Ramos v. Boehringer Manheim Corp.

MEMORANDUM OF DECISION AND ORDER MISHLER, District Judge. Dr. Pedro Ramos instituted this action against Boehringer Manheim Corporation1 seeking damages resulting from the defendant’s alleged infringement of claims one, — 1067 — three and five of United States Patent No. 4,380,090 (“090 patent”), a patent issued to Dr. Ramos and owned by him throughout the pendency of this action. DePuy denies the validity of the patent and further denies infringement. DePuy also asserts the affirmative defense of prosecution history or file wrapper estoppel. The case was tried to the court. This memorandum contains findings of fact and conclusions of law pursuant to Fed.R.Civ.P. 52(a). For the reasons discussed herein, the complaint is dismissed and judgment is entered in favor of the defendant. Background Pedro Ramos is an orthopedic surgeon who invented an artificial hip prosthesis for which a United States Patent was issued on April 19, 1983, and for which a re-examination certificate was issued on September 21, 1993. DePuy is a manufacturer of prosthetic devices for use in orthopedic surgery, including, but not limited to, the DePuy Self-Centering Universal Hip, the presently accused device. The claims of the patent which are relevant to the disposition of this case are set out below: Claim 1: An artificial hip joint comprising an artificial hip socket having a first cavity and an opening in a surface of the socket communicating with the first cavity, an annular groove formed in the first cavity adjacent said opening, a sectionalized bearing insert registerably positioned in the first cavity inwardly of said groove, said bearing insert having a second cavity of spherical configuration greater in scope than hemispherical, a femoral component having a ball extending from a neck of reduced diameter, said neck extending through said socket opening positioning said ball in operative, retained engagement in said second cavity, said bearing insert having inner and outer sections, each section being formed with a complementary component of said second cavity, the cavity component of said inner section being approximately hemispherical, said outer section being annular in shape and having an outer surface portion opposite the cavity component thereof adapted to align with said annular groove when said outer section is in operative position in said first cavity, and an open annular spring locking ring having opposite ends formed with tool engaging openings, said locking ring being removably engaged in said annular groove in abutment with said outer surface portion of the bearing insert outer section, said locking ring, when in said groove engagement being visibly exposed and removable retaining said annular outer section in said first cavity in operative engagement with the inner section whereby said ball is retained in said second cavity, said annular outer section being formed to expand over said ball when the latter is removed from said first cavity. Claim 3: The artificial hip joint defined in claim 1 in which said annular outer section of the bearing insert is formed as a split ring for said expansion over the ball. Claim 5: The artificial hip joint defined in claim 1 in which said hip socket has a smooth outer surface to engage a patient’s acetabulum in serving as a poly-centric prosthesis. The Development of Bipolar Prostheses The design and manufacture of hip prostheses has evolved over the years into a multi-million dollar industry. As surgeons have become more familiar with the shortcomings and limitations of various prosthetic designs, manufacturing interests such as De-Puy have sought to maintain a reactive posture in the market, continually exploring new ideas and designs in an attempt to produce prostheses responsive to the needs of the surgical community. The desire to maximize the benefits and reduce the complications associated with joint replacement helped lead to the development of bipolar prosthetic hips, of which both the Ramos and DePuy self-centering hips are examples. Bipolar hip replacement is a type of hip replacement that allows relative movement between the outer shell of the insert and the acetabulum, or hip socket, as well as between the artificial femur head, or ball, and the — 1068 — interior of the insert. Unlike in a total hip replacement, the shell of the insert is not adhered to the acetabulum in a bipolar replacement, but rather the shell is permitted freedom of movement within the acetabulum. The first bipolar device was designed in the 1970’s by Dr. Giliberty. The goal of the bipolar hip was to eliminate or at least decrease the amount of wear on the acetabulum which resulted from partial hip replacements. During a partial hip replacement, only the femur head is replaced and fitted into the acetabulum, resulting in wear on the natural hip socket over time. The bipolar hip was specifically designed to limit the wearing effect of the artificial femur head by cushioning it with an insert and a shell. Thus, the bulk of the movement would occur between the artificial femur head and the insert as opposed to between the artificial femur head and the acetabulum. Among the bipolar prostheses whose development bears on this ease is the Bateman device, a hip prosthesis manufactured pursuant to the Averill patent (U.S. Patent No. 3,863,273). As one of the earliest bipolar devices, the Bateman device forged new inroads into the prosthesis arena that were replete with new difficulties. Among them was the tendency for dislocation. The Bate-man device consisted of a metal acetabular cup and a one piece plastic insert which fit into the cup. During certain movements the insert would become dislodged from the cup and further surgery would be required to repair the hip. Another problem with the Bateman device was the dislocation of the cup from the acetabulum. Yet another difficulty was the tendency for insert breakage. It was not long before the industry sought to improve on the Bateman device. One such improvement came by way of the creation of self-centering prostheses, designed to mitigate the dislocation problems associated with the Bateman device. By the 1980’s self-centering bipolar devices had achieved significant success in the market place, and such devices were manufactured by most major orthopedic manufacturers, including but not limited to Osteonics, Howmedica, Zimmer and DePuy. DePuy’s device was manufactured pursuant to a license