---

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

Non-Reportable

Case no: 928/2017

In the matter between:

MEC FOR HEALTH, WESTERN CAPE                                                                  APPELLANT

and

S Q obo I                                                                                                                     RESPONDENT

 

Neutral citation: MEC for Health, Western Cape v Q.

(928/2017) [2018] ZASCA 132 (28 September 2018)

Coram: Lewis, Ponnan, Majiedt, Dambuza and Mathopo JJA

Heard: 16 August 2018

Delivered: 28 September 2018

Summary: Delict – medical negligence – onus on plaintiff to prove damage causing event – reverse reasoning impermissible - inference of negligence may only be drawn where a negligent event is the most plausible inference.

Expert evidence – expert opinion must be logical - probative value of expert opinion evidence depends on reasons furnished therefor.

ORDER

On appeal from: Western Cape Division of the High Court, Cape Town (Le Grange J sitting as court of first instance):

1 The appeal is upheld with costs, such costs to include those consequent upon the employment of two counsel.

2 The order of the Western Cape Division of the High Court is set aside and substituted with the following order:

‘The plaintiff’s claim is dismissed with costs, such costs to include those consequent upon the employment of two counsel’.

JUDGMENT

Dambuza JA (Lewis, Ponnan, Majiedt and Mathopo JJA concurring):

Introduction

[1] This appeal, with the leave of this court, relates to heartrending events which took place in 2011. On 7 December a baby girl, Baby I. Q. (I.), was born at Paarl Hospital in the Western Cape. At birth she presented with clinical features of spastic quadriplegic cerebral palsy. This had been caused by damage to her brain.

[2] In May 2013 the respondent, I.’s mother, instituted an action for damages in the Western Cape Division of the High Court, Cape Town (the high court), against the appellant, the Member of the Executive Council for Health in the Western Cape (the MEC). Her claim was based on negligent medical treatment to herself and the baby by health services employees employed at various health State facilities which she attended during her pregnancy and at the time of the baby’s birth. The claim was defended by the MEC.

[3] On the first day of the trial an order was granted by the high court separating the issues of liability from the determination of the quantum of damages. The trial then proceeded on the issue of liability. The high court found that the injuries suffered by the baby were a direct consequence of the negligent conduct of the MEC’s employees. The court ordered the MEC to pay 100 per cent of proved or agreed damages. It is against this order that the MEC appeals.

 

Background Facts

[4] The following events were not in issue before the high court. On 27 July 2011 the respondent visited Zwelihle Clinic in Hermanus. There, her suspicion that she might be pregnant was confirmed. The nursing staff told her that she was five months into her pregnancy and that her anticipated date of delivery was 16 November. That initial visit was followed by further routine antenatal clinic visits, the first of which was on 24 August.

[5] On 20 September a medical examination revealed that the respondent had contracted a urinary tract infection (UTI). An antibiotic (Amoxicillin) was prescribed and she was told to return to the clinic two weeks later. The infection was still persistent when she returned to the clinic on 4 October. Treatment was extended and she was advised to return to the clinic a week later, on 10 October, for ‘urine check & Dr’s opinion’. On her return her urine still had nitrites despite continued treatment. She was told to return to the clinic two days later.

[6] Although leucocytes were found to be present in her urine when she returned to the clinic on 12 October, she was not experiencing any difficulty or pain when discharging urine (no dysuria) and was ‘asymptomatic’. The doctor ordered that her urine sample be sent for microscopy, culture and sensitivity testing (MST). There is no record of the result of this investigation in the medical records.

[7] Two weeks thereafter, on 16 November, her due date, the respondent returned to the clinic. After assessment by a doctor she was told to return on 23 November. By then the UTI had resolved. Between the 16 and 23 November she relocated from Hermanus to Mbekweni Township in Paarl. Consequently, on 23 November she attended the Phola Park Clinic in Paarl. There she was told, without assessment, to return to the clinic the following Tuesday, 29 November, that being the next antenatal clinic day. On her return to that clinic, on 29 November, her urine showed no abnormalities and she was referred to Paarl Hospital for the following day, as she was assessed to be post mature.

