Errors in interpretation of Gram stains from positive blood cultures

Performance evaluation of machine-assisted interpretation of Gram stains from positive blood cultures

Manual microscopy of Gram stains from positive blood cultures (PBCs) is crucial for diagnosing bloodstream infections but remains labor intensive, time consuming, and subjective. This study aimed to evaluate a scan and analysis system that combines fully automated digital microscopy with deep convolutional neural networks (CNNs) to assist the interpretation of Gram stains from PBCs for routine laboratory use. The CNN was trained to classify images of Gram stains based on staining and morphology into seven different classes: background/false-positive, Gram-positive cocci in clusters (GPCCL), Gram-positive cocci in pairs (GPCP), Gram-positive cocci in chains (GPCC), rod-shaped bacilli (RSB), yeasts, and polymicrobial specimens. A total of 1,555 Gram-stained slides of PBCs were scanned, pre-classified, and reviewed by medical professionals. The results of assisted Gram stain interpretation were compared to those of manual microscopy and cultural species identification by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). The comparison of assisted Gram stain interpretation and manual microscopy yielded positive/negative percent agreement values of 95.8%/98.0% (GPCCL), 87.6%/99.3% (GPCP/GPCC), 97.4%/97.8% (RSB), 83.3%/99.3% (yeasts), and 87.0%/98.5% (negative/false positive). The assisted Gram stain interpretation, when compared to MALDI-TOF MS species identification, also yielded similar results. During the analytical performance study, assisted interpretation showed excellent reproducibility and repeatability. Any microorganism in PBCs should be detectable at the determined limit of detection of 105 CFU/mL. Although the CNN-based interpretation of Gram stains from PBCs is not yet ready for clinical implementation, it has potential for future integration and advancement.

Test Performance and Potential Clinical Utility of the GenMark Dx ePlex Blood Culture Identification Gram-Negative Panel

The test performance and potential clinical utility of the ePlex blood culture identification Gram-negative (BCID-GN) panel was evaluated relative to matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry on bacterial isolates and conventional antimicrobial susceptibility testing. The majority (106/108, 98.1%) of GN bacteria identified by MALDI were on the BCID-GN panel, and valid tests (107/108, 99.1%) yielded results on average 26.7 h earlier. For all valid tests with on-panel organisms, the positive percent agreement was 102/105 (97.2%) with 3 false negatives and the negative percent agreement was 105/105. Chart review (n = 98) showed that in conjunction with Gram stain results, negative pan-Gram-positive (GP) markers provided the opportunity to discontinue GP antibiotic coverage in 63/98 (64.3%) cases on average 26.2 h earlier. Only 8/12 (66.7%) Enterobacterales isolates with resistance to third-generation cephalosporins harbored the CTX-M gene. In contrast, 8/8 CTX-M⁺ samples yielded a resistant isolate. Detection of 1 Stenotrophomonas maltophilia (18 h), 1 OXA23/48⁺ Acinetobacter baumannii (52.4 h), and 3 CTX-M⁺ Enterobacterales isolates on ineffective treatment (47.1 h) and 1 on suboptimal therapy (72.6 h) would have additionally enabled early antimicrobial optimization in 6/98 (6.1%) patients. IMPORTANCE The GenMark Dx ePlex rapid blood culture diagnostic system enables earlier time to identification of antimicrobial-resistant Gram-negative bacteria causing bloodstream infections. Its ability to rule out Gram-positive bacteria enabled early discontinuation of unnecessary antibiotics in 63/98 (64.3%) cases on average 26.2 h earlier. Detection of bacteria harboring the CTX-M gene as well as early identification of highly resistant bacteria such as Stenotrophomonas maltophilia and Acinetobacter baumannii enabled optimization of ineffective therapy in 6/98 (6.1%) patients. Its implementation in clinical microbiology laboratories optimizes therapy and improves patient care.