issued by Drs. Pappas and Buechel, the owners of United States Patent No. 4,619,658. It consists of a polished metal cup and a two piece insert, consisting of a plastic bearing connected to a plastic collar by a ridge on the bearing which fits into a groove in the collar. The DePuy hip arrives from the factory with the insert sub-assembled, that is, the bearing and collar are attached to one another via the ridge and groove. Although pre-assembled as a sub-assembly, the bearing and collar are two distinct pieces. They are, however, extremely difficult to separate, and in some sized devices, their separation is impossible. Removal of the outer collar from the bearing is impossible when the insert assembly is in place within the shell. The Development of the Ramos Hip (090 Patent) In April of 1978, Dr. Ramos first saw the Bateman device, and started to think of ways in which he could improve on some of the device’s shortcomings, particularly its problem with dislocation. By early 1980, Ramos contacted the law firm of O’Brien & Jacobsen to assist him in securing a patent for the invention that resulted from this intellectual exercise. The first patent application was filed on August 13, 1980, (Serial No. 177791) and differed significantly from the device ultimately patented. The August 13 application disclosed a device which consisted of an outer shell and a one piece liner which contained a finger-like retaining mechanism for femur head retention. The liner was held in the shell by means of a locking ring. All eleven claims of Ramos’ application of August 13, 1980 were rejected on July 27, 1981 as obvious in light of prior art, including the Averill (U.S. Patent No. 3,863,273) and D’Errieo (U.S. Patent Nos. 4,044,403 and 4,172,296) patents, among others. On November 25, 1981 Ramos’ then attorney, Joseph Zallen, submitted to the Patent Office a communication cancelling claims one through eleven of the August 13,1980 application and substituting new claims twelve through fifteen. On February 15, 1982 Zallen further requested that the Patent Office cancel — 1069 — claims twelve through fifteen due to inadvertent errors and substitute in their stead claims sixteen through twenty. The new claims described an artificial hip which also had a one-piece insert, but which retained the femur head in the insert by means of a locking ring. The new claims, like their predecessors, were rejected. The reason for their rejection was obviousness in light of prior art, including the Noiles (U.S. Patent No. 3,848,272), Averill and D’Errieo patents, as well as anticipation by the Noiles patent. Due to Ramos’ failure to respond to the rejection, the claims were deemed abandoned by the Patent Office on August 30, 1982. During the time Ramos was amending his claims and substituting new claims under the August 13, 1980 application, he also filed another application to have his prosthesis patented. The file date for the second application, (Serial No. 286532), which was a continuation in part2 of the initial application, was July 24, 1981. The July 24 application itself contained six claims, describing a hip prosthesis that consisted, among other features, of an outer shell, a two piece liner insert and a locking ring. The claims were rejected as obvious in light of the prior art by means of a communication of the Patent Office mailed June 8, 1982. The application was deemed abandoned by the Patent Office in November of 1982. On March 10, 1982 Ramos filed yet another patent application (Serial No. 356881), which was a continuation in part of his prior continuation in part. This application consisted of six claims and disclosed a hip prosthesis which consisted of an outer shell, a two piece insert which functioned as a bearing, and a locking ring to retain the femoral head within the bearing insert. In a communication mailed on August 5, 1982, the Patent Office indicated its rejection of the application, again due to obviousness in light of the prior art. On August 23, 1982 Ramos submitted a document to the Patent Office requesting that new claims seven through twelve be substituted for the six claims previously rejected. On November 8, 1982 the Patent office wrote back to Ramos and instructed him that the amendment submitted on August 23 was non-responsive to the rejection- because it failed to respond to all grounds of rejection. This prompted Ramos to file a response on December 3, 1982, informing the Patent Office that the substitution of new claims seven through twelve was not meant as a response to the rejection, but as the presentation of a new set of claims for review by the Office. These were the claims upon which the 090 patent was ultimately issued. The Ramos patent discloses a hip prosthesis which consists of a metal shell lined with a two piece plastic bearing insert, each piece of which serves a bearing function. The bearing insert is held in the shell by a metal locking ring, which is removable by inserting a forceps-like tool into holes on the ring. The design of the Ramos hip facilitates disassembly of the prosthesis so that various component parts may be replaced with relative ease. When the locking ring is released, the outer bearing and femur head are able to fall away from the inner bearing, leaving the inner bearing in the outer metal shell. Thus, Ramos’ patent claims require that the bearing be composed of two pieces and may not be a monolithic structure, as disclosed in his previous applications. In his attempt to secure the 090 patent, Ramos made various arguments in support of the patentability of the claim contained therein. Among them was that it was essential that the hip be convertible from a partial to a total hip replacement without having to change the femoral component. This, Ramos argued, distinguished his hip from the prior art disclosed in the Noiles and Shersher patents, which disclosed only total hip prostheses. One argument made by Ramos in support of his patent which is of particular signifi — 1070 — canee to the disposition of this ease concerned the distinct outer bearing and locking ring components of his invention. In distinguishing prior art, including Shersher, Noiles and Buechel, Ramos stated in pertinent part: Noiles bearing 30 and its manner of spitting [sic] is entirely incompatible with the insert of Buechel et al. or insert 5 of Shersher, while ring 13 of the latter is neither an outer bearing section or a locking ring but in a sense is a combination