dissenting.
The majority opinion has reached the conclusion that the district court’s granting of a preliminary injunction against the process in which Baxter was engaged was proper and therefore has affirmed the granting of the preliminary injunction. I am unable to agree and respectfully dissent.
The principal issue on this appeal, as I see it, is whether the reconstitution of fully FDA-approved antibiotics in a manner consistent with the labeling directions, using a diluent meeting the specifications in those directions, and placing the reconstituted antibiotic into an FDA-approved Viaflex plastic container, constitutes the manufacture of a “new” drug according to 21 U.S.C. § 321(p) which must be separately approved pursuant to 21 U.S.C. § 355 or 357. The issue concerns only eight reconstituted antibiotics.
What is not the issue is the repackaging procedure which is controlled by Current Good Manufacturing Practice, 21 C.F.R. Parts 210 and 211. Baxter recognizes this *1413right of control on the part of FDA as to which it does not object. Its contention simply was that there was no policy or legal justification for the imposition of the time-consuming and expensive new drug application requirement.
The facts of this case on which the court has ruled are to be found in the district court’s memorandum opinion and uncontro-verted affidavits filed by Baxter in that court. Practically, every hospital in this country administers antibiotics intravenously every day. In fact, a large percentage of hospital patients receive such drugs at some time during their treatment. In one large representative hospital, approximately 40 percent of the patients received such antibiotics.
The antibiotics that are used in these hospitals are not always received from the drug manufacturer in a finished state. Rather, they are usually delivered as lyo-philized (freeze-dried) powders in order to extend the shelf life of the drug. These lyophilized drugs must be reconstituted into liquid form before they may be administered to patients. The lyophilized powders simply cannot be administered to patients without reconstitution.
Admixture programs (centralized drug preparation programs, for the reconstitution of these antibiotics) began in about 1972. Typically, the reconstitution process is described in the manufacturer’s instructions for use which accompany the drug.
When a single dose of an antibiotic, or a very small number of doses, are to be reconstituted, the reconstitution is done with a needle and syringe. An appropriate amount of sterile reconstituting fluid is drawn into the syringe, the needle is inserted through the stopper of the manufacturer’s vial of lyophilized drug, and the reconstituting fluid is inserted into the vial. The powder and fluid are then agitated to form a solution.
Many intravenous antibiotic drugs are also diluted after reconstitution and before administration to patients. Several manufacturers, including Baxter, produce FDA-approved, pre-filled, diluent containers (an intravenous solution plastic bag) used for the administration of a single dose of a drug.
After reconstitution and dilution in the pre-filled, plastic polyvinyl chloride bags, the antibiotics are ready for intravenous infusion. One of the polyvinyl chloride bags that is used for this purpose is manufactured by Baxter, and sold under the trademarks of Viaflex and Minibag.
The single dose preparation technique for intravenous antibiotics is practiced at virtually all hospitals. Many larger hospitals, however, cannot meet their need for intravenous antibiotic drugs through the single dose preparation method. Many of these hospitals have established centralized drug preparation programs in which they reconstitute, dilute, and repackage the antibiotics in batches. It is not cost-effective for a hospital of more than 300 beds to engage in single dose reconstitution of antibiotics. Hospitals that engage in this type of centralized drug preparation do not necessarily reconstitute the antibiotics in response to an order for a specific patient. Some of these hospitals prepare batches of reconstituted antibiotics sufficient for one day’s needs. Other hospitals prepare large batches intended to be used over a relatively long period of time. At least one hospital has prepared as many as 4,000 doses of reconstituted antibiotics at one time.
Hospitals that prepare single doses of reconstituted and/or diluted antibiotics may encounter numerous problems. These include: selection of the wrong drug, contamination during preparation, improper dosage, incorrect drug concentration, drug deterioration, labeling problems, and incompatible mixtures of drugs or fluids. Moreover, all hospitals experience critical care problems if prepared antibiotic drugs are not available.
Many reconstituted intravenous antibiotics have very short stability periods if maintained at room temperature, or even under refrigeration. In the 1960’s and 1970’s, research demonstrated that freezing could extend the stability period for most such drugs. Soon after the first research results were published, many hospital pharmacies began developing drug *1414freezing programs. It was common for hospitals to reconstitute and freeze antibiotics in anticipation of need.
