RESUMEN

Abstract Objective To identify clinical, microscopic, and biochemical characteristics that differentiate cytolytic vaginosis (CV) from vulvovaginal candidiasis (VVC). Methods The present cross-sectional study analyzed the vaginal contents of 24 non-pregnant women aged 18 to 42 years who were attended at the Genital Infections Clinic at Centro de Atenção Integral à Saúde da Mulher da Universidade Estadual de Campinas (CAISM-UNICAMP). They were diagnosed either with (CV = 8, VVC = 8) or without vulvovaginitis or vaginal dysbiosis (controls). The socio-demographic, clinical, and gynecological data were obtained from a detailed patient interview. Samples of the vaginal contents were collected for analysis of vaginal pH, gram stain, and specific fungal culture. The Kruskal-Wallis and Fisher exact tests were used to compare the differences between the groups. Odds ratios were used to compare the categorical variables. The significance level was considered at p < 0.05. Results Both women with CV and VVC had a lumpy vaginal discharge (p = 0,002) and vaginal hyperemia (p = 0.001), compared with controls. The inflammatory process was more intense in the VVC group (p = 0.001). In the CV group, there was statistical significance for the lactobacillus amount (p = 0.006), vaginal epithelium lysis (p = 0.001), and vaginal pH (p = 0.0002). Conclusion Cytolytic vaginosis and VVC diagnoses rarely differ on clinical characteristics but have different laboratorial findings. The present study highlights the importance of conducting an accurate investigation through laboratory tests rather than clinical criteria to avoid misdiagnosis.

Resumo Objetivo Identificar características clínicas, microscópicas e bioquímicas que diferenciam a vaginose citolítica (VC) da candidíase vulvovaginal (CVV). Métodos O presente estudo de corte transversal analisou o conteúdo vaginal de 24 mulheres não grávidas, com idades entre 18 e 42 anos, atendidas no ambulatório de Infecções Genitais do Centro de Atenção Integral à Saúde da Mulher da Universidade Estadual de Campinas (CAISM-UNICAMP). Elas foram diagnosticadas com (CV = 8, CVV = 8) ou sem vulvovaginite ou disbiose vaginal (controles = 8). Os dados sociodemográficos, clínicos e ginecológicos foram obtidos em uma entrevista detalhada do paciente. Amostras do conteúdo vaginal foram coletadas para análise do pH vaginal, coloração de Gram e cultura específica de fungos. Os testes exatos de Kruskal-Wallis e Fisher foram utilizados para comparar as diferenças entre os grupos. A razão de chances foi utilizada para comparar as variáveis categóricas. O nível de significância considerado foi de p < 0,05. Resultados As mulheres com VC e CVV apresentaram corrimento vaginal irregular (p = 0,002) e hiperemia vaginal (p = 0,001), em comparação aos controles. O processo inflamatório foi mais intenso no grupo CVV (p = 0,001). No grupo VC, houve significância estatística para a quantidade de lactobacilos (p = 0,006), lise do epitélio vaginal (p = 0,001) e pH vaginal (p = 0,0002). Conclusão Os diagnósticos de VC e CVV raramente diferem nas características clínicas, mas apresentam achados laboratoriais diferentes. O presente estudo destaca a importância de conduzir uma investigação precisa por meio de testes laboratoriais, em vez de critérios apenas clínicos, a fim de evitar erros de diagnóstico.

Asunto(s)

RESUMEN

Introducción: la detección del género Corynebacterium en flujo vaginal representa un reto para el estudio de su posible papel patogénico en el tracto genital femenino. La coloración de Gram es la técnica recomendada para la detección de corinebacterias en muestras clínicas; no obstante, algunas características biológicas de estas bacterias dificultan la identificación por medio de esta. Objetivo: evaluar el desempeño de la coloración de Albert para la identificación de Corynebacterium spp. en muestras de flujo vaginal. Materiales y métodos: se evaluó la tinción de Albert como prueba diagnóstica para la identificación de Corynebacterium spp. en 451 flujos vaginales en comparación de la coloración de Gram. Se calculó la sensibilidad, la especificidad, los valores predictivos, las razones de verosimilitud y el índice J de Youden con el software Epidat 3.1. Resultados: la coloración de Albert presentó sensibilidad y especificidad de 61,5% y 74,6%, respectivamente, valor predictivo positivo de 78,3%, valor predictivo negativo de 56,5%, razón de verosimilitud positiva de 2,4 y negativa de 0,5 y un Índice de Youden de 0,4. Conclusiones: la tinción de Albert no mostró un mejor rendimiento que la coloración de Gram en la identificación del género Corynebacterium en muestras de flujo vaginal. Por lo tanto, se recomienda el uso en conjunto de las coloraciones de Gram y Albert para la identificación de corinebacterias en flujo vaginal. (AU)

