United States v. Telectronics, Inc., 658 F. Supp. 579 (D. Colo. 1987)

*580 *581 John Fargo, Commercial Litigation Branch, Civ. Div., Dept. of Justice, Washington, D.C., Henry Solano, Asst. U.S. Atty., Denver, Colo., for plaintiff.

James D. Hinga, Baker & Hostetler, Denver, Colo., for involuntary plaintiff and counterclaim defendant.

Bruce G. Klaas, Klaas & Law, Denver, Colo., Michael I. Rackman, Gottlieb, Rackman & Reisman, New York City, William C. Nealon, Suffield, Conn., Charles Goldberg, Peter L. Edwards, Rothgerber, Appel, Powers & Johnson, Denver, Colo., for defendants.

MATSCH, District Judge.

This is an action for patent infringement under 35 U.S.C. § 271. The court has jurisdiction under 28 U.S.C. § 1338. Plaintiff United States owns Patent No. 3,842,841, entitled "Constant Current Power Pack For Bone Healing And Method Of Use," issued on October 22, 1974 on an application filed on October 29, 1971 in the names of Drs. Carl T. Brighton, Zachary B. Friedenberg, and William Redka, as inventors. For convenience, this patent will be referred to as the '841 patent, or Brighton patent. The plaintiff alleges infringement of claims 1, 3, 4 and 5. The patent resulted from work done in performing a contract between the Office of Naval Research and the University of Pennsylvania. The named inventors were all employed by the University of Pennsylvania.

Involuntary plaintiff Zimmer, Inc. was granted a limited exclusive license in the Brighton patent by the Navy on or about October 31, 1977, for a term of ten years.

The patented invention is a device for expediting the healing of fractures and other bone defects. The devices involved are generally known as bone growth stimulators, and Zimmer's is sold under the trademark Zimmer DCBGS. The Zimmer device is quite similar to the preferred embodiment of the invention shown in the patent.

Defendant Telectronics, Inc. is an Australian-owned medical technology company whose main place of business is in Englewood, Colorado. Defendant BGS Medical, Inc. is a related corporation, also located in Englewood, and is the entity under whose name the business is now conducted. The trademark used by defendants for their product is OSTEOSTIM. While the Model 2000 is now being sold, earlier models were the S-12, HS-12 and XM-12. The OSTEOSTIM-2000 is the same in all relevant respects to all single-anode bone growth stimulator devices designed to expedite the healing of long bone fractures previously manufactured or sold by defendants Telectronics and BGS Medical Corporation, or their predecessors, or Telectronics Proprietary, Limited. The accused device in this case will be referred to as the OSTEOSTIM.

The Brighton patent is directed toward a device for expediting the healing of a fracture or bone defect through the use of a low level constant direct current applied to the site of the fracture via a cathode placed internally at the site of the fracture. Such devices are sometimes referred to as direct current ("DC") bone growth stimulators. The only direct current bone growth stimulators approved for use in the United States are the Zimmer DCBGS (or Quadpak) and the OSTEOSTIM made by defendants.

Normally fractures of bones heal naturally as a result of the body's own reparative process in several stages. First, inflammation *582 occurs at the point of injury followed by the formation of soft callus, which is cartilage. This gives way to hard callus formation, in which the soft cartilaginous tissue is replaced with fiber bone. Finally, remodelling occurs in which the fiber bone is replaced with lamellar bone. In bridging the fracture, the first bone formed is subperiosteal, i.e., below the periosteum which is the outer surface of the cortex of the bone. Endosteal bone also forms to bridge the fracture on the inside of the bone. Then the interfragmentary gap is filled with new bone, uniting the cortices.

DC bone growth stimulators are particularly useful in the treatment of nonunions. A nonunion is a fracture which has not healed naturally within an extended period of time, and in which all natural reparative processes have ceased. Nonunions occur in about 5% of all long bone fractures. Prior to the advent of direct current bone growth stimulators, the conventional treatment for a nonunion was bone grafting, which involved removing a portion of the hip bone of the patient and grafting this portion to the nonunion, in the hope that it will stimulate further bone growth and result in a healed fracture. Bone grafting involves major surgery, is painful to the patient and requires about 4-10 days of hospitalization. If the first bone graft is not successful, another must be tried. If this is not successful, the limb may be amputated.

DC bone growth stimulators have been successful in treating nonunions, even in instances where bone grafting has previously failed. The Zimmer Quadpak and the Telectronics OSTEOSTIM enjoy a success rate of about 70-85% in healing nonunions. This is at least as good as the rate of success using bone grafting and avoids the necessity for removal of a portion of the patient's hip for a graft, along with the attendant pain. Furthermore, use of a bone growth stimulator saves hospitalization time, and its attendant costs, over that required for bone grafting.

The four claims of the patent alleged to be infringed are as follows:

1. A system for expediting the healing of bone fractures and bone defects in a living being comprising:

constant current source means for providing a constant value of current despite changes in load;

means for connecting said constant current means to the living being, such connection acting to produce current flow into said fracture or defect,

said connecting means including further means for application internally of said living being at the fracture or defect site,

said constant current being a selected value within a predetermined microampere range so as to promote bone formation at the fracture or bone defect site and avoid fibrous tissue formation in other areas of the living being.

