FACV No. 11 of 2008

IpsofactoJ.com: International Cases [2009] Part 3 Case 8 [CFA]



Dr C.K. Chan

- vs -

Dr Frank Y.K. Yu






2 MARCH 2009


Justice Bokhary PJ and Justice Chan PJ

  1. This patient went into the operating theatre for surgery to his lower body. Following the operation, he was found to have sustained severe nerve injury to his left arm. There has never been any suggestion that this injury existed before he went into the operating theatre. When sued, the anesthetist suggested that the nerve injury to the patient’s arm occurred after the operation while the patient was conscious and recovering outside the operating theatre. But the trial judge (Suffiad J) found that the nerve injury was sustained while the patient was under anesthetic in the operating theatre. So the nerve injury was one from which the anesthetist was duty-bound to protect the patient as far as all reasonable skill and care could do so. There is an obvious point to be noted in regard to the anesthetist’s stance, exploded by the trial judge’s finding, that the nerve injury was sustained at a time when the patient had ceased to be under his care. Such a stance was hardly the most promising basis for seeking a specific finding that he had taken all reasonable steps to protect the patient while the patient was under his care. And the anesthetist has not obtained any such finding.

  2. What the trial judge did find was that the patient’s nerve injury was sustained in consequence of external compression on it while the patient was unconscious in the operating theatre under the care of the anesthetist. Here was a patient whose left arm was healthy when he came under the anesthetist’s care and was severely injured while he was under such care. It was not an injury inflicted by the surgeon. This is a fact to be mentioned since an anesthetist’s duty to protect the patient from injury during an operation extends to all injuries other than those inflicted by the surgeon. How in these circumstances can it be disputed with any conviction that what befell the patient is, as a starting point of inquiry and subject to any explanation that the anesthetist may offer, to be viewed as more consistent with some negligence on his part than the absence of any such negligence? There was some debate at the hearing before us on whether the external compression was considerable and sustained. The short answer, the only answer available and the only answer necessary is that the compression was so considerable and sustained as to cause this severe injury to the root of an underlying nerve, thus bringing about dreadful physical consequences. Quite plainly, the fact of the injury is more consistent than not with negligence.

  3. How then did it come about that the patient failed at the trial so that it became necessary for the Court of Appeal (Rogers VP, Le Pichon JA and Lunn J) to step in and reverse the trial judge’s dismissal of the patient’s claim and enter judgment for him? The answer to that question is that the trial judge, on the brink of inferring negligence from his finding of external compression of the nerve during the operation, declined to infer negligence: so declining by reason of the absence of evidence on how the external compression came about coupled with the view, to be gathered from the expert evidence, that nerve injury during an operation is not always preventable.

  4. There are two major difficulties, each fatal, with the trial judge’s approach in that regard. 

    • First of all, the judge having found that the external compression came about while the patient was unconscious and under the anesthetist’s case, how the external compression came about was for the anesthetist to explain. 

    • Secondly, it is nothing to the point that nerve injury during an operation is not always preventable. Of course not. If it were always preventable, the fact of such injury would not merely provide evidence of negligence but would be conclusive of negligence. It would then always be merely a matter of summary judgment for damages to be assessed.

  5. The trial judge (able and experienced as he is) made on this occasion a plain and obvious error. He made it when he refrained from making the finding of negligence rightly to be inferred from the primary facts which he found. Rightly, indeed inevitably, the Court of Appeal reversed his dismissal of the patient’s claim and entered judgment for the patient. For the foregoing reasons and the reasons given in greater detail by Mr Justice Ribeiro PJ, we would dismiss this appeal with the order nisi as to costs which he proposes.

    Justice Ribeiro PJ

  6. This appeal involves a claim for professional negligence against an anaesthetist arising out of a surgical operation which, although successful, resulted in the plaintiff suffering from radial nerve palsy in his left arm. The legal issues centre on the applicability of the maxim res ipsa loquitur to the evidence and findings of the Judge.


    A.1 The operation

  7. In 2001, the plaintiff, Dr Frank Yu Yu Kai, then aged 68, was diagnosed to be suffering from prostate cancer. It was decided that a radical prostatectomy (removal of the prostate gland) should be performed. Surgery was scheduled for 26 May 2001 at St Teresa’s Hospital to be carried out by Dr Andrew Chan Yau Tong and Dr Richard Lo, both urologists, with the defendant, Dr Chan Chi Keung, as the attending anaesthetist. At the same time as the prostatectomy, it was intended to perform a bilateral hernia repair and also conveniently to excise a bladder diverticulum (a bulge extending out of the bladder). In the event, only the first two procedures were carried out, the diverticulectomy having been deemed unnecessary.

  8. The plaintiff was himself a surgeon, having practised general surgery since 1958. The defendant was Chief of Anaesthesiology at the Adventist Hospital and had started practising in Hong Kong in 1982. He had known the defendant personally for many years.

  9. The operation took place as scheduled. The plaintiff was laid on the operating table in a supine position and his arms abducted from his body supported by cushioned arm boards at an angle of 80o to 90o , with his palms facing upwards and his forearms secured to the arm boards by adhesive tapes. The defendant administered to the plaintiff an epidural anaesthetic followed by a general anaesthetic. The plaintiff’s blood pressure was monitored using an automated blood pressure cuff, referred to at the trial as a Non-Invasive Blood Pressure cuff (or an “NIBP cuff”), attached to his left upper arm. This piece of equipment was to receive much attention at the trial. It was a major operation which lasted some 6 hours, ending at about 7.00 pm.

  10. It appeared to have gone smoothly and, as the defendant’s expert report pointed out, the anaesthesia record showed that the plaintiff’s vital signs had remained normal throughout the procedure with no untoward events. Having been revived from the anaesthesia, the plaintiff was taken to recover in a hospital room under the care of nurses. However, at about 5.30 am the next morning, the plaintiff discovered that he could not move his left hand. When he visited the plaintiff that day, the defendant noticed that the plaintiff had a left wrist drop and suspected radial nerve palsy. He prescribed a splint and physiotherapy.

    A.2 The injury

  11. The following day, 28 May 2001, the plaintiff was examined by Dr Vincent Ma, a neurologist, who found that there was no active extension movement in his left wrist and in all the fingers of his left hand, together with loss of touch sensation in the region between his left thumb and left index finger. Importantly, he found “local tenderness of left Radial nerve at the spiral groove in the posterior aspect of the mid-shaft of the left humerus.” He concluded:

    This is a typical case of left radial nerve palsy (Saturday night palsy). External compression is usually the cause. Most of the cases would have full recovery within 6 weeks.

  12. The expert evidence was that the radial nerve contains mixed sensory and motor fibres and arises from the brachial plexus, descending along the inner side of the humerus to about mid-point and then curves obliquely around and behind that upper arm bone, occupying a shallow depression known as the musculo-spiral groove. It supplies the muscles which straighten the elbow and those which dorsiflex the wrist. Thus, injury to the radial nerve characteristically results in “wrist drop”, which is an inability to “cock” the wrist back, as is normally done when attempting to grasp an object. “Wrist drop” is characteristic of radial nerve palsy where the specific lesion is in that part of the nerve which occupies the musculo-spiral groove.

  13. Dr Ma’s statement that external compression is usually the cause of radial nerve palsy is reflected in the condition’s nickname of “Saturday night palsy”. As the trial judge, Suffiad J, noted when commenting on similar evidence given by Prof Holland, the plaintiff’s expert – HCPI 230/2004, 8 November 2006 at §69:

    Professor Holland’s evidence is quite straightforward, and is simply that radial nerve palsy is commonly caused by extensive and prolonged compression of the radial nerve usually at the spiral groove of the humerus in the upper arm. Because of that and the fact that such injuries very often occur to persons who fall asleep on a sofa or armchair as a result of intoxicating alcoholic drink, and in those circumstances suffer prolonged compression of the radial nerve at the spiral groove because of their arms being extended over the arms of the sofa or armchair, such injury carries the common name of ‘Saturday night palsy’.

  14. In a later report prepared by Dr Ma on 15 November 2001, he explained the usual mechanism of the injury and related it to the anatomical position of the radial nerve:

    The radial nerve is most vulnerable for injury as it travels from the posterior compartment to the anterior compartment of the arm at the spiral groove of the shaft of the humerus. The spiral groove is situated at the upper 1/3 of the posterior aspect of the humerus. At this point the radial nerve is much [closer] to the bone. When the arm is extended at the elbow with the hand in supine position, the radial nerve may be compressed between the bone and hard surfaces such as the edge of the operating table or the border of the extended board of the operating table especially in thin subjects. When the arm is in this improper position for over 2 hours radial nerve palsy is usually occurred.

  15. Dr Ma’s early diagnosis and conclusion that the site of the lesion was at the musculo-spiral groove were confirmed by a nerve conduction study and electromyography report dated 10 July 2001 prepared by Dr Arthur T T Lee, a specialist in neurology. Dr Lee reported that the injury was severe:

    .... this patient has severe left Radial nerve damage at arm level distal to branches supplying the left triceps muscle. There was conduction block as well as significant axonal loss. The prognosis is guarded and it may take more than four to six months for some recovery to occur.

