T 0610/95 (Dressing/SQUIBB) of 21.7.1999

Summary of Facts and Submissions

I. The appellant is the proprietor of European patent No. 0 092 999 comprising 15 claims. The patent was granted on the basis of European patent application No. 83 302 316.1, filed on 22 April 1983, and divisional application No. 88 201 241, filed on 20 June 1988. Claim 1 reads as follows:

"An occlusive multi-layered dressing [10, 30, 50] comprising an adhesive layer [14] which, in use, contacts the wound and surrounding normal skin, an intermediate layer [12] of semi-open-cell polymeric foam, and an outer moisture impervious polymeric film [11] coated or laminated to the upper surface of said foam layer [12], wherein said wound and skin contacting adhesive layer [14] consists essentially of a homogeneous blend of from about 35% to about 50% by weight of one or more low molecular weight polyisobutylenes which act as pressure sensitive adhesive materials and from about 45% to about 65% by weight of one or more water dispersable hydrocolloids selected from sodium carboxymethylcellulose, calcium carboxymethylcellulose, pectin, gelatin, guar gum, locust bean gum, collagen, and gum karaya."

Dependent claims 2 to 5 are directed to specific elaborations of the dressing according to claim 1.

Dependent claim 6 is worded as follows:

"The dressing of claim 1 wherein said wound and skin contacting adhesive layer [14] is bonded to said open-cell polymeric foam by a second adhesive layer [13] and said second adhesive layer [13] consists essentially of a homogeneous blend of from about 30% to about 70% by weight of one or more pressure sensitive adhesive materials and one or more optional thermoplastic elastomers, from about 10% to about 65% by weight of one or more water dispersable hydrocolloids and up to 50% by weight of one or more optional water swellable cohesive strengthening agents and/or one or more optional hydratable polymers provided that said water dispersable hydrocolloids, water swellable cohesive strengthening agents, and hydratable polymers together are present at no more than about 70% by weight of said adhesive layer, from about 5% to 15% by weight of a plasticizer or solvent, and from about 15% to about 25% by weight of a tackifier."

Dependent Claims 7 to 14 are directed to specific embodiments of the dressing according to claim 6.

Claim 15 relates to a method for preparing a multi-layered dressing according to claim 6.

II. The respondent (opponent) filed an opposition against the grant of the patent on the grounds that the subject-matter of claims 1 to 4 was not patentable under Article 100(a) EPC, because of lack of novelty and inventive step. The original opponent (respondent) died during the appeal proceedings which were thereafter continued by his widow who is also his executor.

III. The respondent's objections are essentially based on the following citations:

(1) US-A-3 972 328

(2) EP-A-0 026 572

(3) US-A-3 339 546

(4) US-A-4 192 785, patent family member of FR-A-2 393 566

(5) Current Medical Research and Opinion, Vol. 1, No. 10, 1973, pages 603 to 604

(6) Ars Medici, 34, 1979, 734-752

(6a) English translation of (6)

(7) Nursing, 76, April, page 13

IV. In its decision notified on 27 June 1995 the opposition division reached the conclusion that claim 1 of the opposed patent did not involve an inventive step, contrary to the requirements of Article 56 EPC and, accordingly, decided to revoke the patent pursuant to Article 102(1) EPC. The substance of its reasoning was as follows:

The dressing defined in claim 1 of the patent in suit differed from the dressing disclosed in (1) by the composition of the wound and skin contacting adhesive layer [14]. While this layer [14] consisted in the patent in suit of a blend of from about 35% to about 50% by weight of low molecular weight polyisobutylenes and from about 45% to about 65% by weight of certain water dispersable hydrocolloids defined more precisely in claim 1, the corresponding adhesive layer [11] was described in citation (1) as containing, in addition to a rubbery elastomer (eg polyisobutylene or a mixture of polyisobutylenes) and a hydrocolloid, a tackifier and a plasticiser or solvent.

Although the minimum proportions specified in claim 1 for both components of layer [14], ie the low molecular weight polyisobutylene (35% by weight) and the water dispersable hydrocolloids (45% by weight), added up to a total amount of 80% by weight only, the language of claim 1 of the contested patent could not, on the basis of the disclosure of the invention in the paragraph bridging pages 2 and 3 of the description, be interpreted as including in layer [14] a tackifier and/or a plasticiser, at least not at the levels used in citation (1).

Citation (2) disclosed a dressing comprising a wound and skin contacting adhesive layer [B], which had the same consistency and composition as layer [14] of the claimed dressing in the contested patent, and also an outer moisture impervious polymeric film coated to the upper surface of an intermediate foam layer. The apertures [20] extending through the layer [B] of the curative and absorbent material [11], which contacts the wound in (2), indicated, however, that (2) related to a non-occlusive rather than to an occlusive dressing.

Consequently, neither citation (1) nor (2) was prejudicial to the novelty of the claimed dressing in the patent in suit.

Starting from citation (1) as the closest state of the art, the technical problem could be seen as that of providing an improved dressing that adhered as strongly as the known dressing to the wound and the surrounding skin but was more readily removable without running the risk of re-injuring the wound when the dressing was removed. It was obvious to a person skilled in the art to solve this problem by simply replacing the adhesive layer used in citation (1) by an improved commercially available adhesive material sold under the tradename Stomahesive, which was specifically shown in both citations (3) and (7) to exhibit good healing properties and concurrently good adhesion to the wound and surrounding skin, while it was easily removable due to the absence of both a tackifier and a plasticiser in the adhesive layer.

Alternatively, starting from citation (2) as the closest state of the art, the problem was that of providing a dressing which was occlusive and showed good healing properties and good adhesion to the skin surrounding the wound. It was similarly obvious to a person skilled in the art to solve this problem by plugging the apertures [20] extending through the layer [B] of curative and absorbent material [11] used in citation (2), so as to obtain a continuous adhesive layer instead forming a closed moist wound treatment environment and, thus, to arrive at the invention.

V. The appellant lodged an appeal against the decision of the opposition division and requested in the statement of grounds that the impugned decision be set aside and the patent be maintained unamended.

In the course of appeal proceedings the appellant filed auxiliary requests I to VI:

V/A Claim 1 of auxiliary request I differs from claim 1 as granted by the insertion of the numerical range of the viscosity average molecular weight ["from about 36 000 to about 58 000 (Flory)"] of the polyisobutylenes acting as the pressure sensitive adhesive material and binder in the skin contacting adhesive layer [14].

