Unresolved issues in the epidemiology and diagnosis of bacteremia: an opinion paper.

Bacteremia is an important cause of morbidity and mortality worldwide and, despite the diagnostic and therapeutic advances of the last decades, the evidence supporting many diagnostic aspects of bacteremia is scarce. Information on the epidemiological evolution of this entity is limited and many methodological aspects of blood culture collection and analysis are under discussion. Furthermore, the recommendations of the main scientific societies on many of these aspects are variable and, in many cases, have not been updated recently. In this scenario, we have arranged a series of questions on different aspects of bacteremia and reviewed the literature trying to find proper answers for them. We offer our opinion on the topics where the evidence was weak. The topics covered include epidemiological aspects of bacteremia, indications for blood culture extraction, methods for obtaining and incubating samples, or ways of transmitting results from the microbiology laboratory. We do not intend to summarize the current clinical practice guidelines, nor will we deal with the therapeutic management of this entity. The aim of this paper is to review the current perspective on the diagnosis of bacteremia with a critical approach, to point out the gaps in the literature, to offer the opinion of a team dedicated to infectious diseases and clinical microbiology, and to identify some areas of knowledge on which future studies should focus.

Estimation of missed bloodstream infections without the third blood culture set: a retrospective observational single-centre study.

OBJECTIVES: The question of whether to obtain two or three sets of blood cultures to assess the aetiology of bloodstream infections (BSIs) remains open to debate. Few studies have assessed the proportion of BSIs missed without the third blood culture set (BCS). The aim of our study was to determine the proportion of BSIs that would be missed without the third BCS in a hospital where obtaining three BCSs is the standard of care. METHODS: We performed a descriptive retrospective study in Hospital Gregorio Marañón (Madrid) from 2010 to 2013. We included all episodes of BSI in which three BCSs were systematically obtained. RESULTS: We included 4000 episodes of BSI between 2010 and 2013. Without the third BCS, we would have missed 298 (7.5%) episodes of BSI: 141 (47.3%) by gram-positive microorganisms, 147 (49.3%) by gram-negative microorganisms, and 10 (3.4%) by yeasts. In 132/298 (44.3%) of the episodes another clinical sample was obtained within a week of the BCS extraction; in 101/298 (33.9%), the same microorganism was present in a significant clinical sample other than blood. CONCLUSIONS: Our data suggest that performing a third BCS is useful, as not doing this could result in an unacceptable number of BSIs going undetected.

Isavuconazole is highly active in vitro against Candida species isolates but shows trailing effect.

OBJECTIVES: Isavuconazole is a triazole previously shown to have potent in vitro activity against Aspergillus spp., Mucorales and Candida spp. Unlike other azoles, it is unclear whether isavuconazole induces a trailing effect. We studied isavuconazole MICs for a large collection of Candida isolates from blood samples and determined the extent of the trailing effect when using the European Committee on Antimicrobial Susceptibility Testing (EUCAST) E.Def 7.3.1 method. METHODS: A total of 762 molecularly identified Candida isolates from blood samples of 743 patients admitted to hospital (January 2007 to September 2017) were evaluated and further tested for in vitro susceptibility to isavuconazole following the EUCAST E.Def 7.3.1 test method. RESULTS: C. albicans showed the highest susceptibility, followed by C. parapsilosis and C. tropicalis (geometric mean MIC 0.0029 vs. 0.0049/0.0052, respectively; p <0.001). In contrast, C. glabrata and C. krusei had significantly higher MIC values (geometric mean MIC 0.171 vs. 0.117, respectively). Isavuconazole MIC distributions were not truncated at the lowest concentration tested except for C. albicans. Overall, the mean percentage of trailing was 13.6%, but differences among species were observed: C. glabrata, C. albicans and C. tropicalis exhibited higher trailing compared to C. parapsilosis and non-Candida yeasts (p <0.001). The percentage of non-wild-type C. albicans (considering the heavy trailer isolates as wild type), C. parapsilosis and C. glabrata isolates were 1.1% (4/357), 1.5% (3/201) and 1.1% (1/86), respectively. CONCLUSIONS: Isavuconazole showed high in vitro activity against Candida spp., particularly against C. albicans. A trailing effect is commonly observed with isavuconazole, particularly with C. glabrata.

Biofilm production is not associated with poor clinical outcome in 485 patients with Staphylococcus aureus bacteraemia.

