Originally published February 1996
Peter Kleinhenz and Christine Vogdes
Used in conjunction with an electrosurgical unit (ESU) that supplies the necessary power, electrosurgical devices are a routine part of both laparoscopic and open surgical procedures. These devices can cut, cauterize, and coagulate tissue by means of radio- frequency (RF) electrical energy. Typically, some form of insulation is needed to ensure that energy is directed at the target tissue. Heat-shrink tubing is a preferred method of providing such insulation on electrosurgical tools, particularly on high-volume disposable products such as graspers, scissors, hooks, and probes. Unfortunately, defects in the insulation can allow the full amount of RF energy to be applied in unintended and unseen areas, causing burns or cuts that may not be detected at the time of surgery.
The potential for the unintended and unseen application of RF energy is especially great in laparoscopic surgeries because the surgeon's field of view is restricted to the viewing angle of the laparoscope. Most laparoscopic instruments are approximately 35 cm long and the images viewed on the monitor show less than 5 cm of the distal end of the device. Although the electrode used to deliver RF energy has an insulated covering, 90% or more of this insulation is outside the user's viewing range.1 It has been reported that the incidence of recognized injuries for laparoscopic surgeries is between one and two patients per 1000 operations.2 In addition, several medical journals have indicated that the majority of such injuries go unrecognized at the time of surgery, either because the affected site was not observed during the procedure or because there was no immediate clinical evidence of the in- jury--the time from injury to onset of symp-toms can vary from 18 hours to 14 days.35
When injuries go unnoticed during surgery, the reasons for their occurrence can only be inferred. The most common explanations include: inadvertent grasping or touching of tissue while the device is energized, direct coupling of the device to surrounding tissue while the device is energized, and insulation failures or defects.6 One clinical study has concluded that the greatest contribution to accidental injury is probably inappropriate or inadequate insulation.2
The potential for insulation-related injury can be minimized by selecting insulating materials based on the results of standard tests. With the goal of improving insulation quality, the authors' companies worked together to create a test apparatus and a test protocol to evaluate various shrink tubing materials used by the electrosurgical product industry. The results of that collaboration are reported below.
THE HF18 STANDARD FOR ELECTROSURGICAL INSTRUMENTS
The Association for the Advancement of Medical Instrumentation (AAMI) has established minimum safety and performance requirements for electrosurgical devices, which were adopted by the American National Standards Institute (ANSI) and are detailed in the document designated ANSI/AAMI HF18-1993.7 Because the collaborative project began in 1993, prior to the latest revision of this standard, and concluded in early 1995, the experimental test methodology and protocol was based on both HF18-1993 and HF18-1986.8
In paragraph 22.214.171.124 (6), the HF18-1986 standard requires that insulated shafts pass a 30-second test at 4000 V peak to peak at 1 MHz. This combination of high voltage and high frequency represents a very severe challenge and therefore provides a high margin of safety. A typical ESU operates in the 3000-V range at 500 kHz. One of the difficulties in applying HF18-1986 has been that there was no standard test equipment that operates in the suggested 4000-V, 1-MHz range. Consequently, there has been some confusion on the part of instrument manufacturers on how to comply with this requirement. It was in order to resolve some of this confusion that Raychem, a supplier of insulating shrink tubing, decided to undertake a benchmark study of various competitive materials and their ability to comply with HF18-1986. Progenics, an experienced designer of ESU devices, was selected to build the test apparatus and generator. In Phase I of the study the key questions addressed were: (1) Which of the commonly used insulating materials meets the AAMI standard? (2) What is the effect of varying the wall thickness of an insulating material? and (3) What is the effect of having air rather than normal saline surrounding the material during testing?
Based on the results of Phase I, a second-phase test plan was developed to study the effects of multiple energizing cycles, gamma and ethylene oxide (EtO) sterilization, installation errors, and stresses created by a step transition. Compared with the first-phase efforts, this expanded test plan more closely paralleled the potential stresses that insulating materials encounter during instrument manufacture and use. While the test program was under way, the 1993 revisions to HF18 were issued. Paragraph 126.96.36.199 changed the output-voltage test requirement to 1.5 times the output voltage at the specified operating frequency of the ESU. Therefore, Progenics selected 6000 V peak to peak as the new voltage setting of the test apparatus because that is 1.5 times the 4000 V recommended for the test generator. A 500-kHz frequency was selected as representative of the average operating frequency for an ESU.
