Fisher v. Victoria Hospital, 2008 ONCA 759

CITATION: Fisher v. Victoria Hospital, 2008 ONCA 759

DATE: 20081112

DOCKET: C46830

COURT OF APPEAL FOR ONTARIO

Rosenberg, Feldman and Simmons JJ.A.

BETWEEN:

Wendy Fisher, Alecia Fisher by her Litigation Guardian, Wendy Fisher and Darrell Fisher

Respondents

and

Dr. D. Barry Atack, Dr. John M. Muir, Dr. J. MacMillan, Dr. Frederick Sabga, J. Hutchison and Nancy Stephens, T. Grimminck, P. Rawlins and W. Wanless, Deborah Craven, Jane Doe and Victoria Hospital

Appellant

D. H. Rogers, Q.C. and Anita M. Varjacic for the appellant

Paul J. Pape for the respondents

Heard: April 3, 2008

On appeal from the judgment of Justice Helen Rady of the Superior Court of Justice dated February 12, 2007.

Simmons J.A.:

I. Overview

[1] Alecia Fisher was born at Victoria Hospital in London at 10 p.m. on February 4, 1991. At around 9 p.m., while Mrs. Fisher was in active labour, nurses were unable to reliably detect a fetal heart rate for a period of between twelve and seventeen minutes. Despite this problem, doctors did not intervene, and Mrs. Fisher proceeded to a vaginal delivery. Alecia was not breathing spontaneously at birth and had to be resuscitated. Doctors eventually determined that she suffered asphyxia during labour, which led to cerebral palsy with all its tragic and devastating effects.

[2] Alecia’s parents sued the obstetrician who attended Alecia’s birth, the anaesthetist, a paediatrician who assisted in the resuscitation efforts, an obstetrical resident, an intern, most of the nurses who were present during labour and delivery, and the hospital. In addition to saying that the doctors were negligent in failing to expedite Alecia’s birth, Alecia’s parents claimed that the hospital’s fetal heart monitoring protocol was below standard, that the nurses were negligent in carrying out the monitoring, and that proper monitoring every 15 minutes would have detected the asphyxia before 9 p.m., leading to intervention.

[3] The parties agreed on damages prior to trial and, by the end of the 61-day trial, only the obstetrician and the hospital remained as defendants. Rady J. dismissed the negligence claim against the obstetrician, but held the hospital vicariously liable for the negligence of its nurses. Although she determined that the hospital’s protocol requiring intermittent monitoring of the fetal heart every thirty minutes during the active phase of the first stage of labour met an acceptable standard of practice, the trial judge found that the nurses breached the standard of care by failing to auscultate[^1] the fetal heart as required at 6:30 and 7:00 p.m.

[4] Concerning causation, the trial judge found that Alecia’s brain injury was the result of the cumulative effect of a period of acute near total asphyxia that occurred at around 9 p.m., and an earlier period of prolonged partial asphyxia that occurred in the hours before birth. As no irregularities in the fetal heart rate were detected at other times when the fetal heart was being monitored properly, the trial judge concluded that the prolonged partial asphyxia likely occurred between 6:00 p.m. and 7:30 p.m. Had the nurses auscultated the fetal heart as required at 6:30 and 7:00 p.m., electronic fetal heart monitoring would likely have commenced, and that, in turn, would inevitably have led to an expedited delivery, “thereby sparing Alecia the magnitude of her injuries.”

[5] On appeal, the hospital does not contest the trial judge’s findings that the nurses breached the standard of care by failing to auscultate the fetal heart at 6:30 and 7:00 p.m.; that the factual cause of Alexia’s brain injuries was the cumulative effect of prolonged partial asphyxia and acute near total asphyxia; and that the acute near total asphyxia occurred at around 9 p.m. Rather, the hospital claims that the trial judge misapprehended important evidence about when the prolonged partial asphyxia occurred and reached her conclusions on that issue and on the liability of the hospital based on a theory that was not pleaded, and based on a series of findings and assumptions that are not supported by the evidence. The hospital raises the following issues on appeal:

i) Did the trial judge err in law and make a palpable and overriding error by making factual findings and inferences in support of a judicial theory of negligence that are unsupported by the evidence and by inadvertently applying the benefit of hindsight?

ii) Did the trial judge err in law by finding the hospital liable based on a theory of negligence that was not raised by the respondents at trial?

iii) Did the trial judge err in law by misapplying the legal test of causation?

[6] For the reasons that follow, I would allow the appeal and order a new trial.

II. Background

i) The events surrounding Alecia’s birth

[7] Mrs. Fisher was admitted to hospital at 4:37 p.m. on February 4, 1991. Alecia was Mrs. Fisher’s second child, and Mrs. Fisher was considered a low risk patient.[^2] The hospital protocol required that intermittent fetal heart monitoring increase from every hour to every 30 minutes once a low risk mother reached the active phase of the first stage of labour. Although the hospital led evidence that Mrs. Fisher did not reach that phase until 7:30 or 8:00 p.m., the trial judge found that Mrs. Fisher commenced active labour at 6 p.m.

[8] There is no dispute that fetal heart monitoring occurred by auscultation at 4:45, 4:55, 6:00, 7:30, 7:55; 8:00, and 8:35 p.m., and that on all these occasions, the fetal heart rate was in the normal range of 120 to 160 b.p.m.[^3] It is also clear that auscultation did not occur as required at either 6:30 or 7:00 p.m.

[9] The respondents took no issue at trial with the nurses’ actions at 9 p.m., when they were unable to detect a fetal heart rate. Indeed, the trial judge found that the nurses took “all appropriate steps, including the application of oxygen, turning Mrs. Fisher from side to side and increasing the intravenous level.”

[10] In the minutes after 9 p.m., the nurses made multiple attempts to locate a fetal heart rate. As many as four different nurses tried to locate the fetal heart rate through auscultation. In addition, they tried using a transducer and a fetascan.[^4] The transducer briefly displayed a fetal heart rate of 96-113 b.p.m. However, the nurses could not hear an audible fetal heart beat and were unable to maintain the fetal heart rate on screen. Although they thought they picked up the fetal heart on the fetascan, one of them described the fetal heart as “barely fluttering”.

[11] When the nurses were unable to locate a fetal heart rate, they paged the obstetrical resident, Dr. MacMillan. At around 9:05 p.m., they re-paged Dr. MacMillan and also paged an obstetrician present in the hospital, Dr. Atack. Mrs. Fisher’s obstetrician, Dr. Maynard, was also paged but was unable to attend as he was at another London hospital, attending another birth.

[12] At the time that he was paged, Dr. MacMillan was in the operating room assisting Dr. Atack with surgery on a patient with an ectopic pregnancy. Dr. MacMillan left the operating room and went to Mrs. Fisher’s labour room within minutes, while Dr. Atack remained in the operating room to complete the surgery.

[13] At 9:10 p.m., the nurses charted the fetal heart rate at 60-90 b.p.m. and, at about 9:12 p.m., the fetal heart rate was recorded by auscultation at 130 b.p.m. Dr. MacMillan arrived, ruptured the amniotic membranes, and placed a fetal heart rate monitor on Alecia’s scalp (a scalp clip). At 9:17 p.m.[^5], the fetal heart rate was detected by the scalp clip and was recorded as improved at 145-150 b.p.m.

[14] The trial judge found that Dr. Atack arrived in the labour room at about 9:25 p.m. The fetal heart rate was around 150 b.p.m. at that point. Although there had been two variable decelerations, down to between 90 and 100 b.p.m., since electronic monitoring began, they were brief in duration. Dr. Atack monitored Mrs. Fisher over about five minutes during two contractions and observed no decelerations in the fetal heart rate. As labour appeared to be progressing fairly rapidly, he approved administration of an epidural.

[15] Dr. Atack observed three more decelerations in the fetal heart rate before Mrs. Fisher was transferred to the delivery room, one just before 9:38, one at 9:39, and one at 9:41 p.m. He felt that the first two were not concerning and that the third was associated with a vaginal examination in which he confirmed that Mrs. Fisher was fully dilated.

[16] After Mrs. Fisher began pushing, Dr. Atack observed further decelerations in the fetal heart rate, but always with recovery. During the final two minutes before birth the fetal heart rate dropped to 50-60 b.p.m. Alecia was born at 10:00 p.m. by spontaneous vaginal delivery. She was limp and not breathing at birth. Her umbilical cord was wrapped loosely around her body and her head.

[17] Alecia was ultimately diagnosed as suffering from a dyskinetic form of cerebral palsy, marked by dystonic and athetoid movement in all four of her limbs.[^6] She has increased spasticity in her legs, and is quadriplegic. She is also slightly microcephalic, meaning she has a smaller than average head. Alecia exhibits some cognitive deficits and learning problems.

