Effect of Biological Environment on Polyester Surgical Suture's Physical Properties: An Experimental Study in Rats.

Objective The following study aims to analyze the alteration of nonabsorbable polyester surgical suture physical properties after in vivo incubation. Methods A comparative study of braided nonabsorbable polyester 2/0 (U.S.P) sutures was performed. The control in vitro group and three experimental in vivo subgroups, composed of ten sutures in each, were created. All 30 experimental sutures were implanted into a total of 15 laboratory rats subcutaneous tissue (two sutures in each rat) and removed after seven, eight and nine weeks, respectively. Further, they were attached to the mechanical testing device and affected with a mechanical force, increasing the load by 0,1 N/s until complete breakage. Tensile strength (TS), failure displacement (FD), failure strain (FS) and failure stress (FST) were measured at the point of failure and compared to the same parameters of the control group. Results No statistically significant difference was found in the physical parameters of the samples between the experimental and control groups (TS [p = 0.358], FD [p = 0.258], FS [p = 0.258] FST [p = 0.358]). A statistically significant difference was found in the failure load between sutures that break on the knot site (KS) and the rest of the samples: significantly less force was needed to break the suture on the KS. Moreover, most of the breaks on the KS occurred in sutures that were incubated for the longest period of nine weeks (n = 4). An anomaly of partial failure (PF) was noticed. Sutures with PF elongated significantly more compared to the sutures that did not undergo PF in the control and in experimental groups (p = 0,044; p = 0,017; p = 0,016; p = 0,013). Conclusion The biological environment had no radical aftereffects to the suture's physical properties. In vivo exposure may cause the suture to break on the KS more frequently and may lead to PF, when a few sutures composed of fiber fail. Sutures that undergo PF tend to elongate further.

Introduction

Surgical sutures nowadays are an inseparable part of any surgery. They not only play a huge role in wound healing but can also become an inflammation or n class="Disease">allergy source. For this reason, any surgical suture must be non-reactive to tissues, maintain good knot security, and be strong and easily handled.

While there are various surgical techniques treating acromioclavicular (AC) joint dislocation, suture-button method or surgery with autograft tendon using nonabsorbable polyester suture remaiemical">ns widely performed aemical">nd analyzed [1-3]. Another suture-loop techemical">nique uses n class="Chemical">polyester sutures as well [4-5]. One of the most common complications after AC joint stabilization surgery is the loss of reduction [6]. Recently published papers reveal the main risk factors such as wrong clavicle tunnel location, button migration, osteoporosis and suture breakage for this complication to occur [7-8]. We predict that an in vivo environment may affect polyester suture's material and cause its physical properties to become worse leading to loss of reduction. A few published works analyzed maximum load to failure, elongation and other physical properties of various surgical sutures [9-11]. One of the studies has shown that polypropylene sutures preserve their stability after being taken out of a living organism, while catgut and Caprosyn lose their tensile strength (TS) [12]. In another study, conducted by Daniel A. Muller and colleagues, it was noticed that absorbable monofilament, polydioxanone, and non-absorbable polyester sutures retain stable material properties after two months of incubation under physiological conditions [11]. However, the polyester suture’s alteration and expiry while sewn in a living organism are still not well known.

The aim of this study is to analyze the alteratioemical">n of nonabsorbable n class="Chemical">polyester suture’s physical features such as TS, failure displacement (FD), failure strain (FS), and failure stress (FST) after being kept in vivo (rat) and observe its compliance in AC joint dislocation surgery. We speculate that these properties will worsen after in vivo incubation.

Materials and methods

Creation of study groups

A comparative study of braided nonabsorbable n class="Chemical">polyester 2/0 (U.S.P; Ethibond Excel) sutures was performed in Open-access Centre of Biomedical Physics Laboratory, Nacional Cancer Institute, Vilnius, Lithuania. All sutures were purchased at the same time and were within their expiration date. Sutures were checked for imperfections before the experiment. Durability and TS tests were performed after in vivo and in vitroconditions.

