INTRAVITREAL SILICONE OIL DROPLETS
`AFTER INTRAVITREAL DRUG
`INJECTIONS
`
`SOPHIE J. BAKRI, MD, NOHA S. EKDAWI, MD
`
`Purpose: To present the finding of tiny silicone oil droplets in 15 eyes of 15 patients after
`intravitreal injections of an anti–vascular endothelial growth factor agent or triamcinolone
`acetonide and to discuss the likely source of silicone oil.
`Methods:
`In an observational case series, charts of patients who had undergone
`intravitreal injections by one surgeon were reviewed retrospectively. The finding of intra-
`vitreal silicone oil droplets was noted. The following information was also documented:
`number and type of injections before the appearance of silicone oil droplets, symptoms
`and evidence of ocular inflammation, visual acuity before and after silicone oil droplets,
`length of follow-up, and visual acuity at the last examination.
`Results: Fifteen eyes of 15 patients were found to have silicone oil droplets docu-
`mented after a various number of injections (range, 1–16). Patients were asymptomatic,
`and there were no adverse side effects associated with the presence of silicone oil droplets
`at examination.
`Conclusions: Silicone oil droplets may occur in the vitreous cavity after intravitreal drug
`injections. There were no adverse effects found associated with silicone oil in the vitreous
`after injections of anti–vascular endothelial growth factor agents or triamcinolone ace-
`tonide. The likely source of silicone oil is the needles and syringes used for the injections.
`RETINA 28:996 –1001, 2008
`
`The use of intravitreal injections for the treatment
`
`of retinal disorders has become a much more
`common procedure, in light of the development of
`drugs specifically designed to block vascular endothe-
`lial growth factor and the efficacy of intravitreal ste-
`roids. Intravitreal drugs that are US Food & Drug
`Administration–approved for the treatment of choroi-
`dal neovascularization associated with age-related
`macular degeneration include pegaptanib (Macugen;
`OSI/Eyetech, New York, NY)1 and ranibizumab (Lu-
`centis; Genentech, Inc., South San Francisco, CA).2,3
`In addition, intravitreal bevacizumab (Avastin; Ge-
`
`From the Department of Ophthalmology, Vitreoretinal Diseases
`and Surgery, Mayo Clinic, Rochester, Minnesota.
`Supported by Research to Prevent Blindness, Inc. (New York, NY).
`Reprint requests: Sophie J. Bakri, MD, Department of Ophthalmol-
`ogy, Vitreoretinal Diseases and Surgery, Mayo Clinic, 200 First Street
`SW, Rochester, MN 55905; e-mail: sbakri@hotmail.com
`
`nentech, Inc.) is used off label for the treatment of
`choroidal neovascularization, diabetic retinopathy, reti-
`nal venous occlusive disease, and other vascular en-
`dothelial growth factor–mediated diseases.4 Intravit-
`real triamcinolone has been shown to be effective in
`the treatment of macular edema.5– 8 Known complica-
`tions of intravitreal injections include sterile and in-
`fectious endophthalmitis, retinal detachment, intraoc-
`ular inflammation, vitreous hemorrhage, retinal tears,
`and damage to the crystalline lens. Most of these com-
`plications are procedure related. However, the cumu-
`lative damage from repeated injections has not been
`addressed. We present a series of 15 eyes of 15 patients
`with intravitreal silicone droplets as a consequence of
`intravitreal injections with an anti–vascular endothe-
`lial growth factor agent or triamcinolone and discuss
`the likely source of silicone oil.
`
`996
`
`Novartis Exhibit 2007.001
`Regeneron v. Novartis, IPR2020-01317
`
`

`

`INTRAVITREAL SILICONE OIL USAGE AFTER INTRAVITREAL DRUG INJECTIONS ● BAKRI AND EKDAWI
`
`997
`
`Methods
`
`After receiving approval from the Mayo Clinic Insti-
`tutional Review Board (Rochester, MN), a retrospective
`review of charts of patients who had received intravitreal
`injections of pegaptanib, bevacizumab, triamcinolone, or
`ranibizumab by one surgeon (S.J.B.) between July 2005
`and November 2007 was performed. The charts were
`reviewed for the presence of intravitreal silicone oil
`droplets. The number and type of injections before the
`appearance of silicone oil droplets were documented.
