The use of computers in teaching clinical laboratory science.

The Transfusion Medicine Tutor (TMT) has been designed to study the use of computers in teaching concepts and problem- solving skills important in the field of clinical laboratory science. This system provides students with an opportunity to gain experience by solving a wide range of actual cases, and coaches these students when they are having difficulties. This system is designed specifically to detect and respond to a variety of errors that students may make while solving cases, and to suggest more advanced problem-solving methods when appropriate. This article describes the concepts behind the design of TMT.

An empirical evaluation of the performance of antibody identification tasks.

Four empirical studies were conducted for better understanding of the nature of problem-solving activities by medical technologists and medical technology students when performing antibody identification tasks. The results indicated the importance of strategies that ensure the collection of converging evidence, as these strategies protect against the fallibility of commonly used heuristics and against errors due to simple slips. The results also indicate that not only do students make significant numbers of errors, but so do practicing technologists. In one of the studies covering a 1-year period, for instance, a group of 16 technologists made a total of 41 errors in 1057 cases. On the basis of these findings, several alternatives are proposed to reduce errors.

Procedural errors in antibody identification.

In experimental studies of students and line technologists performing antibody identification procedures, both groups made errors. These errors included, at times, either failing to identify an antibody or misidentifying the specficity(ies)A. prospective study was undertaken to identify errors made in a laboratory setting. Errors were classified as 1) failing to follow protocol (procedural error) or 2) arriving at the wrong answer (misidentification error). Over a 1-year period, 1,057 workups were reviewed. There were 41 (3.88%) procedural errors and no misidentification errors. In 25 workups (61% of errors), the selection of cells m rule out underlying alloantibody(ies) was in error. The remaining 16 involved various "slips" (minor mistakes or memory lapses) and clerical errors. Based on an analysis of the probable causes of these errors, potential solutions include 1) developing computer aids to detect "rule-out" errors or missing tests results; 2) providing timely, careful review of workups before transfusion; and 3) designing better panel layout and cell selection.

Qualitative representation of behavior in the medical domain.

In medical knowledge-based systems, there is a need for models of the human body which can predict and explain behavior based on qualitative information about the structure and behavior of the body. We present a framework for representing the structure and behavior of physical systems and apply it to the human cardiovascular system. Our framework allows for hierarchical representation of physical systems and facilitates two kinds of reasoning processes: composition of behaviors and qualitative simulation.

RED: a red-cell antibody identification expert module.

We describe a software module in an expert system RED, which interprets data related to red cell antibody identification. There are three portions to this module: the problem-solving component, which incorporates the knowledge required for antibody identification as a hierarchy of programs. The programs in the hierarchy organize within themselves small pieces of knowledge represented in the form of production rules, which are capable of making judgments concerning a specific hypothesis; an intelligent data base for storage of patient data, red cell attributes, and test results; the "overview critic" portion, which combines the atomic hypotheses judged favorably by the antibody programs into a unified judgment concerning the case. Overview makes the decision to terminate processing with a conclusion about which antibodies are actually present and what specific further tests need to be performed to resolve any remaining ambiguities.

Filamentous Histoplasma capsulatum endocarditis involving mitral and aortic valve porcine bioprostheses.

We report a case of histoplasmosis capsulati endocarditis simultaneously infecting a mitral and aortic porcine bioprostheses. Histologically, the fungus demonstrated a diversity of morphologies, ranging from typical, intracellular, 2- to 5-micron yeast cells, to septate and branching hyphae, to bizarre hyphae with vesicular swellings. Cultures obtained from both valves were positive in four days. The combination of unexpected histologic appearance and rapid growth in culture may pose problems in diagnosis.

Mycobacterium fortuitum infection of a Hickman catheter site.

A patient with histiocytic lymphoma who developed a Hickman catheter exist site infection due to Mycobacterium fortuitum is described. Due to the risk of dissemination in immunosuppressed patients and the resistance to antibiotic therapy, rapid-growing mycobacteria should be considered when gram-positive bacilli are associated with infections in patients with these catheters.

Evaluation of the Anaerobe-Tek system for identification of anaerobic bacteria.

The Anaerobe-Tek system of Flow Laboratories, Inc. (McLean, Va.), is a new bacterial identification system which uses Gram morphology, sporulation, and reactions from 15 agar-based biochemical tests to generate a six-digit code number for identification of anaerobic bacteria. Supplemental tests are recommended when necessary to complete species identification. Individual test and identification performance of this system was evaluated by testing 216 anaerobic bacteria representing 31 species and one Centers for Disease Control unnamed group in parallel with a routine clinical laboratory identification system. Most of the tests in the Anaerobe-Tek system performed well; 85% of the organisms were correctly identified. The 32 (15%) failures in identification were due to omission from the identification code data base (38%), false-negative indole reactions (22%), and other incorrect biochemical reactions (40%). The replacement of the recommended indole test with an extraction method using the inoculum broth and an expansion of the data base of the system could raise the correct identification for this organism population to over 90% with no change in the test materials.

Improvement of human myocardial mitochondria after dobutamine: a quantitative ultrastructural study.

Dobutamine is a new inotropic agent that may induce prolonged clinical improvement in patients with congestive cardiomyopathy. Sixteen patients with severe heart failure but without obstructive coronary disease were studied by serial functional class determinations and by noninvasive measurements of left ventricular function before and after bedrest control period and then after a 3-day infusion of dobutamine. An endomyocardial biopsy procedure was performed on 11 patients before and after the bedrest and on all 16 patients before and after dobutamine. Quantitative ultrastructural analysis of the biopsies was performed on electron micrographs (31,200 X) by a grid technique. The number of electron dense particles per 100 mitochondria did not change after bedrest but there was a significant decrease from 84 +/- 14 to 65 +/- 12 (P < .005) in the pre- to postdobutamine biopsies. The cristae to matrix ratio of the mitochondria did not change with bedrest but improved from 0.37 +/- 0.04 to 0.47 +/- 0.05 (P < .02) after dobutamine. The average mitochondrial size did not change significantly in the bedrest control nor in the dobutamine biopsy samples. However, when the 16 patients were divided into those who had a good clinical response (by functional class and noninvasive measurements of left ventricular function) and those who had no or little response, the 10 responders did decrease their average mitochondria size from 0.26 +/- 0.03 mu 2 to 0.23 +/- 0.02 mu 2 (P < .02). The mechanism by which a 3-day infusion of dobutamine induces a prolonged clinical improvement is not well understood. The use of quantitative ultrastructural technique in this study has demonstrated that there is a morphologic basis to the improvement.

The importance of request and report forms in the interpretation of therapeutic drug monitoring data.

The interpretation of therapeutic drug monitoring data depends not only on the measurements themselves but also on additional data collected on the request forms and the manner in which the results are displayed on the report forms. A study was conducted to learn how presently used request/report forms affect the interpretation of therapeutic drug level measurements. Letters were sent to 232 directors of laboratory services requesting them to forward examples of forms that they use in therapeutic drug monitoring and toxicology. Replies were received from 77 directors of whom 53 enclosed examples of their therapeutic drug monitoring request/report forms. The characteristics of the 53 request/report forms were analyzed, and the results indicate that request/report forms are not optimally configured to facilitate effective therapeutic drug monitoring. Many request forms are not designed to allow entry of even the most basic data essential to the interpretation of therapeutic drug level measurements. Many report forms do little more than provide a drug level measurement and a therapeutic range. Improved request/report forms, as well as systems to ensure their proper use through adequate numbers of qualified personnel are necessary. Data/word processing computers should prove helpful.