Claims 2 to 15 correspond to those of the patent as granted.

V/B Claim 1 of auxiliary request II differs from claim 1 as granted by the insertion of

(i) the wording "which, in use, results in a closed moist wound treatment environment" as an additional technical feature characterising the function of the occlusive multi-layered dressing [10, 30, 50] in the preamble of claim 1, and

(ii) the wording "forming a fluid-tight bond with the healthy skin around the wound so as to seal the dressing to the skin" as an additional technical feature characterising the function of the adhesive layer [14].

Claims 2 to 15 correspond to those of the patent as granted.

V/C Claim 1 of auxiliary request III differs from claim 1 as granted by the insertion of both the amendment to claim 1 suggested in auxiliary request I ["from about 36 000 to about 58 000 (Flory)"] and the amendments (i) ["which, in use, results in a closed moist wound treatment environment"] and (ii) ["forming a fluid-tight bond with the healthy skin around the wound so as to seal the dressing to the skin"] suggested in auxiliary request II.

Claims 2 to 15 correspond to those of the patent as granted.

V/D Auxiliary request IV differs from the claims as granted in several respects, more specifically

(i) by the limitation of claim 1 of the granted patent to an occlusive three-layered dressing, wherein the wound and skin contacting adhesive layer [14] is bonded directly to the bottom surface of the foam layer [12];

(ii) by the introduction of newly filed independent claim 5 including the features of claim 1 and dependent claim 6 of the patent as granted; and

(iii) by the addition of newly filed dependent claims 14 to 16.

V/E Claim 1 of auxiliary request V corresponds to claim 1 of auxiliary request IV with the sole exception that the numerical range of the viscosity average molecular weight ["from about 36. 000 to about 58 000 (Flory)"] of the low molecular weight polyisobutylenes has been introduced in claim 1 (see claim 1 of auxiliary request I).

Claims 2 to 17 of auxiliary request V are identical with claims 2 to 17 of auxiliary request IV.

V/F Claims 1 to 13 of auxiliary request VI correspond to claims 5 to 17 of auxiliary request IV.

VI. Oral proceedings were held before the board on 21 July 1999. The appellant's submissions both in the written procedure and at the oral proceedings can be summarised as follows:

Citation (1) concerned a wound dressing which was admittedly similar to that claimed in the patent in dispute except that the adhesive layer [11] of (1) comprised a mixture of lower and higher molecular weight polyisobutylenes, as well as high levels of tackifier and plasticiser. This disclosure was actually in contrast to the claims of the contested patent which required the adhesive layer [14] of the dressing to comprise from about 35% to about 50% by weight of low molecular weight polyisobutylene and from about 45% to about 65% of one or more water dispersable hydrocolloids, in the absence of tackifier and plasticiser. These requirements were not met by citation (1).

Citation (2) differed from the invention in that it contained a plurality of apertures and required a layer of deodorising material (D) and an outer layer (E) having an adhesive coating which secured the dressing to the body. Since the apertures extended through the curvature and absorbent layer [11], which was in contact with the wound, through the foam layer [13] and through any optional backings [12], [14] that may be present as well, citation (2) did not disclose an occlusive dressing as required by claim 1 of the patent in dispute.

Whilst citation (1) failed to give any hint or indication that the adhesive used in the known dressing caused an injury problem when the dressing was removed, it was in fact the Reilly declaration submitted by the appellant, which contained an extensive comparison between the dressing according to (1) and the dressing claimed in the patent in suit and revealed for the first time the problem of re-injury when the dressing was to be changed. Since this declaration was not available to the public prior to the priority date of the contested patent, the skilled man would not have been aware of the problems caused by the prior art of (1). The knowledge contained in the declaration was thus already part of the appellant's invention.

In considering citation (1) and having knowledge of Stomahesive (citations 3 to 7) a person skilled in the art could only conclude that the addition of tackifier and plasticiser to the prior art adhesive was required to hold the dressing firmly in place and Stomahesive per se did not have sufficient adhesion to meet the requirements of the dressing used in (1). One had also to assume that the common inventor of both citation (1) and Stomahesive considered the latter to be too weak an adhesive for the purpose of the dressing disclosed in citation (1). Whether in the end this assumption was right or wrong was irrelevant since the only question was what one was to conclude from the state of the art, and one could only conclude that Stomahesive was too weak an adhesive for the intended purpose.

In considering citation (2) and the question of whether the adhesive used therein would have sufficient stick to hold the bandage by itself, one could not avoid a negative conclusion, since layer [E] which had an adhesive coating which secured the dressing to the body was absolutely necessary. Hence, citations (1) and (2) clearly discouraged the skilled person from the modification proposed in the patent-in-suit.

In none of the citations was Stomahesive adhesive wafer applied alone without something else such as an elastic bandage [see (5)], gauze dressing [see (6)] or tape strips [see (7)] required for holding the Stomahesive adhesive wafer in place and securing it to the body. The conclusion drawn in the decision of the opposition division that citation (7) taught in connection with the treatment of patients suffering from stage 5 ulceration the possibility of applying Stomahesive without the need to employ some additional means such as an adhesive tape for securing it to the body, was the result of a clear misinterpretation of the state of the art. Thus, (7) referred in connection with stage 5 ulceration to the treatment of severe ulceration with undermining from one ulcer to another and chronic purulent drainage. Therefore, (7) taught, in reality, that in such severe cases of ulceration the specific use of an adhesive tape was to be avoided in order to prevent those portions of intact skin, which were already undermined by decubiti, from injury when the dressing was to be changed. This could not, however, be understood as excluding the need for using some other suitable means, such as a bandage, to hold the Stomahesive adhesive wafer firmly in place.

Moreover, as evidenced by the declaration of Peter C. Kallos, there had been several changes to the original Stomahesive formula which had altered the attributes of the resulting end products such that the trademark Stomahesive denoted a range of formulae rather than a single one and did not reveal anything about the composition of the product. The analysis of the products, in particular the adhesive mixtures of hydrocolloids and gum-like substances were, however, extremely difficult.

