Chadwick v Hollingsworth (No. 2)

This is an appeal from the judgment and order of HH Judge Knight QC dated 8 January 2010, whereby he gave permission to the claimant, Mr Donald Chadwick, to amend his particulars of claim in medical negligence, but on terms that he pay all the costs of the action to date (other than costs ordered on a previous occasion). His judgment is contained in the following brief passage of the proceedings:

“I disagree for two reasons. I think that the way this case has been repleaded does advance a substantially new case on both the allegations of breach and causation. I also accept the defendant’s point that the particulars of claim in so far as the causation evidence is concerned, would have failed. I will not say anything more about the extent to which the amended particulars of claim do or do not give rise to a claim based on Bowman [sc Bailey]and the correct causation test. Those are probably issues to come, but it does seem to me that the inevitable consequence of the amendments you make mean that the defendant now faces a substantially different case with the inevitable result that the trial must go off, but I will give you permission to appeal.”

The consequent appeal was made, as the judge had himself anticipated, to the court of appeal. However, the question arose whether it should have been made to a high court judge, on the ground that the judge’s decision had not been a “final decision” for the purposes of the Access to Justice Act 1999 (Destination of Appeals) Order 2000. At the hearing of the appeal on 4 October 2010, the court of appeal held that it had no jurisdiction and remitted the appeal to the high court (see the judgment of Lord Justice Moore-Bick [2010] EWCA Civ 1210). I was presiding over that hearing and then directed, with the consent of the parties, that, sitting as a member of the high court and exercising its powers, I would complete the hearing of Mr Chadwick’s appeal then and there.

This is my judgment on that appeal.

On 11 July 2003 Mr Chadwick, to whom I will refer as the appellant, underwent a replacement of his left knee under the care of the defendant, Mr Robin Hollingsworth, a consultant orthopaedic surgeon (the “respondent”). The wound became infected, and this led to further complications, including a two-stage revision of the replacement in November 2003 and January 2004, and to the appellant’s allegation of chronic infection, on-going pain and disablement.

The appellant commenced his claim in these proceedings on 29 June 2006. His claim form was served on 25 October 2006, at that time based on the expert advice and a report from another consultant orthopaedic surgeon, Mr Sean Curry FRCS. No complaint was then made about the original operation, but the alleged negligence was said to have commenced from 19 September 2003, when the appellant saw the respondent again as an outpatient. There was swelling under the scar on the knee which had been operated on. The swelling was aspirated and found to be infected by coagulase-negative staphylococcus aureus. The appellant was admitted for a course of intravenous antibiotics. This was not successful and on 24 October 2003 the appellant was admitted again for a debridement, washout and the insertion of a drain, and for a further course of intravenous antibiotics. On 18 November 2003 the respondent performed the first stage of a second knee replacement, by removing the components inserted in July; and on 22 January 2004 the respondent performed the second-stage of that second replacement or “revision” as it is called, when the new components were inserted.

In his original particulars of claim the appellant, basing himself on the advice of Mr Curry, complained (a) of the respondent’s failure to move directly on discovery of the infection in September 2003 to a debridement and washout of the knee and the removal of the then present insert; and (b) of the respondent’s further failure to ensure that the infection had been eliminated before proceeding to the second-stage of the revision in January 2004.

The respondent had instructed his own expert orthopaedic surgeon, Mr Andrew Jackson FRCS. Mr Curry and Mr Jackson met and produced a joint report dated 13 November 2008. As a result it became clear that Mr Curry had changed his mind and no longer felt able to support the appellant’s case. He now conceded that even if a debridement and washout had been performed immediately in September 2003, there would only have been a “small chance of success” in eradicating the infection; and secondly, that the decision to proceed with the second-stage replacement would have depended on the appearance of the soft tissues but that it was unlikely that the respondent would have proceeded in the presence of a suspected wound.

Following the joint report, the appellant’s solicitors attempted to obtain an explanation from Mr Curry for his change of mind, but there was no response to any of their letters. They therefore approached another orthopaedic surgeon, Mr John Newman FRCS, who produced a first report dated 21 September 2009. This was favourable to the appellant, essentially reinstating Mr Curry’s original opinion. Mr Newman concluded his report as follows:

“In conclusion, Mr Chadwick was unfortunate to develop infection following his original knee replacement. His management was unsatisfactory in that once the acute infection was recognised an urgent wash-out was not performed. In addition, when debridement and a wash-out were carried out, this was done in an inadequate way. His management was also unsatisfactory in that the second stage of the revision procedure was performed too early and I believe a wrong decision about the timing was made, though to be fair, it did seem as though his treating surgeon was exercising reasonable care in his decision-making but succumbed to the temptation to move matters forward when further delay would have been wiser. This has resulted in recurrent infection with a substantial increase in Mr Chadwick’s ultimate disability.”