TEST METHODS AND RESULTS
The test generator and apparatus were designed to meet the test specifications in the 1986 HF18 standard, which called for laying an insulated mandrel on a ground plane and testing the sample in air or with a saline-soaked sponge covering it. The associated test equipment includes power supplies, a 1-MHz amplifier with a 4000-V output transformer, and an oscilloscope to measure current and voltage (see Figure 1).
During both phases of testing, the shrink tubing was installed on steel mandrels with a heat gun to remove entrapped air and then finished with a 3-minute oven soak. In Phase I tests, single and multiple layers of material were installed to test the effect of varying the material thickness. Because HF18 requires high-voltage testing, the samples were slowly powered to the peak voltage and then held at that voltage for the required 30 seconds. For all test runs, observations were recorded on the presence and location of a luminous discharge (corona) at the surface of the insulating material and the electrical conductor. The voltage at breakdown and the time to failure were also recorded.
Phase I. The Phase I tests evaluated six different insulating materials from various manufacturers: polyvinylidene fluoride (PVDF), low-density polyethylene (LDPE), a blend of polyolefin and ethylene acrylic acid copolymer that is partially zinc or sodium neutralized (ionomer), high-density polyethylene (HDPE), two types of fluorinated ethylene propylene (FEP), and polyvinyl chloride (PVC). For each material, 12 or 15 samples (3 each of 4 or 5 thicknesses between 0.008 and 0.016 in.) were tested at 4000 V peak to peak, 1 MHz in both air and normal saline.
The results of those initial tests are shown in Table I. The data indicate that the PVC and PVDF materials did not pass the 4000-V, 1-MHz test even at 0.016 in., the largest thickness tested, and the other materials required a thickness of 0.015 in. or greater to pass. There was no significant difference in the results when air and saline were used as the media surrounding the test materials. If a material passed the test in air, it was also a good insulator in normal saline. This capability is important because electrosurgical instruments come in contact with such saline formulations as blood and human tissue and certain bipolar instruments are used specifically in conjunction with saline washes.
The test results were found to be consistent with the known dielectric properties of the individual materials. In addition, they were consistent from sample to sample within a group, which indicated the materials were uniform in quality. The fact that the test data suggest that candidate insulators should be at least 0.015 in. thick raised a significant question, because many electrosurgical instrument manufacturers currently use insulation 0.008 in. thick without adverse effects.
Phase II. Based on the success of the initial test model, a more rigorous testing matrix was developed for a Phase II program, which explored the effects of repeated energizing, stress flaws, poor installation techniques, and sterilization on insulating ma-terials. Stress flaws in the insulation were simulated by press-fitting rings of copper tubing on the steel test mandrels to create an area of sharp transition. To simulate poor installation techniques, samples with visible air bubbles or poor contact with the mandrel were created through underheating. In general, the goal of this phase was to understand why many commercially available surgical instruments perform acceptably even though they have thinner insulation than the results of the first round of testing would suggest were required for safety.
The Phase II study only evaluated samples of three materials that had performed well in the Phase I tests: LDPE, the ionomer blend, and FEP. Unless otherwise noted, all of the tests used the 4000-V, 1-MHz test apparatus for a 30-second duration. The first series of tests involved testing the materials before and after they underwent EtO or gamma sterilization cycles, and it was determined that there was no significant decrease in the performance of any samples based on these sterilization conditions. Next, the samples created with deliberate defects such as air bubbles, pockets, and wrinkles were evaluated, and, for all three materials, samples with a wall thickness of 0.015 in. or greater passed. The performance of thicker samples was not evaluated because experience indicated that they would pass the test.
The test results for the third set of samples, which used a stepped mandrel to simulate sharp transitiontype stresses, were not conclusive. Stepped mandrel samples that were 0.011 in. thick performed approximately the same as nonstepped samples with a similar thickness. The fourth area for investigation was the effect of repeatedly energizing the same area of a sample; even after each sample mandrel had undergone 10 test cycles there was no deterioration in the performance of the insulating materials. The results of these Phase II tests simulating manufacturing and use conditions are summarized in Table II.