VII. The respondent disagreed and argued in the written procedure and at the oral proceedings in essence as follows:

Although the period available to the appellant for replying to the respondent's statement filed on 20. April 1998 was more than one year, auxiliary requests III to VI were filed only on 21 June 1999, one month before the date set for oral proceedings, and should therefore be disregarded as being late filed.

Citation (1) disclosed a three-layered occlusive wound dressing, the outer and intermediate layer of which were identical with the corresponding layers of the dressing in the patent in dispute. Since the lower limits of hydrocolloids and isobutylene present in the adhesive layer [14] added according to claim 1 to a total amount of 80% only, the claim could not be read so as to exclude the presence of tackifier and plasticiser in the adhesive layer [14] of the patent-in-suit. The content of (1) was therefore prejudicial to the novelty of the claimed dressing.

The outer layer [14], the intermediate foam layer [13] and the wound and skin contacting layer [11] of the dressing disclosed in (2) were, with regard to their composition and consistency, in fact identical with the corresponding layers of the dressing in the patent-in-suit. In spite of the fact that the wound contacting layer [11] contained a plurality of apertures, at least the outer flexible layer of the dressing disclosed in (2) was continuous and provided overall protection of the wound. Since such a dressing would similarly be considered to be occlusive, the prior art of (2) was likewise novelty-destroying.

Even if the board came to the conclusion that, in spite of the foregoing, novelty could be acknowledged, the patent would nevertheless have to be revoked on the grounds of lack of inventive step.

Citations (3) to (7) all referred to a product which had been on the market for more than 20 years under the tradename Stomahesive. Thus, for example, citation (5) described Stomahesive as consisting of gelatin (20% by weight), pectin (20% by weight), sodium carboxymethylcellulose (20% by weight) and polyisobutylene (40% by weight). The fact that the polyisobutylene used in Stomahesive was a matter of public knowledge and was indeed of low molecular weight had been confirmed by the appellant itself in it's letter filed on 4 October 1989.

The notional skilled person having realised that the dressing disclosed in (1) caused in certain cases a re-injury problem would have known, for example from citation (7), that Stomahesive was successfully used as a readily removable wound dressing for the treatment of severe ulceration and would, accordingly, in the first place consider solving the problem by modifying the adhesive layer [11] in (1) so as to correspond to the known composition of Stomahesive.

Similarly, it was obvious to a skilled person to modify the dressing disclosed in (2) by closing the holes in order to obtain a dressing providing a fluid-tight seal of the wound.

The appellant's argument that, on the basis of the disclosure in the state of the art referring to the necessity of Stomahesive being secured by means of a bandage or adhesive tape strips or in a similar appropriate way to the body, the skilled person would have considered Stomahesive to be too weak an adhesive for the intended purpose, was irrelevant. Apart from the fact that, on the one hand, (7) clearly taught the possibility of applying Stomahesive without using any additional adhesive means, such an adhesive tape, the scope of the claims of the patent-in-suit certainly did not, in the absence of a proper limitation, exclude dressings in accordance with the patent in suit, even if they were secured by means of an adhesive tape or a bandage or in a similar appropriate way to the body. Moreover, as had already been pointed out by the respondent in his letter filed on 22 July 1994, a number of documents submitted by the appellant itself in support of its argumentation recommended fixing additionally the claimed dressing in the contested patent with an elastic tape whenever the effect of mechanical stresses on the dressing was to be expected.

VIII. The appellant requested that the decision under appeal be set aside and the patent be maintained unamended (main request) or in amended form on the basis of one of the auxiliary requests I to VI submitted by fax on 21. June 1999.

The respondent requested that the appeal be dismissed.

Reasons for the Decision

1. The appeal is admissible.

2. Admissibility of the appellant's requests

As is apparent from paragraph V above, the text of the alternative sets of claims, filed by the appellant as auxiliary requests I to VI, was modified so as to incorporate a plurality of substantial amendments in the text of the granted patent. The first question to be decided is, therefore, whether such alternative sets of claims can be admitted into the proceedings.

2.1. The EPC does not guarantee a patent proprietor the right to have proposed amendments incorporated in opposition or subsequent appeal proceedings. According to the established jurisprudence of the boards of appeal, the admissibility of amendments to the text of the granted patent during such proceedings is a matter that is for the instance in question to decide in its discretion under Rules 57(a) and 58(2) EPC. To be admissible, proposed amendments should be "appropriate" and "necessary" having regard to the nature of the grounds for opposition and the issues raised thereby.

On the basis of the criteria laid down in decision T 295/87 (OJ EPO, 1990, 470, see especially reasons, point 3) amendments to the text of a granted patent during opposition or subsequent appeal proceedings should only be considered "appropriate" and "necessary" within the meaning of Rules 57(a) and 58(2) EPC and therefore admissible, if they can fairly be said to be occasioned by grounds for opposition laid down in Article 100 EPC.

The competent board emphasised in the cited decision that the opposition procedure provided for under Articles 100 to 102 EPC and the relevant Implementing Regulations, in particular Rules 57, 57(a) and 58 EPC, is designed to provide an examination of the validity of a patent on the basis of the objections to validity raised under Article 100 EPC. Opposition proceedings are not an opportunity for the patentee to propose amendments to the text of a patent for purposes which are not clearly related to meeting a ground of opposition raised under Article 100 EPC.

In particular, the addition of claims to the text of the granted patent during opposition or subsequent appeal proceedings, which have no counterpart in the granted version of the claims of the patent in suit, cannot normally be regarded as an attempt to respond to an objection under Article 100 EPC and is, therefore, not admissible (see T 295/87, especially reasons, end of point 3).

2.2. On the basis of the above considerations, the main request and auxiliary requests I to III are, in the board's judgment, admissible, while auxiliary requests IV to VI are not. The reasons for this finding are as follows:

(a) The main request, which is the sole request filed with the grounds of appeal, refers to the maintenance of the patent unamended in the form as granted.

(b) The proposed amendment to claim 1 of auxiliary request I (see paragraph V/A above) can fairly be regarded as an appropriate attempt on the part of the appellant to define more precisely the specific type of low molecular polyisobutylenes used in the contested patent so as to counter more effectively the respondent's objections to the validity of the patent on the grounds of lack of novelty and lack of inventive step over the prior art of citation (1)

(c) Both amendments to claim 1 of auxiliary request II (see paragraph V/B above) can be regarded as a suitable attempt on the part of the appellant to define more precisely the specific function of the claimed dressing, which is said to be occlusive in claim 1, and, similarly, to define more precisely the specific function of the adhesive layer [14] in the contested patent, so as to challenge more effectively the respondent's objection to the validity of the patent on the grounds of lack of novelty and lack of inventive step over the prior art of citation (2).

