Abstract
Purpose.
Expanded polytetrafluoroethylene suture is commonly used for chordal replacement in mitral valve repair, but due to material characteristics, knots can unravel. Our aim was to determine the knot security, including how many throws are necessary to prevent knot failure, with Gore-Tex (W.L. Gore and Associates, Elkton, MD) and the newly available Chord-X (On-X Life Technologies Inc, Austin, TX).
Description.
Knots were evaluated for maximal load based on: number of throws (6, 8, 10, and 12), tension to secure each throw (10%, 50%, and 85%) and suture type (Gore-Tex CV-5 and Chord-X 3–0). A physiologic force of 2 N was used for comparison.
Evaluation.
We evaluated 240 knots. For all knots, the mean load to failure was 11.1 ± 5.8 N. Failure occurred due to unraveling in 141 knots (59%) at 7.1 ± 4.1 N and to breaking in 99 (41%) at 16.7 ± 2.0 N (p < 0.01). Gore-Tex failed at higher loads (12.6 ± 6.0N vs 9.5 ± 5.2 N, p < 0.01); however, an equivalent number, 6 Gore-Tex and 6 Chord-X, unraveled at 2 N, all with fewer than 10 throws.
Conclusions.
Expanded polytetrafluoroethylene has adequate strength to prevent breakage; however, a risk of knot unraveling at physiologic conditions exists when fewer than 10 throws are performed.
Mitral valve repair often involves chordae tendineae replacement. First proposed in the 1980s by Frater and colleagues [1], this is now commonly performed with an expanded polytetrafluoroethylene (ePTFE) suture [2]. ePTFE, used for its strength, has shown excellent results for chordal replacement [2]. Owing to the handling characteristics, however, there exists a risk of knots unraveling. Knot failure acutely postoperatively has dramatic consequences, requiring a repeat operation with cardiopulmonary bypass. Some compensate for this risk with multiple additional throws or using a clip to form a secure knot. However, no studies have evaluated the effect of the number of throws on knot unraveling. Furthermore, newer ePTFE suture has been introduced with slightly different handling characteristics. We aimed to evaluate the effect of the number of throws and the tension used to secure each throw on a knot’s security and evaluate the knot security of two different ePTFE sutures.
Technique
Knot Groups: Number of Throws, Tension, and Suture Type
There were four throw groups: 6, 8, 10, or 12 throws to form each knot. There were three tension groups: 85%, 50%, and 10% of the knot holding power. The 85% group represents a snug knot as tied by a surgeon, the 10% group represents a loosely tied knot, and the 50% group provides an intermediate value [3, 4]. Two suture types were evaluated, Gore-Tex CV-5 (W.L. Gore and Associates, Elkton, MD) and Chord-X 3–0 (On-X Life Technologies Inc, Austin, TX). Owing to different extrusion techniques, these sutures have unique surface characteristics (Fig 1). Gore-Tex does not use standard suture sizing; therefore, suture sizes were selected by diameter. Gore-Tex CV-5 measures on average 0.245 mm in diameter, and Chord-X 3–0 measures between 0.2 and 0.245 mm in diameter. Each throw/tension/suture group consisted of 10 knots; therefore, 240 total knots were evaluated.
Fig 1.
Scanning electron photo-micrographs of Chord-X (On-X Life Technologies Inc, Austin, TX) and Gore-Tex (W.L. Gore and Assoc, Elkton, MD) sutures demonstrate variations based on different extrusion techniques. Chord-X at original magnification (A) ×300, (C) ×600, and (E) ×1,200. Gore-Tex at original magnification (B) ×300, (D) ×600, and (F) ×1,200.
Knot Holding Power
The knot holding power was determined for each suture as previously described [4]. Tension was applied to a hand-tied loop secured around a mandrel, with the maximal force before failure representing the knot holding power. This was performed using 10 modified identical sliding knots with each suture type. The maximal load to failure was determined by placing the loop around 2 hooks on opposite ends of the tensiometer (Instron series 5542 Universal Testing System; Instron, Norwood, MA) and increasing the distance at 20 mm/min. The mean was halved, half the tension was distributed to each of the 2 arms within the loop. The lower of the two was considered 100% for both, this allowed for both suture types to be secured with the same absolute tension [5].
