The Chinese finger trap (or finger trap) suture pattern is commonly used to secure the external portion of various types of tubing and catheters placed in dogs and cats. Common applications for finger trap sutures include enteral feeding catheters; pleural, peritoneal, and wound drainage catheters; and urethrostomy catheters. Secure fixation of catheters is important for preventing both minor and potentially life-threatening complications associated with premature or inadvertent removal; these complications are well described elsewhere in the literature but include stomal infection, drainage, pneumothorax, sepsis, and peritonitis. Even in the absence of direct negative consequences, replacement of dislodged catheters increases patient morbidity, exposure to anesthetics, and total cost of veterinary care; therefore, a secure means of affixing these materials is desirable.1–12
Multiple descriptions of finger trap–type sutures are present in the literature. The ability of the finger trap to grip the underlying catheter primarily appears to be a function of 2 factors: the rough surface of the finger trap increasing the coefficient of friction between the suture and tube and the ability of the arms of the finger trap to slide past each other at their intersections. Nevertheless, there is considerable variation among recommended suture materials, patterns, and the number of knots used, and almost all described finger trap variations include some element of knotting. Mitchell and Clark13 described a technique to secure biliary drainage catheters in people by use of 2–0 silk suture. Their technique involved placing a skin suture adjacent to the catheter with equal-length ends that were then crisscrossed around the catheter for approximately 6 cm; a tight triple knot was tied around the catheter, and then the pattern was repeated back to the skin, tied to the catheter, and then repeated again to the skin.13 Dixon et al14 described a modification to the pattern reported by Mitchell and Clark13 in which hairpiece tape was used to improve suture hold to the catheter material in various human drainage catheters. Smeak15 explained the pattern in the veterinary literature, recommending nonabsorbable suture ranging in size from 2–0 to size 2; a preference for braided versus monofilament suture was not reported. Smeak15 described a technique of placing a 0.5- to 1.0-cm section of suture through the skin approximately 1 to 2 cm away from the tube exit site and tying a square knot with equal suture end lengths. The ends were tied to the tube by use of a surgeon's knot, then crisscrossed down the tube, with 4 to 6 surgeon's knots spaced 0.25 to 0.50 cm apart that slightly compressed the tube without compromising the lumen.15 The tension placed on the tube by the knots reportedly resulted in constriction of the suture pattern, resulting in increased catheter hold.15 More recently, Radlinsky10 described the finger trap as a suture pattern initiated with a purse-string suture through the skin around the base of the tube, tied with 2 square knots, leaving 2 equal lengths of suture. The pattern proceeds away from the patient along the tube where suture ends are crossed on the side of the tube adjacent to the patient and tied on the contralateral side with a surgeon's throw that slightly compresses the tube.10 The suture crossing on the patient side and contralateral surgeon's throws are spaced at least as wide as the tube diameter, and ≥ 6 crossings and surgeon's throws are placed.10 The pattern is finalized with 2 square knots.10 Anecdotally, we have encountered numerous other variations on the theme.
Other investigators have biomechanically evaluated and compared finger trap sutures with 4 knots to the 4 friction suture technique with nonabsorbable suture and a variety of tube materials and diameters in a cadaver model.16 However, to our knowledge, an objective, biomechanical study evaluating the biomechanical properties of alternate knot placement, fewer knots, and suture and catheter material influence on security of catheters affixed with finger trap sutures has not been performed. The aims of the study reported here were to examine the influence of catheter material, suture material, and finger trap knot patterns on catheter security as well as to describe an alternate model for testing. The null hypothesis was that there would be no significant difference among finger trap styles, suture types, or catheter materials in resistance to axial load.
Materials and Methods
Leather swatches (40 × 20 × 3 mm; n = 132) were fabricated with 2-mm-diameter holes positioned 10 and 14 mm from 1 end on midline of the long axis. For each suture type, catheter type, and finger trap variant, a single square knot tied between the 2 holes secured the suture to the leather swatch, leaving 2 suture ends of approximately equal length. All finger trap variants were tied by 1 surgeon experienced with the techniques; 2–0 GLC,a monofilament polypropylene,b and braided nylonc suture materials were used to create finger traps on 5F polypropylened and polyvinylchloridee catheters placed so that 1 end extended 110 mm through the swatch (Figure 1). A permanent marking pen was used to create reference lines on the proximal and distal sides of the final knot to facilitate evaluation for slippage.
