Fatigue testing of three peristernal median sternotomy closure techniques
© Wangsgard et al; licensee BioMed Central Ltd. 2008
Received: 19 August 2008
Accepted: 24 September 2008
Published: 24 September 2008
Failure of a sternotomy closure because of closure system fatigue is a complication that may result in dehiscence and put the individual at risk for serious complications. The purpose of this study was to assess the fatigue performance of three peristernal median sternotomy closure techniques (figure-of-eight stainless-steel wires, figure-of-eight stainless-steel cables, or Pectofix Dynamic Sternal Fixation [DSF] stainless-steel plates) in order to quantify the potential risk of fatigue failure of these devices when subject to cyclic loads in physiologically relevant loading directions.
All tests were conducted on polyurethane foam sternal models. A cardiothoracic surgeon divided each sternal model longitudinally and repaired it with a closure device. Tests were performed using a materials testing system that applied cyclic loading in a uniaxial direction until the test model catastrophically broke or data run-out occurred. For each loading direction (lateral distraction and longitudinal shear), five trials of each closure technique were tested. Life data and location of device failure (if present) were evaluated. Statistical analysis was performed using regression with life data allowed for correlation between life data and the various closure techniques to develop risk assessment curves for each device.
The data show that the figure-of-eight stainless-steel cable and the DSF plate systems are considerably less likely to fail under both lateral distraction and longitudinal shear cyclic loading conditions as compared to the figure-of-eight stainless-steel wire system. Moreover, the figure-of-eight stainless-steel cable system is the most resistant to failure, particularly for high cycle counts.
This study in addition to Cohen and Griffin's earlier published biomechanical comparison of the ultimate strength of these same three closure techniques provide extensive experimental evidence regarding the mechanical differences among these three peristernal median sternotomy closure techniques. All data support the hypothesis that both the DSF plate system and the stainless-steel cable system offer important advantages over figure-of-eight wire closure techniques; although twisted wires are the weak-link in the systems we tested.
Any physical disruption or infection in the sternal region following median sternotomy closure results in compromised wound healing and the risk of additional complications. Although these issues are reported in only 0.3–5% of cases, the associated risks are severe, resulting in mortality between 14 and 47% of the time [1, 2]. Moreover, studies have indicated that physical forces generated by coughing or patient movement can cause dehiscence, the separation of the sternal halves due to bone cutting, or closure device failure [3–5]. This separation can lead to life-threatening complications and is the most common complication of sternotomies, reported in 0.30–0.90% of patients [6–8].
In materials science, mechanical fatigue refers to a mode of material failure that occurs in structures subjected to fluctuating and recurring stresses, occurring even when stress levels are considerably less than the ultimate strength of the material under a static load . Observing that forces generated in the thoracic cavity such as coughing can occur frequently over time, a successful sternal closure device must be able to survive cyclic loading from repetitive physiological forces. Additionally, current literature shows that complications associated with sternotomies often occur within the first two weeks after surgery, establishing the possibility that fatigue is a prevalent cause of closure device failure and consequent sternotomy complications . Thus, fatigue testing may be critical to understanding the wear and cause of complications in various sternal closure devices.
There have been several studies of fatigue loading sternotomy closure devices in cadaveric sterna, in which the closure fails in less than 500 cycles due to the cutting of the wire into the sterna, which is certainly a clinically important common failure mode . However, there appear to be no biomechanical studies that reliably demonstrate failure of the closure device by fatigue, which is also a clinically relevant failure mode . Failure of a closure system is seen clinically in the common finding of broken sternal wires following sternal closure.
Due to the variability in bone density, sternal size, and thickness between both animal and human cadaver specimens, a foam model sternum was used order to minimize confounding variables, minimize subject variability and focus solely on the mechanical performance of the respective sternal closure devices. The polyurethane foam models used in this study simulate the mechanical properties observed in human cadaveric sterna and minimize variability among sternal specimens . Moreover, unlike bone, polyurethane foam mechanical properties, such as stiffness and strength, has been shown to be unaffected by how fast the loads are applied . Therefore, long-term fatigue testing with these models can occur at higher than physiologic rates – shortening test duration – and still reliably simulate fatigue performance in-vivo.