  16. On the same day, the plaintiff was examined again by Dr Ma who shared Dr Lee’s view of the severity of the injury, noting that there had been no improvement after three weeks of physiotherapy. He added:

    Prolonged external compression is most likely the cause of the damage of the left radial nerve.

  17. Another specialist in neurology, Dr Bell Tse Sing Sang, opined in his medical certificate dated 23 July 2001, that the plaintiff’s condition “was caused by external compression during the course of the surgery.” The same view was taken by an anaesthetist, Dr Karl Kai Shu, who certified on 24 July 2001, as follows:

    In view of the history and nature of the lesion, the most likely cause in my opinion is compression trauma to the [left distal radial nerve] sustained during the course of the recent operation.

  18. Dr Fali Shroff, a neurosurgeon who conducted neurological examinations on the plaintiff several times between September 2001 and September 2002 and who gave evidence at the trial, was in no doubt as to the cause and severity of the injury. His testimony was as follows:


    .... And did you have any doubt as to what caused that injury to the radial nerve?


    I have no doubt whatsoever.


    And what would that have been?


    It’s what is normally known as Saturday Night Palsy, and I’ve seen several cases of that.


    Yes, and can you describe that, what the nature of the injury to the ....


    The nature is compression against the bone on one side and against a hard object on the other side, and that is compression of the radial nerve.


    And having regard to the extent of the disability that you saw, .... could you come to a view as to the type of compression required to cause that?


    It would need to be a hard object. Otherwise .... there can be no compression. It needs to be a hard object and it needs to be prolonged.

  19. In Dr Shroff’s report of 27 September 2001, he noted that while in general one could expect improvement over time, “this is likely to be minimal in view of the results of the nerve conduction and EMG tests.”

  20. However, on 30 May 2002, he noted significant improvement had in fact occurred, enabling the plaintiff to perform light surgery. And on 25 September 2002, he recorded that the plaintiff stated that he no longer suffered from any symptoms and felt physically and mentally capable of performing major surgery.

  21. As Dr Ma had indicated, most radial nerve palsy symptoms clear up within about 6 weeks. The injury in the present case was plainly severe in that it took a year from the date of the operation before the plaintiff could take up light surgery and some 16 months before the plaintiff could be considered to have recovered fully.


    B.1 The plaintiff’s pleaded case

  22. It was not in dispute that the defendant owed to the plaintiff a duty of care in contract or in tort to exercise reasonable skill and care in managing the anaesthesia and protecting the plaintiff from harm (other than any harm that the surgeon might inflict) while he was in an unconscious state in the course of the operation.

  23. The plaintiff’s case was that his left radial nerve palsy was caused by prolonged and excessive localized external compression during the operation and that this would not have occurred in the ordinary course of events if the requisite skill and care had been exercised by the defendant so that negligence should be inferred.

  24. He went on to plead that “further or in the alternative” the defendant had been positively negligent in three possible ways:

    1. By permitting the plaintiff’s arm to fall off the arm board and so to be subjected to prolonged compression causing the injury;

    2. By incorrectly applying the NIBP cuff, so that it had a kink, wrinkle, fold or defect bringing injurious pressure to bear on the nerve during the long operation; and/or,

    3. By using the NIBP cuff to monitor his blood pressure at all “instead of using a cannula in the radial artery to measure arterial blood pressure directly.”

    B.2 The NIBP cuff

  25. A NIBP cuff enables arterial blood pressure to be measured indirectly by a microprocessor detecting fluctuations in pressure which occur in the cuff when it is inflated then slowly deflated. The cycles of inflation and deflation are automated and in the present case, were programmed to occur every 5 minutes. Thus, every 5 minutes, the cuff would take about 10 seconds to inflate to a pressure higher than the patient’s systolic arterial pressure and then deflate over about 30 seconds, giving a reading on the machine. It follows that over a 6 hour operation, the cuff will have inflated and deflated about 70 times. It is standard practice, as adopted in the present case, to line the cuff with a soft protective bandage known as a “Velband”.

  26. Accordingly, the plaintiff’s alternative positive case in relation to use of the NIBP cuff had two aspects. First, the court was invited to find that the compression causing the injury came about due to the improper use of the cuff or the Velband lining; and secondly, that even proper use of the cuff was a possible source of the pressure causing the injury.

    B.3 The defendant’s pleaded case

  27. The defendant denied that any injurious compression had occurred in the operating theatre. He suggested that the injury might just as well have occurred “during the plaintiff’s sleep between 1.00 am and 5.30 am on 27 May 2001” as a result of “fluid shift, external compression and/or stretching .... in isolation or in any combination”. In other words, the defendant was denying that pressure alone was the cause and, if it was, that it had not occurred during his watch.

  28. He asserted that he had used all reasonable care, and indeed, that since he knew the patient personally, he had exercised special care during the operation. He denied that the plaintiff’s arm had been allowed to slip off the arm board or that the NIBP cuff had been improperly applied, saying that he had regularly checked various pressure points, including the plaintiff’s upper arms without finding any compression. The defendant also asserted that use of the NIBP cuff was usual and accepted practice and posed no additional risk of radial nerve palsy.


    C.1 The trial and the Judge’s findings

  29. The trial took place, as indicated above, before Suffiad J. The plaintiff and the defendant both gave evidence. The plaintiff called as his expert witnesses, first, Prof Ross Holland, an anaesthetist and formerly Professor of Anaesthesia in Hong Kong and at Newcastle in Australia; and secondly, Dr Fali Shroff, a neurosurgeon. The defendant’s expert witness was Dr John Matthew Low, an anaesthetist. Apart from the medical reports referred to above, Suffiad J also heard evidence from some of the nurses and had witness statements from others involved. Articles from the medical literature concerning perioperative nerve injuries were also placed before the court and discussed in some detail by the experts. The complex trial was conducted with great ability by counsel on both sides, Mr Mohan Bharwaney appearing for the plaintiff and Mr Russell Coleman appearing for the defendant, both having since been elevated to the rank of Senior Counsel.

  30. Suffiad J rejected the theory that the plaintiff’s arm had been allowed to slip off the arm board ( HCPI 230/2004, 8 November 2006 at §§31-32); as well as the theory that the NIBP cuff had been improperly applied (ibid §49). He also found that use of such a cuff was not in itself negligent (ibid §43).

  31. His Lordship expressly made the following positive findings which are of central importance in the present appeal :

    1. The plaintiff was suffering from radial nerve palsy, as conceded by the defendant: ibid §79.

    2. The left radial nerve injury to the plaintiff came about due to external compression at the spiral groove at the posterior aspect of the plaintiff’s left upper arm: ibid §82.

    3. The finding as to the cause of the injury was justified by the unchallenged evidence of Dr Ma that he had elicited tenderness at that site (ibid §80) and that this was a typical case of Saturday night palsy for which external compression is the usual cause (ibid §81).

    4. The external compression which caused the left radial nerve palsy took place during the period when the plaintiff was under anaesthetic in the operating theatre and not, as suggested by the defendant, when he was recovering in his hospital room after the operation: ibid §83 and §84.

  32. In addition, I think it clear that Suffiad J also implicitly found that the external compression was prolonged and continuous or unalleviated.

    1. Thus, his Lordship accepted the uncontradicted evidence of Dr Vincent Ma who, as noted in Section A2 of this judgment, stated in his report dated 15 November 2001 that such an injury is likely to have involved the nerve being compressed between the bone and a hard surface for over 2 hours.

    2. Secondly, the reasons given by the Judge for rejecting the defence thesis that the compression may have occurred post-operatively, were first, that even if sleepy or drowsy during that time, the plaintiff would probably “have changed position to alleviate or remove such external compression, instead of letting it culminate” in the radial nerve palsy (ibid §87); and secondly, because the nurse taking care of him post-operatively had given evidence of manually taking his blood pressure using a cuff on his left arm making it “unlikely that the plaintiff’s left arm would have been subject to prolonged external compression when he was in his hospital room that night”. The words in quotations demonstrate that the Judge accepted that the palsy occurred as a result of prolonged and unalleviated pressure.

  33. Having made the foregoing findings, Suffiad J felt he could take his investigation no further, stating (ibid §90): “.... there is in effect no evidence, and therefore no finding, as to how that external compression came about”.

  34. He posed for himself the crucial question in the following terms:

    Given these findings and the circumstances of it, the question that I need to ask is whether this was such an event which in the ordinary course of things was more likely than not to have been caused by the negligence of the defendant as the anaesthetist having charge of the plaintiff during the course of the operation itself. In other words, am I able to infer negligence on the part of the defendant by reason only that such an injury was caused to the plaintiff.

  35. His Lordship concluded that he could not draw the inference for the following reason:

    Having regard to the medical literature put before me, in particular those medical literature indicating that while it was previously thought that all perioperative nerve injuries were preventable, but that is no longer the general view today, I am unable to infer negligence, given the facts of this case, on the part of the defendant, even though I was able to come to the finding that the external compression causing the left radial nerve palsy likely took place at the time when the plaintiff was having the operation under anaesthetic.

