- Email: [email protected]
Variations in epidural catheter manufacture: Implications for bending and stiffness
Variations in Epidural Catheter Manufacture: Implications for Bending and Stiffness David M. Eckmann, Ph.D., M.D. Background and Objectives: There is no formal evaluation method used to relate epidural catheter design and manufacture to clinical outcomes, such as subarachnoid or intravascular catheter placement. We analyzed catheter bending stiffness to determine the range of stiffness of catheters commonly used. We hypothesized that catheter material has a greater influence on stiffness than does cross-sectional shape. Methods: We determined the elastic modulus by axial load testing and the area moment of inertia using calibrated microscopic measurements of cross-sectional geometry for 6 different catheter types, including 2 types of wire styletted catheters. We calculated bending stiffness as the product of the elastic modulus and the area moment of inertia. Results: Catheters had similar area moments of inertia, but markedly different elastic moduli. Nylon and polyurethane catheters had the same bending stiffness, which was twice as high as that of coil reinforced catheters (P ⬍ .05), but 35% lower than that of radiopaque catheters (P ⬍ .05). Nylon and radiopaque wire styletted catheters had similar bending stiffness, which were 23-fold to 90-fold greater than that of the nonstyletted catheters (P ⬍ .05). Conclusions: Catheters currently available establish the range of bending stiffness that should not be exceeded, only optimized to clinical outcome. Clinical studies are needed to correlate the incidence of unintentional intravascular or subarachnoid catheter placement or migration and bending stiffness. Catheter technology improvements may enhance safety and increase the likelihood of successful catheter insertion, maintenance, and removal. Reg Anesth Pain Med 2003;28:37-42. Key Words:
Bending stiffness, Catheter: Epidural, Mechanics: Buckling, Modulus of elasticity.
I
ntroduction of an epidural catheter can be complicated by intravascular or subarachnoid placement of the catheter tip.1 A catheter may also migrate from the epidural space into an epidural vein or into the subarachnoid space.2 Thus, local anesthetics unexpectedly may be delivered directly into the bloodstream or the cerebrospinal fluid, causing drug toxicity (seizure or cardiac arrhythmia) or sudden high spinal anesthesia. Unintended catheter penetration through a blood vessel wall or the meninges also increases the risks of bleeding, develop-
From the Department of Anesthesia, University of Pennsylvania Health System, Philadelphia, Pennsylvania and Institute of Medicine and Engineering, University of Pennsylvania, Philadelphia, Pennsylvania. Accepted for publication November 1, 2002. This work should be attributed to the University of Pennsylvania Department of Anesthesia. Reprint requests: David M. Eckmann, Ph.D., M.D., Department of Anesthesia, The University of Pennsylvania, 3400 Spruce Str, Philadelphia, PA 19104. E-mail: eckmanndm@ uphs.upenn.edu © 2003 by the American Society of Regional Anesthesia and Pain Medicine. 1098-7339/03/2801-0008$35.00/0 doi:10.1053/rapm.2003.50016
ment of a postdural puncture headache, or traumatic penetrating injury to other tissues. A design and manufacture objective is the creation of a catheter sufficiently firm to pass through the needle easily and advance a few centimeters into the epidural space. The catheter must flex to avoid tissue puncture. No objective criteria pairing clinical outcomes and formal mechanical testing exist to establish measures quantifying appropriate ranges of “firmness” or “flexibility” satisfying both ease of placement and minimization of patient risk. An ideal catheter should advance into and through the needle rather than flex excessively when resistance is encountered, and yet still bend and deflect off tissue. A goal is to establish a framework for evaluating catheter mechanical characteristics that can be directly correlated with measures of clinical performance to assess ease of use and safety. Interplay between evaluation and outcome can guide improvements in design and manufacturing that serve both clinician and patient needs. The mechanical behavior of a catheter bending during advancement beyond the tip of epidural needle is well described by the mechanics of buckling of a column loaded axially.3 The column (or
Regional Anesthesia and Pain Medicine, Vol 28, No 1 (January–February), 2003: pp 37–42
37
38
Regional Anesthesia and Pain Medicine Vol. 28 No. 1 January–February 2003
Fig 1. (A) Axial loading and (B) buckling of a slender column of length L. L, length; P, compressive axial force; Pcr , critical buckling load.
catheter) will buckle at its critical load, which depends on the column shape and its material properties. The Euler Buckling Formula is a mathematical expression for determining the maximum axial load that a long linear elastic column can withstand.4 Swiss mathematician, Leonard Euler (1707 to 1783), derived the formula, and it has broad applications in mechanical engineering and materials analysis. The aim of this study was to determine the range of critical loads borne by 6 types of epidural catheters (clear nylon, radiopaque nylon, coil reinforced, styletted clear nylon, styletted radiopaque nylon, and polyurethane) used commonly in clinical practice. We hypothesized that catheter material has a greater influence on bending stiffness than does cross-sectional shape. No prospective or retrospective data are available to compare complication rates associated with the catheters studied. The methods we used serve as a model for evaluation of catheter mechanical characteristics in relation to the study of clinical complications (e.g., insertion and migration) and in the development of catheters made of new materials or having a novel crosssectional shape.