3. The system as defined in claim 1 wherein said constant current means comprises miniature solid state means suitable for mounting on the living being in close proximity to said fracture or defect.

4. The system as defined in claim 1 wherein said cathodic electrode is positionable within said fracture or defect.

5. The system as defined in claim 1 wherein said current is in the range of from substantially 5 microamperes to substantially 20 microamperes.

Exhibit 1.

The magnitude of the constant current in the embodiment of the invention illustrated in the patent is 10 microamperes, although a range of 5-20 microamperes is said to be useful. Defendants use a constant current of 20 microamperes. When using the product of either party, the cathode (negative terminal) is placed in the defect site. The Zimmer cathode is made of stainless steel, the material described in the patent. The OSTEOSTIM cathode is made of titanium. The major difference between the products of the parties pertains to the anode (positive terminal). As disclosed in the patent drawing and accompanying description, and as marketed by Zimmer, the anode is placed on the skin of the patient. So is the power pack (current source) itself. The *583 only internal element is the cathode a pin which is inserted through the skin into the defect site. This technique avoids the need for surgery; after several months of treatment, the cathode pin is simply pulled out. The OSTEOSTIM device, on the other hand, is completely implanted, an embodiment which while not shown in the patent drawing is nevertheless described. The power pack and the anode of the OSTEOSTIM are placed in soft tissue near the bone. The original OSTEOSTIM S-12 had a power pack from which two wires extended, the wires terminating respectively at a titanium cathode for placement in the defect site, and a platinum anode for placement in the soft tissue. In all of the later models, including the OSTEOSTIM-2000, the anode wire was omitted. The anode is the case itselftitanium with a patch of platinum.

The defendants' denial of infringement in this case is based solely on the defendants' anode and case being used internally. Accordingly, the critical question in the case is whether the language of claim 1 (and with it, the dependent claims) is limited to a skin anode.

The defendants' contention is that an internal anode could not come within the literal language of claim 1 because fibrous tissue formation inevitably results from such an implant. Therefore, fibrous tissue formation could not be avoided in the dictionary sense of "keep away from" or "stay clear of". Webster's Third New International Dictionary, 151 (1971).

The patent is for a medical device an apparatus which will promote healing. To accomplish that purpose, it must, of course, avoid injury or adverse medical results. It must not do damage. Fibrous tissue forms as a response to a foreign body. That is called encapsulation. Fibrous tissue also results from any trauma. Electrical current produces fibrous tissue. Fibrinoid necrosis is the term used to describe dead fibrous tissue. Fibrinoid necrosis is damage which must be avoided.

The defendants made a conscious decision not to avoid the formation of fibrous tissue, electing to tolerate the amount formed from implanting the case/anode because its effect is not so deleterious as to outweigh the benefit of greater freedom of movement for the patient. Additionally, an implanted case/anode does not have a wire penetrating the skin and thus the risk of infection is reduced. Drs. Brighton and Friedenberg, on the other hand, chose to avoid the two surgical procedures necessary for implant and explant.

The claim limitation directed to the avoidance of fibrous tissue means what it plainly says. Accordingly, there is no literal infringement because in the context of the patent, even minimal fibrous tissue formation is not its avoidance.

Although literal infringement has not been proven, the plaintiff claims infringement under the doctrine of equivalents contending that the OSTEOSTIM performs substantially the same function in substantially the same way to obtain the same result. The plaintiff recognizes the ambiguities in the language of claim 1 and asserts that the last phrase should be read to mean: "said constant current being a selected value within a predetermined microampere range so as to promote bone formation at the fracture or defect site, and minimize fibrous tissue formation and subsequent necrosis of that tissue that is related to current." Plaintiff's Requested Findings of Fact and Conclusions of Law, p. 30 n. 5.

The anode's function is to return current so that the circuit is completed and, in this regard, the anode may be placed anywhere as long as this function is served. There are three possible positions for placement of the anode: on the skin, in soft tissue, and in the bone. Placement within the bone must be done carefully to avoid the effect of insulation from the cortical bone.

The inventors of the Brighton patent conducted animal experiments which were described in the patent and in which the cathode was inserted into the fracture or bone defect site, and the anode fastened in *584 soft tissue. Using stainless steel, the current levels used produced different results. At 1 to 5 microamperes there was little reaction at either the anode or the cathode site. At 10 to 20 microamperes there was a dark discoloration around the anode, but no discoloration at the cathode. At 50 to 100 microamperes, marked discoloration occurred at both electrode sites.

Analysis of these results demonstrated that fibrous tissue formation at the anode reached a maximum at ten microamperes and at current levels higher than ten tissue destruction occurred with a maximum being attained at twenty microamperes and persisting through one hundred microamperes. That tissue destruction was that of fibrinoid necrosis.

