England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM NORWICH COUNTY COURT
HIS HONOUR JUDGE DARROCH

		Royal Courts of Justice Strand, London, WC2A 2LL
		16th April 2003

Lady Justice Hale:

1. This is the defendant's appeal against the order of His Honour Judge Darroch in the Norwich County Court dated 22 August 2002. He gave judgment for the claimant in an action for personal injuries against his employer in a sum to be agreed or decided. Judgment was entered on 7 October 2002 for £211,149.
2. The claimant suffers from palmar arch disease (PAD). This is the occlusion of a blood vessel known as the palmar arch, located in the palm of the hand and supplying the blood vessels going to the fingers. The main issue in this appeal is whether the judge was wrong to find that the claimant had proved that his PAD was caused by the use of vibratory tools at work (the 'medical causation' issue). A secondary issue is whether the judge was wrong to find that it was caused by excessive (ie negligent) exposure to vibration (the 'legal causation' issue).

Factual summary

3. The claimant was employed by the defendant at their Norwich depot as a distribution fitter from 1974 to 1998. Throughout that time he routinely used a variety of vibratory tools, including road breakers and whackers and other drills. In May 1998 he was diagnosed with Hand-Arm Vibration Syndrome (HAVS) and taken off his duties as a fitter. He was medically retired in November that same year. At that stage both his employer's occupational health service and his GP were agreed that he had HAVS. In February 1999 Dr Merry, a consultant rheumatologist, agreed that the most likely diagnosis was vibration white finger (VWF) which is part of HAVS. Proceedings were begun on 31 October 2000 on the basis that he was suffering from 'HAVS formerly known as vibration white finger' (VWF). A number of standard allegations of negligence and/or breach of statutory duty were made. In essence these amounted to having no system for preventing or detecting the disease. The engineers instructed on each side reported that the claimant's exposure to vibration had been enough to cause VWF and that the risk was foreseeable. The defendant therefore admitted breach of duty.
4. The dispute was over the diagnosis. The claimant had instructed Mr FW Cross, a consultant vascular surgeon. He had originally opined that the claimant was suffering from HAVS. The defendant instructed Dr RA Cooke, a consultant in occupational medicine. He thought that the claimant's symptoms were not typical of the vascular effects of vibration - they had not improved once exposure had ceased (indeed they had got worse) and were near permanent rather than episodic - and needed further investigation. Mr Cross therefore carried out a battery of further tests (costing some £2000) which would not have been thought necessary or appropriate in an ordinary HAVS claim. These showed that the claimant was not suffering from VWF.
5. A second joint statement of the medical experts, signed on 26 March 2002, agreed that 'Raynaud's phenomenon is the predominant symptom of the vascular component of hand arm vibration syndrome' and the claimant was not suffering from Raynaud's phenomenon. They also agreed that 'Mr Griggs has a problem affecting the blood flow through the palmar arch of his hand'. It was Mr Cross's opinion that this was palmar arch disease (PAD) rather than a similar condition known as hypothenar hammer syndrome (HHS); also that 'in the absence of any signs of vascular disease elsewhere, the most likely cause of this in Mr Griggs case is the use of vibrating machinery'; and that 'vibrating machinery can cause palmar arch disease (ref Honma H et al, Int J Occup Med Environ Health 2000; 13: 275-286)'. It was Dr Cooke's opinion that Mr Griggs might be suffering from HHS but that generalised or diffuse PAD would also account for his symptoms; and that vibration had not been shown to cause PAD and the reference quoted by Mr Cross did not confirm such a relationship. At trial, however, Dr Cooke was content to adopt the diagnosis of PAD and the judge found that the claimant was in fact suffering from PAD.

The judgment

6. The judge first asked himself whether the claimant's PAD was caused by vibration and concluded:

"I have absolutely no doubt in my mind that this is a vibration induced complaint. I say that because I have considered the evidence and what Mr Cross says is that he is 80% or so sure that it is caused by vibration. He has found a case in the literature, a thin case if you like about motocross, which appears to connect the two. His opponent says that he has suspicion but is not prepared to make the jump. So at one extent 80%, at the other 'suspicion'. I think therefore that the answer to the primary question 'Is this vibration induced?' is that it is. I have borne in mind the evidence of the expert engineers; the claimant's own evidence; the previous diagnoses, and the extent to which the claimant expert can go, and the maximum concession made by the defendant.