(d) The amendments to claim 1 of auxiliary request III (see point V/C above) are considered admissible and necessary in the sense outlined above for the reasons given in foregoing points (b) and (c).

The amendment to claim 1 of auxiliary request I is derivable from page 4, lines 6 to 9, of the application as filed; the first amendment [feature (i)] added to claim 1 of auxiliary request II is derivable from page 29, lines 1 to 3, the second amendment [feature (ii)] from page 2, lines 14 to 16 of the application as filed.

In view of the foregoing, the board judges auxiliary requests I to III not only admissible but also acceptable under the terms of Articles 84 and 123(2) and (3) EPC.

(e) As can be seen from a comparison of paragraphs I and V/D above, auxiliary request IV differs from the set of claims as granted by a number of substantial amendments referred to under items (i), (ii) and (iii) in paragraph V/D.

re (i):

With reference to the various major amendments introduced in the set of claims forming auxiliary request IV, as a preliminary point it should be emphasised that claim 1 is the sole independent claim opposed in the granted version of the claims and, consequently, on the basis of the observations set forth in point 2.1 (above), only an amendment to claim 1, such as the limitation of claim 1 resulting from amendment (i), could normally be said to arise from the grounds of opposition and could, therefore, possibly be considered admissible provided that the proposed amendment was appropriate and necessary and in compliance with the provisions of Articles 84 and 123(2) and (3) EPC.

re (ii):

By contrast, with amendment (ii), the appellant proposes a set of claims which incorporates in addition to amended claim 1 a newly-filed independent claim 5 resulting from the combination of the broader (unamended ) claim 1 and dependent claim 6 of the granted patent.

Since claim 1, which is the sole independent claim relating to a dressing in the granted version of the patent and which is concurrently the sole independent claim opposed, has been maintained in auxiliary request IV, it appears evident, in the board's judgment, on the basis of the principles set out in T 295/87 (loc. cit.), that the addition of new independent claim 5, which as such has no counterpart in the granted version of the claims of the patent in suit, cannot be regarded as an attempt to respond to an objection under Article 100 EPC. Further, during the hearing before the board, the appellant failed to provide a reasoned argument that the filing of new independent claim 5 was indeed necessitated by a ground of opposition and the issues raised thereby and, accordingly, that this amendment to the text of the granted patent was appropriate and necessary within the meaning of Rules 57(a) and 58(2) EPC.

In this respect reference should also be made to decision G 1/84 (OJ EPO 1985, 299, see especially reasons point 9). In that decision the Enlarged Board of Appeal made it clear that the opposition procedure is not designed to be, and is not to be misused as, an extension of the examination procedure. It would, in the board's opinion, contravene those principles set out in G 1/84, if it was considered admissible to amend the text of a granted patent during opposition proceedings, while maintaining the sole independent claim under opposition, by incorporating an additional new independent claim which as such has no counterpart in the granted patent and, accordingly, was neither the subject of substantive examination in the examination procedure nor open to opposition owing to its non-existence in the granted patent.

re (iii):

As regards amendment (iii), newly filed dependent claims 14 to 16 which depend on claim 5 concern preferred embodiments of the occlusive multi-layered dressing according to newly filed independent claim 5 which requires that the wound and skin contacting adhesive layer [14] is bonded to the open-cell polymeric foam [12] by a second adhesive layer [13]. This specific embodiment of the invention claimed in new claim 5 was mentioned in dependent claim 6 of the granted patent. Since claim 6 of the granted patent contains no reference other than to claim 1, (see paragraph I above) new dependent claims 14 to 16 which combine the features of claims 2 to 4 as granted with those of new independent claim 5 have no counterpart in the granted patent and, therefore, cannot be regarded as an attempt to respond to an objection under Article 100 EPC. The above mentioned criteria set out in decision T 295/87 (loc. cit.) fully apply to dependent claims 14 to 16. It is clearly stated in the cited decision that such claims represent, in effect, amendments which go beyond the objections to validity actually raised and are, therefore neither necessary nor appropriate under Rules 57(a) and 58 EPC. Consequently, auxiliary request IV is for this reason also not admissible.

(f) The observations and objections made in point (e) above apply mutatis mutandis

re (i): to independent claim 1,

re (ii): to independent claim 5 , and

re (iii): to dependent claims 14 to 16

of auxiliary request V (see paragraph V/E above). Consequently, auxiliary request V is similarly not admissible.

(g) Newly-filed claims 10 to 12 of auxiliary request VI (see paragraph V/F above) correspond to claims 14 to 16 of auxiliary request IV. The observations and objections to claims 14 to 16 of auxiliary request IV (see re (iii) in point (e) above) apply mutatis mutandis to dependent claims 10 to 12. Consequently, auxiliary request VI is likewise not admissible.

3. In the following paragraphs 4 to 6 reference is made to the main request.

4. The closest state of the art; the technical problem and its solution

4.1. Citation (1) discloses a medical dressing which is similar in construction to the claimed wound dressing in the patent in dispute in that it comprises at least the following three components or layers:

(i) a pressure sensitive adhesive composition layer [11] comprising a pressure sensitive rubbery elastomer adhesive material having intimately dispersed therein a water soluble or swellable hydrocolloid or mixture of hydrocolloids, a tackifier, and a plasticiser or solvent; according to the disclosure in column 1, lines 59 to 61, suitable rubbery elastomers include, inter alia, polyisobutylene, with a mixture of polyisobutylenes of a molecular weight of 10 000 to 11 700 and 81 000 to 99 000 being preferred; the hydrocolloids or mixtures of hydrocolloids comprise more than 30% by weight of the pressure sensitive adhesive composition with 40 to 50% by weight being preferred (see especially column 1, lines 61 to 64; column 2, lines 8 to 12);

(ii) a semi-open cell elastic or flexible foam layer [12] attached to the adhesive layer [11]; and

(iii) an outer water-impervious polymeric elastic or flexible film coating [13] attached to the opposite side of the foam layer [12].

