Arthrocare Corp. v. Smith & Nephew, Inc.

United States Court of Appeals for the Federal Circuit

04-1323, -1487

ARTHROCARE CORPORATION,

Plaintiff/Counterclaim Defendant-Appellee,

and

ETHICON, INC.,

Counterclaim Defendant-Appellee,

v.

SMITH & NEPHEW, INC.,

Defendant/Counterclaimant-Appellant.

Jared Bobrow, Weil, Gotshal & Manges LLP, of Redwood Shores, California,
argued for plaintiff/counterclaim defendant-appellee. With him on the brief were
Matthew D. Powers and Perry R. Clark. Of counsel on the brief were Timothy E.
DeMasi, of New York, New York; and Jack B. Blumenfeld and Karen Jacobs Louden,
Morris, Nichols, Arsht & Tunnell, of Wilmington, Delaware.

George F. Pappas, Venable LLP, of Washington, DC, argued for counterclaim
defendant-appellee. With him on the brief were Vicki Margolis and Rebecca G.
Lombard.

Ruffin B. Cordell, Fish & Richardson P.C., of Washington, DC, argued for
defendant/counterclaimant-appellant. With him on the brief were Lauren A. Degnan and
Tina M. Chappell. Of counsel on the brief were Mark J. Hebert and Thomas M.
Johnston, of Boston, Massachusetts. Of counsel was William J. Marsden, Jr., of
Wilmington, Delaware.

Appealed from: United States District Court for the District of Delaware

Chief Judge Sue L. Robinson
 United States Court of Appeals for the Federal Circuit

04-1323, -1487

ARTHROCARE CORPORATION

Plaintiff/Counterclaim Defendant-
Appellee,

and

ETHICON, INC.,

Counterclaim Defendant-Appellee,

v.

SMITH & NEPHEW, INC.,

Defendant/Counterclaimant-Appellant.

___________________________

DECIDED: May 10, 2005
___________________________

Before MAYER, LOURIE, and BRYSON, Circuit Judges.

BRYSON, Circuit Judge.

The term electrosurgery refers to a surgical technique in which high frequency

electrical current is applied to cut or ablate body tissue. There are two forms of

electrosurgical devices, monopolar and bipolar. In monopolar devices, electric current

passes from a single exposed electrode into the body tissue that is to be ablated. The

current then passes through the body to a return electrode, usually attached to the

outside of the patient’s body. In bipolar devices, both electrodes are inserted into the
body. The current passes from one electrode, through the targeted body tissue, and

then back to the return electrode.

Electrosurgery has the benefit of reducing patient bleeding and trauma.

However, there are disadvantages to applying high voltages within the patient’s body,

including the risk that the electrical discharge will cause damage other than to the target

tissue. For that reason, the path of the electrical current through the body needs to be

carefully controlled. Moreover, surgeons prefer to cleanse the surgical area during

arthroscopic procedures with fluids that conduct electricity, such as saline. Therefore,

electrosurgical devices need to be usable in such fluids. The patents at issue in this

case sought to address the problems of controlling the electrical path and enabling

electrosurgical instruments to function in the presence of conductive fluids.

The three patents at issue, U.S. Patent Nos. 5,697,536 (“the ’536 patent”),

5,697,882 (“the ’882 patent”), and 6,224,592 (“the ’592 patent”), are owned by

ArthroCare Corporation. ArthroCare sued Smith & Nephew, Inc., in the United States

District Court for the District of Delaware claiming that Smith & Nephew was liable for

infringement of those patents based on its manufacture of certain electrosurgical probes

and the use of those probes in surgery. In response, Smith & Nephew filed a

counterclaim alleging that ArthroCare and Ethicon, Inc., had violated the antitrust laws

by entering into a conspiracy in restraint of trade. Smith & Nephew’s theory of antitrust

liability was that ArthroCare and Ethicon had settled an earlier dispute in a manner

designed to restrain other competitors from entering the market for electrosurgical

devices and that ArthoCare had brought this action, although knowing it to be

04-1323,-1487 2
objectively baseless, as part of an unlawful conspiracy with Ethicon to interfere with

Smith & Nephew’s business.

