No abstract

Antimicrobial resistance is a major global health threat and its development is promoted by antibiotic misuse. While disk diffusion antibiotic susceptibility testing (AST, also called antibiogram) is broadly used to test for antibiotic resistance in bacterial infections, it faces strong criticism because of inter-operator variability and the complexity of interpretative reading. Automatic reading systems address these issues, but are not always adapted or available to resource-limited settings. We present an artificial intelligence (AI)-based, offline smartphone application for antibiogram analysis. The application captures images with the phone’s camera, and the user is guided throughout the analysis on the same device by a user-friendly graphical interface. An embedded expert system validates the coherence of the antibiogram data and provides interpreted results. The fully automatic measurement procedure of our application’s reading system achieves an overall agreement of 90% on susceptibility categorization against a hospital-standard automatic system and 98% against manual measurement (gold standard), with reduced inter-operator variability. The application’s performance showed that the automatic reading of antibiotic resistance testing is entirely feasible on a smartphone. Moreover our application is suited for resource-limited settings, and therefore has the potential to significantly increase patients’ access to AST worldwide.

BackgroundWar-wounded civilians in Middle East countries are at risk of post-traumatic osteomyelitis (PTO). We aimed to describe and compare the bacterial etiology and proportion of first-line antibiotics resistant bacteria (FLAR) among PTO cases in civilians from Syria, Iraq and Yemen admitted to the reconstructive surgical program of Médecins Sans Frontières (MSF) in Amman, Jordan, and to identify risk factors for developing PTO with FLAR bacteria.MethodsWe retrospectively analyzed the laboratory database of the MSF program. Inclusion criteria were: patients from Iraq, Yemen or Syria, admitted to the Amman MSF program between October 2006 and December 2016, with at least one bone biopsy sample culture result. Only bone samples taken during first orthopedic surgery were included in the analysis. To assess factors associated with FLAR infection, logistic regression was used to estimate odds ratio (ORs) and 95% confidence intervals (CI).Results558 (76.7%) among 727 patients included had ≥1 positive culture results. 318 were from Iraq, 140 from Syria and 100 from Yemen. Median time since injury was 19 months [IQR 8–40]. Among the 732 different bacterial isolates, we identified 228 Enterobacteriaceae (31.5%), 193 Staphylococcus aureus (26.3%), 99 Pseudomonas aeruginosa (13.5%), and 21 Acinetobacter baumanii (2.8%). Three hundred and sixty four isolates were FLAR: 86.2% of Enterobacteriaceae, 53.4% of Pseudomonas aeruginosa, 60.5% of S. aureus and 45% of Acinetobacter baumannii. There was no difference in bacterial etiology or proportion of FLAR according to the country of origin. In multivariate analysis, a FLAR infection was associated with an infection of the lower extremity, with a time since the injury ≤12 months compared with time > 30 months and with more than 3 previous surgeries.ConclusionsEnterobacteriaceae were frequently involved in PTO in war wounded civilians from Iraq, Yemen and Syria between 2006 and 2016. Proportion of FLAR was high, particularly among Enterobacteriaceae, regardless of country of origin.Electronic supplementary materialThe online version of this article (10.1186/s12879-019-3741-9) contains supplementary material, which is available to authorized users.

BackgroundThe cause of past plague pandemics was controversial but several research teams used PCR techniques and dental pulp as the primary material to reveal that they were caused by Yersinia pestis. However, the degradation of DNA limits the ability to detect ancient infections.MethodsWe used for the first time immuno-PCR to detect Yersinia pestis antigens; it can detect protein concentrations 70 times lower than the standard ELISA. After determining the cut-off value, we tested 34 teeth that were obtained from mass graves of plague, and compared previous PCR results with ELISA and immuno-PCR results.ResultsThe immuno-PCR technique was the most sensitive (14 out of 34) followed by the PCR technique (10 out of 34) and ELISA (3 out of 34). The combination of these three methods identified 18 out of 34 (53%) teeth as presumably being from people with the plague.ConclusionImmuno-PCR is specific (no false-positive samples were found) and more sensitive than the currently used method to detect antigens of ancient infections in dental pulp. The combination of three methods, ELISA, PCR and immuno-PCR, increased the capacity to identify ancient pathogens in dental pulp.

No abstract

No abstract

To reduce the delay in diagnosis of Q fever, we have adapted the ultrasensitive immuno-PCR method for the detection of Phase II IgM anti-Coxiella burnetii. We compared its performance to ELISA, IFA and PCR using 31 acute Q fever sera and 50 control sera. The best sensitivity was obtained by iPCR (27 out of 31) followed by PCR (18 out of 31), ELISA (12 out of 31) and IFA (10 out of 31). A specificity of 92% was found by iPCR (3 false positive out of 40), 92% for ELISA (3 false positive out of 40) whereas PCR and IFA exhibited a specificity of 100%. Among the 31 Q fever sera, we compared the four methods for the detection of the early sera sampled during the two first weeks after the onset of symptoms and found a sensitivity of 90% by iPCR, 55% for PCR, 35% for ELISA and 25% for IFA. The results presented in this study suggest that iPCR is a promising, sensitive and specific method that can be used for the early diagnosis of acute Q fever and more generally for acute infections where traditional methods lack sensitivity.

Tropheryma whipplei has long been considered as a rare bacterium causing a rare disease, Whipple's disease. However, recent advances now suggest that T. whipplei is a ubiquitous environmental bacterium that may cause gastroenteritis, commonly associated with viral pathogens. We developed an animal model to support this hypothesis. We found that orally given T. whipplei induced diarrhea in mice, without spreading into the intestines. Aggravating factors, such as damage to the intestinal mucosa, favored bacterial spreading. Indeed, bacterial presence was prolonged in stools of dextran sulfate-treated mice, and bacteria were detected in the colon. This resulted in an immune response, with T. whipplei-specific serum IgM and IgG and fecal IgA, as measured by newly introduced immuno-polymerase chain reaction technique. Our results confirm that T. whipplei is an agent causing gastroenteritis and suggest that existing mucosal damage may favor bacterial invasion of tissues.