Methods Figure 1A depicts a column (catheter) of length L subjected to a compressive axial force P. For a sufficiently large load, the column buckles (Fig 1B). The critical buckling load, Pcr, for a simply supported uniform column under end compressive loading is: P cr ⫽ 2 EI / L 2 (1) in which E is the Young’s modulus or modulus of elasticity of the material from which the column is made and I is the area moment of inertia of the column.3 The modulus of elasticity is the ratio of stress to strain for a slender rod undergoing tensile
stress. The larger the value of E, the less elastic is the material. The area moment of inertia measures the column’s ability to resist bending. The larger the value of I, the less the column bends when loaded. The relevant length of catheter that is loaded into compression during insertion is the length of catheter that is beyond the tip of the needle (i.e., the distance from the tip of the needle to the tissue encountered by the catheter tip). This is independent of catheter type (the clinical correlate is that needle placement into the epidural space does not depend on which catheter is being inserted after the needle is set). Considering a fixed value of L for all catheters simplifies the problem: Pcr is directly proportional to EI. Catheter buckling characteristics can be determined by evaluating both the material properties (i.e., measuring E) and the geometric characteristics (i.e., determining I). We studied 6 types of epidural catheters: clear nylon, clear nylon with a wire stylette, radiopaque nylon, radiopaque nylon with a stylette, polyurethane, and coil reinforced. Six different manufacturers produced the catheters: Becton Dickinson (Franklin Lakes, NJ), Arrow International (Reading, PA), B. Braun (Bethlehem, PA), Abbott Laboratories (North Chicago, IL), Rusch (Duluth, GA), and SIMS Portex (Keane, NH). We obtained 6 samples of each type of catheter. Catheters were stored and studied under the same environmental conditions and were not exposed to extremes of temperature or humidity. The elastic modulus is temperature dependent. Although catheter temperature does equilibrate with body temperature, all testing was performed at room temperature, 23 ⫾ 1°C, because this represents catheter temperature at the moment of insertion. Determination of E, Modulus of Elasticity Axial tensile load-deformation testing was conducted on 10-cm long sections of catheter clamped at least 1 cm away from the edge to minimize effects of any deformation caused by cutting. Testing was performed on a fully automated PC-controlled test frame (Model Sintech 20/G; MTS, Eden Prairie, MN) using a 2.23 kN (500 lb) load cell. The gauge length was 25.4 mm. Load and crosshead displacement were sampled at 3.3 Hz and recorded automatically (TestWorks software, MTS). The software calculates many common mechanical properties and automatically constructs stress-strain plots. The program determined E, the elastic modulus, from the slope of the initial linear elastic region of the stress-strain plot generated for each catheter sample tested. The crosshead speed was selected, in part, based
Epidural Catheter Bending Stiffness
•
David M. Eckmann
39
Table 1. Measured Values and Calculated Parameters for Catheters and Catheter/Stylette Assemblies Catheter Type
Outer Diameter D (mm)
Inner Diameter d (mm)
Moment of Inertia I ⫻ 102 (mm4)
Elastic Modulus E (MPa)
Bending Stiffness EI (N 䡠 mm2)
CR PU CN RN CNS RNS
1.092 ⫾ 0.002 0.838 ⫾ 0.003 0.864 ⫾ 0.004 0.829 ⫾ 0.002 0.835 ⫾ 0.002 0.834 ⫾ 0.004
0.467 ⫾ 0.003 0.426 ⫾ 0.002 0.421 ⫾ 0.001 0.423 ⫾ 0.001 0.424 ⫾ 0.001 0.423 ⫾ 0.001
6.75 ⫾ 0.05 2.26 ⫾ 0.03 2.58 ⫾ 0.05 2.16 ⫾ 0.03 2.25 ⫾ 0.03 2.25 ⫾ 0.05
7.26 ⫾ 0.39 48.8 ⫾ 1.9 45.6 ⫾ 3.1 74.4 ⫾ 3.8 2591 ⫾ 402 2780 ⫾ 466
0.49 ⫾ 0.03* 1.10 ⫾ 0.03 1.18 ⫾ 0.08 1.61 ⫾ 0.07* 58.3 ⫾ 9.5† 62.5 ⫾ 10.9†
NOTE. n ⫽ 6 catheters per group. Abbreviations: CR, coil reinforced; PU, polyurethane; CN, clear nylon; RN, radiopaque nylon; CNS, clear nylon styletted; RNS, radiopaque nylon styletted. *P ⬍ .05 compared with PU or CN. †P ⬍ .05 compared with CR, PU, CN, or RN.
on the elastic modulus of the material being tested. The speed was slow enough for the linear portion of the stress-strain relationship to be evident, but fast enough for the experiment to be completed in a reasonable time. Under these conditions, the values of E obtained are reliable and independent of the speed at which the sample is stretched. For testing sections of catheter alone, the crosshead speed was set at 50 mm/min. With the wire styletted catheter, the crosshead speed was reduced to 2.5 mm/min to accommodate the wire’s greater elastic modulus. Measurements were made on both styletted catheter sections containing the wire and on separate sections of the wire alone. This was done because the catheter/wire assembly occasionally slipped during testing and to identify the contribution of the wire to the overall bending stiffness. Determination of I, Area Moment of Inertia The area moment of inertia, I, is a geometrical property that depends on the beam’s cross-sectional shape. We assume that catheters are axisymmetric, that they have a uniform annular cross section, and that they bend about the centroid. Under these conditions, the area moment of inertia Ic for the catheter alone is given by I ⫽ Ic ⫽ (D4 ⫺ d4)/64, in which D and d are the outer and inner diameters of the catheter, respectively.4 Is for the stylette is calculated as Is ⫽ ds4/64, in which ds is the measured wire stylette diameter.4 For the catheter/stylette assembly, I ⫽ Ic ⫹ Is. This assumes the 2 bodies remain coaxial. A section of each catheter was cut with a scalpel. The cross-sectional shape was examined under a Wild M3Z stereomicroscope (Leica Microsystems AG, Wetzlar, Germany) fitted with a calibrated eyepiece graticule. D and d were measured and recorded. Values of ds for the stylettes were measured separately. Ic and Is were calculated for each sample examined. Values of EI were calculated for each
catheter or catheter plus stylette tested. No significant deformation of cross-sectional shape resulted from cutting. Values of the measured and calculated parameters for each of the 6 types of catheters and 2 types of wires studied were averaged (n ⫽ 6 per group) and the standard deviations were determined. Analysis of variance and post hoc Tukey’s t test were used for statistical analysis to identify differences in bending stiffness between groups. Results were considered statistically significant at P ⬍ .05.
