buckle when the chest is closed.

Questions were raised about technical issues. In their joint report, Cornel and Duncan wrote:

Despite the fact they are not regarded by some as difficult, an aorto-pulmonary shunt is one of the most technically demanding surgeries. The second shunt, despite having been done on bypass, was inadequate at the aortic orifice. (Exhibit 354, page 15)

In his report, Cornel wrote, "The cause of failure of the central shunt is not clear but a technical problem is the most likely." (Exhibit 353, page 72)

This operation was extremely long and accompanied by extensive bleeding. Erin's death was one of these that led Soder to conclude in his report that the:

skill and dexterity of the surgeon performing these operations were insufficient for the challenges of successfully repairing infant hearts with complex malformation. [boldface in the original] (Exhibit 345, page 8)

Should Dr. Andrew Hamilton have assisted in this operation?

Finding

It was in response to concerns about Odim's skills that Blanchard informed Postl (and, through Postl, other persons involved in the program) that Hamilton would assist Odim in all neonatal surgery. Hamilton should have been involved in this operation.

What was the cause of death and was it preventable?

Erin had been in the operating room for 12 hours because of all the problems and complications associated with the surgical procedure. This would have left her in a weakened condition. In addition, she experienced severe problems with both clotting and bleeding. The clotting of the central shunt, which cannot be definitively explained, led to a loss of blood flow to her lungs and her eventual heart failure.

Finding

Cornel's conclusion that the failure of the central shunt was due to technical problems, combined with Soder's conclusion that the skill and dexterity of the surgeon were not adequate to the challenge of this procedure, clearly suggests that the death was possibly preventable.