[8] On 30 November the respondent’s urine had leukocytes, but she had no dysuria. A Non Stress Test (NST)[1] returned a ‘non-reactive’ result. Dr Bilson, who attended to her, instructed that a second NST be done and that if the result was reactive the respondent was to return in a week for a follow up assessment. If it was non-reactive she was to be admitted to the labour ward. The test was redone on the same day and was assessed, mistakenly, as reactive. The respondent was told to return in a week. When she did so, on 7 December, the NST again returned a non-reactive result with a ‘poor baseline variability’.[2] She was admitted to hospital and labour was induced with Misoprostol inducing cream (also referred to as Cytotec). At 23h15 I. was born.

[9] At birth the baby presented with microcephaly.[3] Her head circumference of 32cm was abnormally small compared with her above average length of 54cm and normal weight 3,48kg. She had sustained a cephalohaematoma[4] and suffered an intra cerebral bleed. Her blood pH was 7.22. Her Apgar score[5] at birth measured four; at five minutes after birth it was seven, and at ten minutes after birth it had improved to nine. The medical staff recorded that the baby had suffered Hypoxic Ischaemic Encephalopathy (HIE), a brain injury that occurs when an infant’s brain does not receive sufficient oxygen and blood. Damage to I.’s brain had rendered her a spastic cerebral palsy quadriplegic. The following day she started having seizures.

[10] By twelve days after birth I.’s head had enlarged to 38cm. A sonar showed hydrocephalus.[6] She was transferred to Tygerberg Hospital. The results of a CTG[7] scan done at that hospital and the radiologist report confirmed the HIE and revealed cortical laminar necrosis with bilateral hygromas (cysts). These results are discussed more fully later in this judgment. I. was discharged from Tygerberg Paediatrics Hospital a day later and returned to Paarl Hospital from where she was discharged on 21 December.

 

Pleadings

[11] In the summons the respondent’s claim was based on various acts of alleged negligence by the medical staff during her pregnancy, at the time of birth, and after birth. Paraphrased, the allegations were that they failed to perform investigations necessary for the proper detection of danger to the child and to timeously intervene by performing a Caesarean Section (c-section). They also failed to monitor the respondent’s labour contractions and the baby’s heart rate. After delivery they failed to place the baby under the cooling-off system due to lack of beds at Paarl Hospital. As will be seen below, the evidence led on behalf of the respondent somewhat extended the ambit of the alleged grounds of negligence.

[12] In respect of the antenatal period, the allegations of negligence related to three periods. Firstly, weeks 32 to 34 of the respondent’s pregnancy, when she was suffering from the UTI,[8] secondly, on 30 November, when the two NST tests were done, and thirdly, during the week following 30 November until 7 December when, due to non-intervention, the respondent’s pregnancy was prolonged. In relation to 30 November, it later became evident that the assessment of the repeat NST done on that day as reactive was incorrect. The test had, in fact, returned a non-reactive result. The respondent’s contention was that the incorrect recordal was further negligent conduct which resulted in her pregnancy being prolonged, thus further increasing the risk of harm to the baby.

[13] Although in the summons the negligence during delivery was pleaded as an alternative to the negligence during the antenatal period, the evidence was presented in such a manner that composite consideration of all the alleged acts or omissions was contended for. The respondent persisted in this approach on appeal before us. The MEC denied any negligence on the part of the employees.

 

Issues

[14] The issues were the determination of the damage causing occurrence, when such damage occurred, whether there was negligence on the part of the MEC’s employees and if so, the causal link, if any, between the negligent conduct and the injury occasioned to I..

[15] The respondent relied on three expert witnesses: Prof Andronikou, a radiologist who performed a cranial MRI on the baby on 17 September 2013, Dr Mtsi, an obstetrician and gynaecologist, then a consultant at the Chris Hani Baragwanath Hospital, and Dr Reid, a clinical neurologist and part-time lecturer in neurology at Tygerberg and Groote Schuur Hospitals. The MEC relied on two witnesses: Prof van Toorn, a paediatric neurologist at Tygerberg Hospital and a part time lecturer at the University of Stellenbosch, and Prof Odendaal, a gynaecologist and obstetrician in private practice, also employed as a part-time lecturer at Stellenbosch University.