`In addition, symptoms and evidence of ocular inflam-
`mation were noted. Visual acuity was also docu-
`mented before and after silicone oil droplets were
`noted during examination; length of follow-up and
`final visual acuity were measured as well.
`
`Results
`
`A total of 1,529 intravitreal injections between July
`2005 and November 2007 were reviewed. Fifteen eyes
`of 15 patients had documented intravitreal silicone oil
`droplets noted during examination. All 15 eyes had
`tiny silicone oil droplets in the vitreous that were
`visible by slit-lamp or funduscopic examination (Figs.
`1–3). At examination in various head positions, the
`silicone oil droplets were found to float superiorly. All
`patients were asymptomatic. There was no ocular in-
`flammation, as evidenced by anterior chamber or vit-
`reous cell throughout the follow-up period. The num-
`ber of injections before silicone oil droplets were
`noted varied from 1 to 16 (Table 1) of pegaptanib,
`bevacizumab, triamcinolone, or ranibizumab. Patients
`were being treated for choroidal neovascularization
`and cystoid macular edema due to branch or central
`retinal vein occlusions. As expected from the course
`of disease treatment, visual acuity fluctuated through-
`out treatment, depending on the activity of choroidal
`
`Fig. 2. Patient 5. A tiny silicone oil droplet noted in the vitreous cavity
`during slit-lamp examination. The large yellow area is a light reflection.
`The red reflex is shown.
`
`neovascularization or macular edema. Visual acuity
`was not affected by the presence of silicone oil drop-
`lets but was consistent with the course of the disease
`(Table 1). The presence of silicone oil was confirmed
`with ultrasonography (Fig. 4 [Patient 1]). The high
`echogenicity matched that of silicone oil. No interven-
`tion was necessary in our group of patients, and the
`patients were observed without complications for the
`follow-up period (range, 0 –13.9 months; median, 7.0
`months).
`
`Discussion
`
`Intravitreal silicone oil droplets may occur after
`intravitreal injection. In our case series, there was no
`association with a specific agent. As shown in Table 1,
`10 of 15 patients had received bevacizumab injections
`before silicone oil droplets were noted during exami-
`nation. The other five patients had received injections
`
`Fig. 1. A tiny silicone oil droplet noted just anterior to the optic nerve.
`
`Fig. 3. Patient 5. A tiny silicone oil droplet noted in the vitreous cavity.
