Investigation of capacitor failures in an automated external defibrillator.

Over the past 2 years, the Canadian Medical Devices Bureau has received a number of reports of capacitor failures on the high voltage board of an automated external defibrillator. Twenty-five cases of broken capacitor leads were found during routine preventive maintenance by the biomedical engineering staff of the institutions reporting the incidents. The Bureau has carried out a laboratory investigation to determine the effect of missing capacitors on the energy delivered by the defibrillator and to assess whether these capacitor failures represent a significant risk to patients. Our findings indicate that the automated external defibrillator will not perform acceptably with two broken capacitors. They further suggest that, during preventive maintenance, operators should use a defibrillator analyser to measure the delivered energy output rather than using the internal energy measurement circuit within the automated external defibrillator.

Repassivation of a high chromium stainless steel orthopaedic alloy.

The repassivation of Ortron90, a high chromium stainless steel alloy, was studied to assess the effect of passive layer removal on corrosion levels. Cathodic dissolution of the passive layer followed by potentiostatic experiments within the passive region were employed to study the repassivation process. The scratch method, carried out for comparison purposes, yielded current densities similar to those obtained by cathodic dissolution. The detailed analysis of the slopes of plots of log I/log t indicated that the repassivation did not follow any of the existing growth models. It also suggested that the slope depended on the extent of repassivation and reached a minimum when three to five monolayers of oxide were completed. The virtual potentiodynamic plots suggested the presence of some form of protective layer within 5-10 seconds and the change in open circuit potential vs time indicated that the passive layer present after 10 seconds of repassivation continued to evolve for about 1000 to 1500 seconds possibly representing a transition between two different forms such as oxide and a oxyhydroxides. Estimation of the effect of localized removal of passive layer on the corrosion indicated a dramatic increase in local corrosion density. These results show that a fast repassivation rate is an important requirement for biocompatibility of alloys used to manufacture dental and orthopaedic devices.

Galvanic corrosion of selected dental alloys.

Samples prepared from three different amalgam brands were coupled to two gold alloys and orthodontic brackets. In the resulting galvanic cells, the amalgam coupled to gold were anodic, exhibiting galvanic current densities about one order of magnitude higher than the uncoupled corrosion current densities of 0.2-0.5 microA. Coupling amalgams to orthodontic brackets resulted in galvanic current densities of the same magnitude as the uncoupled samples. Corrosion current densities at the anode were found to be up to six times higher than the measured galvanic current densities. Brushing caused transient increases in galvanic current densities that decayed within 100 s to the previous steady state levels. Brushing of amalgam/bracket couples, caused an anodic peak followed by brief polarization reversal during which the brackets were anodic.

Effect of reuse on surface characteristics of balloon angioplasty catheters.

Surfaces of reused angioplasty catheter balloons were characterized by scanning electron microscopy (SEM), image analysis, Fourier transform infrared (FT-IR) spectroscopy and X-ray photoelectron spectroscopy (XPS). The surfaces were found to have longitudinal ridges, cracks, wrinkled regions and particulates. Most of the particles were carbon-based, and contained traces of Cl and Si. The average number of particles ranged between 11 and 38 per mm2, while the average particle size ranged from 14 micron to 28 micron. About 70%-90% of the particles were larger then 10 micron which is approximately the diameter of the fine blood capillaries. The particles appeared to be firmly attached to the outer surface of the balloon. FT-IR and XPS analysis revealed the presence of Si, the absence of proteins, and suggested that the surface of the reused balloon catheters was oxidized. The study highlights the large number of particles created and released during angioplasty.

Effects of a wireless local area network (LAN) system, a telemetry system, and electrosurgical devices on medical devices in a hospital environment.

Concerns have been raised about interference of wireless local area network (LAN) systems and telemetry systems with medical devices in hospitals. The authors have investigated the susceptibility of 65 electromedical devices to a wireless LAN system and a telemetry system in preselected areas of a hospital. Testing was based on the American National Standards Institute Standard C63.18. The wireless LAN system operated at 2.42 GHz with an output power of 100 mW. The telemetry system operated at 466 MHz with an output power of 4 mW. Of the 65 devices tested, only two hand-held Doppler units, a Mini Doppler Model D900 (Huntleigh Healthcare Ltd) and a Ultrasonic Doppler Model 811 (Parks Medical Electronics, Inc.), were affected by the LAN system. Placed within 10 cm of the LAN system in standby mode, both units emitted periodic high-pitched beating sounds, which could be misinterpreted as normal beating sounds from the patient. These changed to random static noise during data transmission by the LAN. Under normal conditions of use, a LAN system would never be placed this close to a medical device. The quality of data transmission from the LAN system changed from "good" to "acceptable" in the colonoscopy room. This deterioration in transmission quality could have been caused by the lead shielding in the room. Electrosurgical devices operating at 0.5 to 1 MHz did not affect the LAN system at distances up to 3 m. None of the devices was affected by the telemetry system. These findings suggest that wireless LAN systems and telemetry systems can be acceptable for use in hospitals. Nevertheless, other systems should be tested on potentially susceptible devices by the hospital before use.