Performances of automated digital imaging of Gram-stained slides with on-screen reading against manual microscopy

The objective of this study was to evaluate the performances of the automated digital imaging of Gram-stained slides against manual microscopy. Four hundred forty-three identified Gram-stained slides were included in this study. When both methods agreed, we considered the results as correct, and no further examination was carried out. Whenever the methods gave discrepant results, we reviewed the digital images and the glass slides by manual microscopy to avoid incorrectly read smears. The final result was a consensus of multiple independent reader interpretations. Among the 443 slides analyzed in this study, 101 (22.8%) showed discrepant results between the compared methods. The rates of discrepant results according to the specimen types were 5.7% (9/157) for positive blood cultures, 42% (60/142) for respiratory tract specimens, and 22% (32/144) for sterile site specimens. After a subsequent review of the discrepant slides, the final rate of discrepancies dropped to 7.0% (31/443). The overall agreement between the compared methods and the culture results reached 78% (345/443) and 79% (349/443) for manual microscopy and automated digital imaging, respectively. According to culture results, the specificity for automated digital imaging and manual microscopy were 90.8% and 87.7% respectively. In contrast, sensitivity was 84.1% for the two compared methods. The discrepant results were mostly encountered with microorganism morphologies of rare occurrence. The results reported in this study emphasize that on-screen reading is challenging, since the recognition of morphologies on-screen can appear different as compared to routine manual microscopy. Monitoring of Gram stain errors, which is facilitated by automated digital imaging, remains crucial for the quality control of reported Gram stain results.

Reclassification of Catabacter hongkongensis as Christensenella hongkongensis comb. nov. based on whole genome analysis

The genera Catabacter (family 'Catabacteraceae') and Christensenella (family Christensenellaceae) are close relatives within the phylum Firmicutes. Members of these genera are strictly anaerobic, non-spore-forming and short straight rods with diverse phenotypes. Phylogenetic analysis of 16S rRNA genes suggest that Catabacter splits Christensenella into a polyphyletic clade. In an effort to ensure that family/genus names represent monophyletic clades, we performed a whole-genome based analysis of the genomes available for the cultured representatives of these genera: four species of Christensenella and two strains of Catabacter hongkongensis. A concatenated alignment of 135 shared protein sequences of single-copy core genes present in the included strains indicates that C. hongkongensis is indeed nested within the Christensenella clade. Based on their evolutionary relationship, we propose the transfer of Catabacter hongkongensis to the genus Christensenella as Christensenella hongkongensis comb. nov.

Evaluation of MicroScan Bacterial Identification Panels for Low-Resource Settings

Bacterial identification is challenging in low-resource settings (LRS). We evaluated the MicroScan identification panels (Beckman Coulter, Brea, CA, USA) as part of Médecins Sans Frontières' Mini-lab Project. The MicroScan Dried Overnight Positive ID Type 3 (PID3) panels for Gram-positive organisms and Dried Overnight Negative ID Type 2 (NID2) panels for Gram-negative organisms were assessed with 367 clinical isolates from LRS. Robustness was studied by inoculating Gram-negative species on the Gram-positive panel and vice versa. The ease of use of the panels and readability of the instructions for use (IFU) were evaluated. Of species represented in the MicroScan database, 94.6% (185/195) of Gram-negative and 85.9% (110/128) of Gram-positive isolates were correctly identified up to species level. Of species not represented in the database (e.g., Streptococcus suis and Bacillus spp.), 53.1% out of 49 isolates were incorrectly identified as non-related bacterial species. Testing of Gram-positive isolates on Gram-negative panels and vice versa (n = 144) resulted in incorrect identifications for 38.2% of tested isolates. The readability level of the IFU was considered too high for LRS. Inoculation of the panels was favorably evaluated, whereas the visual reading of the panels was considered error-prone. In conclusion, the accuracy of the MicroScan identification panels was excellent for Gram-negative species and good for Gram-positive species. Improvements in stability, robustness, and ease of use have been identified to assure adaptation to LRS constraints.