Results Mean values and standard deviations of the measured catheter dimensions, Young’s modulus, calculated area moments of inertia, and bending stiffness appear in Table 1. The small standard deviation of the diameters is expected, with catheter production conforming to very specific manufacturing tolerances and quality control. Calculated values of I are clustered tightly around the mean for each catheter type. Values of I in Table 1 for the 2 types of styletted catheters include the stylette in the calculation. Separate values of Is are given in Table 2. The wire’s overall contribution to I is approximately 1%. The range of values of I (Table 1) is narrow because the catheters are nearly the same dimension, except the coil reinforced catheter whose outer diameter is about 25% larger than that of the other catheters tested. The catheter types have markedly different material properties. Values of E range from low (coil reinforced) to very high (both styletted catheters), with 3 intermediate results (clear nylon, radiopaque nylon, polyurethane). Bending stiffness, EI, was no different (P ⬎ .05) between the clear nylon and polyurethane catheters (Table 1). In comparison to these groups, EI was significantly lower (P ⬍ .05) for the coil reinforced group and significantly
40
Regional Anesthesia and Pain Medicine Vol. 28 No. 1 January–February 2003 Table 2. Measured Values and Calculated Parameters for Catheter Wire Stylettes
Stylette
Stylette Diameter ds (mm)
Moment of Inertia Is ⫻ 104 (mm4)
Elastic Modulus Es ⫻ 10⫺5 (MPa)
Bending Stiffness EsIs (N 䡠 mm2)
CNS RNS
0.258 ⫾ 0.001 0.268 ⫾ 0.001
2.17 ⫾ 0.03 2.54 ⫾ 0.03
1.77 ⫾ 0.10 1.72 ⫾ 0.10
38.4 ⫾ 2.3 43.7 ⫾ 2.5
NOTE. n ⫽ 6 wire stylettes per group.
higher (P ⬍ .05) for the radiopaque nylon group. This primarily reflects a material effect (influence of E) rather than a geometric effect (influence of I). The values of EI obtained for the styletted catheters were 35 to 125 times larger than the values determined for the other 4 groups (Table 1). This is due to the inelastic wire, although this result may not accurately describe the composite catheter/wire assembly. If the relatively inelastic wire dominates the elastic behavior, then the appropriate value of I for calculating the bending stiffness is Is. If the catheter material, compared with the wire, contributes minimally to E, then I for the composite is too large to determine the bending stiffness accurately. The values in Table 1 are actually the upper limit of the bending stiffness. Separate calculations of EsIs (Table 2) isolate the contributions of the stylettes. These values are ⬃ 30% to 35% smaller than those of the catheter/wire assemblies and can be treated as the lower limit of the bending stiffness. These values are still 23 to 90 times greater than the values obtained for the other groups.
Discussion The multiple types of epidural catheters available to clinicians should be used with an eye kept toward improving catheter design for safer and more effective patient care. It is important to establish methods of assessing catheter performance to interpret clinical complications associated with epidural catheter use. Complications can occur at any time from insertion to removal. Catheters can kink,5 occlude,6 or knot7 to prevent epidural drug delivery. Prior work on catheter mechanical behavior has focused on difficult catheter removal, breakage, and retention.8-10 One recent study shows that the modulus of elasticity, E, contributes greatly to catheter breakage and retention.11 Those studies were motivated by case reports, but the material properties described relate catheter design and manufacture with clinical outcomes. Interpreting more frequent complications, such as rates of intravenous insertion or failure rates in threading a catheter, is not currently possible because of 2 major limitations in correlating clinical and labora-
tory studies. One is the absence of prospective or retrospective clinical studies quantifying complication rates associated with specific catheters. The other is the lack of appropriate metrics for assessing designs that contribute to complications. An aspect of this is the limited understanding of the biomechanics of tissues at risk to be punctured. A recent tissue mechanics study revealed the tangential (shear) stress properties of the lumbar dura mater.12 No studies report measures of the perpendicular (normal) stress properties important in tissue “failure” occurring with puncture. The incidences of epidurovasal (11.5%) and epidurosubarachnoid (0.9%) malpositioning of epidural catheters have been reported.13 Case reports document delayed subdural14,15 and subarachnoid16 catheter migration. Burstal et al.2 reported 3 occurrences of intravenous catheter migration and 1 occurrence of subarachnoid catheter migration in 1,291 continuous epidurals. The type of catheter used was not reported. Holmstrom et al.17 found in cadavers that unstyletted catheters perforated the dura in 45% of cases after a single dural puncture with an 18-gauge Tuohy needle and in 5% of cases if the tip was guided toward a cluster of multiple dural punctures made with a 25-gauge spinal needle. It is unknown if the complication rates reported would have been different using a catheter with a larger or smaller bending stiffness or in the presence of a stylette. The analysis presented establishes the mechanical basis for the manufacturer’s recommendation of withdrawing the stylette a few centimeters before advancing the catheter. The stylette markedly increased the bending stiffness ⬃ 25-fold to 30-fold (compare the EI results for unstyletted clear nylon and radiopaque nylon catheters in Table 1 to the stylette wire results in Table 2). This only applies to the portion of the catheter in which the stylette resides. Withdrawing the stylette dramatically lowers the bending stiffness. This simple maneuver reduces the potential for penetrating tissue trauma or catheter tip placement in an undesirable location. As a safe practice guideline, only the catheter, and
