Member of the Executive Council for Health, Eastern Cape v Z M (576/2019) [2020] ZASCA 169 (14 December 2020)

SAFLII Note: Certain personal/private details of parties or witnesses have been redacted from this document in compliance with the law and SAFLII Policy

THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

JUDGMENT

Not Reportable

Case no: 576/2019

In the matter between:

THE MEMBER OF THE EXECUTIVE COUNCIL

FOR HEALTH, EASTERN CAPE

APPELLANT

and

Z M obo L M

RESPONDENT

Neutral citation: The Member of the Executive Council for Health, Eastern Cape v Z M obo L M (576/2019) [2020] ZASCA 169 (14 December 2020)

Bench: PETSE DP, ZONDI and NICHOLLS JJA and EKSTEEN and UNTERHLATER AJJA

Heard: 13 November 2020

Delivered: This judgment was handed down electronically by circulation to the parties' representatives via email, publication on the Supreme Court of Appeal website and release to SAFLII. The date and time for hand-down is deemed to be 09h45 on 14 December 2020.

Summary: Action in delict – acute profound brain injury during labour causing the child to suffer from cerebral palsy – whether the trial court could depart from the expert opinion of the radiologist agreed to by the parties – no basis found to depart from the causal factors implicated in an acute profound brain injury.

ORDER

On appeal from: Eastern Cape Division of the High Court, Bhisho (Hartle J sitting as court of first instance): judgment reported sub nom L v Member of the Executive Council for Health, Eastern Cape [2019] ZAECBHC 7

(a) The appeal is upheld.

(b) The order of the high court is set aside and replaced by the following order:

‘The plaintiff’s claim is dismissed.’

Unterhalter AJA (Petse DP, Zondi and Nicholls JJA, and Eksteen AJA concurring)

[1] On 3 December 2005, the respondent, Ms M, gave birth to a son, L (the child), at the Madwaleni Hospital. L was born with severe brain damage, as a result of which he suffers from cerebral palsy. Ms M instituted an action against the appellant in which she alleged that the negligence of the staff at the hospital failed to prevent the brain injury that occurred during her labour. That injury was caused by a lack of oxygen and blood flow to L’s brain. The technical description of reduced brain oxygen is hypoxic ischaemic encephalopathy.

[2] Ms M began to experience labour pains at 03h00 on 3 December 2005. Together with her mother, Ms M went to the hospital. They arrived at about 07h00, and Ms M’s membranes ruptured. She was admitted to the labour ward at about 08h00. There a nurse performed a vaginal examination, checked the foetal heart rate and informed Ms M that the birth of her child was still far from occurring, and that she should wait in a separate room. There are different accounts as to what Mpetshini was told in the course of the day and what further examinations took place. However, at 21h00 Ms M began to feel great pain and started screaming. She was then examined by a nurse to check the position of the foetus and the nurse informed her that she would give birth soon. A drip was inserted. Ms M was attended upon by two nurses. One of the important factual issues at trial was precisely what these nurses did during this final stage of labour. But it is clear that this stage of her labour lasted some 60 minutes before the child was delivered. During this time, and while Ms M was pushing, so too did a nurse push on her abdomen. How often, and with what amount

of force, was disputed. In addition, one of the nurses performed an episiotomy. L was born at 22h00. He did not cry. A doctor was called and, upon investigation, it was determined that L had suffered a severe brain injury during labour.

[3] At the trial, two issues were agreed by the parties. First, that the period in contention was the second stage of labour, though the parties were at odds, as I shall explain, as to how long that stage endured. Second, the parties stated that the report of Dr Pretorius, an expert radiologist, and its addendum, was admitted into evidence by agreement. Dr Pretorius considered an MRI scan of L’s brain. He concluded that the image indicated an acute profound hypoxic ischemic event that occurred intrapartum.

[4] The significance of this conclusion is an important matter. It was explained, in the course of the trial, by the expert obstetrician and gynaecologist, Dr Buchmann, who testified on behalf of the appellant. He testified that there is a distinction between an intrapartum acute profound brain injury (‘an acute profound injury’) and an intrapartum prolonged partial brain injury (‘a prolonged partial injury’). An acute profound injury is severe, with total or near-total asphyxia (deficient supply of oxygen); it is of short duration, and sudden onset, and generally occurs 30 minutes before delivery. A prolonged partial injury is less severe, with partial asphyxia; it develops slowly over several hours; it is often preceded by a deteriorating foetal heart rate that gives a warning of developing hypoxia, that is, lack of oxygen. Professor Buchmann stated that an acute profound injury is caused by a catastrophic sentinel event, most likely, in the case of L, the event was the compression of the umbilical cord. Dr Pretorius identified the injury suffered by L as an acute profound injury.

