OMI Holdings Incorporated (OMI) seeks review of the decision of the United States District Court for the District of Kansas, Manildra Milling Corp. v. Ogilvie Mills, Inc., 878 F.Supp. 1417 (D.Kan.1995), which awarded Manildra Milling Corp. (Manildra) costs in the amount of $82,754.93. We affirm.
I
Because a detailed recitation of facts is available in the district court’s opinion, Manildra, 878 F.Supp. at 1419-20, we present only a summary of those facts pertinent to this appeal. Manildra sued OMI seeking a declaratory judgment that OMI’s patents were invalid and damages for OMI’s violations of the Lanham Act, Kansas common law, and federal antitrust laws. OMI responded by filing a $17 million counterclaim for patent infringement.
After eighteen days of trial, the district court judge was forced to order a mistrial because of improper ex parte communication between three jurors and Manildra’s expert witness, John Howell. Upon the completion of the second trial, the jury determined that OMI’s patents were invalid, and awarded a total of $5 million in damages to Manildra for OMI’s violations of the Lanham Act and Kansas common law. On appeal, this court affirmed the judgment of invalidity but reversed the judgment on the Lanham Act and Kansas common law. Manildra Milling Corp. v. Ogilvie Mills, Inc., 1 F.3d 1253, 30 USPQ2d 1012 (Fed.Cir.1993), cert. denied, — U.S. -, 114 S.Ct. 1188, 127 L.Ed.2d 538 (1994).
On November 3, 1993, Manildra submitted a bill of costs seeking $190,393.74. OMI moved to deny costs, contending that: (1) Manildra was not a prevailing party under Fed.R.Civ.P. 54(d)(1); (2) the court should exercise its discretion to deny costs; and (3) costs of certain items in the bill of costs should be disallowed. Manildra, 878 F.Supp. at 1424.
The district court determined that Manil-dra’s success was sufficient to make it a prevailing party. Accordingly, the district court awarded partial costs of $82,754.93 to Manildra. Id. at 1424-25. OMI now appeals to this court for review of the district court’s award. We have jurisdiction under 28 U.S.C. § 1295(a)(1) (1994).
II
The first issue on appeal is whether the district court properly decided that Manildra is a prevailing party in this case. This case comes to us from a district court in the Tenth Circuit. In that circuit, the court seems to approach the issue of prevailing party in two different ways.
Some cases suggest that the entire decision of awarding costs, including the determination of whether a party prevailed, is left to the discretion of the district court. See, e.g., Roberts v. Madigan, 921 F.2d 1047, 1058 (10th Cir.1990), cert. denied, 505 U.S. 1218, 112 S.Ct. 3025, 120 L.Ed.2d 896 (1992). Under this view, an appellate court may reverse the district court’s prevailing party determination only for abuse of discretion. Id.
Other cases, however, state that although the ultimate decision to award fees rests within the district court’s discretion, any statutory interpretation or other legal conclusions that provide a basis for the award are reviewable de novo. See Supre v. Ricketts, 792 F.2d 958, 961 (10th Cir.1986).1 Under this view, rule interpretation, such as the meaning of “prevailing party,” like statutory interpretation, is a question of law that is reviewable de novo, while the ultimate deci*1181sion to award costs remains within the district court’s discretion. See id.
In the Tenth Circuit, then, there is some ambiguity as to whether the meaning of “prevailing party” is a question of law. A large part of the ambiguity arises from the fact that many opinions lump together analysis of the prevailing party issue with the ultimate decision to award costs. This makes it difficult to determine the standard of review.
In the present case, this ambiguity is exacerbated by a choice of law issue. We must not only decide which of these two approaches to use, but also whether to apply Tenth Circuit law or Federal Circuit law.