Walkie-talkies and cellular telephones: the hazards of electromagnetic interference in hospitals.

This article presents an overview of the effects of electromagnetic interference (EMI) on medical devices, with emphasis on portable telecommunication devices. In addition, the authors discuss existing voluntary national and international standards and policies governing the use of cellular telephones and walkie-talkies in hospitals, possible solutions to the susceptibility of medical devices to EMI, and recommendations to lessen EMI's potentially hazardous effects.

Temperature distribution beneath pediatric electrosurgical dispersive electrodes: a model study.

Concerns have been raised about the safety and the thermal performances of commercially available electrosurgical dispersive electrodes. A model system was designed to simulate the thermal effects observed in humans. The following critical factors affecting the thermal performances of pediatric electrodes were investigated: the intensity and duration of the applied current, the electrical resistivity of the test medium, the contact interface area between the electrode and the test medium, and the water content of the adhesive gel of the electrode. The maximum temperature rises (delta Tm) under the electrodes were 0.6 +/- 0.1 degrees C, 2.8 +/- 0.2 degrees C, and 5.2 +/- 0.3 degrees C when 60 s of 1 A, 2 A, and 3 A currents, respectively, were applied. delta Tm increased linearly with the increase of the duration of applied current and the resistivities of the test medium. When dispersive electrodes with reduced areas or reduction of water content of the adhesive gel of the electrodes were used, delta Tm increased with the areas or the water contents decreased. The currents and durations used in this study were in excess of those routinely used in electrosurgery. The study confirmed that the maximum temperature increase under the electrode is inversely proportional to the square of the perimeter of the electrode, rather than the square of the surface area. The test methods described can be used to evaluate the thermal performances of other electrosurgical dispersive electrodes. Monitoring the impedance at the interface between the dispersive electrode and the tissue in contact would reduce the risk of burns.

Rapid detection of group A streptococci: comparative performance by nurses and laboratory technologists in pediatric satellite laboratories using three test kits.

Rapid tests for detecting group A streptococci in throat swabs are often performed outside hospitals or commercial laboratories by individuals with little or no technical training. We compared the abilities of nurses and technologists to perform and interpret three commercial kits (Directigen 1-2-3, ICON Strep A, and Culturette Brand 10-Minute Strep A ID) in three hospital satellite locations (the emergency department, a walk-in emergency clinic, and a general pediatric clinic). When the three tests were compared with culture, the sensitivities of the tests as performed by nurses and technologists, respectively, were 39 versus 44% for Directigen, 55 versus 51% for Culturette, and 72 versus 39% for ICON. A significant difference in sensitivity was found only with ICON tests. This result was largely explained by the tendency of technologists to test moist swabs, while nurses generally processed dry swabs; ICON test sensitivity was significantly greater with dry swabs. The specificities of Directigen and ICON tests performed by nurses and technologists were high (97 to 100%). The difference in the specificities of the Culturette test as determined from results obtained by nurses and technologists (80 versus 98%) was due to the tendency of one nurse to overinterpret the latex agglutination reaction. Analysis of the accuracies of the tests during practice periods compared with the accuracies of the tests during the study periods revealed statistically significant improvement in test performance. We conclude that these tests are specific but not sensitive when performed by nurses and technologists in satellite laboratories. With one exception, nurses and technologists performed the tests with comparable accuracy after brief training periods.

In vitro degradation of a poly(ether urethane) by trypsin.

In vitro enzymatic degradation of non-porous films of segmented poly(ether urethane) (Pellethane 2363-80AE) was investigated by incubating the biomaterial in concentrated trypsin solutions for 5 months at room temperature. Chemical degradation of films was monitored by surface analysis techniques such as Fourier transform infrared spectroscopy-attenuated total reflectance and electron spectroscopy for chemical analysis. This latter technique proved to be much superior in detecting chemical changes. Extraction of films with methanol and characterization of the extracts by gel permeation chromatography revealed the presence of low-molecular-weight polymers. Results have shown that trypsin has the ability to induce degradation in PEU, the soft segment being most affected, particularly the CH2-O bond of the ether linkages.