[5] The expert called to testify on behalf of Ms M was Dr Ndjapa-Ndamkou. He is also an obstetrician and gynaecologist. Dr Ndjapa-Ndamkou

readily recognised that he was not qualified to express an opinion as to the conclusion reached by Dr Pretorius, since he is not a radiologist, and deference was thus warranted. While there are passages of his testimony where Dr Ndjapa-Ndamkou appeared to stray from his area of expertise to suggest that there were features of the case consistent with a prolonged partial injury, ultimately, he accepted that L suffered an acute profound injury.

[6] The court below was not convinced. Hartle J considered that there could be a ‘mixed injury pattern’,[1] connoting an injury pattern that has some of the characteristics of a prolonged partial injury and some of the characteristics of an acute profound injury. This postulate was advanced for two reasons. First, Hartle J’s ‘own recourse to literature makes it plain that mixed patterns do exist and that when energy substrates are depleted in partial prolonged asphyxia … there can be a further assault in the form of near total collapse’.[2] Second, Hartle J considered that the absence of reasons offered by Dr Pretorius for his opinion was unhelpful and did not exclude the possibility of a mixed injury pattern.

[7] Hartle J proceeded to consider the evidence led at the trial. Hartle J criticised the evidence of Professor Buchmann. In particular,

Hartle J considered Professor Buchmann’s approach to be biased as to how fundal pressure (that is, applying pressure by pushing

downward on the mother’s uterus during delivery) was conducive to cerebral palsy; that his denial of high risk factors that had occurred during Ms M’s labour was unwarranted; and that his unwillingness to make concessions that deviated from his view that this was an unforeseeable acute profound injury lent support to an assessment of Professor Buchmann’s testimony as lacking objectivity. Hartle J considered Dr Ndjapa-Ndamkou’s evidence to be more probable and credible. The central finding is this:

‘Each of the risk factors highlighted by Dr Ndjapa-Ndamkou in a cascading manner point to a causal connection between the inappropriate

management of the plaintiff’s labour, including ( more especially) the injudicious and untimely application of the fundal

pressure in a situation where the child was already in a compromised position, and the superimposed acute profound hypoxic result.’[3]

Consequently, the high court held that the appellant was liable in delict to the respondent. Aggrieved by this result, the appellant appeals against that order with the leave of the high court.

[8] The appellant submitted that the court below fell into error. There was no basis to depart from the finding of Dr Pretorius that this was a case of an acute profound injury. Such an injury was properly characterised as a catastrophic sentinel event by Professor Buchmann. It was an unforeseeable injury, occurring in the final stage of labour, in all likelihood less than 30 minutes before delivery; and, in this case, it was mostly likely the complete occlusion of the umbilical cord for a period of 15 to 20 minutes that caused the injury. Neither the length of the labour, nor the recognised failure by the hospital staff properly to monitor and inspect the progress of the labour, nor the rupture of Ms M’s membranes, nor the application of fundal pressure, was causally implicated in the injury suffered by L. There was also no warrant, so it was contended, for the court below to find that Professor Buchman gave biased testimony.[4] Indeed, it was submitted on behalf of the appellant that the only bias that may be found was exhibited by the court below itself.

[9] Counsel for Ms M acknowledged the agreement made concerning the evidence of Dr Pretorius and the concessions made by Dr Ndjapa-Ndamkou, in particular, that L had sustained an acute profound injury. Nevertheless, it was contended that upon a reasonable extrapolation as to the length of Ms M’s labour, there may have been intermittent hypoxia (oxygen deprivation), compromising the ability of the foetus to recover. Hypoxia of this kind could have been detected had proper foetal monitoring been done, which it was not, and steps could then have been taken to avoid serious injury, for example by performing a timely caesarean section. This, together

with the negligent manner in which fundal pressure was administered, resulted in the acute profound injury.

[10] The first issue to be determined is whether the court below was justified in departing from the opinion of Dr Pretorius, that L had suffered an acute profound injury. It will be recalled that Hartle J did so on the basis of her own recourse to the literature and the absence of reasons for the opinion expressed by Dr Pretorius.

