Paediatric community-acquired pneumonia (CAP) is common in North America. It is often treated with beta-lactam antimicrobials targeting S. pneumoniae, the most important cause of CAP in young children. Current guidelines recommend 10 days of therapy for paediatric CAP, regardless of severity; in contrast, mild CAP in adults is routinely treated with only 5 days of antimicrobials. There have been no definitive studies of 5-day vs. 10-day therapy for CAP in children. The objective of this thesis was to conduct a pilot RCT comparing 5 to 10 days of amoxicillin for the treatment of mild paediatric CAP and then design the multicentre follow-up trial. Children aged 6 months - 10 years with no significant past medical history presenting to the McMaster Children's Hospital emergency department with mild CAP were eligible for enrollment. All participants were randomized to either 10 days high-dose amoxicillin (90 mg/kg/day divided bid) or 5 days of high-dose amoxicillin + 5 days placebo. The primary outcome was clinical cure at day 14-18 post-enrollment. In total, 61 participants were recruited. The median participant age was 2.64 y. Only 60% of chest radiographs were reported by the radiologist as showing evidence of pneumonia. There were six treatment failures; one participant failed to defervesce on day 4, one participant had recurrent fevers leading to re-presentation to the emergency, and the other four participants did not meet clinical cure criteria but were essentially well at the time of follow-up. Study blinding has been maintained. The majority of previously healthy children with mild CAP who are well enough to be treated as outpatients appear to do well, regardless of duration of antimicrobial treatment. Feasibility and safety of the trial protocol have been demonstrated; the follow-up multicentre trial is slated to begin in mid-2015. / Thesis / Master of Science (MSc)

Factors that affect liquid chemical intervention methods of controlling pathogens on fresh-cut produce were investigated. The relationship between produce tissue structure (intercellular space, cell size, and cell distribution) and the sanitizing effectiveness of liquid chemical treatment was studied. Experiments determined if sanitizer contact with bacteria could be improved through the use of surfactants and different application methods (drop application method, negative pressure differential, and sonication). To test these factors, a model sanitizer, H2O2, and a model microorganism: Salmonella Typhimurium, along with various fresh-cut produce (apple, pear, carrot, and potato) were tested. Microscopic analysis revealed a very complicated pore structure consisting of irregular capillaries. S. Typhimurium was found to survive in all produce tested, and washing did not significantly reduced inoculated bacteria regardless of the bacterial incubation time or produce type. The results showed that a 3% H2O2 solution reduced S. Typhimurium in produce and the solution’s efficiency varied in the following descending order: potato>apple>carrot>pear. In seven min treatments, bacteria were reduced by 2.5 CFU/ml in potato, 2.3 CFU/ml in apple, 1.5 CFU/ml in carrot, and 0.7 CFU/ml in pear. There was no direct evidence on how intercellular space, its percentage or cellular distribution and shape affected efficiency, but some possibilites were discussed. The rate and extent of liquid penetration, and how varying pore diameter in each cell or air space prevent complete chemical treatment penetration were also analyzed. It was determined that bacterial density has a slight effect in bacterial reduction but this depends on type of produce inoculated. The use of surfactants did not improve bacterial reduction in either washing or chemical treatments, and neither did the use of drop application method or temperature differential. On the other hand, applying the chemical treatment with a surfactant while using a sonicator did improve the treatment’s efficiency. This thesis provides a number of factors to be considered when designing a chemical treatment and a guideline for further research in areas such as rate and extent of liquid chemical treatment penetration into fresh-cut produce.