In the patent description, the inventors described the work with human subjects as being limited to external anodes and power pack.

The plaintiff's position is that the meaning of claim 1 is that because all fibrous tissue formation cannot be prevented where the anode is implanted, care must be taken to use a sufficiently-high level of current within the predetermined healing range to promote bone formation at the cathode in the defect site, and yet not be so high as to cause substantial fibrous tissue formation at the anode, and thereby avoid the development of fibrinoid necrosis.

Stated simply, the plaintiff claims that the adverse effect to be avoided is fibrinoid necrosis; that it is avoided in the Brighton patent by a constant controlled current within a predetermined range which is 5 microamperes for stainless steel electrodes, and that the defendants avoid fibrinoid necrosis by using a platinum electrode with a constant current of 20 microamperes. Thus, since the defendants' devices are admittedly substantially similar in all other pertinint aspects, there is infringement under the doctrine of equivalents.

The resolution of this issue requires careful consideration of the prosecution history of the application for the Brighton patent because the defendants contend that the inventors narrowed the scope of the claims by excluding the internal anode to avoid prior art and thereby obtain their patent.

The Brighton patent issued from application Serial No. 193,728 filed on October 29, 1971. It contained the following claims:

Application Claim 1

A means for supplying a constant current across bone fractures and defects in a living body to promote more rapid healing thereof comprising:

a power source adapted to provide a substantially constant current supply to the region of said fracture or defect notwithstanding changes in impedance there across during healing thereof.

Application Claim 2

The means as defined in Claim 1 wherein said power source includes an electrode implanted at the site of the fracture or defect and an electrode applied to the surface of said body.

Application Claim 3

The means as defined in Claim 2 wherein said current is in the range of substantially 5 microamperes to substantially 20 microamperes.

Exhibit 2, p. 10.

In an Office Action mailed January 8, 1973, the Examiner rejected the claims both for failure to distinctly claim the invention as required by 35 U.S.C. § 112 and for anticipation and obviousness under 35 U.S.C. §§ 102 and 103. Claims 1-3 were rejected as unpatentable over Gershen patent 3,317,817 and Warner patent 3,179,892. The Examiner also cited an article written by Dr. Andrew Bassett, Robert Pawluck and Robert Becker entitled "Effects of Electrical Currents on Bone In Vivo," published in the 1964 edition of Nature [hereinafter referred to as the 1964 Nature article], in rejecting other claims.

Amendment B was filed in response to this Office Action on March 13, 1973. The claims were cancelled and new claims 14-26 presented. The following application claims 14-18 are pertinent to the claims in dispute:

Application Claim 14

A system for expediting the healing in vivo of bone fractures and bone defects comprising: *585 a source of constant dc current; and

means connecting said source of constant current to the living being so as to produce current flow into said fracture or defect,

only one of said connecting means applied to the skin surface of the living being,

said constant current into said fracture or defect being maintained within a selected microampere level so as to promote bone formation at the fracture or bone defect site and avoid fibrous tissue formation in other areas of the living being.

Application Claim 15

The system as defined in claim 14 wherein said connecting means applied to the skin surface is an anode,

said connecting means including a cathode implanted at the site of said fracture or defect.

Application Claim 16

The system as defined in claim 15 wherein said source of constant current is a miniature solid state device mounted on the living being in close proximity to said fracture or defect.

Application Claim 17

The system as defined in claim 16 wherein said cathode is positioned in said fracture or defect.

Application Claim 18

The system as defined in claim 17 wherein said current is in the range of from substantially 5 microamperes to substantially 20 microamperes.

Exhibit 2, pp. 28-30.

The Navy argued that the 1964 Nature article did not concern bone healing but only taught that "some new bone formation may be produced within the medullary canal by current flow between a pair of electrodes inserted thereinto." Exhibit 2, p. 36. It distinguished the Nature article from the amended claims on three grounds; namely,

No reference teaches or suggests the present inventive concept, namely that current flow across a fracture or defect may be induced to effectively enhance healing. The references also do not teach or suggest inserting one electrode in the being at the site of the fracture and applying the other electrode to the skin of the being in the vicinity of the fracture. A further deficiency in the references is an identification of the current levels within which healing is enhanced and outside of which undesirable effects are encountered.

Exhibit 2, p. 36.

The Examiner replied to Amendment B by again rejecting the claims as unpatentable in an Office Action mailed June 8, 1973. The Examiner stated that the 1964 Nature article teaches a bone growth stimulator "employing [a] D.C. (battery) source with a current within the claimed ranges and substantially constant after an initial transient response (Fig. 2), located near the defect (i.e. drill hole) of the bone, at least one electrode implanted in the defect and another circuit-completing electrode connected between the body and the source." Exhibit 2, p. 41.