In the meantime, well after the particulars of claim had been served but before Mr Curry had departed the scene, the litigants had also instructed consultant microbiologist experts. The respondent had instructed a microbiologist first, Professor K A V Cartwright, whose report is dated March 2008. He supported the respondent’s defence, which was that the infection could not be cured other than by a two-stage revision of the knee, which had been properly performed. The appellant’s microbiologist, Professor S J Eycken, produced her report on 21 August 2008. She opined that the original operation had not been carried out with an acceptable prophylactic regime, but she doubted that this would have altered the outcome. She agreed with Mr Curry that there should have been an urgent debridement and washout in September 2003. She also said that the second-stage revision had been performed too soon, when the wound had barely healed and thus was prone to infection: this went beyond Mr Curry’s criticisms. The microbiologists’ disagreement survived their joint report dated 14 October 2008.

Following Mr Curry’s non-cooperation and the obtaining of a new report from Mr Newman, the appellant applied on 28 September 2009 for permission to substitute Mr Newman and his report for Mr Curry and his, and also for the vacation of the trial date which had been fixed for 30 November 2009.

That application came before HH Judge Bailey on 5 November 2009, and he granted it. It was contemplated that the trial might only have to be postponed for some two months, and it was relisted for 25 January, for 2 days (the previous listing had been for four days). Directions were given to accommodate that timetable, which included an updated report from Mr Newman, and a joint report. It was not then expressly contemplated that the particulars of claim might have to be amended. Judge Bailey awarded the respondent 75% of the costs of the application in any event, but the appellant was awarded 25% in the case, ie if he won at trial. The total costs were summarily assessed at £7,000. There was no request for costs thrown away as a result of the adjournment of the trial, then imminent. One reason for the shorter trial estimate was that it was then agreed that the microbiologist experts need not be called.

On 12 November 2009 Mr Newman produced his second report. He now recast his opinion somewhat, and in particular added new allegations of negligence dating back to the original operation in July 2003 and encompassing fresh inadequacies of the debridement procedure of October 2003 and the second-stage revision of January 2004. He concluded:

“Mr Chadwick developed infection following a knee replacement which has resulted in him having suffered substantially for a number of years and being left with major permanent disability. Although no one specific element of his management has caused the infection, there are five aspects of his management which I believe have not been in accordance with good practice, all of which have contributed to increasing the risk of him developing major infective problems. These are:-

1.

Failure to provide adequate and timely prophylactic antibiotics at the time of his primary knee operation.

2.

Failure to institute aggressive surgical treatment at the time his infection was first recognised on 19.09.03.

3.

Failure to react to persistent infection when reviewed on 30.09.03.

4.

Undertaking a totally inadequate debridement procedure on 24.10.03.

5.

Performing the second stage revision arthroplasty on 22.01.04 which was too short a period after persistent problems with his leg and knee had been noted.

I believe that all these actions have increased the risk of an unsatisfactory outcome.”

To some extent Mr Newman was there building on or agreeing with criticisms raised by Professor Eycken.

As for complaint 4, Mr Newman accepted that aggressive early therapy only succeeded in the absence of a second replacement in a minority (20-30%) of cases. Nevertheless, his approach generally was that it was the accumulation of these defects which had caused, or increased the risk, of the appellant’s infection, ongoing pain and disability.

For these purposes, Mr Simon Butler, on behalf of the appellant, relies on what this court said about causation in Bailey v. Ministry of Defence [2008] EWCA Civ 883 at [46] (viz, “In a case where medical science cannot establish the probability that ‘but for’ an act of negligence the injury would not have happened but can establish that the contribution of the negligent cause was more than negligible, the ‘but for’ test is modified, and the claimant will succeed”). That was formally pleaded in further information given, pursuant to a request, on 15 March 2010.

Following Mr Newman’s second report, Mr Butler drafted amended particulars of claim so as to plead the whole of what Mr Newman was now saying. The draft amendment replaced the brief particulars of breach in their entirety and were now stated in 13 sub-paragraphs (a) to (m) at para 16. The earlier pleading on causation had incorporated passages which might have been argued to have had the effect of narrowing the plea, but these were now dropped, and the causation pleading was also extended with the intention of incorporating the Bailey approach, viz – “18. The failures particularised at paragraph 16 above added materially to the risk of an unsatisfactory outcome.”

There is no doubt that this was a wholesale revision of the pleaded case. However, the original case of breach was still retained as part of the newly pleaded particulars: see para 16 (j) (in part), (k), and (l). Thus –

“(j)

The delay in instituting surgery, opening of the old lateral wound [and applying a vacuum suction pump]” – this is new – “all contrived to increase the soft tissue damage around the knee. The damage caused major problems and was in part responsible for the need for a plastic surgical procedure at the time of the second two-stage revision.