 

The evidence – Joint minute by Dr Mtsi and Prof Odendaal

[16] Prior to the start of the trial Dr Mtsi and Prof Odendaal concluded a joint minute. They agreed that the respondent should not have been sent home after 30 November when the NST results were non-reactive. Instead a c-section should have been done on that day. The different reasons held by each for this opinion are discussed below. They also agreed that labour should not have been induced on 7 December, when the NST was non-reactive. Furthermore, the CTG monitoring of the baby should have continued during delivery because of the assessment of poor variability. They agreed that the recordal of observations during labour was poor.

[17] Disagreement related, firstly, to Dr Mtsi’s acceptance of the anticipated date of delivery (16 November) as recorded on the respondent’s clinic card. Prof Odendaal was of the view that, deriving from the respondent’s last menstruation, on 15 February, the correct date of delivery was 22 November.

[18] Further, according to Dr Mtsi an NST done on 23 November was assessed as non-reactive. Prof Odendaal could not find such a test on the medical records. At the trial Dr Mtsi admitted that her reference to such a test done on this day was erroneous.

[19] Lastly, it was Dr Mtsi’s opinion that, had the baby been delivered by a c-section on 7 December, it would have been saved from harm or the extent thereof would have been ameliorated. Prof Odendaal’s opinion, however, was that the damage to the brain had occurred weeks prior to induction of labour. The c-section would not have improved the baby’s condition as it was already severely microcephalic when labour was induced.

[20] At the trial the respondent gave evidence. As would be expected, apart from narrating the history of her pregnancy and delivery, as set out above, she could not say much about the cause of the brain injury to her baby or when such injury occurred. Only two of her qualified experts testified on her behalf: Dr Mtsi, and Dr Reid. Both Prof van Toorn, and Prof Odendaal testified on behalf of the MEC.

 

Negligence during the antenatal period 20 September – 12 October

[21] Dr Mtsi’s opinion, based on a consultation she had had with the respondent, and examination of the respondent’s medical records, was that when the UTI persisted until 4 October, further investigations should have been done to determine the cause thereof. An MST test should have been done as early as 4 October, when the infection still persisted. Further, on that day the respondent should have been admitted into hospital instead of simply continuing with the antibiotics. This would have provided a more reliable diagnosis of the problem. At week 34 of the pregnancy, a c-section should have been done as the baby could have survived outside the hostile uterus. Failure to take these steps constituted substandard medical treatment by the medical staff. They ignored ‘signals of something having gone wrong’.

[22] Prof Odendaal, on the other hand, was of the opinion that although infection during pregnancy could result in cerebral palsy, it was improbable that the UTI could have had such a result. He referred to specific medical literature of a study conducted in this regard, the results of which revealed no link between UTI during pregnancy and cerebral palsy developing in a uterus. This evidence was undisputed.

 

30 November

[23] As stated above, two NST tests were conducted on this day. The first result was flat, showing very low beat to beat variability. The second appeared to show significant improvement from the first; the baseline variability appeared to have improved compared to the first. This is the result that was incorrectly recorded. Dr Mtsi’s evidence was that were it not for the negligent misreading of the repeat test, a c-section could have been performed to save the baby.

[24] Prof Odendaal’s opinion was that even if a c-section had been performed on this day it would not have prevented the injury as the brain damage and microcephaly had already occurred. But both Dr Mtsi and Prof Odendaal held the view that the respondent should not have been sent home on this day.

[25] Much of the MEC’s case regarding the microcephaly and brain damage is based on Prof van Toorn’s opinion drawn from the medical records of the CTG scan done at Tygerberg Hospital. The relevance of this opinion straddles the period starting from week 34 of the pregnancy to delivery of the baby on 7 December.

 

7 December

[26] Again, as already stated, both Dr Mtsi and Prof Odendaal were of the opinion that a c-section should have been done in view of the non-reactive NST. Both criticised the sparsity of recordings of the progress of the respondent’s labour and the baby’s heart rate. However, both Prof Odendaal and Prof van Toorn insisted that the microcephaly had already occurred by this time.