Before trial, the district court bifurcated the case. The first phase encompassed

the patent issues of infringement, invalidity, and inequitable conduct. The second

phase addressed damages, willfulness, and the antitrust counterclaim. The court

stayed the second phase until after completion of the trial on the first.

At the conclusion of the patent trial, the jury determined that Smith & Nephew

had directly or indirectly infringed the three patents and that none of the patents were

invalid. Smith & Nephew then moved for judgment as a matter of law and a new trial.

ArthroCare meanwhile moved to dismiss Smith & Nephew’s antitrust counterclaim for

failure to state a claim upon which relief could be granted. Before Smith & Nephew’s

response to that motion was due, the district judge stayed all proceedings on the

antitrust counterclaim while she considered Smith & Nephew’s motions for judgment as

a matter of law and a new trial. The court eventually denied Smith & Nephew’s motions

and entered a permanent injunction against Smith & Nephew. ArthroCare Corp. v.

Smith & Nephew, Inc., 310 F. Supp. 2d 638, 681 (D. Del. 2004). On the same day, the

court granted ArthroCare’s motion to dismiss the antitrust counterclaim before Smith &

Nephew responded to the motion. Smith & Nephew sought reconsideration of the

dismissal order, but the court denied the motion for reconsideration. ArthroCare Corp.

v. Smith & Nephew, Inc., 315 F. Supp. 2d 615, 618 (D. Del. 2004). In denying

reconsideration, the court stated that the order staying proceedings on the antitrust

counterclaim had stayed only discovery and trial of the counterclaim and did not affect

the motion to dismiss. The court further stated that Smith & Nephew’s reliance on a

04-1323,-1487 3
statement by the court in a June 2003 telephone conference was misplaced, and that if

Smith & Nephew “believed that ArthroCare’s motion [to dismiss] was premature and

inconsistent with” the court’s stay order, it should have indicated so, presumably in a

more formal manner. Id. at 318 n.3. The court added that it was “not persuaded that

any argument from Smith & Nephew about the basis for its antitrust allegations will

change the court’s decision.” Id. at 319.

I

On appeal, Smith & Nephew first argues that the district court erred in dismissing

the antitrust counterclaim without giving it an opportunity to respond to the motion to

dismiss or to amend its counterclaim. Following the trial on the patent issues, the

district court continued the stay of the antitrust proceedings pending the disposition of

Smith & Nephew’s motions for judgment as a matter of law and a new trial, and

ArthroCare’s request for an injunction. After the court ruled on those matters, however,

the court dismissed the antitrust counterclaim even though it had not received a

response to the motion to dismiss from Smith & Nephew. The court noted that Smith &

Nephew had not filed a response to the motion and from its silence “presume[d] that

Smith & Nephew does not oppose the motion.” Moreover, the court concluded that the

“sham litigation” aspect of Smith & Nephew’s antitrust counterclaim was baseless. The

court did not address the other ground for the antitrust counterclaim, namely, the

allegation that ArthroCare and Ethicon had entered into a settlement of their dispute that

was designed to exclude other competitors, including Smith & Nephew, from the

relevant market.

04-1323,-1487 4
 Smith & Nephew contends that, because of the stay of proceedings on the

antitrust counterclaim, it never had an opportunity to respond to the motion to dismiss.

In the absence of an opportunity to respond, Smith & Nephew contends that it was error

for the court to grant the motion to dismiss.

In its opinion on reconsideration, the district court characterized the pretrial order

staying proceedings on the antitrust counterclaim as staying discovery and trial but not

the motion to dismiss. While it is true that the written stay order referred only to

discovery and trial, the court elaborated on that order in a June 9, 2003, telephone

conference, in which the court stated that proceedings on the pending motion to dismiss

the antitrust counterclaim were stayed. In response to a question about the pending

motion to dismiss, the court stated that “everything is stayed and we’ll deal with the

antitrust issues later. . . . So the pending motion [to dismiss] on antitrust is stayed and

everything having to do with the antitrust counterclaims, discovery, substantive motions,

et cetera, is stayed pending further order of the court.” In light of that colloquy, it was

reasonable for Smith & Nephew to conclude that the stay order extended to the

proceedings on the motion to dismiss and that it would not be required to respond to the

dismissal motion until the stay was lifted. Thus, the effect of this sequence of events

was that the court granted ArthroCare’s motion to dismiss the antitrust counterclaim

without giving Smith & Nephew an opportunity to respond to the motion.