Knot Tying
All knots were modified identical sliding knots, consisting of 2 identical sliding throws, followed by alternating throws until completion [6]. Each knot was tied around a plastic mandrel 0.5 inches in diameter. The initial 2 throws were performed without applying any tension. The knot was secured by placing the tails in a tensiometer and increasing the distance by 50 mm/min until the predetermined tension was reached (Fig 2A) [3]. This was repeated for each additional throw until completion. Tails were cut to 3 mm [4]. The knot was soaked in 0.9% saline for 15 minutes before evaluation.
Fig 2.
(A) Each knot was tied over a plastic mandrel and secured with a tensiometer. (B) After the appropriate number of throws, each knot was then attached to the tensiometer and pulled apart, increasing the distance at 20 mm/min.
Knot Evaluation
The loop was placed around 2 hooks on opposite ends of the tensiometer (Fig 2B). Tension was applied by increasing the distance at 20 mm/min, as previously described [3]. The failure method, unraveling or breaking, and maximal force before failure were recorded. The force was halved to account for both arms of suture in the loop.
Physiologic Force Comparison
The data were analyzed to determine the likelihood of failure at clinically significant forces, which have been demonstrated to be 0.5 to 2 N [7–9]. With asymmetric ventricular wall motion, as in functional mitral regurgitation, some chordae may become tethered and experience force increases of as much as 200% [10]. Therefore, data were analyzed for chordae experiencing 2 and 4 N. Because it is sutured as a loop when used for chordal replacement, each chordae is effectively replaced by 2 lengths of suture; therefore, the number that failed at 1 and 2 N was analyzed.
Statistical Analysis
Continuous data are expressed as mean ± standard deviation. Categoric data are expressed as counts, and percentages. The primary outcomes were failure by unraveling or breaking. Logistic regression was used to determine the effects of number of knots, tension to secure each throw, and suture type on the maximal load at knot failure. An analysis of variance was performed to determine the interaction between suture type and the other variables on maximal load. A p value of less than 0.05 was used as the critical level for statistical significance.
Clinical Experience
Knot Holding Power Determination
The knot holding power was 14.2 N for Chord-X and 18.5 N for Gore-Tex. For both suture types, each throw in the 85%, 50%, and 10% groups was secured with 12.1 N, 7.1 N, and 1.4 N, respectively.
Knot Failure
maximal load.
The mean maximal load before failure for all 240 knots was 11.1 ± 5.8 N. The maximal load at failure for each throw, tension, and suture group is reported in Table 1. Failure was by unraveling in 141 knots (59%) and by breakage in the remaining 99 (41%), all of which occurred at the knot (Fig 3). The maximal load before failure for knots that unraveled was lower, 7.1 ± 4.1 N, than for those that failed by breaking, 16.7 ± 2.0 N (p < 0.01).
Table 1.
Maximal Load at Failure for Each Suture Type Grouped by the Number of Throws and the Tension Used to Secure Each Throwa
| Maximal Load at Failure (N) | |||
|---|---|---|---|
| Throws | Tension (N) | Chord-Xb Mean ± SD | Gore-Texc Mean ± SD |
| 6 | 1.4 | 3.2 ± 1.7 | 4.3 ± 4.3 |
| 6 | 7.1 | 2.5 ± 0.9 | 5.0 ± 3.0 |
| 6 | 12.1 | 4.9 ± 1.5 | 7.3 ± 2.4 |
| 8 | 1.4 | 7.6 ± 3.6 | 11.3 ± 3.1 |
| 8 | 7.1 | 5.7 ± 2.0 | 11.4 ± 5.0 |
| 8 | 12.1 | 6.4 ± 2.3 | 7.6 ± 2.7 |
| 10 | 1.4 | 13.8 ± 3.7 | 16.1 ± 3.1 |
| 10 | 7.1 | 12.5 ± 3.3 | 17.5 ± 1.6 |
| 10 | 12.1 | 13.4 ± 1.9 | 18.4 ± 2.2 |
| 12 | 1.4 | 14.4 ± 1.6 | 17.3 ± 1.5 |
| 12 | 7.1 | 15.4 ± 0.6 | 18.0 ± 1.7 |
| 12 | 12.1 | 14.4 ± 3.0 | 17.2 ± 4.3 |
10 knots per group.
On-X Life Technologies Inc, Austin, Texas.
W.L. Gore and Associates, Elkton, Maryland.
