Ambiguity in the identification of Streptococcus pneumoniae. Optochin, bile solubility, quellung, and the AccuProbe DNA probe tests.

We prospectively evaluated 639 sequential clinical isolates of alpha-hemolytic gram-positive cocci as possible Streptococcus pneumoniae. On the basis of results of tests for optochin susceptibility, tube bile solubility, and the quellung reaction, 74 strains (11.6%) were categorized as unequivocal pneumococci (optochin positive, tube bile solubility positive, quellung reaction positive). Among 450 optochin- and tube bile solubility-negative organisms, a subset of 56 strains was tested for quellung reaction (all negative); these isolates were categorized as unequivocal nonpneumococci. A final 115 organisms with an inconsistent or discordant combination of susceptibility to optochin, tube bile solubility, and quellung reaction were categorized as equivocal strains. With the unequivocal isolates, a commercial molecular probe for S pneumoniae (AccuProbe; Gen-Probe, San Diego, Calif) showed 100% sensitivity (74/74) and 100% specificity (56/56). Among the 115 equivocal strains, however, 33 (28.7%) reacted with the AccuProbe, whereas only 3 (2.6%) showed a capsule that reacted in the quellung test. A subset of the equivocal strains identified in this group of primarily respiratory isolates may have been S pneumoniae that only partially expressed their classic phenotype of optochin susceptibility and bile solubility and only rarely expressed capsular antigens. A practical, cost-sparing algorithm is proposed to facilitate the routine clinical identification of S pneumoniae.

Radical improvements in the display of clinical microbiology results: a Web-based clinical information system.

OBJECTIVE: To evaluate the performance of a novel computerized system for reporting clinical microbiology results. The new system provides a summarized overview of a patient's current (or past) microbiological status, with the option to selectively explore in greater depth. It is deployed using World Wide Web technology, which supports virtually any kind of computer and allows physicians to obtain results via the internet using personal computers in the office or at home. METHODS: In an unblinded crossover study at a university-affiliated medical center, participants used both the new system and a conventional display system to retrieve selected microbiology results for two actual patients, according to standardized questionnaires, with balanced allocation of sequence of system use. Participants also subjectively rated the two systems. The participants were 16 physician, pharmacist, and nurse volunteers. Outcome measures included completion time and number of errors (categorized as major and minor) associated with results retrieval, and participants' ratings of the new system. RESULTS: Mean completion time was 45% shorter (13.9 versus 25.5 minutes; P < 0.001), and there were fewer associated major errors (0 versus 13; P = 0.01) and minor errors (10 versus 21; P = 0.003) with the summarized display system. All participants rated the new system as easier to learn and use than the conventional system. CONCLUSIONS: A system that appropriately summarizes and groups microbiology results can significantly shorten retrieval times and reduce interpretive errors, while providing users with information needed for cost-effective therapy. Such a system can be deployed by leveraging the rapidly evolving technology of the World Wide Web.

Closing the clinical laboratory testing loop with information technology.

Information technology can enhance the effectiveness of the part of the laboratory testing loop that occurs outside of the clinical laboratory. That external component involves the presentation of laboratory results and related information to physicians and the reception of physician requests for follow-up testing. Improvements made in the clinician-computer interface can conceivably enhance the quality of information transfer of this part of the testing loop and thereby improve the effectiveness of the entire loop. Our experience has been that results presentation is easier to address than order reception and that the economic benefits from improved result presentation can be substantial. Improving order reception requires more finesse, especially with the limitations of today's health systems and their information systems in mind. For instance, effective computer-based management of prospectively developed orders (eg, clinical protocols) not only can have a substantial positive impact on test use, but is actually welcomed by clinicians. Application of information technology at the clinician-computer interface has good potential to foster more appropriate use of clinical laboratory resources.

The clinical workstation as a means of improving laboratory use.

Driven to make timely decisions, doctors may act with less than all of the relevant patient data. Sub-optimal use of clinical laboratory resources can result. Our clinicians' workstation (CWS) is meant to provide doctors and nurses ready access to laboratory results in a form that makes the data easy to review and use. In addition, the workstation provides immediate feedback regarding blood orders, to encourage appropriate clinical practice. Feedback is based on the medical staff's clinical guidelines that have been incorporated into an embedded expert system. CWS also helps blood bank physicians to monitor, review and control requests for special needs such as irradiated products. Challenges to system acceptance await those trying to bring functional decision support to the clinical environment. Among these are barriers to understanding and cooperation posed by departmental boundaries and interacting professional cultures as well as the politics of hospital-based information systems.