Abstract Prosthetic joint infection (PJI) is one of the most devastating complications of hip or knee arthroplasty. Treatment options for PJI include prosthesis retention, prosthesis exchange and salvage procedures (e.g., arthrodesis or amputation). The purpose of this retrospective study is to assess the impact of shortening antibiotic treatment durations in PJIs treated with debridement, antibiotics and implant retention (DAIR) and to evaluate the predictors of DAIR treatment failure for PJIs. A second aim is to evaluate the outcomes and reimplantation microbiologies of PJIs treated with two-stage revision. The main data consist of 197 patients with PJI treated in Oulu University Hospital, Finland, between February 2001 and August 2009. Patients were identified retrospectively using the hospital’s patient databases. The study shows that, if antibiotic treatments for PJI are completed as planned, then DAIR treatment success rates (88%) are excellent. Prolonging antibiotic treatment over three months in total knee arthroplasty PJIs or over two months in total hip arthroplasty PJIs does not seem to offer any additional benefits. The failure of DAIR in the treatment of PJI is independently associated with high leucocyte counts at admission and with ineffective empirical antibiotics. Rifampin combination therapy, especially the combination of rifampin and ciprofloxacin, is significantly associated with successful DAIR treatments of staphylococcal PJIs. Our study also shows that a six-week course of antibiotics between stages is sufficient for treating PJIs with two-stage revision. Positive reimplantation cultures do not seem to be associated with significantly worse outcomes. / Tiivistelmä Tekonivelinfektio on yksi vakavimmista lonkan ja polven tekonivelleikkauksen komplikaatioista. Tekonivelinfektion hoitovaihtoehtoja ovat tekonivelen säästävä hoito, tekonivelen vaihto ja ns. salvage-toimenpiteet (esimerkiksi jäykistys tai amputaatio). Tämän retrospektiivisen tutkimuksen tavoitteena oli selvittää lyhennetyn mikrobilääkehoidon toimivuutta tekonivelinfektion säästävässä hoidossa. Lisäksi pyrittiin löytämään tekonivelinfektion hoidon epäonnistumiselle altistavia tekijöitä. Tavoitteena oli myös selvittää hoidon tuloksia silloin, kun hoidetaan tekonivelinfektiota kaksivaiheisella revisiolla, sekä tutkia revision toisessa vaiheessa otettavien mikrobinäytteiden merkitystä hoidon onnistumiselle. Tutkimusaineisto koostui 197 potilaasta, joilta hoidettiin tekonivelinfektiota Oulun yliopistollisessa sairaalassa helmikuun 2001 ja huhtikuun 2009 välisenä aikana. Potilastiedot saatiin sairaalan hoitotietojärjestelmästä. Tutkimuksessa todettiin, että lyhyemmällä hoitoajalla voidaan saavuttaa erinomaiset hoitotulokset (88 %) tekonivelinfektion säästävässä hoidossa, jos suunniteltu antibioottihoitoaika voidaan toteuttaa. Yli kahden kuukauden (lonkan tekonivelinfektio) ja kolmen kuukauden (polven tekonivelinfektio) hoitoajasta ei näytä olevan hyötyä säästävässä hoidossa. Lisäksi todettiin, että sairaalaantulovaiheessa mitattu veren leukosyyttiarvo > 10×109/l ja tehoton empiirinen antibiootti ovat itsenäisesti hoidon epäonnistumiselle altistavia tekijöitä. Rifampisiini-yhdistelmähoidon, erityisesti rifampisiini yhdistettynä siprofloksasiiniin, todettiin olevan merkittävästi yhteydessä hoidon onnistumiseen silloin, kun hoidetaan stafylokokki-infektiota tekonivelen säästävällä hoidolla. Tutkimuksemme osoitti myös, että kuuden viikon antibioottihoito on riittävä hoidettaessa tekonivelinfektiota kaksivaiheisella revisiolla. Positiivinen mikrobiviljelynäyte toisen vaiheen leikkauksessa ei näytä olevan yhteydessä huonompaan hoitotulokseen.

