Permissible Repair versus Impermissible Reconstruction

by Dennis Crouch

I was looking forward to the oral arguments in the interesting case of Karl Storz Endoscopy v. STERIS Instrument Management  Services (IMS) focusing the threshold between permissible repair and impermissible reconstruction.  Oral arguments were scheduled for December 4, 2023, but the parties filed a joint dismissal on December 3.

Storz is a medical device manufacturer and sells its patented endoscopes throughout the united States that included a specialized optical system covered by two patents. US7530945 and RE47044.  The patents claim an endoscope with an optical relay components surrounded by transparent shrink wrap, which allows the components to be pre-assembled and optically tested before insertion in the endoscope body.  Hospitals hire IMS to repair its medical devices, including the Storz endoscopes — a process that could involve breaking open the endoscope, removing the optical relay, replacing damaged lenses/spacers with new or recycled ones, inserting the rebuilt optical relay back into the endoscope body, and resealing the device.

A manufacturer’s patents covering its product are “exhausted” as soon as the company places product in the chain of commerce.  Those particular authorized products can be resold, used, and repaired by third parties without creating any patent infringement liability.  But, the buyers do not have permission to create new copies of the patented work, and that includes wholly reconstructing the product after it has been “spent.” The seminal Supreme Court case addressing  the distinction between permissible repair and impermissible reconstruction is Aro Manufacturing Co. v. Convertible Top Replacement Co., 365 U.S. 336 (1961), which held that “replacement of unpatented parts, one at a time, whether of the same part repeatedly or different parts successively, is no more than the lawful right of the owner to repair his property.” In Wilbur-Ellis Co. v. Kuther, 377 U.S. 422 (1964), the Court found extensive refurbishing and resizing of multiple unpatented components was repair because it extended the useful life of the original patented article. The Federal Circuit expanded on the scope of repair in Dana Corp. v. American Precision Co., 827 F.2d 755 (Fed. Cir. 1987), permitting rebuilding of worn unpatented parts, and in Jazz Photo Corp. v. ITC, 264 F.3d 1094 (Fed. Cir. 2001), allowing invasive modifications so long as there was no complete reconstruction. However, in Sandvik Aktiebolag v. E.J. Co., 121 F.3d 669 (Fed. Cir. 1997), the court found impermissible reconstruction where a patented drill bit was recreated with a new cutting tip after the drill was spent.

Before the district court, IMS moved for summary judgment that its activities constitute permissible repair and not infringing reconstruction. The district court agreed, holding that IMS was exercising its customers’ right to repair the devices they had purchased and therefore did not infringe.  The district court particularly found that IMS’s repairs preserve the useful lifespan of the endoscope and do not “reconstruct” or make a wholly new article.  On appeal, Karl Storz raised three main grounds for reversing the district court’s decision:

  1. Readily Replaceable Parts: Karl Storz argued that replacing “readily replaceable” parts can be repair, but replacing parts that are not readily replaceable constitutes impermissible reconstruction. From the briefing it appears undisputed that the optical relay in its endoscopes and covered by its patents is not a readily replaceable part.
  2. Novel/Inventive Aspect: The district court found the optical relay assembly was the “novel and distinguishing” part of Karl Storz’s patents. Karl Storz contended that by fully replacing this inventive aspect, IMS has impermissibly reconstructed the device, even though the patent does not have a claim directed solely to the optical relay assembly.
  3. New Article: Case law says repair does not allow a party to construct an essentially new article. Karl Storz argues that IMS’s replacement of nearly all the patented components results in an endoscope that is so different from the original that IMS has created a new device.

The would have been an important one in the evolving area of patented medical device repair and third-party servicing rights.

32 thoughts on “Permissible Repair versus Impermissible Reconstruction

  1. 6

    The repair vs reconstruction arguments here of “readily replaceable” strikes me as novel and ambiguous. There are other repair vs reconstruction cases where worn out or dead parts were not easily removed. As to “fully replacing this inventive aspect” or having “impermissibly reconstructed” the device, “to construct an essentially new article” should that not be directly tied to actual claims, especially
    claim elements argued as novel in the spec or in application prosecution?
    [At least this issue got addressed here pre-trial, and lead to settlement, before further patent litigation costs.]

    1. 6.1

      At least….?

      Spoken like a litigator who has forgotten the value of decisions resolving legal matters and only (myopically) looking at costs.

