CambridgePolymer Group, Inc.

Testing, Consultation, and Instrumentation for Polymeric Materials

Assessing the Cleanliness of Medical Devices

Stephen Spiegelberg

AMI’s Medical Grade Polymers 2010

Background of Cambridge Polymer Group

• Contract research company formed in 1996• Core competencies

– polymer R&D– test methodology development– instrumentation design

• Biomedical community– bone cement– crosslinked polyethylene– spine– eye– vascular– product recalls– cleanliness of medical devices

Sterile is not the same as Clean

• Sterile: live microorganisms content is below acceptable levels– Bacteria, yeast, fungi, molds, viruses– Sterility Assurance Limit (SAL): probability that an implant will

remain nonsterile following sterilization • 10-6 (one in a million)

• Clean: non-live residue content is below acceptable levels– Pyrogens – dead but deadly– Chemicals– Particulate matter

Sterile is not the same as clean

Methods of Sterilization

Detergent Wash

Alcohol Wash

Acid Passivation

Air blasting

High pressure rinses

Sonication

Methods of Cleaning

Glutaraldehyde

Ethylene oxide

Steam/Dry Heat*

Plasma

Ionizing radiation*

*issues with polymeric devices

Case Study I

• In 2000, Sulzer Orthopedics noticed higher than normal revision surgeries on their InterOp Acetabular Shell

• High failure rate in isolated manufacturing group

Explanted hip components showed little tissue ingrowthinto the porous titanium backing, even after 11 months of in vivo use.

Spiegelberg, Deluzio, Muratoglu, Trans Orthopedic Research Society, 2003

InterOp Acetabular ShellCementless fixation: relies on osseointegration in porous titanium structure

What Happened?

Independent Research Team

Pathologists Manufacturing Experts Analytical Labs

Believed to be related to a manufacturing residue

Try to identify type of residue in order to determine best analytical technique

Design sample preparation procedure to extract and quantify residue

Validate extraction and analysis technique

Determine resolution levels

Preliminary Information

• Suspected that a residue was on implants

• Introduction believed to be from machining lubricants

• Received sample lubricants from manufacturer

Protocols

• Extract residue from component

– solvent selection

• Analyze mass of residue with quantified technique

• Identify composition

• Look for trends with manufacturing

Solvent Selection

-1

0

1

2

3

4

5

400900140019002400290034003900

wavenumber [cm-1]

abso

rban

ce [A

.U.]

oilCarbon tetrachlorideChloroform

Select solvent that does not have an IR absorption band in the region of the target materialGood solvent for target material

C Cl

Cl

Cl

Cl

oil

Carbon tetrachloride

HH

Extraction Protocol

Protocol on web site: www.campoly.com

Carbon tetrachloride

Sonicate for 1 hour

Rinse component

Concentrate solution(77 C)

Calibration Curve for Oil

A = 325.27cR2 = 0.9967

0

10

20

30

40

50

60

70

0 0.05 0.1 0.15 0.2 0.25

hydrocarbon concentration [wt.%]

A28

19-2

992

cm-1

Oils A, I, K, F, D

1 mm path lengthAreas baseline-corrected

A = αbc

Validation StudyPorous titanium test samples coated with known amounts of oil

Oil Measurement Validation Study

0

5

10

15

20

25

30

35

40

45

1 2 3 4 5 6 7 8 9 10

sample

oil c

onte

nt [m

g]

Applied Oil

0

5

10

15

20

25

30

35

40

45

1 2 3 4 5 6 7 8 9 10

Meas ured Oil C on ten t

Oil

cont

ent [

mg]

Corrections for hydrocarbons in solvent

Oil in Shelf-Stored Components

• Examined oil content in over 500 acetabular shells

• Grouped component lots by manufacturing history

Manufacturing Procedure

Machine titanium

shell

Apply porouscoating

Group1

Nitric acidpassivation

High temperaturesintering

Detergentwash&rinse

Group2

Peg chamfer

Group3

ID turn

Group4

No passivation

Manufacturing History Analysis

0

10

20

30

40

50

60

70

80

1260000 1280000 1300000 1320000 1340000 1360000 1380000 1400000 1420000 1440000 1460000

Lot Number

Oil

Con

tent

(mg)

Group1Group2Group3Group4

passivation

88% of failures

no passivation

No statistical dependence of oil content on manufacturing history

Oil Removal by Nitric Acid Passivation

0

10

20

30

40

50

60

70

Control Passivation Only

Res

idua

l Oil

Con

tent

[mg]

A SeriesK SeriesAIK Series

1 hour soak in 27 vol.% nitric acid

• Nitric acid passivation does not remove measurable quantities of oil

Histopathology of Tissue from 113 InterOp Shells

• Acute and chronic inflammation in periprosthetic tissue, with an abundance of lymphocytes, granulation tissue, neutrophils, and giant cells. Staining was highly positive for IL-1b and Il-6 activity [1].