The Examiner said that it would be obvious to substitute a solid state constant current source for the Bassett device's "large value resistor" and that it would further be obvious to use a surface electrode in light of Mirowski patent 3,614,955. Exhibit 2, pp. 41-42. The Mirowski patent is directed to a "Standby Defibrillator and Method of Operation." A standby defibrillator is a device for monitoring the rhythm of the heart and delivering a short high voltage shock to counteract fibrillation. It uses one electrode positioned in the right ventricle of the heart and a second electrode which is either placed on the skin, sutured under the skin or applied to the outer surface of the heart. When heart fibrillation is sensed, it delivers a shock of approximately 2500 volts to the heart. The Examiner noted that this rejection was a final action.

The applicants submitted amendment C on August 9, 1973. New claims 27-39 were presented. Claim 27, corresponding to claim 1 of the patent, explicitly defined a skin electrode, and the last limitation once again dealt with the avoidance of fibrous tissue formation. The remarks which accompanied *586 the amendment included several arguments which distinguished the claimed invention from the prior art because of the use of a skin anode. Notably, the Navy lawyers said:

It is submitted that the foregoing references are completely deficient insofar as anticipating the present invention. In the invention, current is induced across a fracture by an electrode inserted at the site thereof and a second electrode applied to the skin of the living being. There is no teaching in, or continuity among, Bassett et al, Hurd, III and Mirowski et al by which a person, having only these three references, could arrive at applicant's invention.

Exhibit 2, p. 56.

In an Advisory Action of September 6, 1973, the Examiner informed the applicants' attorney that the new claims had not been entered, and that old claims 14-26 still stood rejected. On September 21, 1973, Amendment D was filed. Application claims 14-26 were submitted. Application claims 14-18 are identical to claims 1-5 of the issued patent. There is no dispute that the claims of Amendment D were entered. The amendment, as it appears in the prosecution history produced by the Patent and Trademark Office contains a handwritten note "OK to enter WEK" and handwritten renumbering of the claims from 14-26 to 27-39, with the initials "WEK" appearing next to some of the renumbered claims. Apparently, the Examiner renumbered the application claims. The remarks section of the amendment specifically states that the claims were being amended as agreed at an interview held with the Examiner on September 16, 1973.

In an Office Action mailed November 6, 1973, the Examiner withdrew his final rejection and rejected the claims under 35 U.S.C. § 102 as unpatentable over the Friedenberg 1970 Surgery, Gynecology and Obstetrics article, and under 35 U.S.C. § 103 and § 102 as unpatentable over the Friedenberg 1968 article dealing with stimulation of the growth plate and an article by Assimacopoulous dealing with the use of current to heal wounds. There is confusion as to which set of claims was being examined, the claims numbered 27-39 submitted in Amendment C or the amended claims numbered 14-26 submitted in Amendment D, which the Examiner renumbered.

The issue is important because responsive Amendment E, Exhibit 2, pp. 73-79, filed on December 7, 1973, while not containing any changes to the claims, did contain an argument directed to claims 27-39 which focused on a skin anode. Defendants assert that the argument was directed to the claims of Amendment D; since claim 27 of Amendment D was identical to patent claim 1, patent claim 1 is to be interpreted as being limited to the use of a skin anode. Plaintiff asserts that the argument in Amendment E was directed to claim 27 of Amendment C, this claim explicitly defining a skin anode; it was later that the claims of the patent were submitted again, and the previous argument about a skin anode applied only to claim 27 of Amendment C which explicitly defined a skin anode.

On conflicting evidence, this court finds that the Office Action of November 6, 1973 rejected the claims of Amendment D, not the claims of Amendment C. The claims of Amendment C were not entered and the applicants' attorney was told this in an Advisory Action. It would be highly unusual for an Examiner to decide to enter claims in such a case and not tell the applicants' attorney. There is no question that the claims of Amendment D were entered. Amendment C was filed prior to the interview which was held on September 16, 1973, Amendment D was not only filed after the interview, but specifically states that the claims were based on what transpired at the interview, and the remarks of Amendment E also refer to the interview. The claims in Amendment C were not rejected in view of the prior art (they were simply not entered), while the claims of Amendment D were rejected in view of the three new references; the argument of Amendment E distinguished claims 27-39 from these references so it must be the claims of Amendment D which the attorney had in mind because they were the only claims which had been rejected in view of *587 these references. Although no argument was presented in Amendment D, the remark was made that the claims were being amended "in accordance with the changes which the Examiner agreed would remove his objections." Exhibit 2, p. 66.

It is unlikely that an attorney, who thought that he had reached an agreement with the Examiner on a second group of claims but then discovered that the Examiner was still not convinced, would have argued in Amendment E about another completely different first group of claims without even telling the Examiner. At any rate the question is not what the applicants thought they were construing. What they were thinking is irrelevant. It is what was said that counts; an estoppel would arise based upon the effect of the argument on the Examiner. It is clear from the Office Action which followed the filing of Amendment E that the Examiner had previously had in mind the claims of Amendment D. The Examiner continued to reject the claims as unpatentable over the 1968 Friedenberg article in light of the Assimacopoulous article. He reasoned that the 1968 Friedenberg article taught a means of expediting growth at the fracture site including using a substantially constant current, an external anode and an internal cathode at the point of desired growth. He said that while the Friedenberg 1968 article did not involve the use of constant current, it would be obvious to modify the device disclosed in that article to include a constant current source in light of the Assimacopoulous article. The Examiner made this rejection final.

