Edwards Lifesciences Corporation & Anor v Meril GMBH & Anor

I now need to deal with the first issue in the non-private, in the public part of this case, concerning the consequences from my judgment that was handed down in September in this action ([2020] EWHC 2562 (Pat)). It is the question of costs. The first point is being dealt with separately because until one knows the answer to it, it really is not sensible to have further submissions from either party. The question is the basis on which the costs order will be made, and that in turn turns on the familiar question of who won the case. It is now traditional at the end of every patent case to have a debate, which seems to take a remarkably long time, to find out who won.

To be fair, the reason this matter is important is because it can have a profound effect on very substantial amounts of money, that is to say the costs of patent litigation. Also, to be fair, patent cases in particular can raise some tricky issues on this topic when, as in this case for example, more than one patent was tried together and the two patents are legally and factually distinct to a significant extent; although in saying that I am not pre-judging issues where there is overlap in the costs. This case is a good example because one patent is about 19½ years old and the other is about 10 years old, and they relate to different aspects of this technology. One patent relates to the valve, the other relates to the delivery system. Most important of all the outcomes at the trial are completely different. For the '753 patent it is fair to say that Meril won; and on the other patent, that is to say the '929 patent, although Meril won in part, it is fair to say Edwards won.

The question is whether that statement that I have just made is the governing conclusion for the decision on costs or whether, as Mr. Purvis submits, I should stand back and make an assessment of who the overall winner in the proceedings was. He submits that the answer to that question is that it was his clients and, therefore, the basic order that should be made is a single order essentially starting from the premise that there is to be a payment by Meril to Edwards. Then one applies the familiar principles on issue-based costs orders in patent actions which may involve deductions from Edwards’ costs, including notional deductions where a deduction is made as a notional payment to Meril for various aspects.

Mr. Purvis refers to the familiar authorities which start with the words of Lord Bingham in Roache v News Group Newspapers [1998] EMLR 161 at 168 referred to by Floyd J in Omnipharm v Merial [2012] EWHC 172 (Pat) and then also picked up by me in Merck v Ono [2015] EWHC 3973 (Pat). The Merck v Ono case is particularly instructive because it was a case in which there were two patents in issue and the outcomes on the two patents were the opposite; in other words, one party won on one patent and the other party won on the other patent. In fact, in Merck v Ono the result on one of the patents had effectively happened before the trial, but that makes no difference. It was clear that one way of looking at it, just as it is in this case, was victory for one side on one patent and for the other side on the other one. Nevertheless, in that case I decided that the right approach was to consider if an overall winner could be identified and if it could then to do so. That is what I did, and the consequences flowed from it (see the judgment as a whole (paras 1-20) and in particular paragraphs 12, 13 and 17).

In my judgment, the principle, which is nothing more than what I said in Merck v

The question is what to do in this case. One aspect of the matter is an argument about whether this is really about two separate devices or one device. The point is that one patent relates to the valve and the other relates to the delivery system.

Mr. Purvis says, however, that one cannot sell one without the other, and therefore even though the valve part of Meril's device has been found not to infringe any valid claims because the '753 patent is invalid, nevertheless as the evidence shows, the product in its current form cannot be sold because the delivery system part in its current form infringes the valid claims of the '929 patent.

Mr. Turner, on the other hand, points out that there is a way of selling the Myval valve with a delivery system which does not infringe the '929 patent and that is using device A, which has no flex indicating device at all. I think it was agreed even before trial that device did not infringe the '929 patent at all. So, Mr. Turner is entitled to point out that his clients have technology which would allow them to sell the valve and delivery system without infringing '929.

On the other hand, Mr. Purvis points out that as matters stand the CE mark approval which is the basis on which Meril are starting to conduct trials in this jurisdiction, does not encompass the device A product. Therefore the only product which can currently go in the clinical trials would be prevented from sale by the '929 patent; and that of course would prevent the delivery system but also effectively the valve.

