SAN DIEGO – About one-fourth of survivors after critical illness are substantially affected by symptoms of posttraumatic stress syndrome up to 1 year after discharge from the ICU, results from a large meta-analysis demonstrated.
"This incidence is as high as PTSD following other traumatic exposures, such as wartime combat," Dr. Ann Parker said in an interview in advance of an international conference of the American Thoracic Society, where the research was presented. "It is important for clinicians to recognize that patients with preexisting psychological symptoms, receiving benzodiazepines for sedation in the ICU or reporting memories of ‘frightening’ ICU experiences are at increased risk of developing PTSD following critical illness."
Dr. Ann Parker
In a study led by Dr. Dale M. Needham, medical director of Johns Hopkins University’s critical care physical medicine and rehabilitation program, Dr. Parker and fellow first author Dr. Thiti Sricharoenchai searched PubMed and four other databases to perform a systematic review and meta-analysis of the prevalence of and risk factors for PTSD in survivors of critical illness.
"The number of studies investigating PTSD among critical illness survivors has doubled since the publication of prior reviews [in 2007 and 2008]," said Dr. Parker, who is a fellow in pulmonary and critical care medicine at Johns Hopkins University, Baltimore. "With these additional publications, there is greater similarity between studies regarding the timing of and instruments used for PTSD symptom assessment. As a result, we provide the first meta-analysis yielding a pooled prevalence of clinically important PTSD symptoms."
The researchers evaluated the databases from inception through July 15, 2012, for studies that included adult ICU survivors, used a validated PTSD instrument 1 month or more post-ICU discharge, focused on general ICU populations, and included at least 10 patients with substantial PTSD symptoms. In all, 28 articles on 25 unique cohorts representing a total of 3,437 patients were identified. The most common validated PTSD instrument used in the studies was the Impact of Events Scale (IES), a scoring system that ranges from 0-75, with higher scores indicating greater symptoms.
Dr. Parker reported that among 429 patients who were assessed 1-6 months post-ICU discharge, the pooled mean IES score was 19 and the pooled prevalence of clinically important PTSD symptoms ranged from 23% to 42%. Among 698 patients who were assessed 7-12 months post-ICU discharge, the pooled mean IES score was 17 and the pooled prevalence of clinically important PTSD symptoms ranged from 17% to 34%. In other studies the prevalence of PTSD symptoms ranged from 5% to 62%.
Risk factors for PTSD in critical illness survivors included patient-specific factors, such as younger age and preexisting mental health disorders, as well as ICU-specific factors, including sedation with benzodiazepines and memories of "frightening" experiences in the ICU. In addition, PTSD symptoms were associated with worse quality of life.
"It seems that patients’ memories of their ICU experiences after ICU discharge may play a more important role than the duration of their stay in the ICU or the severity of their illness," Dr. Parker said. "Patients who recalled ‘frightening’ memories were more likely to have substantial PTSD symptoms. These memories may be related to delirium in the ICU, but few studies have attempted to evaluate this theory. There are very few interventions with proven efficacy for reducing PTSD symptoms in critical illness survivors. One intervention that has shown promising results in two European studies is the ICU diary, which has not been rigorously evaluated in North America."
She acknowledged certain limitations of the study, including the fact that differences between studies regarding sample populations and PTSD symptom instruments "make direct comparison difficult. However, the use of meta-analysis to pool the results of the 10 studies using the IES for PTSD symptom assessment strengthens the assertion that PTSD symptoms are highly prevalent among general critical illness survivors."
Neither Dr. Parker nor her associates had relevant financial disclosures.