Gram Staining: a Comparison of Two Automated Systems and Manual Staining

Various Gram staining automated systems are available to accelerate and standardize the staining process, but a systematic comparison of different systems is largely lacking. The objective of this study was to evaluate two devices in comparison to manual Gram staining. Clinical samples (n = 500; University Hospital Münster, Germany; May to June 2020) were simultaneously Gram stained manually and with two automated Gram stainers (Previ Color Gram, bioMérieux, and ColorAX2, Axonlab). The quality was assessed based on four criteria: (i) homogeneous staining of bacteria/fungi, (ii) uniform staining of the background, (iii) absence of staining artifacts, and (iv) congruency between culture and microscopy. Each criterion was rated with 0 (absence) or 1 (presence) point to calculate a quality score (0 to 4 points). The costs for each staining procedure were calculated based on consumables and hands-on time (applying the average wage of a laboratory technician in the public service for Germany and the United States). The mean (± standard deviation [SD]) quality scores were comparable for manual staining (3.06 ± 0.91) and Previ Color Gram (3.04 ± 0.90; P = 0.6), while significantly lower scores were achieved by ColorAX2 (2.57 ± 1.09; P < 0.0001). The total cost per Gram stain was €1.13/$1.34 for Previ Color Gram, €0.80/$0.83 for manual, and €0.60/$0.71 for ColorAX2, respectively. The quality and costs per slide vary significantly between instruments of different manufacturers.

Diagnostic performance of the Sputum Gram Stain in predicting sputum culture results for critically ill pediatric patients with pneumonia

BACKGROUND

The sputum Gram stain is an inexpensive, rapid, and convenient laboratory method that predicts the bacterial pathogens in patients with pneumonia. This study aimed to evaluate the diagnostic performance of this method in predicting sputum culture results for critically ill pediatric patients.

METHODS

From June 2008 to June 2018, patients with pneumonia with an endotracheal or a tracheostomy tube in place in the Pediatric Intensive Care Unit at Changhua Christian Hospital were enrolled retrospectively. Sputum was collected from each patient via the artificial airway for Gram stain and culture evaluations of bacterial pathogens. Mixed culture results were excluded. A successful prediction was defined as a match of the sputum Gram stain and culture results.

RESULTS

A total of 622 records were reviewed, of which 542 were analyzed. Haemophilus influenzae, Pseudomonas aeruginosa, and Streptococcus pneumoniae were the three most common pathogens found. The overall prediction success rate of the sputum Gram stain was 59.23%. The sensitivity of the method in predicting gram-negative bacilli (GNB), gram-negative cocci (GNC), and gram-positive cocci (GPC) was 0.45, 0.67, and 0.61, respectively. Its specificity in predicting GNB, GNC, and GPC was 0.87, 0.98, and 0.87, respectively. Its positive likelihood ratio in predicting GNB, GNC, and GPC was 3.46, 33.50, and 4.69, respectively. The highest prediction success rate among all pathogens was for GNC.

CONCLUSION

The sputum Gram stain had high specificity and relatively low sensitivity in predicting the bacterial pathogens in critically ill pediatric patients. Its high specificity in predicting sputum culture results means that clinicians can confidently use sputum Gram stain results to guide their antibiotic choice for treatment.

Is Lactobacillus Gram-Positive? A Case Study of Lactobacillus iners

Lactobacillus iners is the most prevalent bacterial species in the human vaginal microbiome, and there have been few reports of its Gram-negative stain appearances despite the fact that the genus Lactobacillus is universally described as Gram-positive. Here, using transmission electron microscopy, we reveal that the thinness of the cell wall (17.39 ± 0.8 nm) gives the Gram-negative stain appearance, which can lead to over-diagnosis of bacterial vaginosis. Moreover, comparative genome analysis identified four genes commonly absent in L.iners genomes that might contribute to this phenotypic difference. We suggest that, along with the several niche-specific attributes identified, this unique feature may contribute to the species’ distinguished capability to thrive as the predominant species in the fluctuating vaginal environment as well.