Photograph of a finger trap created with 2–0 braided nylon suture tied to a polypropylene catheter. The catheter was placed through a 40 × 20 × 3-mm leather swatch fabricated with two 2-mm-diameter holes positioned 10 and 14 mm from 1 end on midline of the long axis. The side of the catheter facing the hole 14 mm from the end was defined as the proximal surface.
Citation: Journal of the American Veterinary Medical Association 246, 5; 10.2460/javma.246.5.515
Finger trap variant techniques—Three finger trap variants were created (Figure 2). The first (finger trap variant 1; n = 52) was a traditional type of finger trap pattern currently used in the authors’ practices similar to the technique described by Radlinsky.10 The 2 suture ends affixed to the leather swatch were crossed under and around the catheter, and a surgeon's throw was placed on the surface of the catheter facing the swatch hole located 14 mm from the end on midline (ie, proximal surface), with sufficient tension to slightly indent the tube. The strands were crossed under and around the tube, and another surgeon's throw was placed on the proximal surface with sufficient tension to slightly indent the tube. This was repeated 4 more times and finalized with a square knot on the sixth repetition. The distance between locations where the sutures crossed was equal to or greater than the tube diameter.
Photographs depicting 3 finger trap variants assessed in an in vitro study evaluating the use of various finger trap patterns and suture materials for securing 5F polyvinylchloride and polypropylene catheters. Finger traps were created with 2–0 GLC, monofilament polypropylene, and braided nylon suture; the image shows 2.38-mm multifilament braided rope for illustration purposes. A—Finger trap variant 1 was a traditional style of finger trap suture currently used in the authors’ practices. Two suture ends affixed to a leather swatch with a square knot were crossed under and around the catheter, and a surgeon's throw was placed on the proximal surface with sufficient tension to slightly indent the tube. The strands were crossed under and around the catheter, and another surgeon's throw was placed on the proximal surface with sufficient tension to slightly indent the tube. This was repeated 4 more times and finalized with a square knot on the sixth repetition. B—Finger trap variant 2 was initiated as described for variant 1 with suture secured to the swatch and ends crossed under and around the catheter, but surgeon's throws were replaced by crossing the suture ends on the proximal surface for the first 5 repetitions. The 6th repetition was completed with a surgeon's throw that slightly indented the tube, and the pattern was finished with a square knot. C—Finger trap variant 3 was a hybrid between finger trap variants 1 and 2. Ends of the suture affixed to the leather swatch were crossed around the catheter as described in variant 1, with a surgeon's throw that slightly indented the tube placed for the first and sixth repetitions; the pattern was finished with a square knot.
Citation: Journal of the American Veterinary Medical Association 246, 5; 10.2460/javma.246.5.515
The second finger trap variant (finger trap variant 2; n = 44) was created by affixing suture to the leather swatch and crossing the suture ends under and around the catheter as described in variant 1, but without surgeon's throws on the proximal surface for the first 5 repetitions (these were replaced by simple crossing of the suture ends). A surgeon's throw that slightly indented the tube was used to complete the sixth repetition, and the pattern was finished with a square knot (Figure 2).
A hybrid between finger trap variants 1 and 2 was also created (finger trap variant 3; n = 36). The 2 ends of suture affixed to the leather swatch were crossed around the catheter as described in finger trap variant 1, with a surgeon's throw that slightly indented the tube used to complete the first repetition and the sixth repetition; suture was crossed on the proximal side for the intervening repetitions, and the pattern was finished by tying a square knot (Figure 2).