Cette étude s'est concentrée sur la performance d'un filtre antimicrobien en fibres de polypropylène contenant de la pyrithione de zinc (PP/ZPT) au laboratoire et comparée à celle d'un filtre similaire (PP) avec la même classification F7 (EN779:2002). La performance de filtration à l'échelle du laboratoire des 2 filtres testés pendant le colmatage avec des particules PM10 a été quantifiée dans un dispositif expérimental avec mesure de la perte de charge du filtre et comptage des particules en amont et en aval des filtres. La croissance microbienne sur des filtres neufs et usagés, tous deux contaminés par aérosolisation avec un consortium microbien composé de deux bactéries (Staphylococcus epidermidis Gram positif, Serratia marcescens Gram négatif et spores fongiques (Penicillium chrysogenum). L'influence de trois paramètres sur la survie microbienne sur les filtres a été examinée : l'humidité relative de l'air, la présence ou l'absence de pyrithione de zinc (ZPT) en tant que substances antimicrobiennes et la présence de particules organiques. Des analyses quantitatives par unité formant colonies ont été utilisées pour déterminer la survie après 8 jours du consortium bactéries-champignons collecté par le filtre. Les deux filtres ont présenté des performances de filtration similaires en termes de variation de perte de charge et d'efficacité de collecte des particules pendant leur colmatage avec des particules PM10, ce qui signifie que le traitement antimicrobien n'a pas dégradé les performances de filtration du filtre. A faible valeur d'humidité de conditionnement (50% RH), avec des filtres neufs ou usagés, avec ou sans traitement antimicrobien, la population microbienne sur les filtres diminue et éventuellement ne survivra pas (Serratia). Lorsque l'humidité relative du conditionnement est élevée (90% RH), les bactéries ne se développent pas sur les filtres neufs, et seuls les champignons ont pu se développer. L'effet du traitement antimicrobien avec la pyrithione de zinc est confirmé pour les filtres neufs, en particulier en ce qui concerne les champignons Penicillium. Pour les filtres usagés, les résultats indiquent que le traitement antimicrobien n'est pas plus efficace avec une croissance significative du Penicillium, l'espèce endémique des particules de riz micronisées (PM10) collectées par les filtres ; les deux populations de bactéries diminuent significativement avec ou sans traitement antimicrobien.Dans une deuxième étape, les performances de filtration des 2 filtres testés précédemment ont été étudiées dans des conditions réalistes avec un air extérieur semi-urbain sur une période de 7 mois. Le comportement des microorganismes extérieurs sur les filtres (croissance/mortalité) a été observé. Deux unités de filtration fonctionnaient à IMT Atlantique, l'une contenant le filtre PP/ZPT et l'autre le filtre PP. Chaque unité filtrait le même air semi-urbain. Les deux unités de filtration ont fonctionné en continu et plusieurs paramètres ont été surveillés tout au long de la période d'exploitation : température, humidité relative, chute de pression du filtre, efficacité de collecte des particules du filtre, concentration massique des particules à l'entrée et concentrations microbiennes ; en outre, la concentration microbienne sur les filtres a été quantifiée 3 fois (tous les 2 mois) à partir d'une méthodologie innovante basée sur des coupons de média. Les performances de filtration des deux filtres testés en termes de changement de perte de charges et d'efficacité de collecte de particules étaient différentes de celles obtenues à l'échelle du laboratoire. La méthodologie des coupons a permis d'étudier le comportement des micro-organismes tout au long de l'étude. L'effet antimicrobien de la pyrithione de zinc a été confirmé concernant l'inhibition des champignons sans influence du niveau d'encrassement du filtre (dépôt de masse de particules). / This study focused on the performance of a marketed antimicrobial polypropylene fibers filter containing zinc pyrithione (PP/ZPT) at the laboratory and compared to those of a similar filter (PP) with same classification F7 (EN779:2002). The filtration performance at laboratory scale of the 2 tested filters during clogging with PM10 particles was quantified in an experimental set-up with filter pressure drop measurement and particle counting up and downstream of the filters. The microbial growth onto new and used filters, both contaminated by aerosolization with a microbial consortium composed of two bacteria (Staphylococcus epidermidis Gram positive, Serratia marcescens Gram negative and fungal spores (Penicillium chrysogenum). The influence of three parameters on the microbial survival onto filters was examined: the air relative humidity, the presence or absence of Zinc pyrithione (ZPT) as antimicrobial substances and the presence of organic particles. Quantitative analyses by colony forming unit were used to determine the survival after 8 days of the bacteria–fungi consortium collected by the filter. The two filters revealed similar filtration performance in terms of change in pressure drop and particle collection efficiency during their clogging with PM10 particles, meaning that the antimicrobial treatment did not degrade the filtration performance of the filter. At low humidity value of conditioning (50% RH), with new or used filters, with or without antimicrobial treatment, the microbial population onto the filters decreases and possibly will not survive (Serratia). At high humidity value of conditioning (90% RH), the bacteria do not grow onto the new filters, and only the fungi was able to develop. The