1 23

Journal of Failure Analysis andPrevention ISSN 1547-7029Volume 13Number 6 J Fail. Anal. and Preven. (2013)13:678-683DOI 10.1007/s11668-013-9749-3

Cervical Stent Failure Analysis

Wayne Reitz

1 23

Your article is protected by copyright and

all rights are held exclusively by ASM

International. This e-offprint is for personal

use only and shall not be self-archived in

electronic repositories. If you wish to self-

archive your article, please use the accepted

manuscript version for posting on your own

website. You may further deposit the accepted

manuscript version in any repository,

provided it is only made publicly available 12

months after official publication or later and

provided acknowledgement is given to the

original source of publication and a link is

inserted to the published article on Springer's

website. The link must be accompanied by

the following text: "The final publication is

available at link.springer.com”.

CASE HISTORY—PEER-REVIEWED

Cervical Stent Failure Analysis

Wayne Reitz

Submitted: 9 July 2013 / in revised form: 11 September 2013 / Published online: 28 September 2013

� ASM International 2013

Abstract Harrington rods failed after a short period in

service. Metallurgical analysis showed (1) notches were

present on the rods, (2) small cracks present in the bent

regions of the rod, and (3) the fractures occurred at

clamped locations. All of these conditions can shorten the

fatigue life by eliminating the crack initiation stage of

fatigue and allowing corrosion fatigue to occur.

Keywords Annealing � Biomaterials � Failure analysis �Titanium

Introduction

The Harrington rod, developed in 1953 by Paul Harrington,

a professor of orthopedic surgery at Baylor College of

Medicine in Houston, Texas, was implanted along the

spinal column to treat lateral curvature of the spine. Har-

rington rods were intended to provide a means to reduce

the curvature and to provide more stability to a spinal

fusion. The device was implanted and secured onto the

vertebral laminae [1].

A Harrington rod cervical stent fractured while in-ser-

vice. The device was implanted in 2005 and retrieved in

2006 and then submitted for metallurgical examination to

determine cause of failure.

Investigation

The investigation included visual inspection at 91, mac-

roscopic inspection, Knoop microhardness, chemical

analysis, scanning electron microscopy and energy dis-

persive spectroscopy (SEM/EDS), and metallography.

Metallographic etch consisted of immersion in a solu-

tion of 10 ml KOH, 5 ml H2O2, and 20 ml H2O for 10 h to

highlight grain size.

Discussion

The components for this cervical stent were shown in

Fig. 1. Indentations were present at the clamp positions;

the Harrington rod fractured adjacent to a retaining clamp.

The fracture occurred on both rods in essentially the same

location based on clamp marks.

Semi-quantitative chemical analysis was performed

using SEM/EDS and the results were listed in Table 1. The

data showed that rod ‘‘B’’ had low aluminum and that iron

was present, which might be a surface contaminant. Rod

‘‘A’’ chemical analysis met the chemistry specification for

ASTM F136.

Microhardness measurements were conducted on the

longitudinal and transverse cross-sections with all samples

exhibiting the same hardness of 30 Rockwell C. The

mechanical properties of the rods and the ASTM specifi-

cation and typical annealed values were presented in

Table 2. All the hardness values were in reasonable

agreement. The low ductility in ASTM F136 was for cold

worked material, while annealed material was typically

30%.

Figures 2 and 3 showed the fractured rods and their

mating surfaces. The general location of the crack initiation

W. Reitz (&)

Talbott Associates, Inc., 7 SE 97th Ave., Portland,

OR 97216, USA

e-mail: wreitz@talbottassociates.com

123

J Fail. Anal. and Preven. (2013) 13:678–683

DOI 10.1007/s11668-013-9749-3

Author's personal copy

sites were indicated by black arrows. Rod ‘‘A’’ exhibited a

large amount of rubbing/sliding as one broken end moved

out of place and slid over the other end that was fixed in

place (see Figs. 4, 5). This phenomenon indicated that the

Rod ‘‘A’’ failed first and then Rod ‘‘B’’ failed due to

overloading.

Metallographic examination showed longitudinal

grooves, or pits, in both rods when examined at a sample

location about 1.5 in. from the fracture, as shown in Fig. 6.