The Navy attorney, in Amendment F, filed another set of claims, this time numbered 40-52, but otherwise identical with the claims of Amendment D; it was these claims which were ultimately allowed.

The remarks which accompanied the claims in Amendment F included the following argument:

... Applicants take strong exception to this analysis of the [Friedenberg-Kohanim article, Exhibit 4] reference. Nowhere in this article is there either stated or suggested that one of the electrodes need simply be applied to the surface and the other introduced into the fracture site....

Exhibit 2, p. 92.

The argument for allowability of the submitted claims continued as follows:

Applicants throughout the prosecution of this case have repeatedly attempted to convey to the Examiner the important differences between their technique where only one of the electrodes need pierce the skin and enter the fracture site and the other prior art arrangements where two electrodes have to pierce the skin and then fit into prescribed locations formed in the bone structure under study.

Exhibit 2, p. 93.

Together with Amendment F there was submitted an affidavit by several of the co-authors of the Friedenberg 1970 Surgery, Gynecology & Obstetrics article which the Examiner had cited. That affidavit states that "the concept of disposing a cathode electrode in the fracture site and positioning the anode electrode on the skin surface of the subject originated with our co-authors, Drs. Friedenberg and Brighton ..." Exhibit 2, p. 97. It is clear from the file history that what convinced the Examiner to allow the claims over the prior art was the argument that a skin anode was used in the invention. The doctrine of prosecution history estoppel "bars a patentee from construing its claims in a way that would resurrect subject matter previously surrendered during the prosecution of the patent application, and thus prevents a patentee from enforcing its claims against otherwise legally equivalent structures if those structures were excluded by claim limitations in order to avoid prior art." Mannesman Demag Corp., v. Engineered Metal Products Co., 793 F.2d 1279, 1284 (Fed.Cir.1986). Accordingly, the plaintiff is prevented from construing its claims to include an internal anode.

Claim 2 is different from claim 1. Claim 1 is implicitly limited to a skin electrode, but the claim does not say whether it *588 is the anode or the cathode which is on the skin; it is claim 2 which says that the external means of claim 1 is the anode, with the internal electrode being the cathode. The doctrine of claim differentiation has been satisfied.

Moreover, claim 4 does nothing more than say that the cathode is positionable within the fracture or defect. If claim 1 inherently defined the cathode as the internal means, claim 4 would be nothing but a repetition of that portion of claim 1, contrary to the doctrine of claim differentiation. D.M.I. v. Deere & Co., 755 F.2d 1570, 1575 (Fed Cir.1985); Kalman v. Kimberly-Clark Corp., 713 F.2d 760, 770 (Fed.Cir. 1983), cert. denied, 465 U.S. 1026, 104 S. Ct. 1284, 79 L. Ed. 2d 687.

If the word avoid in claim 1 means minimize fibrous tissue formation, as plaintiff claims, there is no infringement because the OSTEOSTIM causes the development of fibrous tissue around the anode, which is not insignificant, and which creates difficulty when explanting the device. On all models, except the original S-12, the anode is a platinum patch on the titanium case. When first implanted, the 20-microampere current splits about evenly between the platinum and titanium materials. As the weeks go by, an oxide layer builds up on the titanium and more and more of the total current flows through the platinum patch. But even after 10 weeks, 4 microamperes still flow through the titanium case. So the titanium case functions as part of the anode structure. The fibrous tissue layer which forms around the case/anode of the OSTEOSTIM is not minimal. It is the equivalent of 8 sheets of yellow paper. Thus, the OSTEOSTIM does not minimize fibrous tissue in the "other areas" called for by the claim. The OSTEOSTIM does not, however, develop fibrinoid necrosis. Accordingly, to find infringement, the Brighton patent must be read to minimize fibrinoid necrosis formation from the electrical current.

The re-phrasing of claim 1 to the broad language now under consideration has evolved during the progression of this lawsuit. In the agreed pre-trial statement submitted to this court on January 17, 1986, the plaintiff's counsel submitted this language:

Plaintiffs further note that claim 1 calls for selecting a current level within a predetermined range so as to promote bone formation in other areas of the living being. This limitation merely refers to the current level used, not where the anode is placed.

Pretrial Statement, pp. 5-6.

