News Release

Antidepressant use associated with increased mortality among critically ill patients?

Peer-Reviewed Publication

American Thoracic Society

ATS 2012, SAN FRANCISCO – Researchers at Beth Israel Deaconess Medical Center, in Boston, and the Massachusetts Institute of Technology in Cambridge, have found that critically ill patients were more likely to die if they were taking the most commonly prescribed antidepressants when they were admitted to the intensive care unit (ICU).

The researchers conducted a retrospective study using the electronic medical records of 10,568 patients to look at in-hospital mortality and mortality a year after being admitted to the ICU.

"We found that mortality was higher overall for patients taking antidepressants prior to admission to the ICU and remained significantly higher a year later," said Katherine. M. Berg, MD, one of the physicians involved in the study. "We also found that certain subgroups of patients, particularly patients admitted for cardiac problems, appeared to be at even greater risk."

The study results will be presented at the ATS 2012 International Conference in San Francisco.

Of the 10,568 patient records studied, 1,876 patients were taking either a selective serotonin reuptake inhibitor (SSRI) or a serotonin and norepinephrine reuptake inhibitor (SNRI) prior to admission to an ICU. They were compared to patients admitted to an ICU who were not on one of these medications. After adjusting for age, gender, ICD-9 diagnosis, disease severity and co-morbidities, the researchers found that patients on SSRI/SNRI's prior to admission to the ICU were 73 percent more likely to die in the hospital (p<0.001), and that the increase in risk persisted at one year.

Among patient groups, risk was highest among patients with acute coronary syndrome and those who had undergone cardiac surgery. For both of these patient groups, the risk of dying in the hospital was more than double if they were taking one of these antidepressants prior to admission (OR 2.41; p<.0020 and 2.08; p<0.001, respectively).

Not all patient groups demonstrated that patients on SSRI/SNRI's were more likely to die. There was, for example, no increase in mortality among patients admitted with sepsis.

Both classes of antidepressants, which work by increasing the levels of neurotransmitters in the brain, are generally thought to have fewer side effects than previous generations of antidepressants. Recent studies, however, have found that SSRIs may increase a patient's risk of bleeding, dizziness, falls and stroke.

"Major depression is a common disorder affecting more than 16 percent of adults in the United States, and SSRI's are the most commonly prescribed medication class for this disease," added Dr. Berg. "The benefits of SSRI's for the treatment of depression are well documented. Due to the practical limitations of clinical trials, however, the long-term risks are unknown."

The researchers acknowledged that this study shows an association, but that the preadmission SSRI/SNRI use may not have been the cause of the increased mortality. The authors attempted to control for other factors that could lead to increased mortality, but were unable to control for some potentially important ones such as smoking and depression itself.

The authors stressed that these results require validation by similar studies utilizing other ICU databases. They stated, however, that this information highlights the need for alternative ways to monitor for potential adverse effects of medications, and the role that large clinical databases may play in this line of investigation.


The study was undertaken using the Multiparameter Intelligent Monitoring in Intensive Care II database (MIMIC II). This public, de-identified database was developed with the support of the National Institute of Biomedical Imaging and Bioengineering. ###

"Pre-Admission Use Of Selective Serotonin Reuptake Inhibitors Is Associated With ICU Mortality" (Session C24, Tuesday, May 22, 2012, 8:15 a.m - 10:45 a.m., Room 309, Moscone Center; Abstract 31998).

* Please note that numbers in this release may differ slightly from those in the abstract. Many of these investigations are ongoing; the release represents the most up-to-date data available at press time.

Abstract 34176
Pre-Admission Use Of Selective Serotonin Reuptake Inhibitors Is Associated With ICU Mortality
Type: Late Breaking Abstract
Category: 04.07 - ICU Management/Outcome (CC)
Authors: K.M. Berg1, M. Ghassemi2, M.W. Donnino1, J. Marshall1, L. Celi1; 1Beth Israel Deaconess Medical Center - Boston, MA/US, 2Massachusetts Institute of Technology - Cambridge, MA/US

Abstract Body

Rationale: Major depression is a common disorder affecting more than 16% of adults in the United States. In 2010, antidepressants were the second most-prescribed medication class in the United States. SSRI's have been regarded as less toxic than earlier generation antidepressants, but in the past several years studies have found several adverse effects, including increased risk of bleeding, falls, stroke and orthostatic hypotension. A high percentage of patients admitted to the intensive care unit (ICU) are on SSRI's, and we therefore sought to determine if pre-admission SSRI use was associated with mortality both in-hospital and at one year.

Methods: We queried the MIMIC II 2.6 database, which includes information from the electronic medical records on 26,870 ICU patients at the Beth Israel Deaconess medical Center (BIDMC) collected since 2001. The exposure of interest was SSRI/SNRI use prior to ICU admission. The primary outcome was in-hospital mortality. Patients' age, gender, ICD9 diagnoses, initial laboratory values, vital signs during the first 24 hours of admission and Simplified Acute Physiology Score (SAPS) were also collected. A combined Elixhauser score was calculated for each patient to represent co-morbidities.

Results: 21,903 patient discharge summaries were available for single-admission ICU patients. After excluding patients for uncertainty regarding pre-hospital medication use, 10,568 patient records were included for analysis, 1876 in the SSRI/SNRI group and 8,692 controls. Female gender and history of COPD or diabetes were slightly more prevalent in the SSRI/SNRI group. After adjustment for age, SAPS, and combined Elixhauser score, SSRI/SNRI use was associated with an OR for in-hospital mortality of 1.73 (95% CI 1.46-2.13, p<0.001). The increased mortality risk was maintained over one year.(Fig.1) Among the patient subgroups, risk was highest among patients with acute coronary syndrome (OR=2.41, 95% CI 1.38-4.21, p=0.002), and those who underwent cardiac surgery (OR=2.08, 95% CI 1.45-3.01, p<0.001). No increased risk was seen among patients presenting with sepsis or diabetes. Mortality appeared to vary by specific SSRI.(Table 4)

Conclusion: We found an increase in in-hospital and one-year mortality in patients being treated with SSRI/SNRI's when compared to controls. Mortality was higher in patients on medications with a higher degree of serotonin reuptake inhibition. Our study is limited by its retrospective nature, and by an inability to control for smoking, severity of depression, and other possible confounders in the depressed population.

Funded by: None reported

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.