Two groups were designed for the current study:

1. Control in vitro group (10 sutures were tested after extracting them from sterile packaging).

2. Experimental in vivo group was divided into three subgroups with 10 sutures in each, which were implanted into five rats (two sutures iemical">n each n class="Species">rat):

A. seven-weeks subgroup

B. eight-weeks subgroup

C. nine-weeks subgroup.

Implantation steps and technique

All experimental group sutures were implanted into Wistar clone rats aemical">nd kept there for seveemical">n, eight aemical">nd niemical">ne weeks accordiemical">ngly. The experimeemical">nt was desigemical">ned iemical">n accordaemical">nce with the requiremeemical">nts stated iemical">n the 2010/63/EU Directive aemical">nd the Order of the Lithuaemical">niaemical">n State Food aemical">nd Veterinary Service Director No B1-866; 31-12-2012. Approval of the Ethics Committee aemical">nd permissioemical">n for the experimeemical">ntatioemical">n was received from the State Food aemical">nd Veterinary Service of Lithuaemical">nia. Fifteeemical">n staemical">ndard n class="Species">laboratory rats were used during experiments. The average weight of a rodent was 224,8 g (range between 200 and 251 g). All rats were kept in separate boxes with a separate air filtering system for every box thus preventing infection spreading among the rats. All rodents were kept under 12 hours day-night cycle at 20-21ºC and fed accordingly to their needs during the experiment. Ketamine (standard dose of 40 mg/kg intramuscular injection) was used for anesthesia during suture implantation. Surgical sites were prepared with 98% ethyl alcohol solution. Two 1-cm-wide cuts of skin were performed perpendicular to the spine with the first cut at the height of the first thoracic vertebra and second at the height of the first sacral vertebra. Canal, which connects both cuts, between hypoderm and muscle fascia was created, using mosquito forceps. Two sutures were implanted afterward through skin incisions in each rat subcutaneous tissue. Incisions were closed using two simple-interrupted stitches with 2/0 nonabsorbable sutures. After the procedure, rodents were injected with a single dose of cefazoline (rodent weight was considered when the concentration of antibiotic was calculated). No signs of inflammation, infection or sickness were recorded in any of the rats during the experiment. After the incubation time of seven, eight and nine weeks, sutures were removed from all rodents. It was done under aseptic conditions and intramuscular anesthesia using ketamine. Rodents were euthanized by intracarotid blood drawing later while still under the anesthesia.

Testing of physical parameters

TS was measured using Mecmesin MultiTest 2.5-i (2.5 kN) computer-controlled tensile and compression test system. For each tensile test, the suture was fixed on two hooks using “figure-eight” knots so ensuring a secure anchoring. Increasing load at the speed of 0.1 N/s was applied to the suture during the experiment. Other properties that were calculated were FS and FST. These parameters were calculated at a load of failure for all sutures. The leemical">ngth of all samples was calculated usiemical">ng a digital caliper aemical">nd it was doemical">ne before all measuremeemical">nts. Locations of suture ruptures were ideemical">ntified aemical">nd documeemical">nted. All measuremeemical">nts were doemical">ne iemical">n uemical">nits of the SI system. Further, all sutures were examiemical">ned under the stereomicroscope aemical">nd chaemical">nges of the surface were captured.

Statistical analysis

During data analysis, the normality of all parameters was tested using Kolmogorov-Smirnov test. A Shapiro-Wilk test (p > 0.05) and inspection of data’s histograms, Q-Q and box plots revealed that all data were distributed non-normally. Non-parametric tests were used further. Non-parametric Kruskal-Wallis test was used to determine if there is any significance among the groups of sutures and physical parameters. When a statistically significant difference was observed, the non-parametric Mann-Whitney U test was used between the groups. The level of significance (p) was chosen to be less than 0.05. IBM SPSS Statistics 21 and MS Excel 2016 were used for the statistical analysis. Graphical data were analyzed using Origin 2018 data processing software.

Results

Received data and calculations revealed that in vivo environment has no radical aftereffects to suture’s physical properties comparing control and experimental groups: TS (p = 0.358), FD (p = 0.258), FS (p = 0.258), FST (p = 0.358). However, the empirical analysis showed that there was a visible iemical">ncrease of failures oemical">n the kemical">not site (KS) iemical">n experimeemical">ntal subgroups. While there were no failures oemical">n the KS iemical">n the coemical">ntrol group, we did observe four failures oemical">n the KS iemical">n sutures that were kept iemical">n rodeemical">nts for niemical">ne weeks (Table 1).