In deciding which law to apply, we must necessarily evaluate competing policy interests. Biodex Corp. v. Loredan Biomedical, Inc., 946 F.2d 850, 856-58, 20 USPQ2d 1252, 1257-59 (Fed.Cir.1991), cert. denied, 504 U.S. 980, 112 S.Ct. 2957, 119 L.Ed.2d 579 (1992). On the one hand, we must remain mindful of “the general policy of minimizing confusion and conflicts in the federal judicial system.” Panduit Corp. v. All States Plastic Mfg. Co., 744 F.2d 1564, 1574, 228 USPQ 465, 471 (Fed.Cir.1984). On the other hand, we cannot forget that one of Congress’ primary purposes in creating the Court of Appeals for the Federal Circuit was to bring about uniformity in the area of patent law. Id. at 1574, 223 USPQ at 470-71.
In resolving these competing policy interests, we have considered several factors including: the uniformity in regional circuit law, the need to promote uniformity in the outcome of patent litigation, and the nature of the legal issue involved. See Biodex, 946 F.2d at 855-59, 20 USPQ2d at 1256-59. As explained below, we believe these factors favor our applying our own law to define the meaning of prevailing party in the context of patent litigation.
We have generally conformed our law to that of the regional circuits when there exists expressed uniformity among the circuits. Id. at 856, 20 USPQ2d at 1256-57. That is not the ease here. The Fifth and Seventh Circuits appear to treat the issue of prevailing party as lying within the district court’s discretion. See Studiengesellschaft Kohle v. Eastman Kodak Co., 713 F.2d 128, 131, 219 USPQ 958, 960 (5th Cir.1983); Northbrook Excess and Surplus Ins. Co. v. Commercial Union Ins. Co., 924 F.2d 633, 642 (7th Cir.1991). In contrast, the First and Ninth Circuits appear to treat the meaning of “prevailing party” as a question of law to be reviewed de novo. See Gamma Audio & Video, Inc. v. Ean-Chea, 11 F.3d 1106, 1113-14, 29 USPQ2d 1257, 1263-64 (1st Cir.1993); Lummi Indian Tribe v. Oltman, 720 F.2d 1124, 1125 (9th Cir.1983).
Therefore, there is no uniformity of the law outside our circuit that argues in favor of discerning and applying Tenth Circuit law to the prevailing party issue. Indeed, the circumstances of this case might cause us to decline to employ Tenth Circuit law, even were the other circuits uniform in following either the wholly discretionary or partially discretionary standard of review. Our declination would avoid the need for us to clarify the apparent ambiguity in Tenth Circuit law on this point.
We have also recognized that our adoption of a single position promotes national uniformity in the outcome of patent litigation. See Biodex, 946 F.2d at 858-59, 20 USPQ2d at 1259. Such uniformity is undercut when the outcome of the patent trial varies depending on the regional circuit in which the case is tried. Id. at 859, 20 USPQ2d at 1259. The prevailing party determination in the present case is a prime example. If we leave to each regional circuit the task of defining when a party in a patent case has prevailed, a party such as Manildra may be eligible for costs in one circuit but not another. If, however, we instead define as a matter of law the meaning of prevailing party in the context of patent litigation, eligibility for costs will be uniform nationwide. Such a single definition will promote uniformity in the application of the term in future trials throughout the regional circuits. See id. Therefore, this factor also favors application of our own law.
Finally, we acknowledge that the degree of our deference to regional circuit law has involved the nature of the legal issue at hand. For this reason, we have generally deferred to regional circuit law when the issue involves an interpretation of the Federal Rules *1182of Civil Procedure, as it does here with the interpretation of “prevailing party” under Rule 54(d). Id. at 857-58, 20 USPQ2d at 1258. Our practice in this regard results from the notion that such rules are integral to the routine conduct of trials, and that practitioners and judges should be free to conduct litigation according to the rules that ordinarily apply to them. We catalogued in Biodex the instances in which we have deferred to regional circuit law when confronted with questions of interpretation of the Federal Rules of Civil Procedure. See id. at 857 n. 10, 20 USPQ2d at 1258 n. 10. In each of those cases, the particular rule of procedure relates to the ongoing trial process.
In the present case, however, the impact on the district court’s management of an ongoing trial will not be great because the prevailing party issue arises only after the trial has been completed. Thus, defining this term as a matter of Federal Circuit law does not interfere with the district court’s ability to manage the majority of the trial according to regional circuit law. Cf. id. at 859, 20 USPQ2d at 1259 (stating that district court would not be required to serve two masters because availability of appellate review is irrelevant to conduct of trial or to any decision on substantive legal issues that may arise during trial).
