`
`YAKUGAKU ZASSHI 121(6) 459―463 (2001)
`
`459459
`
`―Notes―
`
`Occurrence of Coring in Insulin Vials and Possibility of Rubber Piece
`Contamination by Self-Injection
`
`Toshinari ASAKURA,,a Hiroaki SEINO,b Seishiro NOZAKI,a and Ryuzo ABEc
`Department of Pharmacya and Department of Internal Medicine Diabetes Center,b Ohta Nishinouchi
`Hospital, 2520 Nishinouchi, Koriyama, Fukushima 9638558, Japan, and Department
`of Internal Medicine,c Ohta Memorial Hospital, 524 Nakamachi, Koriyama,
`Fukushima 9638004, Japan
`
`(Received February 5, 2001; Accepted March 23, 2001)
`
`Coring is reported to occur because rubber pieces are shaved oŠ from a rubber stopper when a needle is inserted into
`the rubber stopper of transfusion liquid formulation. We veriˆed whether coring really occurs in insulin vials of self-in-
`jecting patients. We collected insulin cartridges from 30 hospitalized patients and used the primary injection (trial injec-
`tion), the secondary injection and the cartridge remaining preparation as samples. We observed the rubber pieces using a
`microscope and measured the shape, number of pieces. The occurrence rate of coring was 73% for the primary injection,
`47% for the secondary injection and 97% for the cartridge remaining preparation. The rubber pieces in the primary in-
`jection and the secondary injection which went through the needle are mostly in aggregate shape and the rubber pieces in
`the cartridge remaining preparation which did not go through the needle are mostly in needle-like shape. A number of
`small rubber pieces are found in both the primary injection and the secondary injection, indicating a high possibility that
`rubber pieces may be injected under subcutaneous tissue. The coring is considered to occur because needles are repeated-
`ly inserted and rotated at the same spot. It is required to improve the structure to mount a needle to the pen-type injector
`in future. Coring is a very serious problem from the medical and pharmaceutical points of view. Further study should be
`made on the implication to latex allergy and lipodystrophy.
`
`Key words―insulin; coring; needle; vial; diabetes
`
`INTRODUCTION
`
`The pen-type insulin injector (hereunder referred
`to as ``pen-type injector'') has been developed so that
`patients can mount and self-inject insulin from car-
`tridges (hereunder referred to as ``cartridge'') several
`ten times. Therefore, in Japan, many patients are us-
`ing them for 10 years because they are easy to use and
`convenient to carry with, compared with the conven-
`tional disposable syringes.
`Vial preparation is sealed with a rubber stopper
`so that the medical liquid can be used partially for
`several occasions. However, coring is reported to oc-
`cur because rubber pieces are shaved oŠ form rubber
`stoppers if an injector needle of 18 gauge (hereunder
`referred to as ``G'') is used.1,2) Therefore, Japanese
`pharmaceutical ˆrms have provided information to
`prevent coring in vials containing transfusion liquid
`and to give warning to the users. We made the coring
`occurrence test using cartridges because we consi-
`dered it necessary to conˆrm if coring occurs in the
`cartridges, and if the rubber piece generated could
`pass through a needle. As a result, we found that cor-
`ing occurred with extremely high frequency and that
`
`rubber pieces could pass through the needle.3)
`In this test, we checked if the same result could
`be obtained in the cartridges actually used by
`patients.
`
`MATERIALS AND METHODS
`
`We explained the objectives of this test and ob-
`tained consent from 30 patients hospitalized at Ohta-
`Nishinouchi Hospital (hereunder referred to as ``this
`hospital'') and collected the actually used Penˆll300
`(manufactured by Novo Nordisk Pharma, Co., Ltd.)
`as samples. Then, we set the cartridges to NovoPen
`300 (Novo Nordisk Pharma) and, using Micro-Fine
`Plus31 gauge 8 mm needle (Nippon Becton Dickin-
`son, Co., Ltd.), we collected sample A: the primary
`injection, 2 units), sample B: the
`injection (trial
`secondary injection (2 units) and sample C: the car-
`tridge remaining preparation (cartridge remaining
`preparation) (Fig. 1). After the samples are ˆltered
`using 0.45 mm ˆlter (Millipore ˆlter Type: HA, Lot
`No.: H7SH24028), we observed the samples ˆltrated
`using a microscope (OLYMPUS SZ-RT), measured
`the shape, number of pieces and the size and analyzed
`by one-way ANOVA and ScheŠ's test.