effect of the antimicrobial treatment with zinc pyrithione is confirmed for new filters in particular regarding the fungi Penicillium. For used filters, the results indicate that the antimicrobial treatment is no more efficient with a significant growth of the Penicillium, the endemic species of the micronized rice particles (PM10) collected by the filters; the two populations of bacteria significantly decrease with or without antimicrobial treatment. In a second step, the filtration performances of the 2 filters tested previously was investigated in realistic conditions with a semi-urban outdoor air over a 7 months period. The behavior of the outdoor microorganisms onto the filters (growth/mortality) was observed. Two filtration units were operating at IMT Atlantique location, one containing the PP/ZPT filter and the second the PP filter. Each unit filtered the same semi-urban air. Both filtration units operated continuously and several parameters were monitored throughout the operating period: temperature, relative humidity, filter pressure drop, filter particle collection efficiency, inlet particle mass concentration, and microbial concentrations; in addition, the microbial concentration onto the filters was quantified for 3 times (every 2 months) from an innovative methodology based on media coupons. The filtration performances of the two tested filters in terms of changes in pressure drop and particle collection efficiency were different than those obtained in the Laboratory scale. The methodology of coupons permitted to study the behavior of the microorganisms throughout the study. The antimicrobial effect of the zinc pyrithione was confirmed regarding the inhibition of the fungi cultivated on the DRBC agar with no influence of the level of clogging of the filter (mass of particles deposit).

O tratamento antimicrobiano pode destruir o equilíbrio da microbiota gastrintestinal, podendo induzir sintomas clínicos, principalmente a diarréia. A influência de Lactobacillus casei Shirota sobre a microbiota intestinal foi avaliada em um estudo prospectivo, randomizado, duplo-cego e controlado. Sessenta e três crianças hospitalizadas com idade de 2 a 14 anos, sob tratamento com antibióticos &#946;-lactâmicos, foram randomizadas para receber o leite fermentado por L. casei Shirota, 108-9 UFC/mL, ou o placebo, durante o tratamento antimicrobiano. As amostras de fezes foram colhidas antes da administração do leite fermentado, durante o tratamento antibiótico e uma semana após o término do tratamento com o antimicrobiano e a ingestão do leite fermentado. O número de L. casei Shirota aumentou significativamente (p<0,001) durante o período de ingestão do leite fermentado. Foi observado na microbiota do grupo que recebeu o placebo um aumento na contagem de Pseudomonas aeruginosa (p<0,05) e Clostridium sp (p<0,05), principalmente no último período da terapia antimicrobiana. A alteração da microbiota intestinal em decorrência do tratamento antibiótico foi constatada pela diminuição de acetato (p<0,05), butirato (p<0,05) e formato (p<0,05). Embora nenhuma criança deste estudo tenha apresentado diarréia, na avaliação geral, a microbiota daquelas que receberam o leite fermentado mostrou uma recuperação precoce da microbiota intestinal. Foi observado que a variação da contagem bacteriana realizada não foi significativa para as crianças do grupo que recebeu o leite fermentado, enquanto que no grupo placebo a contagem bacteriana ficou alterada, mostrando desequilíbrio da microbiota. Cerca de 50% das crianças ainda apresentaram L. casei Shirota nas fezes após uma semana da ingestão do leite fermentado. Este estudo mostrou que a ingestão do leite fermentado contendo L. casei Shirota promoveu um reequilíbrio mais rápido da microbiota intestinal quando comparada com a do grupo que ingeriu o placebo. / Antimicrobial treatment can destroy the balance of gastrointestinal microflora, which may induce clinical symptoms, mainly diarrhoea. The influence of Lactobacillus casei Shirota on the intestinal microflora was assessed in a prospective, randomised, double-blind controlled study. Sixty-three hospitalised children, with ages between 2 and 14 years, under treatment with &#946;-lactam antibiotics were randomised to receive milk fermented by L. casei Shirota, 108-9 CFU/mL, or placebo during the antimicrobial treatment. Stool samples were collected before the administration of fermented milk, during the antibiotic treatment, and one week after the end of treatment with the antimicrobial agent and the ingestion of fermented milk. The number of L. casei Shirota increased significantly (p<0.05) during the period in which fermented milk was ingested. An increase in the Pseudomonas aeruginosa (p<0.05) and Clostridium sp (p<0.05) count was observed in the microflora of the group that received placebo, mainly in the last period of antimicrobial therapy. The alteration of intestinal microflora as a result of antibiotic treatment was found by the reduction of acetate (p<0.05), butyrate (p<0.05) and formate (p<0.05). The variation in bacterial count proved not to be significant for the children under antimicrobial treatment who received fermented milk, while the placebo group showed imbalance of microflora with the result of the altered bacterial count. About 50% of the children still presented L. casei Shirota in their stools after interrupting the ingestion of fermented milk for one week. This study showed that ingestion of fermented milk containing L. casei Shirota promoted a much faster re-balance of the intestinal microflora when compared to the group that ingested a placebo.