Epidural Catheter Bending Stiffness
not the rigid wire stylette, should be advanced beyond the tip of the needle. The wider range and relatively large standard deviations reported for E compared with the data for I likely reflects slight variation in batches of material used for day-to-day catheter manufacturing. This finding may also incorporate the thermal properties of the different materials in the range of temperatures at which they were studied. Ideal sample gripping during the elastic modulus testing would have created a no-slip interface between the wire and the catheter. The evidence was that the wire did slip inside the catheter. This was anticipated because the wire diameter and the catheter inner diameter were not perfectly matched. Furthermore, the friction between the surfaces is low to facilitate stylette removal. If perfect gripping were achieved, the value of E obtained would reflect a metal sample and not a plastic or polymer compound. The values obtained (Table 1) are too large to represent the catheter material alone and too small to represent a metal. Tests performed on stylettes alone result in values of Es consistent with a wire specimen (Table 2). The range of values of I in Table 1 is narrow, indicating that E dominates the large range EI reported (Tables 1 and 2). Designs that reduce I include making a thinner wall, but this may increase kinking or contribute to difficulty advancing a catheter beyond the tip of the needle.18 Another design tactic is to reduce E by using a more elastic material. The coil reinforced catheter accomplishes this. The inclusion of the coiled wire suggests that without its additional mechanical support to stent the catheter open, it might kink easily or be difficult to advance. The cost of lowering the bending stiffness is the extra need for the coiled wire, which adds complexity and expense to their manufacture. Very flexible catheters, such as this coil-reinforced device, have been shown to coil readily with as little as 1 cm of length inserted.19 Catheters having a smaller bending stiffness will be difficult to thread and should be avoided. This study demonstrates how catheter cross-sectional shape and material properties influence the bending stiffness. This is likely an important determinant of catheter safety and effectiveness. Formal methods to assess catheters mechanically will be important to interpret clinical outcomes as new materials and designs are introduced. Improvements in catheter technology may increase patient safety and add to the likelihood of successful insertion, maintenance, and removal of catheters from the epidural space. The methods of testing and analysis used in this study are easily applied and can be
•
David M. Eckmann
41
used to assess relative catheter stiffness, one important aspect of catheter design. Catheters currently available establish the range of bending stiffness that should not be exceeded, only optimized to clinical outcome.
References 1. Wheatley RG, Schug SA, Watson D. Safety and efficacy of postoperative epidural analgesia. Br J Anaesth 2001;87:47-61. 2. Burstal R, Wegener F, Hayes C, Lantry G. Epidural analgesia: Prospective audit of 1062 patients. Anaesth Intensive Care 1998;26:165-172. 3. Singer J, Arbocz J. Buckling Experiments: Experimental Methods in Buckling of Thin-Walled Structures, vol 1, Basic Concepts, Columns, Beams, and Plates. New York, NY: Wiley; 1998. 4. Boresi AP, Schmidt RJ, Sidebottom OM. Advanced Mechanics of Materials. New York, NY: Wiley; 1992. 5. Sage FJ, Lloyd Thomas AR, Howard RF. Paediatric lumbar epidurals: A comparison of 21-G and 23-G catheters in patients weighing less than 10 kg. Paediatr Anaesth 2000;10:279-282. 6. Gramling-Babb P, Miller MJ, Reeves ST, Roy RC, Zile MR. Treatment of medically and surgically refractory angina pectoris with high thoracic epidural analgesia: Initial clinical experience. Am Heart J 1997;133:648655. 7. Browne RA, Politi VL. Knotting of an epidural catheter: A case report. Can Anaesth Soc J 1979;26:142144. 8. Morris GN, Warren BB, Hanson EW, Mazzeo FJ, DiBenedetto DJ. Influence of patient position on withdrawal forces during removal of lumbar extradural catheters. Br J Anaesth 1996;77:419-420. 9. Boey SK, Carrie LE. Withdrawal forces during removal of lumbar extradural catheters. Br J Anaesth 1994;73:833-835. 10. Asai T, Yamamoto K, Hirose T, Taguchi H, Shingu K. Breakage of epidural catheters: A comparison of an arrow reinforced catheter and other nonreinforced catheters. Anesth Analg 2001;92:246-248. 11. Ates Y, Yucesoy CA, Unlu MA, Saygin B, Akkas N. The mechanical properties of intact and traumatized epidural catheters. Anesth Analg 2000;90:393-399. 12. Runza M, Pietrabissa R, Mantero S, Albani A, Quaglini V, Contro R. Lumbar dura mater biomechanics: Experimental characterization and scanning electron microscopy observations. Anesth Analg 1999;88: 1317-1321. 13. Beck H, Brassow F, Doehn M, Bause H, Dziadzka A, Schulte am EJ. Epidural catheters of the multi-orifice type: Dangers and complications. Acta Anaesth Scand 1986;30:549-555. 14. Forrester DJ, Mukherji SK, Mayer DC, Spielman FJ. Dilute infusion for labor, obscure subdural catheter, and life-threatening block at cesarean delivery. Anesth Analg 1999;89:1267-1268.