SD = standard deviation.
Fig 3.
Maximal load before failure by breaking (red) or unraveling (blue) is shown for each individual knot, grouped by method of failure. The middle line represents the median value, with the border of each box representing the first interquartile range and the hinge representing three times this value.
PHYSIOLOGIC FORCES.
At 1 N, 3 knots (1.3%) failed, 2 Gore-Tex and 1 Chord-X (Table 2). At 2 N, 12 knots (5%) failed, 6 Gore-Tex and 6 Chord-X. All knots that failed at less than 2 N did so due to unraveling and occurred within the groups with 6 and 8 throws, with no failures occurring with at least 10 throws with either suture.
Table 2.
Rate of Failure for Expanded Polytetrafluoroethylene Knots Based on the Number of Throws per Knot and Amount of Tension in Each Throwa
| Failure Rate at 1 N | Failure Rate at 2 N | |||||
|---|---|---|---|---|---|---|
| Tension Group | Tension Group | |||||
| No. of Throws | 1.4 N | 7.1 N | 1.4 N | 1.4 N | 7.1 N | 12.1 N |
| 6 | 0.10 | 0.05 | 0 | 0.30 | 0.25 | 0 |
| 8 | 0 | 0 | 0 | 0.05 | 0 | 0 |
| 10 | 0 | 0 | 0 | 0 | 0 | 0 |
| 12 | 0 | 0 | 0 | 0 | 0 | 0 |
20 knots per group.
Number of Throws
An analysis of the knots that failed by unraveling showed that those within the higher throw groups required a higher maximal load to unravel (p < 0.01). Conversely, for knots that failed by breaking, the maximal load did not depend on the number of throws (p 0.94). Overall, the number of throws significantly influenced the maximal load before knot failure (p < 0.01; Fig 4). There was a significant increase in the maximal load before knot failure when increasing from 8 (8.2 ± 3.9 N) to 10 throws (15.3 ± 3.4 N), largely due to the decreased number of knots unraveling within the 10-throw group (p < 0.01).
Fig 4.
Mean maximal load before failure by unraveling or breaking based on the number of throws used to form the knot. The error bars represent the standard error of the mean.
Amount of Tension
The amount of tension used to secure each throw did not affect the maximal load at knot failure (p 0.83). However, when knots with 6 or 8 throws at 2 N were evaluated, the rate of knot failure was dependent on the tension used to secure the knot, with the 10%, 50%, and 85% tension groups failing at a rate of 0.18, 0.13, and 0, respectively (p = 0.02).
Suture Type
Gore-Tex failed at higher loads (12.6 ± 6.0 N) than Chord-X (9.5 ± 5.2 N; p < 0.01). This was a function of the increased force necessary to cause failure by breakage. When the failure rates of each suture were analyzed at clinically significant forces, however, there was no difference at 1 N, with 2 Gore-Tex (1.7%) failing and 1 Chord-X (0.8%) failing (p = 1.00), or at 2 N with 6 (5%) of each failing (p = 1.00). In addition, analysis of variance results confirmed that the interactions between suture type and tension or throws were not significant (p = 0.25), demonstrating similar performance between suture types.
Comment
Mitral valve repair has become a routine intervention, often including chordae tendineae replacement. ePTFE suture is often used; however, its handling characteristics are unique, and the potential for unraveling exists. This study demonstrated that the strength of an ePTFE knot is determined primarily by the number of throws and that a knot with 10 throws is unlikely to fail at physiologic forces.
The forces experienced by individual chordae in vivo differ according to their location and role, leaflet coaptation, or valvular/ventricular competence. Studies have demonstrated that the maximal force experienced by any particular chordae is approximately 0.5 to 1.5 N [7, 8]. In the setting of severe hypertension, this can increase up to 2 N [9].
This study confirms adequate strength of both ePTFE sutures; no sutures broke at forces of less than 10 N. At supraphysiologic levels, the increased force necessary to break the Gore-Tex CV-5 was likely a function of its larger diameter. When knots failed by breaking, the force required to induce failure was relatively consistent. When only knots that unraveled were analyzed, however, there was variability in the force necessary to unravel the knot. This finding indicates that the increase in load required to cause knot failure within the higher throw groups is attributable not only to the reduced frequency of unraveling but also to the higher forces necessary to cause knots with more throws to unravel.