Introduction: The detection of Corynebacterium genus in vaginal discharge represents a challenge for the study of its possible pathogenic role in the female genital tract. Gram's stain is recommended for corynebacteria detection in clinical samples, however, some biological characteristics of these bacteria difficult their identification by this technique. Objetive: to evaluate Albert's stain performans to identify Corynebacterium spp. in vaginal discharge. Materials and methods: Albert's stain was evaluated as a diagnostic test for the identification of Corynebacterium spp. in 451 vaginal swabs in comparison to Gram's stain. Sensitivity, specificity, predictive values, likelihood ratio, as well as Youden's J index were calculated using Epidat software 3.1 version. Results: Albert's stain had a sensitivity and specificity of 61.5% and 74.6%, respectively, positive predictive value of 78.3 %, negative predictive value of 56.5 %, positive likelihood ratio of 2.4 and negative likelihood ratio of 0.5, as well as Youden index of 0.4. Conclusions: Albert's stain did not show better performance than Gram's stain for identification of corynebacteria in vaginal swabs. Thereforeit is recommended the concomitant use of the Gram and Albert stains for the identification of corynebacteria in vaginal discharge. (AU)

Asunto(s)

RESUMEN

La vaginosis citolítica, descrita hace 30 años como citólisis de Döderlein, es frecuente en mujeres en la edad reproductiva y, por las características de flujo vaginal blanquecino y síntomas clínicos, es indistinguible de la vulvovaginitis micótica. Se presenta el caso de una paciente, con diagnóstico clínico presuntivo de vulvovaginitis por Candida spp. a repetición, tratada empíricamente con antifúngicos por año y medio sin ninguna mejoría. Luego de estudios microbiológicos, la coloración de Gram, demostró la presencia de 50 bacilos grampositivos por campo y abundantes núcleos celulares desnudos. El cultivo resultó puro para Lactobacillus spp., lo que permitió confirmar el diagnóstico de vaginosis citolítica. La paciente fue tratada con ampicilina-sulbactam y no ha vuelto a presentar recidivas. En conclusión, es fundamental determinar el pH vaginal de las pacientes en la consulta, así como practicar una coloración de Gram de la secreción vaginal para poner en evidencia los cambios celulares por el exceso de ácido en la vagina y así evitar tratamientos antifúngicos innecesarios que acrecentarán los trastornos de la microbiota vaginal.

Cytolytic vaginosis, described 30 years ago as Döderlein cytolysis, is common in women of reproductive age and, due to the characteristics of whitish vaginal discharge and clinical symptoms, is indistinguishable from mycotic vulvovaginitis. We describe the case of a patient with presumptive clinical diagnosis of recurrent vulvovaginitis by Candida spp. treated empirically with antifungal agents for one and a half years without improvement. After microbiological studies, Gram staining demonstrated the presence of 50 Gram-positive bacilli per field and abundant nude cell nuclei. The culture recovered pure Lactobacillus spp. which permitted the diagnosis of cytolytic vaginosis. The patient was treated with ampicillin-sulbactam and since, has not had recurrences. In conclusion, it is essential to examine the pH of the patient vaginal discharge, as well as to practice a Gram staining of the vaginal secretion to demonstrate the cellular changes produced by the excess of acid in the vagina and therefore avoid unnecessary antifungal treatments that will produce undue changes of the vaginal microbiota.

