Search

Bacteremia and fungemia account for a substantial proportion of all cases of severe sepsis. Antibiotic resistance is a contributing factor in many hospital-acquired infection deaths. Traditional phenotypic methods for the identification of bacteria and yeasts from positive blood cultures and determining antimicrobial susceptibility require 48-72 h, delaying optimal therapy and negatively impacting patient outcomes. Molecular methods, including nonamplified DNA probe panels and peptide nucleic acid probes, and nucleic acid amplification methods such as PCR, proteomic methods (matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry) and direct biochemical tests provide more rapid identification of bacteria and fungi, and in some cases antimicrobial resistance markers, from positive blood cultures, as well as directly from whole blood. These methods vary in the breadth of organisms that they detect, and equally important, their ease of use. This article examines the principles, performance and practicality of the various rapid, nonculture techniques for the detection of bacteremia and fungemia.

A retrospective case-control study was performed to analyze hospital-acquired candidemia in HIV-positive patients. To understand the impact of Highly Active Antiretroviral Therapy (HAART) on the incidence of nosocomial candidemia, two time periods were compared: A (1992-1996) and B (1997-2001). 32 out of 38 (84%) cases of candidemia were hospital-related. A significant reduction in the incidence of all cases of hospital-acquired candidemia has been observed in the post-HAART in respect to pre-HAART period (0.09 episodes vs. 1.1 per 100/py). Multivariate analysis showed that the presence of central venous catheter was the only variable independently associated with the development of nosocomial candidemia. The overall mortality rate was 59%. Univariate analysis indicated three prognostic indicators: presence of concomitant opportunistic infections, isolation of non-albicans Candida species; neutropenia. Multivariate analysis of prognostic indicators showed that isolation of non-albicans Candida species is the only independent variable. Despite the use of HAART, this disease still represents a severe complication of advanced stage of AIDS.

Unlabelled: We retrospectively analyzed the incidence of thrombotic and infectious complications in relation with the use of central venous catheters (CVCs), in a series of patients with hematological malignancies and low platelet and leucocyte counts., Patients and Methods: 126 patients with hematological malignancies were analyzed. A total of 207 CVCs were implanted: 137 centrally (CICCs) and 70 peripherally (PICCs). The median duration of the CVCs was 19 days for a total of 4051 catheter-days. Antithrombotic prophylaxis was unfractionated heparin (UFH), 2,500 IU daily by 24 h continuous infusion in 169 CVCs, low molecular weight heparin (LMWH), 3,800 IU daily by single bolus intravenous injection (i.v.) in 21 and warfarin in one. No prophylaxis was given in 16 CVCs. Thrombotic complications developed in 15.5% of the CVCs (7.9 events/1000 catheter days), and the frequency of infectious complications was 10.6% (5.2 events/1000 catheter days). On multivariate analysis thromboses were more frequent and earlier with PICCs than CICCs (p = 0.0001), and in patients on UFH (16.6%) than in LMWH prophylaxis (4.7%), but the last difference was not statistically significant. In conclusions the incidence of thrombotic complications in our series was comparable to that observed in non-thrombocytopenic patients and was significantly higher in those carrying PICC than CICC (p = 0.0001). There were fewer thrombotic events in the patients receiving i.v. LMWH prophylaxis than in those receiving i.v. UFH. The use of anticoagulants was safe and not associated with hemorrhages.

Long-term central venous catheters (CVC) are necessary in the care of cancer patients. However, catheter-related bloodstream infection (CRBSI) is commonly associated with serious complications resulting in considerable morbidity and mortality. The diagnosis of CRBSI frequently requires catheter removal to confirm the diagnosis by either quantitative or semiquantitative catheter culture method. Differential time to positivity, whereby a nonquantitative blood culture drawn from the CVC that becomes positive at least 2 hr earlier than the peripheral blood culture, is a new method for the diagnosis of CRBSI without removing the catheter. Prevention of CRBSI may be accomplished with the use of strict infection control measures, antimicrobial-impregnated catheters; and antibiotic-lock technique, as well as other methods. Once infection develops, management of long-term CRBSI is dictated by the type of organism, the severity of the infection, and availability of other venous access sites. If the infection is caused by Staphylococcus aureus, gram-negative bacilli, or Candida, the catheter should be removed and systemic antimicrobial therapy given for 10-14 days or longer in cases of complicated or deep-seated infection. In some cases, where there is no other venous access site, the catheter can remain in place, but a combination of systemic antimicrobials and antibiotic-lock therapy should be used.