Effects of temperature and viscosity on prothrombin times of blood.

Accurate prothrombin time tests are important because they are frequently performed on presurgical patients to evaluate their blood-clotting status. We studied the effect of temperature (27-47 degrees C) on PTs obtained with eight different brands of thromboplastin. We also compared the sensitivities of two types of coagulation timers to changes in blood viscosities between 1 and 16 mPa/s. Viscosities were measured with the Brookfield Digital Viscometer. The MLA Eletra 800 and the BBL fibrometer were used to measure PTs. All eight thromboplastins gave convex curves of PT versus temperature, with optimum values lying between 38 and 39 degrees C. The curves were fitted to 4th-degree polynomials which showed that a mean temperature bias of 2 degrees C can increase PTs. Ortho Brain (7.8% change) was affected the most, while thromboplastin C (4.4% change) was affected the least. Plots of PT versus viscosity showed that the BBL fibrometer, which uses an electromechanical sensor, was more affected by viscosity than the MLA Electra 800, with an optical detector. However, above 8.2 mPa/s, all PTs were significantly elevated. Hence, patients with macroglobulinemia, whose plasma viscosities sometimes exceed 8.2 mPa/s, may have falsely elevated PTs. We conclude that temperature and viscosity are critical factors in the test and significantly contribute to within and between laboratory variations in PT measurements.

Leaching of plasticizers from and surface characterization of PVC blood platelet bags.

The leaching of phthalate plasticizers from four types of blood platelet bags was investigated. The anticoagulant solutions used in the blood collection bags had pH values of 5.64 +/- .04 and contained no detectable amounts of phthalates. Platelet bag materials from each bag were soaked in normal salines for up to 5 days. The salines were tested for the leached phthalates from the bags but none could be found. However, di-(2-ethylhexyl) phthalate (DEHP) leached out of the PL-146 and Terumo bags into bovine calf serum used for soaking the bag materials. There was an increase in the amount of DEHP leached from about 1.1 mg at the end of one day to about 3.3 mg per gm of bag material at the end of a five day extraction with the serum. In PL-732 sets, a platelet bag made of a specialty polyolefin, the amount of DEHP leached out was less than 0.02 mg per g of bag material. CLX bags, which contained tri-(2-ethylhexyl) trimelliate (TETM) as a plasticizer, showed a negligible amount of it leaching into the calf serum. Infra-red spectra showed that PL-146 bags had been coated with a layer of a fatty acid amide while the Terumo bags contained a layer of a silicone fluid on their inner surfaces. CLX bags showed a coating of stearates, which were probably soaps of calcium or zinc. Scanning electron microscopy (SEM) showed that the inner surfaces of each brand of the bag were distinctly different morphologically. The two PVC bags were very similar whereas the surfaces morphology of PL-732 was rougher. Terumo bag had a different surface morphology than those of the other bags whereas the CLX bags had a very regular surface pattern. The exact significance of the surface morphology is not certain but excessively rough surfaces may not be desirable for the bags.

Receiver operator characteristic (ROC) curves and non-normal data: an empirical study.

This paper evaluates the performance of several diagnostic kits for assessing levels of serum prostatic acid phosphatase on patients at different stages of cancer of the prostate. Each patient was studied with several kits. We compare results obtained for receiver operator characteristic (ROC) curve methodology with data assumed to follow a normal distribution, with log-transformed data assumed to follow a normal distribution, and when neither of these assumptions holds. There were important differences between the results of the different approaches. For these data, the normal distribution assumption should be used with extreme caution. The log-transformed based results compared favourably with the non-parametric, but unconsidered application of the method should be avoided.

Physician office analyzers. Do they satisfy medically useful criteria for analytical performance of laboratory tests?

The coefficients of variation (CVs) for several tests on physician office analyzers were compared with published data for medically useful criteria of these tests. Medically useful CVs were based on physicians' opinions of significant changes in laboratory tests. Precision studies using control material were carried out on four physician office analyzers by both a trained technologist and nontechnical personnel, such as physicians, nurses, and medical office personnel. This study was carried out during a six-month period. The instruments that were evaluated were the Reflotron, Seralyzer, Vision, and DT60. Low and high concentrations of quality control material, supplied by the manufacturer and those in use in the study laboratory (Omega I and II, Cooper Biomedical, Fairhaven, NJ), were used. With few exceptions, all instruments satisfied the medically useful CVs criteria, ie, the precision obtained by both the technologist and nontechnical personnel either equaled or exceeded the precision requirements set by physicians.