RESUMEN

RESUMEN Objetivos. Determinar la prevalencia de vaginosis bacteriana (VB) y factores asociados en mujeres peruanas de 18 a 29 años de edad en 20 ciudades a partir de datos del proyecto PREVEN. Materiales y métodos. Estudio de tipo transversal, la definición de VB se realizó previa selección de una muestra de secreción vaginal en una lámina portaobjetos. Las láminas fueron teñidas usando la tinción Gram para ser observadas al microscopio usando el puntaje de Nugent, el diagnóstico de VB se aplicó a los puntajes 7-10. Se estimaron razones de prevalencias (RP) y sus intervalos de confianza al 95% (IC 95%) mediante el uso de modelos lineales generalizados. Resultados. Un total de 6322 mujeres contestaron la encuesta epidemiológica y proporcionaron muestras vaginales. La prevalencia de VB fue de 23,7% (IC95%: 22,6-24,7) y se asoció con tener un mayor número de parejas sexuales en los últimos 12 meses (RP: 1,22, IC 95%: 1,03-1,44, p=0,020; para dos parejas y RP: 1,46, IC 95%: 1,23-1,74, p<0,001 para tres o más parejas), no usar condón en la última relación sexual (RP: 1,16, IC 95%: 1,01-1,34, p=0,034), ser residente de la sierra (RP: 1,18, IC 95%: 1,05-1,31, p=0,004) y tener flujo vaginal anormal o con mal olor (RP: 1,20, IC 95%: 1,09-1,33, p<0,001). Conclusiones. La alta prevalencia de VB encontrada remarca la necesidad de fortalecer los servicios de salud para la detección y tratamiento de esta condición.

ABSTRACT Objetives. To determine the prevalence of bacterial vaginosis (BV) and associated factors among 18-29-year-old women in 20 Peruvian cities using PREVEN project data. Materials and Methods. In this cross-sectional study, BV was defined using previously provided vaginal discharge samples on slides, which were Gram stained and observed under a microscope to determine the Nugent scores. A BV diagnosis was applied to samples with scores of 7-10. Prevalence ratios (PR) and 95% confidence intervals (95% CI) were estimated using generalized linear models. Results. A total of 6,322 women participated in the epidemiological survey and provided vaginal swabs. The prevalence of BV was 23.7% (95% CI: 22.6-24.7) and was associated with a greater number of sexual partners in the last 12 months (PR: 1.22, 95% CI: 1.03-1.44, p=0.020 for two partners; PR: 1.46, 95% CI: 1.23-1.74, p<0.001 for three or more partners), not using a condom during last intercourse (PR: 1.16, 95% CI: 1.01-1.34, p=0.034), being a sierra resident (PR: 1.18, 95% CI: 1.05-1.31, p=0.004), and having abnormal vaginal discharge or a bad smell (PR: 1.20, 95% CI: 1.09-1.33, p<0.001). Conclusions. The high prevalence of BV highlights the need to strengthen health services aimed at the detection and treatment of this condition.

Asunto(s)

RESUMEN

BACKGROUND: The Gram stain can be used to direct initial empiric antimicrobial therapy when complete culture is not available. This rapid test could prevent the initiation of inappropriate therapy and adverse outcomes. However, several studies have attempted to determine the value of the Gram stain in the diagnosis and therapy of bacterial infection in different populations of patients with ventilator-associated pneumonia (VAP) with conflicting results. The objective of this study is to evaluate the accuracy of the Gram stain in predicting the existence of Staphylococcus aureus infections from cultures of patients suspected of having VAP. METHODS: This prospective single-center open cohort study enrolled 399 patients from December 2005 to December 2010. Patients suspected of having VAP by ATS IDSA criteria were included. Respiratory secretion samples were collected by tracheal aspirate (TA) for standard bacterioscopic analysis by Gram stain and culture. RESULTS: Respiratory secretion samples collected by tracheal aspirates of 392 patients were analyzed by Gram stain and culture. When Gram-positive cocci were arranged in clusters, the sensitivity was 68.4%, specificity 97.8%, positive predictive value 88.1% and negative predictive value 92.8% for predicting the presence of Staphylococcus aureus in culture (p < 0.001). CONCLUSIONS: A tracheal aspirate Gram stain can be used to rule out the presence of Staphylococcus aureus in patients with a clinical diagnosis of VAP with a 92.8% Negative Predictive Value. Therefore, 7.2% of patients with Staphylococcus aureus would not be protected by an empiric treatment that limits antimicrobial coverage to Staphylococcus aureus only when Gram positive cocci in clusters are identified.

Asunto(s)

RESUMEN

La identificación de microorganismos en tejido es esencial para reconocer un proceso infeccioso. Inicialmente, mediante la coloración de hematoxilina-eosina, se puede identificar el patrón de inflamación asociado y luego, a través de tinciones basadas en plata y la tinción de Gramde tejido se visualizan los microorganismos. La tinción de Gram no solo sirve para bacterias, sino también para algunos hongos y parásitos; no obstante, esta técnica tiene algunos inconvenientes, como la contaminación con otros microorganismos y la imposibilidad de visualizar algunas bacterias, entre ellas Legionella pneumophila, Leptospira spp y Bartonella spp. En este artículode revisión se describirán los fundamentos del Gram de tejido, su contribución en el diagnóstico de infecciones como herramienta adicional para el reconocimiento de microorganismos, y sus limitaciones.