In sum, we conclude that deference to regional circuit law is inappropriate in this case. By establishing a single definition of prevailing party in the context of patent litigation, we promote uniformity in the outcome of patent trials. Moreover, given that this issue arises only after the completion of patent trials, we create a rule of law no more intrusive on trial practice than any other rule of patent law that must be applied.
Ill
We now turn to the task of defining the meaning of prevailing party in the context of patent litigation. Rule 54(d)(1) provides, in pertinent part: “Except when express provision therefor is made either in a statute of the United States or in these rules, costs other than attorneys’ fees shall be allowed as of course to the prevailing party unless the court otherwise directs.” Fed.R.Civ.P. 54(d)(1). In those cases in which one party wins completely on every claim at issue, determining which party has prevailed is a straightforward task. The inquiry becomes more difficult when each party has some claims adjudicated in its favor.
The Supreme Court recently addressed this issue in Farrar v. Hobby, 506 U.S. 103, 113 S.Ct. 566, 121 L.Ed.2d 494 (1992). That case involved the issue of whether a civil rights plaintiff who receives nominal damages is a prevailing party eligible to receive attorneys fees under 42 U.S.C. § 1988. Farrar, 506 U.S. at 105-07, 113 S.Ct. at 570. The court below denied fees on the grounds that the plaintiff was not a prevailing party. Id.
The Supreme Court bifurcated the inquiry by separating eligibility for fees, based on meeting the definition of “prevailing party,” from the discretionary decision to actually award fees. As to the first issue, the Court stated that “a plaintiff ‘prevails’ when actual relief on the merits of his claim materially alters the legal relationship between the parties by modifying the defendant’s behavior in a way that directly benefits the plaintiff.” Id. at 111-13, 113 S.Ct. at 573. Because a judgment for damages in any amount modifies thé defendant’s behavior to the plaintiffs benefit, the Court held that a plaintiff who wins nominal damages is a prevailing party. Id.
Whether a party is the prevailing party, however, is only a threshold inquiry. Having found that the plaintiff was eligible to receive fees, the Court nonetheless affirmed the appellate court’s decision to deny fees. In so doing, the Court noted that the degree of plaintiffs overall success goes not towards whether the plaintiff prevailed, but rather towards the reasonableness of the fee award. Id. at 113-15, 113 S.Ct. at 574. The Court noted that the district court may lawfully award minimal fees or no fees after considering the amount and nature of the plaintiffs success. Id. at 115-16, 113 S.Ct. at 575.
In the present case, the district court was required to apply Farrar to a plaintiff who receives no money damages at all but instead “receives a hard fought declaration *1183that its competitor’s patents are invalid,” and survives the competitor’s counterclaim for $17 million in patent infringement damages. Manildra, 878 F.Supp. at 1425. Although Manildra failed on its Sherman Act, Lanham Act, and Kansas common law claims, its victory on the patent issues was complete. The district court decided that such a victory was sufficient to satisfy the Supreme Court’s definition of “prevailing party.” The district court explained that
the fact remains that Manildra successfully challenged several of [OMI’s] patents winning a declaration of invalidity and now may use its production process free from risk of infringement and without the necessity of obtaining a license. As a result, although Manildra’s suit ultimately produced no money judgment, it did provide Manildra significant latitude with respect to one of its competitors. Thus, essentially, the suit stripped [OMI] of a competitive edge vis-a-vis Manildra with the result that [OMI] and Manildra play on a more level field.
Id.
We agree with the district court’s analysis and now hold that as a matter of law, a party who has a competitor’s patent declared invalid meets the definition of “prevailing party.”