Abdominal infections are an important cause of morbidity and mortality in neutropenic patients. We present a retrospective series of 16 patients, mostly with acute leukemia, who developed severe abdominal infections during chemotherapy-induced neutropenia between 1991 and 1997. The frequency among patients with acute leukemia was 2.35% (13 of 553). Thirteen patients presented with enterocolitis and 3 patients presented with cholecystitis. Eight patients died. Bacteremia was present in 6 patients, 4 patients suffered from proven or strongly suspected fungal infections, and 1 patient suffered from cytomegalovirus infection. Early surgical management was required in a patient with intestinal obstruction, whereas other patients could be managed conservatively. Two patients with acute cholecystitis were treated with antibiotics until the end of neutropenia and then were resected. Severe abdominal injections in neutropenic patients, which are often fatal, were caused by nonbacterial microorganisms in one-fourth of the cases and could be managed conservatively in most instances.

Candida spp. are the fourth leading cause of bloodstream infection. While the literature on neonatal candidemia is abundant, its prevalence in pediatric surgery cases is hardly mentioned. This study was carried out over a 5-year period to evaluate the prevalence of candidemia in pediatric surgery intensive care unit patients (ICU), and to examine both the neonatal and hospital risk factors for developing candidemia in comparison to control groups of patients with either no infection or with bacteremia, type and outcome of therapy. A total of 1,359 pediatric surgery patients admitted to the ICU and high dependency unit (HDU) were included in the study. Using relevant specimens from them, a microbiological survey was carried out on admission and weekly thereafter. Twenty-five patients developed candidemia during the study period. Twenty-one of them were admitted to ICU. Nine were low birth weight and immature neonates. All 25 patients had underlying disease, most involving the gastroentestinal tract and requiring surgical intervention. All patients had been given broad-spectrum beta-lactam antibiotics with or without aminoglycosides and an anti-anaerobic drug prior to candidemia. The data show that patients who were not infected had very few risk factors that could predispose to candidemia. The bacteremic group of patients had more risk factors: mainly ICU stay, prior antibiotic therapy or GI surgery. The candidemia patients outnumbered these two groups in both neonatal and hospital risk factors. Twenty-three candidemia patients had received amphotericin B and 2 had fluconazole. Seventeen of them improved and the rest expired during therapy. Fourteen of the Candida isolated were C. albicans while the rest belonged to other Candida spp. dominated by C. parapsilosis. In conclusion, candidemia was infrequent in pediatric surgery patients. ICU stay, GI surgery and prior broad-spectrum antibiotic therapy were important risk factors.

The risk factors, therapy and outcome of ten cases of fungemia due to Candida krusei, appearing during the last 10 years in a single national cancer institution, are analyzed. Univariate analyses did not find any specific risk factors in comparison to 51 Candida albicans fungemias appearing at the same institution and with a similar antibiotic policy. Association with prior fluconazole prophylaxis was not confirmed because only one case appeared in a patient previously treated with fluconazole. However, attributable and crude mortality due to C. krusei fungemias was higher than for C. albicans fungemia. The authors review 172 C. krusei fungemias published within the last 10 years to compare with the incidence, therapy and outcome of C. krusei fungemia from our cancer institute.