The identification of microorganisms in tissue is pivotal to recognize infectious processes.At first, the hematoxylin-eosin stain is used to identify the pattern of inflammation associated; after that, microorganisms are seen through Gram or silver stains. Gram Stain of tissue biopsy not only stains bacteria, but also a number of fungus and parasites. However, this technique has some disadvantages, such as contamination with other microorganisms, and lack of stain of some bacteria, including Legionella pneumophila, Leptospira spp and Bartonella spp. This review articleaims to describe the fundamentals of Gram stain of tissue biopsy and its assistance in infectious diagnosis as an additional tool for recognition of microorganisms, as well as its limitations.

Asunto(s)

RESUMEN

Background: Vaginal infection is the commonest cause of genital symptoms and has obstetric and gynecological implications. Aim: To compare the Nugent and Spiegel methods for the diagnosis of bacterial vaginosis (BV) and to analyze discordant specimens using Ison and Hay (Ison/Hay) criteria. Material and Methods: After discardingcases with Candidiasis, deficientspecimens or those lacking bacteria, a total of348 Gram-stained smears vaginal specimens receivedfor the diagnosis of BV, were analyzed. Results: Vaginal microbiota was classified as normal in 203 and 237 samples (58 and 68 percent of samples), accordingto Nugent and Spiegel criteria, respectively One hundred andfive (30 percent) and 111 samples (32 percent), were classified as VB accordingto Nugent and Spiegel criteria, respectively. Both criteria were concordant in 308 samples (88.5 percent). The 40 (11.5 percent) discordant specimens were classified as intermedíate microflora by the Nugent system and as normal or BV by Spiegel. Among these, the Ison/Hay procedure identified four categories of microbiota. Ten (25 percent) specimens were classified as grade II microbiota, confirming their categorization by Nugent as intermedíate microbiota, six (15 percent) were classified in the III category, confirming the diagnosis ofBV by Spiegel, 13 (32.5 percent) corresponded to the category III, that does not exist in the Nugent and Spiegel categorization systems. Finally, 11 specimens could not be assigned to one category due to microscopic limitations to distinguish bacterial morphotypes. Conclusions: The systems proposed by Spiegel, Nugent and Ison I Hay are comparable for the diagnosis of BV. However, we recommend the use of Ison/ Hay procedure to evalúate vaginal microbiota, due to its wider range of categories, allowing a better discrimination ofthe vaginal microbiota.

Asunto(s)

RESUMEN

Introducción: la utilidad del Gram y cultivo de esputo para el manejo y seguimiento de los pacientes con neumonía adquirida en la comunidad (NAC) es aun controversial. Objetivo: describir la frecuencia de cambio del manejo antibiótico de acuerdo con el resultado del cultivo de esputo en enfermos hospitalizados con NAC. Tipo de estudio: descriptivo de corte transversal. Métodos: pacientes ingresados al servicio de medicina interna del Hospitalde San José con diagnóstico clínico y radiológico de NAC que no requirieran manejo en UCI. Se incluyeron 86 casos en un período de cuatro meses a quienes se les aplicó al ingreso la escala de pronóstico CURB 65, la presencia o no de comorbilidades, la positividad del Gram y cultivo de esputo, la estancia hospitalaria y el cambio de antibiótico de acuerdo con el resultado del cultivo de esputo si la muestra era significativa. Resultados: el promedio de edad de la población fue 58.8 años (DS 21.7). El 45% tenía comorbilidades al ingreso y el 37% algún tipo de complicación asociada con la NAC. En la escala de CURB-65 el 52% tenía un punto, dos puntos el 27%, tres el 3.5 % y con cuatro el 2%. La respuesta a la terapia antibiótica se observó en el 85% y la frecuencia del cambio antibiótico según el reporte de cultivo ocurrió en el 10% (6/62). La mediana de días de estancia hospitalaria fue de cinco. El 95% egresó del hospital y el 5% restante fue trasladado a otro servicio. Conclusiones: el Gram y cultivo de esputo son ayudas diagnósticas de bajo costo y útiles en pacientes con NAC clasificados segúnlas escalas de severidad de alto riesgo o neumonías severas, en donde el aislamiento del germen es crucial paradirigir la terapia antibiótica; en neumonías no severas no han demostrado mayor utilidad ya que pocas veces hay aislamiento del germen y no generan cambios en el manejo del paciente. Se debe asegurar la adecuada toma y procesamiento para aumentar la probabilidad de aislamiento.