A clinician's workstation for improving laboratory use. Integrated display of laboratory results.

Physicians are often forced to make decisions about the use of laboratory resources without adequate access to earlier results and related supporting information. Less than optimal use of the laboratory may result. The authors developed and deployed a clinical workstation meant to provide ready access to laboratory information that is presented in a format well-matched to the patient monitoring task. The workstation was one element of a multifaceted effort to improve blood component use in adult and pediatric bone marrow transplantation units. It was the sole intervention focused on improving laboratory testing. In the 2 years since the introduction of the workstation, median charges of bone marrow transplantation cases for laboratory test fell by 32%. This reduction in charges has been maintained for 2 years. Better informed physicians appear to use laboratory resources more sparingly.

User interface reengineering. Innovative applications of bar coding in a clinical microbiology laboratory.

Some clinical laboratory departments (such as microbiology) provide extensive reporting of text and other data not generated by instruments that can be interfaced to a laboratory information system. These data are usually entered into the laboratory information system manually by keyboard data entry, which can be cumbersome and time consuming. Bar codes, which are already used in laboratories to facilitate rapid entry of sample-identifying information, have the potential to be used much more broadly as a generalizable data entry technique. We developed a comprehensive system that takes advantage of several applications of bar coding to facilitate the work of our Clinical Microbiology Laboratory. Central to our system is the use of bar code "scripts" to meet many of our complex data entry requirements. Use of these scripts is transparent to the laboratory information system (ie, no special "drivers" are needed) because data are received as if they had been generated by typing the characters on the keyboard. The scripts consist of bar codes that encode the series of keystrokes needed to give the appropriate response at the series of prompts offered by the laboratory information system. Both alphanumeric and other keys, including carriage returns and special characters, can be converted into bar codes and incorporated into scripts. By creating and printing these scripts in the laboratory using standard wordprocessing software and bar code fonts for personal computers, laboratorians without specialized computer training have the tools to substantially improve the data entry efficiency of existing data entry terminals for a variety of laboratory information systems.

Integrating legacy laboratory information systems into a client-server world: the University of Minnesota Clinical Workstation (CWS) project.

The development of an innovative clinical decision-support project such as the University of Minnesota's Clinical Workstation initiative mandates the use of modern client-server network architectures. Preexisting conventional laboratory information systems (LIS) cannot be quickly replaced with client-server equivalents because of the cost and relative unavailability of such systems. Thus, embedding strategies that effectively integrate legacy information systems are needed. Our strategy led to the adoption of a multi-layered connection architecture that provides a data feed from our existing LIS to a new network-based relational database management system. By careful design, we maximize the use of open standards in our layered connection structure to provide data, requisition, or event messaging in several formats. Each layer is optimized to provide needed services to existing hospital clients and is well positioned to support future hospital network clients.

Laboratory detection of Clostridium difficile. A comparison of media and incubation systems.

Parallel testing for culture recovery of Clostridium difficile was performed using three selective media in each of four anaerobic incubation environmental systems. Testing was completed on 67 stool samples from 60 hospitalized patients in whom C difficile-associated diarrhea was suspected. Three different media were evaluated: CCFA (modified cycloserine-cefoxitin-fructose agar), CCFA-PRAS (CCFA, prereduced-anaerobically-sterilized) and CMBA (modified cycloserine-mannitol-blood agar). The incubation systems compared were an anaerobic chamber (Model 800 Anaerobic Environmental System, Anaerobe Systems, San Jose, CA), the anaerobic jar (BBL, Cockeysville, MD), the anaerobic Bio-Bag (BBL), and the anaerobic pouch (BBL). C difficile was found in 16 samples collected from 15 patients. Comparing recovery on the various types of media in all four anaerobic atmospheres, C difficile was recovered on all (64) CCFA plates, 56 of 64 CCFA-PRAS plates, and 43 of 64 CMBA plates (P < .03 comparing CCFA and CMBA). Of the 48 plates in each incubation system that were inoculated with specimens positive for C difficile, the organism was recovered from 43 plates in the anaerobe chamber, 41 incubated in an anaerobe jar, 40 in the Bio-Bag, and 39 incubated in the GasPak pouch, all providing similar recovery of C difficile (P = .08). The CCFA-PRAS and CMBA media demonstrated less inhibition of normal stool flora compared to the CCFA. Overall CCFA that was anaerobically reduced at least 4 hours before use, and contained the original concentration of antimicrobial agents described by George and colleagues, was superior to CMBA for recovery of C. difficile.

The deployment of a World Wide Web (W3) based medical information system.

