This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA
REVIEW
Identifying clinical and acute psychological risk factors for PTSD after critical care: a systematic review D. WADE 1, R. HARDY 2,D. HOWELL 1, M. MYTHEN 3 1The
Critical Care Unit, University College London Hospitals NHS Foundation Trust (UCLH), London, UK; 2MRC Unit for Lifelong Health and Ageing, Division of Population Health, University College London (UCL), UK; 3National Institute for Health Research Biomedical Research Centre, UCL, London, UK
ABSTRACT Background. Patients may suffer extreme psychological reactions in intensive care units (ICU), and post-traumatic stress disorder (PTSD) after leaving hospital. Previous systematic reviews of studies up to 2007 found that the true prevalence of and consistent risk factors for PTSD after ICU were not established, due to methodological shortcomings of studies. Therefore we aimed to conduct a systematic review of observational studies of post-ICU PTSD from 2008-2012, and to compare them to 1997-2007 studies, with regard to quality, prevalence estimates and risk factors. Methods. We used a pre-specified protocol, and systematic, explicit methods to identify, select and critically appraise studies. Studies in general ICU settings with mixed-diagnosis patients (N.>30) were included. Risk of bias was assessed, with lower-risk studies given greater weight. No quantitative synthesis was possible due to heterogeneity, therefore ranges of estimates and frequencies of risk factors were examined. Results. The review included 26 papers, 13 from 1997-2007 and 13 from 2008-2012. There were more high quality studies in the latter period. The range of prevalence estimates from high-quality studies was similar; 8% to 27% (1997-2007) and 9% to 27% (2008-2012). Clinical risk factors consistently identified over the two periods were use of benzodiazepines, duration of sedation and mechanical ventilation. Psychological risk factors include stress and fear experienced acutely in ICU, and frightening memories of the admission. Conclusion. The quality and number of post-ICU PTSD studies has increased over time, and we can be more confident in the accumulated findings. Evidence from both periods suggests that up to 27% of ICU survivors suffer from PTSD. There is also increasing evidence that use of benzodiazepines and duration of sedation, along with fear, stress and delirium in the ICU are likely risk factors for subsequent PTSD. (Minerva Anestesiol 2013;79:944-63) Key words: Critical care - Stress disorders, traumatic - Risk factors.
A
s critical care medicine advances and more patients survive, attention has turned to the quality of their survival. It is now recognised that physical and psychological recovery of critical care patients may be poor.1 Since the late 1990s there has been particular concern about the prevalence of post-traumatic stress disorder (PTSD) after intensive care.2, 3 PTSD is an “anxiety disorder that often follows exposure to an extreme stressor that causes injury, threatens life or physical integrity”.4 The person’s immediate response involves intense fear, helplessness or horror. The disorder
944
is characterised by three clusters of symptoms: re-experiencing, avoidance and hyper-arousal, that persist for more than a month and cause distress or impaired functioning. As well as mental suffering, patients with PTSD are more likely to endure worse physical health, quality of life and mortality.5 Critically ill patients are at high risk of PTSD for a number of reasons. They suffer life-threatening illness, a known traumatic stressor that may precede PTSD.4 They undergo treatments and procedures that may save lives, but can also
MINERVA ANESTESIOLOGICA
August 2013
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
be invasive and disturbing, including intubation, ventilation, catheterisation and haemofiltration. Patients describe the cumulative stress of pain, thirst, hunger, sleep deprivation, noise and light, inability to communicate, fear of dying and loss of control in the intensive care unit (ICU) as overwhelming.6 Furthermore, many patients receive a cocktail of psychoactive drugs including benzodiazepines, inotropes, anti-psychotics and corticosteroids that have poorly understood effects on the brain and emotions of a critically ill patient. Consequently, patients may experience anxiety, panic, low mood and delirious symptoms including hallucinations and terrifying delusions during their stay in an ICU.7, 8 Three systematic reviews of PTSD after critical care appraised studies up to 2007. Prevalence of post-ICU PTSD was reported as being in the range of 0-64% 9, 10 or as having a median point prevalence of 22%.11 Two reviews 9, 11 