OBJECTIVES: Staphylococcus aureus biofilm may constitute a major cause of virulence. Our main objective was to analyse whether there was an association between biofilm production and poor outcome in patients with S. aureus bacteraemia. METHODS: We studied 485 S. aureus strains isolated from the blood of patients with bacteraemia from 2012 to 2015. We assessed in vitro biomass production using crystal violet assay and metabolic activity using tetrazolium salt assay. Strains were classified in tertile ranks as follows: low biomass producers, moderate biomass producers, high biomass producers, low metabolic activity, moderate metabolic activity and high metabolic activity. We excluded from analysis strains with moderate crystal violet and tetrazolium salt values. We defined poor outcome as fulfillment of one or more of the following conditions: 30-day attributable mortality, infective endocarditis, persistent bacteraemia and recurrent bacteraemia. RESULTS: Outcome was poor in 199 (41.0%) of 485 S. aureus bacteraemia episodes. The distribution of poor outcome with respect to biomass production and metabolic activity was as follows: low biomass producers, 36.6% vs. high biomass producers, 43.2% (p 0.26); and low metabolic activity, 43.5% vs. high metabolic activity, 36.2% (p 0.91). The presence of methicillin-resistant S. aureus was the only characteristic that was more likely to be present in the high metabolic activity group (17.4% vs. 39.3%, p < 0.001). CONCLUSIONS: Biofilm production, as determined by any of the methods used in the present study, is not associated with poor outcome in patients with S. aureus bacteraemia.

Should non-bacteraemic patients with a colonized catheter receive antimicrobial therapy?

OBJECTIVES: The impact of antimicrobial therapy on the outcomes of patients with colonized catheters and no bacteraemia has not been assessed. This study assessed whether targeted antibiotic therapy is related to a poor outcome in patients with positive cultures of blood drawn through a non-tunnelled central venous catheter (CVC) and without concomitant bacteraemia. METHODS: This was a retrospective study involving adult patients with positive blood cultures drawn through a CVC and negative peripheral vein blood cultures. Patients were classified into two groups: those with clinical improvement and those with a poor outcome. These two groups were compared. The outcome was considered poor in the presence of one or more of the following: death, bacteraemia or other infection due to the same microorganism, and evidence of catheter-related bloodstream infection. RESULTS: A total of 100 patients were included (31 with a poor outcome). The only independent predictors of a poor outcome were a McCabe and Jackson score of 1-2 and a median APACHE score of 5. No association was found between the use of targeted antimicrobial therapy and a poor outcome when its effect was adjusted for the rest of the variables. CONCLUSIONS: This study showed that antimicrobial therapy was not associated with a poor outcome in non-bacteraemic patients with positive blood cultures drawn through a CVC.

Uncommon aetiological agents of catheter-related bloodstream infections.

The clinical and microbiological characteristics of catheter-related bloodstream infection (CR-BSI) due to uncommon microorganisms was assessed in a retrospective case-control study over a 9-year period in a tertiary teaching hospital. Uncommon microorganisms were defined as those representing <0·5% of all CR-BSI. Diagnosis of CR-BSI required that the same microorganism was grown from at least one peripheral venous blood culture and a catheter tip culture. Thirty-one episodes of CR-BSI were identified due to 13 different genera and these accounted for 2·3% of all CR-BSI in the hospital. Although these infections were not associated with increased mortality, they occurred in patients with more severe underlying conditions who were receiving prolonged antibiotic therapy.

Is catheter-related candidemia a polyclonal infection?

Diagnosis of catheter-related candidemia (CRC) requires the simultaneous isolation of Candida spp. from both blood and catheter samples. We previously observed that in most CRC cases, the genotype of the yeast found in catheter samples is also recovered from blood. However, it is not clear whether CRC is a polyclonal infection. We prospectively studied 20 patients with CRC caused by Candida albicans, C. parapsilosis, or C. glabrata to analyze whether their infections were polyclonal. As many as 10 colonies per sample (n = 475) isolated from blood (n = 220) and catheter (n = 255) specimens were studied using species-specific microsatellite markers. Genotyping always revealed matches between the Candida spp. from blood and catheter samples. However, 15% of patients had a polyclonal pattern of infection or catheter colonization that was species specific. An additional genotype was found exclusively in the catheters of two patients infected with C. albicans, whereas an additional genotype was noted in the blood culture of a patient infected with C. parapsilosis. Considering only the presence of different genotypes in blood samples, 5% of patients had polyclonal infections. We conclude that most cases of CRC are caused by a single genotype.

Direct identification of pathogens from positive blood cultures using matrix-assisted laser desorption-ionization time-of-flight mass spectrometry.