The Supreme Court has stated that under Rule 12(b)(6) of the Federal Rules of

Civil Procedure, “a plaintiff with an arguable claim is ordinarily accorded notice of a

pending motion to dismiss for failure to state a claim and an opportunity to amend the

complaint before the motion is ruled upon.” Neitzke v. Williams, 490 U.S. 319, 329

04-1323,-1487 5
(1989). The purpose of such a procedure is to enable the plaintiff “meaningfully to

respond by opposing the motion to dismiss on legal grounds or by clarifying his factual

allegations so as to conform with the requirements of a valid legal cause of action.” Id.

at 329-30. Providing the plaintiff with an opportunity to respond “crystallizes the

pertinent issues and facilitates appellate review of a trial court dismissal by creating a

more complete record of the case.” Id. at 330.

The Third Circuit, whose law applies to this procedural issue, has extended that

principle by adopting a categorical rule that “a Rule 12(b)(6) motion for dismissal . . .

may be disposed of only after a hearing, which affords an opportunity to present legal

arguments either orally, in writing, or both at the District Court’s discretion.” Dougherty

v. Harper’s Magazine Co., 537 F.2d 758, 761 (3d Cir. 1976); see also Oatess v.

Sobolevitch, 914 F.2d 428, 430 n.5 (3d Cir. 1990) (a district court may not dismiss a

complaint under Rule 12(b)(6) sua sponte without giving the plaintiff a chance to

respond). In this case, Smith & Nephew did not have an opportunity to respond to the

motion to dismiss, in contravention of that rule.

ArthroCare and Ethicon claim that Smith & Nephew was given the opportunity to

contest the motion to dismiss in the form of its motion for reconsideration, which the

district court denied. That argument is flawed for several reasons. In Dougherty, the

plaintiff also petitioned the district court for reconsideration, yet the Third Circuit

reversed the district court for dismissing the case without giving the plaintiff an

opportunity to respond. 537 F.2d at 761; see also Jordan v. County of Montgomery,

Pa., 404 F.2d 747, 748 (3d Cir. 1969) (finding that “the district court erred in dismissing

[the plaintiff’s] complaint on the defendants’ motions without affording him an

04-1323,-1487 6
opportunity to submit a written statement in opposition to the motions” even though the

plaintiff made a motion for relief from judgment under Rule 60(b)(1)). Additionally, when

it denied Smith & Nephew’s motion for reconsideration, the district court did not conduct

a de novo analysis of the motion to dismiss, but instead applied the highly restrictive

standard applicable to reconsideration motions. See ArthroCare, 315 F. Supp. 2d at

618. The reconsideration process thus did not satisfy the requirement that Smith &

Nephew be given the opportunity “meaningfully to respond” to the motion to dismiss.

On the merits, ArthroCare and Ethicon argue that Smith & Nephew’s

counterclaim should fail because the claim does not describe the antitrust injury

sufficiently and does not provide enough specificity in describing the antitrust violation.

Third Circuit precedent indicates, however, that if a claim fails for lack of specificity, the

district court should grant leave to amend the complaint, regardless of whether the

complainant asks for it. Shane v. Fauver, 213 F.3d 113, 116 (3d Cir. 2000); Borelli v.

City of Reading, 532 F.2d 950, 951 n.1 (3d Cir. 1976). The court should dismiss only if

the complainant is unable or unwilling to amend the complaint. Dist. Counsel 47 v.

Bradley, 795 F.2d 310, 316 (3d Cir. 1986). We therefore vacate the district court’s

dismissal of the antitrust counterclaim and direct the court to allow Smith & Nephew to

respond to the motion to dismiss. If the court concludes, as urged by ArthroCare and

Ethicon, that Smith & Nephew’s antitrust counterclaim fails for lack of specificity, Smith

& Nephew should be given the opportunity to amend.