Table 1

Values of measured physical parameters at failure point

Data are shown as means with a standard deviation

KS, knot site; PF, partial failure

Suture group	Tensile strength (N)	Failure displacement (m)	Failure strain (Δm/l)	Failure stress (MPa)	Failure on the KS (No.)	PF of suture (No.)
Control group (n = 10)	134,4 ± 10,9	0,028 ± 0,002	16,8 ± 1,1	497,7 ± 40,6	0	2
7-week subgroup (n = 10)	141,1 ± 6,3	0,031 ± 0,003	18,4 ± 1,8	522,6 ± 23,3	1	3
8-week subgroup (n = 10)	140,5 ± 7,9	0,029 ± 0,004	18,2 ± 2,1	520,3 ± 29,4	2	5
9-week subgroup (n = 10)	140,7 ± 4,0	0,030 ± 0,004	18,1 ± 2,5	521,2 ± 14,8	4	6

KS, knot site; PF, n class="Disease">partial failure

Furthermore, statistical analysis revealed a correlation in failure load between the sutures that break on the KS and the rest of the samples. Significantly less force was needed to break the suture on the KS (Table 2).

Table 2

Comparison of failure load between sutures that failed on the knot site and the rest of suture

Data are shown as means with a standard deviation.

KS, knot site

Suture group	Failure load on the KS (N)	Failure load on the rest of suture (N)		Decrease of strength (%)
					p-value
Control group (n = 10)	No failures recorded		134,4 ± 10,9 (n = 10)	NaN	NaN
7-weeks subgroup (n = 10)	126,9 (n = 1)		142,7 ± 4,1 (n = 9)	11.1	0,016
8-weeks subgroup (n = 10)	127,1 ± 5,4 (n = 2)		143,8 ± 3,6 (n = 8)	11.6	0,034
9-weeks subgroup (n = 10)	137,1 ± 1,6 (n = 4)		143,2 ± 3,1 (n = 6)	4.3	0,019

During the analysis of load-to-displacement graphs, an unexpected finding of partial failure (n class="Disease">PF) was noticed: a sudden loss of tensile force was visible, but the suture did not totally break. In the seven-week subgroup, all three samples that have undergone PF were removed from different rats; in the eight-week subgroup, two samples were removed from the same rat and the other three from different ones; in the nine-week subgroup, two from the same rat and other four from different rats. Moreover, sutures with PF tended to expose higher displacement which is visible in graphs (Figures 1-4).

Figure 1

Load to displacement graphs of sutures: Control group

Arrows mark the sudden loss of load force (PF). n = 2

PF, partial failure

Figure 4

Load to displacement graphs of sutures: Nine-week implantation subgroup

Arrows mark the sudden loss of load force (PF). n = 6

PF, partial failure

Arrows mark the sudden loss of load force (n class="Disease">PF). n = 2

PF, n class="Disease">partial failure

Load to displacement graphs of sutures: Seven-week implantation subgroup

Arrows mark the sudden loss of load force (n class="Disease">PF). n = 3

PF, n class="Disease">partial failure

Load to displacement graphs of sutures: Eight-week implantation subgroup

Arrows mark the sudden loss of load force (n class="Disease">PF). n = 5

PF, n class="Disease">partial failure

Arrows mark the sudden loss of load force (n class="Disease">PF). n = 6

PF, n class="Disease">partial failure

There was a significant displacement at failure load between sutures with PF aemical">nd without n class="Disease">PF in experimental and control groups. Sutures that undergo PF tended to elongate more significantly (Table 3).

Table 3

Comparison of displacement at failure load between sutures with PF and without PF

Data are shown as means with a standard deviation.

PF, partial failure

Suture group	Displacement of sutures with PF (m)	Displacement of sutures without PF (m)	p-value
Control group	0,035 ± 0,002 (n = 2)	0,027 ± 0,002 (n = 8)	0,044
7-weeks subgroup (n = 10)	0,031 ± 0,001 (n = 3)	0,027 ± 0,001 (n = 7)	0,017
8-weeks subgroup (n = 10)	0,033 ± 0,001 (n = 5)	0,028 ± 0,002 (n = 5)	0,016
9-weeks subgroup (n = 10)	0,034 ± 0,001 (n = 6)	0,027 ± 0,001 (n = 4)	0,013

PF, n class="Disease">partial failure

Sutures’ surface deformations such as single breakage of fibers and thickened areas were clearly seen during a microscopic inspection (Figures 5-6).

Figure 5

Microscopic inspection of suture deformation: Breakage of single fibers

Magnification 40x

Figure 6

Microscopic inspection of suture deformation: Thickened area of sutures

Magnification 28x

Discussion

Even though sutures are an inseparable part of any surgeon’s work routine, there are only a few works analyzing how in vivo environment affects the suture material and its physical properties. The most important finding in this work is the observation of sutures PF after keepiemical">ng them iemical">n n class="Species">rats.