  36. The plaintiff’s case both in the Court of Appeal and in this Court is that the learned trial judge adopted the wrong approach. Mr James Badenoch QC who appears with Mr Bharwaney SC for the plaintiff (respondent) against Mr Coleman SC for the defendant (appellant) on this appeal, submits that once such findings were made in the context of the undisputed legal responsibilities of the defendant, a prima facie case had been established requiring him to provide a plausible non-negligent explanation for the pressure-induced injury. That, he argues, the defendant wholly failed to do.

    C.2 The Court of Appeal

  37. In reaching its decision to allow the appeal and to hold the defendant liable, the Court of Appeal ([2007] 4 HKLRD 837, Rogers VP, Le Pichon JA and Lunn J) essentially accepted the foregoing argument as advanced by Mr Bharwaney. The leading judgment was delivered by Le Pichon JA, with whom Lunn J agreed.

  38. Her Ladyship pointed to the findings made by the Judge as to the cause of the injury being external compression applied to the spiral groove when the plaintiff was in the operating theatre (ibid §18), comprising pressure of a duration and intensity sufficient to cause the local tenderness elicited by Dr Ma and sufficient for the injury to result (ibid §25). Those circumstances indicated prima facie a preventable injury since one would as a matter of commonsense expect external compression to be capable of being avoided. Accordingly, she held (ibid §27 and §30) that “.... it fell to the defendant anaesthetist whose responsibility it was to protect the anaesthetised patient to provide a plausible explanation for that condition.”

  39. Le Pichon JA took the view that the Judge’s reason for declining to infer negligence, namely, that the thrust of the medical literature cited indicated that perioperative nerve injuries should no longer be regarded as preventable, did not displace the prima facie case of negligence where the aetiology of the radial nerve palsy had been found by him to be external compression. Her Ladyship concluded (ibid §30) that “as no explanation was forthcoming from the defendant, much less any plausible explanation, negligence has to be inferred.”

  40. Although the decision of the Court of Appeal was unanimous, it should perhaps be regarded in substance as a majority judgment with Rogers VP dissenting. While agreeing with the other members of the Court his Lordship (ibid §§1-4) expressed “no little anxiety” as to whether the decision was correct since he considered there to be much force in the Judge’s reasons for refusing to infer negligence on the findings made. When the matter went back before the Court of Appeal on the plaintiff’s application for leave to appeal, Rogers VP’s doubts had increased, particularly as to whether the doctrine of res ipsa loquitur was applicable in cases of medical negligence like the present. He took the unusual course of dissenting from the majority’s decision not to grant leave.


  41. Use of the phrase “res ipsa loquitur” is sometimes viewed with a degree of disapproval. The classic statement of the evidential rule going back to 1865 is by Erle CJ in Scott v London and St Katherine Docks Co [18] (1865) 3 H & C 596 at p 601; 159 ER 665 at p 667, as follows:

    .... where the thing is shewn to be under the management of the defendant or his servants, and the accident is such as in the ordinary course of things does not happen if those who have the management use proper care, it affords reasonable evidence, in the absence of explanation by the defendants, that the accident arose from want of care.

  42. Res ipsa loquitur is, as Hobhouse LJ pointed out in Ratcliffe v Plymouth and Torbay Health Authority (1998) 42 BMLR 64 at 82, “no more than a convenient Latin phrase used to describe the proof of facts which are sufficient to support an inference that a defendant was negligent and therefore to establish a prima facie case against him.”

  43. Whether one uses the label “res ipsa loquitur” or one speaks (as Hobhouse LJ would have preferred) of establishing a prima facie case, one is concerned with a rule regarding the proper approach to the evidence. It is an approach whereby, in cases where the plaintiff is unable to say exactly how his injury was caused but, consonant with his duty of care, one may expect the defendant to know, one asks whether the evidence has raised a prima facie case against the defendant and if it has, whether the defendant has, at the end of the day, dispelled that prima facie case by providing a plausible explanation for the plaintiff’s injury which is consistent with the absence of negligence on his part.

  44. Thus, in Lloyde v West Midlands Gas Board [1971] 1 WLR 749 at 755, Megaw LJ explained the approach as follows:

    I doubt whether it is right to describe res ipsa loquitur as a ‘doctrine’. I think that it is no more than an exotic, although convenient, phrase to describe what is in essence no more than a common sense approach, not limited by technical rules, to the assessment of the effect of evidence in certain circumstances. It means that a plaintiff prima facie establishes negligence where (i) it is not possible for him to prove precisely what was the relevant act or omission which set in train the events leading to the accident; but (ii) on the evidence as it stands at the relevant time it is more likely than not that the effective cause of the accident was some act or omission of the defendant or of someone for whom the defendant is responsible, which act or omission constitutes a failure to take proper care for the plaintiff's safety. I have used the words ‘evidence as it stands at the relevant time'. I think that this can most conveniently be taken as being at the close of the plaintiff's case. On the assumption that a submission of no case is then made, would the evidence, as it then stands, enable the plaintiff to succeed because, although the precise cause of the accident cannot be established, the proper inference on the balance of probability is that that cause, whatever it may have been, involved a failure by the defendant to take due care for the plaintiff's safety? If so, res ipsa loquitur. If not, the plaintiff fails. Of course, if the defendant does not make a submission of no case, the question still falls to be tested by the same criterion, but evidence for the defendant, given thereafter, may rebut the inference. The res, which previously spoke for itself, may be silenced, or its voice may, on the whole of the evidence, become too weak or muted.

  45. Mr Coleman, relying on a dictum of Stuart-Smith LJ in Delaney v Southmead Health Authority (1992) 26 BMLR 111 at 118, suggested that this approach is not applicable to medical negligence cases. I can see no reason in principle why that should be so, particularly bearing in mind the purpose of the rule which is well-recognized. As Lord Normand stated in Barkway v South Wales Transport Co Ltd [1950] 1 All ER 392 at 399, “its purpose is to enable justice to be done when the facts bearing on causation and on the care exercised by the defendant are at the outset unknown to the plaintiff and are or ought to be within the knowledge of the defendant.”

  46. Although the approach will not be important in medical negligence cases where the issues of causation and negligence are wholly fought out on competing evidence, it seems to me obvious that in a significant number of such cases – particularly where the patient is unconscious when the injury is incurred – the res ipsa loquitur or prima facie case approach will be indispensable. As Hobhouse LJ stated in Ratcliffe v Plymouth and Torbay Health Authority (1998) 42 BMLR 64 at 83:

    Medical negligence cases have the potential to give rise to considerations whether the plaintiff has made out a prima facie case and whether or not the defendant has provided an adequate answer to displace the inference to be drawn from the plaintiff's prima facie case. Further, it is commonplace that the plaintiff will not, himself or herself, have fully known what occurred, particularly if the relevant procedure was an operation carried out under anaesthetic. The procedures were under the control of the defendant and what the defendant did or did not do is exclusively within the direct knowledge of the defendant.

  47. In my view, adoption of the prima facie case or res ipsa loquitur approach would have been wholly apposite in the present case. A prima facie case would have arisen against the defendant if the occurrence of the injury is something which would not have happened in the ordinary course of events without someone’s negligence and if the injury falls within the sphere of the defendant’s responsibility to take due care of the plaintiff.

  48. To dispel such a prima facie case, the defendant would have to point to evidence supporting a plausible explanation consistent with the absence of negligence. As to the meaning of “plausible”, Buxton LJ pointed out that – Lillywhite v University College London Hospitals’ NHS Trust [2005] EWCA Civ 1466 at §89:

    .... mere assertion will not do; but neither need the explanation be shown to be the probable or likeliest answer. An explanation to that modest standard has to be reasonably available on the evidence taken in the round.

  49. Similarly, in his summary of the law, Brooke LJ in Ratcliffe (1998) 42 BMLR 64 at 80 stated:


    The position may then be reached at the close of the plaintiff's case that the judge would be entitled to infer negligence on the defendant's part unless the defendant adduces evidence which discharges this inference.


    This evidence may be to the effect that there is a plausible explanation of what may have happened which does not connote any negligence on the defendant's part. The explanation must be a plausible one and not a theoretically or remotely possible one, but the defendant certainly does not have to prove that his explanation is more likely to be correct than any other. If the plaintiff has no other evidence of negligence to rely on, his claim will then fail.

  50. I would add that in the context of an appeal, the plausibility of the proffered explanation must be assessed in the light of the Judge’s unchallenged findings and in the light of what the evidence as a whole may fairly be taken to have established.


  51. In examining the Judge’s findings and the evidence to determine whether a prima facie case of negligence was made out by the plaintiff in the present case, it is important to distinguish between the cause of the injury and the cause of the compression which caused the injury.