After the 1993 revision of HF18 was issued, it was decided to retest some of the thinner-wall samples that had marginal-to-poor performance during the first series of tests. The materials that were reevaluated were LDPE, the ionomer blend, HDPE, and FEP. Tests were performed at 6000 V peak to peak and 500 kHz to satisfy the new requirements of HF18-1993, and all of these materials passed, even at wall thicknesses of 0.008 in. The materials passed at the lower frequency and higher voltage because reducing the frequency increased the capacitive reactance of the materials. When the frequency decreases, the potential for leakage current decreases, thereby reducing the chance for the insulating materials to break down. Table III compares the results of tests at 500 kHz with the original results at 1 MHz.
The creation of appropriate test equipment and a test protocol made it possible to quickly screen typical insulating materials to the specifications of HF18-1986. Results were consistent for samples of the same material and correlated well with the predicted material performance. The test equipment and protocol also permitted the researchers to study areas of particular concern, such as the effects of improper installation and sterilization degradation of insulating materials. Finally, retesting the materials to the different electrical conditions specified in HF18-1993 led to an understanding of the discrepancy between previous test results and the actual performance of instruments currently in use. Several conclusions were drawn from these efforts, including the following:
* HF18-1986 specifies a very severe test for insulating materials used in electrosurgical devices. Only materials such as FEP, LDPE, HDPE, and the ionomer blend that are >= 0.015 in. thick can pass this test unconditionally. By separating these better high-frequency insulating materials from poorer insulators such as PVC and PVDF, the HF18-1986 test provides a high margin of safety.
* There is no significant difference in electrical performance between FEP and polyolefins such as LDPE, HDPE, and the ionomer blend. This comparability is important to instrument manufacturers because polyolefins are easier to install, less costly, and more able to withstand radiation sterilization than FEP is.
* The medium used to surround the insulating materials during testing has no significant effect on their performance. Results were the same for tests at 4000 V, 1 MHz in both air and saline. Various sterilization methods and poor installation practices did not significantly influence the materials' insulating capabilities either, nor did cycling up to 10 times at 4000 V, 1 MHz.
* Reducing the test frequency from 1 MHz to 500 kHz had a very strong effect on the results. FEP and polyolefins such as LDPE, HDPE, and the ionomer blend with wall thicknesses down to 0.008 in. were capable of withstanding 6000 V peak to peak for 30 seconds at 500 kHz. This capability correlates well with the known instrument-manufacturing practice of using insulating materials between 0.005 and 0.015 in. thick.
* The test specifications in HF18-1993 more closely parallel instrument insulation requirements in actual use than did those in HF18-1986. However, engineers might consider using a higher-frequency test like the HF18-1986 standard to screen candidate materials during the design phase of electrosurgical instrument development.
1. Odell RC, "Laparoscopic Electrosurgery," in Minimally Invasive Surgery, Hunter JG (ed), New York, McGraw-Hill, pp 3341, 1993.
2. Nduka CC, Super PA, Monson JR, et al., "Cause and Prevention of Electrosurgical Injuries in Laparoscopy," J Am Coll Surg, 179: 161170, 1994.
3. Strasberg SM, Sanabria JR, and Clavien PA, "Complications of Laparoscopic Cholecystectomy," Canadian J Surg, 275280, 1992.
4. Leahy AL, Bouchier-Hayes DB, and Hyland JM, "Early Experiences of Laparoscopic Cholecystectomy in Five Irish Hospitals," Irish J Med Sci, 161:410413, 1992.
5. Voyles CR, and Tucker RD, "Education and Engineering Solutions for Potential Problems with Laparoscopic Monopolar Electrosurgery," Am J Surg, 164(1):5762, 1992.
6. Voyles CR, and Tucker RD, Essentials of Monopolar Electrosurgery for Laparoscopy, Electrosurgical Concepts, 1992.
7. Electrosurgical Devices, ANSI/AAMI HF18-1993, Arlington, VA, Association for the Advancement of Medical Instrumentation (AAMI), 1993.
8. Electrosurgical Devices, ANSI/AAMI HF18-1986, Arlington, VA, AAMI, 1986.
Peter Kleinhenz is president and CEO of Progenics Corp. (Columbus, OH). Christine Vogdes is principal scientist at Raychem Corp. (Menlo Park, CA).