APSXAS 2 (2) (65-128) (1990)
`
`airsCoe
`
`faageel
`
`Byfe)geile)
`
`Editors:
`ma10ale)Sxelc1¢@
`R.B.Sund
`
`RON
`
`ADAMIS EXHIBIT 1007
`Page A
`Page A
`
`ADAMIS EXHIBIT 1007
`
`

`

`Racemisation and oxidation in adrenaline
`injections
`
`, T. A. Langvik, P. Hasselgard and P. 0. Roksvaag
`G. Fyllingen* 1
`Military Pharmaceutical Laboratory, Joint Medical Service, P. 0 . Box 107 Veitvet,
`N-0518 Oslo 5, Norway
`
`Adrenaline injection fluids aged between 3 and 33 years were anaJyzed with respect to
`oxidation and racemization. The oxidation was determined by ion-pair reversed phase
`HPLC, and the degree of racemization was determined by derivi\ti7.ation of the adrenaline
`isomers to diastereomeric forms and subsequently separated by reversed phase HPLC.
`10% adrenaline was oxidized after about 11 years, while 10% L-adrenaline was converted
`to 0-adrenaline after only 4 years. After about 4 years, the injections contained less than
`90% active adrenaline.
`
`For military purposes, drugs may be stored for years past their ordinary expiration
`date. In Norway, the storage conditions vary considerably due to the shifting climate
`and a decentralized storage system with many small depots of varying quality.
`The shelf-life of pharmaceutical preparations may be estimated by accelerated
`studies at high temperatures. In practice, such studies are of limited value where
`extremely long-term storage is concerned. Therefore, shelf-lives should be based on
`retrospective studies of drugs stored under realistic conditions.
`The Norwegian armed forces store their drugs for emergency purposes for up to 15
`years. Earlier studies have shown that several drugs may be stored for this long
`without a deterioration of quality (1, 2].
`Adrenaline is an important drug in military medicine. In Norway, the shelf-life of
`ampoules with adrenaline injections is 3 years. The present study was undertaken to
`document the long-term stability of adrenaline injections under extreme storage
`conditions.
`There are two optical isomers of adrenaline, of which only L-adrenaline is biologi(cid:173)
`cally active and used for injections.
`L-adrenaline is easily racemized in acidic solutions [3]. The kinetics of the race(cid:173)
`mization have been determined, and the reaction rate was estimated to 10% race(cid:173)
`mization at pH 3-3.5 in 3 years (4]. Later, it was shown that adrenaline injections in
`ampoules stored for 7 .5 years at a temperature less than 15°C had racemized only to a
`very small extent [5]. Injections of local anaesthetics containing adrenaline conta(cid:173)
`ined 5% or less D-adrenaline after the expiration date [ 6].
`Adrenaline in solutions is easily oxidized and the reaction is catalyzed by bases
`[7- 10]. The reaction is complex, and only the intermediate degrada tion products
`
`• Correspondence.
`1 Current address: Apothekernes Laboratorium A. S., P. 0. Boks 158 Sk0yen, N-0212 Oslo 2, Norway.
`
`Act.a Pharm. Nord. 2(5)1990
`
`355
`
`ADAMIS EXHIBIT 1007
`
`Page 355
`
`