42
Regional Anesthesia and Pain Medicine Vol. 28 No. 1 January–February 2003
15. Hartrick CT, Pither CE, Pai U, Raj PP, Tomsick TA. Subdural migration of an epidural catheter. Anesth Analg 1985;64:175-178. 16. Barnes RK. Delayed subarachnoid migration of an epidural catheter. Anaesth Intensive Care 1990;18:564566. 17. Holmstrom B, Rawal N, Axelsson K, Nydahl PA. Risk of catheter migration during combined spinal epi-
dural block: Percutaneous epiduroscopy study. Anesth Analg 1995;80:747-753. 18. Seitman DT, Shapiro BE. Inability to thread epidural catheter through epidural needle. Anesth Analg 1989; 69:267-268. 19. Lim YJ, Bahk JH, Ahn WS, Lee SC. Coiling of lumbar epidural catheters. Acta Anaesthesiol Scand 2002;46: 603-606.
Bending stiffness, Catheter: Epidural, Mechanics: Buckling, Modulus of elasticity.
I
ntroduction of an epidural catheter can be complicated by intravascular or subarachnoid placement of the catheter tip.1 A catheter may also migrate from the epidural space into an epidural vein or into the subarachnoid space.2 Thus, local anesthetics unexpectedly may be delivered directly into the bloodstream or the cerebrospinal fluid, causing drug toxicity (seizure or cardiac arrhythmia) or sudden high spinal anesthesia. Unintended catheter penetration through a blood vessel wall or the meninges also increases the risks of bleeding, develop-
From the Department of Anesthesia, University of Pennsylvania Health System, Philadelphia, Pennsylvania and Institute of Medicine and Engineering, University of Pennsylvania, Philadelphia, Pennsylvania. Accepted for publication November 1, 2002. This work should be attributed to the University of Pennsylvania Department of Anesthesia. Reprint requests: David M. Eckmann, Ph.D., M.D., Department of Anesthesia, The University of Pennsylvania, 3400 Spruce Str, Philadelphia, PA 19104. E-mail: eckmanndm@ uphs.upenn.edu © 2003 by the American Society of Regional Anesthesia and Pain Medicine. 1098-7339/03/2801-0008$35.00/0 doi:10.1053/rapm.2003.50016
ment of a postdural puncture headache, or traumatic penetrating injury to other tissues. A design and manufacture objective is the creation of a catheter sufficiently firm to pass through the needle easily and advance a few centimeters into the epidural space. The catheter must flex to avoid tissue puncture. No objective criteria pairing clinical outcomes and formal mechanical testing exist to establish measures quantifying appropriate ranges of “firmness” or “flexibility” satisfying both ease of placement and minimization of patient risk. An ideal catheter should advance into and through the needle rather than flex excessively when resistance is encountered, and yet still bend and deflect off tissue. A goal is to establish a framework for evaluating catheter mechanical characteristics that can be directly correlated with measures of clinical performance to assess ease of use and safety. Interplay between evaluation and outcome can guide improvements in design and manufacturing that serve both clinician and patient needs. The mechanical behavior of a catheter bending during advancement beyond the tip of epidural needle is well described by the mechanics of buckling of a column loaded axially.3 The column (or
Regional Anesthesia and Pain Medicine, Vol 28, No 1 (January–February), 2003: pp 37–42
37
38
Regional Anesthesia and Pain Medicine Vol. 28 No. 1 January–February 2003
Fig 1. (A) Axial loading and (B) buckling of a slender column of length L. L, length; P, compressive axial force; Pcr , critical buckling load.
catheter) will buckle at its critical load, which depends on the column shape and its material properties. The Euler Buckling Formula is a mathematical expression for determining the maximum axial load that a long linear elastic column can withstand.4 Swiss mathematician, Leonard Euler (1707 to 1783), derived the formula, and it has broad applications in mechanical engineering and materials analysis. The aim of this study was to determine the range of critical loads borne by 6 types of epidural catheters (clear nylon, radiopaque nylon, coil reinforced, styletted clear nylon, styletted radiopaque nylon, and polyurethane) used commonly in clinical practice. We hypothesized that catheter material has a greater influence on bending stiffness than does cross-sectional shape. No prospective or retrospective data are available to compare complication rates associated with the catheters studied. The methods we used serve as a model for evaluation of catheter mechanical characteristics in relation to the study of clinical complications (e.g., insertion and migration) and in the development of catheters made of new materials or having a novel crosssectional shape.