`
`MYLAN INST. EXHIBIT 1073 PAGE 1
`
`MYLAN INST. EXHIBIT 1073 PAGE 1
`
`
`
`hon p.2 [100%]
`
`
`
`460460
`
`Vol. 121 (2001)
`
`Fig. 1. Sampling Methods
`Sample A: the primary injection (trial injection, 2 units), Sample B: the secondary injection (2 units) and Sample C: the cartridge remaining preparation (car-
`tridge remaining preparation).
`
`We selected the rubber pieces of 0.01 mm or larg-
`er in size and measured the longer diameter of ag-
`gregate rubber pieces and the length of needle-shaped
`or spiral-shaped rubber pieces. The test was conduct-
`ed in November 1999. The cartridges were mounted
`on pen-type injectors and stored at room temperature
`(22―25°C), shielded from light, when they were used
`by the patients and when the observation was made.
`Each time the patients made self-injection,
`they
`replaced the needle with new one and wiped the rub-
`ber stopper with alcohol-sterilized cotton before the
`tip of the needle was inserted into the rubber stopper.
`The needle was inserted into the pen-type injector car-
`tridge formulation in accordance with the ``Novo
`Pen300 Utilization Guide'' prepared by Novo Nor-
`disk Pharma, Co., Ltd.
`The average age of the 30 patients from whom
`the samples were collected was 59±18, HbAlc was 8.0
`±1.6%, the number of injections made per day was 3
`±1, the unit of injection per day was 29±16 units,
`the number of insertion made to the collected car-
`tridges was 26±15 and the amount of insulin collect-
`ed was 0.9±0.4 ml.
`
`RESULTS
`
`The
`1. Coring Occurrence Rate (Table 1)
`coring occurrence rate of the 30 samples was 22 sam-
`ples (73%) for Group sample A, 14 (47%) for
`Group sample B and 29 (97%) for Group sample C.
`Most of the rubber pieces in Group sample A and
`sample B were in aggregate shape while the rubber
`pieces in Group sample C are mostly in needle-like
`shape (Fig. 2). Spiral-shaped rubber pieces were
`found only in Group sample C.
`Many of the rub-
`2. Size of Rubber Pieces
`ber pieces of Groups sample A and sample B were
`smaller than 0.01 mm in diameter, while the rubber
`pieces of Group sample C were relatively larger in di-
`ameter, 0.01 mm or more. When we compared the
`rubber pieces of 0.01 mm or more in diameter, we ob-
`tained the following results: the average size of Group
`sample A was 0.07±0.06 mm in diameter. The
`average size of Group sample B was 0.04±0.02 mm
`and the average size of Group sample C was 0.18±
`0.15 mm in diameter (Fig. 3). The size of rubber
`pieces of Group sample C was signiˆcantly larger in
`diameter than those of Groups sample A and sample
`
`MYLAN INST. EXHIBIT 1073 PAGE 2
`
`MYLAN INST. EXHIBIT 1073 PAGE 2
`
`
`
`hon p.3 [100%]
`
`No. 6
`
`Table 1. Coring Occurrence Rate
`
`Samples
`Coring occurrence
`rate
`Rubber pieces
`
`aggregate shape
`Needle-like shape
`Spiral shape
`
`Sample A
`
`Sample B
`
`Sample C
`
`30
`22(73%)
`
`30
`14(47%)
`
`30
`29(97%)
`
`29(pieces)
`
`22(pieces)
`7
`0
`
`26
`
`22
`4
`0
`
`58
`
`25
`32
`1
`
`Sample A: the primary injection (trial injection, 2 units), sample B: the
`secondary injection (2 units) and sample C: the cartridge remaining prepa-
`ration (cartridge remaining preparation). The ``samples`` and ``the coring
`occurrence rate`` gives insulin vial number. Except for these, gives rubber
`pieces number.
`
`Fig. 2. Needle-Like Shape Rubber
`(magniˆcation×140)
`
`B (p<0.01 by ScheŠ's test).
`3. Observation of the Inside of Rubber Stopper
`By observation of the inside of rubber stoppers, we
`found that the rubber pieces were pushed out by a s-
`eries of needle insertion and depressions were made
`by the dropped rubber pieces (Fig. 4).