One hundred and eighteen (118) episodes of bacteremia and fungemia in children with cancer were compared to 401 episodes of bacteremia and fungemia in adults with cancer to assess differences in etiology, risk factors and outcome. A retrospective univariate analysis was performed of all episodes of bacteremia in national pediatric and adult cancer institutions appearing in 1990-1996. A total of 519 episodes of bacteremia were assessed and compared. Both cancer centers differed in prophylactic antibiotic policies. About 50% of adults but less than 5% of children received quinolone prophylaxis during neutropenia, even though the empiric antibiotic therapeutic strategy was similar. There were differences in etiology between the groups: staphylococci and Stenotrophomonas maltophilia were more frequently observed in children (P<0.01), Pseudomonas aeruginosa and Acinetobacter spp. in adults (P<0.05). Gram-positive bacteremia was surprisingly more commonly observed in adults (65.7% vs 33.3%, P<0.01). Mixed polymicrobial bacteremia occurred more commonly in adults (31.8% vs 7.6%, P<0.001) than in children. Analysis of risk factors did not observe differences in risk factors except for underlying disease (acute leukemia was more frequently observed in children -48.3% vs adults 33.7%, P<0.05 and prophylaxis: (prior prophylaxis with quinolones was more common in adults (47.5%) than in children (2.5%) P<0.0001). Overall and attributable mortality in pediatric bacteremia was significantly lower than in adults (P<0.03).

Objective: To elucidate the incidence, determinants, and consequences of severe sepsis after cardiac surgery., Design: Prospective study., Setting: Cardiac surgical unit, Greece., Subjects: 2615 adult patients having cardiac operations., Main Outcome Measures: Microbiological evidence of sepsis, mortality, and duration of stay in the intensive care unit (ICU) and hospital., Results: Severe sepsis developed in 41/2615 patients (2%), all during their stay in the ICU: there were 30 men and 11 women, mean (SD) age 65 (10) years. It was most common after combined coronary artery bypass grafting and valve-related operations (7/95, 7%), followed by miscellaneous cardiac operations (7/147, 5%), valve replacement (8/359, 2%), and coronary artery bypass grafting (19/2014, 1%). When the 41 patients who developed severe sepsis were compared with those who did not (n = 2574) by univariate analysis, there were significant differences in age (p = 0.004); type of operation (p < 0.0001); duration of operation (p < 0.001); bleeding that necessitating either reoperation or significantly more blood transfused (p < 0.0001); and the incidence of low cardiac output syndrome (p = 0.0001). Of the 41 patients with severe sepsis, 19 (46%) had serious operative complications, 40 (98%) had severe complications in the ICU, and 16 (39%) required reintubation for hypoxaemia. Among the 41 there were 54 bacteraemic episodes of which 37 (69%) were caused by gram positive cocci, 6 (11%) by gram negative bacteria, and 11 (20%) by Candida albicans. Staphylococcus epidermidis was the most common pathogen isolated (n = 26, 48%). Sepsis associated with bacterial infection usually developed during the first two weeks, and that caused by fungal infection was most common after the twentieth postoperative day. Patients with severe sepsis required longer mechanical ventilation (31 (21) days compared with 0.9 (0.1) days); longer stay in the ICU (40 (25) days) compared with 2 (1) days); longer stay in hospital (48 (27) days compared with 10 (2) days); and significantly more of them died (13 (32%) compared with 41 (2%), p < 0.0001 in each case)., Conclusions: We concluded that severe sepsis mainly developed in cardiac surgery patients with serious operative and postoperative complications and was associated with a longer stay in both ICU and hospital, and a higher mortality.

Invasive fungal infections occur frequently in neutropenic patients although only in recent years has the role of emerging fungi been clearly established. We describe two cases of fungemia caused by Trichosporon beigelii and Rhodotorula glutinis respectively in two neutropenic patients with hematological malignancies who were treated with amphotericin B. The first patient, with refractory multiple myeloma, died following massive pneumonia despite therapy with amphotericin B and granulocyte-colony stimulating factor (G-CSF); the second patient, with relapsed acute lymphatic leukemia and persistent fever without any other clinical evidence, finally recovered. Amphotericin B continues to be considered the "gold standard" in the treatment of invasive mycoses although other approaches need to be tested for refractory infections.