I have handed down judgment in this case and I must now decide the appropriate order for costs. As is apparent from my judgment, Edwards sought revocation of Cook’s patent EP (UK) 1 255 510. Cook defended the claim and counterclaimed for infringement. I have found the patent is invalid and not infringed. Accordingly, the claim has succeeded and the counterclaim has failed. Edwards is the commercial winner and has defeated an attempt by Cook to have its SAPIEN product removed from the market. In these circumstances Edwards asks for its costs of the action.

Cook does not dispute that Edwards is entitled to an order for costs in its favour but says I should apply a discount to reflect the fact that Edwards relied on a number of pieces of prior art which it either abandoned or upon which it failed. It also says that Cook’s costs are too high for a case of this length and complexity.

The principles which I must apply are well established and set out in CPR 44.3 which provides, so far as relevant:

I would add the following observations. First, as Jacob LJ explained in SmithKline Beecham Plc v. Apotex Europe Ltd. (No. 2)[2005] FSR 24, an issue by issue approach is one that should be applied so far as it reasonably can, but such an approach is not the be all and end all. Whether or not it was reasonable for a party to raise, pursue or contest a particular allegation remains a relevant fact to be taken into account as part of the conduct of the parties. Second, the exercise upon which the court must engage is necessarily imprecise as Jacob LJ himself recognised in the SmithKline case. Third, one of the matters to which the court must have regard is proportionality. Finally, the court has a wide discretion to further the overriding objective.

In applying these principles to the present case, the following matters appear to me to be of particular relevance: first, as I have indicated, Edwards is the successful party. I consider it also relevant that the consequence to Edwards of failure would have been extremely serious. Its commercial product, the SAPIEN, was threatened with removal from the market. By contrast, Cook does not manufacture or sell a heart valve and there is no evidence before me that it brought this action to protect a market position.

Second, Cook originally maintained that all 34 claims of the patent were independently valid. When ordered by the court to serve a revised set of independently valid claims it reduced the number to 25. It persisted in this position up to the point of exchange of experts’ reports but, at the last moment and after Edwards’ experts had substantially completed the preparation of their reports, it reduced the number still further to 10. In the end I considered a total of nine claims at trial. I have no doubt that the failure by Cook to address the question of independent validity of the subsidiary claims more purposefully at an earlier stage has caused an unnecessary increase in the costs incurred by Edwards.

Third, Cook persisted in seeking disclosure of documents relating to the development of the SAPIEN. It was explained to Cook that such documents could have no possible relevance to the case, not least because the developers of the SAPIEN had no prior experience of designing heart valves. Nevertheless, Cook maintained these documents were relevant to the allegations of insufficiency, obviousness and lack of novelty. A hearing before Mann J did not resolve the issue and he directed Cook to provide a statement of case or a skeleton argument setting out the relevant disclosure issues. Pursuant to that direction, Cook sent a letter to Edwards dated 13 March 2009 contending that documents relating to any consideration of Andersen and its limitations, the valve stent connection and the choice of valve material should be disclosed. In response Edwards indicated by letter dated 19 March 2009 it would search for documents in the first two categories but maintained it did not believe them to be relevant to the proceedings because they could not be said to be representative of documents created by the notional skilled team.

In the following weeks Edwards provided to its solicitors over 10,000 documents and they set about completing the disclosure exercise at an overall cost of some £90,000. At trial Mr. Stanton Rowe of Edwards provided a witness statement. He was made available for cross-examination and was in a position to speak to the documents. In the event, however, Cook decided not to cross-examine him and not one single document was referred to.

It is quite clear now that this whole disclosure exercise has been a monumental waste of time and money. It provides a salutary reminder that disclosure in patent actions costs a great deal and is rarely of any use. Plainly, Cook must bear the costs but should it do so on an indemnity basis? Edwards urges me to say “Yes” because the facts of the case and the conduct of Cook are such as to take the situation away from the norm.

Not without some hesitation I have reached the conclusion that an order for indemnity costs is not appropriate. It is true that Cook persisted in requesting disclosure after the hearing before Mann J but it did so in accordance with the direction of the judge by setting out the issues to which it believed the documents were relevant and having considered the matters raised by Mr. Rowe in his witness statement. Edwards then acceded to the request, albeit under protest. In these circumstances the fact that such a large number of documents were subsequently produced by Edwards and examined by its solicitors is not in itself a reason for ordering indemnity costs against Cook. Nor is the fact that Cook did not deploy any of the documents at trial. No doubt it considered the documents and formed the view that they did not assist its case.

The answer, as it seems to me, is for the court and the parties to be much more robust in ensuring that this kind of futile exercise is not engaged upon in the first place and that any disclosure is both necessary and proportionate. In this case it was neither.

Fourth, Cook applied to amend the patent under the centralised procedure of Article 105A EPC. Specifically it did so by introducing the requirement that the valve material must comprise pericardium. That amendment was granted by the EPO on 28 November 2008, the date the experts’ reports were due to be exchanged. In the meantime, Edwards had requested Cook’s acknowledgement that the patent as granted was not being asserted. But that acknowledgement was not forthcoming. Accordingly, Edwards says, its experts had to prepare their reports both in relation to the original and proposed amended claims – another unnecessary and wasteful exercise. I do not think there is much in this point. The limitation to pericardium was always in a subsidiary claim and Edwards maintained its attack based on all of the art after the date of the amendment.

Fifth, Edwards certainly did not win on all issues. Of those fought through to the end of the trial it lost the novelty attack based on Thorpe and the obviousness attack based on Moll and the common general knowledge. Further, during the course of the trial it abandoned another novelty attack based on an item of prior art referred to as Bailey and another obviousness attack based on an item of prior art referred to as Block. Edwards contends that almost all the work on Thorpe and Bailey was done before the amendment was allowed by the EPO on 28 November 2008. But, for the reasons I have given, I do not think that this is a matter to which it is right to attach any significant weight.

Edwards estimates that it incurred about 10% of its costs in relation to the issues on which it lost and arrives at that estimate taking into account the costs incurred in relation to Thorpe and Bailey before 28 November 2008. Cook, not surprisingly, puts its costs incurred in relation to the issues on which Edwards lost rather higher, at about 35%. I have reached the conclusion that it is appropriate to deduct a proportion of Edwards’ costs to reflect its failure on the issues to which I have referred. Having considered all of the evidence before me and my own impressions of the case, I think a fair deduction is 20%.

Finally, I must refer to the fact that the total costs incurred by Edwards in defending this action are estimated at in excess of £2,360,000. By contrast, the costs incurred by Cook are estimated at about £1,540,000. I have to say that both these costs estimates seem to me to be extraordinarily high for a patent action of no great complexity lasting only seven days in court and the costs estimate provided by Edwards particularly so. I appreciate that the case was of great commercial significance to Edwards and I must also make allowance for the disclosure exercise to which I have referred. Nevertheless, I have formed the view on the materials before me that a proportionate gross, that is to say, pre-assessment figure for Edwards’ costs would not be in excess of about £1.5 million. The difficulty in arriving at a precise figure prevents me from imposing a costs cap, but I hope the costs judge will give due consideration to this expression of my opinion when the costs come to be assessed.

In all the circumstances I conclude that a fair and just result would be achieved if Edwards recovers 80% of its costs to be assessed in the light of the matters to which I have referred and I will so order.

I must now consider what sum to award by way of an interim order for costs. In arriving at this figure I must consider what is likely to be the irreducible minimum that Edwards will receive upon assessment.

I believe that an appropriate figure to award in the light of all the matters to which I have referred is £650,000.