Several complications related to user or manufacturer errors concerning epidural catheters have been reported (1–3). Although these catheters have been designed resist a certain amount of flexion and traction, unintended trauma to the catheter during insertion or removal may result in serious damage to the catheter and eventual breakage (2,4).
Standardized data regarding the mechanical properties of epidural catheters have not been published. One study has addressed the breaking strain of clear and radiopaque catheters (5), but the method was based on “dead weight loading” rather than the use of a constant rate of extension. Computerized tensile testing by a constant rate of extension mimics the clinical situation more closely and is more reproducible (6,7). The only study using a tensile machine to measure the breaking strength of epidural catheters tested a single catheter type (clear nylon) (8). The clinical study part of the same investigation does not provide data on the speed at which the catheters were pulled out of the body; therefore, the clinical experiment is not reproducible (8).
The aim of this study was to compare the mechanical properties of three types of epidural catheters (clear nylon, polyurethane, and radiopaque marked clear nylon) when they were intact or traumatized by a needle bevel or a surgical blade in a controlled setting. We measured the resistance of these traumatized catheters to a traction force and compared it with that of nontraumatized catheters. We also planned to investigate which trauma model would render the catheter more prone to breakage. These data may be used as reference points for further studies using these catheters or in the development of catheters made of new materials.
Epidural catheters were obtained from three manufacturers: Abbott (Abbott Ireland Ltd., Sligo, Ireland), Portex (Portex Ltd., Hythe, Kent, UK), and B. Braun (Perifix-B. Braun, Melsungen AG, Germany). All catheters were designed for use with 18-gauge Tuohy needles. Fourteen catheters from each manufacturer were included in the study. Each catheter type was handled as a group, referred to as A (polyurethane), B (radiopaque), and C (clear nylon) groups. Each type was obtained from a single manufacturer, and therefore, it is reasonable to assume that the catheters in each group had the same geometric and material properties. All catheters were kept in the same environmental conditions and were not exposed to excessive heat, cold, or humidity before the study. Only single-use catheters with undamaged packages were included.
The study was performed in a blinded manner. The investigator performing the trauma models to the catheters and the investigators performing the mechanical tests were blinded to the catheter types. All catheters were cut into 30-cm pieces (minimal requirement for the present tensile tests). The study groups were:
- Group A (1): Polyurethane catheter control (nontraumatized)
- Group A (2): Polyurethane catheter needle bevel trauma
- Group A (3): Polyurethane catheter surgical blade trauma
- Group B (1): Radiopaque catheter control (nontraumatized)
- Group B (2): Radiopaque catheter needle bevel trauma
- Group B (3): Radiopaque catheter surgical blade trauma
- Group C (1): Clear nylon catheter control (nontraumatized)
- Group C (2): Clear nylon catheter needle bevel trauma
- Group C (3): Clear nylon catheter surgical blade trauma
Each group contained 10 catheter specimens (30-cm each). One of the authors performed all catheter trauma under operative microscopic magnification. Carl Zeiss, Opmi-CS Varioscope AF microscope (Carl Zeiss Geschaftsbereich Medizinisch-Optische Gerate D-73446, Oberkochen, Germany) was used with 40× magnification. Each catheter was taped to a ruler and was traumatized at the 15-cm mark, exactly at the midpoint. Care was taken to puncture only one wall of the catheter (with a needle bevel or a surgical blade) and to advance to the midst of the lumen (depth = 0.40 ± 0.01 mm for all groups). Both trauma models were created in the axial direction of the catheter (length = 0.013 ± 0.001 mm for needle bevel traumas and 0.034 ± 0.002 mm for blade traumas) in all groups measured under microscope after trauma by a digital caliper sensitive to 0.001 mm. Single-use needles (22-gauge × 1 1/4 inches) and blades were used. Each needle and each blade were discarded after performing a single puncture.
The tensile tests of the epidural catheters were performed at 23 ± 1°C on a fully automated Lloyd LS500 material testing machine (Lloyd, Southampton, UK), which was controlled via a personal computer with DAPMAT-1.4 software. Figure 1 shows the Lloyd LS500 and the personal computer while testing a Group A catheter. The load during the testing was detected through a 500-N load cell, and the displacement data were obtained from the cross-head. Data sampling rate was 3.3 Hz. The loading was a displacement-controlled type, and the cross-head speed was 50 mm/min. Maximal length of extension was 1000 mm. The gauge length was 34.5 mm. Load versus elongation data were collected through the tests, which, afterward, were converted to stress versus strain data by using the mechanics of materials formulation given below (9,10). By presenting the results in terms of stress-strain diagrams rather than load-elongation diagrams, we implicitly include the catheter cross-sectional area in the analysis without having to consider that as an additional variable. The use of such stress-strain diagrams for comparative evaluation puts the analysis on a more uniform basis. The necessary formulas are as follows:
MATH where ς = stress, P = load applied, Ao = original cross-sectional area of the specimen, ε = strain, l = final length of the specimen, and lo = initial length of the specimen.