• Inflammation was found in the capsule as well, and was not therefore relegated to tissue in direct contact with the device.

• Concluded that a substance in the oil, rather than the oil itself, was responsible for the inflammation [1, 2].

1. Campbell, P.M., J; Catelas, I. Examination of Recalled Inter-Op Acetabular Cups for Cause of Failure. in Society for Biomaterials. 2002. Tampa, FL.2. Campbell, P.M., J; Catelas, I. Histopathology of tissues from Inter-Op acetabular sockets. in 48th Annual Meeting of the Orthopaedic Research Society. 2002.

3 Week Rabbit Study

Tissue response in rabbits injected with Oil I

Tissue response in InterOppatients

Acute Inflammation No Yes

Chronic Inflammation 82.1% Extensive

Eosinophils 96.4% Minimal

Giant Cells 14.3% Abundant

Fibrunous Exudate No Yes

Lipogranuloma 82.1% No

Granulation Tissue No Abundant

Lipid Droplets 46.4% No

Metal No Yes

Other Foreign Body 3.6% Yes

Fibrous Tissue 42.9% Yes

Necrosis 3.6% Yes

Only 2 pathological markers were shared in the two studies

Bloebaum, R.D., E.L. Whitaker, J. Szakacs, and A. Hofmann. The tissue response to an injection of gamma sterilized mineral oil in rabbits. in 49th Annual Meeting of the Orthopaedic Research Society. 2003. New Orleans, LA.

Nitric Acid + Oil

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

10001100120013001400150016001700180019002000

w avenum ber [cm -1]

abso

rban

ce [A

.U.]

residue-top layer (primarily oil)residue-bottom layer (primarily acid)oil A

removed

new

removed

• There is a modest chemical change in the oil with exposure to acid• GC/MS analysis on residues was inconclusive•Cytotoxicity testing on the residues came back negative

Could Endotoxins be the culprit?• Histopathology of endotoxins produced a similar tissue response as

that observed in the Inter-Op tissue [1].

• Nitric acid passivation can reduce the levels of endotoxins adhered to titanium samples [2].

• Endotoxins were found in the sump water of the machine shop

• Trace amounts could be stationed at the oil-tissue interface, enough to prevent osseointegration

a lipopolysaccharide (LPS) produced from Gram-negative bacteria

1. Greenfield, E.M., Y. Bi, A.A. Ragab, V.M. Goldberg, J.L. Nalepka, and J.M. Seabold, Does endotoxin contribute to aseptic loosening of orthopedic implants? J. Biomed. Mater Res, Part B: Appl. Biomater., 2005. 72B: p. 179-185.2. Merritt, K., S.A. Brown, and V.M. Hitchins. Ability of nitric acid or acetone to inactivate bacterial lipopolysaccharide (LPS). in 28th Annual Meeting Transactions of the Society for Biomaterials. 2002

Conclusion of Case I Study

• Oil present on all manufactured lots tested, including those with successful outcomes

– Specific manufacturing history associated with failed implants

• Explanation of clinical response

– not related to absolute level of oil

– appears to be related to nitric acid passivation step

• Most likely culprit was an adherent endotoxin that was delivered via the oil, and was not inactivated by nitric acid passivation

Case Study II: Residue Identification

• What manufacturing residues are introduced?

– Release agents

– Lubricants/coolants

– Polishing compounds

– Cleaning agents

– Particulate debris

– Eluants from packaging

– Chemicals from gloves

• Identification allows the manufacturer to identify and control the source of the residue production

Further Identification of Residues

Analytical techniques

– FTIR indicates quantity of oil, but not identity

• Sum of all parts

– Soluble residue: Gas Chromatography/Mass Spectroscopy

– Insoluble residue: Energy dispersive spectroscopy (EDS)

GC/MS Analysis of Residues

• Catalogued all compounds that may have come into contact with machined components

• Identified which residue composition was present on component

• Identify what manufacturing step introduces residueRoughing Pump Oil approximatechemical composition concentration %octadecane 1nonadecane 11-hentetracontanol 11-chloro-octadecane 1117-pentatriacontene 17(1-butylhecadecyl)-cyclohexane 191-octadecanethiol 6(2-hexyloctyl)-cyclopentane 9N-methyl-N-[4-[4-methoxyl-1-hexahydropridyl]-2-butynyl]-acetamide 71',4 dihydroxy-2,3' dimethyl-, (-)-[1,2'-binaphthalene]-5,5',8,8'-tetrone 251-docosene 3

*GC/FID: approx. 1-10 ppm

Insoluble Residue

• Scanning Electron Microscopy

• Energy Dispersive X-ray Spectroscopy (EDS or EDXA)

– Qualitative/semi-quantitative

– particulate matter adhered to carbon tape, then analyzed in Phillips XL-30 FEG SEM.

