Becky McCall

April 13, 2012

April 13, 2012 (London, United Kingdom) — Elderly patients treated with central catheter and/or mechanical ventilation devices in intensive care units (ICUs), admitted from the emergency department or as an urgent case, are at very high risk for hospital-acquired infection (HAI), according to the results of research presented here at the 22nd European Congress of Clinical Microbiology and Infectious Diseases.

In-hospital mortality in ICU patients with HAI was 4 times higher than in those without HAI, according to this analysis of an American hospital database.

Florence Joly, PharmD, from global evidence and value development at Sanofi in Paris, France, led the study and presented the findings.

"A device, namely a central catheter [odds ratio (OR), 3.4] or mechanical ventilation [OR, 2.8], trebled the risk of having a HAI, including bloodstream infection, nosocomial pneumonia, and surgical-site infection," Dr. Joly told Medscape Medical News. Higher risks were associated with coming from the emergency department and being older than 75 years of age, he said.

She added that mortality was 4 times higher in patients with a HAI than in those without (18.5% vs 4.5%). In patients with HAI, length of stay in the ICU doubled, from a mean of 8.1 days to 15.8 days.

Bloodstream infection was the greatest driver of mortality, at 24.7%, followed by hospital-acquired pneumonia, at 16.7%, and surgical-site infection, at 10.9%.

The analysis was a retrospective cohort study, using the Premier Perspective hospital database, of adults older than 18 years with an ICU stay of at least 48 hours in 2007. This database covers 20% of hospital discharges in the United States. The researchers also analyzed data from 2008 and 2009 and found similar results, although the poster presented covered only the 2007 results.

Other data are available on HAI in the United States, but according to Dr. Joly, they do not always permit comparison with uninfected patients. She explained that the strength of these results is in the large number of ICU patients — more than 460,000.

"Real-life data analyzed from this kind of database can provide a quick snapshot of a national situation, enable trends to be followed over time, and allow a controlled analysis, such as risk analysis and cost effectiveness, to be done," she said.

"We wanted to get a better idea of the burden of disease and infection — especially in the ICU, where these infections represent a real challenge — and the economic impact incurred. We wanted to know the potential impact of a preventative intervention on these infections."

Over a 1-year period, Dr. Joly and her colleagues analyzed data from 511,815 ICU stays. Most patients had 1 ICU stay, but 7.0% had 2, and 1.5% had 3 or more. Hospital-acquired pneumonia was associated with 16.9% of ICU stays, ventilator-acquired pneumonia with 3.7%, bloodstream infection with 14.5%, and surgical-site infection with 1.5%. Overall, 26.7% of ICU stays were associated with at least 1 HAI, which is 3 times higher than the general infection rate in the hospital.

"We found that hospital-acquired infection triggered a significant extra 7.5 days in intensive care," reported Dr. Joly. "The longest stay was observed for surgical-site infections, with a mean of 23.4 days."

A cost analysis was also conducted, based on direct variable costs and fixed overhead costs.

These infections cost an extra $16,000 per ICU stay — from $21,500 for a patient without a HAI to $37,500 for a patient with a HAI, reported Dr. Joly.

Christian Brun-Buisson, MD, PhD, professor of medicine and intensive care at Université Paris-Est Créteil, and director of the medical intensive care unit at Hôpital Henri Mondor in Créteil, France, moderated the session and commented on the study in an interview with Medscape Medical News.

"These data...confirm the current burden of HAI on ICU patients and the role of known risk factors, mainly invasive devices, to which this population is highly exposed, although in this study, the exposure rate to mechanical ventilation is unusually low," he said.

"Limiting exposure to devices is one way to reduce the risk of HAI..., as is the development of devices less prone to colonization," Dr. Brun-Buisson noted.

However, he advised caution when interpreting this study's outcomes. "The consequences of HAI on morbidity and mortality have often been overestimated, which may lead to unrealistic expectations of the impact from preventive strategies," said Dr. Brun-Buisson.

Adding insight on the results, Kenneth B. Christopher, MD, from Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, said that the 25.8% of ICU patients in this study who developed a HAI were more likely to have a predisposition to infection related to supportive measures, such as endotracheal tube placement and intravascular lines.

"The studies emphasize the importance of infection-control practices in critically ill patients that are currently implemented at many hospitals," he told Medscape Medical News.

He pointed out that a Centers for Disease Control and Prevention report published in 2011 on ICUs in the United States showed that from 2001 to 2009, there was a 58% reduction in central-line-associated bloodstream infections, and that the initiation of a bundle protocol was shown to be an effective method to reduce ventilator-acquired pneumonia.

"The author's demonstration of a high prevalence of HAI and the ability to reduce HAI with the adoption of best practice underscore the importance of changing behavior in the clinical setting," he concluded.

Adding his interpretation of the results, Jordi Vila, MD, head of the Department of Clinical Microbiology, Hospital Clinic, Barcelona, Spain, said the study confirms that the 3 nosocomial infections in ICU patients — sepsis, surgical-site infection, and hospital-acquired pneumonia — generate an economic and clinical burden in the hospital setting.

"Future measures need to be taken to decrease the prevalence of these infections. The additional value of this study is the high number of patients included," he said.

Dr. Joly is an employee of Sanofi, France. Dr. Brun-Buisson, Dr. Christopher, and Dr. Vila have disclosed no relevant financial relationships.

22nd European Congress of Clinical Microbiology and Infectious Diseases (ECCMID): Poster 1128 and abstract O312. Presented April 1, 2012.