After obtaining the stress-strain diagrams for all control and trauma groups, two variables from these diagrams were calculated, allowing us to draw conclusions for the comparative analysis. The first variable is the modulus of elasticity (E) which is the slope of the initial linearly elastic portion of the stress-strain diagram, and it represents the stiffness of the material. In this linearly elastic zone, stress and strain are related through Hooke’s law (9): ς =E ε.
In calculating the moduli of elasticity, a first-order polynomial was fitted to the data belonging to the initial linear portion of each of the stress-strain diagrams of the catheter groups. The slope of each of these line equations represented the modulus of elasticity.
The second variable obtained from the stress-strain diagrams was toughness, defined as the ability of absorbing energy in the plastic deformation zone and usually characterized by the total area under the stress-strain diagram (10). Toughness values of different catheter groups were calculated by finding the total area under each one of the stress-strain diagrams numerically.
The results were expressed as mean ± SD. Analysis of variance for completely randomized groups and post hoc Tukey’s t-test were used for statistical cross comparison of data, and the differences of P < 0.05 were considered statistically significant.
One important fact about the tensile tests performed is that the nontraumatized, control group of polyurethane catheters (Group A) did not break within the elongation limit of the tensile testing machine. Thus, neither the exact breaking strength nor the exact toughness value is available for group A (Control). In the presentation of the results concerning Group A, the last collected load and elongation data have been used, meaning that some additional straining before breakage can be expected for this group. Apart from this, all the other specimens broke before the elongation limit of the machine was reached, and it was clearly observed that the traumatized specimens broke at load and elongation levels considerably smaller than the corresponding control levels, as expected. Table 1 shows mean maximal load values before breakage and the mean values of maximal elongation before breakage in the study groups.
In Figure 2, the stress-strain diagrams of different catheter types, allowing comparison of the behavior of the traumatized and control groups within each type, are given. The rearranged stress-strain diagrams, allowing comparison of different catheter types under the same testing conditions, are given in Figure 3.
In Table 2, the modulus of elasticity values are presented as mean ± SD to demonstrate the differences in the elastic behavior of the catheters and in Table 3, the toughness values are given as mean ± SD to emphasize their plastic deformation characteristics.
Several types of epidural catheters are available for clinical use, and their physical properties show considerable variations. These properties become more important when complications related to the epidural catheter occur, such as kinking, curling, occlusion, knotting, defects in construction, damage of catheter at insertion, severance, or breakage on removal (11,12). Data on mechanical properties of epidural catheters can be used to investigate the complications encountered during clinical use of different catheter types. In this study, we compared the break strengths of three types of epidural catheters during intact and traumatized conditions.
Davies et al. (8) studied epidural catheters produced by Portex (16-gauge). The material of the catheter tested was not mentioned; therefore, we could not compare their data with ours. In the same study, the strain was applied at a rate of 10 mm/min (laboratory study), whereas the rate of strain applied in the clinical study was not recorded (8). In this study, the rate of strain application was 50 mm/min. Although this may still be slightly slower than the rate applied in clinical conditions, it is more relevant to the clinical rate when compared with the previous investigations (8). We have not encountered any data on the break strength of intact or traumatized polyurethane epidural catheters. Difficulties encountered during the clinical use of wire-reinforced (2), Teflon® (DuPont, Wilmington, DE) (13), or radiopaque (1) catheters have been reported, but there have been no reports of complications related to the clinical use of polyurethane catheters, perhaps because polyurethane catheters have been only recently available for clinical use.
One of our most important findings is the extreme difference between physical characteristics of polyurethane catheters and the other two types of catheters. It has been shown that clear nylon catheters can be stretched by 30% of their original length, which was considerably more than the radiopaque catheters (5). In this study, polyurethane catheters were stretched by more than 300% of their original length without breaking.
From a mechanical point of view, it is appropriate to assess the experimental results in terms of the elastic and plastic behavior of the specimens. The modulus of elasticity, E, being the slope of the initial linear portion of the stress-strain diagram, is the definite conclusive variable for an elastic behavior analysis. It should be noted that, in this elastic region, the material undergoes no permanent (plastic) deformation, i.e., the specimen recovers its entire elongation when the load is removed. The modulus of elasticity represents the stiffness of the material, relating the stress to strain through Hooke’s law (9). A higher value of E means a smaller elastic strain for a given stress. When the control groups are considered, it can be seen in Table 2 that Groups A and C have very close E values, whereas that of Group B is considerably greater. Trauma, whether of needle bevel or surgical blade type, does not affect the modulus of elasticity values in Groups A and C; however, the modulus of elasticity of the traumatized Group B is lowered considerably, especially for the needle bevel trauma case. In light of the mechanical considerations given above, the present experimental results indicate that radiopaque catheters have considerably higher stiffness to elastic deformation under intact conditions. Under traumatized conditions, the elastic stiffness of radiopaque catheters approaches that of the other two catheter types, although still slightly higher, the differences being statistically significant. Greater stiffness to elastic deformation clinically means the catheters break before stretching happens. However, less stiffness may be beneficial, because stretching will lower the external diameter of the catheter and facilitate removal. However this hypothesis remains to be tested clinically.