of the two and does not lend itself to splitting as defined in claim 9 or forming as two separate members as defined in claim 10. We decline the plaintiffs invitation to interpret this argument as limited to claims 9 and 10 of the patent. Rather we understand this statement by Ramos to distinguish the prior art from all of his claims. The fact that the prior art design would not lend itself to the modifications discussed in claims nine and ten is merely an example of why the Ramos invention, as disclosed in what is now claim one of the 090 patent, could not consist of a one piece structure. Were we to interpret the argument differently, we would reach the illogical conclusion that for purposes of claim one, the two components could be merged and be distinguishable from the prior art, but for purposes of claims nine and ten they could not. If the two piece design is distinguishable from the prior art for one claim, it must be for all claims, otherwise it is not really different from the prior art at all. Thus, as discussed in a subsequent portion of this opinion, we conclude that in light of Ramos’ argument, the 090 patent requires the locking ring and outer bearing component be two distinct pieces and a unitary structure does not fall within the scope of the patent. On June 5,1992, DePuy requested that the 090 patent be reexamined in view of prior art. The Patent Office issued a rejection of the claims contained in the 090 patent on December 11, 1992. Ramos challenged the rejection on February 16, 1993. The Patent Office reversed its rejection in a communication dated April 17, 1993. One of the justifications for holding the patent valid was Ramos’ ability to convince the Patent Office of the uniqueness of utilizing a locking ring to hold the outer piece of the two piece bearing in place. In holding the claims valid on April 17, the patent examiner stated in pertinent part: “[I]t was never the patent owner [sic] intention to claim a unitary liner but rather a liner having two separate inner and outer components---- This rejection failed to appreciate the invention as a whole. (Emphasis in original). Dr. Ramos utilize [sic] subtle changes which collectively allowed quick and easy interchangeability between a polycentric joint and a total joint. These changes include an inner bearing component that has a cavity greater than hemispherical and an open annular spring locking ring which engages the outer liner (emphasis added) for removably retaining the outer liner in the first cavity in operative engagement with the inner liner. Among the factors considered by the Examiner in finding the claims patentable was Ramos’ argument that the bearing had to consist of separate inner and outer components. He argued that “[i]n order to interpret the sectionalized bearing insert of claim 1 as allegedly encompassing a one-piece structure (Averill), the emphasized language of claim 1 has to be ignored. In considering the patentability of a claim its language must be considered as a whole, including all limitations therein, even functional limitations which define the relationships among the recited elements. It is improper to ignore limitations in a claim.” He emphasized that the unique design of his prosthesis permitted the surgeon to release the locking ring and remove the ball and outer insert from the inner insert. Furthermore he stated that “the claim 1 language must be interpreted as requiring a separate, removable outer section, as shown in Fig. 1 of the 090 patent, to achieve the advantages attributed to the presently claimed invention....” The Ramos—DePuy Correspondence Ramos argues that DePuy’s infringement of his patent stemmed from its misappropriation of his invention, which he revealed to them through correspondence beginning in the latter half of 1979, and conversations with a DePuy representative which occurred — 1071 — around the same time. It was in August of 1979 that Ramos first officially approached DePuy and inquired whether it would be interested in pursuing commercial production of the bipolar device Ramos had conceived. DePuy did not yet have its self centering bipolar hip on the market. Although Ramos’ invention was not yet patented at the time, a confidentiality agreement was executed between DePuy and Ramos. Throughout 1979 and 1980 Ramos continued to modify his design, changing his prime focus at various times from retention mechanisms to bearing design. Throughout the period during which Ramos worked on the development of his design, he kept in contact with DePuy either directly or through letters from his attorney; however, the designs for the prosthesis as eventually patented were never sent to DePuy. On October 1,1980, Ramos’ attorney wrote to Mr. Keeven, DePuy’s general counsel, informing him that Ramos was working on an improved design. Crary included with his letter copies of the engineering drawings that had been submitted along with the initial patent application. Keeven responded on November 6, 1980 with a letter to Crary expressing DePuy’s concern that the design proposed by Dr. Ramos appeared to infringe on existing patents, including the Averill and Giliberty patents. Several letters were sent by Crary to Keeven updating him on Ramos’ attempts to secure a patent for his invention, and assuring him that Ramos’ ultimate design would be unique and would not infringe on any existing patents. An April 1, 1981 letter from DePuy to Crary, Ramos’ attorney, informed Ramos that DePuy was itself in the process of developing a new hip prosthesis. In a letter dated October 28, 1981 DePuy made clear to Ramos through its attorney Mr. Keeven that it was not interested in Ramos’ design and it was going to pursue the development of bipolar prostheses independent of Dr. Ramos’ contributions. In November of 1982 Ramos attended a convention of the Academy of Orthopedic Surgeons in New Orleans where DePuy was an exhibitor. It was at this convention that Ramos first saw the DePuy device and felt it was in fact his invention “with a facelift.” Ramos conveyed his concerns to DePuy and DePuy responded by a letter from Mr. Keeven, DePuy’s counsel, to Mr. Crary, Ramos’ attorney. In that letter, Keeven included dated drawings indicating that DePuy had been working on a bipolar prosthesis, including a retaining ring, since April of 1980, some six months before Ramos disclosed his invention to DePuy. A second letter was sent to Ramos on May 9, 1983 indicating DePuy’s assumption