For many years, training programs have been conducted across the United States to teach hospital pharmacists how to develop drug admixture programs. Information on freezing antibiotics has been included in such programs. Virtually every hospital in the United States which has more than 300 beds has a pharmacy which operates an admixture program and many of these programs engage in the freezing of reconstituted antibiotics.
Once freezing programs were developed, drug storage times were extended. The extensions were determined by reference to published literature, the institution’s own studies, the manufacturer’s labeling, or technical letters from manufacturers. Most hospitals, however, do not possess the facilities to engage in sophisticated analytical testing of these products. In fact, very few of these hospitals conduct stability testing on these frozen reconstituted antibiotics.
Baxter initiated its TRC program in 1982. The concept was that the TRC would perform the same functions as the hospital pharmacy, but more efficiently, more safely, and at a reduced price.
In June 1982, Baxter transmitted a letter to the FDA Chicago District Office describing the TRC process contemplated by Baxter. FDA raised no regulatory concerns or objections to the Baxter program, and the TRC commenced operation in July 1982 at Morton Grove, Illinois.
When the TRC program was commenced, only antibiotics were reconstituted. Baxter purchased antibiotics that had been fully approved by the FDA, and those antibiotics were received in the same form as they were supplied to hospitals. The drugs are accompanied by FDA-approved labeling which includes instructions for reconstitution and/or dilution. Each of the diluents and reconstituting fluids used by Baxter is fully approved by the FDA. Even the director of the FDA Chicago District Office has acknowledged that “the powders and liquids that the TRC currently uses as starting materials are themselves FDA approved finished pharmaceuticals.”
Baxter reconstitutes these antibiotics according to the manufacturer’s instructions. The antibiotics are reconstituted with a fluid recommended in the manufacturer’s FDA-approved labeling. After agitation to assure that the dry powder has dissolved, single doses of the reconstituted drug are aspetically transferred to FDA-approved Viaflex containers filled with an appropriate diluent.
The reconstitution and dilution steps performed by the TRCs are the same as those performed by a hospital. The environment and controls present in the TRC’s operations, however, are more sophisticated than those in virtually any hospital or physician’s office. In the TRCs, drug preparation occurs in a special environmentally controlled clean room. The room is supplied with filtered air at a pressure above that present in surrounding rooms. This prevents airborne contamination from entering the room from outside sources. Personnel working in the room are attired in special clean room clothing. They have been specifically trained for their work activities. Preparation steps are carried out in vertical laminar flow hoods to further control the environment in which actual drug preparation occurs. They also protect the workers from hazardous drug vapors and spills. Significant process controls are also applied, including the regular testing of sterile media fills to verify aseptic techniques. Essentially, the TRC performs the same function as a hospital pharmacy but on a much larger scale. This process, however, is accompanied by substantial advantages in safety, effectiveness, and reduced cost.
After the antibiotics have been reconstituted, they are frozen in the Viaflex containers. This is in essence the same operation that is performed at the admixture centers of larger hospitals. Baxter has conducted stability studies on each of these frozen antibiotics, and expiration dates are assigned to each product based upon these studies.
*1415The freezing of the antibiotics is specifically mentioned in the package inserts of six of these products. For the other two, Baxter has received letters from the manufacturer which support use of the freezing process. For three of the antibiotics under consideration, ceftazidime, cefuroxime and cefamandole, the manufacturer’s package insert specifies both freezing and an expiration date which is consistent with that used on the TRC product. For three other antibiotics, nafcillin, piperacillin, and ticar-cillin, the manufacturer’s labeling mentions freezing, but the TRC expiration date is somewhat extended based on Baxter’s own stability studies. For the remaining two antibiotics, penicillin G and tobramycin, the package insert does not specifically mention freezing. Both the freezing of these products and the extended expiration dates, however, are justified by Baxter’s own stability studies. FDA is specifically authorized to review the stability testing and information that has been conducted at Baxter, and has used this authority on several occasions.
In the first seven years that the TRC has reconstituted and shipped these products, it has enjoyed an excellent safety record. During this time period, approximately seven million products have been produced at the TRCs, and there have been only 14 medical complaints. Moreover, none of these complaints have been attributed to the reconstitution and repackaging process at the TRC.
The district judge, Brian Barnett Duff, who granted the preliminary injunction, held that these antibiotics were misbranded pursuant to 21 U.S.C. § 352(f)(1). This conclusion was based solely on the determination that they required new drug approval, and that no such approval had been obtained.