[11] This judicial assiduity strays beyond permissible bounds. The trial court will ordinarily consider the assistance given by an expert’s evidence and come to its own conclusions. But here the parties had agreed to accept the conclusion reached by Dr Pretorius and ran the trial on this basis. In these circumstances, the trial court should exercise very great caution. If Hartle J considered that Dr Pretorius’s conclusion was questionable, it was essential to give notice of this to the parties, as soon as possible, in the course of the trial, in turn securing the attendance of Dr Pretorius to explain his position. Absent this, the trial court’s judgment is not the place to reject Dr Pretorius’s opinion, nor to introduce a theory of injury at variance with the parties’ agreed position.

[12] Furthermore, a judge’s independent recourse to literature should also be undertaken with circumspection. The admissibility of expert evidence is predicated upon the special knowledge and skill of the expert to form an opinion that the judge, unassisted, cannot. A judge with an interest in a field of expert knowledge should not ordinarily undertake an independent search for relevant literature. And if it is done for some special reason, the yield of such explorations must be put to the parties and their experts. To do otherwise leaves the judge open to the very criticism made in this case by the appellant: that the court below had recourse to undisclosed literature to call into question an agreed conclusion that formed the basis upon which the trial was run - that L suffered an acute profound injury.

[13] The high court was not willing to accept the opinion of Dr Pretorius and characterised the injury as a ‘mixed injury pattern’. That was not a position available to the trial court. Without notice to the parties, the literature reviewed by the learned judge, not introduced into evidence, could not be used to justify a judicial characterisation of the injury. Nor could the learned judge safely reach any such characterisation, given the parties’ acceptance of the opinion of Dr Pretorius. Doubts as to that opinion required of Hartle J that she take up the matter with the parties and enjoy the benefit of testing her scepticism by securing the assistance of Dr Pretorius. This is not simply a question of fairness. To make a finding on a matter that requires an expert opinion, and to do so in contradiction of the opinion of an acknowledged expert, agreed upon by the parties, without testing the matter with Dr Pretorius, was to run a risk of error that cannot be allowed to stand.

[14] It follows that the matter must be considered from the agreed premise that the parties adopted at the trial. That premise was that L suffered an acute profound injury. The consequences of accepting this premise are far reaching. As I have explained, the nature of an acute profound injury and a prolonged partial injury are different. Those differences were set out by Dr Buchmann. Under cross examination, Dr Ndjapa-Ndamkou agreed that this was a case of an acute profound injury and he concurred with the proposition that there was no indication of a prolonged partial injury. Further, he agreed that an acute profound injury would require a complete occlusion of the umbilical cord for a period between 15 and 20 minutes.

[15] What emerges from the differences between the two types of injury is that their causes are also distinct. As their descriptions make plain, an acute profound injury results from a catastrophic event, as Professor Buchmann explained. A prolonged partial injury results from cumulative deprivations of oxygen over time. The injury develops over hours and it may be observed by a deteriorating foetal heart rate. Once this distinction is accepted, then very different time periods and causes are implicated for the purposes of understanding what gave rise to the injury suffered by L.

[16] Precisely because the court below appreciated that the distinctive nature of an acute profound injury excluded from consideration

aspects of the negligent treatment given to Ms M, the court questioned the conceptual distinction between the two types of injury

and adopted the notion of a mixed injury pattern. This permitted the court below to consider the cumulative risks borne by Ms M of the rupture of her membranes, the length of her labour, the failure to undertake proper monitoring, and the fundal pressure applied to her. But without the premise of a mixed injury pattern, what of these cumulative risks has any bearing upon the catastrophic event that caused the acute profound injury?

[17] Counsel for Ms M sought to resurrect the case on this score, even accepting that this was a case of an acute profound injury. He did so in two ways. First, he contended that the active phase of Ms M’s labour endured for considerably longer than Professor Buchmann allowed. There may have been intermittent hypoxia, over a considerable period of time, which would have weakened the foetus.

The nurses failed to do proper monitoring to determine whether this risk was present. In addition, the rupture of the membranes may have led to the reduction of amniotic fluid, and the added risk of cord compression. Taken together with the fundal pressure that was applied, the acute profound injury was a probable consequence. I shall refer to this as the cumulative risk argument.

[18] Second, counsel contended that the duration of the fundal pressure was in itself a sufficient explanation of the acute profound injury. The nurses applied this pressure in a manner that was both wrongful and negligent and hence the liability of the appellant was proven. I shall refer to this as the excessive fundal pressure argument.

[19] The cumulative risk argument rests on the factual premise that the active phase of labour was lengthy which gave rise to risks,

undetected by the nurses responsible for Ms M, that cumulatively contributed to the acute profound injury suffered by L.