Development of a standardized Gram stain procedure for bacteria and inflammatory cells using an automated staining instrument

Gram stain is a subjective and poorly controlled test, and the resultant errors often perplex laboratory scientists. To reduce errors and make Gram stain a precisely controllable and meritorious test, a standardized Gram stain procedure for bacteria and inflammatory cells was developed using an automated staining instrument in this study. Freshly expectorated sputum specimens, used as the optimized targets, were smeared on slides by laboratory technicians, defining each slide loaded with uniform matrix and monolayer cell. And then, the staining and decolorizing time, as well as the stain and decolorant volume, were optimized as 15, 105, 1, and 25 s and 1.1, 1.4, 0.3, and 0.7 ml, respectively. Culture‐positive blood specimens and original purulent fluids were used for confirming the developed standardized Gram stain procedure. Distinct tinctures of bacteria and inflammatory cells adhered to slide uniformly in a monolayer were observed, and the obtained staining results of these samples were highly consistent with their cultured results. Furthermore, according to the staining results under different staining conditions, an updated molecular mechanism of Gram stain for bacteria and the probable staining mechanism for inflammatory cells were also proposed in this study.

Clostridium ramosum rapidly identified by MALDI-TOF MS. A rare gram-variable agent of bacteraemia

Clostridium ramosum is an enteric anaerobic, endospore-forming, gram-positive rod with a low GC content that is rarely associated with disease in humans. We present a case of C. ramosum bacteraemia. To the best of our knowledge, this is the second case of C. ramosum bacteraemia in an elderly patient presenting with fever, abdominal pain and bilious emesis. We highlight the Gram stain variability, the lack of visualization of spores and the atypical morphology of the colonies that showed C. ramosum in a polymicrobial presentation that initially appeared to show monomicrobial bacteraemia. The microorganism was rapidly identified by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). We present a comprehensive literature review of 32 cases of clinical infections by C. ramosum in which we describe, if available, sex, age, clinical symptoms, predisposing conditions, other organisms present in the blood culture, other samples with C. ramosum , identification methodology, treatment and outcome.

Clinical Performance of the Novel GenMark Dx ePlex Blood Culture ID Gram-Positive Panel

Rapid identification from positive blood cultures is standard of care (SOC) in many clinical microbiology laboratories. The GenMark Dx ePlex Blood Culture Identification Gram-Positive (BCID-GP) Panel is a multiplex nucleic acid amplification assay based on competitive DNA hybridization and electrochemical detection using eSensor technology. This multicenter study compared the investigational-use-only (IUO) BCID-GP Panel to other methods of identification of 20 Gram-positive bacteria, four antimicrobial resistance genes, and both Pan Candida and Pan Gram-Negative targets that are unique to the BCID-GP Panel.

Rapid identification from positive blood cultures is standard of care (SOC) in many clinical microbiology laboratories. The GenMark Dx ePlex Blood Culture Identification Gram-Positive (BCID-GP) Panel is a multiplex nucleic acid amplification assay based on competitive DNA hybridization and electrochemical detection using eSensor technology. This multicenter study compared the investigational-use-only (IUO) BCID-GP Panel to other methods of identification of 20 Gram-positive bacteria, four antimicrobial resistance genes, and both Pan Candida and Pan Gram-Negative targets that are unique to the BCID-GP Panel. Ten microbiology laboratories throughout the United States collected residual, deidentified positive blood culture samples for analysis. Five laboratories tested both clinical and contrived samples with the BCID-GP Panel. Comparator identification methods included each laboratory’s SOC, which included matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) and automated identification systems as well as targeted PCR/analytically validated real-time PCR (qPCR) with bidirectional sequencing. A total of 2,342 evaluable samples (1,777 clinical and 565 contrived) were tested with the BCID-GP Panel. The overall sample accuracy for on-panel organisms was 89% before resolution of discordant results. For pathogenic Gram-positive targets (Bacillus cereus group, Enterococcus spp., Enterococcus faecalis, Enterococcus faecium, Staphylococcus spp., Staphylococcus aureus, Staphylococcus epidermidis, Staphylococcus lugdunensis, Listeria spp., Listeria monocytogenes, Streptococcus spp., Streptococcus agalactiae, Streptococcus anginosus group, Streptococcus pneumoniae, and Streptococcus pyogenes), positive percent agreement (PPA) and negative percent agreement (NPA) ranged from 93.1% to 100% and 98.8% to 100%, respectively. For contamination rule-out targets (Bacillus subtilis group, Corynebacterium, Cutibacterium acnes, Lactobacillus, and Micrococcus), PPA and NPA ranged from 84.5% to 100% and 99.9% to 100%, respectively. Positive percent agreement and NPA for the Pan Candida and Pan Gram-Negative targets were 92.4% and 95.7% for the former and 99.9% and 99.6% for the latter. The PPAs for resistance markers were as follows: mecA, 97.2%; mecC, 100%; vanA, 96.8%; and vanB, 100%. Negative percent agreement ranged from 96.6% to 100%. In conclusion, the ePlex BCID-GP Panel compares favorably to SOC and targeted molecular methods for the identification of 20 Gram-positive pathogens and four antimicrobial resistance genes in positive blood culture bottles. This panel detects a broad range of pathogens and mixed infections with yeast and Gram-negative organisms from the same positive blood culture bottle.