identified potential risk factors for PTSD including delusional memories of ICU, use of sedation, psychiatric history, younger age and female gender. However few risk factors were consistently identified across studies. PTSD remains a topic of interest in the critical care literature, with 63 articles published in 2011-2012, and more continuing to appear after this review was completed. One 2012 study found that 39% of patients screened positive for PTSD at 12 months.12 All were acute respiratory distress syndrome (ARDS) patients, a sub-group of ICU patients who may have higher than average PTSD rates. As well as PTSD in patients, there has been increasing recent research interest in PTSD among relatives and ICU staff. Symptoms of PTSD have been detected in up to 33% of family members 13 and 24% of ICU nurses.14 Since 2007 there have also been a small number of trials of ICU interventions to prevent PTSD. Patient diaries were pioneered in Scandinavia to fill in memory gaps and help patients come to terms with their experience.15 A pan-European RCT found that patients receiving a diary intervention had a significantly lower incidence of PTSD than a control group (5% vs. 13%).16 Another study investigated the effect of early support by psychologists in critical care.17 The intervention group had a significantly lower risk of PTSD
Vol. 79 - No. 8
WADE
(21% vs. 57%) at 12 months, but the study had a historical control rather than RCT design. As yet there are too few intervention studies to warrant a systematic review. However more information is needed about important, consistent risk factors for PTSD so that high-risk patients can be monitored and followed-up. Furthermore, it is recognized that for psychological interventions to be successful, they should be designed to target modifiable risk factors.18 All three systematic reviews to 2007 concluded that the studies they reviewed had deficiencies in design, methodology and reporting, as well as conflicting results, that limited the conclusions that could be drawn about either prevalence of, or risk factors for post-ICU PTSD. For example, Griffiths et al.10 suggested that more rigorous, larger studies focusing on general ICU patients, rather than sub-groups, should be carried out. They also recommended carrying out longitudinal studies with assessments over multiple timepoints. Other recommendations from the reviews were that studies should examine the effects of pre-ICU psychopathology, sedation strategies, and in-ICU delirium on subsequent PTSD.9, 11 In order to assess whether the evidence on post-ICU PTSD had improved in the suggested ways since publication of the three systematic reviews, we aimed to carry out a systematic review of observational studies published from 2008 to 2012. Second, we aimed to compare the newer studies, with the studies published up to 2007, in terms of size and methodological quality. Third, unlike the previous systematic reviews, which included studies of sub-groups and very small studies, we aimed to include only studies of general mixed-diagnosis ICU patients and studies with at least 30 patients. Finally, our review aimed to detect new evidence about clinical and acute psychological risk factors for, as well as prevalence of PTSD, since 2007. We were aware that increased recognition of psychological stress in the ICU, as well as improvements in ICU therapies might have reduced the prevalence of PTSD and altered risk factors since 2007. The quality of more recent studies might be higher and estimates more accurate. Our overall aim was to compare 2008-2012 studies with 1997-2007 studies, with regard to PTSD prevalence, risk factors and quality.
MINERVA ANESTESIOLOGICA
945
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA WADE
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
Review questions
Types of outcome measures
1. What percentage of survivors of general intensive care treatment suffers from post-ICU PTSD in the months after intensive care? 2. What are the clinical and acute psychological risk factors for post-ICU PTSD? 3. Has the quality of evidence regarding the first two questions improved between the 19972007 period and the 2008-2012 period?
Studies were selected if they used reliable, validated questionnaires or interviews for PTSD. Studies that used single item measures or unvalidated bespoke questionnaires for PTSD were excluded.
Materials and methods The systematic review was conducted according to PRISMA recommendations;19 It was based on a pre-specified protocol, and used systematic and explicit methods to identify, select and critically appraise studies. The risk of bias in studies was assessed and higher quality studies were given greater weight.