[20] It was argued that this factual premise is supported by the evidence led at the trial. Counsel relied upon the evidence of Dr Ndjapa-Ndamkou. He testified that, during his consultation with Ms M, she had told him that a nurse had informed her at 13h00 that she would give birth at 16h00. If this was what a nurse believed at the time, it gives rise to the possibility that Ms M entered the active phase of her labour in the afternoon of 3 December 2005. Given the time of L’s birth at 22h00, Ms M would have been in the active phase of her labour for a lengthy time, with attendant risks.

[21] The difficulty with this evidence is that Ms M, who testified both before and after Dr Ndjapa-Ndamkou, made no mention of what the nurse had said to her. This is an important omission. What Dr Ndjapa-Ndamkou recounted of his consultation is hearsay. The

difficulty is not simply technical. Absent some cogent evidence that what was allegedly said by the nurse is supported by some other evidence that Ms M had entered the active phase of her labour during the afternoon, the hearsay evidence, even if considered, carries very little weight.

[22] The supporting evidence relied upon was this. Dr Ndjapa-Ndamkou placed some reliance on the guidance that dilation in the latent

phase of labour takes places at 1-2 cm per hour. An extrapolation using this guidance might suggest that Ms M would have given birth at 16h00, and this coincided with what she reported to Dr Ndjapa-Ndamkou the nurse had indicated to her. This was also consistent with the evidence of Ms M that at 13h00 she was feeling such pain that it did not permit her to lie on the bed until the contractions abated. Given when it was that the birth in fact occurred, Ms M endured a long active phase of labour that could have contributed to the risk of foetal distress.

[23] The difficulty with this evidence is that it rests upon the assumption that the progress of Ms M’s labour followed the

course of the guidance. There are no records as to the progress of her labour. There was some evidence of an examination of Ms M at 13h00, though this was not confirmed in the testimony of Ms M. There was no evidence given as to what the examination revealed. Professor Buchmann testified that there are very great variations as to the rate of dilation between women, and there is thus little to be learnt from the extrapolation exercise relied upon by Dr Ndjapa-Ndamkou.

[24] Absent some direct evidence as to the course of Ms M’s labour, the sparse details given by Ms M and the speculative generalisations

offered by Dr Ndjapa-Ndamkou do not provide evidence of sufficient weight to support the nurse’s estimate of 16h00, even if made, as being a plausible time at which the birth might have been expected to take place.

[25] What the evidence does establish is that Ms M cried out in pain at 21h00. She was then taken to the labour ward, a drip inserted, and she was attended upon by two midwives. Fundal pressure was applied by one of the midwives. There was some difficulty in the birth taking place. An episiotomy was performed, and L was born at 22h00.

[26] Without clear evidence that there was a long period of active labour that might have compromised the foetus and thereby contributed, in some way, to the injury suffered by L, there is no factual basis for the cumulative risk argument. Mere supposition will not suffice.

[27] The cumulative risk argument flounders for another reason. Professor Buchmann was very clear as to the nature of an acute profound injury. It is of short duration, and sudden onset, and generally occurs 30 minutes before delivery. In the evocative language of Professor Buchmann, it is caused by a catastrophic sentinel event. Dr Ndjapa-Ndamkou was constrained to accept this.

[28] The efforts by the court below to characterise the injury as being a mixed injury pattern cannot be supported on the evidence before the court. Dr Ndjapa-Ndamkou’s evidence was at odds with itself. He accepted that this was a case of an acute profound injury, yet he sought to implicate features of the case that are consistent with a prolonged partial injury. In particular, his efforts to suggest that there was a prolonged labour was the basis for suggesting that asphyxia might have developed slowly over several hours. But that is the very hallmark of a prolonged partial injury. Precisely what Dr Pretorius said it was not.

[29] There is simply no adequate basis upon which the factual evidence and a proper appreciation of the expert evidence supports the cumulative risk argument.

[30] I turn to the excessive fundal pressure argument. The issue here is well defined: could the midwife have applied such excessive fundal pressure so as to cause the catastrophic sentinel event that so grievously injured L?

[31] Professor Buchmann’s evidence was this. An acute profound event represents a complete shut-off of blood flow to the brain. After 10 minutes, brain damage starts; and after 25 minutes, most of the brain is damaged. Dr Ndjapa-Ndamkou was in agreement that an acute profound event would mean a complete occlusion of the cord for a period of between 15 and 20 minutes.