that in the absence of further correspondence Ramos’ concerns of infringement were alleviated. The patent under which DePuy’s accused device is licensed was not issued until October of 1986. The Development of the DePuy Device In the fall of 1979 Dr. Bueehel convinced DePuy to permit him to develop a bipolar hip prosthesis for the company. Dr. Pappas worked with Dr. Bueehel throughout the design process. DePuy produced a set of drawings at trial which were made in connection with the development of the prosthesis and bore a date' of October 2, 1979, initialled by Drs. Pappas and Bueehel. The drawings included several possible designs which were later refined. These initial drawings revealed prostheses consisting of split bearing inserts held in place by metal locking rings. As the design process progressed, various additional sketches were submitted to DePuy executives detailing the progress in the design development. Periodically the design shifted from one containing a locking ring to a ringless version with a split collar similar to the one contained in the accused device. Ramos’ Shifting Understanding of the Essence of His Invention Throughout the course of Ramos’ dealings with DePuy and throughout the design process in general, Ramos’ notion of the nub of his invention varied periodically. Ramos testified that the essence of his original invention was the locking ring mechanism for retaining the bearing insert in the acetabular cup, although he acknowledged that prior to his invention several other orthopedic manufacturers had already developed locking ring devices for their prostheses. According to — 1072 — Ramos, his was more effective than the ones previously designed. Ramos also testified, when questioned about DePuy’s notification of possible infringement of the Averill and Giliberty patents, that his subsequent designs constituted a whole new prosthesis as opposed to a mere locking device. He conceded, however, that the designs for the entirely new prosthesis were never sent to DePuy. The only designs Ramos sent to DePuy were designs which were eventually rejected by the patent office and which concentrated more on the locking mechanism than on a unique prosthetic device. Once Ramos improved his invention from a mere locking mechanism to an entirely new prosthesis, Ramos’ concept of the essence of his invention seems to have shifted from the locking device to the structure of the bearing as well as the removal holes in the locking ring which permitted the removal of the outer bearing while keeping the inner bearing and ball in place. Such a notion was confirmed by the testimony -of Ramos’ attorney, Mr. Crary. He testified that based on his discussions with Ramos, he understood the essence of the Ramos invention to be the unlocking procedure. The presence of the holes for forceps removal of the plastic insert from the metal shell was the critical component of the invention. In fact, Crary testified that if the DePuy device did not have the holes, in his opinion there would be no infringement of the Ramos patent. This design was supposed to facilitate the conversion of a bipolar hip to total hip replacement. Finally, in arguing in favor of the patentability of the claims in the 090 patent, Ramos urged that the essence of his invention was the ability to convert from a bipolar to a total hip replacement with ease. He wrote the following to the Patent Office: One object of applicant’s invention is to provide a hip prosthesis construction of ultimate simplicity which will facilitate a onversion [sic] from a polycentric prosthesis to a total hip by disassembling the femoral component from the smooth surfaced socket of the polycentric, cementing a total hip socket replacement into position and reassembling the two components without requiring replacement of the femoral component. Discussion I. Validity Our initial task is to address the validity of the 090 patent. The validity of this patent previously has come under scrutiny both before the Patent Office, as earlier discussed, as well as in the course of another law suit. In Ramos v. Biomet, Inc., 828 F.Supp. 1570 (S.D.Fla.1993) appeal pending, No. 94-1004 (Fed.Cir.1994) Judge Aronovitz held the 090 patent valid. His determination of validity was not appealed to the Federal Circuit. Early in this ease, DePuy’s counsel informed the court that it had no intention of pursuing the validity issue as it had been litigated before Judge Aronovitz. Thus, there would be no attempt to prove Ramos’ invention was obvious in light of the prior art, but, rather defense counsel assured the court that the challenge to the 090 patent’s validity would be limited to an attempt to prove prior invention by DePuy’s designers.3 All patents enjoy a statutory presumption of validity pursuant to 35 U.S.C. § 282. A patent which has survived a reexamination proceeding is entitled only to the same presumption of validity which attaches to any patent; there is no heightened presumption of validity awarded to a patent surviving re-examination. Kaufman Co., Inc. v. Lantech Inc., 807 F.2d 970, 974 (Fed.Cir.1986). The burden which must be overcome when challenging the validity of a patent is that of clear and convincing evidence. American Hoist & Derrick Co. v. Sowa & Sons, Inc., 725 F.2d 1350, 1360 (Fed.Cir. — 1073 — 1984), cert. denied, 469 U.S. 821, 105 S.Ct. 95, 83 L.Ed.2d 41 (1984). Notwithstanding the prior decision of this court or the outcome of the re-examination proceeding, the validity of the 090 patent is dubious at best when the invention contained therein is considered in light of the prior art. The fact that the claims were initially rejected and were rejected again during the initial phase of the re-examination proceeding is telling. Nonetheless we limit our analysis of validity to the priority of the invention as between Pappas and Buechel on the one hand and Ramos on the other. We find that the defendant has failed to carry its burden of proof that the patent is invalid based on the prior invention of Drs. Pappas and Buechel. While evidence was presented which tended to prove Pappas and Buechel were working on the development of prostheses similar to that disclosed in the 090 patent contemporaneously with Ramos, we remain unconvinced that Pappas and Buechel’s invention pre-empted Ramos’ work such that Ramos’ patent would thereby be rendered invalid. Hence, we conclude that for the purposes of this litigation, Ramos’ 090 patent is