O tratamento antimicrobiano pode destruir o equilíbrio da microbiota gastrintestinal, podendo induzir sintomas clínicos, principalmente a diarréia. A influência de Lactobacillus casei Shirota sobre a microbiota intestinal foi avaliada em um estudo prospectivo, randomizado, duplo-cego e controlado. Sessenta e três crianças hospitalizadas com idade de 2 a 14 anos, sob tratamento com antibióticos &#946;-lactâmicos, foram randomizadas para receber o leite fermentado por L. casei Shirota, 108-9 UFC/mL, ou o placebo, durante o tratamento antimicrobiano. As amostras de fezes foram colhidas antes da administração do leite fermentado, durante o tratamento antibiótico e uma semana após o término do tratamento com o antimicrobiano e a ingestão do leite fermentado. O número de L. casei Shirota aumentou significativamente (p<0,001) durante o período de ingestão do leite fermentado. Foi observado na microbiota do grupo que recebeu o placebo um aumento na contagem de Pseudomonas aeruginosa (p<0,05) e Clostridium sp (p<0,05), principalmente no último período da terapia antimicrobiana. A alteração da microbiota intestinal em decorrência do tratamento antibiótico foi constatada pela diminuição de acetato (p<0,05), butirato (p<0,05) e formato (p<0,05). Embora nenhuma criança deste estudo tenha apresentado diarréia, na avaliação geral, a microbiota daquelas que receberam o leite fermentado mostrou uma recuperação precoce da microbiota intestinal. Foi observado que a variação da contagem bacteriana realizada não foi significativa para as crianças do grupo que recebeu o leite fermentado, enquanto que no grupo placebo a contagem bacteriana ficou alterada, mostrando desequilíbrio da microbiota. Cerca de 50% das crianças ainda apresentaram L. casei Shirota nas fezes após uma semana da ingestão do leite fermentado. Este estudo mostrou que a ingestão do leite fermentado contendo L. casei Shirota promoveu um reequilíbrio mais rápido da microbiota intestinal quando comparada com a do grupo que ingeriu o placebo. / Antimicrobial treatment can destroy the balance of gastrointestinal microflora, which may induce clinical symptoms, mainly diarrhoea. The influence of Lactobacillus casei Shirota on the intestinal microflora was assessed in a prospective, randomised, double-blind controlled study. Sixty-three hospitalised children, with ages between 2 and 14 years, under treatment with &#946;-lactam antibiotics were randomised to receive milk fermented by L. casei Shirota, 108-9 CFU/mL, or placebo during the antimicrobial treatment. Stool samples were collected before the administration of fermented milk, during the antibiotic treatment, and one week after the end of treatment with the antimicrobial agent and the ingestion of fermented milk. The number of L. casei Shirota increased significantly (p<0.05) during the period in which fermented milk was ingested. An increase in the Pseudomonas aeruginosa (p<0.05) and Clostridium sp (p<0.05) count was observed in the microflora of the group that received placebo, mainly in the last period of antimicrobial therapy. The alteration of intestinal microflora as a result of antibiotic treatment was found by the reduction of acetate (p<0.05), butyrate (p<0.05) and formate (p<0.05). The variation in bacterial count proved not to be significant for the children under antimicrobial treatment who received fermented milk, while the placebo group showed imbalance of microflora with the result of the altered bacterial count. About 50% of the children still presented L. casei Shirota in their stools after interrupting the ingestion of fermented milk for one week. This study showed that ingestion of fermented milk containing L. casei Shirota promoted a much faster re-balance of the intestinal microflora when compared to the group that ingested a placebo.