Methods and materials
The sternal models used in this study were fabricated using 20-lb/ft3 rigid polyurethane foam (1025-2, Pacific Research Laboratories Inc, Vashon, Washington) and simulate the manubrium, sternal body, and xiphoid process of the sternum including approximately 2.5 cm of the surrounding ribs extending from the sternum. The sternal models were sawed in half longitudinally and were re-approximated by a board-certified cardiothoracic surgeon familiar with all three closure devices. The sterna were closed with either three peristernal figure-of-eight No. 5 stainless-steel wires (Ethicon Inc, Somerville, NJ), three figure-of-eight stainless-steel cables (Pioneer Surgical Technology, Marquette MI), or three pairs of stainless-steel plates (DSF Dynamic Sternal Fixation System, Pectofix Inc, South Plainfield, NJ).
Relevant geometry parameters published by Losanoff et al were used in the calculations using equation 1 . The average radial distance of the chest (r) in our calculations was 0.15 m, the average length of the sternum (l) was 0.20 m, and the pressure applied to the chest walls (P) was 175 Pa. Locke et al examined respiratory pressures and associated rib cage mechanics in patients that had recently undergone a median sternotomy procedure . In that study, maximum inspiratory pressures were reported: 61 mmHg one week after surgery; 59 mmHg three months following surgery; 74 mmHg for subjects with no history of a sternotomy. Losanoff et al reported a range of pressures within the lungs of 100–300 mmHg during coughing episodes . Using Equation 1, resultant lateral forces for this range of thoracic pressures (59–300 mmHg) as reported by Locke and Losanoff were found to be 236–1200 N. This range is approximately 0.2 – 1.0 times the reported yield loads for these three sternotomy closure techniques in the lateral distraction loading mode in our previous study using this model .
Longitudinal shear simulates lateral flexion stretching, lying on one's side, or supporting the body on one vertically oriented arm (Figure 2). McGregor and coworkers reported that forces up to 325 N in longitudinal shear are physiologic . Because the events that cause longitudinal shear occur in nearly equal frequencies between the left and right sides of the body, the models were subject to a sinusoidal force of ± 300 N resulting in the complete cycling of the sternal halves between one another. The peak value of 300 N is approximately 20 percent of the average ultimate load in the longitudinal shear direction using Cohen and Griffin's ultimate strength values for these same three sternal closure devices . A cyclic wave period of 1 Hz was used for this loading mode in order to allow for complete displacement of the sternal halves, approximately 1–2 mm, and maintain proper load control. Fifteen models (5 of each closure technique) were tested in longitudinal shear.
After each test was completed, the number of cycles was recorded. For analysis purposes, if the sternal closure device did not fail by fatigue during testing, the number of cycles tested with the model was recorded and annotated as run-out data, which means that failure did not occur during testing, but presumably would at some time in the future if the test had continued.
Regression with life data was performed (Minitab 15.2, Minitab Inc, State College, PA) for both the lateral distraction test and the longitudinal shear test. This regression was performed for various probability distributions commonly utilized in accelerated testing: smallest extreme value, Weibull, exponential, normal, lognormal, logistic, and loglogistic. Run-out data were defined as right-censored and included in this analysis. The probability plot for standardized residuals in addition to an adjusted Anderson-Darling statistic was calculated for each distribution and the best fitting distribution was selected. Finally, using best fitting distribution, a table of survival probabilities was generated for the various models tested, from which relative risk curves were generated. The relationship of how closure technique affects fatigue life was also assessed using a post-hoc statistical analysis of the mean fatigue life (50% probability of survival) using Tukey's HSD. Statistically significant differences were reported when p < 0.05.