Taking into account that the wound dressing disclosed in citation (1) is not only very similar in construction to the occlusive multi-layered dressing claimed in the patent in dispute but was also shown in the course of the proceedings to exhibit similar properties and advantages, the content and technical teaching of citation (1) is considered to be the closest state of the art available in the proceedings.

4.2. More specifically, in citation (1) it is said (see especially column 2, line 61, to column 3, line 56) that the hydrocolloids present in the adhesive layer [11] absorb moisture such as perspiration and wound exudate and transfer such moisture from the surface of the skin to the layer of open-cell foam where it can evaporate through the sides of the bandage. By regulating the moisture level at the surface of the skin the adhesive layer of the dressing disclosed in (1) enables the bandage to remain firmly in place for long periods and reduces or eliminates the need for the dressing to be changed.

As can be derived from the disclosure in the contested patent (see especially page 8, line 54, to page 9, line 7), the same or similar effects are achieved when wounds are treated with the dressing of the invention.

K. Reilly reports in her declaration filed on 4 October 1989, that a dressing corresponding to that disclosed in citation (1) (designated "dressing (C)" in the said declaration) exhibits satisfactory properties, with regard, for example, to (i) wound protection, (ii) adhesiveness to the normal skin surrounding the wound, (iii) absorbency of wound exudate and leakage, (iv) overall appearance of the dressing prior to its removal, (v) amount, viscosity, and colour of the wound exudate (residue) at the wound site after removal of the dressing, and (vi) quality of wound healing.

In spite of the clear statement in (1) that "since the hydrocolloids within the adhesive layer become mucilaginous when contacted with the burn exudate, removal of the bandage is possible without damage to the surface of the injured skin and with a minimum of pain" (see column 3, lines 54 to 57), K. Reilly found in the experiments reported in her declaration (see especially pages 14 to 17, item VII: residue on and mechanical injury to the normal skin upon removal of the dressing; pages 19 to 20, item IX: re-injury) that the adhesive layer in "dressing C" adhered to the wound bed and upon removal caused irritation to new tissue growth and in two cases resulted in re-injury of the wound.

4.3. Citation (2) discloses a three or multi-layered wound dressing or bandage comprising an outer layer [14], an intermediate foam layer [13] and a wound and skin contacting layer [11] the composition and consistency of which is similar to that of the claimed dressing in the patent-in-suit. This dressing is likewise useful in the treatment of open wounds such as decubitus ulcers. Citation (2), which was considered by the opposition division and the respondent as an alternative starting point for the assessment of inventive step, is, however, specifically concerned with the problem of deodorising gas escaping from a wound without impeding its passage and is therefore, in the board's judgment, considered to be less closely-related prior art than (1).

4.4. Consequently, in the light of the closest state of the art according to (1), the technical problem may be seen to be that of providing an improved dressing which, with regard to its beneficial properties and advantages, such as those referred to under items (i) to (vi) in point 4.2 above, at least meets the standards reported for the dressing disclosed in (1) but which, at the same time, permits easy removal of the dressing without causing irritation to the new tissue growth and without bearing the risk of re-injuring the wound on removal.

According to the patent in suit the appellant essentially proposes to solve this problem by the modification of the consistency of wound contacting adhesive layer [11] in the dressing disclosed in (1) ("dressing C"). This modification essentially involves the steps of increasing in the adhesive layer the proportion of lower molecular weight polyisobutylenes and removing from the adhesive layer the tackifier, plasticiser and higher molecular weight polyisobutylenes.

On the basis of the comparative results reported in the Reilly declaration for "dressing A" (corresponding to that described in example 1 of the patent in suit) and "dressing B" (corresponding to that described in example 1 of the patent in suit, except that it does not include the second adhesive layer [13]), the board has no reason to doubt that the claimed dressing provides an effective solution to the stated problem. This was not contested by the respondent.

In fact, from the experiments VII (residue on, and mechanical injury to, the normal skin upon removal of the dressing, pages 14 to 17 ) and IX (re-injury to the wound, pages 19 to 20) in th e Reilly declaration it can be derived that neither "dre ssing A" nor "dressing B" caused a substantial irritat ion of the normal skin or the wound on removal and bot h these dressings were readily removable without ca using re-injury.

5. Novelty (Article 100(a) in conjunction with Article 54 EPC)

In view of the respondent's prevailing objection to lack of novelty, the question to be decided is whether or not the proposed solution to the stated technical problem is derivable directly and unambiguously from the disclosure of either citation (1) or citation (2) as a whole including any features which a person skilled in the art would find implicit in what is expressly mentioned in these citations.

5.1. Example 1 is the only disclosure in (1) (see especially line 60 in column 3 to line 16 in column 4) which specifically describes the wound dressing disclosed in (1). Closer inspection reveals that this example refers to a three-layered dressing comprising a pressure sensitive adhesive layer [11], the quantitative and qualitative composition of which substantially differs from the composition of the corresponding adhesive layer [14] of the dressing claimed in the contested patent in several respects, more specifically in that

- the total hydrocolloid content is 33% by weight in the example in (1), whereas claim 1 of the contested patent requires an amount of 45% by weight minimum;

- the combined content of hydrocolloids and polyisobutylenes is 71% by weight in the example in (1), whereas claim 1 of the contested patent requires an amount of 80% by weight minimum;

- the adhesive layer [11] in the example in (1) contains a tackifier and plasticiser in a total amount of 28,5 % by weight, while the presence of a tackifier and/or a plasticiser in the adhesive layer [14] is not mentioned at all in the contested patent.

5.2. Turning now to the content of (1) as a whole, the following points appear relevant for the correct interpretation of the technical teaching imparted to a person skilled in the art by the cited document. While citation (1) discloses that the hydrocolloid content should comprise more than about 30% by weight and preferably about 40% to about 50% by weight of the pressure sensitive adhesive layer [11] (see especially column 2, lines 8 to 12; claims 3 and 4), the description and the claims are entirely silent about the proportion of polyisobutylenes present in this adhesive layer [11]. The only reference in the complete citation to the possible proportion of polyisobutylenes in the pressure sensitive adhesive layer [11] is that found in the above-mentioned example, wherein a mixture of two types of polyisobutylenes of strikingly different molecular weights of 10 000 to 11 700 and 81. 000 to 99 000 in a total amount of 38% by weight is used.