Mechanical testing—Six to 10 constructs/group were fabricated for testing (Appendix). Each construct was mounted in a universal materials testing machinef by anchoring the leather swatch with a universal steel clamp attached to a 15 kN/150 N•m load cell. The catheter was affixed to an eyed adaptor on the materials testing machine by creating 5 circumferential passages of the catheter through the eyed adaptor and then securing it with a steel clamp to the eyelet. The distance between the eyed adaptor and leather swatch was 43 mm for all constructs. A transducer was connected to the universal materials testing machine computer for data acquisition via the serial port, and the softwareg recorded force and distraction distance as a function of time. Sampling frequency was 50 Hz with rate of pull-out at 5 mm/s. Displacement began as force was applied to the catheter and stopped at failure or after reaching 100 mm of distraction. Failure was defined as suture breakage, suture slippage, or catheter breakage. Suture breakage included any detectable unraveling or breakage of any portion of the finger trap suture. Suture slippage was defined as suture visibly sliding over the tube beyond the reference marks. Catheter breakage was defined as plastic deformation of the catheter. Plastic deformation of the polypropylene catheters was associated with a visible color change of the material. If multiple modes of failures occurred, the first was considered the failure mode. All testing was digitally video recordedh to enable repeated evaluation. Constructs with incomplete data collection or technical malfunction during construct testing were excluded from analysis.
Statistical analysis—Statistical analyses were conducted with the aid of statistical software.i A competing risks model was used assuming a 3-factor factorial model in a randomized complete block design, with suture type, finger trap, and catheter as factors of interest and replicate as the blocking factor. Because an upper limit was assumed for the distance response (100 mm), Cox proportional hazards regressioni for competing risks was used to model the right-censored data and perform tests of the interactions and main effects. Medians and ranges for the effects evaluated are reported. Proportions of censored values for the levels of the factors of interest were compared by means of contingency tables and χ2 tests. Values of P < 0.05 were considered significant.
Results
Incomplete data collection due to errors during testing resulted in the removal of 3 constructs from the data analysis. A single construct of fingertrap variant 1, tied with GLC suture on a polypropylene catheter (from a total of 10), and 2 constructs of fingertrap variant 2, tied with GLC suture on polypropylene catheters (from a total of 10), were removed from analysis. Evaluation of proportions of right-censored data for each level of finger trap variant, suture, and catheter type revealed that 35 of 51 (68.6%) constructs that involved finger trap variant 1 failed during testing (ie, at or below the maximum of 100 mm distraction), resulting in censoring of the remaining 16 (31.4%) constructs. Twenty-six of 42 (61.9%) constructs that involved finger trap variant 2 failed during testing, with the remaining 16 (38.1%) censored. Eighteen of 36 (50%) that involved finger trap variant 3 failed during testing, and 18 (50%) were censored. These were not significantly (P = 0.213) different. Denominators for the proportions of tests varied because of the differing number of construct variants tested in the analysis.
Every polypropylene catheter construct failed during testing, with 0 of 63 results censored. Sixteen of 66 (24.2%) polyvinylchloride catheter constructs also failed, resulting in censoring of the remaining 50 (75.8%) constructs; the difference between catheter types was significant (P < 0.001). For suture type comparisons, 31 of 49 (63.3%) GLC-tied constructs failed during testing (18 [36.7%] censored), 29 of 44 (65.9%) braided nylon–tied constructs failed during testing (15 [34.1%] censored), and 19 of 36 (52.8%) monofilament polypropylene–tied constructs failed during testing (17 [47.2%] censored). These were not significantly (P = 0.455) different. When test results for constructs created with polyvinylchloride catheters only were considered, 8 of 26 (30.8%) GLC-tied constructs failed, leaving 18 (69.2%) censored; 7 of 22 (31.8%) braided nylon–tied constructs failed, with 15 (68.2%) censored; and 1 of 18 (5.6%) monofilament polypropylene–tied constructs failed, with 17 (94.4%) censored. These values did not differ significantly (P = 0.095). The competing risks model for finger trap data for all replicates and all combinations of variables revealed no significant differences among the 3 variants (Table 1). However, the competing risks model revealed some significant interactions of suture type and catheter material (Tables 2 and 3). For polypropylene catheters, braided nylon and GLC suture withstood greater distraction distance and force, respectively. Every polypropylene catheter tested failed through plastic deformation of the catheter material at the sites where the finger trap engaged the catheter, resulting in shorter distraction distances, compared with polyvinylchloride catheters. However, all polyvinylchloride catheter failures were attributable to slippage of the catheter through the finger trap beyond the reference marks.