Methods Figure 1A depicts a column (catheter) of length L subjected to a compressive axial force P. For a sufficiently large load, the column buckles (Fig 1B). The critical buckling load, Pcr, for a simply supported uniform column under end compressive loading is: P cr ⫽ 2 EI / L 2 (1) in which E is the Young’s modulus or modulus of elasticity of the material from which the column is made and I is the area moment of inertia of the column.3 The modulus of elasticity is the ratio of stress to strain for a slender rod undergoing tensile
stress. The larger the value of E, the less elastic is the material. The area moment of inertia measures the column’s ability to resist bending. The larger the value of I, the less the column bends when loaded. The relevant length of catheter that is loaded into compression during insertion is the length of catheter that is beyond the tip of the needle (i.e., the distance from the tip of the needle to the tissue encountered by the catheter tip). This is independent of catheter type (the clinical correlate is that needle placement into the epidural space does not depend on which catheter is being inserted after the needle is set). Considering a fixed value of L for all catheters simplifies the problem: Pcr is directly proportional to EI. Catheter buckling characteristics can be determined by evaluating both the material properties (i.e., measuring E) and the geometric characteristics (i.e., determining I). We studied 6 types of epidural catheters: clear nylon, clear nylon with a wire stylette, radiopaque nylon, radiopaque nylon with a stylette, polyurethane, and coil reinforced. Six different manufacturers produced the catheters: Becton Dickinson (Franklin Lakes, NJ), Arrow International (Reading, PA), B. Braun (Bethlehem, PA), Abbott Laboratories (North Chicago, IL), Rusch (Duluth, GA), and SIMS Portex (Keane, NH). We obtained 6 samples of each type of catheter. Catheters were stored and studied under the same environmental conditions and were not exposed to extremes of temperature or humidity. The elastic modulus is temperature dependent. Although catheter temperature does equilibrate with body temperature, all testing was performed at room temperature, 23 ⫾ 1°C, because this represents catheter temperature at the moment of insertion. Determination of E, Modulus of Elasticity Axial tensile load-deformation testing was conducted on 10-cm long sections of catheter clamped at least 1 cm away from the edge to minimize effects of any deformation caused by cutting. Testing was performed on a fully automated PC-controlled test frame (Model Sintech 20/G; MTS, Eden Prairie, MN) using a 2.23 kN (500 lb) load cell. The gauge length was 25.4 mm. Load and crosshead displacement were sampled at 3.3 Hz and recorded automatically (TestWorks software, MTS). The software calculates many common mechanical properties and automatically constructs stress-strain plots. The program determined E, the elastic modulus, from the slope of the initial linear elastic region of the stress-strain plot generated for each catheter sample tested. The crosshead speed was selected, in part, based
Epidural Catheter Bending Stiffness
•
David M. Eckmann
39
Table 1. Measured Values and Calculated Parameters for Catheters and Catheter/Stylette Assemblies Catheter Type
Outer Diameter D (mm)
Inner Diameter d (mm)
Moment of Inertia I ⫻ 102 (mm4)
Elastic Modulus E (MPa)
Bending Stiffness EI (N 䡠 mm2)
CR PU CN RN CNS RNS
1.092 ⫾ 0.002 0.838 ⫾ 0.003 0.864 ⫾ 0.004 0.829 ⫾ 0.002 0.835 ⫾ 0.002 0.834 ⫾ 0.004
0.467 ⫾ 0.003 0.426 ⫾ 0.002 0.421 ⫾ 0.001 0.423 ⫾ 0.001 0.424 ⫾ 0.001 0.423 ⫾ 0.001
6.75 ⫾ 0.05 2.26 ⫾ 0.03 2.58 ⫾ 0.05 2.16 ⫾ 0.03 2.25 ⫾ 0.03 2.25 ⫾ 0.05
7.26 ⫾ 0.39 48.8 ⫾ 1.9 45.6 ⫾ 3.1 74.4 ⫾ 3.8 2591 ⫾ 402 2780 ⫾ 466
0.49 ⫾ 0.03* 1.10 ⫾ 0.03 1.18 ⫾ 0.08 1.61 ⫾ 0.07* 58.3 ⫾ 9.5† 62.5 ⫾ 10.9†
NOTE. n ⫽ 6 catheters per group. Abbreviations: CR, coil reinforced; PU, polyurethane; CN, clear nylon; RN, radiopaque nylon; CNS, clear nylon styletted; RNS, radiopaque nylon styletted. *P ⬍ .05 compared with PU or CN. †P ⬍ .05 compared with CR, PU, CN, or RN.
on the elastic modulus of the material being tested. The speed was slow enough for the linear portion of the stress-strain relationship to be evident, but fast enough for the experiment to be completed in a reasonable time. Under these conditions, the values of E obtained are reliable and independent of the speed at which the sample is stretched. For testing sections of catheter alone, the crosshead speed was set at 50 mm/min. With the wire styletted catheter, the crosshead speed was reduced to 2.5 mm/min to accommodate the wire’s greater elastic modulus. Measurements were made on both styletted catheter sections containing the wire and on separate sections of the wire alone. This was done because the catheter/wire assembly occasionally slipped during testing and to identify the contribution of the wire to the overall bending stiffness. Determination of I, Area Moment of Inertia The area moment of inertia, I, is a geometrical property that depends on the beam’s cross-sectional shape. We assume that catheters are axisymmetric, that they have a uniform annular cross section, and that they bend about the centroid. Under these conditions, the area moment of inertia Ic for the catheter alone is given by I ⫽ Ic ⫽ (D4 ⫺ d4)/64, in which D and d are the outer and inner diameters of the catheter, respectively.4 Is for the stylette is calculated as Is ⫽ ds4/64, in which ds is the measured wire stylette diameter.4 For the catheter/stylette assembly, I ⫽ Ic ⫹ Is. This assumes the 2 bodies remain coaxial. A section of each catheter was cut with a scalpel. The cross-sectional shape was examined under a Wild M3Z stereomicroscope (Leica Microsystems AG, Wetzlar, Germany) fitted with a calibrated eyepiece graticule. D and d were measured and recorded. Values of ds for the stylettes were measured separately. Ic and Is were calculated for each sample examined. Values of EI were calculated for each
catheter or catheter plus stylette tested. No significant deformation of cross-sectional shape resulted from cutting. Values of the measured and calculated parameters for each of the 6 types of catheters and 2 types of wires studied were averaged (n ⫽ 6 per group) and the standard deviations were determined. Analysis of variance and post hoc Tukey’s t test were used for statistical analysis to identify differences in bending stiffness between groups. Results were considered statistically significant at P ⬍ .05.