The development of Web technologies has revolutionized information dissemination on the Internet. The University of Minnesota Hospital and Clinic's Web Clinical Information System (CIS) demonstrates the use of the Web as an infrastructure for deploying a medical information system at a fraction of the developmental cost of more traditional client server systems. This Web CIS has been deployed since December 1994. It makes available laboratory results, including a radically improved clinical microbiology reporting system, ad hoc laboratory order entry, and an embedded expert system protocol laboratory ordering system. It provides these services to any physician or patient care area with TCP (or SLIP/PPP) connection to our hospital network backbone, whether the client computer is running MS Windows, the Macintosh OS, or X-Windows. A formal evaluation of one of this systems subcomponents, the display of clinical microbiology information, demonstrated a significant savings in clinician time (43% p < .001) and substantial reduction in interpretive errors (0 vs 15 p < .01).

Activity of twenty-one antimicrobial agents including l-ofloxacin against quinolone-sensitive and -resistant, and methicillin-sensitive and -resistant Staphylococcus aureus.

There is a need to identify alternative agents to vancomycin for the treatment of infections with methicillin-resistant Staphylococcus aureus (MRSA). One candidate is the l isomer of ofloxacin (DR-3355). We tested 520 frozen MRSA isolates, 248 fresh MRSA isolates, and 375 fresh methicillin-susceptible S. aureus (MSSA) isolates from Minnesota, and 600 clinical isolates of S. aureus (150 MRSA and 450 MSSA) from Illinois. Over 90% of the MRSA strains were resistant to 32 micrograms/ml of oxacillin. Of the 520 frozen MRSA, 24% were susceptible to < or = 2 micrograms/ml ofloxacin, and an additional 74% were susceptible to ofloxacin between 8 and 16 micrograms/ml. More than 98% of all strains were susceptible to < or = 16 micrograms/ml ofloxacin or l-ofloxacin. All the quinolones had a bimodal distribution of in vitro activity, but for only ofloxacin and l-ofloxacin was activity confined to a very narrow range.

GyrA sequence analysis of Staphylococcus aureus and methicillin-resistant S. aureus strains selected, in vitro, for high-level ciprofloxacin resistance.

Four methicillin-susceptible Staphylococcus aureus and eight methicillin-resistant S. aureus (MRSA) isolates, all of which were ciprofloxacin susceptible (MIC < 2.0 micrograms/ml) were manipulated, in vitro, to achieve high-level ciprofloxacin resistance by means of up to 14 passages onto media containing increasing concentrations of ciprofloxacin. Resistance to ciprofloxacin at a concentration of at least 512 micrograms/ml was achieved in all 12 isolates tested. This resistance was continually detected during weekly passage on antibiotic-free media for 12 weeks. The parent and daughter cells from four strains had their gyrA sequenced from amino acid (aa) codons 70-100, the region of previous mutations in high level quinolone-resistant S. aureus. Mutations at aa codon 84 were seen in three of four strains, but appeared at varying levels of ciprofloxacin resistance. High-level resistance of S. aureus and MRSA to ciprofloxacin can be developed in vitro using multiple exposures to incremental concentrations of the drug. It is apparently due to multiple mechanisms and, once established, remains stable over time.

Utility of slide centrifuge gram's stain versus quantitative culture for diagnosis of urinary tract infection.

The slide centrifuge (cytospin) Gram's-stain technique has been shown in previous studies to be a sensitive technique for detecting bacteriuria when compared to culture. The method concentrates urine sediment in a small defined area on a glass slide for Gram's staining. A positive test provides morphologic information about suspected pathogens. This study evaluated the cytospin technique using 788 urine specimens, on which routine culture was simultaneously performed, and compared both with clinical evidence for urinary tract infection. One hundred twelve of these specimens, which were cytospin positive and had a culture growing more than 100,000 CFU/mL, were assumed, by definition, to represent true urinary tract infection. Five hundred twenty-six specimens had negative cytospin and negative culture results (less than 1,000 CFU/mL) and were assumed, by definition, to rule out the diagnosis of urinary tract infection. Clinical data were evaluated for 56 cytospin-positive specimens in which culture results were less than 100,000 CFU/mL. Of these specimens, 37 were false positive (no clinical evidence of urinary tract infection), 9 had clinical evidence of urinary tract infection, and for the remaining 10, data regarding clinical status could not be interpreted. Seventy-one specimens were cytospin negative, with cultures growing more than 1,000 CFU/mL. Of these, only one patient had clinical evidence of a urinary tract infection, and his culture result was less than 10,000 CFU/mL. The predictive value of a negative cytospin test was 99.8% compared to clinical information, whereas the predictive value of a negative culture (less than 100,000 CFU/mL) was 98.4%.