Introduction: the applicability of sputum Gram stain and culture to define management and follow-up of patients with community-acquired pneumonia (CAP) remains controversial. Objective: to describe the frequency of changes in antimicrobial therapy based on results of sputum culture in hospitalized patients with CAP. Type of study: this is a descriptive cross sectional study. Methods: patients admitted to the Internal Medicine Department at the San José Hospital with radiologic and clinic diagnosis of CAP who did not require ICU care were selected. Eighty-six (86) cases in a four-month period were included. They were graded with the CURB-65 prediction scale upon admission considering comorbid factors, positive sputum Gram stain and culture, hospital stay and change of antimicrobial therapy based on the result of the sputum culture when sample was substantial. Results: the mean age of the study population was 58,8 years (SD 21,7). Forty-five percent (45%) had comorbidities upon admission and 37% had some type of complication associated to CAP. In the CURB-65 scale 52% rated one point, 27% rated two points, 3,5 % three points and 2% four points. Response toantimicrobial therapy was observed in 85% and a change of antimicrobial regime based on culture results occured in 10% (6/62). The median number of days in hospital was five. Ninety-five percent (95%) were discharged from the hospital and the remaining 5% were transferred to another department. Conclusions: sputum Gram stain and culture are low-cost diagnostic tools and are useful in patients with CAP classified acording to the high-risk severity scales or severe pneumonias, in which isolation of the pathogen is crucial to tailor antimocrobial therapy; in non-severe pneumonias they have not shown mayor applicability for in a very small number of cases the etiologic organism is isolated and no changes in patient management are generated...

Asunto(s)

RESUMEN

Introducción: La neumonía asociada a la ventilación (NAV) es una complicación frecuente de la ventilación mecánica (VM) y se asocia a alta mortalidad. Su adecuado manejo se basa en detección precoz y terapia con antibióticos (ATB) empírica eficaz, siendo esenciales los métodos que permitan predecir la NAV y ayuden a seleccionar los ATB. Objetivo: valorar la eficacia del examen directo por tinción de Gram de muestras obtenidas del tracto respiratorio en la detección de NAV y como ayuda a la selección de ATB. Materiales y Métodos: Se incluyeron prospectivamente 95 episodios de sospecha de NAV en 69 pacientes, considerando sólo los casos sin modificación de los ATB por más de 48 hs o sin ATB. Las muestras fueron obtenidas por medio de Lavado bronquioloalveolar a través de Fibro- broncoscopia (BAL) o minibal por catéter envainado a ciegas (miniBal). La confirmación se efectuó por el análisis bacteriológico cuantitativo (> 10 elevado a la cuarta y > 10³ para el BAL y minibal respectivamente). Se calculó la sensibilidad (S), especificidad (E), valor predictivo positivo (VPP) y negativo (VPN) de la técnica de tinción directa de Gram, considerando como referencia el resultado microbiológico del cultivo del BAL o mini-BAL. Se comparó además la concordancia entre el tipo de germen observado en la tinción directa de Gram, bacilo Gram negativo (BGN) o coco Gram positivo (CGP) y el tipo de bacteria aislada en el cultivo. Se analizaron también las variables predictivas específicas o cualitativas, esto es la eficacia del Gram en predecir VAP por CGP o por BGN. Resultados: Se diagnosticó NAV en 52 casos. La S de la tinción de Gram fue de 60%, E: 100%, VPP Positivo: 100%, VPN: 67%. En cuanto a la eficacia para predecir VAP por BGN se observó: S: 52%, E: 100%, VPP: 100%, VPN: 72%. Para CGP: s: 91 %, E: 100%, VPP 100% y VPN: 99%. Conclusión: Si bien un examen bacteriológico directo por tinción de Gram predice presencia de VAP, su VPN es bajo por lo que un directo negativo...