Because we dispose of the counterclaim issue on a procedural ground, we take

no position on the merits of the counterclaim. However, we note that the district court

did not intend to issue a permanent injunction until after it disposed of the antitrust

04-1323,-1487 7
counterclaim. Because the district court must reconsider that counterclaim on remand,

the permanent injunction against Smith & Nephew must be vacated pending the

disposition of the antitrust counterclaim. See Tegal Corp. v. Tokyo Electron Am., Inc.,

257 F.3d 1331, 1351 (Fed. Cir. 2001).

II

Smith & Nephew next appeals the denial of its motion for judgment as a matter of

law that the asserted claims of the ’536 patent (claims 46, 47, and 56) were anticipated

by a prior art patent, U.S. Patent No. 4,116,198 (“the Roos patent” or “the ’198 patent”),

and an article by the inventor of that patent, Eberhard Roos and a co-author, E.

Elsässer.

As an initial matter, ArthroCare argues that Smith & Nephew is precluded from

arguing invalidity on appeal. ArthroCare maintains that Smith & Nephew did not specify

the basis on which it sought judgment as a matter of law after presenting its evidence at

trial, as required by Rule 50 of the Federal Rules of Civil Procedure. Because of that

failure, ArthroCare claims that Smith & Nephew may not assert invalidity now. That

argument has no merit, however, because the district judge acknowledged that she

precluded argument on the motions for judgment as a matter of law at trial and indicated

that Smith & Nephew’s rights were preserved.

On the merits, the ’536 patent is directed to an electrosurgical system. The three

asserted claims of the ’536 patent all recite an electrosurgical probe “comprising a shaft

having a proximal end and a distal end, an electrode terminal disposed near the distal

end, and a connector near the proximal end of the shaft electrically coupling the

electrode terminal to the electrosurgical power supply.” The claims also recite “an

04-1323,-1487 8
electrically conducting fluid supply for directing electrically conducting fluid to the target

site such that the electrically conducting fluid generates a current flow path between the

return electrode and the electrode terminal.” ArthroCare maintains that neither the

Roos patent nor the Roos and Elsässer article teaches either an electrically conducting

fluid or an electrosurgical probe with a connector at the proximal end connecting the

probe to the power supply. We disagree and hold that the evidence at trial clearly

established that the prior art references disclose both of those features.

A

With respect to the “electrically conductive fluid” limitation, claim 1 of the Roos

patent recites that the claimed electrode is “adapted to be filled with liquid to provide

electrical conductance.” ’198 patent, col. 7, ll. 60-62. ArthroCare posits that there is a

legally tenable distinction between a fluid that provides electrical conductance and “an

electrically conducting fluid.” In particular, ArthroCare argues that while all materials

provide some electrical conductance, most do not possess a sufficiently high level of

conductivity for a person of skill in the art to consider them “electrically conductive.”

ArthroCare’s distinction is belied by the description of “electrically conducting

fluid” in the ’536 patent and by the prosecution history of the Roos patent, which

together make clear that both patents recite a fluid that provides a path for the electrical

current between the electrodes of the electrosurgical devices. The ’536 patent explains

that the conducting fluid provides a “current flow path between the target site and the

return electrode.” ’536 patent, col. 3, ll. 27-30; id., col. 7, ll. 35-46. The inventor of the

’536 patent affirmed that he used the term “conducting fluid” in the ’536 patent to

indicate that the fluid “provides the pathway between the active electrode or electrodes,

04-1323,-1487 9
plural, and the return electrode.” Furthermore, the description of the fluid in the patent

indicates that the conducting fluid facilitates the passage of current by providing a low

electrical impedance current path between the two electrodes. Id., col. 7, ll. 40-43.

The prosecution history of the Roos patent makes clear that the fluid “provid[ing]

electrical conductance” recited in claim 1 of the Roos patent reads on the “electrically

conducting fluid” of the ’536 patent. The Roos patent prosecution history notes that the

washing fluid recited in claim 1 of the Roos patent must “provide the necessary

electrical conductor” between the electrodes and that “there is always a well-defined

current path . . . through the washing (and tissue) fluid.” Thus, the Roos patent

describes a fluid that creates a “current flow path.” That description of the fluid makes

sense given the language of claim 1 of the Roos patent, which recites that the liquid

“provides electrical conductance between said electrodes.” That language means that

the fluid is introduced during electrosurgery to provide conductance and to help

generate a “current flow path.” While it is true that, given enough voltage, an electrical

current can be made to flow through any substance, it would be bizarre to say that a

non-conductor was introduced to “provide electrical conductance.” Consequently, we

conclude that the Roos patent discloses an electrically conducting fluid.