Studies were excluded if: 1. their sample size was smaller than 30 participants; 2. they were published only as conference papers or abstracts; 3. full text was not available in English; 4. they were published pre-1997; 5. they were set in neo-natal or pediatric ICUs. Search strategy
Criteria for study selection Three criteria were used to select studies for inclusion in the systematic review. Type of studies Prospective cohort studies, retrospective cohort studies, and cross-sectional surveys were included in the review. Data from the control groups in RCTs of interventions to reduce psychological morbidity in ICU patients were also considered eligible. Types of participants The study populations were adult, mixed-diagnosis ICU patients who received intensive care >24 hours in general, medical or surgical ICUs. Studies of ICU sub-groups such as patients with ARDS or pancreatitis were not eligible as they are not representative of general ICU patients. However studies of ICU patients receiving mechanical ventilation were included. Patients who receive mechanical ventilation have many different underlying conditions so they are approximately representative of the general ICU population.
946
Exclusion criteria:
Studies were identified on June 28, 2012 from the following databases: —— Medline (Ovid SP 1946 -present) —— Embase (Ovid SP, 1947-present) —— PsycINFO (Ovid SP, 1806 -present) —— CinahlPlus (EBSCO Host, 1937-present) The initial search was carried out on Medline using the strategy outlined in Table I. Similar searches were carried out on the other three databases. On September 12, 2012 the searches were re-run. There were nine new references in the Medline search and none in the others. Only one new study20 was eligible and has been added to the review. A further study was identified in October 2012.6 Assessment of risk of bias (quality assessment) It should be noted that the PRISMA statement 19 recommends that reviewers should use the phrase “assessment of risk of bias”, rather than “quality assessment”. The terms are used interchangeably in this review. We based our quality assessment on methodology checklists 21 for study designs including cohort studies. Quality criteria for robustness of outcome data were used. For example a study that used the Impact
MINERVA ANESTESIOLOGICA
August 2013
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
WADE
Table I.—Medline search strategy. 1. #1 #2 #3 #4 #5 #6 #7 #8 #9 #10
MEDLINE Search terms 1946-Present Exp critical care/ (critical care or intensive care or critical illness or ICU or ITU).ti, ab. 1 or 2 (neonatal or pediatric or baby or babies or infant or PICU or child).ti, ab. 3 not 4 Limit 5 to (english language and yr= “1997-Current” and (clinical trial or comparative study or journal article or multicenter study or randomized controlled trial)) Stress disorders, traumatic/ or stress disorders, post-traumatic/ or stress disorders, traumatic, acute/ (PTSD or “posttraumatic stress disorder” or “post-traumatic stress disorder” or “post traumatic stress disorder” or “posttraumatic stress” or “post-traumatic stress” or “post traumatic stress”).ti, ab. 7 or 8 6 and 9
Table II.—Quality criteria used in assessment of risk of bias. The sample 1. A clear definition of source population and clear eligibility criteria for selection of subjects are used, to ensure the sample is representative. 2. Comparison is made between full participants and those lost to follow up. 3. A power calculation is reported. If not, sample size is small, medium or large. Outcome 4. The likelihood that some subjects might have the outcome at baseline is accounted for. 5. The outcomes are clearly defined. 6. Evidence is used to demonstrate that measure of outcome is valid and reliable. 7. Follow-up is long enough for outcome to occur. Risk factors-outcome analysis 8. The study addresses an appropriate and clearly focused question. 9. Any measures of risk factors are reliable. 10. Main potential confounders are identified and taken into account in design and analysis. 11. Confidence intervals have been provided. 12. Appropriate statistical analyses have been carried out. Overall assessment How well was study done a) to minimise risk of bias and b) to establish a causal relationship between exposure and effect.? Adapted from Scottish Intercollegiate Guidelines Network checklist for cohort studies
of Events-Revised Scale 22 scored more highly than one using the Impact of Events Scale 23 as the latter includes only two of three symptom clusters necessary for a diagnosis of PTSD.4 Another criterion – controlling for confounding factors – is important in evaluating the quality of observational studies. A further criterion was use of appropriate statistical analysis. The quality criteria are shown in Table II. Using the recommended system,21 the majority of 1997-2007 studies were assessed as medium quality, with few poor or high quality studies. To provide better differentiation between studies (Table II), numerical scores were
Vol. 79 - No. 8
assigned to each study based on poor (0), adequate (1) and good (2) ratings for each of four selected parameters. These were: Representativeness of the sample (based on criteria 1 and 2); Power of the study/sample size (criterion 3), Robustness of outcome assessment (criteria 4-7), and Appropriateness of statistical analysis used (criteria 8-12). For sample size, studies with 3059 participants were rated poor/small; studies of 60-150 adequate/medium and studies with 150 or more participants, or that included a power calculation, were good/large. The first three parameters applied to prevalence estimation; the fourth to risk factor-outcome analysis. The
MINERVA ANESTESIOLOGICA
947
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA WADE
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
range of scores was 0-6 out of 6 for prevalence, and 0-2 out of 2 for the risk factor analysis. This numerical system was then used to assess the 2008-2012 studies. Inter-rater reliability DW carried out quality assessment of 19972007 papers and ER of 2008-2012 papers. Three other raters assessed three papers each to assess inter-rater reliability. There was 100% agreement between all raters in the quality assessment of the 9 papers.