Mr. Purvis submits that standing back and looking at it overall this is a case in which one can say that Edwards is the overall successful party, albeit that the outcome on '753 will have to be taken into account on assessment.

Mr. Turner, on the other hand, says that this is all happenstance, the '929 patent came late in the day, only three months before the trial of what was then fixed for '753, and if '929 had never come into that case and been dealt with separately then his clients would have simply had a trial on '753 which they would have won and they would have had their costs and it cannot be said otherwise and so it would be quite unfair to prejudice his clients in that respect.

It seems to me two things overall are important. It is important to have in mind that this discussion is all about costs. This is not some PR exercise about who won. It is about analysing the appropriate order for costs and using an assessment of the overall success or failure of parties in the proceedings as a tool for reaching an assessment of a fair costs order; nothing more and nothing less.

The other matter, which is part of an answer in my judgment to the point made by Mr. Turner, is that experience shows -- and this case will be no different -- that there are significant common costs in actions even though there are distinct issues. Part of the answer to the point about '929 coming in late is that in fact the trial did involve both '929 and '753. It was not a trial only of '753. Some of the costs relating to that trial

The question at the end, therefore, is can I find who is an overall winner from the point of view of carrying out a costs assessment? In my judgment, I can and the answer is Edwards. The result of this trial means that the current Meril product in its current form cannot be sold in this jurisdiction because it infringes the patent rights of Edwards. Therefore, it is fair to say that Edwards is the overall winner of these proceedings.

Nevertheless, in reaching that conclusion, I want to emphasise that does not mean that the fact that Meril won on the '753 patent, which they clearly did, is not going to be an important aspect of the costs assessment. The right approach, in my judgment, in awarding the costs of this action, is to start from the premise that there is to be a payment to Edwards of its assessed costs subject to deductions which may be very significant and they could even be so significant that it might even end up with a

This is my decision relating to the apportionment of costs in this case. I have already dealt with the first principle to be sorted out, which is the basis on which this decision is being made and I have decided that it has been made on the basis that Edwards is the successful party and I am going from there.

Edwards contends that the right order to make in these circumstances is that they should be awarded 90% of their costs and that the only deduction, of about 10%, is one that should be made given that they lost on the obviousness of the '753 patent. Meril contends that is not right and that much more substantial reductions should be made overall for various reasons.

Edwards's total costs in these proceedings in very round figures, excluding costs relating to the '457 patent, come to a total of about £2.5 million, that is for both '753 and '929 together. Meril's, by contrast, come to a somewhat lower figure of £1.5 million for '753 and '929. That difference is a matter I will return to in due course.

In order to carry out this exercise, the first question I have to resolve is the relative share of the costs of Edwards for the '753 patent, on which essentially speaking Edwards did not win, and the '929 on which essentially speaking Edwards did one.

I have the information from Meril. Meril's £1.5 million total is made up of essentially in very round figures £900,000 from '753 and £600,000 for '929. By contrast, according to the evidence of Mr. Kusumakar, the solicitor for Edwards, his estimate is that 60% of Edwards's costs were on '929 and 40% were on '753. Meril are sceptical about that and submit, particularly bearing in mind that the '929 patent only came into the case relatively late in the day, it is inherently unlikely that that is the true split.

A point made by Edwards, which I believe has some force, is that a significant amount of costs was spent before trial in the '929 case relating to prior uses called Retroflex, which even though they did not in the end figure greatly at trial, were responsible for a significant amount of costs.