Interestingly, when the entire collection of knots was considered, the tension used to secure each throw was not a significant factor. However, tension was significant when the rate of failure at 2 N within the 6 and 8 throw groups was analyzed. This finding indicates that if a surgeon uses only 6 or 8 throws, each throw should be secured with the maximal amount of tension. This study suggests that 10 throws may prevent the unraveling of ePTFE suture at clinically significant forces, even when insufficient tension is used to secure the knot.
Limitations
This study had some important weaknesses. A tensiometer was used to secure the knot. Although this allowed for consistency and objectivity, it was not possible to simulate exact clinical conditions. Also, this study could be underpowered, with only 10 knots per group. Lastly, the forces were applied continuously rather than pulsatilely, as occurs physiologically, which may affect the tendency to unravel. However, results were consistent, with increasing throws showing decreased failures, suggesting accuracy.
Conclusions
ePTFE is commonly used for chordal replacement, with excellent results. This study demonstrates that ePTFE has sufficient strength and further demonstrates that, at physiologic conditions, knot failure occurs by unraveling. By performing at least 10 throws for either suture type, this risk of unraveling can be decreased.
Disclosure and Freedom of Investigation
The study was supported by NIH grant T32 HL007776. Funding was provided by On-X Life Technologies, Austin, Texas. Chord-X suture was donated, and Gore-Tex suture was purchased. The authors had full control of the design of the study, methods used, outcome variables and results, analysis of data, and production of the written report.
Footnotes
Publisher's Disclaimer: Disclaimer
The Society of Thoracic Surgeons, the Southern Thoracic Surgical Association, and The Annals of Thoracic Surgery neither endorse nor discourage the use of the new technology described in this article.
References
- 1.Frater RW, Vetter HO, Zussa C, Dahm M. Chordal replacement in mitral valve repair. Circulation 1990;82(5 Suppl): IV125–30. [PubMed] [Google Scholar]
- 2.David TE, Armstrong S, Ivanov J. Chordal replacement with polytetrafluoroethylene sutures for mitral valve repair: a 25-year experience. J Thorac Cardiovasc Surg 2013;145:1563–9. [DOI] [PubMed] [Google Scholar]
- 3.Dang MC, Thacker JG, Hwang JC, Rodeheaver GT, Melton SM, Edlich RF. Some biomechanical considerations of polytetrafluoroethylene sutures. Arch Surg 1990;125: 647–50. [DOI] [PubMed] [Google Scholar]
- 4.Thacker JG, Rodeheaver G, Moore JW, et al. Mechanical performance of surgical sutures. Am J Surg 1975;130: 374–80. [DOI] [PubMed] [Google Scholar]
- 5.Faulkner BC, Gear AJ, Hellewell TB, Mazzarese PM, Watkins FH, Edlich RF. Biomechanical performance of a braided absorbable suture. J Long Term Eff Med Implants 1996;6:169–79. [PubMed] [Google Scholar]
- 6.Schubert DC, Unger JB, Mukherjee D, Perrone JF. Mechanical performance of knots using braided and monofilament absorbable sutures. Am J Obstet Gynecol 2002;187:1438–40; discussion 1441–2. [DOI] [PubMed] [Google Scholar]
- 7.Jimenez JH, Ritchie J, He Z, Yoganathan AP. Mechanics of the mitral valve: in vitro studies. Conf Proc IEEE Eng Med Biol Soc 2004;5:3727–9. [DOI] [PubMed] [Google Scholar]
- 8.Nielsen SL, Soerensen DD, Libergren P, Yoganathan AP, Nygaard H. Miniature C-shaped transducers for chordae tendineae force measurements. Ann Biomed Eng 2004;32: 1050–7. [DOI] [PubMed] [Google Scholar]
- 9.Prot V, Haaverstad R, Skallerud B. Finite element analysis of the mitral apparatus: annulus shape effect and chordal force distribution. Biomech Model Mechanobiol 2009;8: 43–55. [DOI] [PubMed] [Google Scholar]
- 10.Padala M, Gyoneva L, Yoganathan AP. Effect of anterior strut chordal transection on the force distribution on the marginal chordae of the mitral valve. J Thorac Cardiovasc Surg 2012;144:624–633 e2. [DOI] [PMC free article] [PubMed] [Google Scholar]