FDA’s regulations state that a drug will be exempt from the requirements under § 352(f)(1) relating to adequate directions for use “to the extent to which such an exemption is claimed in an approved application with respect to such drug....”
Because of my opinion that it is inconceivable under the facts and circumstances of this case that the TRC program itself is subject to the new drug procedure requirements, it would follow that Baxter is exempt from the requirements of paragraph 352(f)(1) and accordingly they are not mis-branded within the meaning of that section of the statute. It is of interest to note that the FDA was informed of Baxter’s intention to commence the TRC program even before it became operational. This was by letter to the district director of FDA in Chicago on June 16, 1982. FDA did not raise any regulatory objection to this program and Baxter began reconstituting drugs in July 1982. FDA investigators conducted inspections at the TRC in June 1983 and June 1985. Neither of these inspections resulted in any expressions of concern by the FDA. Other inspections were conducted but it was not until October 1986 that the FDA issued a Form 483, Inspectional Observations, which is FDA’s method of informing the inspected facility of the agency’s belief of a violative practice. Even that notice of observation, however, did not raise questions concerning safety or efficacy, nor did it suggest that Baxter’s reconstitution and repackaging activities required the submission of a new drug application. Thereafter a series of meetings were held between representatives of Baxter and FDA followed by a seizure complaint filed in May 1987, which was itself replaced a year later by the complaint for injunction in April 1988.
At all times since July 1982 Baxter has engaged in the reconstitution process for the eight antibiotics in question in the manner I have described earlier. The fact that it is still doing so is attributable to stays in which four federal judges have been involved one way or another. Judge Duff issued his order and memorandum opinion granting the preliminary injunction on April 26, 1989. Because apparently 1200 hospitals relied on Baxter for these reconsi-tuted antibiotics and because they could not begin to reconstitute them immediately, Baxter sought a modification of the preliminary injunction to allow continued shipment during a 60-day transition period. Over the FDA’s objection, the court stayed the preliminary injunction until May 16, *14161989. On that date, the court issued an amended memorandum correcting a misstatement in the original order that antibiotic reconstitution had already been terminated before the preliminary injunction order was issued.
Baxter moved for stay of the injunction pending appeal. In Judge Duffs absence, the motion was initially presented to the designated emergency judge, Marvin E. Aspen, on May 30, 1989. Judge Aspen entered an interim order staying the injunction until June 10, 1989. On June 1, 1989, Judge Duff vacated the interim order and denied Baxter’s motion for stay. Baxter then moved in this court for a stay pending appeal, the notice of appeal having been filed on May 23, 1989. This motion was supported by a memorandum totaling 133 pages of which 17 pages were a legal memorandum. While this was pending before this court, the FDA filed an interim response to Baxter’s request for a stay consisting of 17 pages objecting to a stay. Two days later the FDA filed a supplemental response consisting of 14 pages, again vigorously opposing the granting of a stay pending an appeal. On June 14, 1989, a three-judge panel consisting of Judges Bauer, Cummings, and Kanne granted Baxter’s motion for a stay of the preliminary injunction, which effectively permitted Baxter to continue reconstituting and shipping antibiotics during the pendency of the appeal, which it currently is continuing to do.
I am unable to find any case squarely on point on the issue of whether the TRC’s reconstituting process, as I have outlined it earlier, constitutes in itself the creation of a new drug, as that term is used in the Federal Food, Drug, and Cosmetic Act in relation to the entire drug product. I do find, however, support for the position I have taken in its recognition by the United States Supreme Court in United States v. Generix Drug Corp., et al., 460 U.S. 453, 103 S.Ct. 1298, 75 L.Ed.2d 198 (1983). The Court reserved the question of whether two demonstrably bioequivalent products containing the same active ingredients, but different excipients, might under some circumstances be the same drug. 460 U.S. at 460-61, 103 S.Ct. at 1302. Here in the present case we have only one drug with an active ingredient. Here the excipient is an inert substance used to give the active ingredient, FDA approved, a suitable form for use by hospitals. There is no argument that the active ingredient itself cannot be administered to a patient in its liquid or powder form.
In the case at bar, the TRC products contain the same active and inactive ingredients as the approved product, and are of equal bioavailability and are bioequivalent to those approved products. Thus, it appears to me, the TRC products are the same “drug” as that which the FDA has approved.
The term “bioavailability” is defined in 21 C.F.R. § 320.1(a). The regulation states that bioavailability of a drug means the:
Rate and extent to which the active drug ingredient or therapeutic moiety is absorbed from a drug product and becomes available at the site of drug action.