`
`Novartis Exhibit 2007.002
`Regeneron v. Novartis, IPR2020-01317
`
`

`

`998
`
`RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● 2008 ● VOLUME 28 ● NUMBER 7
`
`2.4
`
`0.0
`0.0
`
`0.0
`0.0
`
`7.4
`
`13.9
`5.6
`
`5.1
`3.3
`13.7
`
`5.4
`
`12.4
`
`5.3
`5.6
`
`WasNoted
`SiliconeOil
`
`Follow-upAfter
`Duration(mo)of
`
`20/70
`
`20/25
`20/40
`
`20/40
`20/50
`
`20/100
`
`20/60
`20/30
`
`20/80
`20/30
`20/40
`
`20/25
`
`20/40
`
`20/40
`20/25
`
`20/70
`
`20/25
`20/40
`
`20/40
`20/50
`
`20/150
`
`20/40
`20/30
`
`20/30
`20/20
`20/40
`
`20/30
`
`20/40
`
`20/30
`20/20
`
`Follow-up
`atFinal
`
`VisualAcuity
`
`WasNoted
`SiliconeOil
`
`After
`
`20/60
`
`20/30
`20/50
`
`20/50
`20/40
`
`20/100
`
`20/30
`20/50
`
`20/40
`20/20
`20/40
`
`20/25
`
`20/100
`
`20/25
`20/20
`
`WasNoted
`SiliconeOil
`
`Before
`
`VisualAcuity
`
`VisualAcuity
`
`pqatriamcinolone⫻5
`
`triamcinolone
`
`Bevacizumab⫻2;
`ranibizumab⫻3
`
`bevacizumab⫻1
`dexamethasone⫻1;
`ranibizumab⫻8;
`
`Pegaptanib⫻12;
`Ranibizumab⫻6
`
`Pegaptanib⫻6;
`Bevacizumab⫻11
`bevacizumab⫻3
`ranibizumab⫻5;
`Bevacizumab⫻2;
`
`bevacizumab⫻4
`
`ranibizumab⫻10
`bevacizumab⫻7;
`
`Pegaptanib⫻1;
`Bevacizumab⫻5
`
`Pegaptanib⫻3;
`Pegaptanib⫻3
`Bevacizumab⫻7
`ranibizumab⫻9
`
`dexamethasone
`ranibizumab,
`
`bevacizumab
`
`ranibizumab
`bevacizumab,
`
`ranibizumab
`
`triamcinolone⫻1
`Bevacizumab⫻10;
`
`triamcinolone
`bevacizumaband
`ranibizumab⫻9;
`Bevacizumab⫻2;
`Bevacizumab⫻1
`
`Pegaptanib⫻3;
`
`Follow-up
`
`UsedbyFinal
`
`IntravitrealAgent(s)
`
`SiliconeOilNoted
`
`UsedBefore
`
`CME,cystoidmacularedema;CRVO,centralretinalveinocclusion;CNV,choroidalneovascularmembrane;AMD,age-relatedmaculardegeneration;BRVO,branchretinalvein
`
`occlusion.
`
`Bevacizumab,
`ranibizumab
`
`Pegaptanib,
`Ranibizumab
`
`Pegaptanib,
`Bevacizumab
`
`ranibizumab
`Bevacizumab,
`
`Pegaptanib,
`Bevacizumab
`
`Pegaptanib,
`Pegaptanib
`Bevacizumab
`
`Pegaptanib,
`
`Bevacizumab
`
`ranibizumab
`Bevacizumab,
`Bevacizumab
`
`7
`
`15
`
`6
`
`16
`11
`
`10
`
`5
`
`4
`
`20
`
`7
`
`3
`
`12
`
`8
`
`12
`
`1
`
`6
`
`15
`
`5
`
`15
`10
`
`3
`
`3
`
`4
`
`16
`
`1
`
`8
`
`3
`
`7
`
`2
`
`3
`
`retinopathy
`
`CMEduetoradiation
`
`CNVduetoAMD
`CNVduetoAMD
`
`CNVduetoAMD
`CMEduetoBRVO
`
`CNVduetoAMD
`
`CNVduetoAMD
`CMEduetoBRVO
`
`CNVduetoAMD
`CNVduetoAMD
`CMEduetoCRVO
`
`CNVduetoAMD
`
`retinopathy
`
`CMEduetoradiation
`
`CNVduetoAMD
`CMEduetohemi-CRVO
`
`15
`
`14
`13
`
`12
`11
`
`10
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`Table1.DescriptionofPatientsWithIntravitrealSiliconeOilBubblesSeenDuringExamination
`
`IntravitrealAgent(s)
`
`Follow-up
`Throughout
`Injections
`TotalNo.of
`
`OilNoted
`
`BeforeSilicone
`No.ofInjections
`
`Disease
`
`Patient
`
`Novartis Exhibit 2007.003
`Regeneron v. Novartis, IPR2020-01317
`
`

`

`INTRAVITREAL SILICONE OIL USAGE AFTER INTRAVITREAL DRUG INJECTIONS ● BAKRI AND EKDAWI
`
`999
`
`investigators questioned whether this is due to the
`silicone oil lubricant on the syringes or needles used
`for intravitreal injections. Pegaptanib is packaged in a
`sterile, 1-mL, USP type 1, graduated, prefilled glass
`syringe fitted with a sterile 27-gauge needle. All nee-
`dles used in our series were of varying gauges, man-
`ufactured by Becton Dickinson and Company (Frank-
`lin Lakes, NJ). Triamcinolone is drawn up with a
`20-gauge needle and injected with a 27-gauge needle.