Introduction: Ventilator Associated Pneumonia (VAP) is a frequent complication in mechanically ventilated patients, and is associated with a high mortality. Early diagnosis and appropriate use of empiric antibiotic therapy (ATB) is essential to its treatment. Tests that help to make an early diagnosis and select the initial ATB are useful. Objective: to evaluate the accuracy of the Gram stain examination of respiratory tract samples in the prediction of VAP and its potential role in the selection of empiric ATB. Materials and methods: prospective evaluation of 95 episodes in 69 patients with clinical suspicion of VAP who had not had any changes in the ATB for more than 48 hours or had not been treated with ATB. Samples were taken by bronchoalveolar lavage (BAL) through flexible bronchoscopy or blind mini bronchoalveolar lavage with protected catheter. VAP was confirmed when the quantitative cultures showed =10 to the fourth ufc/ml for BAL and =10³ ufc/ml for mini BAL. A comparison was made between the type of bacteria observed by the direct Gram staining (Gram negative bacteria or Gram positive bacteria) and the type of bacteria isolated by culture. Specific or qualitative predictive variables were analyzed to determine the value of Gram staining to predict a Gram-negative VAP or Gram-positive VAP. Results: VAP was confirmed in 52 cases. The sensitivity of the Gram stain for VAP was 60%, specificity 100%; positive predictive value: 100% and negative predictive value: 67%. The sensitivity of Gram negative bacteria for VAP was: 52%; specificity: 100%; positive predictive value: 100%; negative predictive value: 72%. The sensitivity of Gram positive bacteria for VAP was: 91 %; specificity: 100%; positive predictive value: 100%; negative predictive value: 99%. Conclusions: A positive finding in the Gram staining examination of respiratory tract samples was highly predictive of VAP but a negative result did not mean absence of VAP, because the negative...

Asunto(s)

RESUMEN

OBJETIVO: evaluar la utilidad del citoquímico de orina y la coloración de Gram en el diagnóstico de la infección del tracto urinario en pacientes hospitalizados. DISEÑO Y MUESTRA: estudio de corte transversal; 10.917 urocultivos. MÉTODOS: se hicieron el citoquímico de orina y la coloración de Gram de la orina sin centrifugar según la solicitud de los médicos tratantes. Cuando no se solicitó este último examen, la coloración de Gram se hizo del sedimento urinario. Se compararon los resultados obtenidos con el resultado del urocultivo. Resultados: se obtuvo el resultado del citoquímico de orina en 6.762 de las muestras. Se solicitaron 2.762 coloraciones de Gram de orina sin centrifugar, y en 8.008 orinas se hizo la coloración de Gram del sedimento urinario. En comparación con el urocultivo, el citoquímico de orina presentó sensibilidad del 99%, especificidad del 51%, valor predictivo positivo del 60%, y valor predictivo negativo del 98%. La coloración de Gram de la orina sin centrifugar tuvo sensibilidad, especificidad y valores predictivos positivo y negativo de 98%, 73%, 82% y 97%, respectivamente. Para la coloración de Gram del sedimento urinario los valores correspondientes a la sensibilidad, especificidad y valores predictivos positivo y negativo fueron 95%, 78%, 67%, y 97%. CONCLUSIONES: el citoquímico de orina, bajo los criterios propuestos, y la coloración de Gram tienen un alto valor predictivo negativo, que permitiría utilizarlos como pruebas de tamización para determinar la necesidad de solicitar el urocultivo.

OBJECTIVE: To evaluate the usefulness of urinalysis and of urine Gram stain in the diagnosis of urinary tract infection in hospitalized patients. DESIGN AND SAMPLE: Urinalysis and Gram stain of non centrifuged urine were carried out according to the requests of physicians, in specimens submitted to the laboratory for urine culture. When this modality of Gram stain had not been requested, it was done with urine sediment. Results were compared with those of the culture. RESULTS: Ten thousand nine hundred and seventeen urine cultures were carried out, as well as 6.762 urinalyses, 2.762 Gram stains of non centrifuged urine, and 8.008 Gram stains of urine sediments. In comparison with the urine culture, urinalyses had 99% sensitivity, 51% specificity, 60% positive predictive value, and 98% negative predictive value. Gram stain of non centrifuged urine had 98% sensitivity, 73% specificity, 82% positive predictive value, and 97% negative predictive value. Gram stain of urine sediment had 95% sensitivity, 78% specificity, 67% positive predictive value, and 97% negative predictive value. CONCLUSIONS: Urinalysis, following well defined criteria, and Gram stain have high negative predictive values, that may allow to use them as screening tests to determine the need to request urine culture.

Asunto(s)