Buthelezi v Ndaba (575/2012) [2013] ZASCA 72 (29 May 2013)

THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

JUDGMENT

REPORTABLE

Case No: 575/2012

In the matter between:

GABRIEL BUTHELEZI APPELLANT

v

PRISCILLA ZANELE NDABA RESPONDENT

Neutral citation: Buthelezi v Ndaba (575/2012) [2013] ZASCA 72 (29 May 2013).

Coram: Brand, Lewis, Cachalia, Majiedt JJA et Saldulker AJA

Heard: 13 May 2013

Delivered: 29 May 2013

Summary: Alleged medical malpractice – determination of negligence on the part of the surgeon – not to be inferred from concession by surgeon that something must have gone wrong during the operation – dependent on evaluation of the reasoning underlying conflicting expert opinions

________________________________________________________________

ORDER

________________________________________________________________

On appeal from: KwaZulu-Natal High Court, Pietermaritzburg (Jappie J sitting as court of first instance):

1 The appeal is upheld with costs.

2 The order of the court a quo is set aside and replaced with the following:

‘The plaintiff’s claim is dismissed with costs.’

________________________________________________________________

JUDGMENT

________________________________________________________________

BRAND JA (LEWIS, CACHALIA, MAJIEDT JJA et SALDULKER AJA CONCURRING):

[1] This is a case about alleged medical negligence. The appellant, Dr Gabriel Buthelezi, practises as a specialist gynaecologist and obstetrician in Pietermaritzburg. On 13 February 2006 he performed a surgical operation known as a total abdominal hysterectomy on the respondent, Ms Priscilla Ndaba. Some time after the operation, she began to suffer from urine incontinence. Further examination by a urologist revealed the cause of her problem to lie in what is known in medical parlance as a vesico-vaginal fistula which can be described in layman’s terms as a hole in the posterior wall of the bladder and the contiguous anterior wall of the vagina. What happened in these circumstances, so the medical experts explained, was that urine passed from the bladder to the vagina above the control mechanisms which caused the urine incontinence. Relying on these facts, the respondent instituted an action against the appellant in the court a quo for the damages she suffered in consequence of the vesico-vaginal fistula (the fistula) which resulted from his alleged negligence in performing the hysterectomy operation.

[2] When the matter came before Jappie J in the court a quo he was asked, by agreement between the parties, to determine the issues pertaining to the appellant’s liability first while the quantum of the respondent’s alleged damages stood over for later determination. During the preliminary proceedings that followed, it became apparent that the only relevant dispute hinged on whether or not the appellant was negligent and that this question in turn depended on an election between the opposing views expressed by two medical experts in this regard. On the one hand Prof R W Green-Thompson testified, on behalf of the respondent, that in his view the appellant was negligent in performing the hysterectomy operation. On the other hand, the appellant’s expert witness, Prof L Snyman, took the position that he was not. At the end of these proceedings Jappie J upheld the respondent’s claim on the merits with costs. In consequence he ordered the appellant to pay the respondent such damages as either agreed on or as the respondent may prove at the second stage of the trial. The appeal against that judgment is with the leave of the court a quo.

[3] I find it convenient to start the recital of the facts with an introduction of the medical specialists involved in the order that they testified. First, there was Dr Abdul Dada, a specialist urologist who was called to testify on behalf of the respondent and who was responsible for diagnosing and eventually repairing the fistula that caused the incontinence problem. Secondly, there was Prof Green-Thompson, the former head of the department of obstetrics and gynaecology at the Faculty of Medicine, University of KwaZulu-Natal, to whom I have already referred. Thirdly, there was the appellant himself who testified on his own behalf. Albeit that he was not called as an expert witness, it transpired that when he performed the operation on the respondent, he had practised as a specialist in his field for nearly 20 years during which period he had carried out about 1 000 hysterectomy operations. Finally there was the appellant’s expert, Prof Snyman, who held a joint appointment as professor of medicine at the University of Pretoria and the principal specialist at the department of obstetrics and gynaecology at the Kalafong Hospital in Atteridgeville, Pretoria.

[4] Before I turn to the differences in opinion between these medical specialists that emerged during the trial, I propose to deal with the large areas of their testimony which proved to be common cause. First amongst these was the fact that the fistula of about one centimetre in diameter, which was observed by Dr Dada some six weeks later during the remedial operation, did not occur during the hysterectomy, nor immediately thereafter, but that it progressively developed over the ensuing period. At the same time the specialists agreed that the development of the fistula did not start spontaneously, but that it was triggered by something that happened during the hysterectomy. On this aspect they specifically agreed that the following factors, singularly or in combination, could not in themselves cause the fistula: the fact that the respondent was HIV positive; that she suffered from diabetes; that she had previously undergone a caesarean section; and that she had been struggling with chronic pelvic infection at the time. Yet, they also concurred that because of these factors, the respondent was at an increased risk of the formation of a fistula.