Having regard to the technical information of (1) in its entirety, there is, however, absolutely no reason for a person skilled in the art to associate the specific proportion of the mixture of polyisobutylenes of 38% in the adhesive layer disclosed in the example of (1), which, with respect to the proportion of all the other components, is outside the scope of the present claims, with the preferred proportion of 40% to 50% of hydrocolloids disclosed in column 2, lines 8 to 12. of (1). It follows necessarily that citation (1) does not, contrary to the respondent's opinion, directly and unambiguously make available to the public a dressing comprising an adhesive layer which consists of a homogeneous blend of from 35% to 50% by weight of low molecular weight polyisobutylenes and from 45% to 65% of water dispersable hydrocolloids. For the assessment of novelty of the claimed dressing vis-à-vis citation (1) it is therefore not necessary to discuss and decide the disputed question of whether or not the low molecular weight polyisobutylenes as such used in the adhesive layer [14] of the contested patent differ with respect to their molecular weight from the mixture of polyisobutylenes used in citation (1).

5.3. As to citation (2), the curative and absorbent layer [B] of the dressing disclosed in (2) is described on page 3 as comprising a blend of hydrocolloids and a natural or synthetic viscous substance which acts as a binder. The list of suitable viscous substances includes natural rubber, silicone rubber, acrylonitrile rubber, polyurethane rubber, and polyisobutylene of entirely unspecified molecular weight (see especially page 3, lines 11 to 14). Consequently, the mere reference in (2) to a list of viscous substances including polyisobutylenes of an unspecified molecular weight as one option amongst a series of natural and synthetic rubbers does not, contrary to the respondent's submissions in this respect, make available to the public the use of a low molecular weight polyisobutylene as the binder for the curative and absorbent layer [B] in citation (2).

5.4. Regarding the preparation of the curative and absorbent homogeneous cohesive mass [11] forming the layer [B], citation (2) contains on page 4, lines 23 to 26 cross-references to the whole content of three US patent specifications, viz. US-A-3 972 328 [current citation (1)], US-A- 3 339 546 [current citation (3)] and US-A-4 192 785 [current citation (4)].

The respondent, relying on decision T 153/85 (OJ EPO 1988, 1, see especially reasons, point 4.2), argued that the above-mentioned cross-references in (2) had the effect of incorporating in the disclosure of (2) the specific portion of the prior art of US-A-4 192 785 (see especially column 3, lines 41 to 51) wherein reference is made to the use of low molecular weight polyisobutylenes as a binder for the pressure sensitive adhesive component disclosed in (4).

The present case is, however, substantially different from the case considered in the above-cited decision. Apart from the fact that the competent board emphasised in paragraph 3 of point 4.2 of decision T 153/85 that "when assessing novelty, the disclosure of a particular prior document must always be considered in isolation", the so-called "primary document", corresponding to citation (2) in the present case, contained in T 153/85 a single specific reference to a single second prior document. By contrast, citation (2) refers in the context of the preparation of the curative and absorbent homogeneous cohesive mass [11] to the entire content of three different US patent specifications without giving priority to any of these references. Each of them offers a plurality of different options for preparing pressure sensitive adhesive layers having different compositions. Hence, it cannot be said that the use of a low molecular weight polyisobutylene as the adhesive and binder for the curative and absorbent layer [B] is directly and unambiguously derivable from the wholly general reference to the three different prior documents quoted in (2) and was therefore already made available to the public in citation (2) within the meaning of Article 54(2) EPC.

For the assessment of novelty it is therefore not necessary to decide the disputed question of whether or not citation (2) discloses an occlusive dressing as required in claim 1 of the patent in dispute.

In conclusion, the proposed solution of the stated technical problem satisfies the criteria for novelty within the meaning of Article 54(1) EPC.

6. Inventive step (Article 100(a) in conjunction with Article 56 EPC)

The appellant relied, inter alia, on the argument that the Reilly declaration was not available to the public before the priority date of the patent in suit and the skilled person would thus not have been aware of the problem caused by the prior art of (1) which forms the basis of the contested patent and which was exposed for the first time by that declaration. It submitted that the knowledge contained in the Reilly declaration was part of their invention and this should at least contribute to the acknowledgment of an inventive step in the present case.

6.1. It is true that none of the cited documents discloses or otherwise explicitly refers to a certain drawback or disadvantage associated with application of the dressing disclosed in (1) possibly resulting from its adhesion and aggressiveness towards the wound site and the consequential risk of causing irritation to the new tissue growth or even re-injury to the wound on its removal.

Notwithstanding this, any person, let alone any person skilled in the art, using the dressing disclosed in (1) would, in the board's judgement, normally recognise immediately the irritation to the new tissue growth or even the risk of re-injury to the wound as a serious problem, when it comes to the need for the dressing to be changed or removed for good. Since overcoming such a perfectly obvious, readily identifiable drawback and the achievement of an improvement resulting therefrom must be considered to be the normal task of the skilled person, the board cannot share the appellant's view that the identification of the particular technical problem as defined in point 4.4 above can be seen in the present case as a contribution to inventive step.

The remaining consideration is therefore whether the claimed solution is obvious to a skilled person in view of the prior art.

6.2. Although there is no suggestion in (1) that the adhesive layer [11] should be replaced or modified, the skilled person faced with the stated technical problem would already, in the board's judgment, on the basis of his specialist background knowledge, plausibly conclude that only the composition or consistency of the skin and wound contacting adhesive layer [11] and, more specifically, the relatively high proportion of tackifier and plasticiser (28% by weight) present in said layer [11] of the dressing disclosed in (1), was responsible for the re-injury problem when the dressing is removed. Hence, the skilled practitioner would, from his pre-existing knowledge, have reason to consider the possibility of solving the stated problem by removing the tackifier and plasticiser or at least reducing their proportion in the adhesive layer [14] of the claimed dressing.

6.3. For the assessment of the inventive step of the claimed dressing in the contested patent the development of the relevant state of the art appears important in the present case. The skilled person who had followed this development with the intention of finding in the state of the art a solution to the stated technical problem would have paid particular attention to citation (4), which had been published some four years later than citation (1). This document discloses an improved adhesive composition which is adapted to be used, inter alia, in the ostomy field and consists of a mixture of a hydrocolloid gum, a pressure sensitive adhesive material, and an agent which increases the cohesive strength of the composition. A polymeric film may be applied on one side of the mixture.