`

`·- . ' . .. .. ______ _
`
`have been identified [8J. The end products are characterized as coloured melanins.
`10-12% oxidation has been shown in 7.5 year old solutions [5].
`The most common method of protecting adrenaline in injections against oxidation
`is by using antioxidants such as sodium bisulfite or sodium metabisulfite. But the
`bisulfite may also attack adrenaline. and the reaction product is adrenaline sulfonic
`acid [11-13]. The rate of bisulfite addition is normally low compared to the oxida(cid:173)
`tion rate.
`HPLC has been widely used for analysis of adrenaline in the recent years. Total
`adrenaline may be determined by reversed phase ion-pair liquid chromatography
`[14]. Recent developments in enantiomeric separation have made it possible to
`determine D- and L-adrenaline by reversed phase liquid chromatography after
`derivatization of the isomers to diastereomeric forms [6, 15].
`
`Experimental
`
`Chemicals
`Methanol and citric acid monohydrate, both of analytical grade, Merck (Darmstadt, F. R. G.);
`potassium dihydrogenphosphate, analytical grade, Riedel-de Haen (Seelze, F. R. G.); L(cid:173)
`adrenaline hydrogen tartrate. Boeringer Ingelheim (lngelheim, F. R. G.); isoprenaline
`sulfate, NMD (Oslo, Norway); DL-adrenaline hydrochloride and hydrazine hydrate 80%,
`TCI (Tokyo, Japan); 2,3,4,6-tetra-O-acetyl-~-D-glucopyranosyl isothiocyanate (GITC), Po(cid:173)
`lysciences (Warrington, PA, U.S.A.); dimethylformamide, analytical grade, BDH Chemicals
`(Poole, England).
`
`Solutions
`Buffer pH 3: 1.05 g citric acid was dissolved in 250 ml methanol and 700 ml water. The pH was
`adjusted to 3.0 and the solution was diluted to 100 ml.
`
`Buffer pH 2.9: 1.36 g KH2P04 was dissolved in 900 ml water. The pH was adjusted to 2.9 and
`the solution was diluted to 1000 ml.
`
`Internal standard: 200 mg isoprenaline sulfate was dissolved in 100.00 ml buffer, pH 3.
`
`Apparatus
`The Shimadzu LC 4 A (Kyoto, Japan) liquid chromatograph with gradient mixer and variable
`UV-spectrophotometric detector was used.
`
`Total adrenaline: The samples were chromatographed on a Hibar Lichrocart RP 18 column,
`250x4 mm, 7 µm particles. Merck (Darmstadt, F. R. G.). The detector was operated at 280
`nm. The mobile phase was 20 ml PIC B7 reagent, Waters (Milford, MA, U.S.A.), 250 ml
`methanol and water to 1000 ml. 20 ml PIC B7 reagent provides 0.005 M heptasulfonic acid and
`buffer pH 3 when diluted to 1000 ml. The flow rate was 1.5 ml/min at ambient temperature.
`
`356
`
`Acta Phann. Nord. 2 (5) 1990
`
`ADAMIS EXHIBIT 1007
`
`Page 356
`
`

`

`D- and L-adrenaline: The samples were chromatographed on a Spherisorb ODS 2 column,
`250x4.6 mm, 5 µm particles, Phase Sep (Queensferry, U. K.). The detector was operated at
`254 nm. The mobile phase was 630.00 ml aqueous buffer pH 2.9 and methanol to 1000.00 ml.
`After 22 min, the mobile phase was changed by a gradient mixer to 300.00 ml aqueous buffer
`pH 2.9 and methanol to 1000.00 ml. After 36 min, the original concentrations of the mobile
`phase was reconstituted. The flow rate was 1.5 mUmin at ambient temperature.
`
`Samples
`
`Samples of 18 batches of adrenaline injections stored in 6 different military depots for 3-30
`years were kindly supplied by the regional military pharmacies in Norway. Altogether, 24
`different combinations of batches and depots were analyzed. The adrenaline injections
`contained L-adrenaline hydrogentartrate corresponding to 1 mg/ml of L-adrenaline base.
`
`Sample preparation
`
`Total adrenaline: 1.00 ml adrenaline injectio n and 1.00 ml internal standard solution were
`diluted to 10.00 ml with buffer pH 3. 20 µl of the solution was injected into the chromatogra(cid:173)
`phic system.
`
`I
`
`I I
`
`0
`
`2
`
`4
`
`6
`TIME (min.)
`
`Fig. 1. A chromatogram of total adrenaline (/) and inter(cid:173)
`nal standard, isoprenaline (//).
`
`Acta Phann. Nord. 2 (5) 1990
`
`357
`
`ADAMIS EXHIBIT 1007
`
`Page 357
`
`