`Ranibizumab is drawn up with a 19-gauge filter needle
`(5-␮m filter) and injected intravitreally using a 30-
`gauge 0.5-in needle (PrecisionGlide, Becton Dickin-
`son and Company, Franklin Lakes, NJ). Bevacizumab
`is fractionated by our hospital pharmacy from a large
`vial into 0.15-mL doses supplied in a tuberculin sy-
`ringe (Becton Dickinson and Company). Either a 30-
`gauge 0.5-in needle is placed on this or it is transferred
`into a 31-gauge needle attached to a syringe (Ultrafine II,
`Becton Dickinson and Company, Franklin Lakes, NJ).
`During the manufacturing process, these needles were
`extruded with silicone oil as well as lubricated. The
`tuberculin syringes were also lubricated with silicone oil.
`The lubricant used was most likely 365 Dow Corning
`surfactant according to a discussion with representatives
`from Becton Dickinson and Company. Material sup-
`plied by the Dow Corning Corporation states that the
`components of the surfactant and preservative pack-
`age used in the emulsion are present at levels of
`fractions of a percent in the final product. They esti-
`mated that the product had 2% to 3% of silicone oil
`before application of the lubricant in the syringe. They
`report no adverse effects due to the extremely low
`levels of silicone oil. They further highlight that these
`additives are used in other industries that have human
`contact including food applications. They, however,
`recommend that users determine if these additives
`create a problem in their specific application.11
`There have been many previous reports, especially
`in the diabetes literature, regarding the contamination
`of skin and insulin with silicone oil found on needles
`and syringes.12 A granulomatous reaction was found
`in one series of three cases.13 Miller et al14 reported on
`the amount of silicone oil extruded from syringes after
`flushing with distilled water. They found that more
`silicone oil was flushed from 10-mL syringes when
`exhibiting accumulation of lubricant on the tips of
`plungers (10.7–11.3 mg) than when none was seen (0
`mg). Most notably, the plunger tip was thought to
`have 24% of the silicone lubricant. This report also
`confirmed that washing the same syringe multiple
`times continued to provide an equal amount of sili-
`cone oil in each washing. Speculating that pumping
`the syringe back and forth caused more silicone oil to
`be added into the washings, Chantelau et al15 in 1986
`
`Fig. 4. Patient 1. Silicone oil droplets shown by A-scan and B-scan
`ultrasonography. A, B-Scan showing three droplets of silicone oil in the
`vitreous cavity. B, A-Scan showing the highly echogenic pattern.
`
`with pegaptanib, ranibizumab, or a combination. Be-
`cause none of the patients had received triamcinolone
`alone, we cannot implicate this drug to be associated
`with this finding. The agents used did not have sili-
`cone oil in their drug vehicle. There were no adverse
`effects on vision, and there was no ocular inflamma-
`tion during the follow-up period. Intravitreal silicone
`oil is used commonly in retinal detachment surgery,
`and its ocular adverse effects include migration of
`silicone oil into the anterior chamber, emulsification
`of silicone oil, and keratopathy. Other known side
`effects are more applicable in the presence of a near-
`complete silicone oil fill in the vitreous and include
`pupillary block, angle closure without pupillary block,
`and idiopathic open-angle glaucoma even with the
`absence of silicone oil in the anterior chamber.9 None
`of these side effects were seen in our series.