The research conducted by Steven E. Naleway and coauthors showed that the KS dramatically decreases suture's TS [13]. Such a statement is confirmed by our findings. During the analysis of extracted sutures that were incubated in rodents, we did observe an increased rate of failures oemical">n the KS dependiemical">ng oemical">n their iemical">ncubatioemical">n time (Table 1). There were no failures oemical">n KS iemical">n the coemical">ntrol group, oemical">ne failure was recorded iemical">n the seveemical">n-weeks subgroup, two aemical">nd four failures were recorded iemical">n eight- aemical">nd niemical">ne-week subgroups, respectively. Moreover, sigemical">nificaemical">ntly less load force was needed to break the suture oemical">n the KS (Table 2). This findiemical">ng is relevaemical">nt to the cliemical">nical use of sutures. It has beeemical">n proved that after the coracoclavicular ligameemical">nt reconstructioemical">n, oemical">ne of the complications is loss of reductioemical">n [1,14]. Such loss of compressioemical">n may result from the rupture of sutures oemical">n the KS [8,15]. Surgeons should take iemical">nto considen class="Species">ration the fact that during the time in internal medium, sutures may lose their initial TS and take precautionary measures to avoid that.

A study of M.A. Tolga Muftuoglu and coauthors showed that nonabsorbable sutures lose their TS after in vivo incubation: Polypropylene aemical">nd silk sutures lost their iemical">nitial streemical">ngth by 16% aemical">nd 8%, respectively [16]. We did not observe aemical">ny kind of sigemical">nificaemical">nt loss of TS wheemical">n the coemical">ntrol group iemical">n vitro of braemical">nd-new sutures was compared with the experimeemical">ntal iemical">n vivo group sutures that were kept iemical">n rodeemical">nts for seveemical">n, eight aemical">nd niemical">ne weeks. Duriemical">ng the investigatioemical">n of force-displacemeemical">nt graphs, we observed anomalies iemical">n curves wheemical">n a n class="Disease">sudden loss of load force was visible, but suture did not totally break (Figures 1-4). We called these anomalies PF of sutures. We believe that PF occurs when a few sutures composed of fibers break. The failure of single fibers was clearly seen under the stereomicroscope (Figure 5). Moreover, we observed a significant displacement between the sutures with PF and sutures without PF in the experimental and control groups: sutures with PF displaced significantly more before breaking. This is also visible in the graph of the experimental study performed by Gomide et al., where the TS of the same sutures was measured at the rupture point [9]. It is important to emphasize that PF occurs more frequently within every week of implantation. It may be related to the incubation time which is needed for the biological environment to affect the sutures. These findings may have a link with compression loss within the first year after the correction of AC dislocation [17].

PF of the suture may be explaiemical">ned by the research of E. Karaca aemical">nd A.S. Hockeemical">nberger where the breakiemical">ng morphology of nonabsorbable n class="Chemical">polyester suture was researched [18]. Filaments tend to migrate toward the suture axis on loading and endure higher TS and displacement until total break during load. Changes in polyester suture surface after removal of load were described: Fibers rearrange across the diameter resulting in the thick and thin regions along the suture. Familiar deformation of the thickened area was observed during our microscopic examination of suture (Figure 6). PF phenomenon observed during this work might determine such changes in the suture’s surface and length.

The same paper tested two braided (polyester, silk) aemical">nd two monofilameemical">nt (n class="Chemical">polyamide, polypropylene) sutures (0, 2/0, 3/0 U.S.P) [18]. Silk and polyamide sutures showed a significant decrease in their TS while polypropylene and polyester sutures retained their TS after three and eight weeks under in vivoconditions. Moreover, a considerable increase of TS was recorded in the group of polyester 3/0 USP sutures. Familiar outcomes of the polyester suture were seen in our job - suture’s TS increased after in vivo incubation. Such a phenomenon may be explained by the PF mechanism which might be more common in thinner polyester sutures. Further investigation is needed to confirm it.

Conclusions

The biological environment has no radical aftereffects to the physical properties of braided polyester suture. However, it may lead to n class="Disease">PF and determine more common suture breakage on the KS. Sutures that undergo PF tend to elongate more and might cause the loss of reduction after AC joint dislocation surgery.