[18] The electronic fetal heart rate tracings for the period between 9:17 p.m. and 10:00 p.m. were filed as an exhibit at trial. There was no issue at trial that Alecia’s brain injury was caused by asphyxia during labour. However, the precise timing and mechanism of the injury was vigorously disputed. Much of the evidence at trial related to these issues and to the question of whether Dr. Atack fell below the standard of care by failing to expedite Alecia’s delivery based on what he was told by the nurses about what happened at around 9 p.m., and based on the fetal heart rate tracings, both before and after he arrived.

III. The Positions of the Parties at Trial

[19] The thrust of the respondents’ position at trial was that Alecia suffered brain damage either in the period of five to ten minutes before birth or in the period of about two minutes before and two minutes after birth, and that Dr. Atack caused or materially contributed to her injuries, primarily by failing to recognize the need to take steps to expedite her delivery.

[20] The respondents claimed that, at some point during labour, an unknown process that led to asphyxia commenced. Although they could not pinpoint the commencement of the asphyxial insult with certainty,[^7] the respondents’ causation experts (primarily Drs. Perlman, MacGregor, and Armstrong but also Drs. Hodd and Schifrin to some extentopined that the process probably began somewhere between 40 to 50 minutes and up to one to two hours before birth, and culminated either in the five to ten minute period or in the two minute period prior to delivery. The process resulted in either a prolonged partial asphyxia that became increasingly severe; an acute near total asphyxia; or more likely a combination of both forms of asphyxia. Whatever the form of the asphyxia, the respondents claimed that it did not cause injury until the final minutes before Alecia’s birth and that her injuries could have been avoided had Dr. Atack expedited her delivery.

[21] In addition to their allegations against Dr. Atack, the respondents claimed that the hospital’s fetal heart rate monitoring protocol for low risk patients requiring intermittent monitoring of the fetal heart every thirty minutes during the active phase of the first stage of labour was below standard and that proper monitoring every fifteen minutes would have detected the asphyxia before 9 p.m., and would have led inevitably to intervention. The respondents also claimed that the method of auscultation used by at least one of the nurses was below standard. Finally, the respondents claimed that the nurses breached the standard of care by failing to monitor the fetal heart rate in accordance with the hospital’s protocol.

[22] The doctors’ position at trial was that all doctors met the standard of care, that the timing of Alecia’s brain injury could not be established with precision, and that, in any event, her brain injury did not occur within the final minutes before birth.

[23] The thrust of the hospital’s position at trial was that its fetal heart rate monitoring protocol met an acceptable standard of practice and that the nursing care provided to Mrs. Fisher equalled or exceeded the standard of care.

IV. The Trial Judge’s Reasons

[24] The trial judge began the analysis portion of her reasons by considering the negligence claim against Dr. Atack. She concluded that Dr. Atack met the standard of care in deciding to monitor the fetal heart rate and in permitting Mrs. Fisher to continue to vaginal delivery.

[25] The trial judge then turned to the question of whether any breach of the standard of care that Dr. Atack may have committed caused or materially contributed to Alecia’s brain injury. She began this portion of her reasons by stating, “[c]ritical to the analysis of causation is the timing of Alecia’s brain injury and the precision with which it can be measured.”

[26] In a section entitled Conclusions: Causation, the trial judge noted that there were three scenarios advanced to explain Alecia’s brain injury: i) acute near total asphyxia; ii) prolonged partial asphyxia; and iii) a mixed form of asphyxial exposure. She concluded that Alecia suffered both a prolonged partial asphyxia and an acute near total asphyxia in the hours before birth, and that the events around 9 p.m. represented the end of an acute near total asphyxia. As Dr. Atack did not assume responsibility for Alecia’s care until after these events, the trial judge found no causal link between any possible breach by him of the standard of care and Alecia's injury.

[27] The trial judge summarized her reasons for these findings as follows:

Having considered the evidence, I am not satisfied that it is possible to pinpoint the timing of Alecia’s injury with minute accuracy and certainly not to within minutes of birth, particularly in the absence of a sentinel event. Moreover, there is no objective evidence of an acute near total event at this time. There are only two minutes of bradycardia which by itself is insufficient to cause the extent of Alecia’s injury. I am not satisfied that there is evidence pointing to a sentinel event during this interval.

I am satisfied that the following findings of fact are supported by the evidence on a balance of probabilities. Alecia most probably sustained intrapartum asphyxia. There was, more likely than not, a period of prolonged partial asphyxia and a period of more acute asphyxia, the cumulative effect of which accounts for the pattern of brain injury that she sustained and the type of cerebral palsy that she exhibits. While it is not possible to pinpoint the timing of the asphyxial insult(s) with complete precision, I accept Dr. Watts' evidence that there was likely both partial prolonged asphyxia and an acute near total event that occurred in the hours before birth. I accept Dr. Watts’ opinion that the partial prolonged asphyxia occurred in the hours before birth and that the events at [9:00 p.m.] represented the tail end of a bradycardia representing a near total acute insult. In this regard, I accept Ms. Rawlins’ testimony about her observation that the fetal heart was barely fluttering. There is no doubt in my mind that Ms. Rawlins, an experienced labour and delivery nurse, knew what she was seeing-a heart that was beating very slowly.

The foregoing events all occurred before Dr. Atack assumed responsibility for Mrs. Fisher’s care. As a result, if I had found Dr. Atack breached the standard of care, I can find no causal connection between the breach and the damages sustained.

[28] The trial judge then turned to the respondents’ claim against the hospital. She rejected the respondents’ submissions concerning the protocol for auscultation and the method of auscultation used by at least one of the nurses. She found that the hospital’s protocol for low risk mothers requiring intermittent monitoring of the fetal heart every thirty minutes during the active phase of the first stage of labour and the methods of auscultation used by the nurses met an acceptable standard of practice. However, based on a finding that Mrs. Fisher was in the active phase of the first stage of labour at 6 p.m., the trial judge concluded that the nurses breached the standard of care by failing to auscultate the fetal heart at 6:30 and 7 p.m.

[29] With respect to causation, the trial judge concluded that the respondents had established that Alecia’s injuries were caused by the nurses’ negligence for which the hospital is vicariously responsible. She found that there was either a bradycardia[^8] or decelerations of the fetal heart during the period of up to 17 minutes after 9 p.m. during which the nurses were unable to auscultate the fetal heart, which was consistent with what Dr. Watts thought was an acute near total event. As she had already concluded that there was an earlier period of prolonged partial asphyxia, the trial judge found that it likely occurred between 6 p.m. and 7:30 p.m. when the fetal heart was not being properly monitored. The fact that no irregularities in the fetal heart rate were detected at other times when proper monitoring occurred supported this finding.

[30] The trial judge then concluded:

[I]t seems reasonable to infer that if the nurses auscultated the fetal heart between [6:00 and 7:30 p.m.], they would have detected decelerations or perhaps a bradycardia of the fetal heart, raising concern that some form of asphyxial event was occurring. It further seems reasonable to conclude that they would have conducted the appropriate nursing interventions and they would have likely summoned assistance. It is likely that Dr. MacMillan would have responded in the same fashion as he did at [9:15 p.m.], namely by rupturing the membranes and attaching a fetal scalp clip.

In my view, the electronic tracing would have detected the magnitude of the bradycardia (or decelerations) occurring at [9:00 p.m.] and would have shown that an acute event was occurring, consistent with Dr. Watt’s view. There can be no doubt that in such circumstances, steps would have been taken to expedite delivery, thereby sparing Alecia the magnitude of her injuries.

V. Analysis

[31] The hospital’s initial submission under its first ground of appeal is that the trial judge made a palpable and overriding error in finding that Alecia suffered a prolonged partial asphyxia that preceded the events around 9 p.m. I do not accept this submission and I will address this issue first.

[32] In addition, the hospital submits that the series of findings on which the trial judge relied in order to find that the hospital caused Alecia’s injuries reflect a misapplication of the law of causation and, in any event, lack an evidentiary foundation. I will address these issues in turn. As I have concluded that these issues are dispositive of the appeal, it is unnecessary that I address the remaining issues raised by the hospital.

1. Did the trial judge made a palpable and overriding error in finding that Alecia suffered a prolonged partial asphyxia prior to 9 p.m.?

[33] The hospital contends that, in making her findings about the timing of Alecia’s brain injury, the trial judge rejected the respondents’ theory that the injury occurred in the minutes before birth and relied instead on the doctors’ theory of causation, with particular emphasis on the evidence of Dr. Watts. The hospital claims that in finding that Alecia suffered a prolonged partial asphyxia that preceded the events around 9 p.m., the trial judge misapprehended Dr. Watts’ opinion.

[34] The hospital points in particular to two portions of the trial judge’s reasons. In the section of her reasons setting out her conclusions on causation in relation to Dr. Atack, the trial judge said the following:

I also accept Dr. Watts’ evidence with respect to the type of insult Alicia likely endured and his testimony with respect to timing.