This, it appears to me, is the degree to which a drug or other substance becomes available to the target tissue after administration.
In the case at bar, both the TRC products and the products as approved by FDA are intended for intravenous infusion. Because of this route of administration, they are presumed by FDA to be of equal bioa-vailability. See 21 C.F.R. § 320.22(b). Thus, there would seem to be no issue concerning the bioavailability of the TRC products and the FDA-approved products.
In addition, the Code of Federal Regulations also defines the term “bioequivalent.” Bioequivalent drugs are those whose rate and extent of absorption do not show a significant difference when administered at the same molar dose of therapeutic moiety under similar experimental conditions. 21 C.F.R. § 320.1(e).
In the case at bar, the TRC label for each of these products does not change the circumstances or type of use of the product. In fact, the TRC label specifically incorporates the instructions and prescription directions in the approved package insert. Thus, the TRC does not change anything *1417about the product that would affect the rate and extent of absorption, and thus the TRC products are bioequivalent to the products that have already been approved by FDA.
In essence, the TRC products are an example of the products discussed in the Generix decision that exhibit equal bioa-vailability and are bioequivalent to one another and, according to Generix, may very well be the same “drug” for purposes of the Act. On the facts of this case, the TRC products are, in my opinion, the same product as the drug purchased. They are not imitations or variations of the eight FDA antibiotics. They are the eight approved antibiotics now ready to be administered to a patient. They are simply brought farther along in their necessary preparation for use than were the drugs formerly. They are not, as the FDA seems to contend, generic drugs.
The characteristics that will be reviewed by FDA to determine whether a drug product has become a “new drug” which requires its own approval under 21 U.S.C. § 355 are described in 21 C.F.R. § 310.3(h)(l)-(5). That regulation states in pertinent part:
(h) The newness of a drug may arise by reason (among other reasons) of: (1) the newness for drug use of any substance which composes such drug, in whole or in part, whether it be an active substance or a menstruum, excipient, carrier, coating, or other component.
(2) The newness for drug use of a combination of two or more substances none of which is a new drug.
(3) The newness for drug use of the proportion of a substance in combination, even though such combination containing such substance and other proportion is not a new drug.
(4) The newness of use of such drug in diagnosing, curing, mitigating, treating, or preventing a disease, or to affect a structure or function of the body, even though such drug is not a new drug when used in another disease, or to affect another structure or function of the body.
(5)The newness of a dosage, or method or duration of administration or application, or other condition of use prescribed, recommended or suggested in the labeling of such drug, even though such drug when used in other dosage, or other method or duration of administration or application or different condition, is not a new drug. (AA 90-91).
Clearly, the freezing or assignment of an expiration date to a drug that has already been approved is not described by this regulation as an event or characteristic which will require a determination that a “new drug” has been created. I can find no support for FDA’s position.
The freezing and expiration dating of a drug product are events which are controlled by FDA’s extensive regulations concerning compliance with current good manufacturing practices — not by the new drug regulations. The requirement that manufacturers perform and maintain data concerning stability testing is included in 21 C.F.R. § 211.166. This provision and the applicable portions of the Act accord FDA full authority to conduct inspections, review the applicable data, and initiate legal proceedings to compel compliance with current good manufacturing practices if that inspection and review indicates that there has been a serious divergence from the requirements of the regulation. As I have previously stated, Baxter concedes this right of inspection and the agency’s right to compel compliance with good manufacturing practices.
In addition, FDA’s Compliance Policy Guide for repackaging of drugs indicates that the preparation and retention of stability data is a matter to be controlled under the CGMP regulations. FDA’s Compliance Policy Guide, at § 7132.13 states:
Generally, we do not consider the CGMP regulations (21 C.F.R. Parts 210 and 211) to require repackers of finished dosage form drugs to perform analytical testing such as chemical identity tests or assays, or to require receipt of reports of analysis, on a batch-by-batch basis for drug products which are repacked under the following circumstances....
*1418The policy in the Compliance Policy Guide applies to the question of adequate batch-to-batch — testing examination criteria for routine acceptance and release of drug products which are repacked. It does not alter any testing which repack-ers may be required to perform on drug products from other standpoints, such as any stability testing required in order to establish appropriate expiration dates in the container-closure system used by the repacker, testing which may be required to determine the suitability of repacker’s drug product containers and closures,....