Harmonization of microbiology processes and standards: work in progress

Clinical microbiology is a highly challenging technical discipline, which makes it difficult to harmonize processes and protocols. In addition, the lack of scientific consensus on some diagnostic algorithms and the need to address the diverse needs of different patient populations contribute to the lack of interlaboratory consistency. Laboratories utilize a number of measures and metrics, but the lack of standardized approaches and data collection means that they cannot effectively evaluate their performance against their peers. Coordinated efforts are required to develop tools that can be used across laboratories regardless of size or complexity.

Automated Interpretation of Blood Culture Gram Stains by Use of a Deep Convolutional Neural Network

Microscopic interpretation of stained smears is one of the most operator-dependent and time-intensive activities in the clinical microbiology laboratory. Here, we investigated application of an automated image acquisition and convolutional neural network (CNN)-based approach for automated Gram stain classification. Using an automated microscopy platform, uncoverslipped slides were scanned with a 40× dry objective, generating images of sufficient resolution for interpretation. We collected 25,488 images from positive blood culture Gram stains prepared during routine clinical workup. These images were used to generate 100,213 crops containing Gram-positive cocci in clusters, Gram-positive cocci in chains/pairs, Gram-negative rods, or background (no cells). These categories were targeted for proof-of-concept development as they are associated with the majority of bloodstream infections. Our CNN model achieved a classification accuracy of 94.9% on a test set of image crops. Receiver operating characteristic (ROC) curve analysis indicated a robust ability to differentiate between categories with an area under the curve of >0.98 for each. After training and validation, we applied the classification algorithm to new images collected from 189 whole slides without human intervention. Sensitivity and specificity were 98.4% and 75.0% for Gram-positive cocci in chains and pairs, 93.2% and 97.2% for Gram-positive cocci in clusters, and 96.3% and 98.1% for Gram-negative rods. Taken together, our data support a proof of concept for a fully automated classification methodology for blood-culture Gram stains. Importantly, the algorithm was highly adept at identifying image crops with organisms and could be used to present prescreened, classified crops to technologists to accelerate smear review. This concept could potentially be extended to all Gram stain interpretive activities in the clinical laboratory.

Waltzing around Sacred Cows on the Way to the Future

Our mostly manual, agar-based clinical microbiology laboratory is slowly but steadily being redefined by automation and innovation. Ironically, the oldest test, the Gram stain test, is still manually read and interpreted by trained personnel. In a proof-of-concept study, Smith et al. (J. Clin. Microbiol. 56:e01521-17, 2018, https://doi.org/10.1128/JCM.01521-17) used computer imaging with a deep convolutional neural network to examine and interpret Gram-stained slides from positive blood culture bottles. In light of the shortage of medical technologists/microbiologists and the need for results from positive blood culture bottles 24/7, this paper paves the way for the next innovations for the clinical microbiology laboratory of the future.

