Financial impact of healthcare-associated infections on intensive care units estimated for fifty Brazilian university hospitals affiliated to the unified health system.

BACKGROUND: Studies show that healthcare-associated infections (HAIs) represent a crucial issue in healthcare and can lead to substantial economic impacts in intensive care units (ICUs). AIM: To estimate direct costs associated with the most significant HAIs in 50 teaching hospitals in Brazil, affiliated to the unified health system (Sistema Único de Saúde: SUS). METHODS: A Monte Carlo simulation model was designed to estimate the direct costs of HAIs; first, epidemiologic and economic parameters were established for each HAI based on a cohort of 949 critical patients (800 without HAI and 149 with); second, simulation based on three Brazilian prevalence scenarios of HAIs in ICU patients (29.1%, 51.2%, and 61.6%) was used; and third, the annual direct costs of HAIs in 50 university hospitals were simulated. FINDINGS: Patients with HAIs had 16 additional days in the ICU, along with an extra direct cost of US$13.892, compared to those without HAIs. In one hypothetical scenario without HAI, the direct annual cost of hospital care for 26,649 inpatients in adult ICUs of 50 hospitals was US$112,924,421. There was an increase of approximately US$56 million in a scenario with 29.1%, and an increase of US$147 million in a scenario with 61.6%. The impact on the direct cost became significant starting at a 10% prevalence of HAIs, where US$2,824,817 is added for each 1% increase in prevalence. CONCLUSION: This analysis provides robust and updated estimates showing that HAI places a significant financial burden on the Brazilian healthcare system and contributes to a longer stay for inpatients.

Costs of healthcare-associated infections to the Brazilian public Unified Health System in a tertiary-care teaching hospital: a matched case-control study.

BACKGROUND: Little is known about the economic burden of healthcare-associated infections (HAIs) in Brazil. AIM: To analyse the costs of hospitalization by reimbursement from the Brazilian government, via the Brazilian Unified Health System (SUS) affiliation, and direct costs in the adult Intensive Care Unit (ICU). METHODS: The matched-pairs case-control study (83 patients with HAIs and 83 without HAIs) was performed at a referral tertiary-care teaching hospital in Brazil in January 2018. In order to calculate the HAI costs from the perspective of the payer, the total cost for each hospitalization was obtained through the Hospital's Billing Sector. Direct costs were calculated annually for 949 critical patients during 2018. FINDINGS: The reimbursement cost per hospitalization of patients with HAIs was 75% (US$2721) higher than patients without HAIs (US$1553). When a patient has an HAI, in addition to a longer length of stay (15 days), there was an extra increase (US$996) in the reimbursement cost per hospitalization. An HAI in the ICU was associated with a total direct cost eight times higher compared with patients who did not develop infections in this unit, US$11,776 × US$1329, respectively. The direct cost of hospitalization in the ICU without HAI was 56.5% less than the reimbursement (US$1329 × US$3052, respectively), whereas for the patient with an HAI, the direct cost was 111.5% above the reimbursement (US$11,776 × US$5569, respectively). CONCLUSION: HAIs contribute to a longer stay and an eight-fold increase in direct costs. It is necessary to reinforce programmes that prevent HAIs in Brazilian hospitals.

Multi-hospital point prevalence study of healthcare-associated infections in 28 adult intensive care units in Brazil.

BACKGROUND: Healthcare-associated infection (HAI) represents a major problem for patient safety worldwide. AIM: To provide an up-to-date picture of the extent, aetiology, risk factors and patterns of infections in intensive care units (ICUs) in 28 Brazilian hospitals of different sizes. METHODS: A one-day point prevalence survey in 2016 enrolled the ICUs of hospitals from the 12 meso-regions in Minas Gerais state, south-east Brazil. Hospitals were classified as university or non-university hospitals. All patients with >48 h of admission to the study ICUs at the time of the survey were included. FINDINGS: In total, 303 patients were studied; of these, 155 (51.2%) were infected and 123 (79.4%) had at least one ICU-acquired infection. The most common ICU-acquired infections were pneumonia (53.0%) and bloodstream infection (27.6%). One hundred and nineteen bacterial isolates were cultured; the most common were Acinetobacter baumannii (27.1%), Pseudomonas aeruginosa (27.1%) and Staphylococcus aureus (39.0%). According to type of infection, the most common pathogens were P. aeruginosa (30.4%) in pneumonia, coagulase-negative staphylococci (23.4%) and Enterobacteriaceae (23.4%) in bloodstream infections, and Enterobacteriaceae (47.6%) in urinary tract infections. CONCLUSION: This study found that the overall prevalence of ICU-acquired infections in surveyed Brazilian hospitals was higher than that reported in most European countries and the USA. A greater proportion of infections were caused by non-fermenting Gram-negative bacteria. These observations, along with a high rate of antimicrobial use, illustrate the urgent need for HAIs to be a priority in the public health agenda of Brazil.

Outbreaks of colistin-resistant and colistin-susceptible KPC-producing Klebsiella pneumoniae in a Brazilian intensive care unit.