FDA Compliance Policy Guide, § 7132.13 (emphasis added). Thus, FDA’s own regulations concerning repackers clearly indicate that the CGMP regulations are to be used when there is a question concerning stability data and the assignment of expiration dates.
I can conceive of no sound policy argument to support the conclusion that these factors must be controlled by means of the new drug application process. A new drug application, filed under 21 U.S.C. § 355(b)(1), (j)(l) or 357, must contain extensive data and information about a variety of issues. Rather than employ this very extensive, expensive, and protracted procedure to deal specifically with the issues of stability in freezing and expiration dating, FDA is fully justified in utilizing the more specific and limited authority conferred by the CGMP regulations.
The Government seems to take an inconsistent position on exactly what is the issue in this case. It acknowledges on page 34 of its brief that “FDA does not charge that the TRC antibiotics are new drugs,” but also includes extensive arguments about why the TRC products are new drugs requiring approval.
I agree with Baxter that these are not new and different products, and therefore no legal or policy need exists to implement the whole regulatory approval process twice for each of these drug products. The TRC products should be exempt from requirements for batch certification and approval because they are the actual approved products. Hence, these products are not, as argued by the Government, misbranded under 21 U.S.C. 352(i) because they are the products for which the original manufacturers have been granted exemption as provided for in 21 C.F.R. § 433.1.
It appears to me that the Government’s brief placed incorrect reliance on certain statutory provisions and regulations to support its argument that the TRCs are engaged in “manufacturing.” Gov’t’s Brief, pp. 42-43. It cites 21 U.S.C. § 360(a), 21 C.F.R. §§ 207.3(a)(8), 210.3(b)(12), and 21 C.F.R. § 201.1(b)(10). These statutory and regulatory sections, however, do not define “manufacturing” as such. They are all related to matters peripheral to the drug manufacturing process, such as registration of producers, (21 U.S.C. § 360(a); 21 C.F.R. § 207.3(a)(8)) or good manufacturing practices, (21 C.F.R. § 210.3(b)(12)), or labeling (21 C.F.R. § 201.1(b)(10)). None can be considered an authoritative definition, for each is obviously intended for purposes of its specific context only. Ironically, Baxter does not even meet the definition of “manufacturer” in 21 C.F.R. § 201.1(b)(10), because it performs only one of the ten requisite activities listed in that definition. 21 C.F.R. § 201.1(b). Under this regulation, Baxter could not label the TRC products as having been manufactured by Baxter, but in fact, pursuant to this regulation, labels the products “Repackaged by Baxter Healthcare Corporation.” Gov’t’s Brief, p. 42.
It is also my opinion that the district court erred in its opinion by asserting that it owed deference to the FDA on issues of scientific interpretation and application. These are legal issues concerning the legal application of the Federal Food, Drug, and Cosmetic Act and implementing regulations. These are legal issues which are entirely within the province of that, and this, court to decide.
I come away from reviewing this appeal with the unfortunate feeling that we have a case of bureaucratic overreaction to Baxter’s declination to engage in a process which it did not deem legally necessary, in which position I fully concur. I do not *1419intend to denigrate or minimize the importance of the Federal Drug Administration in protecting the interest of the public’s safety in the use of drugs necessary in the continuation of good health. The diligence of FDA is testified to obviously by the periodic references in the media to drugs which are being administered in other parts of the world with effectiveness and apparent safety but which have not yet been approved for use in this country. I am putting aside in preparing this dissent any idea that the FDA in this case has succumbed to the necessities of its annual presentation of a budget to preserve, and ordinarily increase, that which Congress gives to it. Such budget requests, of course, are predicated to a considerable extent on justification of personnel needs, and I would be most disappointed in this respected governmental agency if it was devoting itself unnecessarily to an easily found target such as an established pharmaceutical company rather than to fly-by-night generic substitute operators.
I am not unmindful of the disclaimer in the majority opinion that this court has not prejudged the merits of the controversy which will be before the district court after trial. With respect, however, I must observe that the holding on the legal matters involved seems supportive of the legal position taken by the district judge in granting the preliminary injunction and I cannot believe that the able district judge would do otherwise than to apply the same legal principles he did in granting the FDA’s motion. If the disclaimer is to be effective, it is my opinion, reached reluctantly, that on remand, if there is no subsequent action in this court, the ease should be assigned to a different judge of the district court.