[32] When Ms M first gave her evidence, she explained that after 21h00 she was taken to the labour ward. There, in the course of her labour, one of the midwives pressed on the top side of her abdomen to assist her, as Ms M was pushing. Dr Ndjapa-Ndamkou referenced in his report that he had been told by Ms M that excessive forceful pressure was applied. At trial, when this evidence was sought to be led, counsel for the appellant objected on the basis that the evidence was hearsay. This led to the recall of Ms M. Upon her recall, Ms M’s evidence was that the midwife applied hard pressure to her abdomen and slapped Ms M on her right flank. The duration of this pressure was 5 minutes on and 5 minutes off (the 5/5 regime). Under cross-examination, when asked how Ms M could recall this timing so long after the event and given the pain of childbirth, Ms M reported that she had heard that the 5/5 regime was being applied from the midwives. Ms M also recalled that the 5/5 regime was applied three times before the baby was delivered.

[33] The appellant submitted that the evidence of hard pressure and the 5/5 regime should not be believed. It did not figure when first Ms M gave evidence, and it was implausible that Ms M could remember what she claims the midwives said, given that she was in the final stage of labour. Certainly, the trial court had some difficulty in believing this evidence, which was described as ‘bizarre’.[5]

[34] It is unnecessary to express a firm view on this aspect of the matter. Here, too, Professor Buchmann’s evidence is clear and not refuted by Dr Ndjapa-Ndamkou. An acute profound injury requires a complete cut-off of blood flow in excess of 10 minutes to start causing significant brain injury. The account given of the 5/5 regime does not do this. Professor Buchmann considered that this fundal pressure could not have caused the acute profound injury and he was sceptical that any midwife was capable of applying significant pressure for 5 minutes. Dr Ndjapa-Ndamkou, while accepting that complete occlusion was required, nevertheless maintained that the 5/5 regime might have compromised the recovery of the foetus. Here, too, the evidence of Dr Ndjapa-Ndamkou appears to be internally inconsistent. If the type of injury occurring in cases of acute profound injury requires a complete cut-off of blood flow in excess of 10 minutes, then it is unclear how pressure that does less than this, on three occasions, compromised recovery so as to bring about the same result.

[35] For these reasons, the excessive fundal pressure argument cannot be sustained.

[36] Once this is so, the following conclusions follow. First, Dr Pretorius had determined that the injury suffered by L was an acute profound injury. Second, an acute profound injury, as Professor Buchmann explained, with the ultimate concurrence of Dr Ndjapa-Ndamkou, was a catastrophic sentinel event of short duration and sudden onset, occurring in all likelihood 30 minutes before the birth. Third, what was done by the nursing staff in the last hour of Ms M’s labour was not causally implicated in the acute profound injury suffered by L. Fourth, there is neither sufficient evidence nor a defensible basis on the expert evidence to find that the risks arising from a lengthy period of labour occurred in this case. Even if they had, these risks were not causally operative in bringing about the brain injury suffered by L.

[37] It follows that the appeal must succeed.

[38] A few final observations are warranted.

[39] First, Professor Buchmann’s opinions were clearly stated and firmly held. I can discern no taint of bias in the manner in which Professor Buchmann defended his position. The court below was not justified in attributing bias to Professor Buchmann. Quite the contrary, his opinions have been of considerable assistance in determining this appeal. A court may decide that an expert opinion is incorrect, but a court should be slow to attribute bias to an expert of high standing because of the conviction with which the expert defends his opinions.

[40] Second, counsel for the appellant submitted that the court below exhibited bias. In the light of the conclusion to which I have come, there is no need to traverse the detail of this submission. What is at stake in a trial of the kind before Hartle J could not be of greater consequence. In these circumstances, it is of great importance that counsel and the court place a premium upon the dignity and reciprocal respect that allows courts to determine difficult cases.

[41] Third, counsel for the appellant made it plain that in the event that the appellant prevailed in the appeal, no costs were sought, either in the appeal or in respect of the trial. This was a proper stance to adopt.

[42] The following order is made:

________________

D N Unterhalter

Acting Judge of Appeal

APPEARANCES

For appellant: P de Bruyn SC (with him N D Ngadlela)

Instructed by: State Attorney, East London

State Attorney, Bloemfontein

For respondent: D Brown (with him T Mduba)

Instructed by: Dudula Yandis Attorneys, Johannesburg

[1] L v Member of the Executive Council for Health, Eastern Cape [2018] ZAECBHC 7, especially para 121.

[2] Ibid para 69.

[3] Ibid para 122.

[4] Ibid para 114.

[5] L (above fn 1) para 87.