Apart from the fact that the adhesive composition in (4) is related to the adhesive layer [11] used in the dressing of (1) in that it likewise includes a hydrocolloid and a pressure sensitive adhesive material, the skilled practitioner would find it particularly significant that citation (4) refers, in the context of the pressure sensitive adhesive material, which provides dry adhesion to the body and holds the entire composition disclosed in (4) together, in other words, which acts as the adhesive and the binder, to a class of commercially available low molecular weight polyisobutylenes having a viscosity average molecular weight from about 36 000 to about 58. 000 (Flory) as the preferred material for this purpose (see column 3, lines 41 to 53). This avoids the need to add tackifier and/or plasticiser to the low molecular weight polyisobutylene, which act in (4) as the pressure adhesive material, so as to ensure its proper functioning as a sufficiently strong adhesive.

As regards the appellant's assertion that citation (4) concerned an adhesive composition specifically adapted for use with an ostomy appliance and was therefore unrelated to the claimed dressing in the patent in suit, this submission fails to take due account of the fact that the adhesive composition is explicitly described in (4) as likewise being useful for related medical purposes. For example, the adhesive composition disclosed in (4) can be employed to fix various devices to the body and in particular, can also be applied directly to a subcutaneous ulcer (see especially column 5, lines 15 to 23).

Hence, the prior art of (4) shows, on the one hand, that low molecular weight polyisobutylenes having a viscosity average molecular weight from about 36 000 to about 58 000 (Flory), which are described as being the preferred pressure sensitive adhesive material per se of the adhesive component disclosed in (4), have sufficient stick, even in the absence of tackifier and plasticiser, to affix firmly to the body not only a dressing but also some heavier devices such as a catheter, an electronic probe or a wound drainage system (see column 5, lines 17 to 20).

Additionally, in the board's judgment, the recommendation that the adhesive composition be applied directly to a subcutaneous ulcer suggests to one skilled in the art that, due the absence of tackifier and plasticiser, the adhesive composition disclosed in (4), which uses as the pressure sensitive adhesive material low molecular weight polyisobutylenes having the molecular weight indicated above, can easily be removed without causing irritation to the new tissue growth and without running the risk of re-injuring the wound on removal.

Thus, the skilled person being aware of the above mentioned technical teaching in the state of the art, would have, in the board's opinion, reasonably considered the stated technical problem to be solvable by using low molecular weight polyisobutylenes having the viscosity average molecular weight indicated in (4) as the pressure sensitive adhesive material and binder, in the absence of tackifier and plasticiser, for the type of dressing disclosed in the patent-in-suit and citation (1). Determination of the proportion of low molecular weight polyisobutylenes required in the adhesive layer [14] to ensure sufficiently strong adhesion of the claimed dressing to the normal skin surrounding the wound and, at the same time, to facilitate easy removal from the wound and surrounding skin would then have been merely a matter of routine experimentation for the skilled practitioner.

6.4. The skilled person, seeking in the state of the art a confirmation of his conclusions drawn from the prior art of (4), would focus his interest on the hydrocolloid surgical dressing material termed Stomahesive which is the subject of numerous publications in the prior art and which was commercially available at the priority date of the patent in dispute. Thus, as an example only, Stomahesive is described in citation (6a) as a surgical dressing material consisting of gelatin, pectin, sodium carboxymethylcellulose and polyisobutylene. One side of this dressing is said in (6a) to be covered with a film of polyethylene, while the other face is naturally adhesive.

Citation (5) contains a similar information as to the composition of Stomahesive and indicates moreover the proportions of the individual ingredients: gelatin 20%, pectin 20%, sodium carboxymethylcellulose 20%, polyisobutylene 40% (see especially middle of page 603).

This information on the composition of Stomahesive in citations (5) and (6a), which were published in 1973 and 1979 respectively, is coincident with that provided some 10 or 16 years later by the appellants themselves in their letter filed on 4 October 1989 (see page 2, end of third full paragraph), wherein the adhesive layer of Stomahesive is reported to be an equal weight mixture of gelatin, pectin, and sodium carboxymethylcellulose dispersed in low molecular weight polyisobutylene and applied to a polyethylene film backing. From the last two pages of the Reilly declaration (the attachment) it is similarly derivable that the adhesive layer of Stomahesive is a homogeneous blend of 40% by weight low molecular weight polyisobutylene, 20% by weight of gelatin, 20% by weight pectin, and 20% by weight sodium carboxymethylcellulose (see adhesive layer 1 in dressing D) and as such corresponds exactly to the skin and wound contacting adhesive layer [14] of the claimed dressing in the patent-in-suit.

Both citations (5) and (6a) report that the application of the dressing Stomahesive to decubitus ulcers and other injuries affords good and accelerated healing of the wounds and permits easy removal of the dressing without causing irritation of re-injury of the wound. Hence, in considering what is known from (4) about the advantages of using low molecular weight polyisobutylenes as the pressure sensitive adhesive material and binder in hydrocolloid containing adhesive compositions, and having knowledge of the composition and properties of Stomahesive, it was, in the board's judgment, obvious to a person skilled in the art that replacing the adhesive layer [11] used in (1) by Stomahesive as the skin and wound contacting layer [14] of the claimed dressing in the patent-in-suit provides an adequate solution to the stated problem.

6.5. The appellant, relying on the declaration of P. C. Kallos, argued that there had been several changes in the formula of Stomahesive and that the Stomahesive trademark therefore denoted a range of formulae rather than one specific formula. Against that, the respondent filed a declaration of R. Bradley, a former employee of the appellant, which stated that the adhesive composition of Stomahesive was from its commercial introduction until his resignation in 1989 a matter of public knowledge and in fact identical to the composition reported in the Reilly declaration. The appellant did not in its written submissions provide any further information as to the composition of Stomahesive and, when asked about this at the oral hearing, would only confirm that the composition has changed but not saying when or how. When the appellant's attention was drawn by the board to the composition of the adhesive layer of Stomahesive indicated in the Reilly declaration (see 6.4 above), the appellant would only confirm that this was the composition at the time of the tests described in that declaration. It was also said by the appellant that Bradley was not in its employment at the priority date of the patent in suit so could not have known the composition of Stomahesive at that date.