Post-hoc statistical analysis of the means (50% survival) for the fatigue tests.
The purpose of this study was to investigate the mechanical fatigue performance of three peristernal median sternotomy closure techniques with a view toward evaluating the relative risks of fatigue failure. Our hypothesis was that a sternotomy closure system that uses either stainless-steel cables or stainless-steel plates to distribute local stresses of the wires on the sternum will be less prone to fatigue-based failure in-vivo when compared to stainless steel wires.
The number of cycles for distraction used in this study is based upon the idea that 95% of bone healing occurs at 6 weeks (42 days). An estimate of in-vivo loading for lateral distraction at a rate of 3 coughs/min would be 181,440 cycles over 6 weeks. While these values may initially seem to over estimate the number of coughing events, it should be considered that a cough is rarely an isolated event. Furthermore, it is expected that chest pressures for some coughing events will be greater than those used to estimate our peak force. Since the relationship between fatigue life and stress is highly non-linear, high stresses produced by high forces will substantially lower the fatigue life of materials. Therefore, we believe that our values used provide a reasonable approximation for the longitudinal distraction loading.
Under distraction loading, only 43% of figure-of-eight stainless-steel wire closure devices are predicted to survive for 50,000 cycles and by 100,000 cycles, nearly all these constructs will have failed (probability of survival = 8.3%). For longitudinal shear loads, only 69.4% of figure-of-eight stainless-steel wire closure devices are predicted to survive for 500 cycles. By 10,000 cycles, virtually all these devices fail (probability of survival < 0.001%).
Dynamic sternal fixation plates
The present results demonstrate that the cables are superior to the DSF and then wires. The conclusions of this study are different from a previous biomechanical study on the same three constructs where it was reported that the DSF plates were favored over the cables. In that study, however, other parameters, such as stiffness, were considered which are thought better promote better healing by stabilizing the sternotomy closure; whereas the present study examines the durability under repetitive loads . Adding additional wires to a figure of eight closure will improve the fatigue life, yet the sternal loading is complex and demanding, thus the materials will still fail with time. Therefore, the differences between studies are important since the both rigidity and durability are necessary for proper healing.
Most fatigue studies only focus on the notion of wires cutting into the sternum which leads to failure. It is extremely important to realize that while the repeated loads may cause the system to damage the sternum, the closure system itself is made of materials that will fail under repetitive loading. The value of this study is that we have shown clearly that the durability must also be considered when selecting a closure system.
The present study is biomechanical in nature, and each loading mode is considered separately. In an in-vivo situation, all loading occurs simultaneously. This suggests that the fatigue life in-vivo would occur in fewer cycles than predicted for either individual loading since the residual strength of the system is affected by each load cycle. That being said, the results of this study clearly demonstrate consistency among the system ratings between loading method and therefore, it is not expected that the ratings would change.
Furthermore, the testing we report on was done in air, without the additional environmental concerns, such as corrosive (saline) bodily fluids, that are known to shorten fatigue life [17, 18]. Stainless steel alloys are known to be particularly susceptible to such environmental attacks. Therefore, while these fatigue lives may seem very long, it is quite possible that wires, cables, or other hardware associated with sternotomy closure may have the fatigue life shortened by at least a factor of 10, which then accounts for why there are reports of broken wires in the literature as well as the wires cutting into the sternum, leading to dehiscence [2, 6, 17–20]. Additionally, even if the closure method does not fail by fatigue, the debris generated by fatigue process could lead to metallosis or other foreign body reactions that can produce complications [21–23].
One of the most significant aspects of this study is that it provides a reliable means to evaluate the potential durability of a sternal closure system under fatigue. This information, when coupled with cadaveric studies, can provide valuable insight into how the system may affect the clinical outcomes. Obtaining long-term closure system durability information using cadavers would be more challenging since the test would span several days, and preserving the tissue from decomposition over the test duration would be difficult.