While it is not possible to pinpoint the timing of the asphyxial insult(s) with complete precision, I accept Dr. Watts' evidence that there was likely both partial prolonged asphyxia and an acute near total event that occurred in the hours before birth. I accept Dr. Watts’ opinion that the partial prolonged asphyxia occurred in the hours before birth and that the events at [9:00 p.m.] represented the tail end of a bradycardia representing a near total acute insult. [Emphasis added.]

[35] Later, in the section of her reasons setting out her conclusions on causation in relation to the hospital, the trial judge said this:

It seems clear that during this latter interval [from 9:00 p.m. for a period of up to 17 minutes], there was either a bradycardia or decelerations of the fetal heart, consistent with what Dr. Watts thought was an acute near total event. As I have already accepted on the basis of Dr. Watts’ testimony, there was an earlier episode of partial prolonged asphyxia that occurred in the hours before birth. [Emphasis added.]

The trial judge then stated that it was reasonable to infer that the prolonged partial asphyxia occurred between 6:00 and 7:30 p.m.

[36] According to the hospital, Dr. Watts did not opine that Alecia suffered a period of prolonged partial asphyxia that preceded the events around 9 p.m. Rather, his opinion was that the most significant period relating to Alecia’s injury was around 9 p.m., that Alecia likely suffered a prolonged bradycardia representing a near total acute asphyxia between about 8:45 p.m. and 9:08 p.m., that brain damage began in that time frame and likely occurred predominantly between 9:00 and 9:08 p.m., and that a period of prolonged partial asphyxia continued thereafter.[^9]

[37] By way of illustration, the hospital points to the following excerpts of Dr. Watts’ evidence, given while he was being cross-examined by the hospital’s trial counsel:

Q. Dr. Watts, I just have a few questions for you. As I understand your evidence, in coming to your opinion, you took a retrospective look at all of the evidence that you had before you and concluded that Alecia suffered both a near total acute hypoxic event and a partial prolonged hypoxic event. Is that correct?

A. That's correct.

Q. And if I understood your evidence, and once again, looking retrospectively at both, you looked at the clinical evidence together with the obstetrical evidence and you concluded that the near total acute event likely occurred at or prior to [9:00 p.m.]?

A. Yes. I think that's...

Q. Is that a fair statement?

A. ...that's, that's reasonable.

Q. And again, looking at that same evidence, being both the clinical evidence and the obstetrical evidence, you came to the conclusion that the most likely scenario was that, to some degree, there was a prolonged, there was prolonged partial hypoxia that continued after [9:00 p.m.]...

A. Yes.

Q.... for whatever the timing of the actual injury was in and around the time of [9:00 p.m.]. Is that. ..

A. That's right.

Q....a fair statement.

A. Yes.

Q. Would it be fair to say, given the 13 minutes that we have the objective evidence of either bradycardia and/or deep decelerations or a rather slow return to normalcy, that the bradycardia would not have started too long before [9:00 p.m.]?

A. Not too long before?

Q. If it started before [9:00 p.m.], it wouldn't have been 20 minutes before [9:00 p.m.]?

A. It's not likely to be that long before.

Q. And the actual timing of the injury would have been probably close to [9:00 p.m.] on either side of it?

A. The timing of the start of -yes.

Q. It, and it could have been after [9:00 p.m.]?

A. Yes, it could.

Q. And at [9:00 p.m.]?

A. Yes.

Q. All right. And as I also understand, again, using your clinical judgment and opinion based on all of the evidence you've reviewed, that you came to the opinion that in addition to the near total acute that you believe Alecia suffered as evidenced by the events surrounding [9:00 p.m.], that I think you used the term -or based on your sort of differential diagnosis -you believe the most likely scenario was that a prolonged partial continued after..

A. After...

Q. [9:00]?

A....that time, yes.

(Emphasis added by the hospital.)

[38] I do not accept the hospital’s submissions. I have read both the trial judge’s reasons and Dr. Watts’ evidence several times. Although Dr. Watts may not have expressed a medical opinion that a prolonged partial asphyxia preceded the events around 9 p.m., he was not in a position to do so because there was no clinical evidence of a prolonged partial asphyxia before 9 p.m. to support such a hypothesis. However, he clearly recognized the possibility of an earlier prolonged partial asphyxia as illustrated by the following extract from his examination-in-chief:

Q. Dr. Perlman did concede in cross-examination that Alecia Fisher could have had a brain disabling exposure earlier than the last five minutes, sorry, five to ten minutes - and I emphasize the phrase “could have” - and still exhibited these blood gases that we see in Exhibit 4. Do you agree or disagree with that proposition?

A. Yes. I, I, I do agree.

Q. And how does that proposition relate, if at all, to your opinion as to how Alecia Fisher's brain injury occurred?

A. I think it's somewhat similar to my opinion.

Q. In what way?

A. In, in, in that I think there, there is likely to have been a significant event at around nine o'clock. There may have been a contribution of the, of subsequent prolonged partial or intermittent asphyxia to the, to the base deficit or to the blood gases, but it seems unlikely to me, looking at the rest of the picture that that contribution would have changed her outcome in a significant way.

Q. And in your opinion, is it possible that there was prolonged partial asphyxia before the serious episode that happened, as you posited, at around [9:00 p.m.]?

A. Yes. Or, it, it's certainly possible. It's really unknown.

[39] As I read her reasons, the trial judge was fully cognizant of this nuance in Dr. Watts’ testimony. I am not persuaded that she misapprehended Dr. Watts’ evidence. Read fairly, in my view, the trial judge’s reasons indicate that she relied on Dr. Watts’ evidence concerning the possibility of a prolonged partial asphyxia before 9 p.m. together with her analysis of the totality of the causation evidence to conclude that Alecia suffered a period of prolonged partial asphyxia that preceded the events around 9 p.m. In my view, it was open to the trial judge to do so.

[40] It is important to consider the trial judge’s findings on causation in context. The trial judge conducted her analysis of the respondents’ claims in ten main sections: three sections dealing with whether Dr. Atack met the standard of care; three sections on causation in relation to Dr. Atack; three sections dealing with whether the hospital and the nurses met the standard of care; and a concluding section addressing causation in relation to the nurses.

[41] Because of the way in which the trial judge structured her reasons, she reviewed the bulk of the law in relation to causation and the bulk of the causation evidence when addressing the claim against Dr. Atack. When reviewing the law, she noted the limits on an expert’s ability to express an opinion on causation as compared to a trier of fact’s ability to take a “robust and pragmatic approach” to the evidence and to find causation proven on a balance of probabilities even in the absence of a firm medical opinion:

Causation often cannot be demonstrated with scientific precision. It is essentially a question of fact that can be determined by the trial judge through the application of ordinary common sense. [Citation omitted.] [Emphasis added.]

However, in some circumstances… an inference of causation may be drawn from the evidence without positive scientific proof. This is known as the “robust and pragmatic” approach to causation. [Citation omitted.]

It is not necessary to secure a firm medical expert opinion on the existence of causation. Medical experts tend to speak in terms of scientific certainties that are more stringent than the standards required by the law. As noted above, the legal standard only requires demonstration on a balance of probabilities. [Citation omitted.]

[42] The trial judge reviewed the expert causation evidence extensively when dealing with the claim against Dr. Atack. This section of her reasons consists of 109 paragraphs spanning 24 pages. As noted above, the focus of her inquiry at this stage was evidence concerning the timing of Alecia’s brain injury. She began with a brief overview of the opinions of the main causation experts and then reviewed the evidence relating to each of the factors affecting their opinions on timing.

[43] In her overview of Dr. Watts’ opinion, the trial judge referred both to his evidence about the possibility that a prolonged partial asphyxia occurred before 9 p.m. and to his evidence that a prolonged partial asphyxia likely occurred after 9 p.m. She said:

Dr. Watts testified that he would expect to see a bradycardia lasting 10-15 minutes if there were a prolonged partial asphyxia. He felt that the asphyxia occurred before the onset of electronic monitoring. He thought there was likely a significant event at around [9:00 p.m.] and that it was certainly possible that there was a partial prolonged asphyxia before the serious episode at around [9:00 p.m.]

He thought the most likely cause of Alecia’s injury was a combination of an acute near total event and prolonged partial asphyxia.

Dr. Watts testified as follows:

A. Essentially, my, my summary was that when one puts together the findings on the, the heart rate pattern, and her physical findings, and the investigations, including the CT scan, blood tests and everything else, it points to the most likely form of injury being an acute near total insult with, probably with a prolonged partial component. And the absence of evidence that that prolonged, that acute near total insult took place at the end of labour makes it considerably more likely, in my mind, that that acute near total insult took place earlier. And one logical time for that is the period immediately prior to [9:00 p.m.], [9:05 p.m.], and that that, that, the observations at that point represented the tail end of such an acute near total event. There may well have been a combination of prolonged partial asphyxia, particularly in, in, either in the form of cord compression and/or changes in uterine blood flow that occurred subsequently. But it seemed to me that, looking at the pattern on the, on the fetal heart tracing that that's so unlike what one sees in severely injured infants at birth. In fact, it's a pattern that is seen commonly in, in the, the second stage of labour in babies who are otherwise normal, that the substantive injury is likely to have been that that occurred earlier, as, as I say, probably at around [9:00 p.m.] hours. [Emphasis added.]