      Litigators seem to be becoming ever more dull.

      1. 6.1.1

        It is clients that pay the costs of patent litigation, which is mostly the costs of the litigators. To extend litigation just to resolve legal matters is not normally acting in the interests of the client.


          Nice – but has nothing to do with the myopic nature here (and lack of clarity of an issue may well ALSO be more costly to not just the clients, but society writ large).

    1. 5.1

      Have to agree – I don’t think Storz would care if the Hospital lab tech repaired the scope. The fact that a third party, i.e., IMS, is making money off repairs appears to be the driving factor.

      Storz just need to “warranty” its product and let the hospitals ship the broken scopes back to Storz who repairs them (for a warranty fee).

    1. 4.1

      Definitely a bit of that going on, with the blurry boundaries between “repair” and “rebuild”. I note that one of the many proposed “resolutions” to the paradox (on Wikipedia) is “group consensus” which is what we’re likely to see in this case (the group being a group of judges).

      I’m curious as to whether this issue has ever arisen in the context of an invention whose advantage over the prior art was stated to be that it was “more easily repairable.” Also it seems to me that when buying an item the manufacturer has exhausted its rights to not only the item but also all of the component parts. In other words, if I buy two copies of a patented widget, I should be able to take them both apart, mix the pieces up and then rebuild one or both of them from scratch without worrying about a lawsuit, regardless of how many parts each newly built widget were shared with its “parent”.

      1. 4.1.1

        We have home “renovations” that preserve a single wall of the previous structure. For all intents, it’s new construction, but not for new home permitting, environmental, etc.

        If I had to split this baby, that’s what I would probably do; split it.

        If a “repair” involves 50.1% new material/labor content of the original, the item is probably being remanufactured. If it involves 49.9%, its probably being repaired.


          Are those weight percentages? 😉

          In the minimal example, the invention Z is a combination of two prior art components, called X and Y. Let’s say I buy Z. If X breaks off, I can surely re-attach it without fear of infringement liability. What if I have a drawer full of spare X’s? Can I repair Z using one of those without infringing?

          Now let’s say Z is a combo of three prior art products: a large main body B with two fragile small components X and Y attached independently to B. While hiking, X and Y bust off …


            I think the percentage of allowable rebuild would have to be measured not in units of mass or parts count, but rather the economic value required to complete the process, reasonably scoped to the patented article.

            But then again, we know that proportionality is not a concept that rides well with the patent maximalists.


              marty’s glomming to the words “patent maximalists” says far more about him than he most likely realizes.

              There can NEVER be too many (valid) patents.


                “ There can NEVER be too many (valid) patents”


                Comments from patent crack addicts should be rare!

                1. I would agree that they SHOULD be rare – but not for the reasoning that you likely had “in mind.”

                  Most people who have studied innovation would agree with Lincoln, Clemens, as well as the original architects of this country who postulated that no reasonable person would doubt the value of a system for promoting innovation (and with the further understanding that the Quid Pro Quo is ALWAYS a good deal).

                  But you are not anti-patent, as you like to attempt to gaslight, eh Malcolm?

  2. 3

    In order to protect against such “repairs / replacements,” instead of outright sales of their scopes, Storz should license their use; with a “we must repair” provision (ala’ the widespread use of software licensing).

    This would also virtually guarantee all repairs and replacements are done according to the product’s original specs.

    Which also makes sense given the criticality of such medical devices.

    1. 3.1

      They want to sue the third party company doing the repairs, not each individual customer. Can’t do that with a licensing agreement.

  3. 2

    Can someone explain why it should matter whether a replaceable component of a repairable device is “readily” replaceable versus merely “replaceable”?

    When the replaced component itself is not patented (or is separately purchased), then repair of a multi-component device with that replaceable component should generally be permitted.

    Storz’s real problem here seems to be a failure to successfully compete in the market for repair of its own devices. I’d be curious to know how that happened, especially if the companies doing the repairs have their own patents which cover an aspect of the repair process and which keep Storz out of the box of repairing its own products.

    1. 2.1

      Storz runs arguments 1), 2) and 3). You mention 1). I am more interested in 2) because that is an argument that is used also in “repair” cases where I am, in Europe.