Results of analysis with a competing risk model evaluating in vitro security variables for 3 finger trap constructs created with 2–0 GLC, braided nylon, and monofilament polypropylene suture on 5F polypropylene and polyvinylchloride catheters.
| Finger trap variant | ||||
|---|---|---|---|---|
| Variable | 1 | 2 | 3 | P value |
| Distance (mm) | 58.2753 (14.0826–100.060) | 53.1821 (18.9220–100.058) | 63.6991 (15.0970–100.055) | 0.959 |
| Force (N) | 29.2008 (19.3615–33.2092) | 30.6419 (21.9335–33.9616) | 29.4765 (25.8096–34.0972) | 0.105 |
Values reported are median (range).
A leather swatch with two 2-mm-diameter holes (10 and 14 mm from 1 end on midline of the long axis) was used for each construct, with a single square knot securing the suture to the leather swatch. The side of the catheter facing the hole 14 mm from the end was defined as the proximal surface. Finger trap variant 1 was a traditional finger trap pattern currently used in the authors’ practices; the 2 suture ends were crossed under and around the catheter, and a surgeon's throw was placed on the proximal surface with sufficient tension to slightly indent the tube. The strands were crossed under and around the tube, and another surgeon's throw was placed on the proximal surface with sufficient tension to slightly indent the tube. This was repeated 4 more times and finalized with a square knot after the sixth repetition. Finger trap variant 2 was initiated as described for variant 1 with suture ends crossed under and around the catheter, but surgeon's throws were replaced by crossing the suture ends on the proximal surface for the first 5 repetitions. The sixth repetition was completed with a surgeon's throw that slightly indented the tube, and the pattern was finished with a square knot. Finger trap variant 3 was a hybrid between finger trap variants 1 and 2. Suture ends were crossed under and around the catheter as described for variant 1, with a surgeon's throw that slightly indented the tube placed for the first and sixth repetitions; the pattern was finished with a square knot. Six to 10 constructs/group (132 constructs in total) were constructed and tested with a universal materials testing machine to the point of failure or a maximum of 100 mm of distraction. Data for constructs that did not fail at or before the test limit were censored.
Catheter-by-suture type interactions for distraction distances in a competing risk model comparing in vitro security variables for 3 finger trap variants tied on polypropylene or polyvinylchloride catheters with various types of 2–0 suture material.
| Catheter | Suture | Distance (mm) | P value |
|---|---|---|---|
| Polypropylene | GLC | 23.569 (14.0826–30.186) | < 0.001 |
| Braided nylon | 27.676 (19.3241–35.685) | ||
| Monofilament polypropylene | 21.267 (16.8429–25.398) | ||
| Polyvinylchloride | GLC | 99.998 (54.6550–100.060) | 0.261 |
| Braided nylon | 99.996 (51.7092–100.058) | ||
| Monofilament polypropylene | 100.027 (77.3750–100.049) |
See Table 1 for key.
Catheter-by-suture type interactions for force in a competing risk model comparing in vitro security variables for 3 finger trap variants tied on polypropylene or polyvinylchloride catheters with various types of 2–0 suture material.
| Catheter | Suture | Force (N) | P value |
|---|---|---|---|
| Polypropylene | GLC | 31.4443 (27.5745–33.2092) | 0.002 |
| Braided nylon | 30.4525 (27.0658–34.0972) | ||
| Monofilament polypropylene | 29.9298 (28.1173–31.7939) | ||
| Polyvinylchloride | GLC | 28.2445 (21.9335–32.3499) | 0.444 |
| Braided nylon | 28.7081 (19.3615–33.9616) | ||
| Monofilament polypropylene | 27.7035 (23.8559–31.7129) |
See Table 1 for key.