Rod ‘‘A’’ exhibited 37 notches, while Rod ‘‘B’’ exhibited

Fig. 1 Partial assembly of components; note fractures at black

arrows

Fig. 2 Fractured Rod ‘‘B’’; crack initiated at black arrow

Fig. 3 Fractured Rod ‘‘A’’; crack initiated at black arrow; notice

deformation (at white arrows) due to repeated impacts

Fig. 4 Surfaces that were in rubbing contact; the part on the right has

been rotated 90� to show mating rubbing surfaces; the part on the left

was clamped at the point of the fracture (Rod ‘‘A’’)

Table 1 Chemical analysis [2, 3]

Rod ‘‘A’’ Rod ‘‘B’’ ASTM F136

Ti 89.7 90.9 89–91

Al 6.7 4.6 5.5–6.5

V 3.5 3.9 3.5–4.5

Fe 0 1.9 \0.25

Table 2 Mechanical properties [2–4]

Hardness,

Rc

UTS,

ksi

YS,

ksi

%

elongation Comments

Rod A 30 130 ASTM grain

size = 13,

residual CW

Rod B 30 130 ASTM grain

size = 16

ASTM

F136

Specification

(min)

26 125 115 8 Annealed or

cold worked

Typical

Annealed

35 135 125 30 Annealed

J Fail. Anal. and Preven. (2013) 13:678–683 679

123

Author's personal copy

only 18 notches, as listed in Table 3, for the same

approximate surface areas. Circumferential notches were

most critical in shortening fatigued life [5]. These notches

could emanate from rod processing and/or poor surface

finishing or from indentations from the clamping

mechanism [6–8]. One researcher has stated that fatigue

occurs at the clamp. Additionally, different surface finishes

between the rod and the clamp (even when they are of the

same alloy) can cause galvanic corrosion [9].

Figures 7 and 8 showed the overall microstructure. Rod

‘‘B’’ possessed very fine grains of alpha ? beta micro-

structure, typical of this alloy. The equiaxed grains and

hardness indicated the metal was annealed (see Table 2).

Rod ‘‘A’’ possessed slightly larger grains and the same

hardness. There were no noteworthy differences between

the two microstructures.

SEM/EDS results were presented in Figs. 9, 10, 11, 12,

13, 14, 15, 16, and 17. The locale for the crack initiation

site on Rod ‘‘A’’ was shown in Fig. 9. Crevice corrosion

between the rod and clamp could accelerate fatigue failure

via corrosion-fatigue [10–12]. The dark semicircle at the

red arrow suggested long-term exposure to the environment

Fig. 5 Rotated to proper orientation (Rod ‘‘A’’)

Fig. 6 Circumferential groove on rod that act as notches and shorten

fatigue life

Fig. 7 Fine-grained microstructure of Rod ‘‘B’’; grain diameter is

1 lm

Fig. 8 Grain microstructure of Rod ‘‘A’’; grain diameter is 3 lm

Table 3 Summary of rod observations

Rod ‘‘A’’ Rod ‘‘B’’

Surface morphology Rough Smooth

Grain diameter Small (3 lm) Fine (1 lm)

Rubbing of fractured ends Yes No

Number of notches on perimeter

greater than 2.5 lm deep

37 18

Cracks present on surface in

bent regions of rods

Numerous Some

680 J Fail. Anal. and Preven. (2013) 13:678–683

123

Author's personal copy

due to the discoloration, which can be a sign of corrosion.

Additional analysis would be required to definitively

characterize this phenomenon.

The fractured surface morphology for both rods was

shown in Figs. 10 and 11; both samples revealed grain

boundary fractures, akin to grain boundary decohesion.

The general surface morphology in the straight regions

of the rods was shown in Figs. 12 and 13. Work by Sittig

et al. [13] has shown that the roughness of this alloy

increases with pickling time in HNO3–HF. Research by

Hur [14] on cold bending Ti–6Al–4V tubes showed that

limited ductility exists for this material when deformed at

room temperature. The combination of rough surfaces,

probably due to processing, and the potential for cold

bending these rods would shorten the fatigue life by

Fig. 9 SEM image of Rod ‘‘A’’; note discoloration at top; crack

initiated near red arrow

Fig. 10 Fracture surface morphology of Rod ‘‘A’’

Fig. 11 Fractured surface morphology of Rod ‘‘B’’

Fig. 12 Surface morphology of straight section of Rod ‘‘A’’

Fig. 13 Surface morphology of straight section of Rod ‘‘B’’