The district court provided three reasons for concluding that the Roos patent

does not teach an electrically conducting fluid. First, the court reasoned that the Roos

patent does not disclose such a fluid because it does not list either saline or Ringer’s

lactate as an example of an electrically conducting fluid. That rationale is unconvincing,

however, because there is no requirement that an anticipating reference must provide

specific examples; rather, the reference need only “be enabling and describe the

04-1323,-1487 10
applicant’s claimed invention sufficiently to have placed it in possession of a person of

ordinary skill in the field of the invention.” In re Paulsen, 30 F.3d 1475, 1479 (Fed. Cir.

1994). Second, the district court focused on the fact that the Roos patent specification

does not distinguish between the fluid used in monopolar devices and the fluid used in

bipolar devices. The court reasoned that because most monopolar devices use

nonconducting fluid, the Roos patent does not clearly teach conducting fluid. That

inference, however, is contradicted by the claim language and prosecution history of the

Roos patent reviewed above. Finally, the court looked to an embodiment described in

the Roos patent in which the probe touches the tissue. The court concluded that there

would be no need for electrical contact with the patient’s tissue if the fluid were

conducting. The court’s analysis, however, focused on only one embodiment in the

Roos patent. There are other embodiments in the patent as to which it is clear that no

such direct contact is necessary, see, e.g., ’198 patent, col. 3, ll. 9-15, and it was error

for the district court to limit the disclosure of the prior art reference to a preferred

embodiment. See Ultradent Prods., Inc. v. Life-Like Cosmetics, Inc., 127 F.3d 1065,

1068 (Fed. Cir. 1997). ArthroCare makes a similar point, contending that another

embodiment described in the Roos patent uses non-conducting fluid. In support of its

argument, ArthroCare points to a later patent by Roos, which describes the foreign

patent to which the Roos patent claimed priority. Like the district court’s analysis,

however, ArthroCare’s argument fails because it addresses only a single embodiment in

the Roos patent.

The Roos and Elsässer article also teaches an electrically conducting fluid. The

article describes how problems with prior art monopolar devices can be eliminated by

04-1323,-1487 11
providing “the high-frequency current a path . . . offering such low resistance that

aberrant currents or leakage currents do not even occur.” The article describes a way

of accomplishing that goal by placing a neutral electrode close to the active electrode in

an irrigation liquid so that current flows through the liquid. The article states that

creating such a current path with the irrigation liquid creates “very good electrical

conditions.” Furthermore, the diagrams in the Roos and Elsässer article depict current

“directly flowing” along a path through the fluid. The description of the role of the

irrigation liquid is quite similar to the description of the role of the conducting fluid in the

’536 patent, which is to provide a “current flow path between the target site and the

return electrode.” ’536 patent, col. 3, ll. 27-30.

ArthroCare maintains that the article does not teach an electrically conducting

fluid because the article uses the term “irrigation liquid” in describing the liquid used in

both the bipolar and the monopolar procedures. As we have noted, most monopolar

procedures use nonconducting fluids. Because the article does not use different names

for the liquids used in the two procedures, ArthroCare contends that there is no way of

knowing if the irrigation liquid is a conducting fluid. ArthroCare’s argument fails,

however, because the article pays little attention to the nature of the irrigation liquid

used in the monopolar prior art. It is unclear whether the liquid in the monopolar

procedure is nonconductive or whether it is even the same liquid that is used in the

bipolar case. What is clear is that, in describing bipolar devices, the Roos and Elsässer

article describes the liquid as providing a path for the current, thus serving as a

conducting fluid. Even giving ArthroCare the benefit of all reasonable inferences, the

04-1323,-1487 12
fact that the article uses the same term to refer to the fluid in both procedures does not

justify an inference that the fluid described in the bipolar procedure is nonconductive.

B

ArthroCare also maintains that the Roos patent and the Roos and Elsässer

article do not disclose “a connector near the proximal end of the shaft for electrically

coupling the electrode terminal to the electrosurgical power supply.” Both the patent

and the article clearly show that the electrodes are coupled to a power source. See,

e.g., ’198 patent, col. 5, ll. 30-35. Hence, in arguing that the prior art does not

anticipate, ArthroCare focuses on the term “connector near the proximal end.”