[44] After summarizing the opinions of the other main causation experts, the trial judge turned to a review of the evidence relating to each of the factors affecting their opinions on the timing of Alecia’s brain injury. In the course of this review, the trial judge identified several factors pointing to the unlikelihood that significant brain injury occurred in the final minutes before birth, and to the likelihood of an acute near total asphyxial event around 9:00 p.m. that was preceded by a period of prolonged partial asphyxia in the hours before birth:

▪ The absence of evidence of a sentinel event. Dr. Watts testified that the term “sentinel event” is most commonly used to refer to an acute evident physical cause for an asphyxial insult, such as rupture of the uterus, premature and massive separation of the placenta from the uterus, premature major abruption, or prolapse of the umbilical cord.

The trial judge noted that a sentinel event will cause a prolonged bradycardia of the fetal heart and, in the absence of a sentinel event, “timing brain injury becomes more difficult.”

The trial judge observed that only Dr. Perlman thought there was evidence of a sentinel event between 9:17 and 10:00 p.m. He thought that the last few moments of the fetal heart rate tracing showed evidence of a severe bradycardia, i.e., a fetal heart rate less than 60 b.p.m., indicating a sentinel event. The other experts disagreed. None of them thought there was evidence of a sentinel event during this period.

▪ The emergence of seizures between 9:17 p.m. and 10:00 p.m. or shortly after birth. The trial judge noted that the development of seizures is one of the known sequelae of asphyxia during labour; that Dr. Perlman testified that Alecia began having seizures at 40 minutes of age; and that Dr. Watts testified that seizures do not begin until hours after an asphyxial event.

The trial judge also noted Dr. Schifrin’s evidence about an ominous sawtooth pattern on the fetal heart rate tracing between 9:17 p.m. and 10:00 p.m., suggesting that Alecia was having seizures in utero. Although other experts did not agree with Dr. Schifrin’s opinion, the trial judge observed that if there was a sawtooth pattern on the tracing shortly after it began, “that would seem to indicate brain injury having occurred before the electronic tracing began”, which was “consistent with Dr. Watts’ evidence on the issue.”

▪ The distribution of Alecia’s brain injury. The radiological evidence at trial demonstrated that Alecia sustained injury both to the cortex or hemispheres of the brain and to the central area of the brain (the basal ganglia, thalami and putamen). In addition, one expert testified that Alecia may also have suffered brain stem injury. A significant portion of the expert evidence at trial dealt with whether there is a correlation between particular patterns of brain injury and the different form(s) of asphyxial insult.[^10]

The trial judge noted that Drs. Perlman, MacGregor, Watts, Gagnon and Low[^11] all agreed that Alecia need not have suffered an acute near total asphyxia to explain the pattern of her brain injury. She also highlighted Dr. Perlman’s evidence that Alecia’s brain injury could have been caused by any one of the three forms of asphyxia (acute near total, prolonged partial, or a combined form of asphyxia); that the more typical clinical picture of a combined form of asphyxia is prolonged partial followed by acute near total; and that 40 minutes of prolonged partial asphyxia at the more severe end of the spectrum followed by acute near total asphyxia would be a perfect model to explain Alecia’s injuries.

Other expert evidence addressed whether there is any correlation between the type of cerebral palsy Alecia suffers and the likely form(s) of asphyxial insult. The trial judge concluded that the evidence on this issue “would seem to suggest both types of asphyxia occurred.”

▪ Multisystem involvement. The trial judge explained the general consensus of the expert witnesses that Alecia’s multisystem injury (injury to parts of the body other than the brain) tends to support the likelihood of a prolonged period of asphyxia. Multisystem injury can occur where the fetus compensates for asphyxia by shunting blood to the brain, heart and adrenal glands at the expense of other bodily systems. Where an acute near total insult occurs, the shunting mechanism does not have time to respond in the same way.

Dr. Perlman, in particular, noted the presence of kidney dysfunction, liver dysfunction and transient cardiovascular abnormality, indicating an asphyxial event in the 12-24 hours before birth, with the caveat that impairment of these organs is not a “fine timing device”. Dr. Watts agreed that the indications of transient cardiovascular abnormality are a reliable marker of prolonged partial asphyxia and also that kidney dysfunction is not a fine timing device.

▪ Lymphocytes and nucleated red blood cells. Drs. Perlman, Watts and Low all discussed the significance of elevated lymphocyte (white blood cell) levels and the presence of nucleated red blood cells (recently formed red blood cells that still contain a nucleus) in Alecia’s blood. The trial judge noted that this was a new area of science but highlighted Dr. Perlman’s evidence that Alecia’s elevated lymphocyte levels might indicate hypoxia one to two hours before birth.[^12] The trial judge explained that the onset of an asphyxial insult and brain injury are not simultaneous; and that injury occurs only at some point following the onset of asphyxia. She noted Dr. Perlman’s evidence that elevated lymphocyte levels are only seen within a few hours after the onset of an asphyxial insult and that the presence of nucleated blood cells, although perhaps not helpful in relation to timing, was in no way inconsistent with that conclusion.

Concerning Dr. Watts’ opinion, the trial judge said he agreed with Dr. Perlman that lymphocyte counts could not be used to pinpoint the timing of a brain injury to within five to ten minutes of birth or even 30 to 50 minutes. She said, “[a]s already noted, Dr. Watts felt the asphyxial event occurred earlier than the ten minutes before birth and perhaps some hours earlier.”

▪ Umbilical cord gases. Arterial and venous umbilical cord blood is routinely drawn when a baby is born in a depressed state to permit analysis of umbilical cord gases (PO2 and PCO2). Cord gases are an indication of fetal oxygenation immediately prior to birth and are used to evaluate a newborn’s acid base status as a measure of fetal response to labour.[^13] The trial judge reviewed the evidence relating to this issue in detail.

Alecia’s blood gas values at birth were 23 mmol. per litre (arterial) and 17 mmol. per litre (venous) respectively. Base deficits in excess of 16 mmol. per litre are known to lead to brain damage.

The trial judge noted that Dr. Perlman relied on the difference between Alecia’s arterial and venous gas levels at birth (the AV difference) in formulating his opinion that Alecia suffered a sentinel event in the last six minutes before birth. He opined that the six mmol. AV difference had accumulated at the rate of one mmol. per litre as a result of complete oxygen deprivation during the last six minutes before birth, indicating a sentinel event.

However, the trial judge observed that there was no evidence of total asphyxia during the period from 9:54 to 10:00 p.m. Rather than an immediate and persistent fall in the fetal heart rate, a pattern of recurrent variable decelerations with a return to baseline continued until the final two minutes before birth when the heart rate went down and stayed down. Dr. Watts testified that the returns to baseline were inconsistent with Dr. Perlman’s theory. He also said that Alecia’s fetal heart rate tracings during the period between 9:17 and 10:00 p.m. are of a type commonly seen in the second stage of labour and are usually associated with normal infants.

The trial judge also noted Dr. Low’s evidence that, despite Alecia’s accumulated base deficit, the actual oxygen levels in the umbilical artery and vein at the time of sampling indicated the blood gas exchange occurring at the time of birth was good. In addition, the trial judge referred to Dr. Watts’ testimony that base deficit accumulates more slowly in cases of prolonged asphyxia as well as Dr. Low’s evidence that there was nothing in the last six minutes of the labour that could be responsible for the six mmol. rise in the arterial base deficit from 17 to 23. Dr. Low testified that this rise could be explained by a prolonged partial asphyxia.

The trial judge concluded that the clinical data in this case does not support “the window of injury postulated by the plaintiffs to be as narrow as either the six minutes before birth or the 3½ to four minutes before birth and the 90 seconds following delivery.”

[45] After reviewing the foregoing evidence, the trial judge set out her conclusions on causation in relation to Dr. Atack. While essentially rejecting the respondents’ theory, she commented favourably on the evidence of Drs. Low, Watts and Schifrin. She said:

Dr. Low’s view was that an acute total or near total exposure just before birth was not likely because there was no evidence of a sentinel event and no evidence of prolonged terminal bradycardia. The renal complications exhibited by Alecia point to asphyxia over time. Similarly, the blood gases at the time of delivery are inconsistent with an acute asphyxial exposure of short duration just before birth.