Our holding is informed by the unique nature of the relationships engendered by patent law. A patentee receives the right to exclude all others from making, using, or selling the patented invention. 35 U.S.C. 154 (1994). In essence, a patent inhibits those who would compete with its owner by warning them not to practice the invention. A judicial declaration that one is free from another’s right to exclude alters the legal relationship between the parties. The patentee no longer can enforce his patent against the would-be infringer litigant, or any other would-be infringer. See Blonder-Tongue Lab., Inc. v. University of Illinois Found., 402 U.S. 313, 349-50, 91 S.Ct. 1434, 1453, 28 L.Ed.2d 788, 169 USPQ 513, 527-28 (1971). The freedom to practice an invention without fear of suit by the patentee is a valuable commercial benefit. By removing the potential threat of the patentee instituting an infringement action, the competitor necessarily alters the patentee’s subsequent behavior to his benefit.
IV
OMI correctly notes that satisfaction of the prevailing party definition does not end the inquiry. As the Supreme Court noted in Farrar, even if a party satisfies the definition of prevailing party, the district court judge retains broad discretion as to how much to award, if anything. Farrar, 506 U.S. at 115-16, 113 S.Ct. at 575. In the present case, OMI argued to the district court that Manildra should be denied costs altogether because: (1) Manildra injected unmeritorious issues which unnecessarily prolonged the trial; (2) the case was close and difficult; (3) Manildra received only insignificant relief; and (4) OMI litigated in good faith. Manildra, 878 F.Supp. at 1425. After examining each of these reasons, the district court declined to exercise its discretion to deny costs across the board. Id.
On appeal, OMI makes the same arguments and points to no evidence which the district court failed to consider. Based on the record before us and in light of the presumption in Rule 54(d)(1) that costs are to be awarded, we cannot say that the district court abused its discretion by awarding partial costs to Manildra.
In this regard, we note that we are deferring to regional circuit law. The decision of whether to award costs to a prevailing party implicates considerations not unique to the patent law, such as the litigants’ behavior at trial. The process of deciding whether to award costs in a patent trial is no different than that employed in any other type of trial. Therefore, we review this aspect of the district court’s decision in the same manner as we believe the Tenth Circuit would.
V
OMI next argues that even if Manildra is a prevailing party, and even though the district court properly declined to deny all costs to Manildra, the district court abused its discretion by awarding certain costs. Specifically, *1184OMI contends that the district court should not have awarded costs for transcripts of the mistrial caused by Manildra and for unused deposition transcripts.
Section 1920 of Title 28 enumerates the expenses that a federal court may award as costs under its Rule 54(d)(1) discretionary authority. 28 U.S.C. § 1920 (1994); Crawford Fitting Co. v. J.T. Gibbons, Inc., 482 U.S. 437, 441-42, 107 S.Ct. 2494, 2497, 96 L.Ed.2d 385 (1987). Section 1920(2) authorizes the awarding of costs for “[flees of the court reporter for all or any part of the stenographic transcript necessarily obtained for use in the case.” 28 U.S.C. § 1920(2). This language embraces both trial transcripts and deposition transcripts. See Laura B. Bartell, Taxation of Costs and Awards of Expenses in Federal Court, 101 F.R.D. 553, 567 (1984).
OMI first contends that the district court abused its discretion by awarding costs for the seventeen days of trial transcripts during the mistrial — amounting to $3,728. As a general rule, daily trial transcript costs should not be awarded absent court approval prior to the trial. Griffith v. Mt. Carmel Medical Ctr., 157 F.R.D. 499, 506 (D.Kan.1994). In the present case, there was no prior approval. However, a district court may overlook the lack of prior approval if the case is complex and the transcripts proved invaluable to both the counsel and the court. Id. Here, the district court specifically found that
the daily transcripts were necessarily obtained and invaluable to both the court and the parties. It is beyond dispute that the issues litigated were complex and the trial lengthy. Daily transcripts helped to focus the issues, avoid repetitive testimony, and expedite the trial. Not only were the transcripts vital to the parties’ preparation and presentation of the case, but they also proved critical to the court’s management of the litigation.
Manildra, 878 F.Supp. at 1427.