However, both the Roos patent and the article disclose such a connector.

The Roos patent states that the claimed invention relates to an electrosurgical

device with electrodes and “an insulated cable which can be passed through an

endoscope, to which can be connected the pole of a high frequency generator.” ’198

patent, col. 1, ll. 5-15; see also id., col. 7, ll. 50-51. The district court construed the term

“connector” to mean “a structure that electrically links the electrode terminal to the high

frequency power supply.” The insulated cable in the Roos patent does exactly that.

Specifically, Figure 4 of the Roos patent provides a schematic diagram for the

electrosurgical probes in the patent, and it illustrates that the electrodes are connected

via “output lines” to a high frequency generator. See id., col. 5, ll. 8-9; id., col. 5, ll. 35-

36.

On appeal, ArthroCare appears to accept the district court’s construction of the

term “connector,” but it asserts that the jury could have rejected the contention that a

wire is a connector for the purposes of the ’536 patent. ArthroCare raises various

04-1323,-1487 13
arguments concerning whether a wire is a connector, but those arguments miss the

point. The district court stated that a connector is a “structure” that electrically links the

electrodes and the power supply. That construction of the term “connector” easily

encompasses a wire between the electrodes and the power supply. Because

ArthroCare does not dispute the district court’s construction, ArthroCare’s attempt to

distinguish “wires” from “connectors” fails. The Roos patent clearly depicts a connector

under the district court’s construction.

Furthermore, the Roos patent indicates that the electrical wires that connect the

electrodes to the power source pass through the probe. The specification of the Roos

patent describes one embodiment as having the two electrical leads to the electrodes

“pass outwards from the cylindrical neutral electrode 11, which at 20 are combined to

form a single cable, leading to the rear end of the endoscope 13.” ’198 patent, col. 7, ll.

3-5. In other words, the electrical leads attach to the power source from near the

proximal end of the endoscope. While Smith & Nephew’s expert agreed that the Roos

patent does not explicitly identify the point at which the wires exit the probe, he stated

that a person of skill in the art would understand that the wires would be attached to the

power source after exiting the back end of the probe. See Helifix Ltd. v. Blok-Lok, Ltd.,

208 F.3d 1339, 1347 (Fed. Cir. 2000) (even if a piece of prior art does not expressly

disclose a limitation, it anticipates if a person of ordinary skill in the art would

understand the prior art to disclose the limitation and could combine the prior art

description with his own knowledge to make the claimed invention).

A “connector near the proximal end” of the electrosurgical probe is also found in

the Roos and Elsässer article. In Figure 8 of the article, the electrosurgical probe is

04-1323,-1487 14
drawn in cross-section. The figure shows the active and neutral probes attached to

wires, and the labels state that those wires go to the power supply. Therefore, a

connector is disclosed in the Roos and Elsässer article as well. Moreover, Figure 10 of

the article illustrates how those wires leave the probe. That figure is a schematic

diagram, which depicts the cutting and neutral electrodes inside the body and projecting

from the “resectoscope shaft” on one end. The wires also project from the other end of

the shaft, where they connect to a high frequency power source. Thus, the connectors

are shown exiting near the proximal end of the probe. The article even provides a

photograph of the instrument depicted in Figure 10. The photograph verifies that the

electrical leads leave the probe at its back end. In contending that the article does not

disclose a connector near the proximal end of the probe, ArthroCare limits its argument

to contesting the veracity of Smith & Nephew’s expert and disputing his conclusions

regarding the presence of the electrical connector in the article. However, the article

speaks for itself, and it clearly discloses such a connector.

In sum, Smith & Nephew has proved by clear and convincing evidence that the

asserted claims of the ’536 patent were anticipated either by the Roos patent or the

Roos and Elsässer article. Because the jury’s determination that the ’536 patent was

not invalid is not supported by substantial evidence, we reverse the district court’s

denial of Smith & Nephew’s motion for judgment as a matter of law on that issue.

III

Smith & Nephew also appeals the denial of its motion for judgment as a matter of

law that the ’882 patent is invalid because the claims of that patent were impermissibly

broadened by a certificate of correction. In particular, Smith & Nephew argues that

04-1323,-1487 15
claim 1 of the ’882 patent required three electrodes when it was originally issued, but

that after the correction the claim required only two electrodes. Smith & Nephew

contends that the change impermissibly broadened the patent’s scope.