valid. II. The Scope of the 090 Patent In an action for patent infringement, the court’s task is two-fold. As an initial step we must determine the scope of the patent. Only then may we turn our attention to the existence of infringement. McGill Inc. v. John Zink, Co., 736 F.2d 666, 671 (Fed.Cir.1984) (citing SSIH Equipment S.A. v. U.S. Intern. Trade Com’n, 718 F.2d 365, 376 (Fed.Cir.1983)), cert. denied, 469 U.S. 1037, 105 S.Ct. 514, 83 L.Ed.2d 404 (1984). The scope of the claims is a question of law, Intellicall, Inc. v. Phonometrics, Inc., 952 F.2d 1384,1387 (Fed.Cir.1992), which the court is to determine by examining the claim language, the specifications contained in the patent and the prosecution history of the patent. SmithKline Diagnostics, Inc. v. Helena Laboratories Corp., 859 F.2d 878 (Fed. Cir.1988); ZMI Corp. v. Cardiac Resuscitator Corp., 844 F.2d 1576 (Fed.Cir.1988). In considering the claim language, words are given their ordinary meaning unless there is evidence that the patentee used the terms differently. Intellicall, Inc. v. Phonometrics, Inc., 952 F.2d 1384, 1387 (Fed.Cir.1992). “It is entirely proper to use the specification to interpret what the patentee meant by a word or phrase in the claim.” E.I. Du Pont de Nemours & Co. v. Phillips Petroleum Co., 849 F.2d 1430, 1433 (Fed. Cir.), cert. denied, 488 U.S. 986, 109 S.Ct. 542, 102 L.Ed.2d 572 (1988). However, this court is cognizant of the admonition of the Federal Circuit relating to our authority to utilize the specification as an interpretive aid. “[I]nterpreting what is meant by a word in a claim ‘is not to be confused with adding an extraneous limitation appearing in the specification, which is improper.’ ” Intervet America, Inc. v. Kee-Vet Laboratories, Inc., 887 F.2d 1050, 1053 (Fed.Cir.1989) (quoting Autogiro Co. of America v. U.S., 384 F.2d 391, 395-96, 181 Ct.Cl. 55 (1967). A court may not read the contents of the specification into the claim, but the specification may be used in interpreting the language of the claims. Sjolund v. Musland, 847 F.2d 1573, 1581 (Fed.Cir.1988). Just as the contents of the specification may be utilized as an interpretive aid, so too may a court look to the prosecution history of the patent for interpretive guidance. McGill Inc. v. John Zink Co., 736 F.2d 666, 673 (Fed.Cir.1984). The contents of the file wrapper may not, however, be used to make the claims into what they are not. Carroll Touch, Inc. v. Electro Mechanical Systems, Inc., 3 F.3d 404, 409 (Fed. Cir.1993). In interpreting the claims presently before the court, we first look to the language of the claims themselves. In so doing, we find as a matter of law that the Ramos 090 patent requires the locking ring be a separate component from the outer bearing. The claim language discloses that the “locking ring be[ ] ... in abutment with said outer surface portion of the bearing insert outer section----” We reject the plaintiff's argument that a single component is in the “ultimate form of abutment” with itself. The ordinary meaning of abutment conveys the understanding that two distinct items are in a certain relation to one another. It describes the position of one component with respect to the position of another component. Interpreting the claim broadly — 1074 — enough to encompasses within its breadth a one piece device would require us to ignore the specific claim limitation that the bearing and ring be in a certain position with respect to one another. Although we base our interpretation that the locking ring and outer bearing be two separate pieces on the ordinary meaning of “abutment”, our holding is further supported by the prosecution history of the 090 patent. When Ramos attempted to distinguish the prior art as disclosed in the Shersher patent, in a letter of December 3, 1982 to the Patent Office, he argued that unlike the ring disclosed in his patent, the ring disclosed in the Shersher patent was “neither an outer bearing section or [sic] a locking ring but in a sense is a combination of the two____” As indicated above, we fail to limit this argument to only certain claims of the 090 patent, but rather interpret it to apply to all the claims. Claim one of the 090 patent also requires that the locking ring be “visibly exposed and removably retain[ ] said annular outer section in said first cavity____” Were we to interpret this claim as encompassing devices where the locking ring and outer bearing were one piece, we would be ignoring completely the language that describes the relationship between the two components. Furthermore, the locking ring is required to be visibly exposed. This is a characteristic of the device which would be conspicuously absent where the ring is integral to another component. We also interpret claim one to extend only to devices in which the outer bearing may be removed from the insert while leaving the inner bearing inside the acetabular shell. Ramos made much of the concept of “removably retaining.” The specification emphasizes that one of the unique features of the Ramos hip is the ability to readily remove and change certain components of the insert as necessary. The improvement over prior art which this represents mandates that the inner and outer bearing insert be separate and held in operative engagement by the removable locking ring. Interlocking the two components such that one may not be removed while leaving the other in place is not within the scope of the claim. Our conclusion is supported by the plain language of the claim: said locking ring ... being visibly exposed and removably retaining said annular outer section [of the bearing insert] in said first cavity [in the acetabular shell] in operative engagement with the inner section whereby said ball is retained in said second cavity. This interpretation of the claim is also supported by the specification, the drawings and the prosecution history of the patent. In defending the 090 patent during the re-examination proceedings, Ramos argued that [C]laim 1 necessarily define[s] the sectionalized bearing insert as comprising a separate inner section and a separate outer section. Claim 1 specifically recites that the locking ring retains the annular outer section in the first cavity [of the acetabular cup] in operative, but removable, engagement with the inner section.... The claim 1 language “in operative engagement” necessarily requires that the annular outer section be removable. During the interview with the