[27] Professor van Toorn gave detailed evidence of microcephaly development. He explained that a significant injury to the brain is followed by swelling, which subsides a day or so later and is accompanied by the death of some of the brain cells, leading to loss of volume of the head. This process happens over weeks. Where brain injury is sustained during delivery, the decline in head growth could be noticeable in the third to fourth month.

[28] Against this background, his opinion was that, given the extent of shrinkage of I.’s head, it was unlikely that it was a result of an occurrence during delivery or a week before then. The CTG scan done after day 12 of the baby’s birth showed severe cortical laminar necrosis (loss of the white brain matter). The whole brain was already severely shrunken. Fluid had collected and filled up the gaps around the brain where the white matter had been, hence the hydrocephaly.

[29] Prof van Toorn also explained that brain injury (resulting in death of brain cells) is usually visible on the CT scan at two weeks. At that stage the necrosis imaging presents as lines on the scan. It then peaks at one to two months after the insult. In this case, at 13 days after birth, the scan already showed ‘global white matter hypodensity’ (whole layers of cortical laminar necrosis), which was out of keeping with an injury at birth. Such extensive brain damage could only have developed over four to five weeks. On the other hand if there had been an intrapartum insult the head circumfirance would have been normal with a decline only becoming evident some months later. Further, the presence of gliosis (scar tissue) shown on the MRI scan was an indication that the insult occurred after 34 weeks as no scar tissue could have been formed prior to that time.

[30] In addition, Prof van Toorn referred to four internationally accepted criteria for diagnosis of an intra partum hypoxic event were met in this case. Only two of these were met in this case, the first being metabolic acidosis with blood acidity of less than pH 7 within the first two hours of birth, resulting from the lack of oxygen during birth. In this case I.’s blood pH of 7.22 at birth, was inconsistent with intrapartum brain injury. The second criterion – symptoms of encephalopathy at birth, and the third – cerebral palsy of a specific kind, were met. The fourth – the exclusion of other causes of brain injury was also not met because an ultrasound done on day 12 after the baby’s birth showed blood breakdown products or remnants of blood clots. This was evidence of previous bleeding in the baby’s brain or the HIE. Such bleeding could also have caused lack of oxygen and breakdown of the brain cells leading to the brain damage.

[31] Professor van Toorn also considered that several other factors militated against brain injury during delivery. The first was that there was no ‘sentinel’ (trigger) event during birth.[9] Further, the improvement in the Apgar score was uncharacteristic of intrapartum brain injury where the norm is that the five minute Apgar score remains at 0 to 3.

 

The high court’s findings

[32] In upholding the respondent’s claim the high court found that the defendant’s medical staff breached their legal duty towards the respondent and the baby:

‘by failing to assess, manage and treat them with the necessary degree of care, skill and diligence as expected of reasonable medical practitioners and or nursing staff in their position, particularly on 30 November 2011 and the subsequent induction on 7 December 2011.’

[33] Apart from this general conclusion the high court made no factual finding as to the cause of the brain damage. Nor did it set out explicitly the reasons for preferring the evidence of Dr Mtsi to that of the other expert witnesses. It also made no finding related to the UTI.

[34] Although the high court did not make a specific factual finding, it seemed to incline in favour of a conclusion that the brain damage to the baby occurred during birth. It reasoned that Prof van Toorn had not been ‘absolutely certain’ that the injury to the patient’s brain occurred prior to the start of labour. The high court remarked that there was a likelihood that it occurred during the onset of birth.

[35] It also found weighty the ‘diagnoses’ (contemporaneous notes) made by clinicians at Paarl and Tygerberg Hospitals, that the baby had suffered ‘birth asphyxia’. These notes, the high court reasoned, belied the conclusion that the injury occurred before labour as postulated by Prof Odendaal and Prof van Toorn. Instead, they supported Dr Mtsi’s opinion.

[36] The absence of a true factual finding as to the actual cause of the damage was not raised in the appeal. The MEC’s approach was to rather rely on the absence of a causal link between its employees’ alleged acts and/or omissions and the injury (and sequelae) suffered by the baby.