Internal quality assurance in diagnostic microbiology: A simple approach for insightful data

Given the importance of microbiology results on patient care, high quality standards are expected. Internal quality assurance (IQA) could mitigate the limitations of internal quality control, competency assessment and external quality assurance, adding a longitudinal insight, including pre- and post-analytical steps. Here, we implemented an IQA program in our clinical microbiology facilities with blind resubmission of routine samples during 22 months. One-hundred-and-twenty-one out of 123 (98.4%) serological analyses and 112 out of 122 (91.8%) molecular analyses were concordant. Among the discordances in molecular biology analyses, 6 results were low positive samples that turned out negative, likely due to stochastic repartition of nucleic acids. Moreover, one identified retranscription error led us to implement automated results transmission from the Applied Biosystems instruments to the laboratory information system (LIS). Regarding Gram stain microscopy, 560 out of 745 (75.2%) of compared parameters were concordant. As many as 67 out of 84 (79.8%) pairs of culture results were similar, including 16 sterile pairs, 27 having identical identification or description and semi-quantification and 24 only showing variations in semi-quantification with identical description or identification of colonies. Seventeen pairs had diverging identification or description of colonies. Culture was twice only done for one member of the pairs. Regarding antibiotic susceptibility testing, a major discrepancy was observed in 5 out of 48 results (10.4%). In conclusion, serological tests were highly reproducible. Molecular diagnosis also revealed to be robust except when the amounts of nucleic acids present in the sample were close to the limits of detection. Conventional microbiology was less robust with major discrepancies reaching 39.5% of the samples for microscopy. Similarly, culture and antibiotic susceptibility testing were prone to discrepancies. This work was ground for reconsidering multiples aspects of our practices and demonstrates the importance of IQA to complete the other quality management procedures.

Direct Gram staining and its various benefits in the diagnosis of bacterial infections

In the era of rapid development of molecular and other diagnostic methods, direct Gram staining (DGS) tends to remain in the background, although it can provide both microbiologists and clinicians numerous benefits. The aim of this review was to emphasize the importance of DGS for the diagnosis of many clinically important infections. A PubMed search was carried out using relevant keywords for articles published primarily since 2010. The DGS can provide early information for a timely diagnosis of infections, can reveal the causative agents of the infections even under suboptimal conditions of specimen collection, transport or identification methods, can detect the presence of rare/unusual pathogens, moreover, the method shows the specimen quality, by distinguishing between contamination and true infection, it can direct or change initial antibiotic treatment before the availability of culture results, can indicate the need of other methods for pathogen identification and, in some cases, can show the need for emergency attention such as urgent antibiotic therapy and surgical measures. Briefly, the DGS remains an easy, rapid, inexpensive and important method, which use should be encouraged in conditions of a standardized and controlled performance to avoid technical or interpretation errors.

Rapid Detection of Bloodstream Pathogens in Liver Transplantation Patients With FilmArray Multiplex Polymerase Chain Reaction Assays: Comparison With Conventional Methods

BACKGROUND

Bloodstream infection (BSI) is an important concern in transplant patients. Early intervention with appropriate antimicrobial therapy is critical to better clinical outcome; however, there is significant delay when conventional identification methods are used.

METHODS

We aimed to determine the diagnostic performance of the FilmArray Blood Culture Identification Panel, a recently approved multiplex polymerase chain reaction assay detecting 24 BSI pathogens and 3 resistance genes, in comparison with the performances of conventional identification methods in liver transplant (LT) patients. A total of 52 defined sepsis episodes (signal-positive by blood culture systems) from 45 LT patients were prospectively studied.

RESULTS

The FilmArray successfully identified 37 of 39 (94.8%) bacterial and 3 of 3 (100%) yeast pathogens in a total of 42 samples with microbial growth, failing to detect only 2 of 39 (5.1%) bacterial pathogens that were not covered by the test panel. The FilmArray could also detect additional pathogens in 3 samples that had been reported as having monomicrobial growth, and it could detect Acinetobacter baumannii in 2 samples suspected of skin flora contamination. The remaining 8 blood cultures showing a positive signal but yielding no growth were also negative by this assay. Results of MecA, KPC, and VanA/B gene detection were in high accordance. The FilmArray produced results with significantly shorter turnaround times (1.33 versus 36.2, 23.6, and 19.5 h; P < .05) than standard identification methods, Vitek II, and Vitek MS, respectively.