examiner, Ramos explained that in order to achieve the benefits of the device, namely the easy replacement of component parts,4 it was necessary to interpret the language of claim one “as requiring a separate, removable outer section, as shown in Fig. 1 of the 090 patent.” Hence, the scope of this claim does not bring within it any devices in which the outer section of the insert may not be removed from the prosthesis while leaving the inner section intact within the acetabular cup. We also interpret Ramos’ use of the term “spherical configuration” as it appears in Claim one of the 090 patent. We reject the defendant’s argument that such a term is properly limited to inserts with only one radius. Instead, we adopt the meaning of the word as understood by plaintiffs expert — 1075 — Dr. Swanger. We interpret the word “spherical” as used in claim one to include those inserts which approximate a sphere and permit rotation in no particular preferred direction. III. The DePuy Device Does Not Infringe Literally on the 090 Patent Once a court has determined the scope of the patent claims, it must then compare the patent claims to the accused device and determine the issue of infringement, which is a factual determination. Morton Intern. Inc. v. Cardinal Chemical Co., 5 F.3d 1464 (Fed.Cir.1993). The finding of infringement is made by determining whether the claims read on the accused device. Loctite Corp. v. Ultraseal Ltd., 781 F.2d 861 (Fed.Cir.1985). In order for this court to find that the DePuy device infringes on the 090 patent, we must determine that the De-Puy hip embodies every limitation in the claims contained in the 090 patent, either literally or by a substantial equivalent. Conroy v. Reebok Intern., Ltd., 14 F.3d 1570 (Fed.Cir.1994). A court must consider all limitations in the claim, including those defining relationships among the recited elements. Pac-Tec, Inc. v. Amerace Corp., 903 F.2d 796, 801 (Fed.Cir.1990).5 Since we find that there are limitations within the 090 patent which are absent in the DePuy hip, we find that the DePuy hip does not embody the claims of the 090 patent, and hence there is no resulting infringement of Ramos’ patent rights. We reach our conclusion by reading the claims of the 090 patent on the DePuy device. It is true that the DePuy device is [a]n artificial hip joint comprising an artificial hip socket having a first cavity and an opening in a surface of the socket communicating with the first cavity, [and] an annular groove formed in the first cavity adjacent said opening. However, we find that the DePuy device does not have a “sectionalized bearing insert ... said bearing insert having inner and outer sections, each section being formed with a complementary component of said second cavity.” We find instead that the accused device has an insert consisting of a one piece bearing and a locking collar. We reject the testimony of Dr. Gordon Hill that the collar in the DePuy device functions as a bearing. While we understand that during certain extremes of motion the collar comes in contact with the ball of the femur, we also find that mere contact between components does not a bearing make. We are persuaded by the testimony of Richard Smith, based in part on the prosthesis retrieval project, that the outer collar does not play a bearing role in the DePuy device. Furthermore, if the collar did serve as a bearing, the DePuy design team, concerned with insert wear, would have restricted its choice of materials to those substances, such as polyethylene, which are appropriate for load bearing purposes. Instead, the DePuy engineers considered a broad range of materials for the collar, including substances such as delrin, eel-con and polysulfone, none of which is appropriate for use in load bearing prosthetic components. If the collar were truly performing a load bearing function, DePuy’s engineers would have limited their choices to materials capable of effective load bearing. Although we agree with Ramos that the limitation in Claim one relating to the “second cavity of spherical configuration greater in scope than hemispherical” is broad enough to read on the DePuy device, this does not save Ramos’ claim from failure. In order for a finding of infringement to be upheld, all of the limitations in the patent claims must be present in the accused device. Conroy, 14 F.3d 1570. The presence of a single limitation in the accused device does not warrant a finding of infringement. The 090 patent also does not read on the DePuy device with respect to the require — 1076 — ment of distinet bearing and locking ring components. The claims call for the two piece bearing insert “and an open annular spring locking ring having opposite ends formed with tool engaging openings, said locking ring being removably engaged in said annular groove in abutment with said outer surface portion of the bearing insert outer section, said locking ring, when is said groove engagement being visibly exposed and removably retaining said annular outer section in said first cavity in operative engagement with the inner section whereby said ball is retained in said second cavity____” This language does not read on the accused device, which does not contain a locking ring in abutment with the outer bearing. As indicated above, we find there is no “outer bearing” in the DePuy device for the locking ring to abut. While it is true the DePuy hip consists of a locking collar in abutment with the device’s one piece bearing, the claim still does not read on the device because of the requirement that the bearing consist of an inner and outer portion. Alternatively, as we previously noted, the collar on the DePuy device is not a bearing, and hence the 090 claim would still not read on the DePuy device. The claim requires two distinct components, namely the locking ring and the outer bearing. Ramos specifically disallowed the merger of the two components into a singular structure when he argued before the Patent Office that his design was patentable over prior art. Additionally there is no infringement by the DePuy device on the Ramos patent due to the removably retaining limitation. The DePuy device may not be disassembled such that the outer piece of the plastic insert, bearing or not, may be removed from the shell while leaving the inner plastic bearing within the shell. There is no locking ring in the DePuy device which “removably retain[s] said annular outer section in said first cavity.” There is