`
`
`
`461461
`
`DISCUSSION
`
`There are two problems with injection: One is
`that glass pieces from an ampoule are entrapped into
`injection liquid when it is cut and second is that cor-
`ing of rubber pieces occurs when a needle is inserted
`into a vial, as discussed above. The mechanism of
`coring is that the rubber stopper is shaved when a nee-
`dle is stuck in a rubber stopper of vial. For this rea-
`son, the insertion angle of the needle and the ap-
`proach angle to the heel of the vial are important. At
`present, Japanese transfusion and medicine manufac-
`turers are collaborating to issue brochures to prevent
`coring and to instruct users how to insert an injector
`needle. The brochures tell you to (i) insert a needle
`slowly and vertically, (ii) not to twist the needle while
`it is inserted and (iii) not to insert it at the same spot.
`In Japan, insulin self-injection charge is not co-
`vered by the health insurance since 1981. Since then,
`with the advent of the pen-type injector and the exten-
`sive education to diabetes patients, it has become
`widely introduced for both type 1 and type 2 diabetes
`patients.
`We investigated the occurrence of coring from
`insulin vial formulation for syringes (29G) and insu-
`lin cartridge formulation for pen-type injector (30G,
`31G) at storage temperature (6―8°C/22―25°C), in-
`sertion angle (90°/60°) and the number of insertion
`(10/30 times). As a result, we have found that coring
`may occur when thin 29-31G needles for subcutane-
`ous injection are used. Coring also occurs when vial
`formulation for syringes is stored at a cool place.
`Coring occurred for the cartridge formulation for
`pen-type injector in both cases.3) Therefore, we con-
`ducted this test to check whether coring occurs in the
`cartridges which was used for actual self-injection by
`the patients. It was found that coring occurred at a
`very high ratio of 97% in the cartridges actually used
`by the patients.
`From the pictures showing the inside of the rub-
`ber stopper, it is judged that the rubber pieces come
`out because the needle shaves oŠ the rubber pieces
`when it is mounted on the pen-type injector. The
`material of rubber stoppers used for insulin formula-
`tion has dual layer structure consisting of butyl rub-
`ber and natural rubber.4) The purpose of using the
`material is to prevent coring and to avoid any eŠect
`on the insulin. To minimize the shaving oŠ of rubber
`pieces, the cutting face of the needle (inner needle)
`
`MYLAN INST. EXHIBIT 1073 PAGE 3
`
`MYLAN INST. EXHIBIT 1073 PAGE 3
`
`
`
`hon p.4 [100%]
`
`
`
`462462
`
`Vol. 121 (2001)
`
`Fig. 3. Size of Rubber Pieces of 0.01 mm or more in Diameter
`Sample A: the primary injection (trial injection, 2 units), Sample B: the secondary injection (2 units) and Sample C: the cartridge remaining preparation (car-
`tridge remaining preparation). Data are presented at means±SD. p<0.01 by ScheŠ's test.
`
`suggested that these small rubber pieces may be in-
`jected into the subcutaneous tissue.
`A ``trial injection'' to release 1 or 2 units of insu-
`lin preparation is performed to take out air after the
`needle is ˆxed before injection and to check if no fault
`with the injector has occurred. Then, the unit of insu-
`lin for injection (2 units in this case) is set and inject-
`ed. At ˆrst, we expected that there would be very little
`chance for the rubber pieces to go through the needle.
`Even if some rubber pieces may have entered in the
`needle when the needle is inserted, they are discharged
`when a trial injection is made and there would be no
`rubber pieces in the actual injection. However, rubber
`pieces were also found in Group sample B in this test.
`Therefore, it is very a serious problem because many
`rubber pieces were found in Group sample B as well
`as in Group sample A, indicating a very high possibil-
`ity that rubber pieces may be injected into the sub-
`cutaneous tissue.
`A number of coring seems to have occurred be-
`cause the needle mounted on Novo Pen300 which is
`used in this test is inserted by rotating it at the same
`spot. However, the pen-type injector currently availa-
`ble in the market employs the same inserting method.
`Therefore, we have to ˆnd out a way to prevent cor-
`ing by enhancing the needle ˆxing method to the car-
`tridge formulation in future.
`In addition, as we found that rubber pieces were
`passed through an injector needle in this test, we have
`to pay attention to the possibility of ``foreign sub-
`stance'' mixture. As the insulin self-injection is sub-
`cutaneous injection into the upper arm, abdominal
`
`Inside of Rubber Stopper
`Fig. 4.