Discussion
To our knowledge, this is the first study to biomechanically evaluate the influence of alternate finger trap patterns with respect to different suture and catheter materials. Our clinical experience is that the suture material chosen for the finger trap is often selected in a number of ways, including surgeon preference and availability (eg, suture remaining from that used for incisional closure or what is available in a limited hospital inventory). We selected 2–0 suture materials for the present study because we believed they would more likely withstand the amount of force required to slightly indent the stiff polypropylene catheter material than a smaller-diameter suture would; 2–0 suture is also the smallest diameter reported in the study by Smeak.15
The purpose of a finger trap is to secure the catheter by the suture binding against the catheter as the suture arms that cross each other are placed under traction and drawn down against the catheter. A general characteristic of braided materials is a higher coefficient of friction (so-called drag), when compared with monofilament materials.17 Our results suggested that this increased coefficient of friction does not always benefit a finger trap's ability to provide overall catheter security, given the lack of significant difference in force and distraction distances for polyvinylchloride catheters tied with braided suture (GLC and braided nylon) when compared with monofilament (monofilament polypropylene) in vitro. The specific method for creating a finger trap can vary tremendously, and our impression is that the chosen method relies little on objective data and much more on personal opinion. Although it may seem to be a minor issue, finger trap patterns that use throws at multiple intersections of the crossing suture take longer to create and use more suture material. Furthermore, the very nature of a throw at each suture intersection would seem to prevent, or at least inhibit, the arms of the suture from sliding past each other and tightening against the underlying catheter. Finger trap variants 2 and 3 were modifications of finger trap variant 1 observed over time by the authors in their clinical practice. These variants had fewer knots than finger trap variant 1 and provided the authors with clinically based alternatives to finger trap variant 1 to evaluate the influence of knots on the finger trap–catheter interaction. Because there were no significant differences among the finger trap variants in this study, it seems most appropriate to use the simplest and most rapid method that results in a secure coupling of the catheter to the animal.
Given the degree of variability and limited objectivity in the literature describing finger trap patterns and their relative security, we sought to determine whether variations in material composition of suture and catheters, as well as variations in the finger trap pattern, had an effect on finger trap security. Although not a substitute for in vivo testing, an in vitro model for testing is a reasonable first step because it avoids patient morbidity while allowing evaluation of the interaction of the suture, suture pattern, and catheter material, all of which are physical variables that do not require animal use for assessment. We used leather swatches instead of cadaveric skin or other material to affix the materials for testing to provide uniformity in the material used to anchor the suture and to avoid some of the inherent variability of the tissues and in the amount of tissue included in the pattern when suturing the constructs to cadaveric skin. Cadaveric skin tends to be more elastic in comparison with leather and exposes the constructs to bodily fluids, which may affect construct behavior during testing and alter results. Our model did not account for the effects of fluids on the constructs; therefore, it is possible that different types and amounts of bodily fluids may alter performance of some or all of the tested constructs in a clinical setting. We chose to create a hole in the leather swatch that was larger than the catheter diameter so that only the interaction of the suture with the catheter was evaluated (vs any potential interaction between the leather swatch and catheter).
Our testing was limited to a maximum displacement of 100 mm by the equipment available for use; therefore, we did not reach a failure point (defined as catheter or suture breakage or suture slippage relative to reference marks) for all constructs tested. Although this was a limitation, we are not aware of any data that describe what a clinically relevant maximum distance of distraction of a catheter is or what amount of force resistance is required to adequately secure catheters with finger traps. We believed that distraction of 100 mm (or 2.33 times the catheter segment) was a subjectively appropriate maximum from a clinical standpoint; however, a standard has not been established. Song et al16 reported distracting larger catheter materials 27 cm in testing similar constructs to failure in a cadaver model. Interestingly, the finger trap sutures used by Song et al16 failed more commonly through suture breakage, whereas all our failures occurred by means of slippage for polyvinylchloride catheters and plastic deformation for polypropylene catheters. All the 5F polypropylene catheters tested in the present study failed within the 100 mm of distraction applied; therefore, we maintain that 100 mm was sufficient to evaluate these constructs to clinically relevant failure. The plastic deformation of the polypropylene catheters resulted in a visible narrowing of the catheter as well as a color change in the polypropylene. Our model did not simultaneously test catheter patency during testing, and it is possible that narrowing could cause occlusion of a catheter, rendering it clinically inoperable or at least more prone to occlusion at the point of narrowing. However, because failure was recorded at the first sign of plastic deformation (a noticeable color change) when the luminal diameter had not changed dramatically, we considered it unlikely that the catheters reached an inoperative state before failure was recorded. Ideally, future study should involve a model that can simultaneously test catheter patency in the midst of distraction.