`Freund et al10 described a series of three eyes of
`three patients where silicone oil was noted after intra-
`vitreal injections. Ultrasonography showed a highly
`echogenic pattern, characteristic of silicone oil. These
`
`Novartis Exhibit 2007.004
`Regeneron v. Novartis, IPR2020-01317
`
`

`

`1000 RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES ● 2008 ● VOLUME 28 ● NUMBER 7
`
`used syringes to draw insulin into a syringe and then
`pushed the plunger in to expel air droplets three times.
`They noted that 0.15 mg to 0.25 mg of silicone oil had
`been expelled from each syringe. Another report by
`Baldwin16 speculated on how silicone oil became in-
`corporated in insulin during vigorous pumping back
`and forth on the syringe. In the diabetes literature, there
`are also multiple instances when insulin contaminated
`with silicone oil was found to have less efficacy.16,17
`Typically, to draw up a medication (e.g., ranibi-
`zumab or triamcinolone), a larger-bore needle is
`placed on the syringe, and the plunger is pulled back
`to withdraw more medication than is needed. Then the
`smaller-bore (e.g., 30-gauge) needle used for the in-
`travitreal injection is placed on the syringe, and the
`plunger is pushed up to the amount needed, which also
`eliminates air droplets drawn up from the vial. If
`residual air droplets are seen, the syringe is tapped to
`force the droplets up toward the hub of the syringe.
`Compounded agents may have been subjected to be-
`ing placed in more syringes, needles, and tubing, with
`more friction occurring between the agent and the
`syringe and needle.
`On the basis of compelling evidence in the diabetes
`literature, the presence of silicone oil as a lubricant in
`our needle and syringes is most likely the source seen
`in our patients. Although the presence of silicone oil
`has not caused any adverse effects, there is a need to
`review the methods of injections and utilization of
`syringes to decrease the amount of silicone oil that is
`inadvertently injected. It has been previously reported
`that silicone oil when left in the eye for long periods
`(⬎12–18 months) can be engulfed by the trabecular
`meshwork and Mu¨ller cells of the retina.18 Sugges-
`tions for reducing silicone oil contamination include
`pulling back the plunger as minimally as necessary to
`withdraw the drug to reduce the risk of unnecessary
`contamination by additional material. In addition, re-
`ducing the amount of manipulation and tubing that the
`drug is subjected to during repackaging or compound-
`ing may be important. It is also important to perform
`drug stability testing in new packaging, especially for
`prefilled syringes. The medication may interact with
`the silicone oil layer, and this may affect drug degra-
`dation. As previously discussed, silicone oil may de-
`tach from the syringe or plunger and form droplets in
`the medication. The use of syringes and needles with
`non–silicone-based lubricants is another option worth
`considering to eradicate this possible adverse effect.
`Because this is a retrospective review of charts of
`patients seen in a busy clinic, it is possible that more
`patients had silicone oil droplets, which are difficult to
`see, than reported. The silicone oil droplets had been
`seen and noted incidentally when the patients were
`
`examined with a 90-diopter or 78-diopter lens on a
`slit-lamp biomicroscope. The droplets were not spe-
`cifically looked for. Therefore, we cannot adequately
`address the true prevalence of silicone oil droplets
`after intravitreal injections. These tiny droplets are diffi-
`cult to see, and the amount of silicone oil is difficult to
`quantify; some droplets may be hidden from view
`anterior to the ora serrata. It is difficult to assess, on
`the basis of this case series, whether certain types of
`syringes are more likely to produce this adverse effect.
`Although we do have ultrasonographic evidence that
`this finding is consistent with silicone oil, we do not
`have any pathologic evidence, because no vitreous
`biopsies were done in any of our cases. However, we
`believe that this not necessary because there has been
`no affect on visual function to substantiate the risk of
`biopsy for these patients.