`(magniˆcation×44)
`
`which is inserted into the rubber stopper of the car-
`tridge is so designed that has no blade part. However,
`we observed no improvement in this test and coring
`occurred with high probability. To ``insert needles
`while rotating them on the same spot'' seems to be
`one of the greatest factors to cause coring, rather than
`the rubber material and the cutting face because a
`number of needle-shaped rubber pieces were found in
`the remaining insulin preparation.
`Many small rubber pieces were found in Group
`sample A and Group sample B. The 31G needle has
`an outer diameter of approximately 0.26 mm and the
`inner diameter of approximately 0.15 mm.5) As the
`diameter of a number of rubber pieces found in the
`sample A and sample B of this test is less than 0.15
`mm and could pass through the needles, it is strongly
`
`MYLAN INST. EXHIBIT 1073 PAGE 4
`
`MYLAN INST. EXHIBIT 1073 PAGE 4
`
`
`
`hon p.5 [100%]
`
`No. 6
`
`wall, femur, etc., and not an intravenous injection,
`there is little possibility of rubber pieces entering into
`blood vessels. However, as reported in many cases of
`latex allergy,6) it is necessary to study the in‰uence of
`coring on insulin allergy7―10) or relationship with
`lipodystrophy.11) Therefore, we should concentrate
`on coring prevention and further tests should be made
`to ˆnd the existence of clinically harmful substance.
`
`ACKNOWLEDGEMENT
`
`We are grateful to Mr. Yoshikatsu Chonan and
`Mr. Masahiko Sato of Koriyama Plant of Novo Nor-
`disk Pharma for cooperation for this test and meas-
`urement.
`
`REFERENCES
`
`1) Sakai T., Kho H., Mastuki A., Jpn. J.
`Anesthesia, 45, 15331535 (1996).
`2) Kubota H., Katahira M., Sumioka T.,
`Wakiichi M., Taniyama E., Miyahara K., ICU
`& CCU, 21, 743745 (1996).
`3) Asakura T., Seino H., Mizuno M., Nozaki S.,
`Abe R., Jpn. J. Hosp. Pharm., 25, 407413
`(1999).
`Internal document of Novo Nordisk Pharma,
`Co., Ltd.
`Internal document of Nippon Becton Dickin-
`son, Co., Ltd.
`6) Akira A., IRYO, 51, 201204 (1997).
`7) Leslie C. G. et al., J. Allergy Clin. Immunol.,
`71, 250254 (1983).
`8) Guntram S., Diabetes Care, 16, 155165
`
`4)
`
`5)
`
`463463
`
`(1993).
`9) Suzuki T., Hikita S., Kizima Y., Okuzumi K.,
`Hurukawa H., Miyaji N., Miyagawa K., Jpn.
`J. Hosp. Pharm., 24, 768773 (1998).
`10) Kastuda E., Shimizu H., Ubukata T., Mori
`M., J. Japan Diab. Soc., 42, 10131015
`(1999).
`11) Sakamoto Y., The Journal of Therapy, 64,
`441447 (1982).
`
`要旨
`注射針を輸液剤等のゴム栓に穿刺するときに,ゴ
`ム栓からゴム片が削り取られるというコアリングが
`報告されている.実際に自己注射を行っている患者
`のインスリンバイアルでもコアリングが発生してい
`るかを確認した.入院患者30名よりインスリンカー
`トリッジを回収し,空打ち液,注入液,カートリッ
`ジ残液を試料とした.発生したゴム片を顕微鏡下で
`観察し形状と個数,大きさを測定した.コアリング
`の発生率は,空打ち液の発生率は73%,注入液は47
`%,カートリッジ残液は97% であった.形状は,
`針を通過している空打ち液と注入液では塊状が多
`く,針を通過していないカートリッジ残液では針状
`が多かった.空打ち液と注入液では小さなゴム片が
`多数確認されたことにより,ゴム片が皮下内に注入
`される可能性が強く示唆された.コアリングの原因
`としては,針を同一個所に回転させて刺すためと考
`えられる.そのため,今後はペン型注射器への針の
`装着について構造上の改良が必要である.コアリン
`グは注射液の異物混入という点で,医学的にも薬学
`的にも非常に重大な問題である.今後はラテックス
`アレルギーやリポディストロフィーなどとの関連も
`検討する必要がある.
`
`MYLAN INST. EXHIBIT 1073 PAGE 5
`
`MYLAN INST. EXHIBIT 1073 PAGE 5
`
`