BACKGROUND: Carbapenem-resistant Enterobacteriaceae (CRE), especially those that produce Klebsiella pneumoniae carbapenemase (KPC) and are associated with colistin resistance, pose a severe health threat due to the limited treatment options. AIM: To describe two outbreaks of KPC-producing K. pneumoniae in an adult intensive care unit (AICU) in Brazil. In May 2015, 14 patients had colistin-susceptible KPC-producing strains (ColS-KPC), and in July 2015, nine patients had colistin-resistant KPC-producing strains (ColR-KPC). METHODS: Between September 2014 and August 2015, we performed surveillance at a university hospital and all CRE were tested for blaKPC genes. Clonality was investigated by pulsed-field gel electrophoresis. Resistance to colistin was confirmed by broth microdilution method. Consumption of carbapenems and colistin was expressed as defined daily doses. FINDINGS: In all, 111 patients with CRE were identified during the surveillance period; K. pneumoniae was the major isolate (77.13%). The two outbreaks were identified when infection rates (KPC per 1000 patient-days) exceeded the background level. Rates of carbapenem and colistin consumption were high. Control measures (bedside alcohol gel, contact precautions, regular rectal swabs) did not curtail the outbreaks. Mortality rates were 42.9% and 44.4% for ColS-KPC- and ColR-KPC-infected patients, respectively. After the death of four infected patients with ColR-KPC, the unit was closed to new admissions. CONCLUSION: Our experience demonstrates the serious risks presented by KPC, and especially ColR-KPC, in Brazilian AICUs. Selective pressure from excessive antibiotic use and transmission on healthcare workers' hands were likely the major factors in transmission.

Bacterial colonization of pressure ulcers: assessment of risk for bloodstream infection and impact on patient outcomes.

BACKGROUND: Pressure ulcers (PUs) represent a major problem for hospitalized patients, health professionals and society. AIM: To evaluate the impact of in-hospital PUs as a reservoir of multidrug-resistant organisms (MDROs), risk factor for bacteraemia and predictor of poor prognosis. METHODS: A prospective cohort study of patients with stage II or greater PUs hospitalized in a tertiary acute care university teaching hospital for more than 48 h was performed to evaluate colonization/infection by potential and/or multi-resistant hospital pathogens from April to December 2005 and from August 2009 to April 2010. FINDINGS: A total of 145 patients with stage II or greater PUs were included. Of these, 76.5% (111/145) had PUs colonized and/or infected with either S. aureus (20.7%), Gram-negative bacilli (32.5%), or both (46.8%) and most were MDROs (64.8%). Bacteraemia was detected in 50.5% (56/111) of the patients. The ulcers were considered to be the probable source of bacteraemia in 53.6% (30/56) of the episodes. Prior administration of antibiotics (P = 0.04) and infected wound (P < 0.001) were the variables independently associated with bloodstream infection as well as associated with a higher 30-day mortality rate; risk factors for the latter included hospitalization in ICU (P = 0.03) and mechanical ventilation use (P = 0.05). CONCLUSIONS: Our results suggest that besides being a major reservoir of MDROs, patients with PUs constitute a high-risk population for bacteraemia with a poor outcome. Broad-spectrum antibiotics and infected wound were independent factors predisposing patients to both bacteraemia and death.

Vancomycin use in a Brazilian University Hospital: comparison with Hospital infection control practices advisory committee guidelines

Emergence of vancomycin-resistant bacteria is of concer. In an effort to reduce this danger, guidelines to ensure proper prescribing of vancomycin have been proposed to the Hospital Infections Control Practices Advisory Committee (HICPAC) of the Centers for Disease Control. To evaluate use of vancomycin at Uberlândia University Hospital in Brazil, each patient who received the drug during a 10 month period had nasal and recal cultures done within 48 hours of initiation of therapy, then at weekly intervals until discharge. Their hospital records were reviewed to obtain demographic and clinical data, and each was scored as to whether or not HICPAC guidelines were followed. Thirty-one patients were enrolled in the study; 15 of whom had been approved for vancomycin by the hospital infectious diseases (ID) specialist, and 16 who had the drug given without approval. During the study, 4 strains of VRE (Vancomycin Resitant Enterococci) and 8 strains of MRSA (Methicillin Resistant S.aureus) emerged. The use of vancomycin did not follow HICPAC guidelines in 21/31 patients (68 percent), in that the drug was prescribed empirically without prior documenation of need. This occurred 13 of 16 (81 percent) times when no approval by an ID specialist was provided, and 8 of 15 times (53 percent) after approval by a ID specialist. In 6 of 8 patients (75 percent), empirical use of vancomycin was approved by an ID specialist in patients with severe illnesses, each of whom subsequently died. his could be considered appropriate use, although outside the guidelines. Only 3 of 13 patients (23 percent) were severely ill when vancomycin was used without ID approval. Most of the vancomycin use which did not follow HICPAC guidelines occurred on the medicine service in patients moderately ill with pneumonia or bacteremia, and associated with invasive procedures. We conclude that there is a special need to improve education regarding the appropriate use of vancomycin and to increase review of its use by an ID specialist, particularly on medicine services of our hospital.