Shortly before trial, Dr. Brighton and plaintiff's counsel first informed the defendants' counsel that it was fibrinoid necrosis that was to be minimized in a "tradeoff" in selecting the specific value of current to be used within the "window" of 5-20 microamperes. The defendants filed a motion in limine to exclude testimony in support of the contention that the "avoid fibrous tissue formation" phrase in the patent should be interpreted as "avoid fibrinoid necrosis formation", claiming surprise and bad faith from plaintiff's late change of position. This court denied that motion. Upon consideration of all the evidence, it is apparent that this rewrite was required to make a tenable claim of infringement. It is this court's finding that the suggested language of interpretation is a litigation claim; not a patent claim. To accept this argument, this court would have to find that the Navy lawyers made a mistake in writing the language of claim 1. That is too much to ask. Fibrous tissue is living tissue. Fibrinoid necrosis is dead tissue. Anyone with minimum skill in the art would avoid confusing the terms. Moreover, this court has no power to correct a mistake in the language of the claim. Super Products Corp. v. D P Way Corp., 546 F.2d 748, 756 (7th Cir. 1976) ("If the claim does not fairly reflect the invention he seeks to protect, the courts are without power to rewrite the claims even though his invention might have justified patent protection, ...").

*589 If claim 1 were to be given the broad meaning which plaintiff asserts, then the patent would be invalid for a failure to comply with the specification requirements of 35 U.S.C. § 112. The plaintiff's reading of claim 1 would make it impossible to determine when sufficient minimization takes place to determine what current range is involved. More importantly, the patent refers only to electrodes made of stainless steel, and it is for that material that the 5-20 microampere range is specified. That range was developed in a dose response experiment with rabbits, described in the patent. The predetermined current range referred to in claim 1 of the Brighton patent will be different for different cathode materials. Dr. Brighton did later studies of platinum electrodes and determined that the optimum range of current for a platinum cathode is 2-5 microamperes. The patent does not contain an adequate description of the methodology for a dose response study of this type. The plaintiff contends that those of ordinary skill in the art would be able to perform dose response studies for different cathode materials.

A dose response study typically costs about $40,000 to $50,000 to perform and requires six to twelve months. It may involve four disciplines: an electrical engineer, a surgeon, a biomechanic and a biologist. The patent does not describe how to conduct such a study and, accordingly, an undue amount of experimentation is required. Indeed, Dr. Brighton admitted in his testimony that only those who were expert in the field and actually working with bone, doing electrical stimulation experiments, such as Bassett and Lavine, would know how to conduct a dose response study to determine the appropriate current to be used with other materials as electrodes. Additionally, while the patent refers to 10 microamperes as the preferred constant current for stainless steel electrodes, Dr. Brighton testified that his experiments showed the maximum amount of fibrous tissue formed at 10 microamperes. Given the breadth of the claims made in the plaintiff's arguments, the patent does not tell a person reasonably skilled in the art how to make and use this invention because it fails to teach how to select a level of current to promote bone formation and avoid fibrous tissue and fibrinoid necrosis formation from such current.

The defendants contend that the '841 patent is invalid for obviousness under 35 U.S.C. § 103, and anticipation under 35 U.S.C. § 102. In determining these issues, the court will limit the patent to an external anode.

The principal prior art involved in this aspect of the dispute is the 1964 Nature article [Exhibit 3], the Friedenberg 1968 article [Exhibit 4], and the Kraus patent [Exhibit 17]. The Shamos and Lavine article [Exhibit D-6] is also asserted. It is this collection of prior art which caused the Navy attorneys to emphasize the skin anode in argument which, in turn, has caused this court to limit the claims of the patent. With that limitation to an external anode, this court essentially agrees with the Examiner and the Patent Office on the validity of the Brighton patent.

In the 1964 Nature article, Dr. Bassett described an experiment in which direct current was applied to the medullary canals of dog bones to study osteogenesis. The device disclosed in the article consisted of a battery and a resistor encapsulated in an epoxy-coated case. Two prongs, which served as the anode and cathode, protruded from the case at a distance of 13 millimeters from each other. The anode and cathode were inserted tightly through two holes drilled in the cortex, or outer wall, of the bone and projected into the medullary canal.

Depending upon the resistor used with the battery, these devices could deliver a current of 1, 10 or 100 microamperes before implantation into the animal. Upon implantation, the current fell rapidly during the first 30 minutes and then appeared to level off. The 100 microampere battery pack fell to a value of about 3 microamperes. The 10 microampere battery pack fell to about 2 microamperes, and the 1 *590 microampere pack fell to about 0.7 microamperes. Of these packs, the 10 and 100 microampere packs produced more bone around the cathode than inactive control electrodes, while the 1 microampere pack was no more effective than the control. However, bone production reached a peak within 14 days and did not increase thereafter. Dr. Bassett concluded that "under the influence of low level, direct electrical currents, bone formation mechanisms are affected in some way that depends upon the polarity of the applied current, to produce new bone growth preferentially in regions of relative electronegativity." Exhibit 3, p. 654.

The "1968 Friedenberg article," written by Dr. Friedenberg and Dr. Kohanim, is entitled "The Effect of Direct Current on Bone" and appeared in the July 1968 issue of Surgery, Gynecology & Obstetrics. Exhibit 4. The experiment reported in this article was directed toward determining if applying direct current to the tibia bone of an immature rabbit could accelerate the growth of that bone, or lengthen it. A passive current source consisting of a battery and a resistor was used to deliver a current to electrodes inserted into holes drilled into the middle of the epiphysis and the metaphysis. In addition, electrodes were inserted into holes drilled near the tibial tubercule, a relatively less dense part of the tibia near its upper portion, to investigate the effect of current on bone tissue near the electrodes.