 

Discussion

[37] It is trite that she who asserts a damage causing event must prove it. The legal duty owed by the medical staff at the various health facilities to the respondent and her baby entailed that they adhere to the general level of skill and diligence possessed and exercised at the time by members of the branch of the profession to which they belonged.[10] They had no duty to provide the highest possible degree of professional skill. Only reasonable care and skill was required.[11] The respondent had to prove, through credible and persuasive evidence that the doctors and nurses failed to adhere to the required standards.

[38] The opinion of the medical experts was central to the determination of the required level of care and whether there was a breach of it. The requirement in evaluating such evidence is that expert witnesses support their opinions with valid reasons. Where proper reasons are advanced in support of an opinion, the probative value thereof is strengthened.[12]

‘It is not the mere opinion of the witness that is decisive but his ability to satisfy the Court that, because of his special skill, training and experience, the reasons for the opinion which he expresses are acceptable.’[13]

In Michael v Linksfield Park Clinic (Pty) Ltd[14] this court held that:

‘The court is not bound to absolve a defendant from liability for alleged negligent medical treatment or diagnosis just because evidence of expert opinion, albeit genuinely held, is that the treatment or diagnosis in issue accorded with sound medical practice. The court must be satisfied that such opinion has a logical basis, in other words that the expert has considered comparative risks and benefits and has reached a “defensible conclusion”’.

[39] The first difficulty with Dr Mtsi’s evidence is that she could not allude to a specific risk associated with the UTI that could have caused cerebral palsy. It was common cause that UTI is common in pregnant women. In her medico-legal report she had referred to its poor response to treatment, without highlighting it as a problem in the respondent’s management as she did in her evidence. She also admitted that the fact that by 4 October the urine still had nitrates did not necessarily mean that the respondent was not responding to the Amoxicillin treatment. In any event, the UTI problem had resolved by 16 October. Significantly, she could not dispute Prof Odendaal’s evidence that there is no study linking UTI to cerebral palsy and that the limited instances in which infection during pregnancy results in cerebral palsy are those in which the infection occurs in the placenta and its membranes. On the common cause facts, that is not what happened here.

[40] On her own evidence Dr Mtsi could not understand those portions of the records which were written in Afrikaans. She relied on the respondent in this regard. All of this probably affected her comprehensive understanding of the medical record, and the admitted to factual errors in her report.

[41] Further, there was no factual basis for her opinion that if the baby had been delivered during the period 32-34 weeks it would have been saved from harm, particularly considered against Prof Odendaal’s evidence that it is best to allow a baby to remain in utero to full term and that delivery of a baby prematurely was not a decision taken lightly.

[42] In relation to her opinion that the microcephaly was not picked up on the two occasions when scans were done on the 30 November, Prof Odendaal disagreed with her opinion that an ultra sound would have led to detection of the problem. His opinion was that there would have been no previous head measurement to compare with the measurement done at that stage. It was common cause that the respondent had missed the 18 week routine scan. Her first clinic visit was during the 24th week. In any event the evidence was that by that stage (30 November) the damage had already been done.

[43] Prof van Toorn was cross examined extensively on entries made by the staff at the paediatric ward. The first was the diagnoses of ‘birth asphyxia’ on I.’s hospital card. The contention on behalf of the respondent was that the notes were a reliable diagnosis of asphyxia during delivery and could not be ignored, particularly in the light of lack of the evidence by the authors thereof. Much was made of the failure by the MEC, to call these witnesses. Prof van Toorn’s evidence in this regard was that loose use of the term ‘birth asphyxia’ was a common occurrence amongst medical staff at Tygerberg Hospital. Further, those who made the diagnosis did not possess his expertise and had had no access to the information which became available to him after the CT scan was done. Hence the note on the same hospital records by those doctors that the baby should be referred to him for further medical attention.

[44] In addition, Dr Mtsi admitted that the term ‘birth asphyxia’ was non-descriptive in relation to the timing of the damage causing event; that it can mean antenatal, intrapartum or postpartum and that the terms ‘birth asphyxia’ and ‘perinatal asphyxia’ were used interchangeably in the medical records at Tygerberg Hostpital.