1. The complicated combined loading induced by breathing, coughing, and body movement of a patient can all lead to failure of sternal closure devices by fatigue [3–5]. While this study examined each loading mode independently, the results should not differ for combined loading, except that the fatigue life is expected to be much shorter than the longitudinal shear tests due to the complex interaction of loading.
2. Adding additional wires or cables may improve the length of the fatigue life, but the weak link in most closure systems is the stainless steel wire. Therefore, any system that uses wires is still more likely to fail than a system that does not use them. Also, the use of more wires will not have a significant impact on the stress concentration induced on the sternum – which is a factor we did not evaluate. Considering this, any benefit that is derived by the addition of hardware to wires (e.g. the DSF), such as improved stiffness, should be considered carefully since the wires are still likely to fail, shifting loads to the remaining construct which may precipitate failure and lead to dehiscence.
3. As compared to the stainless-steel wire and cable systems, cyclic loading of the DSF plate system resulted in considerably more stainless-steel wear debris. Although the effects of metallic debris have not been reported on in the literature with sternotomy closure, current literature does report that metallic debris like this could lead to metal toxicity or corrosion assisted fatigue [17, 20]. This study, in addition to Cohen and Griffin's published biomechanical comparison examining ultimate stress to failure testing of the same three closure techniques, provides extensive experimental evidence regarding the mechanical differences among these three peristernal median sternotomy closure techniques. All data supports the hypothesis that the cables are the best fixation method of these three techniques and that eliminating wires altogether may lead to an improved clinical outcome.
The authors wish to thank Karla Spencer for her contributions to the development of test fixtures used for this study. Test materials were donated for this project by Pioneer Surgical Technology, Marquette, MI and Pectofix Inc., South Plainfield, NJ.
- Satta J, Lahtinen J, Raisanen L, Salmela E, Juvonen T: Options for the management of poststernotomy mediastinitis. Scand Cardiovasc J. 1998, 32: 29-32. 10.1080/14017439850140319.View ArticlePubMedGoogle Scholar
- El Oakley RM, Wright JE: Postoperative mediastinitis: classification and management. Ann Thorac Surg. 1996, 61: 1030-6. 10.1016/0003-4975(95)01035-1.View ArticlePubMedGoogle Scholar
- Losanoff JE, Collier AD, Wagner-Mann CC, Richman BW, Huff H, Hsieh F, Diaz-Arias A, Jones JW: Biomechanical comparison of median sternotomy closures. Ann Thorac Surg. 2004, 77: 203-9. 10.1016/S0003-4975(03)01468-1.View ArticlePubMedGoogle Scholar
- Bruhin R, Stock UA, Drucker JP, Azhari T, Wippermann J, Albes JM, Hintze D, Eckardt S, Konke C, Wahlers T: Numerical simulation techniques to study the structural response of the human chest following median sternotomy. Ann Thorac Surg. 2005, 80: 623-30. 10.1016/j.athoracsur.2005.03.043.View ArticlePubMedGoogle Scholar
- Casha AR, Yang L, Kay PH, Saleh M, Cooper GJ: A biomechanical study of median sternotomy closure techniques. Eur J Cardiothorac Surg. 1999, 15: 365-9. 10.1016/S1010-7940(99)00014-7.View ArticlePubMedGoogle Scholar