The claimant also said through his expert that this was diagnosis by exclusion, which is not entirely satisfactory. It is, however, and I think the authorities show, a possible approach. I can also turn to the criticisms that are made of the defendant's expert, Dr Cooke, and really they amount to saying that at the end of the day he is forced to admit that it is palmar arch disease. He has therefore changed his position as well. I do not think it is the hammer syndrome and I do not think he really seriously argues for that now. It is not Buerger's disease. That is caused by smoking. I can eliminate that. He really, in my view, cannot come up with a more plausible explanation, so the diagnosis by exclusion, the literature and the experience of Mr Cross all lead me to the firm conclusion that this is vibration induced."

7. The judge next asked himself whether it was possible that a non-negligent, acceptable level of vibration might have triggered this complaint. He referred to Fairchild v Glenhaven Funeral Services Limited [2003] 1 AC 32, adopting the principles derived from the opinion of Lord Reid in McGhee v National Coal Board [1973] 1 WLR 1, HL. Lord Reid had said that there was no real distinction between making a material contribution to the injury and materially increasing the risk of injury. The judge asked himself whether the defendant materially increased the risk and held that very clearly they did:

"One looks at the joint experts' report; the history; the degree of involvement; the vibration; the heavy equipment. Asking a man to do that without being monitored medically it seems to me makes a very substantial increase in the risk."

8. Although the second issue was labelled 'apportionment' in the court below, the judge was not in fact asked to apportion the damage between that caused with or without negligence in accordance with the principles in Allen v British Rail Engineering Limited [2001] PIQR Q101. The only issue put to him was legal causation, that is whether any of the injury suffered by the claimant was the result of the defendant's negligence.

Medical causation

9. The judge relied in part on the evidence of the claimant and the engineering experts as to his degree of exposure to vibration, which was on any view very substantial and over a period of many years. The joint statement of the engineering experts, Mr RH Mitchell and Professor MJ Griffin, dated 28 May 2002, to which the judge also referred, contained the following paragraphs:

"1. It is agreed that the magnitudes of vibration experienced by the Claimant while employed by the Defendant were sufficient to cause vibration-induced white finger. . . .

3. It is agreed that the durations of exposure to hand-transmitted vibration as claimed by the Defendant were sufficient to present a foreseeable risk of the Claimant developing vibration-induced white finger.

4. It is agreed that, in addition to vibration-induced white finger, exposure of the hand to vibration can cause a variety of other disorders collectively known as the hand-arm vibration syndrome. . . .

6. We agree that we do not know of a previous case in which the development of palmar arch disease has been attributed to the vibration from hand-held vibratory tools.

7. It is agreed that the development of palmar arch disease, as a specific disease, was not foreseeable, but that the development of hand-arm vibration syndrome was foreseeable."

10. The medical issue, however, is not the degree of the claimant's exposure to vibration. If his symptoms had been typical of VWF there would have been no argument. The issue is whether the symptoms from which he suffers have been caused by his exposure to vibration. There has been a great deal of confusion in this case as to whether they do or do not fall within the generic term HAVS. Although a full explanation was given by Smith J and approved by the Court of Appeal in the Allen case, some basic definition may be helpful.
11. Raynaud's phenomenon is the term given to cold induced episodic whitening affecting the fingers. It can arise constitutionally, but if occupationally induced through the use of vibratory tools, it became known as vibration-induced white finger. The typical symptoms are intermittent reversible blanching, initially of the tips of the fingers most exposed to the vibration source, but eventually spreading to involve all fingers and the tips of the thumbs. There is a direct relationship between the degree of exposure and the severity of the symptoms. It then became recognised that vibration could affect not only the digital vascular tree but also the nervous system and the muscular skeletal system. Hence the term hand-arm vibration syndrome was introduced in 1986.
12. But it is important to remember that a syndrome is simply a group of signs or symptoms which frequently occur together and form a distinctive clinical picture. The underlying causes of the development of those symptoms are another matter. Causal questions are of at least two types. First, what bodily changes are causing the signs and symptoms? Second, what has caused the bodily changes?
13. In VWF there is an interruption of the blood supply to the fingers caused by vasospasm. This is episodic in that the symptoms recede once the exposure to vibration has ceased, although they recur with increasing severity once exposure is resumed. Among many sources, Dr Cooke quoted from Pelmear's contribution to a joint symposium on hand arm vibration syndrome, held in Newcastle, September 1999:

"the predominant health effect is known as hand arm vibration syndrome, defined as a disease entity with the following separate components - circulatory disturbances (vasospasm with local finger blanching - 'white finger') . . . while arterial disruption or spasm is necessary to stop blood flow, vasospasm in the skin venules is required to produce blanching . . . the blanching is restricted to the tips of one or more fingers but progresses as the vibration exposure time increases. The thumbs are usually the last to be affected . . . it is generally agreed that due to the mechanical stimulus, specific anatomical changes occur in the small arteries and arterioles of the digits."

14. Hypothenar hammer syndrome and palmar arch disease are rare examples of damage to the vascular system in the hand. In HHS there is damage to the ulnar artery in the hypothenar (ie opposite the thumb) eminence at the base of the hand. In PAD there is damage to the palmar arch. There are two palmar arches in each hand, superficial and deep, supplied by the radial artery coming up next to the radius and the ulnar artery coming up next to the ulna. They then give off branches to the fingers. If they are obstructed, the blood supply to the fingers is obstructed.
15. The issue between the doctors was whether the damage to the palmar arch could be caused by vibration. Mr Cross gave evidence [Transcript p 33] that the commonest cause of PAD was Buerger's disease, narrowing of the medium to small vessels in the hands, feet and around the ankle, caused by excessive smoking. Other causes would include rheumatoid disease, microvascular disease, and connective tissue disorders [Transcript p 34]. These he had excluded. He had taken a very careful history and done extensive tests. No stone had been unturned. The only aetiological factor he could think of was the exposure to vibration injury. There were no other risk factors. This was 'diagnosis by exclusion' but there was no evidence that there was anything else to cause it. Hence he would say that on the balance of probabilities it was much more likely than not. He would go as high as 85% to 90%.
16. Mr Cross acknowledged [Transcript p 34] that there was not much to be gained from the literature. There was a paper by Stark and others, 'White fingers after excessive motorcycle driving: a case report', VASA, Band 19, 1990, p 257. The patient had symptoms comparable to those in VWF. Angiography revealed an ulnar artery aneurysm, which is a feature of HHS, and damage to the palmar arch. The only explanation the authors could find was the exposure to vibration through excessive motocross driving. Mr Cross acknowledged that there were flaws in this report (Dr Cooke had pointed out, for example, that an earlier accident might have caused the damage). Also, case reports were not very good medical evidence. The best medical evidence was the randomised prospective controlled trial, going down through various other types of study until one reached the humble case report. He accepted that it was 'quite a jump' to conclude that vibration exposure causes PAD but the way conditions come to doctors' notice is by their being reported initially as case reports.
17. He also referred [Transcript p 35] to the 'doyen of vascular surgery textbooks', Rutherford's Vascular Surgery. This states at p 1134:

"Conditions responsible for obstruction of the arteries of the hand and fingers include emboli, vibratory trauma (in chain saw or jackhammer operators), repetitive percussive trauma (hypothenar hammer syndrome, baseball catching), frostbite, auto-immune diseases (scleroderma or rheumatoid arthritis), Buerger's disease, intra arterial drug administration, and exposure to various toxins."

Rutherford then goes on to explain how to detect whether hand symptoms are due to arterial obstruction or to a vasospastic process. If arterial obstruction is found,

"The distribution of the lesions . . . may suggest a cause. An incomplete palmar arch may represent a common congenital variant . . . or it may be due to any of a host of pathological entities, including trauma, atherosclerosis, emboli, and collagen diseases."

Mr Cross's evidence was that congenital absence could be ruled out, and everything else had been excluded but for trauma, in which he would include vibration exposure. Vibration was multiple trauma.