Synthesis of extracted evidence It was not possible to carry out a meta-analysis of PTSD outcomes due to heterogeneity of both results and methods of studies, as well as lack of consistency in reporting results between studies. Therefore we examined ranges of estimates and identified reasons for variation in results, using quality (risk of bias) criteria. Synthesis of information about risk factors was difficult as few studies reported results in a comprehensive manner, particularly in the pre-2007 period. Therefore we summarised the number of times associations were found or not found across studies.
Data extraction strategy Data were systematically extracted for each study using a data extraction form by DW and ER.
Results A total of 503 papers were retrieved in the search outlined above (Figure 1).
Total papers retrieved N.=503
Title screening
Total without duplicates N.=339
Abstract screening
Eligible on title N.=177
Eligible for full paper review N.=49
Full paper review
Synthesis N.=26
1997-2007 N.=13
62 duplicates automatically removed; 102 manually removed 164 excluded: (Inclusion criteria not met e.g. neonatal/paediatric settings; reviews; PTSD in staff not patients; qualitative methods; end of life patients; physiological or mechanistic studies) 128 excluded: (Inclusion criteria not met e.g. editorials, reviews, sub-groups of ICU patients)
23 excluded: (Inclusion criteria not met: e.g. abstracts only, editorials, studies of ICU sub-groups, unvalidated PTSD questionnaires used)
2008-2012 N.=13
Figure 1.—Flowchart of reference retrieval, exclusions and inclusions.
948
MINERVA ANESTESIOLOGICA
August 2013
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
WADE
Exclusion
After removing duplicates, 339 titles and then 177 abstracts were screened, and the full-text versions of 49 potentially eligible papers were reviewed. Of these, 26 papers were eligible for inclusion in the review; 13 from 1997-2007 and 13 from 2008-2012.
In the earlier period, seven studies excluded patients for psychiatric or neurological reasons, but from 2008 almost all (11) studies excluded those patients (Table III).
Characteristics of included PTSD studies
Patient characteristics
There were 13 cohorts of patients in 13 studies in the 1997-2007 period and 12 cohorts in 13 studies from 2008-2012, as two papers 39, 40 covered the same cohort of patients at different time-points. There were 2703 unique patients followed up in the 26 studies, and the separate totals (1119 in 1997-2007, and 1584 in 20082012) show that more patients were assessed in the latter period. There were nine medium-size (N.=60-150) or large (N.>150) studies in the 2008 group, compared to seven in the earlier group. Both groups of studies took place in homogeneous settings (general, medical or surgical ICUs with patients of mixed diagnoses). There were more multi-site studies published in 20082012 (5 vs. 3). From 1997-2007 the majority of studies (7) took place in the UK, with five in mainland Europe and one in the US. Among the 2008-12 studies, seven were in Europe, three in the UK, two in US and one in Australia. Most of the earlier studies were prospective cohort studies (8), with one RCT and four other designs. By the later period, there were more RCTs (4), fewer prospective cohorts (5) and four others. Further details can be seen in Table III.