The current delivered initially was between 7 and 13.5 microamperes but gradually declined during the experiment. The results showed no significant increase in the length of the bone and destruction of tissue around the anode. Bone formation around the cathode was minimal, consisting of sparse intramedullary trabeculae. The article stated that a "controlled system is needed to deliver a constant current during a prolonged period of time to assess what is the most effective current and what is a destructive current." Exhibit 4, p. 102.

The Assimacopoulous article is entitled "Wound Healing Promotion by the Use of Negative Electric Current." Exhibit 5. It reports the use of approximately 100 microampere currents applied to a wound on the skin of an animal to accelerate the formation of fibrous tissue to heal the wound.

While this prior art revealed what may be considered the essential components of the Brighton patent, there was doubt that enough bone could be formed to make electrically-induced osteogenesis clinically useful in the treatment of fractures and bone defects.

U.S. Patent No. 3,745,995 issued to Werner Kraus (the "Kraus patent") on July 17, 1973, based on an application filed in the United States on April 9, 1970. Exhibit 17. The Kraus patent discloses an AC, or alternating current, bone growth stimulator which contains two surface plates on the outside of the fracture, a set of electrodes which extend through screws into the medullary canal and another set of electrodes which enter the fracture site. The apparatus also forms a splint for the fracture.

While the Kraus patent does show a pair of electrodes in a fracture site, it does not suggest that these electrodes could be used without the surface plates and intermedullary electrodes disclosed in that patent. Nor does the Kraus patent suggest that these electrodes should be cathodes if direct current rather than alternating current were used. As of February, 1971, when the Brighton invention was made, those skilled in the art recognized the use of alternating current as distinct from direct current. Some research teams experimented with both DC and AC systems. But, it was not apparent that the electrical parameters used in an AC system could be transferred to a DC system. The manner in which AC and DC systems operated were thought to be distinct from each other.

The claimed invention would not have been obvious to one of ordinary skill in the art at the time Dr. Brighton and his co-workers made their invention. The defendants argue that it would have been obvious to modify the device disclosed in the Nature article to include a power source with self-adjusting, active circuitry to supply a *591 constant current in a range sufficient to heal a fracture, given the suggestion in the 1968 Friedenberg article that a controlled source was needed to permit experimentation with constant current. However, the suggestion that a controlled source was needed appears to have meant different things to the various workers in the field. For example, even after Dr. Brighton and his co-workers made their invention, the Lavine group was still using a power source with a battery and a variable resistor without any compensating circuitry.

These articles show that the Lavine team, while it realized the need to control the current delivered to the defect site, never recognized the desirability of using active circuitry to maintain a constant current despite changes in load and they chose not to use an internal cathode at the fracture site, but rather, tried to provide current flow across the fracture site by using two electrodes squeeze-fit into drill holes on either side of the fracture site.

The Friedenberg article, read in its entirety, describes a failed experiment in which the authors tried unsuccessfully to stimulate the growth plate to lengthen the bone rather than expedite healing of a fracture. The fact that this work was a failure makes it highly uncertain that one of ordinary skill at the time would have seized on the one sentence in that article that defendant emphasizes. Neither the 1964 Nature article nor the 1968 Friedenberg article was concerned with the use of electrically induced osteogenesis to heal a fracture. Rather, they were directed to attempts to study whether electricity could produce bone formation rather than whether electricity could produce sufficient bone to expedite the healing of a fracture. Moreover, when O'Connor attempted to apply the teachings of the 1964 Nature article to the problem of fracture his results were completely negative and disappointing. O'Connor concluded that "[a]lthough our results agree in general with those of Bassett, Pawluk and Becker, the differences between our findings and theirs emphasize the need for caution in clinical applications for there seems to be considerable biological variation in the response of bone to electric current even within a single species." Exhibit 13, pp. 162-163.

Defendants further seek to combine the disclosures of the 1964 Nature article and the 1968 Friedenberg article with the electrodes placed in the fracture site in the Kraus patent. However, the Kraus patent discloses several electrodes without suggesting that the electrodes in the fracture site can be separated from the rest of the device and used in a DC system. While some of the same people experimented with both AC and DC devices in studying electrically induced osteogenesis, they did not apply the electrical features of AC devices to DC devices, or vice versa. The evidence does not show that one of ordinary skill in the art would have found it obvious to adapt the electrode at the fracture site in the Kraus patent to form a cathode in a DC system.

Since it would not have been obvious to have made the invention defined in claim 1, which requires a self-adjusting constant current source and an internal cathode at the fracture site, it would not have been obvious to make the inventions defined in dependent claims 3, 4 and 5.