Le traitement sélectif (TS) des vaches laitières au tarissement (où seuls les quartiers ou les vaches infectées sont traités avec des antimicrobiens) constitue une alternative potentielle au traitement universel (TU, où tous les quartiers de toutes les vaches reçoivent des antimicrobiens, quel que soit leur statut infectieux), pour une utilisation plus judicieuse des antimicrobiens. L'objectif de cette thèse était d’apporter plus de lumière sur les décisions de traitement antimicrobien ciblant les quartiers ou vaches infecté(e)s au tarissement. Différents devis et méthodologies ont été utilisés pour répondre à cet objectif. Un essai contrôlé randomisé a été conçu et 569 vaches (2,251 quartiers) provenant de 9 troupeaux laitiers du Québec avec un comptage de cellules somatiques (CCS) du réservoir <250 000 cellules/mL ont été systématiquement enrôlées et réparties au hasard dans 4 groupes : 1) traitement antimicrobien seul pour tous les quartiers ; 2) traitement antimicrobien combiné avec un scellant interne à trayon pour tous les quartiers ; 3) traitement antimicrobien sélectif seul basé sur les résultats de la culture bactériologique du lait sur Petrifilm® ; et 4) traitement antimicrobien sélectif combiné avec un scellant interne à trayon basé sur les résultats de la culture du lait sur Petrifilm®. Dans les groupes de TS, les quartiers non infectés n'ont reçu qu'un scellant interne à trayon. Aucune différence significative n'a été détectée entre le TS par quartier et le TU des vaches laitières au tarissement, en termes d'élimination des infections intramammaires (IIM) et de prévention de nouvelles IIM pendant la période de tarissement, de risque d'un premier cas de mammite clinique (MC), de production laitière moyenne quotidienne et de CCS au cours des 120 premiers jours de la lactation suivante. Un TS reposant sur les résultats d'une culture de lait de quartier sur Petrifilm® au tarissement a permis de réduire l'utilisation d'antimicrobiens de 52% (IC à 95%: 39 – 64) par rapport à un TU. En plus de cet essai contrôlé randomisé, la culture du lait par quartier à l'aide de Petrifilm® a été comparée à l'historique du CCS par une estimation bayésienne de leur précision pour identifier les quartiers ou les vaches qui devraient être traités avec des antimicrobiens dans des protocoles de TS au tarissement. Compte tenu de la disponibilité des données de CCS, de la facilité d'utilisation du dernier test de CCS pré-tarissement et de la valeur prédictive négative élevée qui pourrait être obtenue, les producteurs pourraient envisager d'utiliser uniquement le dernier test de CCS pré-tarissement comme outil potentiel pour identifier les vaches qui devraient être traitées avec des antimicrobiens au tarissement. Le dernier test de CCS pré-tarissement peut être utilisé seul ou en combinaison avec la culture de lait par quartier sur Petrifilm® sur les vaches avec un CCS élevé pour identifier encore plus spécifiquement les quartiers qui doivent être traités. L'ajout d'une culture de lait par quartier à la ferme sur Petrifilm® pour les vaches identifiées comme infectées à l'aide des données du CCS améliorerait la précision du test (principalement la valeur prédictive positive) et réduirait davantage l'utilisation d'antimicrobiens. Également, une revue systématique et une série de méta-analyses ont été menées pour étudier l'efficacité du TS par rapport au TU, afin de guider les décideurs et les utilisateurs qui s'engagent dans une utilisation plus efficace et judicieuse des antimicrobiens au moment du tarissement. Treize articles représentant 12 essais contrôlés, randomisés ou non, étaient disponibles pour les analyses. Le TS a permis de réduire de 66% (IC à 95%: 49 – 80) l'utilisation d'antimicrobiens au moment du tarissement. Les résultats appuient fortement l'idée que le TS réduirait l'utilisation d'antimicrobiens au moment du tarissement, sans effet négatif sur la santé du pis ou la production laitière au cours des premiers mois de la lactation subséquente, si, et seulement si, les scellant internes à trayons sont utilisés pour les quartiers non traités avec des antimicrobiens. Enfin, le suivi de l'utilisation d'un scellant interne à trayon a été effectué pour déterminer la proportion de quartiers qui ont conservé le bouchon de scellant jusqu’à la première traite après le vêlage et la persistance de résidus de scellant dans le lait après le vêlage. Un bouchon de scellant était présent jusqu'à la première traite pour 83% des quartiers, et nous pourrions émettre l'hypothèse que la perte du bouchon s'est produite près du vêlage secondaire à la tétée ou pour une autre raison (ex., la pression hydrostatique du