J Fail. Anal. and Preven. (2013) 13:678–683 681

123

Author's personal copy

eliminating the crack initiation stage of fatigue [15]. A

typical bend location was shown in Fig. 14. The bent

regions of each rod were examined by SEM and were

shown in Figs. 15 and 16, which exhibited rough, cracked,

grain boundary separation of the surface grains for Rods

‘‘A’’ and ‘‘B’’. These surfaces exhibited numerous small

cracks, especially when compared to the metal fixture

holding these rods, which was shown in Fig. 17 and was

typical of how a metal surface appears, slightly scratched,

maybe oxidized, but no cracking. Each shallow crack could

be a crack initiation site for fatigue. Table 3 summarized

these observations.

These observations suggested that Rod ‘‘A’’ failed first

and then Rod ‘‘B’’ failed due to overloading, since it then

carried all of the forces.

Conclusions

1. The high number of notches/grooves and overall sur-

face roughness of Rod ‘‘A’’, perhaps due to aggressive

pickling, increase the probability of fatigue failure.

2. The rough, shallow cracked, surface in the vicinity of the

bends in Rods ‘‘A’’ and ‘‘B’’ act as crack initiation sites,

which shorten the fatigue life of these components.

3. The clamped regions could have experienced corrosion

fatigue.

References

1. http://www.scoliosis.org/resources/medicalupdates/instrumentation

systems.php. Accessed 20 June 2013

Fig. 14 Location of bend on a rod (black arrow)

Fig. 15 Surface cracks on bent section of Rod ‘‘A’’; black arrows

indicate longitudinal direction; red arrow indicates an incipient crack

Fig. 16 Surface of bent section of Rod ‘‘B’’; surface is rough, but no

apparent cracks

Fig. 17 Surface morphology of SEM sample holder

682 J Fail. Anal. and Preven. (2013) 13:678–683

123

Author's personal copy

2. L.A. Shepard et al., Characterization of a failed spinal implant

(Harrington rod), in ASM Conference Proceedings, Metals Park

(1988), pp. 411–418

3. ASTM F136-02a, Standard specification for wrought titanium–6

aluminum–4 vanadium ELI (extra low interstitial) alloy for sur-

gical implant applications (UNS R56401), 2002

4. R. Boyer, E.W. Collings, G. Welsch (eds.), Materials Properties

Handbook: Titanium Alloys (ASM International, Materials Park,

1994), pp. 483–636

5. H.J. Snyder et al., Fatigue fracture of 316L SS screws employed for

surgical implanting, in Handbook of Case Histories in Failure Analysis,

vol. 1, ed. by K.A. Esakul (ASM International, Materials Park, 1992)

6. M. Prikryl et al., Role of corrosion in Harrington and Luque rods

failure. Biomaterials 10, 109–117 (1989)

7. M. Hahn et al., The influence of material and design features on

the mechanical properties of transpedicular spinal fixation

implants. J. Biomed. Mater. Res. 63, 354–362 (2002)

8. H. Sturz et al., Damage analysis of the Harrington Rod fracture after

scoliosis operation. Arch. Orthop. Trauma Surg. 95, 113–122 (1979)

9. J.S. Kirkpatrick et al., Corrosion on spinal implants. J. Spinal

Disord. Tech. 18, 247–251 (2005)

10. A.C. Fraker, Forms of corrosion in implant materials, in Metals

Handbook, vol 13, 9th edn. (ASM International, Materials Park,

1987), pp. 1324–1335

11. L. Aulisa et al., Corrosion of the Harrington’s instrumentation

and biological behavior of the rod–human spine system. Bio-

materials 3, 246–249 (1982)

12. J.B. Brunski et al., Stresses in a Harrington distraction rod: their

origin and relationship to fatigue fractures in vivo. J. Biomech.

Eng. 105, 101–107 (1983)

13. C. Sittig et al., Surface characterization of implant materials c.p.

Ti, Ti–6Al–4V and Ti–6Al–4V with different pretreatments. J.

Mater. Sci. Mater. Med. 10(1), 35–46 (1999)

14. S. Hur, The 360� cold bending of Ti–6Al–4V large diameter

seamless tube. JOM 51(6), 28–30 (1999)

15. R.W. Hertzberg, Deformation and Fracture Mechanics of Engi-

neering Materials (Wiley, New York, 1976)

J Fail. Anal. and Preven. (2013) 13:678–683 683

123

Author's personal copy