(k)

It was inappropriate and/or unreasonable to re-implant on 22 January 2004, and the Defendant should have waited longer for the knee to settle further to be sure the recurrent infection was not occurring. Early intervention on 22 January 2004 increased the risk of a recurrent problem. The Defendant failed to recognise the risk, or ignored the risk, or alternatively recognised the risk but went on to take it.

(l)

The Defendant performed the second stage revision knee replacement without there being sufficient evidence that the infection had been eradicated. In the light of the history there should have been, shortly before or at the time of the second stage revision an aspiration biopsy and/or debridement and,or tissue sampling.”

Thus, although Mr Butler, correctly, accepted before Judge Knight that the amended particulars “differ significantly in their formulation”, and also correctly accepted that, but for the liberty to replace Mr Curry with Mr Newman as an expert which had already been granted, the appellant’s claim could not have succeeded, he did not accept that, without permission to amend being granted, that claim would have failed.

Moreover, although the draft amendment, both as to particulars of breach and as to causation, raised new matters, the first being of fact and the second being essentially of law, everything arose out of substantially the same factual situation, which had already been covered in previous pleadings and reports.

In those circumstances, and where the trial date had to be vacated as a consequence of the amendments a second time, and the estimate for trial was now extended again, and the respondent was not a well man, and had had this matter hanging over him for a number of years, there was obviously available an argument, I am not necessarily saying a good argument, that might have led a judge, in the exercise of his discretion, to say that permission to amend should be refused, or refused in large part, and the two-day trial which had been refixed for 25 January 2010 should stand or fall by the original pleadings, or at any rate as amended only in such a form as might have allowed the date to be maintained, if that was possible.

The question, however, is whether the judge was right to say that, although permission to amend should be granted and the trial date vacated, that should be at the cost of the appellant paying all the costs to date (other than the costs covered by Judge Bailey’s order). He did so, not essentially as a matter of discretion, but for the two reasons which he stated in his judgment, namely (a) that there was a substantially new case on breach and causation; and (b) that the old case on causation would have failed (see at para 1 above). The judge himself acknowledged this when he went on to discuss his permission to appeal, emphasising that “the two points are short” and that “those are the two reasons I rely on” (proceedings at 26).

However, some months later on 30 April 2010 he was to state that permission to appeal was granted because “if discretion was wrongly exercised it would have important costs consequences” and he also sought to add to the reasons which he had given at the time, viz –

The judge’s judgment (see para 1 above) referred to the case of “Bowman”. No one thinks that is a correct reference. In an email to the Civil Appeals Office dated 28 June 2010 he said that he thought that the reference ought to be to “Beoco”. Although Beoco had been relied on before the judge by Ms Katie Gollop on behalf of the respondent, and he may well have had that case in mind in making the order he made for the reasons he gave, nevertheless I do not think that he intended to refer to Beoco as distinct from Bailey: for only Bailey and not Beoco is concerned with the “correct causation test” and only Bailey makes sense in the sentence in which the judge deployed the reference to “Bowman”.

Beoco is not concerned with causation. The reference is to Beoco Ltd v. Alfa Laval Co Ltd [1995] QB 137. That concerned a contract case where an amendment had been sought at trial. The amendment had been granted, and by the conclusion of the trial the plaintiff had failed on his original case and had succeeded only on his amended case. The trial judge nevertheless ordered the defendant to pay the costs of the action save for the costs attributable solely to the original case. The defendant appealed against leave to amend being granted, and in any event against liability on the new case. The defendant also appealed against the costs order. This court held: (i) that leave to amend had properly been given, since the defendant could be protected in costs; (ii) that the defendant was liable on the plaintiff’s amended case for far less than the judge had awarded; and (iii) that the judge had erred in any event in giving the plaintiff too much in costs. I am concerned with the last holding. As to that, Stuart-Smith LJ said (at 154B):

“As a general rule, where a plaintiff makes a late amendment as here, which substantially alters the case the defendant has to meet and without which the action will fail, the defendant is entitled to the costs of the action down to the date of the amendment. There may, of course, be special reasons why this general rule should not be applied. An example of this is to be found in Kaines (U.K.) Ltd. v. Osterreichische Warrenhandelsgesellschaft (formerly C.G.L. Handelsgesellschaft m.b.H.) [1993] 2 Lloyd’s Rep. 1, 9, where the judge was satisfied that, even if the amendment had been made earlier, the action would have been vigorously resisted.”

The questions therefore are (i) whether the appellant’s amendment “substantially alters the case the defendant has to meet” and (ii) whether without the amendment “the action will fail”. I am prepared to assume that the appellant’s amendment fulfilled the first condition, although I would observe that this case presented nothing like the situation in Beoco, where the amendment was put on a totally different ground. In the present case, the breaches of duty were much expanded, but the original case was preserved. However, would the appellant’s action have failed without the amendment? Such a question is easily answered in a situation like Beoco where the original and amended pleas are entirely separate, the amendment is made at trial and the relevant costs order is made at the conclusion of trial; or in any case where, even if the amendment is made at a late stage before trial, the relevant costs order is made at the conclusion of trial. It can also be easily answered where the amending party accepts that he cannot succeed without his amendment, or where the court is capable of providing a summary answer before trial as to the viability of the unamended case.