`

`D- and L-adrenaline: 1.00 ml adrenaline injection was diluted to 10.00 ml with 35% methanol
`in water. 1.00 ml of this solution was evaporated at lOO°C with N2 till dryness. 100 µJ 2% GITC
`in dimethylfonnamide was added to the residue. After 10 min in a water-bath at 50°C, 20 µI of
`0.5% hydrazine in dimethylformamide was added. After 10 minutes at room temperature,
`1.00 ml mobile phase was added. 20 µl of the solution was injected into the chromatographic
`system.
`
`Reproducibility
`Total adrenaline: A solution containing L-adrenaline hydrogen tartrate corresponding to 0.1
`mg/ml adrenaline base, and 0.2 mg/ml isoprenaline sulfate was injected 10 times into the
`chromatographic system.
`
`D- and L-adrena/ine: 10 aliquots of a solution containing a mixture of L-adrenaline hydrogen
`tartrate and DL-adrenalin hydrochloride corresponding to 1 mg/ml adrenaline base, 75% of
`
`I
`
`I I
`
`0
`
`20
`10
`TIME (min. )
`
`30
`
`Fig. 2. A chromatogram of L- and 0 -adrenaline derivatized with GITC.
`The ratio between L- and 0-adrenaline concentrations is to 3:1.
`I= L-adrenaline, II= D-adrenaline.
`
`358
`
`Acta Pharm. Nord. 2(5) 1990
`
`------------------ ··- ·· ·---·····.
`
`ADAMIS EXHIBIT 1007
`
`Page 358
`
`

`

`the L-isomer and 25% of the D-isomer were treated as described in sample preparation. Each
`sample was injected twice into the chromatographic system.
`
`Standard curves
`Total adrenaline: A 5-point standard curve with adrenaline hydrogen tartrate corresponding
`to the concentrations 0.05, 0.10, 0.15, 0.20, 0.25 mg/ml of adrenaline base was used. The
`concentration of internal standard, isoprenaline sulfate was 0.20 mg/ml.
`
`D- and L-adrenaline: A 6-point standard curve with mixtures of L-adrenaline hydrogen
`tartrate and DL-adrenaline hydrochloride was used. The ratios between the L- and D-isomers
`were 50:50, 60:40, 70:30, 80:20, 90:10 and 100:0, calculated as adrenaline base. The total
`adrenaline concentration was 1 mg/ml in each solution. The lowest L-adrenaline concentra(cid:173)
`tion was 0.5 mg/ml and the highest was 1 mg/ml.
`
`90
`
`-0 c:
`:J
`~ 80
`Cl
`.E
`0 70
`c:
`f
`-0 ? 60
`_J
`lit
`
`50
`
`'40+-~+-~+-~+-~+-~+-----1
`50
`60
`70
`80
`90 100
`'40
`~ L-adrenaline token
`
`l:=J 0-ADRENAUNE
`.:I L-ADRENAUNE
`
`Fig. 3. The standard curve of L- and D-adrenaline.
`Percent of L-adrenaline taken is plotted against percent
`L-adrenaline found. The curve fits the second-order
`equation: y=43.9-0.32x+ a.r10- 3x2
`•
`
`eedbeb bdbb eb debobo boo a
`
`100
`
`80
`
`60
`
`4-0
`
`20
`
`0
`
`c .,
`c 0
`"O .. ...
`0 u .. "O
`
`....
`0
`~
`
`0
`
`38 45 54 66 66 71 88 88 90 96 96 100 109 109 123 123 138138 141157158164 177 360
`Age
`(month,]
`Fig. 4. The content of L- and D- and total adrenaline in the stored samples. A-F denotes
`different depots.
`
`Acta Phann. Nord. 2 (5) 1990
`
`359
`
`ADAMIS EXHIBIT 1007
`
`Page 359
`
`

`

`Results and discussion
`
`The methods
`
`A chromatogram of total adrenaline and internal standard is shown in Fig. 1. No
`degradation products were found in any of the samples by this method. The test for
`reproducibility of this method showed that the error of injection was 1.0% RSD.
`A chromatogram of D- and L-adrenalioe derivatized with GITC is shown in Fig. 2.
`The concentration of methanol in the mobile phase was increased during each run to
`
`I
`
`0
`
`10
`
`30
`
`40
`
`20
`Time (min.)
`Fig. 5. A chromatogram of an adrenaline sample stored in military depot b for 158 months.
`30.4% of the remaining adrenaline had racemized to the inactive D-isomer. Combined with the results
`from the determination of total adrenaline, this sample contains 57.6% of the declared content of active
`adrenaline.
`I:.:: L-adrenaline, II= D-adrenaline.
`
`360
`
`Acta Phann. Nord. 2 (5) 1990
`
`ADAMIS EXHIBIT 1007
`
`Page 360
`
`