[5] An abdominal hysterectomy without complications, so the medical specialists further agreed, involves the removal of the uterus from the upper part of the vagina, known as the vaginal vault, and the subsequent closing of the hole in the vault caused by the removal. The bladder lies anterior to the uterus and although the two organs are in close proximity, they do not share the same wall nor the same blood supply. Of critical importance, so the specialists agreed, is to separate the bladder from the uterus before the operation starts and to protect it from harm during the operation. Normally the separation presents little difficulty, because the connection between the two organs consists of what is described as loose areola tissue. But if the patient had undergone a previous caesarean section, it is likely that there will be some tethering or adhesions between the two organs which the surgeon has to cut through with some sharp surgical instrument.

[6] Prof Green-Thompson’s thesis as to what happened in this case is that, during the hysterectomy operation, some damage was caused to the blood supply to an area of the bladder wall. The compromised area then became devascularised – that is, deprived of blood – which in turn led to a progressive tissue necrosis of the bladder wall and eventually to the formation of the fistula. Because no direct damage to the bladder wall occurred during the surgery, the problem would not have been noticeable at the time. The fistula developed and increased over time through ongoing tissue necrosis in the devascularised part of the bladder wall. That explains why the respondent’s urine incontinence did not manifest itself immediately subsequent to the operation, but only some time thereafter.

[7] As to when and how the damage to the blood supply of the bladder wall occurred, Prof Green-Thompson expressed the view that it happened either when the bladder was separated from the uterus or when the uterus was removed from the vaginal vault. In any event, he said, the problem was caused by the appellant’s failure to ensure that the bladder was kept separate and safe during the course of the operation. If reasonable precautions had been taken, so Prof Green-Thompson concluded, the injury to the bladder would not have occurred. What Prof Green-Thompson’s opinion thus amounted to, as I understand it, was that the fact of the injury to the respondent’s bladder inevitably gave rise to the inference that reasonable precautions were not taken and hence that the appellant was negligent.

[8] The appellant’s independent recollection of the operation on the respondent was virtually non-existent and his contemporaneous notes were not of much assistance. These recorded only that ‘a standard TAH” ie a total abdominal hysterectomy was done; that there was some difficulty due to oozing of the vaginal vault; and that haemostasis was secured, ie that the oozing was stopped. This meant, so the appellant testified, that it was a straightforward operation; that there were no complications and no difficulties, save for the oozing of the vaginal vault, which had soon been resolved. Whether or not there were tetherings or attachments between the respondent’s uterus and bladder, as a result of her previous caesarean section, he could not remember. But if there were, he would not regard it as a problem worthy of mention because he had encountered it on so many occasions in the past.

[9] The reference to a ‘standard TAH’ in his notes also led him to the inference that he used the standard technique which he always applied in the many hysterectomy operations he performed before and after this one. Part of his standard procedure, so the appellant testified, was firstly to ensure the separation of the bladder from the uterus before the operation starts and to protect it from harm. Under cross-examination he confirmed that he had heard of the remedial operation by Dr Dada about six weeks after the hysterectomy and that he then realised that the fistula which necessitated the operation probably resulted from something that happened during the hysterectomy. At that time he therefore gave thought to what he could have done differently from what he routinely did during the literally hundreds of other hysterectomy operations that he had performed, but nothing came to his mind.

[10] He further said under cross-examination that Prof Green-Thompson could be correct in his thesis that the fistula resulted from damage to the blood supply to the bladder during the operation and that this could have happened either when he separated the bladder from the uterus, when he removed the uterus from the vaginal vault, or even when he stopped the blood oozing from the vault after the removal of the uterus by putting in stitches. He also conceded the feasibility of the proposition raised for the first time by the respondent’s counsel in cross-examination, namely, that the fistula could have resulted from infection which was in turn brought about by blood collecting in the area where the oozing occurred and which he did not remove. With reference to this possibility, it was not suggested to the appellant that his failure to remove the accumulated blood would constitute negligence on his part.

[11] Further statements by the appellant which ultimately weighed heavily with the court a quo in concluding that he was negligent, emanated from questions by Jappie J himself. These appear from the following quotation of the evidence in the judgment of the court a quo:

‘You can’t recall as what you did during the operation in regard to the bladder wall except to say you would have performed and used your standard technique? – Yes. Would that be fair to you?

So if the bladder and we must accept because it is common cause that the bladder was compromised, you can only speculate as to how that may well have happened? – Yes.

You can’t take it any further than that? – No I could not. And if it is suggested to you that it should not have happened, what would your response be? – . . . [M]y response would be it should not have happened, I would agree with that. It should not have happened.

But we know it did happen? – . . . Yes.’