[45] The crux of Prof van Toorn’s opinion was that, on a consideration of all the evidence, the baby suffered an antenatal injury at about 34 weeks into the pregnancy or at least some weeks before labour was induced. Dr Reid agreed in the main with Prof van Toorn’s opinion. He accepted, for example, that if the initial blood sample, from which the baby’s pH level was taken within two hours of birth, and the pH level was 7.22, then Prof van Toorn’s opinions were entirely correct. As for the rest, Dr Reid’s evidence was limited to an interpretation of existing medical records. Regarding the timing of the brain injury he postulated two scenarios, antenatally and intrapartum. In his view, both scenarios were perfectly plausible, although he personally favoured an intrapartum asphyxial insult. Moreover and in any event, his field of expertise was adult neurology, unlike Prof van Toorn, who was a paediatric neurologist.

[46] The second aspect on which Prof van Toorn was cross examined at length related to notes made by Dr Friedericks, a paediatric registrar, that the delivery ‘look[ed] like ventouse delivery (using a vacuum); knob on head; has not cried yet’. The contention was that the ventouse delivery of the compromised baby contributed to the brain damage.

[47] Dr Mtsi’s persistence that the delivery was by means of ventouse ignored the records from the obstetrics ward staff, that this was a normal vaginal delivery. It also ignored the fact that the entry by Dr Friedericks was, in its own terms, conjectural. Prof Odendaal provided a logical response that if there had been ventouse delivery, the respondent would have remembered being placed in the lithotomy[15] position to facilitate such delivery. She had made no reference to that fact. Further, Prof Odendaal’s opinion was that even then it was unlikely that such a procedure would have affected the outcome, as microcephaly would not have developed over a week.

[48] The opinion evidence of Prof Odendaal and Prof van Toorn was founded on clearly established facts. It was logical and well reasoned.

[49] Whilst it must be accepted that on the evidence the notional reasonable medical practitioner in the place of the medical staff who attended to the respondent at Paarl Hospital would not have sent her home on 30 November, the evidence shows that the damage to the baby’s head had probably already occurred by that time. Dr Mtsi conceded that the foetus was probably already compromised by these date. There was therefore no causal link between the failure to intervene and the damage that was occasioned. The same conclusion must be reached in relation to reliance on alleged negligence subsequent to the 30 November.

[50] In the end, not only the cause of the damage remained unidentified but also its timing. The fact that harm had been occasioned was not, on its own, proof that the medical staff had caused it, or that they had done so negligently, or even that it had resulted in the brain injury. In Goliath v Member of the Executive Council for Health, Eastern Cape[16] this court warned against reverse reasoning of this kind as follows:

‘… to hold a doctor negligent simply because something had gone wrong, would be to impermissibly reason backwards from effect to cause.’

[51] Consequently the following order is made:

1 The appeal is upheld with costs, such costs to include those consequent upon the employment of two counsel.

2 The order of the high court is set aside and substituted with the following order:

‘The plaintiff’s claim is dismissed with costs, such costs to include those consequent upon the employment of two counsel’.

 

 

_________________

N Dambuza

Judge of Appeal

 

 

APPEARANCE

For Appellant: A Sholto-Douglas SC (with him S K Witten)

(Heads also prepared by H Bawa SC)

Instructed by:

State Attorney, Cape Town

State Attorney, Bloemfontein

For Respondent: D B Ntsebeza SC (with him Z L Mapoma)

Instructed by:

Thamsanqa Phillips Attorneys, Cape Town

Ponoane Attorneys, Bloemfontein

 

[1] This is an antenatal recordal of a foetal heart rate pattern. According to the glossary of terms supplied by the MEC, the result of the test is ‘reactive’ ‘if there are two or more foetal heart rate accelerations within a 20-minute period, with or without foetal movement discernible by the woman’. And it is ‘non-reactive’ if there is ‘a lack of sufficient foetal heart rate accelerations over a 40-minute period’. The evidence was that the acceleration should be 15 beats a minute. The acceleration is then increased over 15 seconds, to at least 20 beats per minute.

---