The magnitude and direction of force that finger trap–secured catheters are subjected to in veterinary patients have not been quantified; therefore, there is presently no recommendation for the force these constructs should be able to resist in the clinical setting. To our knowledge, a model for simulating the forces placed on such catheter materials by veterinary patients in a clinical setting has not been described in the literature, and previous evaluation16 has relied on unidirectional axial loading of materials for test purposes. Forces applied to catheters in vivo occur from a variety of directions; therefore, unidirectional axial testing may not simulate the exact forces acting on these materials in a clinical setting, and correspondingly, the interactions between catheter and suture material may vary in those circumstances. Multiplanar fatigue testing of constructs versus load-to-failure axial testing may result in differing recommendations concerning catheter security in a clinical setting. Similarly, cyclic loading to failure may result in outcomes different from our results. Findings of the present study may have clinical relevance, but the authors caution readers to view these findings as a starting place for examining the clinical utility of these constructs in live patients.
Additionally, the interactions between finger trap pattern, suture material, and catheter material are complex and dependent on factors such as suture size and composition, tube diameter and composition, and the friction generated between catheter and suture during distraction as well as suture tension. Many combinations of suture material, suture pattern, catheter size, and catheter material could have been tested, but we chose to test a limited array of combinations that, in our experience, reasonably mimicked common clinical variations. Variability in the specific dimensions of each finger trap was minimized by having 1 investigator tie all sutures. However, we did not precisely control the length of the finger trap along each catheter or degree of tension applied at each knot to simulate a more realistic clinical scenario. Most descriptions of finger trap sutures call for throws that slightly indent the catheter material; however, the optimum force required to achieve this has not been standardized and likely varies with suture composition, suture diameter, catheter composition, and catheter diameter. Furthermore, the ideal degree of catheter indentation created with the suture has not been quantified, and surgeons have therefore relied on subjective assessment for the degree of appropriate indentation in creating these patterns. These limitations may create bias in the results and are also present in a similar report.16 Repeating the study with > 1 individual creating the finger traps may further define the extent of finger trap construct variability and the impact of such variability on the ability of the finger trap to secure the catheter without unacceptable distortion or damage to the catheter. However, this was beyond the scope and resources available for the present study.
The use of absorbable materials for securing catheters is common in clinical practice, on the basis of the authors’ collective experience in a variety of referral and academic practices. The reasons for this are unknown but speculated to be because of the suture's availability or common use in surgery and the variety of suture inventory, which is often limited. The present study did not account for the complex interaction and degradation of these materials as would occur in products used in vivo. The GLC suture material loses 20% of its tensile strength within 2 weeks and 70% within 3 weeks after placement in living tissue18; therefore, this material should only be selected for securing short-term catheters. Other absorbable monofilament suture materials have longer resorption times but were not evaluated in this study because we were primarily interested in the physical interactions of smooth and braided sutures with the polyvinylchloride and polypropylene catheters. Another reasonable criticism of the present study would be of the comparison of monofilament polypropylene with braided nylon. Because suture materials have different physical properties, a more direct comparison would have been facilitated by the use of braided and monofilament nylon sutures. All polypropylene catheters failed during testing by plastic deformation of the catheter material by permanent elongation and luminal narrowing where the finger trap engaged the catheter, whereas failures in polyvinylchloride catheters were uniformly by slipping of the catheter through the finger trap. In the present study, we considered failure as the point at which the first sign of slippage was observed; therefore, it is possible that the finger trap patterns could have constricted as testing continued to prevent further slippage of the catheter. This phenomenon may enable continued use of a catheter in a clinical setting. We speculate that the amount of clinically acceptable slippage would vary greatly according to catheter purpose, patient size, and possibly other variables.