EDS Analysis of Extracted Residue

vanadium titanium

10 microns

silica iron sodium

Conclusions to Case II

• Manufacturers have techniques to identify type of residue

• Source of residue in manufacturing process

Case Study III: Residue on Explants

• Can we extract and identify manufacturing residues on explanted devices?

• Possibly identify source of implant failure

Explant Analysis

• Problem: biological tissue are hydrocarbon-based

• Look very similar to hydrocarbon-based manufacturing residues under FTIR

O

O

C C

H

H

C

H

H

H

H

H

O

O

C CH

H

CH

H

HH

O

O

C C

H

H

C

H

H

HCH

C

C

H

10

10

10

triglyceride

-1

-0.5

0

0.5

1

1.5

2

2.5

3

3.5

4

24002500260027002800290030003100320033003400

wavenum ber [cm -1]

abso

rban

ce [A

.U.]

oillipids

H-C Stretch

H

H

C

H

H

HCC

H

H

H 10

mineraloil

Explant Analysis

• Remove lipids from solution

• Protocol similar to shelf-stored analysis

– Sonication with carbon tetrachloride

– Evaporation

– Filtration

• Isolates manufacturing hydrocarbons from lipids and most other biological residues

– infra-red analysis

Explant Analysis

Pre-doped devices extracted in presence of 5 grams of acetabular tissue

0

10

20

30

40

50

60

70

80

90

shell 1 shell 2

oil c

onte

nt [m

g]

Applied amountMeasured amountno filtration

Cadaver Analysis

• Shells were doped with known amounts of oil, then placed in cadavers for 24 hours

• Extraction, then filtration

Cadaver Analysis

Cadaver Specimens

0

5

10

15

20

25

30

35

40

45

az197 az198 az199 az200 az201

oil c

onte

nt [m

g]

0

500

1000

1500

2000

2500

3000

3500

4000

4500

resi

due

mas

s [m

g]

0

5

10

15

20

25

30

35

40

45

az197 az198 az199 az200 az201

oil c

onte

nt [m

g]

0

500

1000

1500

2000

2500

3000

3500

4000

4500

resi

due

mas

s [m

g]

ir-measurementApplied oilresidue content-gravimetric

0

5

10

15

20

25

30

35

40

45

az197 az198 az199 az200 az201

oil c

onte

nt [m

g]

0

500

1000

1500

2000

2500

3000

3500

4000

4500

resi

due

mas

s [m

g]

ir-measurementApplied oilresidue content-gravimetric

Identification of Extracted Residue from Explant

• FTIR provides limited information on structure of hydrocarbons– Data provided from labs that show hydrocarbons on all explants

(3000 cm-1)– Immediately conclude that these are manufacturing residues– Presence of carbonyls indicate these are probably fatty acids (1700

cm-1)– Systemic long-chain hydrocarbons(1,2)

• GC/MS identifies specific composition of hydrocarbons

1. Liber, A.F. and H.G. Rose, Saturated hydrocarbons in follicular lipidosis of the spleen. Arch. Path., 1967. 83: p. 116-122.2. Rose, H.G. and A.F. Liber, Accumulation of saturated hydrocarbons in human spleens. J. Lab. & Clin. Med., 1966. 68(3): p. 475-483.

GC/MS on Explant Residues

Nitrazepam: sleep medication

Fenclofenac: non-steroidal anti-inflammatory drug

TricoseneTranylcypromine: is a non-hydrazine monoamine oxidase inhibitor, an antidepressant and antimanic (brand name Parnate)

Elemental Analysis of Explants

• Manufacturing residue and tissue analysis directly on explant

• Scanning electron microscopy/Energy dispersive spectroscopy

• No damage to explant (non-destructive)

Elemental Analysis of Explants

Case III: Explant Analysis Summary

• Validated techniques to quantify manufacturing hydrocarbon content

• Residue composition can be analyzed with GC/MS

• Surface residue can be analyzed directly on devices

• Tissue composition can be identified (elements)

How Clean is Clean Enough?

• Determined by:– Analytical detection limits– Achievable with commercial cleaning

processes– Cost of cleaning

• Not possible to remove all residue– Levels that do not elicit an adverse

biological response• Application area in the body• Size of device

ASTM Activities• ASTM activities on cleanliness issues of

medical devices (F04.15.17)

– Passed first standard on assessing cleanliness (ASTM F2459)

– WK15532: Guide for Assessment of Contamination and Residues on Medical devices

• Compilation of known assays for residues, including endotoxins

– WK13292: Standard Practice/Guide for Shipping Possibly Infectious Materials, Tissues, and Fluids

Conclusions

• Acceptable levels?

Good historical success of medical device = acceptable levels of residues

Acknowledgements

• Dr. Gavin Braithwaite• Dr. Orhun Muratoglu• Denise Saltojanes• Partial funding from Sulzer Orthopedics

Visit our web site www.campoly.com for downloadable application notes and presentations