Adding plastic behavior into consideration changes the picture significantly. Unlike elasticity, the plastic behavior of materials under tensile loading can be characterized by a number of aspects. In the present study, toughness, which is the ability of absorbing energy in the plastic zone, has been considered. The values of the total area under the stress-strain diagrams of each case have been calculated to represent the toughness of the specimens. It should be recalled that the control group of polyurethane catheters did not break within the elongation limit of the testing machine. The stress-strain diagrams in Figures 2A and 3A illustrating the behavior of this group do not, therefore, represent the complete picture, but only the available data. Table 3 shows that the toughness values of Group A (polyurethane) are much larger than the corresponding values of the other two groups, especially under traumatized conditions. Cross-comparison of the toughness values indicates that trauma, whether of needle bevel or surgical blade type, deteriorates the energy-absorbing capacity of all three types of catheters considerably. Although it is not statistically significant, it appears that surgical blade trauma may decrease the toughness value of the catheter more than needle bevel trauma, at least for radiopaque and clear nylon catheters.
Attempts at surgical removal of retained catheters have not been suggested, because catheter pieces are known to be sterile and inert (14). However, this may not be the case in some patients, and neurological complications can be caused by retained epidural catheter parts (13,15,16). Knowledge of the mechanical properties of catheters may be helpful in assessing the risk of complication.
The modulus of elasticity and toughness values we reported can be used as reference data for testing catheters made of new materials. We believe that these data should be available for the clinicians to make a correlation between the complications encountered clinically in relation to the catheter used.
In conclusion, polyurethane is the catheter material least prone to breakage, either intact or traumatized. However, radiopaque catheters have the largest stiffness values in the elastic range. Penetrating trauma, which causes a discontinuation of the catheter wall, does not significantly affect the elastic properties of the catheters considered, but it drastically decreases the energy absorbing capacity (toughness) of all the catheters. Assuming that the breakage of a catheter during removal is the most critical outcome, it would be safe to conclude that it is better to have a higher tensile strength for catheters. Polyurethane catheters seem to be better than radiopaque catheters given this assumption.
1. Tio TO, MacMurdo SD, McKenzie R. Mishap with an epidural catheter. Anesthesiology 1979;50:3:260–2.
2. Ellis JS, Ramamurthy S. More problems with the Arrow-Racz epidural catheter. Anesthesiology 1986; 65:124–5.
3. Seitman DT, Shapiro BE. Inability to thread epidural catheter through epidural needle [letter]. Anesth Analg 1989; 69:267–8.
4. Frankhouser PL. In reply. Anesthesiology 1986; 65:125–6.
5. Hutchison GL. The severance of epidural catheters. Anaesthesia 1987; 42:182–5.
6. Ley SJ, Jones BR. Strength of continuous spinal catheters. Anesth Analg 1991; 73:394–6.
7. Jones BR, Ley SJ. Strength of continuous spinal catheters [letter]. Anesth Analg 1992; 74:779.
8. Davies R, Vaughan RS, Richards J. Epidural catheters: breaking and extraction forces. Anaesthesia 1993; 48:900–1.
9. Beer FP, Johnston ER. Mechanics of materials. New York: McGraw-Hill, 1987.
10. Dieter GE. Mechanical metallurgy. New York: McGraw-Hill, 1988.
11. Ballance JHW. Difficulty in the removal of an epidural catheter. Anaesthesia 1981; 36:71–2.
12. Collins V. Epidural anesthesia. Principles of anesthesiology: general and regional anesthesia. 3rd ed. Philadelphia: Lea & Febigher, 1993: 1571–610.
13. Lenox WC, Kost-Byerly S, Shipley R, Yaster M. Pediatric caudal epidural catheter sequestration: an unusual complication. Anesthesiology 1995; 83:1112–4.
14. Rubin AP. Hazards of local and regional anesthesia. In: Taylor TH, Major E, eds. Hazards and complications of anesthesia. 2nd ed. Edinburgh: Churchill Livingstone, 1993: 591–612.
15. Staats PS, Stinson MS, Lee RR. Lumbar stenosis complicating retained epidural catheter tip. Anesthesiology 1995; 83:1115–8.
© 2000 International Anesthesia Research Society
16. Jongleux EF, Miller R, Freeman A. An entrapped epidural catheter in a postpartum patient. Reg Anesth Pain Med 1998; 23:615–7.