Another set of factors relevant to a determination of invalidity for obviousness under section 103, is objective evidence of nonobviousness, including long-felt need and commercial success. Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S. Ct. 684, 693-94, 15 L. Ed. 2d 545 (1966). Dr. Brighton emphasized that the Zimmer device healed nonunions in a clinically practical and reproducible manner. Prior to the Brighton invention nonunions were treated by bone grafting. If bone grafting proved unsuccessful, then amputation was performed. There is also some merit to the plaintiff's argument that the gross sales of the Zimmer device suggest that it is a commercial success, another indication of nonobviousness. See, W.L. Gore & Associates, Inc., v. Garlock, Inc., 721 F.2d 1540, 1555 (Fed.Cir.1983), cert. denied, 469 U.S. 851, 105 S. Ct. 172, 83 L. Ed. 2d 107.

The defendants contend that the patent should be invalidated because the *592 Navy attorneys committed fraud, or at least, inequitable conduct in the arguments made before the Examiner. This contention is based principally on the statements concerning the reasons for distinguishing the 1964 Nature article and the importance of the constant current findings by the inventors. More particularly, the attorneys are accused of misrepresenting the 1964 Nature article as not being related to healing, but only to osteogenesis, and the principal inventor, Dr. Karl Brighton, testified both in deposition and at trial that the constant current may not be as significant as the current level in the healing process. This court has carefully reviewed the prosecution history, the testimony of Dr. Brighton in deposition and at trial, the deposition testimony of Dr. Friedenberg, and the arguments made in the defendants' briefs. What appears is that the patent was ineptly argued by counsel who apparently had limited understanding of the subject matter and inadequate communication with Dr. Brighton and the other inventors. The plaintiff has already suffered the adverse consequences of the use of a prosecution argument to limit the scope of claim 1. The evidence is insufficient to support the defendants' claims that there was any intent to misrepresent or mislead the Patent Office. The inadequacy of counsel is not equated with fraud or inequitable conduct.

The defendants also claim that William Redka should not have been named as a co-inventor of the Brighton patent because his only contribution was the design of the self-adjusting constant current source disclosed in the patent, and which is the subject of claims 6 through 10, not the claims in suit. Under 35 U.S.C. § 116, when an invention is made by two or more persons jointly they must join in the application.^[1] It is not clear that there is any requirement that all co-inventors must have combined their efforts to each claim in the patent. SAB Industri AB v. Bendix Corp., 199 U.S.P.Q. 95 (E.D.Va.1978). If that is the law, it is this court's view that the circuity design made by Redka is sufficiently related to the patent in its entirety to justify the inclusion of his name as as inventor.

At any rate, if there were misjoinder, it is irrelevant to invalidity in this case because the plaintiff has shown an absence of deceptive intention under 35 U.S.C. § 256. Dr. Brighton testified that he regarded Redka as an inventor, and Redka's particular participation was specified in an affidavit filed with the Patent Office on May 3, 1974. Redka, along with Dr. Brighton, was a University of Pennsylvania employee and his work was part of the performance of the contract with the Office of Naval Research. Naming Redka as a co-inventor had no effect on the ownership of the patent and could confer no financial benefit on Redka. The burden of proof on the issue of deceptive intent is a negative burden on the plaintiff. There is, however, nothing to suggest any motivation to mislead the Patent Office on this point. A finding of deceptive intent would be purely speculative. Under all of the circumstances, this court is persuaded that the naming of Redka as a co-inventor was done either because the attorneys prosecuting the patent believed that his contribution could be considered as a part of all of the claims of the patent or as a result of a failure of those attorneys to appreciate the requirement that the participation of co-inventors must be with respect to the entire invention. Because the court is persuaded that there was no deceptive intent, the patent should be corrected on notice and hearing pursuant to 35 U.S.C. § 256.

The final issue is the defendants' claim of entitlement to an award of attorneys' fees under 35 U.S.C. § 285. Section 285 permits a court to award attorneys' fees to the prevailing party only in exceptional cases. While this court has agreed *593 with the defendants' contentions that there is no literal infringement; that infringement under the doctrine of equivalents has not been sustained because the prosecution history limits the claims to the use of a skin anode, and that if claim 1 is rewritten as the plaintiff contends, it would be invalid, the court is not persuaded that this infringement action has been prosecuted in bad faith. Moreover, the defendants did not prevail on their affirmative defense and counterclaim of invalidity. The plaintiff's position did receive support from expert witnesses and, indeed, this case has been difficult for this court to decide. Accordingly, there is no basis for finding that this is an exceptional case, and the defendants' claim for attorney's fees is denied.

Upon the foregoing, it is

ORDERED, that judgment shall enter for the defendants on the plaintiff's claims of patent infringement and for the plaintiff on the defendants' affirmative defenses and counterclaims of patent invalidity, and it is

FURTHER ORDERED, that all parties shall show cause in writing to be filed on or before Monday, April 27, 1987 why this court should not direct the entry of final judgment on these claims and counterclaims pursuant to F.R.Civ.P. 54(b).

[1] This section was amended by Pub.L. No. 98-622, effective November 8, 1984. The amendment expressly provides that the co-inventors need not make a contribution to the subject matter of every claim of the patent.