    E.1 The cause of the injury

  52. In the present case, there is no mystery at all about what caused the injury. It was prolonged, localized pressure which compressed the radial nerve at the musculo-spiral groove. In the first place, the objective evidence was to this effect. Dr Ma found tenderness at the site two days after the operation, a finding confirmed by the nerve conduction study and the electromyography. The incidence of this kind of injury as a result of external compression is explained by the anatomical vulnerability of the radial nerve.

  53. Secondly, as set out in Section A.2 above, all the doctors who examined the plaintiff concluded without hesitation that this was a classic case of “Saturday night palsy”, that is, radial nerve palsy caused by prolonged external pressure.

  54. Thirdly, and most importantly, the Judge unequivocally found that on the evidence, the injury was due to external compression at the spiral groove at the posterior aspect of the plaintiff’s left upper arm. He found that it was a typical case of Saturday night palsy for which external compression is the usual cause.

  55. It follows that Dr Low’s attempts to question whether compression was indeed the cause were rejected. It is however relevant to the arguments considered below, to examine the nature of Dr Low’s evidence on this matter.

  56. Dr Low sought to question external compression as the cause on the basis of his review of certain medical articles on perioperative nerve injuries. His thesis was:

    1. That the previously held view that all perioperative nerve injuries were preventable no longer holds good.

    2. Quoting from one article by Dr Robert Morell (Current Reviews in Clinical Anesthesia, Lesson 23, Volume 21), that:

      Evidence is rapidly mounting that indicates that mechanisms of perioperative nerve injury are complex, multi-factorial and incompletely understood.

    3. That, on this basis, one has to be “agnostic” about the causes of radial nerve palsy. He puts it thus in his testimony:

      .... from the whole body of literature, we know that the mechanism of neuropathy includes compression, stretching, ischemia, direct trauma, neuritis and metabolic causes, and toxic causes. In .... 10 to 15% of reported neuropathies – peripheral neuropathies in the perioperative period, any other cause. In 85 to 90% the world body of literature says the only position we can take is to be agnostic. We do not know what actually caused the problem.

  57. Both at the trial and before this Court, there was a debate about what the medical literature cited was actually saying.

    1. Mr Badenoch argued with much force that the areas of recognized causal uncertainty arise only in relation to ulnar nerve and not radial nerve injuries. On this, he had the support of the evidence of Prof Holland and Dr Shroff.

    2. Dr Morrell’s article did indeed speak of perioperative nerve injuries being “complex, multi-factorial and incompletely understood”. But, as an examination of the article shows, that was in the context of ulnar nerve injuries. When it came to the radial nerve, he merely pointed out that “Prolonged pressure on the radial nerve as it courses in the spiral groove of the humerus” may cause such injury, without suggesting that any uncertainty arises.

    3. In an article by Dr Mark Warner (“Perioperative Neuropathies”, Mayo Clinic Proceedings, Vol 73(6), June 1998, 567-574), the summary states in general terms that “the cause of perioperative neuropathies is often multifactorial”. However, examination of that article reveals that it concentrates on ulnar and brachial plexus neuropathies, mentioning also median neuropathy and carpal tunnel syndrome. There is only one mention of the radial nerve in a diagram without any discussion.

    4. Mr Coleman argued that the rarity of radial nerve injuries accounts for the absence of their specific treatment in the literature but that we should read the general statements as to uncertain causes as extending to that neuropathy.

  58. It is unnecessary to go further into this debate. There can be no dispute that external compression is generally recognized, in the literature as in the evidence adduced, as the usual cause of radial nerve palsy arising from a lesion in the nerve at the musculo-spiral groove. But even if there may exist a fringe area of uncertainty as to whether other possible causes might exist, the Judge was plainly right to find, on the evidence, that external compression was the cause in the present case. Against Dr Low’s agnosticism, the evidence of the specialist neurologists was that the plaintiff’s neuropathy was unmistakeably caused by external compression. Indeed, Dr Low acknowledged that he is not a neurologist and expressly accepted Dr Ma’s report. It is not surprising that the Judge did not accept Dr Low’s thesis and was able to find definitively that the cause in the present case was indeed external compression.

    E.2 The cause of the compression

  59. The true mystery here is as to how the prolonged external compression came to be applied to the patient in the course of the operation. Since the Judge had rejected the suggestion that the plaintiff’s arm had been allowed to slip off the arm board, or that the NIBP cuff had been improperly applied, or that its use per se caused the injury, what was the source of the external compression which brought about the injury? That is a question about what actually happened in the operating theatre. How did the nerve get pinched between a hard object or surface and the bone at the musculo-spiral groove for a prolonged period during the operation? That is not a question to be resolved by reference to the medical literature. The cause of the pinching is unknown, but it is not unknown because of the current incomplete state of medical science or because of unknown multifactorial causes. It is unknown because the hard object or surface causing the compression was not identified in the factual investigation conducted at the trial.

  60. In my view, with respect, it is in this context that the trial Judge went wrong. As we have seen (in Section C.1 above), when he found himself unable to make a direct finding on the cause of the compression, he asked himself the following question:

    Given these findings and the circumstances of it, the question that I need to ask is whether this was such an event which in the ordinary course of things was more likely than not to have been caused by the negligence of the defendant as the anaesthetist having charge of the plaintiff during the course of the operation itself. In other words, am I able to infer negligence on the part of the defendant by reason only that such an injury was caused to the plaintiff.

  61. He answered his own question in the following terms (Suffiad J at §94):

    Having regard to the medical literature put before me, in particular those medical literature indicating that while it was previously thought that all perioperative nerve injuries were preventable, but that is no longer the general view today, I am unable to infer negligence, given the facts of this case, on the part of the defendant, even though I was able to come to the finding that the external compression causing the left radial nerve palsy likely took place at the time when the plaintiff was having the operation under anaesthetic.

    This led the Judge to conclude that plaintiff had failed to prove his case.

  62. It is unclear whether his Lordship was adopting a res ipsa loquitur approach in reaching his conclusion. In any event, with respect, assuming that he was, the answer which Suffiad J gave was, in my view, the wrong answer. Having identified external compression at the spiral groove accruing during the operation as the cause of the radial nerve palsy, his Lordship erroneously reverted to the debate grounded in the medical literature concerning unknown, multifactorial (and thus unpreventable) causes of certain perioperative nerve injuries. That was a debate about the cause of the injury, not on the particular cause or causes of the compression in the present case. The uncertainties discussed in the literature are irrelevant to the present injury since its aetiology was found to have been clearly established. The debated uncertainties therefore provide no basis for saying that the compression was not preventable and no basis for finding either that there was no prima facie case or that negligence had not been established.

    E.3 A prima facie case?

  63. It therefore falls to the Court to consider whether on the evidence, a prima facie case of negligence against the defendant was disclosed and, if so, whether a plausible non-negligent explanation has been provided.

  64. On the facts found by the Judge, it is my view that he should unhesitatingly have found that there was a prima facie case of negligence which the defendant was required to answer.

  65. It was a particularly serious case of radial nerve palsy whose effects were felt for some 16 months instead of the usual 6 weeks or so. This indicated that there had been a serious injury to the plaintiff’s left radial nerve. It had resulted from pressure exerted by a hard object or hard surface on the plaintiff’s upper arm at the site of the injury for a prolonged period. The pressure had been of an intensity that enabled tenderness to be elicited at the site two days after the operation. The injury-causing compression had occurred during the operation when the plaintiff was in an unconscious state under anaesthesia managed by the defendant. Perioperative compression injuries to the radial nerve are a well-known hazard and measures have been designed to avoid such injuries involving the proper positioning of the patient and cushioning of pressure points, and so forth. While it was not possible for the plaintiff to prove precisely what was the relevant act or omission on the defendant’s part which resulted in the application of the injurious pressure on his upper arm, he was quite entitled to say on the evidence that his injury was an untoward outcome which would not have occurred in the ordinary course without a negligent failure to exercise due care on the part of some person. In the present case, the person prima facie responsible was the defendant since he had, as the attending anaesthetist, a duty to protect the unconscious plaintiff from harm, including harm from the known hazard of compression trauma to the radial nerve. These are circumstances which, in my view, clearly raised a prima facie inference of negligence.


  66. Only two explanations for the pressure exerted on the plaintiff’s arm have been identified as candidates for providing a plausible non-negligent explanation of the injury.

    1. The first is that pressure from the NIBP cuff itself, properly applied without any wrinkle, fold or defect, plausibly resulted in the injury-causing compression.

    2. The second is that relatively light pressure from the NIBP cuff, operating on the nerve which may have been weakened and made more susceptible to injury in the course of the major operation, plausibly resulted in the neuropathy.

    F.1 The cuff alone as a plausible cause

  67. The immediate difficulty faced by the defendant in advancing this argument is that it is, in my view, incompatible with the findings made by the Judge. One of the theories of negligence advanced by the plaintiff at the trial was that it had been negligent to use the NIBP cuff instead of an arterial cannula even if the cuff had been properly applied. The plaintiff’s case at that stage was therefore that a properly applied cuff was a foreseeable cause of the injury resulting from external compression.