In recent years, matrix-assisted laser desorption-ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) has proved a rapid and reliable method for the identification of bacteria and yeasts that have already been isolated. The objective of this study was to evaluate this technology as a routine method for the identification of microorganisms directly from blood culture bottles (BCBs), before isolation, in a large collection of samples. For this purpose, 1000 positive BCBs containing 1085 microorganisms have been analysed by conventional phenotypic methods and by MALDI-TOF MS. Discrepancies have been resolved using molecular methods: the amplification and sequencing of the 16S rRNA gene or the Superoxide Dismutase gene (sodA) for streptococcal isolates. MALDI-TOF predicted a species- or genus-level identification of 81.4% of the analysed microorganisms. The analysis by episode yielded a complete identification of 814 out of 1000 analysed episodes (81.4%). MALDI-TOF identification is available for clinicians within hours of a working shift, as oppose to 18 h later when conventional identification methods are performed. Moreover, although further improvement of sample preparation for polymicrobial BCBs is required, the identification of more than one pathogen in the same BCB provides a valuable indication of unexpected pathogens when their presence may remain undetected in Gram staining. Implementation of MALDI-TOF identification directly from the BCB provides a rapid and reliable identification of the causal pathogen within hours.

Value of matrix-assisted laser desorption ionization-time of flight for routine identification of viridans group streptococci causing bloodstream infections.

Phenotypic tests do not always unequivocally identify some species of viridans group streptococci (VGS). sodA sequence analysis is the most accurate method for identification, although it requires specialized personnel and has not been applied systematically in clinical microbiology laboratory routines. Matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) is emerging as a rapid alternative for bacterial identification. This study assesses the ability of MALDI-TOF and the API 20 Strep system to identify VGS isolates recovered from blood cultures using sodA sequence analysis as the reference method. All clinically significant VGS isolates recovered from blood cultures between January 2007 and January 2010 were identified by sodA sequence analysis and API 20 Strep. The strains were then tested by MALDI-TOF. Agreement between API 20 Strep/MALDI-TOF and sodA sequence analysis was determined. We examined 124 clinical isolates. Sensitivities of API 20 strep and MALDI-TOF for the species level identification of VGS isolates were, respectively, as follows: 60.5% and 73.4%. Sensitivities of API 20 strep and MALDI-TOF for the group level identification were, respectively, as follows: 70% and 94.3%. The turnaround times to identify VGS isolates by sodA sequence analysis, API 20 Strep and MALDI-TOF were 12-24, 24-48 h and 15 min, respectively. API 20 Strep cannot accurately identify all isolates of VGS. MALDI-TOF appeared to be a rapid and reliable alternative for identification of VGS strains to group level, but was not able to discriminate closely related species of certain groups.

Catheter-related bloodstream infection caused by Enterococcus spp.

The role of Enterococcus spp. as a cause of catheter-related bloodstream infections (CR-BSI) is almost unexplored. We assessed the incidence and clinical characteristics of enterococcal CR-BSI (ECR-BSI) over an 8-year period in our hospital. We performed a retrospective study (January 2003 to December 2010) in a large teaching institution. We recorded the incidence, and the microbiological and clinical data from patients with ECR-BSI. The incidence per 10,000 admissions for enterococcal BSI and ECR-BSI was 25 and 1.7, respectively. ECR-BSI was the fourth leading cause of CR-BSI in our institution (6%). A total of 75 episodes of ECR-BSI were detected in 73 patients (6% of all enterococcal BSI). The incidence of ECR-BSI increased by 17% annually (95% CI 19.0-21.0%) during the study period. Nineteen percent of ECR-BSI episodes were polymicrobial. Overall mortality was 33%. ECR-BSI is an emerging and increasingly common entity with a high mortality. This finding should be taken into account when selecting empirical treatment for presumptive CR-BSI.

Differential time to positivity (DTTP) for the diagnosis of catheter-related bloodstream infection: do we need to obtain one or more peripheral vein blood cultures?

The ideal number of blood samples to be obtained from peripheral veins (PVs) when differential time to positivity (DTTP) is being performed is an unresolved issue and most institutions obtain a single set. Our objective was to assess the number of proven central line-associated bloodstream infection (CLABSI) episodes that would have been recovered if blood had been cultured from one or two PVs. We performed a retrospective study in patients with proven CLABSI in which catheter lumens and two or more PV blood cultures were taken simultaneously. We calculated the number of episodes that would have been recovered if the culture of one or more PV blood cultures had been artificially eliminated. During a period of 4 years, we collected 60 episodes of proven CLABSI. Overall, if one PV culture had been eliminated in patients with two or three PV blood cultures, we would have documented 91.8% (p=0.362) and 96.9% (p>0.999) of episodes, respectively. If we had eliminated two PV blood cultures in patients with three PV blood cultures, 90.8% (p>0.999) of episodes would have been documented. When performing the DTTP technique to confirm CLABSI, a single paired PV blood culture was not associated with a significant number of missed CLABSI episodes.