Results Mean values and standard deviations of the measured catheter dimensions, Young’s modulus, calculated area moments of inertia, and bending stiffness appear in Table 1. The small standard deviation of the diameters is expected, with catheter production conforming to very specific manufacturing tolerances and quality control. Calculated values of I are clustered tightly around the mean for each catheter type. Values of I in Table 1 for the 2 types of styletted catheters include the stylette in the calculation. Separate values of Is are given in Table 2. The wire’s overall contribution to I is approximately 1%. The range of values of I (Table 1) is narrow because the catheters are nearly the same dimension, except the coil reinforced catheter whose outer diameter is about 25% larger than that of the other catheters tested. The catheter types have markedly different material properties. Values of E range from low (coil reinforced) to very high (both styletted catheters), with 3 intermediate results (clear nylon, radiopaque nylon, polyurethane). Bending stiffness, EI, was no different (P ⬎ .05) between the clear nylon and polyurethane catheters (Table 1). In comparison to these groups, EI was significantly lower (P ⬍ .05) for the coil reinforced group and significantly
40
Regional Anesthesia and Pain Medicine Vol. 28 No. 1 January–February 2003 Table 2. Measured Values and Calculated Parameters for Catheter Wire Stylettes
Stylette
Stylette Diameter ds (mm)
Moment of Inertia Is ⫻ 104 (mm4)
Elastic Modulus Es ⫻ 10⫺5 (MPa)
Bending Stiffness EsIs (N 䡠 mm2)
CNS RNS
0.258 ⫾ 0.001 0.268 ⫾ 0.001
2.17 ⫾ 0.03 2.54 ⫾ 0.03
1.77 ⫾ 0.10 1.72 ⫾ 0.10
38.4 ⫾ 2.3 43.7 ⫾ 2.5
NOTE. n ⫽ 6 wire stylettes per group.
higher (P ⬍ .05) for the radiopaque nylon group. This primarily reflects a material effect (influence of E) rather than a geometric effect (influence of I). The values of EI obtained for the styletted catheters were 35 to 125 times larger than the values determined for the other 4 groups (Table 1). This is due to the inelastic wire, although this result may not accurately describe the composite catheter/wire assembly. If the relatively inelastic wire dominates the elastic behavior, then the appropriate value of I for calculating the bending stiffness is Is. If the catheter material, compared with the wire, contributes minimally to E, then I for the composite is too large to determine the bending stiffness accurately. The values in Table 1 are actually the upper limit of the bending stiffness. Separate calculations of EsIs (Table 2) isolate the contributions of the stylettes. These values are ⬃ 30% to 35% smaller than those of the catheter/wire assemblies and can be treated as the lower limit of the bending stiffness. These values are still 23 to 90 times greater than the values obtained for the other groups.
Discussion The multiple types of epidural catheters available to clinicians should be used with an eye kept toward improving catheter design for safer and more effective patient care. It is important to establish methods of assessing catheter performance to interpret clinical complications associated with epidural catheter use. Complications can occur at any time from insertion to removal. Catheters can kink,5 occlude,6 or knot7 to prevent epidural drug delivery. Prior work on catheter mechanical behavior has focused on difficult catheter removal, breakage, and retention.8-10 One recent study shows that the modulus of elasticity, E, contributes greatly to catheter breakage and retention.11 Those studies were motivated by case reports, but the material properties described relate catheter design and manufacture with clinical outcomes. Interpreting more frequent complications, such as rates of intravenous insertion or failure rates in threading a catheter, is not currently possible because of 2 major limitations in correlating clinical and labora-
tory studies. One is the absence of prospective or retrospective clinical studies quantifying complication rates associated with specific catheters. The other is the lack of appropriate metrics for assessing designs that contribute to complications. An aspect of this is the limited understanding of the biomechanics of tissues at risk to be punctured. A recent tissue mechanics study revealed the tangential (shear) stress properties of the lumbar dura mater.12 No studies report measures of the perpendicular (normal) stress properties important in tissue “failure” occurring with puncture. The incidences of epidurovasal (11.5%) and epidurosubarachnoid (0.9%) malpositioning of epidural catheters have been reported.13 Case reports document delayed subdural14,15 and subarachnoid16 catheter migration. Burstal et al.2 reported 3 occurrences of intravenous catheter migration and 1 occurrence of subarachnoid catheter migration in 1,291 continuous epidurals. The type of catheter used was not reported. Holmstrom et al.17 found in cadavers that unstyletted catheters perforated the dura in 45% of cases after a single dural puncture with an 18-gauge Tuohy needle and in 5% of cases if the tip was guided toward a cluster of multiple dural punctures made with a 25-gauge spinal needle. It is unknown if the complication rates reported would have been different using a catheter with a larger or smaller bending stiffness or in the presence of a stylette. The analysis presented establishes the mechanical basis for the manufacturer’s recommendation of withdrawing the stylette a few centimeters before advancing the catheter. The stylette markedly increased the bending stiffness ⬃ 25-fold to 30-fold (compare the EI results for unstyletted clear nylon and radiopaque nylon catheters in Table 1 to the stylette wire results in Table 2). This only applies to the portion of the catheter in which the stylette resides. Withdrawing the stylette dramatically lowers the bending stiffness. This simple maneuver reduces the potential for penetrating tissue trauma or catheter tip placement in an undesirable location. As a safe practice guideline, only the catheter, and