  68. When the Judge found that the injury was caused by external pressure but then went on to hold that there was no evidence and no finding as to the cause of that compression, he must have meant that he did not accept that a properly applied cuff could be the relevant cause of the compression. If he had thought otherwise, he would not have said that there was no evidence of the cause of compression and he would have had to go on to consider whether compression injury caused by the cuff was foreseeable so that its use was negligent. He did not do so, but stopped at his finding that there was no discernible cause of the compression.

  69. The second problem the defendant faces in putting forward this argument is that, ironically, in repelling the plaintiff’s aforementioned thesis, he provided powerful reasons for saying that the “cuff alone” explanation is highly implausible.

  70. Having stated, in the course of his testimony, that prolonged pressure on any nerve would cause it injury, Dr Low gave examples of some possible sources of the pressure, such as an “ether screen” which is “a hard metal rod which supports the sterile drapes” and the arm boards extending out of the operating table. He then dismissed the blood pressure cuff as a source of the pressure:


    But a blood pressure cuff is very different. It’s soft. It’s intermittent. It’s above the arterial pressure only for a few seconds before it deflates and it’s .... it is insignificant pressure.

    He stressed that here one is dealing with “a soft intermittent inflation of the blood pressure cuff protected by Velband” which is not a hard object. And he stood by the view stated in his expert’s report that while a NIBP cuff cannot be absolutely excluded as a cause, it is a most unlikely candidate. In other words, the defendant’s own expert was of the view that the cuff alone was not a plausible candidate but, to echo the words of Brooke LJ (see Section D above) “a theoretically or remotely possible one”.

  71. Mr Coleman pointed to two articles in the medical literature from which he sought support for the plausibility of treating the cuff alone as the cause of the injury. The first was an article, translated from the German which had been discussed with the experts but which was not before the Court. It appears that it was a case involving extremely special circumstances where a woman suffered radial nerve injury after a 30 minute operation. However, even though Prof Holland had supported in other respects the plaintiff’s theory that use of the cuff itself had been negligent, he did not consider this article to lend any support to that view:

    .... it’s a very, very special situation, there’s an emaciated woman with severe malignant disease and she is very sick. And .... we know that that wasn’t the case with Dr Yu [the plaintiff] .... I mean she’s so sick that it didn’t take very much at all to injure her radial nerve ....

  72. The second article requires a little more attention. This was a Practice Advisory issued by the American Society of Anesthesiologists in the year 2000 containing the views of a task force on Prevention of Perioperative Peripheral Neuropathies. The task force consisted of 10 members who

    1. reviewed the published research;

    2. got consultants to participate in surveys and to comment on the initial draft report;

    3. surveyed a random sample of active ASA members;

    4. held an open forum on their draft advisory at an ASA convention; and

    5. used “all available information .... to build consensus within the task force”.

  73. The conclusion reached by the task force is in fact directly against accepting the “cuff alone” explanation as plausible. It reported:

    The task force consensus is that the use of properly functioning automated blood pressure cuffs on the arm (ie, placed above the antecubital fossa) does not change the risk of upper extremity neuropathy. Public commentary from an open forum and from Internet correspondence corroborates the task force opinion.

  74. However, the defendant seeks to derive support for the plausibility of his “cuff alone” theory from the somewhat curious survey results obtained by the task force. It seems that the survey respondents were asked whether they agreed with the statement that “use of an automated blood pressure cuff on the arm may increase the risk of radial neuropathy”. Of the 83 consultants who responded, 17 disagreed, 32 agreed and 41 said they were uncertain. Of the 428 ASA members responding, 133 disagreed, 128 agreed and 167 were uncertain.

  75. I do not consider that any weight can be given to the raw survey data which appear to be inconclusive for reasons that are not apparent. The task force was able to reach consensus among themselves that blood pressure cuffs, properly used, pose no increased risk of radial neuropathy. They did so after having reviewed the survey results as well as the published research and other opinion canvassed at its open forum and in correspondence. It was obviously in a far better position than this Court to interpret the data received and to decide on the eventual content of its advisory bulletin.

  76. As the Judge found (§40), the use of an automated NIBP cuff is common practice, at least in Hong Kong if not elsewhere, and its use “cannot be said, by any yardstick, to be outside of the normal and ordinary practice of an anaesthetist”. If it were plausible to suggest that use of the cuff in itself could cause radial neuropathy in a patient who like the plaintiff, was classified as “ASA 1” [ The American Society of Anesthesiologists’ lowest risk classification: “ASA I: Normal, healthy, no limitations of normal activity.”], one would, as Dr Low, testifying for the defendant pointed out, expect “a huge red warning flag on every blood pressure cuff ....” He noted that: “There is no such warning from the FDA or from the ASA representing 35,000 anaesthetists ....”: Transcript B2/540.

  77. For all the above reasons, I do not consider the “cuff alone” explanation plausible on the Judge’s findings and the evidence.

    F.2 The cuff and internal weaknesses

  78. The second suggestion attempts to meet the weaknesses of the “cuff alone” argument. It advances the theory that relatively slight and intermittent pressure from a NIBP cuff, while normally not sufficient to cause a pressure-induced radial nerve palsy, is plausibly sufficient if it operates on weakened underlying nerves.

  79. This argument proved to be somewhat chimerical. It appears first to have been suggested by the defendant himself in cross-examination as follows (see Transcript B1/279):


    .... do you accept that the injury in this case was caused to Dr Yu by excessive and prolonged pressure at the musculo-spiral groove?


    No, I don’t.


    Do you suggest, as part of your case, that it was caused in some other way?


    It could be pressure but may not be excessive, you don’t need excessive pressure if the nerve is vulnerable to compression.

  80. However, on further probing, it became clear that this was a mere assertion which rapidly evaporated (see Transcript B1/279-280):


    I just want to clarify from you that you are not suggesting that he suffered wrist drop from disease, for example.


    No, no.


    You are not suggesting that he suffered wrist drop because of trauma from a surgical instrument?




    The classic and most often recorded cause of radial nerve palsy is compression.


    I agree.


    So do you accept that in Dr Yu’s case he suffered his radial nerve palsy because of compression?


    Compression, possibly.


    You do not put forward any other reason for that radial nerve injury other than compression, do you?


    I cannot answer you this question.

  81. The only other suggestion that the pressure came from the cuff operating on internal weaknesses was made by Dr Low harking back to his “agnostic” approach to causes of perioperative peripheral nerve injuries in general. Thus, in cross-examination, he stated (see Transcript B2/528):


    .... if compression was the mechanism and if the blood pressure cuff was the mechanism, then there must be another factor to make that particular nerve so very vulnerable. And I think .... what’s pointed out in all the literature is that even with the highest standard of medical care and attention in the intra-operative period, peripheral nerve damage and injury occurs in the perioperative period and we do not know why . ...

  82. When asked to clarify, he added (ibid):


    My answer was that there may be factors which make a nerve more vulnerable to an ordinary standard of care, like a blood pressure cuff. Very ordinary light pressure.

  83. This evidence amounts to no more than a general speculative assertion. It is not tied to any of the evidence or findings in the present case. Dr Low was merely suggesting that there might exist some causal factors, presently unknown to medical science, which might so weaken the nerve that even the slight intermittent pressure of a blood pressure cuff, lined with a soft protective layer of Velband, might be sufficient to result in compression-induced radial nerve palsy.

  84. There is simply nothing in the evidence to suggest that any of this is applicable to the plaintiff. He was, as previously noted, classified as an ASA I subject for the operation. None of the doctors who examined him before and after the operation suggested that any special risk factors were present. None of the neurologists suggested that the aetiology of his injury was other than prolonged localized compression at the musculo-spiral groove, classically the cause of Saturday night palsy.

  85. There was some discussion in argument, that the nerve injury might have been contributed to by physiological factors known to cause nerve injuries in some cases. But on the evidence, none of these factors were shown to be relevant to the present case.

    1. Thus, in a document prepared by the defendant in March 2003, he recorded having told the plaintiff at a pre-operative meeting that “massive fluid shift” might cause “nerve dysfunction”. In his report, Prof Holland considered “the fluid shift theory to be untenable, and far-fetched to the point of absurdity”. In his oral testimony, Prof Holland stated that “This just didn’t happen” and that the “numbers don’t support that”. Dr Low’s evidence was that he “didn’t quite understand [the defendant’s] statement” and suggested that it be referred back to him.

    2. Similarly, the defendant indicated that he had informed the plaintiff of a risk of nerve dysfunction caused by oedema bringing pressure to bear on the nervous structures. However, in cross-examination, he accepted that there was no scientific basis for suggesting that oedema could cause radial nerve damage and that he was not thinking about the radial nerve when he mentioned this risk. Prof Holland explained that oedema might cause injuries to nerves located within certain anatomical structures, such as the median nerve in the carpal tunnel. He pointed out that “in the case of the radial nerve there’s no tunnel or rigid compartment that would allow an oedema of the radial nerve, assuming it even happened, to produce injury.”