Comparison of the VIDAS Clostridium difficile toxin A immunoassay with C. difficile culture and cytotoxin and latex tests.

The VIDAS Clostridium difficile toxin A immunoassay (CDA) is a new, automated, enzyme-linked fluorescent-antibody assay for detection of C. difficile toxin A antigen in stool specimens. Simultaneous, parallel testing was performed by using the VIDAS CDA, the Culturette brand CDT latex test for C. difficile antigens, and conventional laboratory cell culture tests for C. difficile, cytotoxicity and C. difficile culture. One hundred ninety-four consecutive fresh soft or liquid stool samples submitted for C. difficile testing between July and September 1990 were evaluated. Of the 194 samples tested, 19 (10%) were from 16 patients who met our case definition for C. difficile-associated disease. The in vitro tests were evaluated in relation to two forms of a clinical case definition. In one form, a positive culture for toxin-producing C. difficile or a positive cytotoxin result obtained directly from the stool specimen was required as laboratory evidence of C. difficile. In the other, a positive result of any of the four laboratory tests was accepted for the laboratory portion of the case definition. No significant difference between the sensitivity of the VIDAS CDA and that of the Culturette brand CDT latex test was found (48 to 58% sensitivity for the CDT latex test and 52 to 63% sensitivity for the VIDAS CDA compared with 93 to 100% sensitivity for culture and 70 to 100% sensitivity for cytotoxin testing). The performance of the VIDAS CDA, however, was hampered by a high percentage of tests (19%) which gave an uninterpretable result.

Detection of gyrA gene mutations associated with ciprofloxacin resistance in methicillin-resistant Staphylococcus aureus: analysis by polymerase chain reaction and automated direct DNA sequencing.

A portion of the gyrA gene from amino acid codons 67 to 129 was sequenced in 34 methicillin-resistant Staphylococcus aureus strains (14 isolated in Minnesota, 10 isolated in Indiana, and 10 isolated in Tennessee). Twenty-eight of these strains were ciprofloxacin resistant. Sixteen of the strains contained a Ser----Leu mutation at codon 84; 3 contained strains a Ser----Ala mutation at codon 84; 3 strains contained two mutations, Ser----Leu at codon 84 and Ser----Pro at codon 85; and 6 strains contained a Glu----Lys mutation at codon 88. Six strains were wild type and ciprofloxacin susceptible. Several mutations from amino acid codons 84 through 88 can be associated with high-level quinolone resistance.

Nonparametric probability density estimation: improvements to the histogram for laboratory data.

The histogram has long been used in the clinical laboratory for the depiction and manipulation of frequency data. We present recent results of refinements to the usual histogram procedures along with modern alternative methods of estimating frequency distributions, including the kernel and discrete maximum penalized likelihood estimation (DMPLE) approaches. We compared these nonparametric methods on 15 different types of simulated distributions, and on several sets (greater than 1000 subjects/set) of real data, including alanine aminotransferase, aspartate aminotransferase, and lactate dehydrogenase levels. Each frequency curve estimation technique was evaluated by measuring the integrated mean square error between each technique's prediction and the true underlying distribution, using Monte Carlo techniques on sample sets with size 49 and 119. The kernel method was the clear method of choice, both in performance (best in 22/36 cases) and in practical usage.

Uptake of 3H-norfloxacin in methicillin-resistant Staphylococcus aureus.

To examine the possibility of a proton-motive efflux pump for quinolones in highly quinolone-resistant clinical isolates of methicillin-resistant Staphylococcus aureus (MRSA), we studied 3H-norfloxacin uptake in two quinolone-resistant and two quinolone-sensitive strains of MRSA whose gyrA region surrounding amino acid codons 84 and 85 had been sequenced. Two strains were related (one sensitive and one resistant) in that both were recovered from a single patient, one before (sensitive) and one after (resistant) ciprofloxacin therapy. Drug uptake was assessed in four separate experiments running triplicate bacterial suspensions with radiolabeled drug added at time = 0. Sampling was performed in 10 min increments up to 50 min by a vacuum filtration method. The ionic uncoupler, carbonyl cyanide m-chlorophenylhydrazone (CCCPH), was added at 40 min to test inhibition of a pump mechanism. The results demonstrated no statistically significant differences in uptake between the sensitive and resistant groups, and the uptake patterns were similar. CCCPH also induced an equivalent surge, or enhanced uptake among these strains, rendering an energy-dependent efflux pump an unlikely contributor to the high levels of resistance seen in our strains. Our findings support parallel studies done on these isolates that implicate mutational changes at amino acid codon 84 and/or codon 85 in the gyrA gene as an explanation for high-level quinolone resistance (MIC to ciprofloxacin greater than or equal to 16 mg/L) in MRSA.