CONCLUSIONS

This study showed that the FilmArray appeared as a reliable alternative diagnostic method with the potential to mitigate problems with protracted diagnosis of the BSI pathogens in LT patients.

One Small Step for the Gram Stain, One Giant Leap for Clinical Microbiology

The Gram stain is one of the most commonly performed tests in the clinical microbiology laboratory, yet it is poorly controlled and lacks standardization. It was once the best rapid test in microbiology, but it is no longer trusted by many clinicians. The publication by Samuel et al. (J. Clin. Microbiol. 54:1442-1447, 2016, http://dx.doi.org/10.1128/JCM.03066-15) is a start for those who want to evaluate and improve Gram stain performance. In an age of emerging rapid molecular results, is the Gram stain still relevant? How should clinical microbiologists respond to the call to reduce Gram stain error rates?

Multicenter Assessment of Gram Stain Error Rates

Gram stains remain the cornerstone of diagnostic testing in the microbiology laboratory for the guidance of empirical treatment prior to availability of culture results. Incorrectly interpreted Gram stains may adversely impact patient care, and yet there are no comprehensive studies that have evaluated the reliability of the technique and there are no established standards for performance. In this study, clinical microbiology laboratories at four major tertiary medical care centers evaluated Gram stain error rates across all nonblood specimen types by using standardized criteria. The study focused on several factors that primarily contribute to errors in the process, including poor specimen quality, smear preparation, and interpretation of the smears. The number of specimens during the evaluation period ranged from 976 to 1,864 specimens per site, and there were a total of 6,115 specimens. Gram stain results were discrepant from culture for 5% of all specimens. Fifty-eight percent of discrepant results were specimens with no organisms reported on Gram stain but significant growth on culture, while 42% of discrepant results had reported organisms on Gram stain that were not recovered in culture. Upon review of available slides, 24% (63/263) of discrepant results were due to reader error, which varied significantly based on site (9% to 45%). The Gram stain error rate also varied between sites, ranging from 0.4% to 2.7%. The data demonstrate a significant variability between laboratories in Gram stain performance and affirm the need for ongoing quality assessment by laboratories. Standardized monitoring of Gram stains is an essential quality control tool for laboratories and is necessary for the establishment of a quality benchmark across laboratories.

Timeliness and accuracy of reporting preliminary blood culture results: a College of American Pathologists Q-probes study of 65 institutions

CONTEXT

The speed and accuracy of preliminary blood culture reports impacts patient management and outcomes.

OBJECTIVE

To evaluate the accuracy and timeliness of preliminary blood culture results among multiple laboratories.

DESIGN

Q-Probes participants collected turnaround time (TAT) data on preliminary Gram stains, compared accuracy of up to 100 preliminary to final culture Gram stain results, and described blood culture laboratory practices.

RESULTS

Sixty-four laboratories and 5031 blood cultures were evaluated. All participants used continuously monitoring blood culture systems. Median TAT from initial growth detection to notification of results was 45 minutes, with the longest component being preparation of Gram stains (median time = 25 minutes). Participants (N = 40) reporting a continuous schedule for processing blood cultures had significantly lower overall TAT (median= 37 minutes) compared with 15 participants with intermittent processing schedules (median= 124 minutes), P= .003. Time to complete Gram stain processing was lower (median time = 21 minutes) for 39 participants using continuous processing schedule compared with 14 others (median time= 67 minutes), P= .03. Goals for total TAT were used by 27 of 56 participants (48.2%). Having goals did not significantly affect TAT. A total of 4962 of 5021 Gram stain results (98.8%) agreed with final culture results. The highest discrepancy rates occurred among gram-positive bacilli (20 of 335; 6.0%) and mixed cultures (22 of 106; 20.8%).

CONCLUSIONS

This study provides benchmarks for assessing blood culture quality performance. Timeliness and accuracy of preliminary blood culture reports were excellent. However, nearly one-third of laboratories did not process blood cultures continuously. This significantly prolonged reporting results, which could affect patient outcomes.