- Hehrlein FW, Herrmann H, Kraus J: Complications of median sternotomy in cardiovascular surgery. J Cardiovasc Surg (Torino). 1972, 13: 390-3.Google Scholar
- Stoney WS, Alford WC, Burrus GR, Frist RA, Thomas CS: Median sternotomy dehiscence. Ann Thorac Surg. 1978, 26: 421-6.View ArticlePubMedGoogle Scholar
- Spencer K: Developoment of a sternum closure fatigue test from an ultmate strength test method. 2003, San Luis Obispo: California Polytechnic State UniversityGoogle Scholar
- Callister WD: Materials Science and Engineering: An Introduction. 2003, NY: John Wiley and Sons, 6Google Scholar
- Wilkinson GA, Clarke DB: Median sternotomy dehiscence: a modified wire suture closure technique. Eur J Cardiothorac Surg. 1988, 2: 287-90. 10.1016/1010-7940(88)90086-3.View ArticlePubMedGoogle Scholar
- Casha AR, Gauci M, Yang L, Saleh M, Kay PH, Cooper GJ: Fatigue testing median sternotomy closures. Eur J Cardiothorac Surg. 2001, 19: 249-53. 10.1016/S1010-7940(01)00584-X.View ArticlePubMedGoogle Scholar
- Locke TJ, Griffiths TL, Mould H, Gibson GJ: Rib cage mechanics after median sternotomy. Thorax. 1990, 45: 465-8.View ArticlePubMedPubMed CentralGoogle Scholar
- Trumble DR, McGregor WE, Magovern JA: Validation of a bone analog model for studies of sternal closure. Ann Thorac Surg. 2002, 74: 739-44. 10.1016/S0003-4975(02)03699-8. discussion 745.View ArticlePubMedGoogle Scholar
- McNee C: A design analysis for fatigue testing sternomy closure techniques. Materials Engineering. 2004, San Luis Obispo: California Polytechnic State UniversityGoogle Scholar
- Cohen DJ, Griffin LV: A biomechanical comparison of three sternotomy closure techniques. Ann Thorac Surg. 2002, 73: 563-8. 10.1016/S0003-4975(01)03389-6.View ArticlePubMedGoogle Scholar
- McGregor WE, Trumble DR, Magovern JA: Mechanical analysis of midline sternotomy wound closure. J Thorac Cardiovasc Surg. 1999, 117: 1144-50. 10.1016/S0022-5223(99)70251-5.View ArticlePubMedGoogle Scholar
- Shih CC, Shih CM, Su YY, Lin SJ: Potential risk of sternal wires. Eur J Cardiothorac Surg. 2004, 25: 812-8. 10.1016/j.ejcts.2003.11.043.View ArticlePubMedGoogle Scholar
- Shih CM, Su YY, Lin SJ, Shih CC: Failure analysis of explanted sternal wires. Biomaterials. 2005, 26: 2053-9. 10.1016/j.biomaterials.2004.07.005.View ArticlePubMedGoogle Scholar
- Harjula A, Jarvinen A: Postoperative median sternotomy dehiscence. Scand J Thorac Cardiovasc Surg. 1983, 17: 277-81.View ArticlePubMedGoogle Scholar
- Jacobs JJ, Gilbert JL, Urban RM: Corrosion of metal orthopaedic implants. J Bone Joint Surg Am. 1998, 80: 268-82.PubMedGoogle Scholar
- Pearle AD, Scanzello CR, George S, Mandl LA, DiCarlo EF, Peterson M, Sculco TP, Crow MK: Elevated high-sensitivity C-reactive protein levels are associated with local inflammatory findings in patients with osteoarthritis. Osteoarthritis Cartilage. 2007, 15: 516-23. 10.1016/j.joca.2006.10.010.View ArticlePubMedGoogle Scholar
- Pearle AD, Crow MK, Rakshit DS, Wohlgemuth J, Nestor BJ: Distinct inflammatory gene pathways induced by particles. Clin Orthop Relat Res. 2007, 458: 194-201.PubMedGoogle Scholar
- Nazzal A, Lozano-Calderon S, Jupiter JB, Rosenzweig JS, Randolph MA, Lee SG: A histologic analysis of the effects of stainless steel and titanium implants adjacent to tendons: an experimental rabbit study. J Hand Surg [Am]. 2006, 31: 1123-30. 10.1016/j.jhsa.2006.03.001.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.