[12] In his testimony Prof Snyman underscored the fact that no one really knew how the fistula came about and that all the theories advanced, including the one by Prof Green-Thompson, amounted to no more than speculation. Although he did not deny that Prof Green-Thompson’s thesis could be valid, his view was that there were other hypotheses that would have an equal claim to validity. Amongst these he seemed to prefer the possibility of a small cut in the bladder wall which occurred when the bladder was separated from the uterus, particularly because of the potential tetherings or adhesions resulting from the respondent’s previous caesarean section. The injury thus caused, he explained, could have been so small that it was not noticeable during the operation. Yet, over time and especially in the light of the respondent’s diabetes and the chronic pelvic infection from which she suffered, the small injury could have developed into the fistula. Another possibility that he recognised was the one suggested to the appellant in cross-examination, to wit, that the fistula resulted from an infection in the vaginal vault which was in turn caused by accumulated blood in the area of the oozing which had not been removed. With regard to the latter possibility Prof Snyman stated, however, that failure to remove the accumulated blood would not constitute negligence per se because ‘[t]here is good evidence that if you do not remove blood after an operation . . . it does not cause any problems . . . [because] the body absorbs [the accumulated blood]’.

[13] What Prof Snyman also emphasized was that the occurrence of a fistula does not in itself justify an inference – as Prof Green-Thompson seemed to suggest – of negligence on the part of the surgeon. On the contrary, he said, the development of a fistula is widely recognised as a consequence of hysterectomy operations that cannot always be avoided, however careful the surgeon might be and whatever precautionary measures he or she may take. In this regard he referred to international medical journals that were placed before the court a quo, which supported his opinion. In one of these it is stated, for example, that:

‘[Even if] the operation . . . [is] performed to an adequate standard . . . it would be generally agreed that inadvertent bladder injury is a recognised complication of hysterectomy and can occur even with careful surgical technique.’

[14] I have said at the beginning that the outcome of the dispute as to whether or not the appellant’s performance of the surgery, which led to the respondent’s injury, could be described as negligent, ultimately turns on an election between the opposing views of two expert witnesses. It is true, of course, as the court a quo accentuated in its judgment, that the determination of negligence ultimately rests with the court and not with expert witnesses. Yet that determination is bound to be informed by the opinions of experts in the field which are often in conflict, as has happened in this case. In that event the court’s determination must depend on an analysis of the cogency of the underlying reasoning which led the experts to their conflicting opinions.

[15] That analysis, so it seems to me, was never undertaken by the court a quo in this case. As I understand its judgment, the court was swayed in favour of the respondent’s expert by what it regarded as an admission of negligence on his part by the appellant himself. As appears from the quotation from the appellant’s evidence by the court a quo, all he admitted, however, was that the injury to the respondent’s bladder should not have happened unless something went wrong during the hysterectomy. As I see it, that does not amount to an admission of negligence. After all, as Lord Denning MR observed in Hucks v Cole [1968] 118 New LJ at 469:

‘With the best will in the world things sometimes went amiss in surgical operations or medical treatment. A doctor was not to be held negligent simply because something went wrong.’

Or as Scott J said in Castell v De Greef 1993 (3) SA 501 (C) at 512A-B:

‘The test remains always whether the practitioner exercised reasonable skill and care or, in other words, whether or not his conduct fell below the standard of a reasonably competent practitioner in his field. If the “error” is one which a reasonably competent practitioner might have made, it will not amount to negligence.’

[16] Turning to the conflicting views of the respective experts, it appears that Prof Green-Thompson’s underlying reasoning departs from the inference that the injury to the respondent’s bladder would not have occurred if the appellant was not negligent. To me that seems reminiscent of an application of the res ipsa loquitur maxim, which the court a quo quite rightly found inappropriate in this case. I say quite rightly because, as was pointed out in the locus classicus on medical malpractice, ie Van Wyk v Lewis 1924 AD 438 at 462, that maxim could rarely, if ever, find application in cases based on alleged medical negligence. The human body and its reaction to surgical intervention is far too complex for it to be said that because there was a complication, the surgeon must have been negligent in some respect. Logic dictates that there is even less room for application of the maxim in a case like this, where it has not even been established what went wrong; where the views of experts are all based on speculation – giving rise to various, but equally feasible possibilities – as to what might have occurred. Moreover, although Prof Green-Thompson did not deny the authority of the international publications, put to him in cross-examination, that contradict his conclusions, he simply seemed to ignore their content. In sum I thus find Prof Green-Thompson’s opinion ill-supported by his reasoning.

[17] By contrast, Prof Snyman’s opinion, based on his expertise and experience in practice, that this type of injury may result from a hysterectomy operation despite reasonable care on the part of the surgeon, appears to be well-supported by views expressed in international journals in the field. In fact, these publications seem to indicate that this type of injury to the bladder is generally accepted as one of the inherent risks of a hysterectomy operation. In these circumstances, I consider that the court a quo erred in finding that negligence on the part of the appellant had been established.

[18] In the result:

1 The appeal is upheld with costs.

2 The order of the court a quo is set aside and replaced with the following:

‘The plaintiff’s claim is dismissed with costs.’

____________________

F D J BRAND

JUDGE OF APPEAL

APPEARANCES:

For Appellant: J Marais SC

Instructed by: MacRobert Inc

DURBAN

Correspondents: Claude Reid

BLOEMFONTEIN

For Respondent: F M Moola SC

Instructed by: Sarawan & Co

PIETERMARITZBURG

Correspondents: Honey Attorneys

BLOEMFONTEIN