Epidural Catheter Bending Stiffness
not the rigid wire stylette, should be advanced beyond the tip of the needle. The wider range and relatively large standard deviations reported for E compared with the data for I likely reflects slight variation in batches of material used for day-to-day catheter manufacturing. This finding may also incorporate the thermal properties of the different materials in the range of temperatures at which they were studied. Ideal sample gripping during the elastic modulus testing would have created a no-slip interface between the wire and the catheter. The evidence was that the wire did slip inside the catheter. This was anticipated because the wire diameter and the catheter inner diameter were not perfectly matched. Furthermore, the friction between the surfaces is low to facilitate stylette removal. If perfect gripping were achieved, the value of E obtained would reflect a metal sample and not a plastic or polymer compound. The values obtained (Table 1) are too large to represent the catheter material alone and too small to represent a metal. Tests performed on stylettes alone result in values of Es consistent with a wire specimen (Table 2). The range of values of I in Table 1 is narrow, indicating that E dominates the large range EI reported (Tables 1 and 2). Designs that reduce I include making a thinner wall, but this may increase kinking or contribute to difficulty advancing a catheter beyond the tip of the needle.18 Another design tactic is to reduce E by using a more elastic material. The coil reinforced catheter accomplishes this. The inclusion of the coiled wire suggests that without its additional mechanical support to stent the catheter open, it might kink easily or be difficult to advance. The cost of lowering the bending stiffness is the extra need for the coiled wire, which adds complexity and expense to their manufacture. Very flexible catheters, such as this coil-reinforced device, have been shown to coil readily with as little as 1 cm of length inserted.19 Catheters having a smaller bending stiffness will be difficult to thread and should be avoided. This study demonstrates how catheter cross-sectional shape and material properties influence the bending stiffness. This is likely an important determinant of catheter safety and effectiveness. Formal methods to assess catheters mechanically will be important to interpret clinical outcomes as new materials and designs are introduced. Improvements in catheter technology may increase patient safety and add to the likelihood of successful insertion, maintenance, and removal of catheters from the epidural space. The methods of testing and analysis used in this study are easily applied and can be
•
David M. Eckmann
41
used to assess relative catheter stiffness, one important aspect of catheter design. Catheters currently available establish the range of bending stiffness that should not be exceeded, only optimized to clinical outcome.
References 1. Wheatley RG, Schug SA, Watson D. Safety and efficacy of postoperative epidural analgesia. Br J Anaesth 2001;87:47-61. 2. Burstal R, Wegener F, Hayes C, Lantry G. Epidural analgesia: Prospective audit of 1062 patients. Anaesth Intensive Care 1998;26:165-172. 3. Singer J, Arbocz J. Buckling Experiments: Experimental Methods in Buckling of Thin-Walled Structures, vol 1, Basic Concepts, Columns, Beams, and Plates. New York, NY: Wiley; 1998. 4. Boresi AP, Schmidt RJ, Sidebottom OM. Advanced Mechanics of Materials. New York, NY: Wiley; 1992. 5. Sage FJ, Lloyd Thomas AR, Howard RF. Paediatric lumbar epidurals: A comparison of 21-G and 23-G catheters in patients weighing less than 10 kg. Paediatr Anaesth 2000;10:279-282. 6. Gramling-Babb P, Miller MJ, Reeves ST, Roy RC, Zile MR. Treatment of medically and surgically refractory angina pectoris with high thoracic epidural analgesia: Initial clinical experience. Am Heart J 1997;133:648655. 7. Browne RA, Politi VL. Knotting of an epidural catheter: A case report. Can Anaesth Soc J 1979;26:142144. 8. Morris GN, Warren BB, Hanson EW, Mazzeo FJ, DiBenedetto DJ. Influence of patient position on withdrawal forces during removal of lumbar extradural catheters. Br J Anaesth 1996;77:419-420. 9. Boey SK, Carrie LE. Withdrawal forces during removal of lumbar extradural catheters. Br J Anaesth 1994;73:833-835. 10. Asai T, Yamamoto K, Hirose T, Taguchi H, Shingu K. Breakage of epidural catheters: A comparison of an arrow reinforced catheter and other nonreinforced catheters. Anesth Analg 2001;92:246-248. 11. Ates Y, Yucesoy CA, Unlu MA, Saygin B, Akkas N. The mechanical properties of intact and traumatized epidural catheters. Anesth Analg 2000;90:393-399. 12. Runza M, Pietrabissa R, Mantero S, Albani A, Quaglini V, Contro R. Lumbar dura mater biomechanics: Experimental characterization and scanning electron microscopy observations. Anesth Analg 1999;88: 1317-1321. 13. Beck H, Brassow F, Doehn M, Bause H, Dziadzka A, Schulte am EJ. Epidural catheters of the multi-orifice type: Dangers and complications. Acta Anaesth Scand 1986;30:549-555. 14. Forrester DJ, Mukherji SK, Mayer DC, Spielman FJ. Dilute infusion for labor, obscure subdural catheter, and life-threatening block at cesarean delivery. Anesth Analg 1999;89:1267-1268.
42
Regional Anesthesia and Pain Medicine Vol. 28 No. 1 January–February 2003
15. Hartrick CT, Pither CE, Pai U, Raj PP, Tomsick TA. Subdural migration of an epidural catheter. Anesth Analg 1985;64:175-178. 16. Barnes RK. Delayed subarachnoid migration of an epidural catheter. Anaesth Intensive Care 1990;18:564566. 17. Holmstrom B, Rawal N, Axelsson K, Nydahl PA. Risk of catheter migration during combined spinal epi-
dural block: Percutaneous epiduroscopy study. Anesth Analg 1995;80:747-753. 18. Seitman DT, Shapiro BE. Inability to thread epidural catheter through epidural needle. Anesth Analg 1989; 69:267-268. 19. Lim YJ, Bahk JH, Ahn WS, Lee SC. Coiling of lumbar epidural catheters. Acta Anaesthesiol Scand 2002;46: 603-606.