    3. In what was probably a variant of the oedema theory, the possibility was also canvassed that pressure from the cuff might have worked on the nerve internally weakened because of pressure from excessive liquid retained from the fluids transfused during the operation. The plaintiff had requested that blood transfusion should, if possible be avoided. The anaesthesia notes showed that he lost some 1,200 ml of blood and was given some 3,500 ml of fluid replacement in the course of the 6 hour operation. However, taking into account the estimated urine loss during the operation, both experts dismissed the load presented by the net fluid gain as insignificant – Dr Low: Transcript B2/526; Prof Holland: Transcript B2/393.

  86. It is accordingly my view that there is no substance to this second proffered explanation. It is, if anything, even less plausible than the first.


  87. I therefore conclude that the defendant has been unable to provide any plausible explanation to dispel the prima facie case of negligence established by the plaintiff and liability must follow. I would accordingly dismiss the appeal and uphold the judgment of the Court of Appeal and its Order dated 24 October 2007 whereby judgment was entered in favour of the plaintiff. I would make an order nisi that the defendant pay to the plaintiff the costs of this appeal and direct that any submissions as to costs be made in writing and lodged with the Court and served within 14 days from the date of the handing down of this judgment, with any written submissions in reply to be lodged and filed within 14 days thereafter.

    Justice Litton NPJ


  88. I have had the advantage of reading in draft Mr Justice Ribeiro PJ’s judgment. The background facts are well set out in that judgment and need not be repeated here.

  89. This case concerns damage to the radial nerve, caused during a very long lower-abdominal operation. The radial nerve is a bunch of sensory and motor fibres which pass down the arm from the brachial plexus to the hands. It curves obliquely round the humerus, passing through a shallow depression known as the musculo-spiral groove in the humerus. Here the nerve gives off branches and is of course fed by blood vessels. It is, to some extent, protected from external pressure by the nerve sheath and the muscles of the upper arm; but if, for a prolonged period, the nerve is compressed against the bone, radial nerve palsy might ensue.

  90. It is common ground in this case that the radial nerve injury suffered by the plaintiff was severe, and its occurrence as a complication arising from lower abdomen surgery was rare.

  91. In the course of a major operation, many complications might arise due to no fault on anyone’s part. Surgery and its associated processes is not a perfect science. It is well-known, for instance, that prostatectomy could lead to urinary incontinence and erectile dysfunction. That lies in the realm of the surgeons. Here we are considering peripheral nerve damage, which comes within the responsibility of the anaesthetist.


  92. Before the trial began, the judge had before him reports (among others) from (a) Professor R. Holland, and (b) Dr John Matthew Low.

  93. Regarding the cause of the injury, Prof. Holland in his report said that the inflation of the non-invasive blood pressure cuff “NIBP cuff” over a period of many hours “would cause the nerve injury if the cuff had a wrinkle, fold or irregularity directly over the underlying radial nerve where it crosses the humerus.” And he added that during the 6 hours in this case (when the NIBP device was operating every 5 minutes), the cuff would have inflated and deflated a minimum of 70 times and possibly more, since the device would “try again” if its first attempt to measure blood pressure was unsuccessful.

  94. In court, Prof. Holland was asked by counsel whether the cuff would be “a potential cause” of injury if there was no wrinkle, fold or irregularity in the Velband (the protective covering underneath the cuff); his answer was this (see Transcript B2 p.389) :

    It’s been reported that in prolonged surgery the use of what’s called an automatic blood pressure monitor, which is again a non-invasive blood pressure machine, has been followed by radial nerve injury.

    And he added that the longer the patient spent in the operating theatre, the greater the chance of nerve injury.

  95. In the light of Dr Vincent Ma’s finding (made two days after the operation) of localized tenderness at the site of the spiral groove, Prof. Holland in testifying before Suffiad J, was firmly of the view that external pressure was the sole cause of the injury. And what might have been the cause of the external compression? This is what Prof. Holland said (see Transcript B2 p.445) :


    So if the anecdotal evidence is dismissed and there is no finding that the arm fell off the table, we’re left with the much less likely mechanism for the radial palsy having occurred through the blood pressure monitor cuff, aren’t we?


    It was the only alternative that I could envisage.


    The only alternative in the operating theatre.



  96. When the American medical literature, referred to in Dr John Matthew Low’s report, was put to Prof. Holland, he said this (see Transcript B2 p.417):

    Well, I think the non-invasive blood pressure machine has the capability, if you like, of in certain circumstances of causing nerve injury in the upper arm, and the nerve most likely to be injured is the radial.

  97. Prof. Holland was the expert witness called by the plaintiff. There can be no doubt regarding the evidence he gave. It was to this effect : (1) External compression, prolonged and severe, at the spiral groove, caused the nerve injury; (2) the NIBP cuff was possibly the cause of the external compression.

  98. Unless some other “hard object” emerged in the course of the evidence, which might have been capable of pressing the radial nerve against the humerus during the 6-hour operation, then the NIBP cuff was the only possible candidate.

  99. What needs emphasis at this point is this : No expert who testified before Suffiad J (or who had made reports on the plaintiff’s condition without having been called to testify) suggested that the NIBP cuff was incapable of causing the external compression. It is common ground that the operation took a long time. Dr Fali J Shroff, a neurosurgeon, said that the compression had to be made by a “hard object” along the spiral groove, against the bone on the other side, but neither counsel asked him whether the NIBP cuff could be regarded as “a hard object”. When Suffiad J asked him whether the compression had to be prolonged to cause the injury in this case, he said “.... it’s difficult but I would think about an hour to .... anything between one and three hours”. Here the operation took 6 hours. He was never asked whether the compression of an NIBP cuff inflated at 5 minute intervals for six hours might have caused the injury.

  100. In cross-examination, Dr Shroff was asked whether, in this case, the period of recovery was “longer than ordinarily occurs in the situation of a radial nerve injury” his answer was “No, I don’t think so.”

  101. Each time the cuff was inflated automatically, the pressure was always 10% to 20% above the highest systolic pressure and, during the operation, there would have been circumstances when the cuff had to “try again” (to use Prof. Holland’s expression). Hence, it was inflated, in all likelihood, well over 70 times. There is no doubt, therefore, that the cuff was capable of exerting pressure externally against the site of the radial nerve. The debate at trial, as I see it, was whether (i) the cuff alone could have caused the injury or (ii) whether there were other factors involved : Factors which, in combination, might have caused the injury. Here, the evidence of Dr John Matthew Low becomes relevant.

  102. Dr Low in his report, introduced a range of medical literature (coming from the U.S.A.) concerning peripheral nerve damage associated with anaesthesia. He said that the medical profession knew very little about the causative factors of peri-operative neuropathies.

  103. Dealing with the matters in Prof. Holland’s report, Dr Low in his own report said this : Although the repeated inflation / deflation cycles of the NIBP cuff could, in theory, cause pressure-induced damage to the nerves, there were no definite studies indicating “a direct cause and effect, by the use of NIBP cuffs and nerve damage”. He adverted to the manufacturers’ recommendation of “careful positioning of the cuff to avoid the high-risk pressure points, where nerves are known to run a superficial course anatomically”. Having studied the peri-operative records in this case, Dr Low concluded that the management of the plaintiff by the defendant anaesthetist could not have been improved upon in any significant way. His conclusion was that the radial nerve palsy resulted from “prolonged bed rest and relative immobility following multiple major surgical procedures, and not primarily due to the planning, technique or conduct of the anaesthesia”.

  104. His “conclusion” rested, of course, on the judge’s findings on the facts at trial. Suffiad J rejected the theory that the injury occurred post-operatively. There was no appeal against that finding. Hence it drops out of our consideration.

  105. In evidence, Dr Low was careful not to identify any specific cause of the nerve injury in this case. He did not suggest that the American medical literature which he had introduced in his written report comprised the whole of the learning on the subject of peripheral neuropathy.

  106. The general tenor of Dr Low’s evidence was to this effect : During the course of a long and complex operation the patient’s body was subjected to considerable physiological stress; there would inevitably be metabolic changes as a result, for instance, of the infusion of what Prof. Holland referred to as a “cocktail” of fluids. Dr Low referred to instances of nerve dysfunction being caused by a “fairly major imbalance in the sodium and potassium balance of the body”, explaining how that might occur. The infusion of fluids might result in oedema (swelling), ischaemia (areas of bloodlessness) which could put internal pressure on nerve sites. He said (see Transcript B2 p.523) :

    As I keep saying, from the whole body of literature, we know that the mechanism of neuropathy includes compression, stretching, ischemia, direct trauma, neuritis and metabolic causes, and toxic causes. In .... 10 to 15% of reported neuropathies – peripheral neuropathies in the perioperative period, any other cause. In 85 to 90% the world body of literature says the only position we can take is to be agnostic. We do not know what actually caused the problem.

  107. Dr Low’s conclusion was that neuropathy arising from anaesthesia was complex, multi-factorial and incompletely understood : and it was so in this case, notwithstanding the fact that in the American material this conclusion was reached by studies largely based on the ulnar nerve, not the radial nerve : Where the material dealt with radial nerve injury, the cause was said to be external compression applied at the site of the spiral groove.