`Key words: anti–vascular endothelial growth fac-
`tor, bevacizumab (Avastin), droplet, intravitreal, in-
`jection, needle, pegaptanib, ranibizumab (Lucentis),
`silicone oil, syringe, triamcinolone.
`
`References
`
`1. VEGF Inhibition Study in Ocular Neovascularization (VISION)
`Clinical Trial Group. Year 2 efficacy results of 2 random-
`ized controlled clinical trials of pegaptanib for neovascular
`age-related macular degeneration. Ophthalmology 2006;
`113:1508.
`2. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for
`neovascular age-related macular degeneration. N Engl J Med
`2006;355:1419–1431.
`3. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab
`versus verteporfin for neovascular age-related macular de-
`generation. N Engl J Med 2006;355:1432–1444.
`4. Michels S, Prager F, Bakri SJ, Wachtlin J. Bevacizumab for
`ophthalmic diseases. Expert Review of Ophthalmology 2007;
`2:369–378.
`5. Ekdawi NS, Bakri SJ. Intravitreal triamcinolone and bevaci-
`zumab combination therapy for macular edema due to central
`retinal vein occlusion refractory to either treatment alone.
`Eye 2007;21:1128–1130.
`6. Mohamed Q, McIntosh RL, Saw SM, et al. Interventions
`for central retinal vein occlusion: an evidence-based sys-
`tematic review. Ophthalmology 2007;114:507–519.
`7. Goff MJ, Jumper JM, Yang SS, et al. Intravitreal triamcino-
`lone acetonide treatment of macular edema associated with
`central retinal vein occlusion. Retina 2006;26:896–901.
`8. Gregori NZ, Rosenfeld PJ, Puliafito CA, et al. One-year
`safety and efficacy of intravitreal triamcinolone acetonide for
`the management of macular edema secondary to central ret-
`inal vein occlusion. Retina 2006;26:889–895.
`9. Costarides AP, Alabata P, Bergstrom C. Elevated intraocular
`pressure following vitreoretinal surgery. Ophthalmol Clin
`North Am 2004;17:507–512.
`10. Freund KB, Laud K, Eandi CM, Spaide RF. Silicone oil droplets
`following intravitreal injection. Retina 2006;26:701–703.
`
`Novartis Exhibit 2007.005
`Regeneron v. Novartis, IPR2020-01317
`
`

`

`INTRAVITREAL SILICONE OIL USAGE AFTER INTRAVITREAL DRUG INJECTIONS ● BAKRI AND EKDAWI
`
`1001
`
`11. Corporation DC. Corning威 365, 35% Dimethicone NF Emul-
`sion. Midland, Michigan: Medical Products Division; 2002.
`12. Kline D, Kuhn T. Needle reuse and tip damage. Diabetes
`Care 2004;27:617.
`13. Travis WD, Balogh K, Abraham JL. Silicone granulomas:
`report of three cases and review of the literature. Hum Pathol
`1985;16:19–27.
`14. Miller JR, Helprin JJ, Finlayson JS. Silicone lubricant flushed
`from disposable syringes: determination by atomic absorp-
`tion spectrophotometry. J Pharm Sci 1969;58:455–456.
`
`15. Chantelau E, Berger M, Bohlken B. Silicone oil released from
`disposable insulin syringes. Diabetes Care 1986;9:672–673.
`16. Baldwin RN. Contamination of insulin by silicone oil: a
`potential hazard of plastic insulin syringes. Diabet Med 1988;
`5:789–790.
`17. Bernstein RK. Clouding and deactivation of clear (regular)
`human insulin: association with silicone oil from disposable
`syringes? Diabetes Care 1987;10:786–787.
`18. Suzuki M, Okada T, Takeuchi S, et al. Effect of silicone oil
`on ocular tissues. Jpn J Ophthalmol 1991;35:282–291.
`
`Novartis Exhibit 2007.006
`Regeneron v. Novartis, IPR2020-01317
`
`