In the face of the district court’s clear exposition of its basis for awarding costs, OMI cites two cases denying fees and costs for portions of the trial which were unnecessary but were caused by the prevailing plaintiff: Baughman v. Wilson Freight Forwarding Co., 583 F.2d 1208, 1216 (3rd Cir.1978) and Advanced Medical, Inc. v. Arden Medical Systems, Inc., Civ. A. No. 87-3059, 1990 WL 39261, at *6 (E.D.Pa. Mar. 30, 1990), rev’d on other grounds, 955 F.2d 188 (3d Cir.1992). These cases are not controlling, however, because unlike the situation in the present case, the courts in those cases did not conclude that the daily transcripts were invaluable. In light of the district court’s explanation and the small amount involved, we cannot conclude that the district court abused its discretion in awarding costs for those transcripts. As with the issue of the court’s discretion to award costs at all to a prevailing party, whether particular costs can be awarded raises an issue going to the manner in which a trial is conducted. That issue also is governed by regional circuit law, and the court’s decision is reviewed for abuse of discretion.
OMI next contends that the district court abused its discretion by awarding costs for eleven deposition transcripts which were never used at trial — amounting to $4,950. Costs for deposition transcripts may be awarded if they were “necessarily obtained for use in the case.” Miller v. City of Mission, Kan., 516 F.Supp. 1333, 1340 (D.Kan.1981). Although use at trial is direct evidence of necessity, an item may still be reasonably necessary for use in the case even if unused at trial. U.S. Industries, Inc. v. Touche Ross & Co., 854 F.2d 1223, 1245-46 (10th Cir.1988). The underlying inquiry is whether the depositions reasonably seemed necessary at the time they were taken. 10 Charles A Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure § 2676, at 341 (2d ed. 1983).
Manildra contended that the eleven deposition transcripts were necessarily obtained for use in the trial because at the time they were taken, the deponents were potential trial witnesses. In support, Manildra submitted the affidavit of its counsel that the eleven depositions “were believed to be necessary for trial at the time they were taken.” Manildra, 878 F.Supp. at 1427. OMI has pointed to no evidence refuting this affidavit. *1185In view of the evidence before us, we cannot conclude that the district court abused its discretion by awarding costs for the eleven deposition transcripts.
VI
Finally, OMI contends that the district court abused its discretion by not sanctioning Manildra for causing the mistrial. Specifically, OMI asserts that the district court used the wrong legal standard by focusing on the harm suffered by Manildra as a result of the mistrial. OMI argues that the district court should have focused instead on the harm caused to OMI and the court.
We disagree with OMI’s characterization of the district court’s analysis. It appears that the district court only addressed the harm suffered by Manildra in an attempt to demonstrate that the relative harm suffered by OMI was not severe enough to warrant sanctions. See Manildra Milling Corp. v. Ogilvie Mills, Inc., 782 F.Supp. 102, 103-04 (D.Kan.1991). In addition, the district court found that the mistrial was not solely caused by Manildra’s expert witness, but also by the jurors’ misconduct. The court also accepted partial blame and recognized that it should have emphasized the importance of avoiding discussion of the case with parties or witnesses. In light of this finding, we cannot conclude that the district court abused its discretion by not imposing sanctions upon Manildra for the mistrial.
VII
Manildra achieved complete victory on the patent issue by having both of OMI’s patents declared invalid, thereby fending-off OMI’s $17 million counterclaim. As a consequence, Manildra satisfied the threshold test of being a prevailing party. As a prevailing party, Manildra was presumptively entitled to costs, and the district court did not err by declining to exercise its discretion to deny costs. The district court also did not err by using its discretion to award costs to Manildra for trial and deposition transcripts. Finally, the district court did not err by declining to exercise its discretion to impose sanctions upon Manildra. As a result, the district court’s decision is affirmed.
AFFIRMED.
No costs.
. Although Supre involves attorney fees under 42 U.S.C. § 1988 (1988) (current version at 42 U.S.C.A. § 1988(b) (West 1994)) instead of costs under Rule 54(d)(1), the meaning of prevailing party is the same in either context. See Farrar v. Hobby, 506 U.S. 103, 118-20, 113 S.Ct. 566, 577-78, 121 L.Ed.2d 494 (1992) (O'Connor, J., concurring); Studiengesellschaft Kohle v. Eastman Kodak Co., 713 F.2d 128, 132, 219 USPQ 958, 960 (5th Cir.1983).