no element of the DePuy hip which satisfies the removably retaining limitation of the Ramos patent. The collar and bearing may not be separated in such a way as to allow the collar to be removed without the head of the femur and the bearing coming out of the shell with it. The ability to remove the outer section of the insert from the inner section of the insert is an essential element of the Ramos patent, which is supposed to facilitate conversion from partial to total joint replacements. Ramos argued as much when he challenged the Patent Office’s determination that his patent was invalid subsequent to DePuy’s request for re-examination. Ramos contended that “[o]nce the annular spring locking ring is removed from the groove engagement, the femoral ball component along with the locking ring and annular outer section of the bearing insert simply slide out from the hip socket and inner section of the bearing insert____ For at least the foregoing reasons, the claim 1 language must be interpreted as requiring a separate, removable outer section, as shown in Fig. 1 of the ’090 patent, to achieve the advantages attributed to the presently claimed invention as explained above and as recognized by the PTO on page 2 of the Office Action.” In light of the properly construed claim language, it cannot be said that the claim one reads on the DePuy device. For the same reasons it does not infringe on claim one, the DePuy hip also does not infringe on claims three and five of the 090 patent. IV. The DePuy Hip Does Not Infringe on the 090 Patent Under the Doctrine of Equivalents Even though we find there is no literal infringement of the 090 patent we nonetheless turn our attention to the doctrine of equivalents, which provides protection to a patentee against devices which perform substantially the same function as the patented device, in substantially the same way to obtain the same result. Graver Tank & Mfg. Co., Inc. v. Linde Air Products Co., 339 U.S. 605, 608, 70 S.Ct. 854, 856, 94 L.Ed. 1097 (1950). Certainly infringement of a patent claim is not necessarily avoided by mere changes or improvements in a invention, Temco Elec. Motor Co. v. Apco Mfg. Co., 275 U.S. 319, 48 S.Ct. 170, 72 L.Ed. 298 (1928); however, “[i]t is only when the changes are so insubstantial as to result in a ‘fraud of the — 1077 — patent’ that application of the equitable doctrine of equivalents becomes desirable.” Slimfold Mfg. Co., Inc. v. Kinkead Industries, Inc., 932 F.2d 1453, 1457 (Fed.Cir.1991). The Federal Circuit has repeatedly cautioned that the doctrine of equivalents is to be applied with care, see, e.g., London v. Carson Pirie Scott & Co., 946 F.2d 1534, 1538 (Fed.Cir.1991). It has taught that the doctrine is not necessarily the second prong of every infringement action. Id. The doctrine should provide relief only where the court determines that the application of the doctrine is proper, i.e., where the accused device performs the same function as disclosed in the patent in substantially the same way to obtain the same result. Carman Industries, Inc. v. Wahl, 724 F.2d 932, 942 (Fed.Cir.1983). A finding of equivalence is a factual determination which must take into account various factors. Coming Glass Works v. Sumitomo Elec. U.S.A., Inc., 868 F.2d 1251, 1261 (Fed.Cir.1989). “ ‘To be a[n] ... equivalent, the element substituted in the accused device for the element set forth in the claim must not be such as would substantially change the way in which the function of the claimed invention is performed.’ ” Dolly, Inc. v. Spalding & Evenflo Companies, Inc., 16 F.3d 394, 397 (Fed.Cir.1994) (quoting Perkin-Elmer Corp. v. Westinghouse Elec. Corp., 822 F.2d 1528, 1533 (Fed.Cir.1987). When an accused device achieves the same result by performing substantially the same function but does so in a substantially different manner, there is no infringement under the doctrine of equivalents. Id. at 400. Moreover relationships among components may not be ignored in determining the existence of equivalents. Id. Even where the tripartite test of equivalency as set out in Carman Industries is satisfied, a court may not find infringement where the “asserted scope of equivalency of what is literally claimed would encompass the prior art.” Wilson Sporting Goods Co. v. David Geoffrey & Associates, 904 F.2d 677, 683 (Fed.Cir.1990), cert. denied, 498 U.S. 992, 111 S.Ct. 537, 112 L.Ed.2d 547 (1990). A patentee is not entitled to greater rights under the doctrine of equivalents than he would be entitled to under his patent. Id. at 684. “The doctrine of equivalents exists to prevent a fraud on a patent, Graver Tank & Mfg. Co. v. Linde Air Prods. Co., 339 U.S. 605, 608, 70 S.Ct. 854, 856, 94 L.Ed. 1097 (1950), not to give a patentee something which he could not lawfully have obtained from the PTO had he tried.” Id. The specific issue which a court must address in determining the existence of equivalents is whether the patentee has proved that a hypothetical claim, similar to the claim issued in his behalf, but broad enough to literally cover the alleged infringer’s device, would have been patentable. Id. In light of this type of analysis, this court’s decision that the DePuy hip does not infringe the 090 patent pursuant to the doctrine of equivalents, is influenced in part by the Patent Office’s rejection of the claims contained in Ramos’ initial patent application. The claims in the first application were broad enough to read on the DePuy device. See, Testimony of Jere Seal’s, P. 153 March 18, 1994. Among the claims made in that application was the following: [a]n artificial hip joint comprising an artificial hip socket having a first cavity in a groove on the inner surface thereof, a bearing insert registerable within said first cavity, itself having a second cavity, a locking ring registerable with said groove, an artificial femoral component having a ball registerable in said second cavity. The patent office rejected this claim as unpatentable due to its anticipation by the Noiles patent as well as its obviousness in light of the Averill and D’Errico patents. Hence, we have before us an example of a claim drafted broadly enough to read on the DePuy device which was not patentable. While this supports our finding that there is no infringement under the doctrine of equivalents, we do not treat it as dispositive of the issue. Just because the claim in Ramos’ first application would not have been patentable does not mean that it is