lait), étant donné que les associations observées entre la présence ou non d'un bouchon de scellant observable et les chances de nouvelles IIM étaient relativement faibles. Les résidus de scellant pouvaient être observés dans le lait jusqu'à 12 jours après le vêlage, quoique 75% des quartiers n’excrétaient plus de scellant au bout de 5 jours en lait. / Selective dry cow therapy (SDCT, in which only infected quarters or cows are treated with antimicrobials) represents an alternative to blanket dry cow therapy (BDCT, in which all quarters of all cows at dry off are treated with antimicrobials, regardless of their infection status), for a more judicious use of antimicrobials. The objective of this thesis was to shed more light on targeted antimicrobial treatment decisions of infected quarters or cows at dry-off. Different study designs and methodologies were used to meet this objective. A randomized controlled trial was designed and a total of 569 cows (2,251 quarters) from 9 dairy herds in Québec with bulk tank somatic cell count (SCC) <250,000 cells/mL were systematically enrolled and randomly allocated to 4 groups: 1) antimicrobial treatment alone of all quarters; 2) antimicrobial treatment combined with an internal teat sealant (ITS) of all quarters; 3) selective antimicrobial treatment alone based on milk bacteriological culture results on Petrifilm®; and 4) selective antimicrobial treatment combined with an ITS based on milk culture results on Petrifilm®. In the selective antimicrobial treatment groups, uninfected quarters received only an ITS. No significant differences were detected between quarter-based selective and blanket dry cow therapies, in terms of elimination of intramammary infections (IMI) and prevention of new IMI during the dry period, risk of a first case of clinical mastitis (CM), daily average milk yield and somatic cell count in the first 120 days of the subsequent lactation. A selective antimicrobial treatment relying on results of quarter milk culture using Petrifilm® at dry off enabled a reduction in antimicrobial use of 52% (95% CI: 39 – 64) as compared to blanket dry cow treatment. In addition to this randomized controlled trial, quarter milk culture using Petrifilm® was compared with SCC history through a Bayesian estimation of diagnostic accuracy to identify quarters or cows that should be treated with antimicrobials in selective treatment protocols at dry off. Considering the availability of SCC data, the easiness of using just the last Dairy Herd Improvement (DHI) test before dry off, and the high negative predictive value that could be achieved, producers may consider using just the last DHI test before dry off results as a potential tool to identify cows that should be treated with antimicrobials at dry off. The last SCC test before 7 dry off may be used alone or in combination with quarter-level on-farm Petrifilm® milk culture on high SCC cows to more specifically identify quarters that need to be treated. Adding quarter-level milk culture using Petrifilm® to cows identified as unhealthy using cow-level SCC data could improve the test accuracy (mainly the positive predictive value) and further reduced the use of antimicrobials. Also, a systematic review and a series of meta-analyses were conducted to investigate the efficacy of SDCT compared with BDCT, to guide decision-makers and users to engage in a more effective and judicious use of antimicrobials at dry-off. Thirteen articles representing 12 controlled trials, randomized or not, were available for analyses. SDCT reduced the use of antimicrobials at dry off by 66% (95% CI: 49 – 80). Evidences strongly support that SDCT would reduce the use of antimicrobials at dry off, without any detrimental effect on udder health or milk production during the first months of the subsequent lactation, if, and only if, ITS are used for healthy quarters untreated with antimicrobials. Finally, a follow up on the use of ITS was performed to determine the proportion of quarters that had retained the sealant plug until the first milking after calving and the persistence of ITS residues in milk after calving. A sealant plug was present at first milking after calving for 83% of the quarters, and we could hypothesize that the loss of the plug occurred closely around calving due to suckling or for another reason (e.g., milk hydrostatic pressure), since the observed associations between the presence or not of an observable sealant plug and the odds of new IMI were relatively small. The sealant residues could be observed in milk up to 12 days in milk, although 75% of the quarters had expelled the last ITS residues by 5 days in milk.