Nevertheless, it does seem to me there is something in Meril's point. I am surprised at the thought that as little as 40% of the overall costs are attributable to '753. My impression of this case is that is too low a figure overall. Doing the best I can in the circumstances, I find that a fair approach is to attribute 45% of the costs of Edwards's costs to '753 and 55% of the costs to '929. I bear in mind of course that there will be a very significant body of costs which are common and not attributable either way, but that is not relevant to the purpose for which I am doing this exercise, which is to allow me to attribute costs as between these two different parts of the case and come to a

Turning to '929 itself, as I said I have come to the view that about 55% of Edwards's costs related to the '929 patent. Edwards's position was there should be no deduction for '929 at all. Meril, on the other hand, submit that there were suitably circumscribed issues on which Meril succeeded on and which should lead to some deduction from Edwards's total. It bears pointing out that the case on infringement was found only on equivalents and not on a literal basis and that claims 1 and 2 of the '929 patent were found to be invalid over one of the embodiments in the Marchand prior art. It is also fair to point out, as Mr. Purvis does, that the embodiment over Marchand which was the successful one in the end was not one which had been addressed in any detail at all, if at all, in the evidence. That is a reason why the amount of costs attributable to that issue, even though it is one that Meril won, should perhaps not be as large as they might otherwise have been.

Meril's figures were on the basis that the literal infringement part was worth about a 7.5% deduction and that the aspect of validity, for which a deduction should be made, makes a total of about 8%, which would be roughly 15% off the '929 costs.

Standing back, in my judgment, overall, a fair figure to deduct from the costs of '929 is about 10% of those costs. I bear in mind that I then have to multiply that by 55% to get an overall deduction, but that is easy to do. That means I will reduce the overall sum due to Edwards from Meril because of the issues related to the '929 patent by 5.5%.

I now turn to consider what to do about the issues relating to the '753 patent. It is quite clear in my judgment that some deduction should be made for '753. As I have already pointed out, Edwards were prepared to accept a 10% deduction off their costs related to the fact that they lost on obviousness on '753.

A point which Meril makes, which they are entitled to make, is that the obviousness costs relating to '753 were not just concerned with Cribier, one of the pieces of prior art, but involved Cribier and Andersen together. I do not think that was quite taken into account in the way Edwards did the sums; but even if it was, that is an important dimension anyway. It is also relevant to take into account that significant aspects of the common general knowledge, and by my recollection significant part of the cross-examination related to the common general knowledge, related only to the '753 patent, although other parts of the common general knowledge of course were common to both.

It is also right, as Mr. Turner points out, that some aspects of insufficiency to the extent they caused any costs, which I suspect were rather low, were things which went in Meril's favour.

It also bears pointing out that essentially Edwards lost on the '753 patent, although, of course, in losing on the '753 patent it did succeed in various aspects; but the fundamental position is that patent was found to be invalid.

As a deduction from Edwards's costs, it seems to me the fair proportion to take of the costs of '753 which should be removed is half of those costs. I will therefore deduct 22.5% from the headline figure in order to deduct from Edwards its costs relating to the issues on which it did not succeed on '753.

The next question is whether I should make a further deduction effectively by treating as a deduction from Edwards a sum which is to be, in effect, treated as Edwards paying Meril for the costs relating to at least some of the issues on which Meril succeeded on '753 in the overall assessment. What I am not prepared to do is to order two assessments, one of Meril's costs and one of Edwards's costs and, as has been done in many patent cases in the past, a fair and proportionate way of dealing with that is to make a further deduction if one thinks it is appropriate by taking a further sum off the costs of the successful party scaled by reference to the costs of the paying party.

In my judgment, it is appropriate in this case to make a further deduction in this way in relation to the costs of validity particularly of the '753 patent and I will do so. It seems to me again that the right proportion is half, but the half is a half of Meril's costs, not Edwards's costs. Meril's costs in very round figures were £900,000 and that means the figure in money would be £450,000; expressed as a percentage of Edwards's total of 2.5 million I make that minus 18%.

Therefore, the total is the aggregate of a deduction of 5.5, 22.5 and 18%, which by my maths is 46%. I will therefore, assuming I have the maths right, award Edwards 54% of its assessed costs of the action to be paid by Meril. That is my decision.