However, where a judge is asked to make such a summary decision, he must be careful that he does so with due circumspection. Moreover, if trial is in any event imminent, he needs to bear in mind that the trial judge will be in a much better position to determine the question post-trial than a case management judge can do pre-trial. It may nevertheless be the case that the case management judge can both safely determine the summary issue, if asked to do so, and feel it necessary to do so, in the interests of justice, fairness to the parties, and the efficient allocation of resources.

In the present case, however, there was no concession by the appellant that he could not succeed without the amendment. Moreover, at the time when Judge Bailey was asked to permit Mr Newman to replace Mr Curry, embarrassing as such an application was for the appellant to have to make to the court, there was no submission from the respondent that the application should not be permitted because the claim could not in any event succeed. Nor did Judge Knight conclude that the original pleaded case of breach of duty could not succeed, only that the original case of causation could not succeed. This was, as I understand it, because Mr Newman in his second report which was before Judge Knight accepted that early and aggressive treatment of an infection by way of debridement only avoided a second replacement in 20-30% of cases. However, that was not an end to the case as originally pleaded; and in any event, a relatively small amendment, to introduce the Bailey causation issue (essentially an amendment of law, even if it had consequences for the factual enquiry), arguably side-stepped that difficulty, in a way which Judge Knight conceded that he could not determine (“Those are probably issues to come…”).

Ms Gollop submits nevertheless that I can and should take into account a concession that the appellant has made since Judge Knight’s judgment. Thus by a consent order dated 10 September 2010 the appellant agreed that paragraph 16(l) of the amended particulars of breach of duty should be struck out forthwith. That was the particular of breach (see at para 16 hereof above) which complained that the second-stage revision had gone ahead in January 2004 without sufficient evidence that all infection had been eradicated, and had been re-pleaded in the self-same words as one of the two main complaints which had gone back to the time of the original claim. Ms Gollop observes that a reason for this concession, although it only came later, was already to be found in Mr Newman’s second report which was before Judge Knight (“I therefore do not think it was negligent to have failed to undertake an aspiration biopsy”). Moreover, it had subsequently become common ground that there had been a biopsy sampling of tissue at the time of the second-stage revision. Ms Gollop therefore submits that if Mr Newman’s concession about the unlikely effectiveness of early aggressive debridement (the 20-30% point) is added to the consent order’s concession as to the need for biopsy evidence that infection was not present before the second-stage revision proceeded, then there is nothing left of the unamended claim and the judge’s approach can be seen to be vindicated.

Ms Gollop presented her submissions most attractively, but ultimately they did not persuade me. A significant part of the unamended pleading was translated into the amended pleading, now contained in particulars (j) and (k), and survives; and despite the difficulty for the argument on causation produced by the 20-30% concession (which the new pleading on causation is designed to meet), I do not think it is possible at this stage, summarily, to calculate that the whole of the costs of the action were wasted until the amended pleading came along. That would include, for the sake of a small example, but it tests the point, the costs of the claim form itself; it would also include the costs of instructing Mr Newman for the purposes of his first and second reports – which may ultimately, on the judge’s own logic, take the appellant to safety and success.

Therefore the judge’s solution may work injustice. Not only that, but there is no reason to think that the judge at trial will be unable to make a costs order which will properly take into account the ultimate result and its logic. If the appellant fails completely at trial, he will presumably pay all the costs. If he succeeds, but solely on the new particulars of breach, the judge may well be entitled to make a costs order which sanctions the appellant in costs up to the time of the amendment. If, however, the appellant were to succeed at trial at least in part because of an allegation of breach of duty which goes back to the original pleading, then the judge may wish to take that into account in a more nuanced way.

In conclusion, I think that the judge erred, not in a mere matter of discretion, but in principle in thinking that he could deal with the matter of costs on the basis that the claim up to the time of the amendment was worthless and could be summarily dismissed: when the real solution was to allow the outcome of trial, which he was prepared to acknowledge must take place despite its adjournment, to deal with the just allocation of costs.

In my judgment, the correct order for costs below was therefore that the appellant should bear the costs of and caused by the amendment, including or in addition any costs thrown away by the need to adjourn the trial on the two occasions on which it was adjourned (to the extent, as appears to be the case, that the costs of the first adjournment were not dealt with by the order of Judge Bailey). I will deal on paper with any submissions as to the order for costs to be made on this appeal, presumably to the extent that they have not been dealt with by the court of appeal on the abortive attempt to appeal to it.