inconceivable that a hypothetical claim could be drafted which would literally read on the DePuy device and yet still be patentable over the — 1078 — prior art. Thus, the proper disposition of the issue requires that we examine whether the DePuy hip performs the same function as disclosed in the 090 patent in substantially the same way to obtain the same result. We find that the DePuy hip is not the infringing equivalent of the hip described by the claims of the 090 patent. While the DePuy hip performs the same function as Ramos’ invention, i.e., replaces the natural hip joint, our attention is focussed on the way in which the function is performed. One of the novel aspects of Ramos’ invention was its ability to resist dislocation. Ramos achieved this goal by utilizing the locking ring mechanism to secure the interior . components in the metal shell. DePuy, on the other hand, .addressed the dislocation problem common in earlier prostheses by utilizing the self-centering mechanism. This is a substantially different way t'o achieve the same result. We find that the hip disclosed in the 090 patent could not incorporate the concept of self-centering without seriously compromising the range of motion the prosthesis could achieve. We also find that Claim one of the 090 patent which calls for the removable retention of the annular outer bearing inside the cavity of the acetabular shell by means of the locking ring does not have an equivalent in the DePuy prosthesis. One of the principle goals of the Ramos design was to facilitate the replacement of component parts while leaving others in place. This cannot be achieved with the DePuy device. Disassembly of the DePuy hip requires the removal of the entire insert assembly. The same result is not achieved by the DePuy hip as would be by the hip disclosed in the Ramos patent. Just as disassembly of the DePuy and Ramos hips differ, so does assembly. The DePuy hip may be assembled by one person in a matter of several seconds. The Ramos design, on the other hand, requires a two person assembly team. Thus, we find that even if the end result of the two prostheses is the same, which it clearly is not, the prostheses achieve the result in substantially different ways. In order to hold that the infringing device is the fundamental equivalent of the device disclosed in the 090 patent, we would have to ignore the relationships that the 090 patent requires. We may not consider as an equivalent a prosthesis which fails to incorporate the relationships among the elements of the invention that are disclosed in the patent. Dolly, Inc. v. Spalding & Evenflo Companies, Inc., 16 F.3d 394, 400 (Fed.Cir.1994). Hence, contrary to the plaintiffs contention, the “outer bearing” and the locking ring are not “in exactly the same positional relationship to each other that the claim calls for” since there is no outer bearing in the DePuy hip, and, a single structure cannot possibly abut itself. Clearly there is no infringement of the 090 patent by the DePuy hip under the doctrine of equivalents, since a finding of infringement requires that every limitation of the patent claims be contained in the accused device, either literally or by a substantial equivalent. For the reasons discussed herein, every limitation in the Ramos patent does not enjoy an equivalent in the DePuy hip, and we thus hold that the accused device does not infringe the 090 patent. Since we find that the DePuy device does not infringe on the 090 patent under the doctrine of equivalents, we need not address the issue of prosecution history estoppel. Prosecution history estoppel is a limiting doctrine to be evaluated when equivalence is found so that the patentee may not be permitted to gain through litigation that which he has relinquished during prosecution of the patent. See, e.g., Haynes Intern., Inc. v. Jessop Steel Co., 8 F.3d 1573, 1577 (Fed.Cir. 1993) (citing Loctite Corp. v. Ultraseal Ltd., 781 F.2d 861, 870 (Fed.Cir.1985); Hughes Aircraft Co. v. U.S., 717 F.2d 1351, 1363 (Fed.Cir.1983). Since we find no equivalence between the DePuy hip and the claims of the 090 patent, there is no need to explore potential limitations. Order The DePuy universal self centering hip does not infringe literally on the claims in U.S. Patent No. 4,380,090, nor does it infringe under the doctrine of equivalents. — 1079 — The Clerk is directed to enter judgment in favor of the defendant, Boehringer Manheim Corp. and against the plaintiff, Pedro A. Ramos, dismissing the complaint. Footnotes [1] The parties have previously agreed that DePuy is the real party in interest, and DePuy has accordingly assumed the defense of Boehringer Manheim. [2] A continuation in part is a term of art in patent law which refers to an application which carries forward some or all of the disclosure of the original application and in addition adds new disclosure. The term may not properly be used interchangeably with the term continuation, although it frequently is. A continuation is a second application which carries forward disclosure from a previous application but which adds nothing new. It merely provides a second opportunity to have the same disclosure reviewed for patentability. [3] During the course of an opening statement, Mr. Knebel stated the following: I do want to point out to the court that we will not in the course of this case be putting on an invalidity case that is similar to what was put on in the Biomet case. The invalidity decision in the Biomet case was based on published prior art that was before the patent office on re-examination. The invalidity case in this case is based on the prior inventorship by Drs. Pappas and Buechel and art that has never been considered. [4] The ease of disassembly which facilitated the replacement of component parts of the prosthesis and enabled the device to be converted from a partial to a total joint replacement was at one time understood by Ramos to be the essence of his invention. [5] Pac-Tec was cited by Ramos for this proposition in his February 16, 1993 communication with the Patent Office when he argued in support of the patentability of his claims over prior art. Although the Pac-Tec case discusses the proper interpretation of claims for the purpose of determining validity, claims are to be interpreted in the same manner in determining infringement as they are in determining validity. Semned, Inc. v. Richard-Allan Medical Industries, Inc., 888 F.2d 815, 818 (Fed.Cir.1989).