The board notes there is a distinct degree of common ground between the parties as to the composition of Stomahesive to be found, on the one hand, in the information submitted by the respondent in the form of prior art documents (5) and (6a) and the Bradley declaration and, on the other hand, from the information submitted by the appellant in the form of the letter filed on 4 October 1989 and the Reilly declaration. As to whether the basic composition of Stomahesive has or has not been the subject of major changes since its introduction, at least as far as its content of hydrocolloids and the use of low molecular weight polyisobutylenes in the pressure sensitive adhesive material are concerned, the board has to make a finding on the balance of evidence before it. In view of the common ground referred to above and the response, at best equivocal, of the appellant to the evidence of the Bradley declaration, the board can only conclude on the balance of evidence that, even it were accepted that the Stomahesive trademark denoted a range of similar formulae, the basic composition of Stomahesive has not been the subject of major changes since its introduction.

The burden of refuting the allegation to that effect in the Bradley declaration, which was put in evidence in May 1996, lay on the appellant which did not discharge that burden. To say there have been changes but give no detail of when and how changes were made does not assist the board at all. Confirming that the composition referred to in the Reilly declaration was that used at the time of the tests described by Reilly does not contradict Bradley's evidence that this was the composition used from the introduction of Stomahesive until 1989. As to the observation that Bradley was not in the appellant's employment at the priority date (22 April 1982), it can be accepted that at that date Bradley did not know the composition of Stomahesive but there is no evidence from the appellant to contradict Bradley's assertion that, from its introduction in the early 1970's until 1989, the composition remained the same. That is information which Bradley, who says he was employed by the appellant in about March 1983 as Director of Research & Development, could well have acquired. Further, it is Bradley's evidence, unrefuted by the appellant, that the composition was public knowledge since 1973, the date of a published article referred to by Bradley.

6.6. The further argument of the appellant that the skilled person would have considered Stomahesive to be too weak an adhesive to hold the claimed dressing firmly in place by itself is untenable, because citation (4) clearly teaches that low molecular polyisobutylenes, when used as the pressure sensitive adhesive material, develop sufficient stick to secure various devices, let alone a dressing, to the body. This teaching does not conflict with the disclosure in citations (5) and (6a).

Closer inspection reveals that in (5) a number of Stomahesive layers were cut to the size sufficient to cover the ulcer only and were put one on the top of the other to build the pad up to the skin level. This means that the layers of Stomahesive forming the pad were, in contrast to the adhesive layer of the claimed dressing in the patent-in-suit, not in contact with the skin surrounding the wound. According to the treatment schedule used in (5) it was apparently considered necessary to apply a compression bandaging with paste in addition to covering the wound with a pad consisting of a number of layers of Stomahesive (see especially top of page 604). This does, however, not allow the conclusion, that Stomahesive would not successfully function as an adhesive, if an adhesive layer was provided which contacts the wound and the surrounding intact skin, as is the case with the claimed dressing. In contrast to the dressing in (5), which contacts only the wound, the claimed dressing is intended to be secured to the body by that part of the wound and skin contacting layer [14] which contacts the intact skin surrounding the wound.

In view of the fact that according to (6a) the slab of Stomahesive is likewise "trimmed to the size of the lesion that is to be covered" (see especially middle of page 2) the above observations equally apply to the prior art of (6a). It is moreover to be noted that according to the disclosure of (6a) an additional covering with gauze dressing is provided on the top of the Stomahesive layer and it is apparently in the first place this gauze dressing which requires to be maintained in place by means of an adhesive tape (see (6a): page 2, line 11 to 12 from the bottom: "and maintain the whole firmly in place").

Moreover, citation (7) teaches the possibility of applying Stomahesive without using an adhesive tape. The argument that other means for securing the dressing to the body were required instead in the case of stage 5. ulceration, is not supported by the disclosure of (7) and is, accordingly, merely an assumption on the part of the appellant.

Finally, the respondent's observations appear correct, that the appellant itself has submitted with its letter dated 3 November 1993 a number of documents showing that the claimed dressing in the patent-in-suit (sold under the tradename Duoderm) is secured to the body in the same way as described in citations (5) or (6a) for Stomahesive, see as examples only:

- Duoderm , instructions for use, page 10, picture 1: "Tape the edges of the dressing using hypoallergenic tape when used under clothing or when there is a risk of peeling at up the edges "; page 10, picture 2: "Duoderm is a convenient dressing for use under a compression bandage in statis ulcer management" [cf. the identical disclosure in citation 5];

- Towards Rapid Tissue Healing, Johnson, Nursing Times, 28 November 1994, page 42, end of column 3: "Hypoallergenic tape was applied over the edges of the dressing to ensure close contact between skin and dressing";

- Military Medicine, 153, April 1988, page 188, left hand column, lines 9 to 12 from the bottom: "Each sore was then covered with a hydrcolloid disc (Duoderm) which was then fixed with an elastic tape".

7. Auxiliary request I

The numerical range of the viscosity average molecular weight ["from about 36 000 to about 58 000 (Flory)] of the polyisobutylenes used as the pressure sensitive adhesive material and binder in the skin contacting adhesive layer [14] corresponds exactly to the range disclosed in (4). This feature cannot, therefore, contribute to the acknowledgement of an inventive step.

8. Auxiliary requests II and III

There can be no doubt that the dressing disclosed in (1) has the same functions as the claimed dressing in the patent in suit, i.e to provide a closed moist wound treatment environment and to form a fluid-tight bond with the healthy skin around the wound so as to seal the dressing to the skin. Consequently, the additional functional features introduced in claim 1 of auxiliary request II cannot serve as a basis for the acknowledgment of an inventive step either.

The conclusions above likewise apply to auxiliary request III, which includes the features of both auxiliary requests I and II.

9. In conclusion, the solution of the technical problem in this case was, in the board's judgment, obviously derivable by the skilled person from the state of the art. Therefore, neither the subject-matter of claim 1 of the main request nor that of claim 1 of any of the auxiliary requests I to III involves an inventive step required for patentability under Article 52(1) in conjunction with Article 56 EPC.

ORDER

For these reasons it is decided that:

The appeal is dismissed