When ArthroCare originally filed the application that matured into the ’882 patent,

the claims recited only an “active electrode” and a “return electrode.” Before any

examination on the merits, ArthroCare changed the claims by making what it termed “a

few minor amendments.” Those amendments changed the term “active electrode” to

“electrode terminal” in three places in claim 1 of the application, but did not make the

change in a fourth place, where the term “active electrode” was left unchanged. The

prosecuting attorney noted the error on the same day that the patent issued and

immediately asked the Patent and Trademark Office to change the remaining reference

from “active electrode” to “electrode terminal.” The prosecuting attorney testified at trial

that the change listed in the certificate of correction was made solely due to a

typographical error. Smith & Nephew did not attempt to rebut that evidence.

The correction of a ministerial error in the claims, which also serves to broaden

the claims, is allowable if it is “clearly evident from the specifications, drawings, and

prosecution history how the error should appropriately be corrected” to one of skill in the

art. Superior Fireplace Co. v. Majestic Prods. Co., 270 F.3d 1358, 1373 (Fed. Cir.

2001). At trial, Smith & Nephew sought to show that the requisite standard was not met

in the case of the correction to the ’882 patent. Smith & Nephew’s proof on that issue

failed to satisfy the jury, and we hold that substantial evidence supports the verdict.

In the first place, claim 1 of the ’882 patent does not make sense if it is

interpreted to contain three types of electrodes instead of two. The claim requires that

04-1323,-1487 16
an electrode terminal and a return electrode be coupled to a high voltage source. The

claim as originally issued then required that an “active electrode” be placed in close

proximity to the target site. High frequency voltage is then applied between the

electrode terminal and the return electrode, which induces the discharge of energy to

the target site. Nothing in the patent suggests any reason to place a third type of

electrode close to the target site. The whole point of the patent is to use the electrode

terminal and return electrode to apply a voltage across the tissue; a third type of

electrode would serve no apparent purpose. Moreover, the specification refers to

“electrode terminal” and “active electrode” interchangeably. See ’882 patent, col. 20, ll.

19-21; id., col. 20, ll. 53-54. That evidence indicates that it was clear how the

typographical error in the original claims should have been corrected.

The prosecution history further supports ArthroCare’s argument that it was

unambiguous how the remaining reference to an active electrode in claim 1 should be

changed. From the beginning, the claims referred to only two electrodes. The change

of the term “active electrode” to “electrode terminal” was made before any examination

on the merits, and the uncontroverted evidence establishes that it was meant to be a

global renaming. In fact, most of the references to “active electrode” in the claims were

changed. Finally, ArthroCare presented unrefuted testimony from an expert who stated

that he understood the term “active electrode” in the uncorrected claim to refer to the

“electrode terminal.”

Smith & Nephew’s only evidence that it remained unclear how to fix the error in

claim 1 is that claim 53, which depends on claim 1, also refers to “the active electrode.”

According to Smith & Nephew, that evidence implies that it cannot be apparent how to

04-1323,-1487 17
fix the remaining instance of “active electrode” in claim 1, because changing it to

“electrode terminal” would leave claim 53 without an antecedent basis. In fact,

however, a simple explanation for the use of the term “active electrode” in claim 53 is

that the prosecuting attorney made another error in claim 53 of the same type that was

corrected in claim 1. The prosecuting attorney’s failure to replace the term “active

electrode” twice in the claims, instead of once, does not demonstrate by clear and

convincing evidence that a person of ordinary skill in the art would not understand how

to correct those errors. Accordingly, substantial evidence supports the jury’s conclusion

that the certificate of correction was valid. We therefore affirm the district court’s denial

of judgment as a matter of law of invalidity of the ’882 patent.