On the other hand, Dr. Watts opined that it was more likely that the injury did involve an acute near total component as well as a prolonged partial component, but he felt that it occurred earlier than at the end of labour. He thought the acute component likely occurred at around [9:00 p.m.].

Dr. Schifrin also testified that there was “surely” some form of oxygen deprivation before [9:00 p.m.]. He further agreed that there could have been a severe, prolonged cord compression between [8:35] and [9:00 p.m.] and that the 17 minutes during which the nurses could not auscultate a fetal heart rate could represent the tail end of such a severe, prolonged cord compression. [Emphasis added.]

[46] The trial judge explained that she preferred Dr. Low’s evidence concerning blood gas values to that of Dr. Perlman. Turning to the pattern of injury and its relationship to the type of asphyxial insult, she rejected Dr. Armstrong and Dr. MacGregor’s evidence as dogmatic and as tied to out-dated studies, and stated:

I also accept Dr. Watts’ evidence with respect to the type of insult Alecia likely endured and his testimony with respect to timing. Dr. Watts gave his evidence in a cogent and straight forward way and his conclusions accord with common sense.

[47] It was only after this analysis and review that the trial judge set out her conclusions on causation as set out in paragraph 27 above (repeated here for ease of reference):

Having considered the evidence, I am not satisfied that it is possible to pinpoint the timing of Alecia’s injury with minute accuracy and certainly not to within minutes of birth, particularly in the absence of a sentinel event. Moreover, there is no objective evidence of an acute near total event at this time. There are only two minutes of bradycardia which by itself is insufficient to cause the extent of Alecia’s injury. I am not satisfied that there is evidence pointing to a sentinel event during this interval.

I am satisfied that the following findings of fact are supported by the evidence on a balance of probabilities. Alecia most probably sustained intrapartum asphyxia. There was, more likely than not, a period of prolonged partial asphyxia and a period of more acute asphyxia, the cumulative effect of which accounts for the pattern of brain injury that she sustained and the type of cerebral palsy that she exhibits. While it is not possible to pinpoint the timing of the asphyxial insult(s) with complete precision, I accept Dr. Watts' evidence that there was likely both partial prolonged asphyxia and an acute near total event that occurred in the hours before birth. I accept Dr. Watts’ opinion that the partial prolonged asphyxia occurred in the hours before birth and that the events at [9:00 p.m.] represented the tail end of a bradycardia representing a near total acute insult. In this regard, I accept Ms. Rawlins’ testimony about her observation that the fetal hear was barely fluttering. There is no doubt in my mind that Ms. Rawlins, an experienced labour and delivery nurse, knew what she was seeing-a heart that was beating very slowly.

[48] Reading the trial judge’s reasons fairly and as a whole, in my view, the trial judge relied on the totality of her review of Dr. Watts’ evidence, together with the evidence of other experts, to reach her conclusions that: i) the timing of Alecia’s brain injury could not be pinpointed with precise accuracy and certainly not to within minutes of birth; ii) Alecia was injured as a result of the cumulative effects of a prolonged partial asphyxia and an acute near total asphyxia that took place in the hours before birth; and iii) the acute near total asphyxia likely occurred around 9 p.m.

[49] Although there was no firm expert opinion evidence or clinical evidence to support the trial judge’s further conclusion that a prolonged partial asphyxia preceded the events around 9 p.m., this conclusion was supported by Dr. Watts’ evidence and the evidence of other experts concerning the following matters:

▪ the possibility that a prolonged partial asphyxia occurred before 9 p.m. and prior to the commencement of electronic monitoring;

▪ the pattern of Alecia’s brain injury and the type of cerebral palsy she suffers are consistent with a combined form of asphyxia;

▪ Alecia’s lymphocyte counts, cord gas levels, radiological results and the presence of multisystem injury following birth are all consistent with brain injury in the hours before birth;

▪ although not necessarily the case, common sense would probably indicate that in a combined form of asphyxia, prolonged partial asphyxia is succeeded by acute near total asphyxia;

▪ the fetal heart rate tracings for the period between 9:17 and 10:00 p.m. were of a type associated with normal infants; and

▪ although a prolonged partial asphyxia occurring after the significant event at 9:00 p.m. may have contributed to the cord gas levels, it is unlikely that contribution would have changed Alecia’s outcome in any significant way.

[50] I recognize that, other than referring to the fact that he gave the opinion, the trial judge did not address Dr. Watts’ evidence that a prolonged partial asphyxia likely followed the events that occurred around 9 p.m. However, I take it as implicit in her reasons that the trial judge concluded that in addition to any prolonged partial asphyxia that may have occurred after 9 p.m., there must also have been an earlier prolonged partial asphyxia. In light of the evidence referred to in paragraph 49, it was open to the trial judge to do so.

[51] In my view, the trial judge did not make a palpable and overriding error in finding that Alecia suffered a prolonged partial asphyxia that preceded the events around 9 p.m.

2. Do the trial judge’s findings on causation reflect a misapplication of the law of causation?

[52] Before addressing this issue, I will make some general comments about the law of causation.

i) The Law of Causation

[53] Causation is an expression of the relationship that must be found to exist between the tortious act of the wrongdoer and the injury to the victim in order to justify compensation of the latter out of the pocket of the former: Snell v. Farrell, 1990 70 (SCC), [1990] 2 S.C.R. 311, at p. 326. The primary test for establishing causation in negligence cases is the “but for” test. This was recently affirmed by the Supreme Court of Canada in Resurfice Corp. v. Hanke, 2007 SCC 7, [2007] 1 S.C.R. 333. The plaintiff must establish that “but for” the negligent act or omission of the defendant, the injury would not have occurred: see Athey v. Leonati, 1996 183 (SCC), [1996] 3 S.C.R. 458, at para. 14. In special circumstances, the court may apply the “material contribution test”. However, the circumstances where this test can be applied are limited: Hanke, at paras. 24 and 25. [^14]

[54] Whatever test for causation is applied, it is clear that scientific precision is not required to support a finding of causation: Snell, at p. 328; see also: Aristorenas v. Comcare Health Services (2006), 2006 33850 (ON CA), 83 O.R. (3d) 282 (C.A.). Accordingly, in medical malpractice cases, an expert capable of providing a firm opinion that supports the plaintiff’s theory of causation is not required: Snell, at p. 330. Rather, the trial judge is entitled to consider all the facts and circumstances established by the evidence at trial, and, where appropriate, to draw an inference of causation through the application of reason and common sense. This approach has been termed “the robust and pragmatic approach”:

The legal or ultimate burden remains with the plaintiff, but in the absence of evidence to the contrary adduced by the defendant, an inference of causation may be drawn although positive or scientific proof has not been adduced. If some evidence to the contrary is adduced by the defendant, the trial judge is entitled to take account of Lord Mansfield’s famous precept.[^15] This is, I believe, what Lord Bridge had in mind in Wilsher when he referred to a “robust and pragmatic approach to the … facts: Snell, at p. 330; see also Aristorenas.

[55] Further, at p. 331 of Snell, Sopinka J. made the following comments concerning the differing roles of the trier of fact and expert witnesses:

The respective functions of the trier of fact and the expert witness are distinguished by Brennan J. of the United States Supreme Court in the following passage in Sentilles v. Inter-Caribbean Shipping Corp., 361 U.S. 107 (1959) at pp. 109-10:

The jury’s power to draw the inference that the aggravation of petitioner’s tubercular condition, evident so shortly after the accident, was in fact caused by that accident, was not impaired by the failure of any medical witness to testify that it was in fact the cause. Neither can it be impaired by the lack of medical unanimity as to the respective likelihood of the potential causes of the aggravation, or by the fact that other potential causes of the aggravation existed and were not conclusively negated by the proofs. The matter does not turn on the use of a particular form of words by the physicians in giving their testimony. The members of the jury, not the medical witnesses, were sworn to make a legal determination of the question of causation. They were entitled to take all the circumstances, including the medical testimony, into consideration.

[56] However, as indicated in the quotation from Snell at paragraph 54 above, the robust and pragmatic approach does not shift the burden of proof away from the plaintiffs. Rather, the plaintiff must still “provide an evidentiary foundation for finding that there is a substantial connection between the injury and the defendant’s conduct”: Barker v. Montfort Hospital (2007), 2007 ONCA 282, 278 D.L.R. (4th) 215, at para. 54.

[57] Further, as this court emphasized at paras. 54 and 60 of Aristorenas, the robust and pragmatic approach offers a method for evaluating evidence. It is not a substitute for evidence that the defendant’s negligence caused the plaintiff’s injury; nor does it change the amount of proof required to establish causation.

[58] Finally, just as the robust and pragmatic approach cannot be used as a substitute for reviewing and making findings about relevant evidence.