18. Under cross-examination he agreed [Transcript p 44] that the literature 'does not really tell us anything'. He could not stand up at a vascular meeting and say that 'palmar arch disease is part of the hand arm vibration syndrome' as a scientifically proven statement [Transcript p 46]. But he stuck firmly to his view that the claimant's condition was probably caused by vibration. He pointed out [Transcript p 48] that if it was caused by pressing on a pneumatic drill for a long period of time you would expect to see very diffuse disease affecting all the fingers of both hands, which was what the claimant had. He thought it very suggestive that 'where you are pressing on the drill is right over the palmar arch of both hands'. He expanded this later [Transcript p 50]:

"The middle area of the palm is where the drill is when you are leaning on it. It is quite a focused area. The palmar arch is quite short. You can see from this arteriogram the superficial and the deep palmar arches are quite short. Mr Griggs was exposed to a lot of vibration over the years and what would you expect is the vessels to become narrowed over quite a long time to the point where the narrowing became critical."

19. Dr Cooke accepted [Transcript p 63] that exposure to vibration over 24 years was likely to have had some effect. But he did not understand how vibration exposure could cause damage to the palmar arch. He had not found convincing evidence in the literature. This did not prove that there was not a link but it was a jump that he was uncomfortable with given the lack of evidence to date.
20. He was referred to a series of papers on HHS: Ferris et al 'Hypothenar hammer syndrome: Proposed etiology', J Vasc Surgery, 2000; 31: 104-13; Wernick and Smith, 'Bilateral Hypothenar Hammer Syndrome: an Unusual and Preventable Cause of Digital Ischaemia', Am J Emerg Med 1989; 7: 302 - 306; Spencer-Green et al, 'Hypothenar Hammer Syndrome: An Occupational Cause of Raynaud's Phenomenon' J Rheumatology 1987; 14: 1048 - 1051; and Kaji et al, 'Hypothenar Hammer Syndrome in Workers Occupationally Exposed to Vibrating Tools', Journal of Hand Surgery (British and European Volume, 1993) 18B: 761 - 766. He agreed [Transcript p 65] that the main recognised cause of HHS was repetitively striking objects with the heel of the hand, using the hand as a hammer. The final article, by Kaji and others, however, had found HHS in a group of workers with long exposure to vibrating tools. They had concluded that high frequency mechanical stress from hand-held vibrating tools which were repeatedly in contact with this part of the hand had led to the development of HHS. Dr Cooke [Transcript p 66] acknowledged that the authors had demonstrated very well an incidence of HHS in this group of workers. But they had not compared them with a control group of workers doing similar heavy work without the vibration exposure. So it was an interesting start but did not take it through to the conclusion. He drew a distinction between vibration and blunt trauma, although he accepted that in certain circumstances the effect might be the same [Transcript p 67].
21. He also agreed [Transcript p 67] that HHS and PAD were very similar. Indeed 'one could argue that HHS is one cause of PAD'. When asked whether he accepted that Mr Cross had excluded all the other known causes of PAD, he replied that two might be worth considering a little further, but he eventually agreed that both of these were most unlikely. When the combination of evidence was put to him to the effect that the claimant's PAD had been caused by vibration, he agreed [Transcript p 70] that there was 'an element of suspicion that requires further investigation' but in his view that was a long way from the balance of probabilities test.
22. Mr Owen QC for the appellant argues that Mr Cross did not inspire confidence as an expert. In his first report, dated 30 September 1999, he had undertaken an outdated test and diagnosed a severe case of HAVS. In the first joint statement following discussion with Dr Cooke, dated 11 July 2001, he had agreed that there were features not suggestive of HAVS. In his second report after the battery of tests had been carried out, dated 13 December 2001, he had diagnosed PAD which was 'likely to be associated with HAVS', but had referred to VWF in his conclusions. This last was a slip of the pen. In a draft of the second joint statement, signed on 22 March 2002, he had agreed that both HHS and PAD were conditions of some rarity and not part of the HAVS although the symptoms were extremely similar. But in a letter dated 26 April 2002, he stated that he did consider PAD to come within the definition of HAVS.
23. It is clear that PAD has not hitherto been recognised as part of HAVS. Further the claimant is suffering from damage to the vascular system in his hands rather than any of the other damage which is a recognised part of that syndrome. But saying that this claimant's PAD comes within the scope of HAVS was simply another way of saying that his symptoms are caused by exposure to vibration. The definition of HAVS given by the Health and Safety Executive in their guidance on 'Hand-Arm Vibration' (1994, HS(G)88) was this:

"In the United Kingdom HAVS is considered to exist if, after prolonged exposure to hand-transmitted vibration, involvement of the vascular and/or peripheral nervous system occurs with or without musculoskeletal involvement."