  108. In one study published by the American Society of Anesthesiologists in April 2000, 39% of consultants expressed the view that an NIBP cuff increased the risk of radial neuropathy; 20% disagreed; 41% were uncertain. Notwithstanding this, Dr Low’s view was that, in this case, the use of the NIBP cuff causing the radial nerve injury was only a “theoretical” possibility.

  109. None of the American material introduced by Dr Low suggested that external compression at the spiral groove was the only possible cause of radial nerve palsy. Dr Low’s opinion, as expressed in the course of cross-examination by Mr Bharwaney SC, counsel for the plaintiff, was that if compression by the blood pressure cuff was “the mechanism” of injury, “then there must be another factor to make that particular nerve so very vulnerable” : A nerve weakened by, say, oedema because of the build-up of excessive fluid would be vulnerable to injury by compression, even though oedema as such would not be the cause; there was also the possibility of ischaemia (areas of bloodlessness) developing.

  110. Dr Low’s general conclusion was that even if proper care was taken in “positioning, and in the application of blood-pressure cuff, etc” there was “still a mystery with all peripheral nerve injuries”. (emphasis added).

  111. As can be seen from paras 95-110 above, the difference in opinion between the two experts regarding etiology was one of degree : Prof. Holland was of the view that the cuff might have been the cause of the external pressure; external pressure alone caused the injury. Dr Low’s opinion was more muted : It was possible that the cuff was the cause of the external pressure, and if that was the cause of the pressure on the nerve, then the injury would have been because of other factors weakening the site of the nerve.


  112. Given the somewhat fluid state of the opinion evidence as summarized above, the trial judge had a difficult task in making a finding on the cause of the radial nerve palsy.

  113. The judge summarized the effect of the medical literature as put to the two expert witnesses in this way (para. 75 of his judgment) :


    While it was previously thought that all perioperative nerve injuries were presentable, that is no longer the generally held view today;


    the invocation of the doctrine of res ipsa loquitur is no longer considered appropriate for perioperative nerve injury;


    whilst peripheral nerve injuries at different sites may have some differences, there is a significant similarity between them;


    today it is believed that perioperative nerve injuries are complex, multi-factorial and incompletely understood; and


    there has been very little investigated evidence collected in relation to radial nerve injuries because incidents of such injuries are rare which lead to inability to collect such evidence.

  114. The judge cannot be taken as having formed his conclusions based upon his own reading of the medical literature. He did not have the necessary medical knowledge to form an independent judgment, and it would have been rash to the extreme if he had attempted to do so. Hence, the judge must be taken to have accepted Dr Low’s opinion, (formed from Dr Low’s reading of the medical literature), that the precise mechanism of peri-operative nerve injury was complex, multi-factorial and not completely understood. Otherwise para. 75 of his judgment had no relevance to the case. And from what the judge said in his para. 75(3) as quoted above, he must be taken to have accepted Dr Low’s statement : “I think a general survey of the literature does not make any distinction between the mechanism of causation for any particular peripheral nerve. And prolonged pressure will affect the ulnar nerve; prolonged pressure will affect the sciatic nerve, prolonged pressure will affect the perineal nerve, just as much as it would affect the radial nerve.”

  115. He found as a fact that it was external compression to the site of the spiral groove which caused the injury to the radial nerve, but he was not prepared to go further on the issue of liability.

  116. At para. 94 he summarized his conclusion thus :


    Having regard to the medical literature put before me, in particular those medical literature indicating that while it was previously thought that all perioperative nerve injuries were preventable, but that is no longer the general view today, I am unable to infer negligence, given the facts of this case, on the part of the defendant, even though I was able to come to the finding that the external compression causing the left radial nerve palsy likely took place at the time when the plaintiff was having the operation under anaesthetic.

  117. The reference to the medical literature in para. 94 harks back to the matters set out in para. 75 of his judgment (quoted in para. 113 above).

  118. There are here two concepts involved :

    1. what caused the external compression, and

    2. what caused the radial nerve palsy.

    In his judgment these have been elided.

  119. It is seldom that, upon close analysis, a trial judge’s findings cannot be better expressed. Nevertheless, Suffiad J’s conclusion is reasonably clear : There might have been a number of factors which combined could be said to be the cause of the nerve palsy. On the evidence the use of the cuff would clearly be one of the factors. But, having regard to the fluidity of the expert evidence, taken as a whole, he was unable to make a clear finding as to what precisely those factors were, and the precise part they played. Hence (using perhaps an unfortunate form of shorthand), he said at para. 91 :

    In that sense, the cause of the injury is unknown since the external compression could have manifested itself in a number of different ways, some of which may not have been readily apparent to the defendant or anyone else present.

    Here he focused upon “the cause of the injury” and unfortunately elided the two concepts; he did not articulate what those “different ways” might have been, but on the basis of the evidence before the court, he must have been thinking of the factors ventilated in the course of the evidence – the use of the cuff, ischaemia, oedema – which, not singly but in combination, might have constituted the cause of the injury : Echoing Dr Low’s evidence that a nerve weakened by the build-up of excessive fluid would be vulnerable to compression over a long period by something like an NIBP cuff.

  120. Hence, the judge’s conclusion in para. 94 of his judgment as quoted above. In short, what the judge was saying was that there were “a number of different ways” (his para. 91) whereby the injury might have been caused without breach of duty of care on the defendant’s part. On all the evidence before him no inference of negligence could be drawn so, as he said in para. 95, the plaintiff failed to prove his case on liability, and the case was dismissed.

  121. Approaching the case in this way, the judge can be said to have simply paid heed to Lord Scarman’s warning with regard to medical negligence cases as expressed in Maynard v. West Midlands Regional Health Authority [1984] 1 WLR 634 at 638 : “Differences of opinion and practice exist, and will always exist, in the medical as in other professions. There is seldom any one answer exclusive of all others to problems of professional judgment. A court may prefer one body of opinion to the other : but that is no basis for a conclusion of negligence.”


  122. In the Court of Appeal the leading judgment was given by Le Pichon JA who focused upon two matters:

    1. Dr Vincent Ma’s report to the effect that when the plaintiff was examined 2 days after the operation he had “local tenderness of left radial nerve at the spiral groove in the posterior aspect of the mid-shaft of left humerus”, and

    2. Prof. Holland’s opinion to the effect that “prolonged and excessive localized external compression of the radial nerve was the only possible cause of the plaintiff’s injury”.

  123. Le Pichon JA held that as, throughout the operation, the plaintiff was under the defendant’s care, and given the two matters as stated in the paragraph above, the burden of proof shifted to him to explain how the external compression causing the injury had come about. As no explanation was forthcoming, negligence had to be inferred.

  124. At para. 22 of her judgment Le Pichon JA, referring to paras 75 and 94 of Suffiad J’s judgment (as quoted verbatim above), said that “the judge must have implicitly rejected the theory that the injury arose ‘without apparent cause’ by expressly finding that the injury was caused by external compression at the spiral groove.”

  125. This, with respect, is to take far too narrow a view with regard to the judge’s findings, and fails to take into account the entirety of the evidence.

  126. It is true that Prof. Holland expressed the opinion that external compression was the only possible cause of the plaintiff’s injury. Full stop. To take this statement as in effect the entire foundation of Suffiad J’s judgment is to distort the effect of that judgment, read as whole, and to ignore the body of evidence as set out in paras 95 to 110 above.

  127. It is true that, in regard to causation, Suffiad J made no specific mention of the NIBP cuff. That can be categorized as a weakness of the judgment. But note the way the plaintiff’s own expert witness understood his case. In his cross-examination by Mr Coleman SC, Prof. Holland said there were only two things he criticized concerning the defendant’s conduct during the entire intra-operative period:

    1. the arm falling off the board, if that was what happened and

    2. the choice of the use of the NIBP cuff.

    It would appear from this that if (i) above was eliminated, Prof. Holland was saying that the cause of external pressure on the radial nerve was the cuff and that this external pressure caused by the cuff was the sole cause of the injury.

  128. Assuming this is the cause of the compressive injury, it would get the plaintiff nowhere. The finding that the use of cuff in the circumstances of this cause did not amount to negligence is unassailable.

  129. Rogers VP was troubled by Le Pichon JA’s approach. He referred to an article which concluded that nerve injuries could occur “spontaneously without any causal relationship to anaesthesia” (para. 2 of his judgment). He repeated the statement that “peri-operative nerve injuries are complex, multi-factorial and incompletely understood”. And he went on to say:

    In short given the complexities of the human body and the comparatively little that is known about its workings, if 2 grounds of complaint are rejected and there is responsible research which concludes that a condition may arise from unknown, and seemingly non-negligent causes, it would seem an inappropriate case to rely on the maxim res ipsa loquitur which is founded on the basis that the condition could not have arisen without negligence.

  130. Rogers VP nevertheless held that it was for the defendant to explain how the external compression at the spiral groove happened, without his fault. He agreed with Le Pichon JA that the appeal should be allowed. Lunn J agreed with the reasons in Le Pichon JA’s judgment. Hence the appeal was allowed and Suffiad J’s judgment on liability was set aside, with costs.