[59] Put another way, the robust and pragmatic approach does not permit drawing inferences concerning either the ultimate issue of causation or links in the chain of causation without reviewing the relevant evidence and without making findings about the range of available inferences. As Rouleau J.A. pointed out at para. 63 of Aristorenas, quoting from Fairchild v. Glenhaven Funeral Services, [2002] 3 All E.R. 305 (H.L.), at para. 150, common sense cannot become a substitute for resort to the evidence:

[E]ven though it is always for the judge rather than for the expert witness to determine matters of fact, the judge must do so on the basis of the evidence, including the expert evidence. The mere application of “common sense” cannot conjure up a proper basis for inferring that an injury must have been caused in one way rather than another.

ii) Application of the Law to the Facts of this Case

[60] The trial judge’s finding that the nurses caused Alecia’s injury is contained in a section of her judgment entitled “Conclusions: Causation”. This section follows immediately after a section dealing with the nurses’ standard of care and comprises five paragraphs. The essence of the trial judge’s reasoning is set out in the following three paragraphs:

It seems clear that during [the period of up to 17 minutes after 9 p.m.] there was either a bradycardia or decelerations of the fetal heart, consistent with what Dr. Watts thought was an acute near total event. As I have already accepted on the basis of Dr. Watts’ testimony, there was an earlier episode of partial prolonged asphyxia that occurred in the hours before birth. I find that it is more likely than not that this occurred between [6:00 and 7:30 p.m.] when the fetal heart was not being auscultated. There is support for this conclusion because no decelerations or bradycardia were detected at the other times when the fetal heart was being auscultated properly. Furthermore, I have already concluded that Alecia’s injuries were caused by this combination of asphyxial insults. As a result, I am satisfied that the plaintiffs have established on a balance of probabilities that Alecia’s injuries were caused by the negligence of the nursing staff for which the hospital is liable.

In particular, it seems reasonable to infer that if the nurses auscultated the fetal heart between [6:00 and 7:30 p.m.], they would have detected decelerations or perhaps a bradycardia of the fetal heart, raising concern that some form of asphyxial event was occurring. It further seems reasonable to conclude that they would have conducted the appropriate nursing interventions and they would have likely summoned assistance. It is likely that Dr. MacMillan would have responded in the same fashion he did at [9:15 p.m.], namely by rupturing the membranes and attaching a fetal scalp clip.

In my view, the electronic tracing would have detected the magnitude of the bradycardia (or decelerations) occurring at [9:00 p.m.] and would have shown that an acute event was occurring, consistent with Dr. Watt’s view. There can be no doubt that in such circumstances, steps would have been taken to expedite delivery, thereby sparing Alecia the magnitude of her injuries. [Emphasis added.]

[61] The trial judge’s conclusion on causation turns on a series of findings and inferences:

▪ Alexia’s brain injury is the result of the cumulative effect of a prolonged partial asphyxia and combined near total asphyxia suffered during labour;

▪ at 9 p.m. nurses detected a bradycardia or decelerations of the fetal heart, consistent with an acute near total event taking place between approximately 8:45 and 9:08 p.m.;

▪ an earlier period of prolonged partial asphyxia occurred in the hours before birth;

▪ Mrs. Fisher was in active labour at 6 p.m., requiring that monitoring of the fetal heart every half hour begin at that point;

▪ prior to 9 p.m., no decelerations or bradycardia were detected during periods when Alecia’s fetal heart rate was being monitored properly;

▪ the prolonged partial asphyxia that occurred prior to 9:00 p.m. likely occurred between 6:00 and 7:30 p.m., the period during which the nurses failed to auscultate the fetal heart according to the hospital’s protocol;

▪ had the nurses auscultated the fetal heart as required at 6:30 and 7:00 p.m. they would likely have detected decelerations or bradycardia;

▪ having detected decelerations or bradycardia, the nurses would have summoned assistance, Dr. MacMillan would have attended and attached a fetal scalp clip, and electronic fetal heart rate monitoring would have commenced; and

▪ electronic monitoring would have detected the events that occurred around 9 p.m. and would have led to an expedited delivery, sparing Alecia the magnitude of her injuries.

[62] The trial judge made the first five of these findings following a detailed review of the relevant evidence. However, her sole reference to the evidence in relation to her final four findings was to the evidence that no decelerations or bradycardia were detected on the occasions when the fetal heart was auscultated properly.

[63] The trial judge’s central findings that the prolonged partial asphyxia occurred between 6:00 and 7:30 and that the nurses would have detected it had they auscultated the fetal heart as required at 6:30 and 7:00 p.m. rest on a conclusion that the prolonged partial asphyxia that occurred before 9 p.m. was detectable by intermittent auscultation every half hour. However, the trial judge made no explicit finding to that effect, stating instead that it was “reasonable to infer that if the nurses auscultated the fetal heart [as required], they would have detected decelerations or perhaps a bradycardia”.

[64] In my view, taken as a whole, the trial judge’s reasons indicate that she made the final four findings based on common sense, applying the robust and pragmatic approach. Although I agree that that approach can be used to evaluate evidence and to draw reasonable inferences concerning the issue of ultimate causation, particularly in a case involving complex medical evidence, it cannot be used as a substitute for reviewing relevant evidence and making necessary findings of fact.

[65] None of the expert witnesses at trial testified that Alecia’s injury was caused in the precise manner determined by the trial judge. As I have said, the respondents’ experts testified that the asphyxial insult began between 40-50 minutes and up to one to two hours before birth and that the injury happened within minutes of birth. The balance of the expert evidence was divergent, but no one expressed an opinion that corresponds to the findings of trial judge.

[66] Further, and in particular, there was no expert evidence specifically relating to the crucial issues of whether the earlier period of prolonged partial asphyxia likely occurred between 6:00 and 7:30 p.m. and concerning whether and when the earlier period of prolonged partial asphyxia would likely have been detected by intermittent auscultation at half hour intervals. However, although there was no expert evidence at trial addressing these issues specifically, there was extensive expert evidence concerning the nature of prolonged partial asphyxia and some expert evidence concerning when and how it can be detected. The trial judge did not review any of this evidence.

[67] Among other things, the evidence relating to prolonged partial asphyxia indicated that it can take several forms. For example, it can be intermittent or persistent; or it can be persistent with intermittent periods of increased severity. In addition, the severity and duration of prolonged partial asphyxia that will lead to injury can vary dramatically. In this respect, there was relative consensus among the experts at trial that prolonged partial asphyxia must last a minimum of about 30 to 35 minutes to itself cause injury. However, it was also clear that a more moderate form of the asphyxia may require several hours to cause damage. Further, Dr. Low testified that short recurrent episodes of asphyxia may have a cumulative effect so that an exposure that may not itself cause any damage could cause damage if repeated over time.

[68] There was also expert evidence concerning whether, in a combined form of asphyxia, the prolonged partial component would be separated from the acute near total component. Dr. Armstrong testified that it is unlikely that the two forms of asphyxia would be separated from each other; physiologically, it is more likely that one form of asphyxia would progress to the other.

[69] Finally, there was conflicting expert evidence at trial concerning the likelihood of prolonged partial asphyxia being detected. For example, although speaking in terms of the likelihood that electronic heart rate monitoring would have picked up Alecia’s distress, Dr. Schifrin testified:

A. …It is medically implausible that a normal baby will have any degree of oxygen deprivation during labour and not reveal it. It is unknown that a baby who starts out normal and who is subjected to any form of oxygen deprivation during labour, will not fail to show there’s a problem. I may have said that confusingly. Will not fail to show it has a problem. Let me try and say that it...

Q. How will it show it?

A. Oh, by having decelerations. There is no known example of a baby, normal to start, subjected to oxygen deprivation during labour who will not immediately manifest a response to that oxygen deprivation, whatever the mechanism of the oxygen deprivation, and will not fail to tell you, the reasonable observer will not fail to tell you that I have a problem.

Q. And so, if there had been fetal heart rate monitoring with the tracing since the beginning, are you saying that would have shown there?

A. Yes. I believe it would have been normal. And on the basis of that, having believing that it would have been normal, it is, I believe, medically implausible, implausible that the fetus would not, would have failed to have decelerations in response to any form of oxygen deprivation during labour.

Q. And so, you are saying that some time before [9:00 p.m.], there was some form of oxygen deprivation?

A. Yes, reasonably, from one mechanism or another. Surely.

Q. Sure. And had there been a tracing, it would have, that deprivation would have shown in the tracing as a deceleration?

A. Immediately, yes. And immediately, and reasonably at a time when something could have been done about it, without question.

[70] Further, although speaking about a standard of monitoring requiring auscultation every 15 minutes, Dr. Bernstein said:

Q. And you would expect that if an asphyxial event was occurring during labour while you're auscultating, if you were listening by auscultation, more likely than not, the fetal heart would at some point exhibit some abnormality - be it bradycardia, tachycardia, or decelerations which would suggest to the listener that this is something that needs to be either assessed electronically, or prompt some other action - but you would expect the listener to observe something?