Mr Cross's opinion was entirely consistent with that definition.

24. When faced with competing expert evidence, it is an important part of the judge's task to assess whose opinions are the more reliable. As the judge here implicitly acknowledged, willingness to change one's opinions in the light of discussion with others and the evidence as it develops may be a sign of strength rather than weakness. Both of these experts had changed their views to some extent. The judge who hears the experts give their evidence will be best placed to assess whether they genuinely hold the opinions they are now expressing or are vainly clutching at straws to bolster a case. This judge had absolutely no doubt about the conclusion reached.
25. Another factor in deciding between experts is their respective qualifications and experience. Dr Cooke is a specialist in occupational medicine with a longstanding interest in vibration related disease. As such it is not surprising that he detected differences between the claimant's presentation and that of typical VWF sufferers although these had escaped all the other doctors who had previously examined him. Mr Cross is a consultant vascular and general surgeon. As the claimant is suffering from damage to his vascular system, it is not surprising that he was better equipped to diagnose precisely where the damage was. He was also equally if not better equipped than Dr Cooke to consider what has caused it.
26. The most important factor in deciding between experts, however, is the quality of the evidence and reasoning upon which their opinions are based. Mr Owen understandably places great weight on the lack of support from the literature, which Mr Cross freely acknowledged. It is scarcely surprising that Mr Cross could not get up at a vascular meeting and state that PAD was part of HAVS. By then he had agreed that the claimant was not suffering from Raynaud's phenomenon which is the principal vascular component of HAVS. PAD is a very rare disease and the link with the use of vibratory tools has not yet been clearly acknowledged. But we are here concerned with the balance of probabilities rather than scientific proof. Mr Cross was careful to limit his opinion to the cause of PAD in this particular claimant.
27. Mr Owen was also critical of the 'diagnosis by exclusion'. But he stopped short of arguing that this was a case of the 'Holmesian fallacy' explained by Lord Brandon in Rhesa Shipping SA v Edmunds, The Popi M [1985] 1 WLR 948 at 955G: 'when you have eliminated the impossible, whatever remains, however improbable, must be the truth'. He does not and cannot argue that vibration is an improbable cause of damage to the palmar arch. Dr Cooke may have drawn a careful distinction between blunt trauma and vibration, but even he acknowledged that they could produce the same effects. If the claimant establishes that all other possible causes of the damage must effectively be excluded, then it does indeed become more probable than not that the possible cause which has not been excluded is indeed the true cause. To adopt that process of reasoning is not, as Mr Owen submits, to reverse the burden of proof.
28. Nor is it possible to say that there is no support at all for Mr Cross's opinion in the literature. Rutherford refers to obstruction of the arteries of the hand and fingers being caused by both 'vibratory trauma' and 'repetitive percussive trauma'. While damage to the ulnar artery in HHS is generally caused by the latter, there is evidence linking it with the use of vibratory tools. Dr Cooke, who indeed came to court ready to argue that the claimant suffered from HHS, acknowledged the close relationship between HHS and PAD. Mr Cross, who knows about blood vessels, produced a mechanism by which the use of vibratory tools could indeed damage the blood vessels in the palm of the hand. The sources on VWF cited in correspondence by Dr Cooke make it clear that the trouble begins in the fingers in closest contact with the vibrating machinery. The extract from Pelmear quoted above shows that it is generally agreed that due to the mechanical stimulus, specific anatomical changes occur in the small arteries and arterioles of the digits. If vibration can cause damage to the arteries in the fingers, and to the artery in the heel of the hand, it would be surprising if it could not also cause damage to the palmar arch.
29. Thus it is completely understandable that Dr Cooke, with his extensive knowledge of the symptomatology of VWF and the conditions recognised to be part of the HAVS, would recognise that this claimant's case was different. The reason why his symptoms were different was then discovered. His use of vibratory tools was a suspect cause of his condition. It is equally understandable that Mr Cross, having eliminated all other causes and being of the opinion that vibratory trauma could cause this sort of damage, concluded that it was 85 to 90% probable that it was the cause in this case. It would have been surprising if the judge had reached any different conclusion.