  131. When it came to the application for leave to appeal to this Court, whilst the majority of the court was for dismissing the application, Rogers VP took the unusual course of dissenting. He said :

    When giving judgment I said, at the beginning, that the decision had caused me no little anxiety because the maxim res ipsa loquitur had been relied upon. That anxiety has not waned since the judgment. In paragraph 5 of my judgment I expressed the view that this case did not involve that maxim in its full sense ....

    Rogers VP then went on to say :

    There can be no doubt that the plaintiff put his case of negligence against the defendant on 2 specific bases. The first was that his arm had fallen off the arm board on which it had been placed for the duration of the operation and the second was that a non-invasive blood pressure cuff had been used and it was said that an arterial cannula should have been used. Right up to the end of the cross-examination of the defendant the questioning was pursued on the basis of the use and positioning of the cuff.

    The judge rejected both the allegations. He had also heard the evidence. In particular, there was the evidence of the defendant as to the care and attention he took in the course of the operation. In this regard it is to be noted that the defendant said in evidence that he had known the plaintiff for many years as a colleague. For that reason he did recall the operation, albeit one of very many that he had taken part in over the years though it might have been.


  132. As Megaw LJ said in Lloyde v West Midlands Gas Board [1971] 1 WLR 749 at 755, the Latin maxim res ipsa loquitur is no more than an exotic though convenient phrase to describe what is in essence no more than a common sense approach, not limited by technical rules, to the assessment of the effect of the evidence in certain circumstances. And, as Lord Evershed MR remarked in Moore v R. Fox & Sons Ltd [1956] 1QB 596 at 614, it must always be a question whether, upon proof of the happening of a particular event, it can with truth be said that the thing speaks for itself.

  133. The event or “thing” must be so clear-cut that a court can say with assurance : Unless the defendant can come forward with some credible explanation, it must be concluded that want of care produced that result. As stated in Clerk & Lindsell on Torts [19th ed.] para. 8-152, common experience suggests that in the absence of negligence, bales of sugar do not usually fall from hoists, barrels do not fall from warehouse windows, cranes do not collapse, trains do not collide and stones are not found in buns. In the medical context because of the complexity of the human body and the fact that medical science is perpetually evolving and changing, things are seldom so clear-cut.

  134. Here, the opinion evidence as to causation covered a wide spectrum : The use of the NIBP cuff was the cause, a potential cause, a contributing factor. From none of this could an inference of negligence, without more, be drawn. What is more, there was simply no other “hard object” as disclosed in the evidence which could possibly have caused the external pressure.

  135. Mr Badenoch QC in his able submission to this Court said this : Given the extent of the injury (as disclosed by the nerve conduction tests), the localized site of the injury and the length of time before the radial nerve was healed, the inference is irresistible that the external compression injury was caused by prolonged contact with a hard object in the operating theatre. For the purpose of this submission the NIBP cuff must be eliminated as a “hard object”. At one point he floated the idea that a surgical instrument might have been left on the arm board during the operation. If the evidence were as cut and dried as that, then it might well have been right for the burden of proof to shift to the defendant, to rebut the inference of negligence. As Hobhouse LJ said in Ratcliffe v Plymouth and Torbay Health Authority (1998) 42 BMLR 64 at 84, (a case which concerned an operation for fixing a joint : “arthrodesis”), if the facts of that case had been that the plaintiff had gone into the operating theatre to have an arthrodesis to his right ankle and had come out with his right ankle untouched and an arthrodesis to his left ankle, clearly no expert evidence would be required to support an inference of negligence on the part of the defendant. ‘In the ordinary course of things’ that does not happen if those conducting the operation had used proper care. But when expert evidence has been adduced on both sides and the cause of the mishap has been explored evidentially at trial, the foundation for applying the “doctrine” of res ipsa loquitur, shifting the burden of proof to the defendant, would seldom exist. And so it was in this case.


  136. On the evidence it is plain that, from the start, the defendant appreciated the risk of nerve palsy arising from the procedures necessitated by the operation : He said he took steps to lessen the risk by lifting the surgical sheet, feeling the cuff, looking at the cuff and checking on other pressure points. He did not say he specifically recalled doing all these things in this particular case. This was routine procedure and, apart from that one narrow finding on which the Court of Appeal pinned the entire case, there was nothing to suggest that the defendant had departed from normal procedure in this case.

  137. As Hobhouse LJ said in Ratcliffe v Plymouth and Torbay Health Authority (supra) at p83, few, if any, medical negligence cases are brought to trial without, among other things, expert reports lodged. Trials do not open in the vacuum of available evidence and explanation as sometimes occurs in road traffic and other accident cases, but with expert evidence on both sides having been laid before the court; the aetiology would have been to an extent at any rate identified.

  138. At the end of the day, this case cannot be decided solely upon a “thing”, a single matter to determine liability. It can only be resolved on all the evidence, including of course a consideration of the defendant’s conduct : His appreciation of the possible risks involved, and the steps he took to lessen the risks.

  139. I would add here a word of caution : It is dangerous to resolve a case like this based upon the court’s own appreciation of such medical literature as is available to the court; the literature in court could well be incomplete; the court could only come to a conclusion guided by the experts’ understanding of such material, as expressed in their evidence in court.

  140. Here the state of the evidence on causation, at the end of the day, was nebulous and incomplete. But, on any view of the evidence, the NIBP cuff played some part in causing the injury : Albeit, as far as Dr Low was concerned, not a decisive part.


  141. It may well be, as a matter of hindsight, that intra-arterial cannulation to monitor blood pressure would have been preferable in this case. There would not have been pressure on the radial nerve exerted by repeated inflation of the cuff. But to hold that the decision to use the NIBP cuff instead of arterial cannulation amounted to negligence would put the standard of care required of the defendant anaesthetist unacceptably high : Particularly bearing in mind the plaintiff’s admission, in effect, that he (himself a highly experienced surgeon) would not have acted differently. In the words of the judge and of Rogers VP, to hold the defendant liable would render him effectively the insurer of the patient’s well being.

  142. The standard of care which the law requires of the anaesthetist is such reasonable degree of knowledge and skill as is ordinarily possessed and exercised by other members of his profession in similar circumstances. The defendant has not been shown to have fallen below those standards.

  143. I would discharge the Court of Appeal’s order and restore Suffiad J’s order, including his order as to costs.

    Lord Hoffmann NPJ

  144. I have not found the answer to this case obvious. In fact, I have changed my mind more than once. The Court of Appeal said that the judge’s findings were inconsistent with each other. He had said on the one hand that the injury had been caused by external compression of the nerve and on the other hand that the cause could have been something else. No doubt the judge’s findings of fact could have been clearer – there can be few judgments of which this cannot be said – but I think any findings of fact should be approached on the assumption that unless the contrary is clearly demonstrated the judge understood the evidence and did not intend to contradict himself. I think that what the judge meant was that external compression was the cause in the sense that the injury would not otherwise have happened, but that it could have been applied without negligence and caused the injury in conjunction with other unidentified causes.

  145. At first I thought that there was evidence upon which the judge could have reached this conclusion. But the careful analysis of Mr Justice Ribeiro PJ has persuaded me that the medical literature upon which the judge relied will not sustain it. The only pressure which could have been applied without negligence was from the use of the NIBP cuff. But the cuff, properly used, could not possibly have caused the severe compression injury which was found to have occurred. The arm must have been in sustained contact with a hard object during the course of the operation. The burden on the plaintiff is to prove that negligence is more likely than not to have been the cause of his injury and, like Mr Justice Ribeiro PJ, I agree that on the facts found by the judge, that burden was discharged.

    Justice Bokhary PJ

  146. By a majority (Mr Justice Litton NPJ dissenting), the appeal is dismissed with the order nisi as to costs proposed by Mr Justice Ribeiro PJ.


Scott v London and St Katherine Docks Co [18] (1865) 3 H & C 596; 159 ER 665

Ratcliffe v Plymouth and Torbay Health Authority (1998) 42 BMLR 64

Lloyde v West Midlands Gas Board [1971] 1 WLR 749

Delaney v Southmead Health Authority (1992) 26 BMLR 111

Barkway v South Wales Transport Co Ltd [1950] 1 All ER 392

Lillywhite v University College London Hospitals’ NHS Trust [2005] EWCA Civ 1466

Maynard v West Midlands Regional Health Authority [1984] 1 WLR 634

Moore v R. Fox & Sons Ltd [1956] 1QB 596

Authors and other references

Dr Robert Morell, Current Reviews in Clinical Anesthesia, Lesson 23, Volume 21

Dr Mark Warner, “Perioperative Neuropathies”, Mayo Clinic Proceedings, Vol 73(6), June 1998

Clerk & Lindsell on Torts 19th ed.


Russell Coleman SC and Zabrina Lau (instructed by Messrs JSM) for the appellant.

James Badenoch QC and Mohan Bharwaney SC (instructed by Messrs Robertsons) for the respondent.

all rights reserved