Load recorded at failure or end of testing may vary in part because of subtle variations in the spacing of the crossings of the finger trap along the segment of catheter undergoing distraction. Although the proportion of censored tests for polyvinylchloride catheters weakly suggested that monofilament polypropylene–tied constructs were less likely to fail at ≤ 100 mm distraction, compared with other suture materials, there was no significant difference in median distraction distance or force at failure among suture types for constructs created with polyvinylchloride catheters. The competing risks analysis revealed that, in constructs created with polypropylene catheters, braided nylon enabled the greatest distraction distance, and GLC resisted the greatest force. Our evaluation of the data suggests that for 5F polypropylene and polyvinylchloride catheters, there was no difference in security (as measured by median force and distraction distances during load testing) among the 3 finger trap variants tested. On polypropylene catheters, braided nylon or GLC outperformed monofilament polypropylene suture.
Although, to our knowledge, previous studies have not reported displacement data, we quantified the amount of distraction the various constructs tolerated. Although we do not currently know the amount of distraction that veterinary patients subject catheters to, future study may reveal this information. This has the potential to affect the clinical relevancy of the data, may help direct materials chosen for future study, and may contribute to future data sets.
Owing to the complex nature of finger trap catheter security, the authors caution that the data obtained in this report should not be extrapolated to catheters of different diameters or materials, patterns tied with other suture sizes, or clinical performance in vivo. Our data would suggest that, for the material combinations we evaluated, a finger trap with fewer knots may provide catheter security similar to that of a finger trap tied with a more traditional pattern. From a practical standpoint, not placing throws at every crossing of the suture speeds creation of the finger trap.
ABBREVIATION
| GLC | Glycolide-lactide copolymer |
Polysorb, Covidien, Mansfield, Mass.
Surgipro, Covidien, Mansfield, Mass.
Surgilon, Covidien, Mansfield, Mass.
Kendall Sovereign Polypropylene 8890, Covidien, Mansfield, Mass.
Kendall Feeding Tube and Urethral Catheter 8890700514, Covidien, Mansfield, Mass.
MTS, model 370, MTS, Eden Prairie, Minn.
Project Manager and RS SigEdit modules, RS LabSite software, version 2.2.17SP1, Instron Structural Testing System Co, Novi, Mich.
JVC Digital Camcorder GZ-MG365BU, JVC, Wayne, NJ.
Proc PHREG, SAS, version 9.3, SAS Institute Inc, Cary, NC.
References
1. Crowe DT. Nutritional support for the hospitalized patient: an introduction to tube feeding. Compend Contin Educ Pract Vet 1990; 12: 1711–1720.
2. Lewis LD, Morris ML, Hand MS. Anorexia, inanition, and critical care clinical nutrition. In: Lewis D, Morris ML, Hand MS, eds. Small animal clinical nutrition III. 3rd ed. Topeka, Kan: Mark Morris Associates, 1987; 1–43.
3. Donoghue S. Nutritional support of hospitalized patients. Vet Clin North Am Small Anim Pract 1989; 19: 475–495.
4. Lippert AC. Enteral and parenteral nutritional support in dogs and cats with gastrointestinal disease. Semin Vet Med Surg (Small Anim) 1989; 4: 232–240.
5. Crowe DT. Enteral nutrition for critically ill or injured patients—part 1. Compend Contin Educ Pract Vet 1986; 8: 603–612.
6. Swann HM, Sweet DC, Michel K. Complications associated with use of jejunostomy tubes in dogs and cats: 40 cases (1989–1994). J Am Vet Med Assoc 1997; 210: 1764–1767.
7. Crowe DT, Devey JJ. Clinical experience with jejunostomy feeding tubes in 47 small animal patients. J Vet Emerg Crit Care 1997; 7: 7–19.