      Dennis writes that it is the instrument that is what is patented whereas Storz asserts that its patented invention lies in the optical relay asembly of the instrument. What if the repairer removes a broken optical relay and replaces it with one of inferior quality. In doing that, has the instrument been repaired or has it become a different instrument?

      Perhaps the repairer’s optical relay assemblies are inferior to Storz relays, but that doesn’t trouble the buyers. That might allow the repairer to undercut Storz, but would also harm the reputation of Storz as a maker of reliable instruments? Shouldn’t patent law help the patent owner when an unauthorized party, under the guise of repair, undercuts the patent owner’s exclusive rights?

      1. 2.1.1

        How is a person’s choice to go with a known non-original place of repair “undercutt[ing] the patent owner’s exclusive rights?”


          The exclusive right includes the right to “make” the patented article. What I am suggesting is that the act of the self-declared repairer (as I have set it out) is going beyond what the Defendant asserts to be nothing more than a mere repair, and that the Defendant’s act is more correctly categorized as an infringing act of making the patented article.

          Here a Link to the landmark UK Supreme Court Decision in Schutz vs Werit where the goods were the IBC’s to be seen in the photo at the start of the decision text. The outcome was close, the Sup Ct reversing the C of A.

          An IBC has a steel cage surrounding and protecting a large square plastic “bottle”. Does replacing that bottle make a new IBC? At least in England, it might depend on what invention the patent is protecting:

          link to


            The item being repaired has already been sold (exhausted).

            This is not about making (from scratch).


              Obviously. What everybody is discussing here is whether there can be a making of a new article under the guise of repairing an old one.

              The dilemma is age-old, going back at least as far as your Theseus example.

              Is it your stance that no amount of “repair” will ever under any circumstances be enough to constitute the making of a new article? If so, it seems to fly in the face of the established case law in the USA and Europe.



                Respectfully, if one has an exhausted item, how is “making” (from scratch) even on the table?

                The funny business is being imported due to business models (and business models ALONE).

                As to your quip of, “Is it your stance that no amount of “repair” will ever under any circumstances be enough to constitute the making of a new article?” – the critical context is to understand exactly WHY you are putting the word ‘repair’ in quotes.

                As you deign to put that word in quotes, I need to understand exactly what you intend to mean with that word.

                On its normal and customary (read: legal) usage, the stance of repair – no matter the amount – MUST be geared to an item that has been sold (and thus protections legally exhausted).

                This really is not a difficult concept to grasp, now is it?

                1. Dennis cites the 1977 Sandvik case, in which spent drill bits were repaired with a new cutting tip. The Fed Ct found that this was not a repair but, rather, an impermissible reconstruction. Sandvik won, despite having exhausted their rights. Dennis doesn’t mention whether Sandvik’s business model was decisive to the outcome. I assume it wasn’t.

                  This is why I put “repair” in quotation marks. Because the entire debate here is (as far as I can see) about where to draw the line between what is a permissible “repair” and what is an impermissible “reconstruction”.

                  Yes, of course a seller can look to its business model to ameliorate any exhaustion of rights. But that’s another issue entirely, isn’t it?

                2. Do not be so wooden and try for some original thought, eh?

                  And no – obviously – the very point at hand is that the “issues” of business models has NOT been separated.

                  To feed your cross-Sovereign tendencies, see also: link to


                By the by, the business concept of “planned obsolescence” is ALSO merely a business model.

                To be cogently critical here, we do need to separate out mere business models from any underlying right.

                1. … and to emphasize this point, MaxDrei’s ham-fisted “but that ‘s a different issue” is woefully inadequate to actually separate that issue.

                  This has a parrallel in the copyright realm of digital goods.

                  In the real world, the process of “buying” a digital good does not include any shipment of the actual physical good. No “electrons,” or “bytes” owned and possessed by the seller are transferred as real property to the buyer.

  4. 1

    The Federal Circuit decision is likely to provide helpful guidance on interpreting the repair doctrine’s boundaries when patented articles involve complex assemblies with multiple integrated components.
    Did you intend to rewrite this? You wrote earlier that both parties filed a joint dismissal, but this comment implies that you anticipate the Federal Circuit issuing an opinion.

    BTW. This is an interesting topic. Do you have any suggested online resources for more information?

    1. 1.1

      Thanks! I had indeed written this post over the weekend and then planned to update after listening to oral arguments. I was disappointed in the last minute settlement.

Comments are closed.