The slide centrifuge gram stain as a urine screening method.

A slide centrifuge Gram stain procedure was performed to screen for bacteriuria 4161 urine specimens submitted in urine preservative tubes for routine culture. For slide centrifuge Gram staining, each urine sample was mixed well. Thereafter, 0.2 mL of each sample was placed, using a pipette, into a slide centrifuge chamber and centrifuged at 2,000 rpm for 5 minutes. The slides were heat fixed, Gram stained, and read by laboratory personnel who scanned 12 consecutive oil-immersion fields using a set pattern. The presence of the same organism in six or more fields was defined as a positive urine screen. Urine samples were cultured using a 0.001-mL loop and a comparison of culture growth with slide centrifuge screening was made. When growth of 100,000 or more colony-forming units per milliliter (CFU/mL) was the reference for comparison, the screen had a sensitivity rate of 98%, a specificity rate of 90%, a negative predictive value of 99%, and a positive predictive value of 65%. When a lower colony count of 10,000 or more CFU/mL was the reference for comparison, the screen had a sensitivity rate of 88%, a specificity rate of 95%, a negative predictive value of 96%, and a positive predictive value of 84%. The slide centrifuge Gram stain is a very sensitive screening method to detect bacteriuria in an adult male population.

An evaluation of three commercial fecal transport systems for the recovery of enteric pathogens.

Twenty-five isolates, including six strains of Shigella species, six strains of Salmonella species, five strains of Yersinia enterocolitica, six strains of Campylobacter jejuni, and two strains of Vibrio parahaemolyticus, were inoculated at a concentration of 1.5 x 10(4) colony-forming units/mL into the following transport systems: Fekal Enteric Plus (Trend Scientific, Inc., St. Paul, MN), Cary Blair Transport Medium (Remel, Inc., Lenexa, KS), and Para-Pak C & S (Meridian Diagnostics, Inc., Cincinnati, OH). The Fekal Enteric Plus system showed a greater than or equal to 50% recovery of the original bacterial inoculum after 96 hours for all Salmonella, Shigella, and Yersinia strains tested and after as long as 72 hours for Vibrio strains. The Cary Blair Transport System showed greater than or equal to 50% recovery of the initial inoculum at 96 hours for five of six Salmonella strains, four of six Shigella strains, all Yersinia strains, and one of two Vibrio strains. With the use of the Para-Pak C & S, greater than or equal to 50% recovery of the original inoculum after 96 h was demonstrated for five of six Salmonella strains, four of six Shigella strains, all Yersinia strains, and one of two Vibrio strains. All three systems did not demonstrate even 10% recovery of the initial C. jejuni inoculum at 24 hours when held at room temperature.

A general ampC active site oligonucleotide probe for gram-negative rods.

We have developed a 33 mer probe that hybridizes to the serine active site of the chromosomal ampC beta-lactamase gene of Pseudomonas aeruginosa and the Enterobacteriaceae. We tested this probe against a variety of Enterobacteriaceae, and a series of 23 P. aeruginosa by dot-blots and selected Southern blots. This probe is an alternative or supplement to enzyme studies for characterizing the class of a Gram-negative rod's beta-lactamase and is a useful tool for studies of Pseudomonas beta-lactamase regulation.

Comparison of traditional gas chromatography (GC), headspace GC, and the microbial identification library GC system for the identification of Clostridium difficile.

Three gas chromatography (GC) methods were compared for the identification of 52 clinical Clostridium difficile isolates, as well as 17 non-C. difficile Clostridium isolates. Headspace GC and Microbial Identification System (MIS) GC, an automated system which utilizes a software library developed at the Virginia Polytechnic Institute to identify organisms based on the fatty acids extracted from the bacterial cell wall, were compared against the reference method of traditional GC. Headspace GC and MIS were of approximately equivalent accuracy in identifying the 52 C. difficile isolates (52 of 52 versus 51 of 52, respectively). However, 7 of 52 organisms required repeated sample preparation before an identification was achieved by the MIS method. Both systems effectively differentiated C. difficile from non-C. difficile clostridia, although the MIS method correctly identified only 9 of 17. We conclude that the headspace GC system is an accurate method of C. difficile identification, which requires only one-fifth of the sample preparation time of MIS GC and one-half of the sample preparation time of traditional GC.