8. Yagil-Kelmer E, Wagner-Mann C, Mann FA. Postoperative complications associated with jejunostomy tube placement using the interlocking box technique compared to other jejunopexy methods in dogs and cats: 76 cases (1999–2003). J Vet Emerg Crit Care 2006; 16: S14–S20.
9. Lipscomb VJ. Bladder. In: Tobias KM, Johnson SA, eds. Veterinary surgery: small animal. St Louis: Saunders, 2012;1985–1986.
10. Radlinsky MA. Thorax. In: Tobias KM, Johnson SA, eds. Veterinary surgery: small animal. St Louis: Saunders, 2012; 1796.
11. Marques AIDC, Tattersall J, Shaw DJ, et al. Retrospective analysis of the relationship between time of thoracostomy drain removal and discharge time. J Small Anim Pract 2009; 50: 162–166.
12. Seim HB, Willard M. Postoperative care of the surgical patient. In: Fossum TW, ed. Small animal surgery. 2nd ed. St Louis: Mosby, 1997; 64–85.
13. Mitchell SE, Clark RA. Finger trap method of suturing biliary drainage catheters to the skin. AJR Am J Roentgenol 1981; 137: 628.
14. Dixon GD, Farthing DG, Longson F. Double coated medical tape modification of finger trap method for anchoring drainage catheters to the skin. Cardiovasc Intervent Radiol 1986; 9: 113–114.
15. Smeak DD. The Chinese finger trap suture technique for fastening tubes and catheters. J Am Anim Hosp Assoc 1990; 26: 215–218.
16. Song EK, Mann FA, Wagner-Mann CC. Comparison of different tube materials and use of Chinese finger trap or four friction suture technique for securing gastrostomy, jejunostomy, and thoracostomy tubes in dogs. Vet Surg 2008; 37: 212–221.
17. Schmiedt CW. Suture material, tissue staplers, ligation devices and closure methods. In: Tobias KM, Johnson SA, eds. Veterinary surgery: small animal. St Louis: Saunders, 2012;189.
18. Syneture. Polysorb braided absorbable sutures. Available at: www.syneture.com/syneture/pagebuilder.aspx?topicID=285#. Accessed Nov 6, 2013.
Appendix
Summary of constructs created in an in vitro study to biomechanically evaluate various finger trap patterns and suture materials for securing 5F polyvinylchloride and polypropylene catheters.
| Catheter type | |||
|---|---|---|---|
| Finger trap variant and suture type | Polypropylene | Polyvinylchloride | All |
| Variant 1 | |||
| GLC | 10 | 10 | 20 |
| Braided nylon | 10 | 10 | 20 |
| Monofilament polypropylene | 6 | 6 | 12 |
| Variant 2 | |||
| GLC | 10 | 10 | 20 |
| Braided nylon | 6 | 6 | 12 |
| Monofilament polypropylene | 6 | 6 | 12 |
| Variant 3 | |||
| GLC | 6 | 6 | 12 |
| Braided nylon | 6 | 6 | 12 |
| Monofilament polypropylene | 6 | 6 | 12 |
| Total | 66 | 66 | 132 |
Finger trap variant 1 was a traditional finger trap pattern in which the 2 suture ends were crossed under and around the catheter, and a surgeon's throw was placed on the proximal surface with sufficient tension to slightly indent the tube. The strands were crossed under and around the tube, and another surgeon's throw was placed on the proximal surface with sufficient tension to slightly indent the tube. This was repeated 4 more times and finalized with a square knot after the sixth repetition. Finger trap variant 2 was initiated as described for variant 1 with suture ends crossed under and around the catheter, but surgeon's throws were replaced by crossing the suture ends on the proximal surface for the first 5 repetitions. The sixth repetition was completed with a surgeon's throw that slightly indented the tube, and the pattern was finished with a square knot. Finger trap variant 3 was a hybrid between finger trap variants 1 and 2. Suture ends were crossed under and around the catheter as described for variant 1, with a surgeon's throw that slightly indented the tube placed for the first and sixth repetitions, and the pattern was finished with a square knot. All constructs were created on leather swatches with two 2-mm-diameter holes; a single square knot tied between the 2 holes secured the suture to the leather swatch.