`

`reduce the retention time for degradation products of GITC. The test for reproduci(cid:173)
`bility of this method showed that the total error of derivatization and injection was
`3.9% RSD. The standard curve for determination of D- and L-adrenaline was not
`linear. Repeated measurements showed that the curve fitted into a second-order
`equation (Fig. 3). This might be due to chiral impurity of either the adrenaline or the
`derivatization reagent.
`Because of the complicated sample preparation, an internal standard was not used
`in the method for D- and L-adrenaline. This method is only able to determine the
`degree of racemization. Both methods are necessary to determine the content of
`active adrenaline.
`
`Long-term stability
`The results of the analyses of adrenaline injections by the two methods are presented
`in Fig. 4. A chromatogram of a sample stored for 158 months is shown in Fig. 5.
`The oxidation seems to be a rather slow process. 10% of the adrenaline has
`oxidized after 11 years. These results indicate that the oxidation is a slower process
`than described in earlier literature [5]. During the manufacturing of ampoules of
`adrenaline the empty volume may be filled with an inert gas such as nitrogen, which
`might explain the slow oxidation rate.
`The racemization seems to be a faster process than the oxidation. 10% of the
`remaining adrenaline had racemized to the biologically inactive D-isomer after
`about four years. These results indicate that the racemization of adrenaline in
`ampoules is faster than described in earlier literature [5]. There may be several
`reasons for this. The samples have probably not been stored at cool temperature all
`the time. Elevated temperatures may result in an increased racemization rate. An
`decrease in pH during storage will lead to a higher racemization rate (3), but because
`of a limited sample volume the pH was not measured.
`The content of active adrenaline is outside the specifications limits(± 10%) after
`about four years. This makes it difficult to store adrenaline injections for military
`purposes in Norway.
`To improve the shelf-life, raising the pH of the injection somewhat should be
`considered. The pH in the injections is 2.4, which is supposed to be the pH at which
`both oxidation and racemization are at a minimum. A rise in pH may increase the
`oxidation rate, but it would at the same time reduce the racemization rate. In this
`way, the shelf-life for military storage could possibly be prolonged.
`
`Conclusion
`
`This study shows that adrenaline injections are not suited for long-term storage
`under the conditions prevailing at Norwegian military depots. The racemization is
`the limiting degradation process, and the injections should not be used after the
`expiration date.
`
`Acta Phann. Nord. 2 (5) 1990
`
`361
`
`ADAMIS EXHIBIT 1007
`
`Page 361
`
`

`

`References
`1. Roksvaag P. 0 ., Brummenaes H. I. and Waaler T. Pharm. Acta Helv. 54, 180 (1979)
`2. Roksvaag P. 0 ., Fredrikson J . B. and Waaler T. Pharm. Acta Helv. 55, 198 (1980)
`3. Haddock L. Quart. J. Pharm. and Pharmaco/. 6, 496 (1933)
`4. Schroeter L. C. and Higuchi T. J. Am. Phann. Assoc. 47, 426 (1958)
`5. Lundgren P. and Str0m S. Acta Pharm. Suecica 3, 273 (1966)
`6. Allgire J. F., Junge E. C., Damo C. P., Sullivan G. M. and Kirchhofer R. D. J.
`Chromatogr. 325, 249 (1985)
`7. Millard B. J., Priaulx D. J. and Shotton E . J. Pharm. Pharmac. 25, Suppl., 24 (1973)
`8. Milano E. A., Waraszkiewicz S. M. and Dirubio R. J. Pharenter. Sci. Technol. 36, 232
`(1982)
`9. Trautner E. M. and Bradley T. R. Aust. J. Sci. Res. 4B, 303 (1951)
`10. Pelizetti E., Mentesti E., Pramauro E. and Gazetta. Chem. Ital. 105, 307 (1975)
`11. Schroeter L. C. , Higuchi T. and Schuler E. E. J. Am. Pharm. Assoc. 47, 723 (1958)
`12. Higuchi T. and Schroeter L. C. J. Am. Pharm. Assoc. 48, 535 (1959)
`13. Higuchi T. and Schroeter L. C. J. Am. Chem. Soc. 82, 1904 (1960)
`14. Williams 0 . A ., Fung E. E. Y. and Newton D. W. J. Pharm. Sci. 71, 956 (1982)
`15. Nimura N., Kasahara Y. and Kinoshita T. J. Chromatogr. 213, 327 (1981)
`
`Received February 5, 1990.
`Revised edition accepted March 2, 1990.
`
`362
`
`Acta Phann. Nord. 2 (5) 1990
`
`ADAMIS EXHIBIT 1007
`
`Page 362
`
`