Laboratory-acquired brucellosis: a Spanish national survey.

A retrospective postal survey was carried out among 1240 clinical microbiology laboratory workers in Spain. Overall, 75 (43 microbiologists and 32 technicians) had suffered from laboratory-acquired brucellosis (LAB). Considering the total number of replies (N=628), the rate of LAB was 11.9%. The risk of suffering from LAB was clearly related to the number of isolates of Brucella spp. per year. A major break in biosafety measures was recognized in 60 cases (80%). In nine cases, processing was considered to be secure, and in six cases, the source of infection was unknown. Diagnosis was based on serology in all cases. In 51 cases (68%), blood cultures confirmed diagnosis. A variety of regimens were used to treat the 75 LAB cases. The combination of tetracycline and streptomycin was the most commonly used regimen (in 35 patients), followed by the combination of tetracycline and rifampicin (in 19 cases). Only 10 patients (13.3%) suffered from complications. No differences in resolution were observed according to the antimicrobial regimen. Microbiological laboratory workers are still at risk of developing LAB. Improvements in safety seem to be the best means of

Risk factors for pre-eclampsia/eclampsia among working women in Mexico City.

This study examined risk factors for pre-eclampsia/eclampsia in a population-based sample of pregnant working women in Mexico City. Over a 3-month period, all women who gave birth at three major hospitals and who had worked for at least 3 months during pregnancy were interviewed. After excluding mothers with multiple gestations or infants with birth defects, and previous diagnoses of hypertension, chronic renal disease or diabetes, 131 of 2,436 women (5.4%) had been diagnosed with pre-eclampsia and/or eclampsia. The frequency was much higher among women of low socio-economic status: 12% of uninsured women (SSA) compared with 4.2% of private sector employees (IMSS) and 1.3% of public sector employees (ISSSTE). After adjusting for education, women working in services (OR = 1.68, 95% CI = 1.01, 2.81) and in retail (OR = 1.99, 95% CI = 1.18, 3.37), primiparae (OR = 2.64, 95% CI = 1.65, 4.21) and women whose pregestational weight was > or = 55 kg (OR = 2.02, 95% CI = 1.34, 3.04) were at increased risk. Efforts to develop and evaluate intervention programmes should target hospitals serving the uninsured (SSA) if reduction in the number of preventable maternal deaths in Mexico is to be achieved. Such programmes should also target service and retail workers and identify women with poor glycaemic control early in pregnancy.

Cytopathic effect inhibition assay for determining the in-vitro susceptibility of herpes simplex virus to antiviral agents.

We compare a rapid dilution method for the determination of antiviral susceptibility of herpes simplex virus (HSV) with the plaque reduction assay. A total of 84 HSV clinical isolates were studied by both methods to detect in-vitro resistance to acyclovir and foscarnet. The rapid method showed for the detection of HSV isolates resistant to acyclovir and foscarnet, a sensitivity of 96. 8% and 100% and specificity of 100% and 100%, respectively. This method provides an easy and accurate screening procedure for the susceptibility testing of HSV to antiviral agents.

Differences in curable STDs between HIV and non-HIV populations in Spain.

The pandemic impact of HIV has changed the clinical spectrum of STDs all over the world. The incidence and frequency of STDs in the different global geographic areas demonstrate the diagnostic and treatment capabilities of various local and national health systems and is simultaneously informing about the sexual behaviours of the population. The purpose of this study was to determine the frequency of curable STDs (herpes, chlamydia, gonorrhoea, syphilis, trichomoniasis) in a hospital-based STD clinic in Madrid, Spain during a 4-year period. Patients were referred mainly from the emergency department, gynecological wards, and family planning (61%) as well as from the HIV-hospital unit (31 beds) and outpatient department (39%). The total number of patients seen was 952 (243 men, 709 women) with an annual average of 238 patients per year. Of these, 139 (14.6%) were HIV-patients and 813 (85.4%) non-HIV patients. In non-HIV patients, STDs were identified in 493 cases (54.2%). In HIV-patients, STDs were diagnosed in 108 cases (77.7%; p < or = 0.001). Two or more STDs were more prevalent in HIV than non-HIV patients. The frequency of STDs in both HIV and non-HIV patients were vulvovaginal candidiasis, 47.8%:57.2%; syphilis, 11.7%:1.4% (p < or = 0.05); gonorrhea, 5.3%:3.9%; Gardnerella vaginosis, 6.3%:4.8%; genital chlamydia, 6.3%:9.06%; trichomoniasis, 17%:6.5% (p < or = 0.05); and genital herpes, 20.2%:5.3% (p < or = 0.05).