IV

Finally, Smith & Nephew appeals the denial of its motion for judgment as a

matter of law that it was not liable for indirect infringement of the ’592 patent. The ’592

patent pertains to a method for conducting electrosurgery. The method comprises

positioning an electrode terminal near the target site in the presence of an electrically

conductive fluid. Next, a return electrode is positioned in the fluid, while ensuring “that

the return electrode is not in contact with the body structure.” ’592 patent, col. 24, ll. 13-

14. Finally, high frequency voltage is applied between the electrode terminal and the

return terminal so as to force current to flow into the target site. Smith & Nephew

argues that it is not liable for contributory infringement or inducement of infringement of

the ’592 patent, because there was no evidence that its probes were ever used in a

manner that directly infringed the patented method. Smith & Nephew maintains that

none of the videotaped surgical procedures using its probes infringed the patented

04-1323,-1487 18
method because in every case the return electrode was shown touching “the body

structure.” Smith & Nephew asserts that the jury erred in finding infringement because

ArthroCare convinced the jury to disregard the district court’s claim construction.

In construing the claims of the ’592 patent, the district court instructed the jury

that the return electrode “is not to contact the body at all during the performance of the

claimed method.” The court noted, however, that “[t]he claimed method does not

contain any time limitations. Thus, the claimed method is performed when each of the

three steps of the claim has been completed.” Smith & Nephew does not challenge that

claim construction.

Based upon the district court’s claim construction, the jury was free to find

infringement if it concluded that the return electrode did not touch the body when each

step of the patented method was being performed. There was no need for the electrode

to be kept apart from the body throughout the entire surgical procedure; nothing in the

claim language or in the court’s claim construction required that the electrode not touch

the body at any time between the performance of the steps of the claimed process.

That is in effect what the district court advised the jury when it instructed that the

claimed method does not contain time limitations and that the claimed method is

performed when each of the three steps is completed.

Smith & Nephew interprets the court’s claim construction to require that the

return electrode never touch the body until all the claimed steps are completed. That

interpretation, however, is not faithful to the claim construction that the trial court

adopted, and it is not a convincing interpretation of the claim language. When the

district court construed the claim language at issue here, it rejected Smith & Nephew’s

04-1323,-1487 19
proposed construction, which was that the return electrode must never touch the body

at any time during the surgery. The court properly rejected that proffered claim

construction on the ground that it imposed an unclaimed temporal requirement on the

method. In effect, Smith & Nephew is now advancing that rejected claim construction,

while maintaining that it has accepted the district court’s construction. We uphold the

district court’s claim construction and reject Smith & Nephew’s argument that the court’s

construction was actually a version of the very construction that the court rejected

before trial.

Substantial evidence at trial showed that Smith & Nephew’s probes were used so

that the return electrode did not touch the body at a time when all the other claim

limitations were met. ArthroCare’s expert stated that during surgery “the return

electrode is positioned back . . . so that you try to make sure that it’s not in contact” with

the body. Even Smith & Nephew’s expert admitted that there were instances in which

the return electrode was not in contact with the body during certain steps of the claimed

method. Additionally, upon viewing the videotaped electrosurgeries, project managers

for two of Smith & Nephew’s accused probes admitted that the return electrodes were

not in contact with body tissue during use.

There was also strong circumstantial evidence that Smith & Nephew’s probes

were used in an infringing manner, and that Smith & Nephew induced users to employ

the probes in that way. Smith & Nephew’s witnesses confirmed that the return

electrodes of the accused probes were not designed or intended to contact the body

tissue when power was being applied to the device. That evidence was supported by

testimony from ArthroCare’s expert. Moreover, the sales literature accompanying one

04-1323,-1487 20
of the accused devices instructs surgeons that “care should be taken to prevent tissue

contact with the return electrode.” That literature explains why the surgeon should

avoid touching the return electrode to the body tissue. Even though the return electrode

on the accused probe is enlarged so as to lower the return current density and thus

reduce the risk of burns, the return electrode of the Smith & Nephew device was still not

supposed to touch the body during the application of power because “[w]hile it will not

be as hot as the active electrode at the distal tip, the return electrode may become

heated. For this reason, it is important to avoid inadvertent contact with the tissue.”

Instruction manuals for the other accused probes similarly confirm that the return

electrode should be completely surrounded by or immersed in saline during use. Thus,

substantial evidence supports the jury’s determination that Smith & Nephew indirectly

infringed the claimed method. We therefore affirm the district court’s denial of judgment

as a matter of law with respect to the ’592 patent.

Each party shall bear its own costs for this appeal.

AFFIRMED IN PART, REVERSED IN PART, VACATED IN PART, and REMANDED.

04-1323,-1487 21