Legal causation

30. In HAVS there is a recognised dose/response effect: so much vibration produces so much VWF. In PAD it is not possible to say how much vibration will produce the damage. Mr Cross acknowledged this in cross-examination [T p 53]. This is scarcely surprising as PAD is so rare and the link with vibration is only just being recognised.
31. Hence it is argued that the claimant could not prove that his PAD was caused by guilty, unacceptable, negligent levels of exposure rather than by innocent, non-negligent, acceptable levels; and that the judge was wrong to regard this as covered by the principles in McGhee and Fairchild. In both of those cases, the risk - that brick dust could cause dermatitis and that asbestos could cause mesothelioma - was well-known. In both of those cases there was a proven breach of duty which could have caused that risk to materialise. In both of those cases the state of knowledge was such that the claimant could not establish that it was that breach of duty which had caused his damage. In those circumstances, relaxation of the strict 'but for' requirement of causation was justified. Those features, it is argued, do not apply here.
32. Against this, Mr Cooksley QC for the claimant argues first that it is too late for the defendant to raise this issue now. The whole trial proceeded on the basis that if medical causation could be established then legal causation would follow. The legal causation point was only clearly made in closing submissions. Had it been made earlier he would have explored other matters in evidence, including the timing of the onset of symptoms. Secondly, he argues that the 'but for' test is satisfied in this case. The evidence before the court was that the claimant first experienced symptoms in the 1980s. The defendant had admitted global breaches of duty in their system for detecting and preventing vibration induced diseases. The court should infer that had they had a proper system the claimant's condition would have been detected earlier and prevented. Thirdly, however, he argues that the case undoubtedly falls within the material increase in risk principle in McGhee and Fairchild. Here we have a global breach of duty in relation to protecting employees from vibration induced disease. We also have a disease which is very difficult to distinguish from the classic HAVS. It is a disease which the court has held to have been caused by prolonged exposure to vibration. Even if the classic 'but for' test cannot be met in the present state of knowledge, it is a clear case of a material increase in risk.
33. Bearing in mind the length and magnitude of this claimant's exposure to vibration, as demonstrated by the engineering experts, and the extensive breaches of duty admitted by the defendant, it is possible to conclude that the but for test is indeed satisfied in this case. If the breach of duty was sufficient to cause one type of vibration induced injury why should it not also be sufficient to cause another? Despite the careful distinction drawn by Dr Cooke between blunt trauma and vibratory trauma, it was eventually agreed between the experts that all causes of this claimant's condition apart from his exposure to vibratory tools at work had been effectively excluded.
34. This sort of case is very different from those of psychiatric injury caused by stress at work. Not only are there a great many other potential contributory causes of psychiatric injury. More importantly, it is a great deal more difficult to establish the precise breach of duty of which the employer is guilty. It then has to be shown that that breach caused or materially contributed to the psychiatric injury suffered. Neither of those difficulties arises in this case.
35. In any event even if the but for test could not be satisfied, there can be little doubt that the employer's failure to have a proper system for detecting and preventing vibration induced diseases materially increased the risk of an employee sustaining such a disease. Once the degree of exposure, the breaches of duty and the medical causation had been established, it would be an unjust legal system which did not hold the employer responsible for what had happened.
36. It is in the circumstances scarcely surprising that Mr Owen regarded this as very much a secondary point. Although it is right to say that it was raised in the defence, and alluded to somewhat obscurely in the defendant's opening submissions, it was certainly not at the forefront of anyone's thinking in preparation for or during the trial. It would be wrong to disturb the judge's conclusion upon it now.
37. For those reasons, I would dismiss this appeal.

Wilson J:

38. I agree.
Order: As per draft order. (Order does not form part of the approved judgment)