Outbreak among HIV-infected patients of Staphylococcus aureus resistant to cotrimoxazole and methicillin.

OBJECTIVE: To describe an outbreak of trimethoprim-sulfamethoxazole (cotrimoxazole)-resistant and methicillin-resistant Staphylococcus aureus (CMRSA) in a unit housing patients infected with the human immunodeficiency virus (HIV). DESIGN: Prospective study involving patients colonized or infected with CMRSA. PATIENTS: 15 hospitalized patients with cultures positive for CMRSA. METHODS: Isolates of CMRSA were collected and characterized. Molecular typing of the epidemic strains was carried out after total DNA extraction by restriction endonuclease analysis and random amplification of polymorphic DNA. RESULTS: The epidemic was brought under control with the reinforcement of nosocomial transmission measures and with systematic nasal decontamination with mupirocin of all patients admitted to the HIV unit. Molecular typing techniques showed the existence of two epidemic strains: strain A was present in the 12 patients admitted to the HIV unit and strain B in the remaining 3 patients hospitalized elsewhere. CONCLUSIONS: Cotrimoxazole may no longer be a reliable and effective alternative for glycopeptides in patients with infection caused by MRSA strains, and HIV units should be particularly alert for CMRSA strains.

Ocular tuberculosis. A prospective study in a general hospital.

Ocular tuberculosis has traditionally been considered uncommon or anecdotal. Imprecise and variable diagnostic criteria have contributed to the confusion surrounding this topic. The increase in extrapulmonary manifestations of tuberculosis during the AIDS era established the need for a prospective study of ocular involvement in patients with all types of tuberculosis using well-defined criteria. During a 15-month period, 300 cases had culture-proven tuberculosis at our institution. We randomly selected 100 for systematic ophthalmologic evaluation. Our criteria for ocular tuberculosis were divided as follows: certainty (isolation of Mycobacterium tuberculosis from ocular specimens), probability (patients with isolation of M. tuberculosis from extraocular samples, with ocular lesions not attributable to other causes that respond to anti-tuberculous treatment), and possibility (same as probability but follow-up impossible). Ocular tuberculosis was present in 18 patients (18%) of which 10 patients fulfilled probability and 8 patients fulfilled possibility criteria. Eleven of 18 patients had HIV infection. In 11 patients, ocular involvement was asymptomatic. Almost all patients (17/18) had choroiditis, and other ocular lesions included papillitis, retinitis, vitritis, vasculitis, dacryoadenitis, and scleritis. Multivariate analysis showed as risk factors independently predicting ocular involvement in patients with ocular tuberculosis the presence of miliary disease (odd ratio 43.92, p = 0.002), ocular symptoms (odds ratio 6.35, p = 0.0143), and decreased visual acuity (odds ratio 0.04, p = 0.012). We observed an unexpectedly high (18%) incidence of ocular involvement, frequently asymptomatic, in patients with tuberculosis. Miliary disease is a clear predisposing factor in both HIV-infected and noninfected populations. Ocular examination should be routinely considered in patients with proven or suspected tuberculosis.

[Conditions of maternal work and low birth weight in Mexico City]. / Condiciones de trabajo materno y bajo peso al nacer en la Ciudad de México.

OBJECTIVE: To study the association between maternal working conditions and low birth-weight in Mexico City. MATERIAL AND METHODS: Interviews of 2623 workers who gave birth in Mexico City hospitals during 1992 were analyzed. Information on the main biologic and social factors associated to low birth-weight was registered. Occupational stress was determined with the instrument designed by Karasek. Logistic regression models to evaluate the relationship between working conditions and low birth-weight were used, controlling by confounding variables. RESULTS: Low birth-weight was more frequent in workers with working periods of more than 50 h/week (OR = 1.6; 95% CI = 1.17, 2.28) and with problems at work (OR = 1.5; 95% CI = 1.0, 2.25). Lack of tangible social support was identified as a risk factor for low birth-weight (OR = 1.7; 95% CI = 1.20, 2.33). Preventive working measures such as changes in tasks, shortening of working hours and leaves of absence due to illness did not show a beneficial effects on birth-weight, except for the maternity leave of absence. Mothers with no right to this had a 2.2 higher probability of giving birth to low weight children (95% CI = 1.66, 2.93). CONCLUSIONS: These results emphasize the importance of identifying the occupational risk factors during pregnancy.