A. Well, I mean, I think that was the belief that auscultation could provide you with that information. So, I would hope that it was true, otherwise there's no point in doing it.

[71] Dr. Watts’ evidence about the indicia of prolonged partial asphyxia is complicated. At one point in his evidence, he seemed to say that a prolonged partial asphyxial insult that causes injury will manifest irregularities in the fetal heart rate for longer than 30 minutes:

Q. Then let's turn to prolonged partial asphyxia. Is it possible to say, as with acute true total asphyxia, it takes roughly this long in partial asphyxia to injure the brain?

A. It's difficult because prolonged partial asphyxia is very, very variable in its severity, so there's going to be a wide difference in the amount of time that it takes for brain damage to occur. One can start with the evidence from the animal experiments that prolonged partial asphyxia of a very severe nature takes a minimum of 30 minutes to cause brain damage. And it's commonly stated as being prolonged hypoxia at a level less than ten to 15 percent of normal. However...

Q. Sorry. Ten to 15 percent of what value?

A. Of, of, of, of the normal oxygen level...

Q. In the blood?

A. ... oxygen saturation.

Q. Very good.

A. It's usually expressed by people like Myers in terms of volumes percent of, of oxygen in the blood and .5 to 1.5 volumes percent through roughly ten, five to 15 percent. And we also know that at levels of hypoxia higher than that, at about 40 percent...

Q. So, less severe?

A. So, less severe.

Q. Yes?

A. One begins to get changes in the fetal blood pressure and the fetal heart rate. And presumably in most infants, the actual asphyxial insult will have started with when the oxygen level is even higher than that 40 percent. So - and it may be easier if I, if I drew this as a graph, but...

Q. Well, you drew a sketch for me quite recently.

A. Yeah.

Q. And that's why we have the easel up. And I wonder if you could go to that sketch for Her Honour now and explain what you're trying to tell us.

A. If, if one plots the level of oxygen in the fetus - and it doesn't matter whether it's percent or volumes against time - what we know is that if it's below...

Q. Can I - I'm very sorry to interrupt you. Are we speaking about an animal experiment now or...

A. We're talking about animal experiments.

Q. Okay. Go ahead.

A. If, if we, if we, if we call it 100 percent, 40 percent and ten percent, at roughly ten percent a period of 30 minutes is sufficient to cause brain damage, but in order to get to the 30 - that ten percent, the oxygen level is, in the fetus has to fall, and it's falling from somewhere up in, in, in this, this range.

Q. When you say “this range”, roughly what?

A. Between 40 and 100.

Q. All right.

A. When it reaches 40, it will start to produce bradycardia. So that, in saying that 30 minutes of severe asphyxia is sufficient to cause brain damage, we're saying that some longer period of bradycardia would be associated with brain damage. And that's why it's difficult to apply this 30 minute figure from the animal experiments directly to what you observe in labour or, labour and delivery because you actually start the timing probably from the point at which you can see something happen to the fetal heart tracing. In fact, the asphyxial insult has even been there for longer than that because there's a period before you start to see changes in the heart rate when the oxygen is still dropping. That period is probably relatively short because that's the point where the oxygen levels change fastest.

[72] However, shortly after giving that evidence, Dr. Watts appeared to suggest that the external manifestations of an injury-causing prolonged partial asphyxia could cover a much shorter period:

Q. Can you answer this question then? If you're seeing a prolonged period of that deep low oxygen level having been produced by some pattern of prolonged partial asphyxia, can you tell us what one would expect to see on the electronic fetal monitor tracing?

A. I would expect to see some evidence of late decelerations at an earlier stage in, in this, and of a persistent bradycardia at a later stage.

Q. What do you mean by “persistent?”

A. A bradycardia lasting at least ten to 15 minutes, maybe longer.

Q. And what sort of frequency of late decelerations would you expect to see? Or, if you can't tell us, I don't want to push it.

A. I'm not sure whether I can tell you in terms of frequency. I certainly would get alarmed if I saw more than four or five.

[73] Further, in cross-examination by trial counsel for the hospital, Dr. Watts agreed that prolonged partial asphyxia can go undetected:

Q. All right. And as I also understand, again, using your clinical judgment and opinion based on all of the evidence you've reviewed, that you came to the opinion that in addition to the near total acute that you believe Alecia suffered as is evidenced by the events surrounding [9:00 p.m.], that I think you used the term - or based on your sort of differential diagnosis - you believe the most likely scenario was that a prolonged partial continued after...

A. After...

Q. ... 21:00?

A. ... that time, yes.

Q. And even though there isn't - I think as you suggested - evidence on the tracing, that that, it continued during the period of time when the fetal monitor was on.

A. Yes.

Q. And it doesn't surprise you, as you've told us, that that could have occurred where the prolonged partial continued and was not evident from the tracing that was generated, because that's something you've seen, I'm not going to say frequently, but it's not uncommon.

A. That's correct.

Q. Is that correct?

A. Yes.

[74] In addition, Dr. Gagnon seemed to suggest that the fetal heart rate will not necessarily reflect an asphyxial episode:

Q. Dr. Gagnon, do you agree with me, to get me back on track, that if there had been auscultation before [9:00 p.m.], every 15 minutes, every five minutes after every contraction, whatever model you want to take, that it would have picked up an asphyxial event?

A. Certainly, it would have picked up abnormality in the heart rate that could be suggestive of an asphyxial event...

Q. Thank you, sir.

A. ... whether or not that heart rate event would reflect an asphyxial episode or not, that's not necessarily true.

[75] When the trial judge set out her initial findings on causation in the section of her judgment dealing with the liability of Dr. Atack she reviewed the relevant expert evidence carefully and in detail before setting out her conclusions. However, in contrast to her approach in that section, in the section of her judgment dealing with the liability of the hospital, the trial judge did not review the relevant expert evidence at all. In fairness to the trial judge, this was a long trial involving complicated medical evidence and the parties did not assist her.

[76] Out of almost 500 pages of written closing submissions, the respondents’ trial counsel (not Mr. Pape) devoted one paragraph to the submission that an asphyxial event occurred between 6:00 and 7:30 p.m. that would have been picked up by auscultation every half hour:

Even if the standard had been every 30 minutes, a period of 1 ½ hours from [6:00 to 7:30 p.m.] would be below the standard, and it could well be that there may have been some hypoxic events during this time which would have been picked up by proper auscultation, therefore alerting the nursing staff and physicians of such possibility.

Trial counsel did not review the evidence capable of supporting this submission. Moreover, despite the obvious risk to the hospital’s position if failure to auscultate during this period was found to be a breach of the standard of care, the hospital (not Mr. Rogers or Ms. Varjacic) did not respond directly to this issue.

[77] Nonetheless, to make a finding of causation based on a robust and pragmatic approach, in my view, it was incumbent on the trial judge to consider and make findings about the evidence relevant to the medical issues. This is particularly the case where there was no opinion evidence dealing directly with the theory of causation determined by the trial judge.

[78] As I have said, there was evidence that was relevant to the crucial questions of the nature of prolonged partial asphyxia and whether and when the earlier period of prolonged partial asphyxia could and should have been detected. At least some of that evidence concerning intermittent prolonged partial asphyxia and how it manifests itself pointed to a realistic prospect that the earlier period of prolonged partial asphyxia may not have been detectable by auscultation at half hour intervals. In my opinion, before drawing an inference of causation against the hospital, the trial judge was required to consider the relevant evidence and to make findings, on the standard of a balance of probabilities, eliminating the prospect that the earlier period of asphyxia was not detectable. I conclude that she misapplied the law of causation by failing to do so.

[79] It is not possible for this court to make the necessary findings. To do so would require findings of fact on conflicting evidence and including issues of credibility. Regrettably, therefore, there must be a new trial, unless the hospital can show that there was no evidence to support a finding of causation based on the nurses’ negligence. I now turn to that issue.

3. Do the trial judge’s findings on causation lack an evidentiary basis?

[80] As I have said, the trial judge’s conclusion on causation turns on a series of findings and inferences.[^16] Apart from the trial judge’s findings that Mrs. Fisher entered active labour at 6 p.m., that Alecia’s injury was the result of the cumulative effect of a prolonged partial asphyxia and combined near total asphyxia suffered during labour, and that Alecia suffered an acute near total event around 9 p.m., the hospital contends that there was no evidentiary foundation for inferences the trial judge drew. I disagree.

[81] I have already rejected the hospital’s claim that the trial judge made a palpable and overriding error in finding that the prolonged partial asphyxia that contributed to Alecia’s brain injury preceded the events around 9 p.m. In addition, I have identified some evidence that may have supported a finding that the prolonged partial asphyxia would have been detectable, which, in turn, may have supported an inference that it occurred between 6 and 7:30 p.m. However, as I have said, before drawing that inference, it was necessary that the trial judge review and make findings about the evidence suggesting that the earlier prolonged partial asphyxia would not have been detectable.