Multicenter evaluation of the Staphylococcus QuickFISH method for simultaneous identification of Staphylococcus aureus and coagulase-negative staphylococci directly from blood culture bottles in less than 30 minutes

A novel rapid peptide nucleic acid fluorescence in situ hybridization (FISH) method, Staphylococcus QuickFISH, for the direct detection of Staphylococcus species from positive blood culture bottles was evaluated in a multicenter clinical study. The method utilizes a microscope slide with predeposited positive- and negative-control organisms and a self-reporting 15-min hybridization step, which eliminates the need for a wash step. Five clinical laboratories tested 722 positive blood culture bottles containing gram-positive cocci in clusters. The sensitivities for detection of Staphylococcus aureus and coagulase-negative staphylococci (CoNS) were 99.5% (217/218) and 98.8% (487/493), respectively, and the combined specificity of the assay was 89.5% (17/19). The combined positive and negative predictive values of the assay were 99.7% (696/698) and 70.8% (17/24), respectively. Studies were also performed on spiked cultures to establish the specificity and performance sensitivity of the method. Staphylococcus QuickFISH has a turnaround time (TAT) of <30 min and a hands-on time (HOT) of <5 min. The ease and speed of the method have the potential to improve the accuracy of therapeutic intervention by providing S. aureus/CoNS identification simultaneously with Gram stain results.

Impact of reporting gram stain results from blood culture bottles on the selection of antimicrobial agents

We assessed the usefulness of reporting direct blood Gram stain results compared with the results of positive blood cultures in 482 episodes and monitored impact on selection of antimicrobial treatment. We found that the reporting groups "Staphylococcus spp," "Pseudomonas spp and related organisms," and "yeasts" identified in this way matched perfectly with later culture identification. When the report indicated Staphylococcus spp or Pseudomonas spp and related organisms, physicians started or changed antimicrobials suitable for these bacteria more frequently than when "other streptococci" and "family Enterobacteriaceae" were reported (P < .05). Incorrect recognition of Acinetobacter spp as Enterobacteriaceae family is still the most challenging problem in this context. Gram stain results that definitively identify Staphylococcus spp, Pseudomonas spp and related organisms, and yeasts reliably can be rapidly provided by clinical laboratories; this information has a significant impact on early selection of effective antimicrobials. Further investigation is needed to assess the clinical impact of reporting Gram stain results in bacteremia.

First notification of positive blood cultures and the high accuracy of the gram stain report

When blood cultures turn positive, the attending physicians are usually notified immediately about Gram stain findings. However, information on the accuracy of Gram staining is very limited. We examined the accuracy of preliminary blood culture reports provided by a regional laboratory in an observational study including the years 1996, 2000 to 2001, and 2003. We used data from computer files and technicians' laboratory notes. The study was restricted to cultures with one morphological type. Using cultural identification as a reference, we estimated the sensitivity, specificity, and positive and negative predictive values (PPV and NPV, respectively) for the following defined morphological groups: gram-positive cocci in clusters, gram-positive cocci in chains or diplococci, gram-positive rods, gram-negative cocci, gram-negative rods, and yeasts. We further evaluated the Gram stain and wet mount findings for the most frequent bacterial species/groups. We obtained 5,893 positive blood cultures and the following results for the defined groups: sensitivity, range of 91.3 to 99.7%; specificity, 98.9 to 100%; PPV, 94.6 to 100%; and NPV, 99.0 to 100%. The sensitivity for the most frequent species was in the range 91.3 to 100%, with nonhemolytic streptococci having the lowest value (sensitivity, 91.3%; 95% confidence interval, 86.2 to 94.9%). Wet mount reports were less accurate (sensitivity of 30 to 70% for species with peritrichous motility), and Enterobacteriaceae (notably Salmonella spp.) accounted for 25% of the reports stating polar motility. In conclusion, we demonstrated a high accuracy of Gram stain reports, whereas wet mount microscopy was generally less accurate.