[82] In my view, there was also some evidence at trial capable of supporting the trial judge’s further inferences that, having detected decelerations or bradycardia, the nurses would have summoned help, electronic fetal heart rate monitoring would have commenced and would have detected the events around 9 p.m., which, in turn, would have led to an expedited delivery alleviating the magnitude of Alecia’s injuries. For example:

▪ Ms. Wood (a nursing expert called by the respondents) testified that as soon as there are risks or abnormalities detected in the fetus through auscultation, that is “an automatic immediate indication for continuous monitoring”;

▪ Dr. Schifrin testified that earlier electronic monitoring would have detected an acute hypoxic ischemic event at its earliest appearance and provided an opportunity to correct it;

▪ the essence of Dr. Watts’ evidence concerning the events around 9 p.m., which was the period when intervention was required, was that they likely represented a period of bradycardia of up to 23 minutes in duration, and that the injury to Alecia’s brain occurred in the last few minutes of that period;

▪ Dr. Low testified that there is no question that an obstetrician would intervene in the face of a prolonged bradycardia; and

▪ several witnesses testified concerning the likelihood that a crash caesarean section can be performed within 10 minutes.[^17]

[83] Although there was little evidence at trial concerning whether Dr. MacMillan would have been available to attach a fetal scalp clip between 6 and 7:30 p.m. and concerning whether Drs. Muir (the anaesthesiologist) and Atack could have been available to perform a crash caesarean section prior to 9 p.m., depending on the trial judge’s findings concerning what happened between 6 and 7:30 p.m., detailed evidence concerning these issues may not have been necessary. Depending on the nature of the earlier problem that was detected, it may have been reasonable to infer that it was incumbent on the medical professionals to take steps so that they were prepared to intervene quickly if required. I accordingly reject the appellant’s submission that the trial judge’s findings concerning causation lacked an evidentiary basis.

VI. Disposition

[84] Based on the foregoing reasons, I would allow the appeal, set aside the judgment against the hospital and order a new trial of the respondents’ claim against the hospital. Although the parties agreed on costs of the appeal of $100,000, taking account of all the circumstances, I would award costs of the appeal in the amount of $50,000.00 inclusive of disbursements and applicable G.S.T. in the cause of the new trial.

RELEASED: November 12, 2008 “MR”

“Janet Simmons J.A.”

“I agree Marc Rosenberg J.A.”

“I agree K. Feldman J.A.”

[^1]: Auscultation involves manual monitoring of the fetal heart rate with a fetone (an ultrasonic stethoscope).

[^2]: One of the respondents’ experts, Dr. Schifrin testified that Mrs. Fisher was a high-risk patient, which mandates electronic fetal heart monitoring from the outset. Her family doctor, her obstetrician and all other experts who addressed the subject at trial testified that she was a low risk patient.

[^3]: Evidence at trial indicated that a normal fetal heart rate can be described either as being between 120 and 160 b.p.m. or as being between 110 and 150 b.p.m.

[^4]: A transducer is a fetal heart rate ultrasound machine with an external monitor that displays the heart rate. A fetascan is a real-time ultrasound machine that can display the fetal heart.

[^5]: There was evidence at trial concerning a possible five-minute discrepancy between the clock in the labour room and the time on the scalp clip tracing. Thus, when the tracing from the scalp clip recorded a time of 9:17 p.m., the clock in the labour room may have read 9:12 p.m. The trial judge did not make a specific finding resolving this discrepancy and nothing turns on it for the purposes of this appeal.

[^6]: Dyskinetic cerebral palsy involves abnormal movements of the body. Dystonic movement is involuntary movement where posture is sustained, and athetoid movement refers to involuntary writing-type movement.

[^7]: As will be explained more fully later in these reasons, it is important to distinguish between an asphyxial insult and an asphyxial injury. An asphyxial insult must persist for some period of time before an asphyxial injury will occur.

[^8]: A bradycardia is a very slow heart rate below the normal range.

[^9]: Dr. Watts’ reference to 9:08 p.m. appears to relate to the time when he thought the fetal heart rate started to recover. Based on some printouts from the transducer, Dr. Watts testified that the fetal heart rate appeared to be starting to recover at around 9:03 p.m. but that this could have been 9:08 p.m. according to the labour room clock.

[^10]: Although not unanimous in the way they expressed it, the expert witnesses agreed that early animal research indicated that acute near total asphyxia causes damage to some areas of the brain while prolonged partial asphyxia causes damage to other areas. For example, Dr. Watts testified that the research showed that acute near total asphyxia causes damage to the brain stem and in some instances to the thalamus but not to the cortex or basal ganglia while prolonged partial asphyxia causes damage primarily to the cortex and possibly the basal ganglia but not to the thalamus or the brain stem. The trial judge accepted the expert evidence indicating that “current thinking has moved away from such hard and fast rules.”

[^11]: Drs. Gagnon and Low were expert witnesses called by the doctors.

[^12]: Hypoxia refers to a lack of oxygen.

[^13]: Low levels of oxygen in the fetal blood lead to a simultaneous increase in carbon dioxide, producing respiratory acidosis, which can develop into metabolic acidosis (lack of oxygen in body tissues). This, in turn, leads to a decrease in pH or base levels, and an increase in acid levels in the fetal blood. A base deficit occurs when the base concentration decreases below normal as the body uses it to attempt to buffer acid. According to Dr. Low, brain damage does not occur until metabolic acidosis develops. Dr. Gagnon testified that metabolic acidosis is what injures the brain.

[^14]: Two requirements must be satisfied for the material contribution test to be applicable. First, it must be impossible for the plaintiff to prove that the defendant’s negligence caused the plaintiff’s injury using the “but for” test due to factors beyond the plaintiff’s control. Second, it must be clear that the defendant breached a duty owed to the plaintiff, thereby exposing the plaintiff to an unreasonable risk of injury, and the plaintiff must have suffered that form of injury: Hanke, at paras. 24 and 25. One situation where it may be appropriate to apply the material contribution test is where it is impossible to say which of two tortfeasors caused the injury; another is where it is impossible to prove what a particular person in the causal chain would have done had the defendant not committed the negligent act or omission. The examples set out in Hanke of when it will be impossible for the plaintiff to prove the defendant’s negligence using the “but for” test due to factors beyond the plaintiff’s control seem to support the following conclusions of Rouleau J.A. in Aristorenas v. Comcare Health Services (2006), 2006 33850 (ON CA), 83 O.R. (3d) 282 (C.A.), at para. 53: Thus it would seem that the “material contribution” test is applied to cases that involve multiple inputs that all have harmed the plaintiff. The test is invoked because of logical or structural difficulties in establishing “but for” causation, not because of practical difficulties in establishing that the negligent act was a part of the causal chain.

[^15]: In Blatch v. Archer (1774), 1 Cowp. 63, 98 E.R. 969, Lord Mansfield stated at p. 970: It is certainly a maxim that all evidence is to be weighted according to the proof which it was in the power of one side to have produced, and in the power of the other to have contradicted: Snell, at p. 328.

[^16]: For ease of reference, I will reproduce the series of findings and inferences set out in para. 49 above: i) Alexia’s brain injury is the result of the cumulative effect of a prolonged partial asphyxia and combined near total asphyxia suffered during labour; ii) at 9:00 p.m. nurses detected a bradycardia or decelerations of the fetal heart, consistent with an acute near total event taking place between approximately 8:45 and 9:08 p.m.; iii) an earlier period of prolonged partial asphyxia occurred in the hours before birth; iv) Mrs. Fisher was in active labour at 6:00 p.m., requiring that monitoring of the fetal heart every half hour begin at that point; v) prior to 9:00 p.m., no decelerations or bradycardia were detected during periods when Alecia’s fetal heart rate was being monitored properly; vi) the prolonged partial asphyxia that occurred prior to 9:00 p.m. likely occurred between 6:00 and 7:30 p.m., the period during which the nurses failed to auscultate the fetal heart according to the hospital’s protocol; vii) had the nurses auscultated the fetal heart as required at 6:30 and 7:00 p.m. they would likely have detected decelerations or bradycardia; viii) having detected decelerations or bradycardia, the nurses would have summoned assistance, Dr. MacMillan would have attended and attached a fetal scalp clip, and electronic fetal heart rate monitoring would have commenced; and ix) electronic monitoring would have detected the events that occurred around 9:00 p.m. and would have led to an expedited delivery, sparing Alecia the magnitude of her injuries.

[^17]: For example, Dr. Hodd testified that a crash C-section can be performed within 5-6 minutes if the anaesthetist is available and everything is set up; Dr. Schifrin testified that if the equipment is ready and an anaesthetist available, a C-section could be performed within approximately 10 minutes; and nurse Rawlins testified that crash C-sections take 10 minutes.