Age
Inclusion/exclusion criteria of included studies Inclusion From 1997-2007, most studies (11) had inclusion criteria for length of stay (LoS, ranging from >1 to >6 days) or mechanical ventilation (MV, ranging from >1 to >3 days). But six of the 2008-2102 studies had no inclusion criteria for LoS or MV. In the other seven later studies inclusion criteria were LoS (>2 to >4 days) and MV (>12 to >36 hours).
Vol. 79 - No. 8
There was little difference between the range of median/mean ages within the two groups of studies; (42-68 years) from 1997 and (48-69 years) from 2008. Sex There were larger percentages of men in the earlier studies (43-76%) compared with the later studies (25-68%). Illness severity Mean or median Apache II scores of cohorts appeared to be similar; (12-25) in 1997-2007 studies, versus (13-28) in 2008-2012. Length of stay in ICU There was a shorter range of average stays (2.5-21 days) in the 2008 - 2012 studies, compared to a range of 5-52 days in the earlier period (Table III). Quality assessment of PTSD studies According to our quality criteria, more recent studies (2008-2012) have a lower risk of bias in estimating prevalence of PTSD, than older ones (1997-2007). Nine new studies had a prevalence rating of four or more (out of 6), whereas only four old studies scored four or more, and nine old studies had a high risk of bias (Table IV). Ranges of response rates (percentage of recruited patients assessed at follow-up) were similar in both groups of studies (24% to 88% vs. 35% to 91%).
MINERVA ANESTESIOLOGICA
949
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA WADE
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
Table III.—Characteristics of post-ICU PTSD studies. First author and year
N assessed for PTSD symptoms
Type of ICU, country, No. sites Inclusion criteria in study
Exclusion criteria in study (psychological, neurological)
1997-2007 studies Capuzzo (2005) 24 Cuthbertson (2004) 25
63 78
General ICU, Italy General ICU, UK
LoS > 3days
Psychiatric history
Girard (2007) 26
43
Medical ICU, US
MV
Griffiths (2006) 27
108
General ICU, UK
Jones (2001) 28
30
General ICU, UK
LoS>3 days, seen at followup clinic LoS>24h MV
Neurologic disease, mental disability
Jones (2003) 29
General ICUs, UK, 3 sites
LoS>48h MV
Jones (2007) 30
44 in control group at 6 m 238
General ICUs, Europe, 5 sites
ICU>48h MV
Nickel (2004) 31
41
Medical ICU, Germany
LoS>24h
Rattray (2005) 32 Richter (2006) 33
60 at 6 m 80 at12 m 37
General ICU, UK, Surgical ICU, Germany
LoS>24h, emergencies LoS > 30 days
Samuelson (2007) 7
226
General ICUs, Sweden, 2 sites
MV>24h
Scragg (2001) 34
80
General ICU, UK
51 at 3 m 45 at 9 m
General ICU, UK
LoS>72h
36 (two control groups, pre- and post-) 299
Medical- surgical ICU, France
LoS ≥4 days
General ICU, Portugal 9 sites
LoS >48h
32 at 3 m 25 at 12 m (control group) 160 (control group)
Medical ICU, USA
MV>12h
Europe, 6 sites
LoS>72h, MV>24h
Myhren (2009)38
255
General ICU, Norway
LoS>24h
Myhren (2010)39 Rattray (2010) 40
194 42
General ICU, Norway UK, 6 sites
LoS>24h
Treggiari (2009) 41
129
MV>12h
Twigg (2008) 42
44
Medical and surgical ICUs, Switzerland General ICUs, UK, 2 sites
Van der Schaaf (2009) 43 Wade (2012) 6
238 100
Mixed ICU, Netherlands General ICU, UK
LoS>48h
Wallen (2008) 44
100
ICU, Australia
Weinert (2008) 45
149 at 2 m 80 at 6 m
Medical/surgical ICUs, US
Sukantarat (2007) 35 2008-2012 studies Garrouste-Orgeas (2012) 20 Granja (2008) 36 Jackson (2010)37 Jones (2010)16
MV>36h
Psychotic illness, Suicide attempt, head injury Psychotic illness, neurosurgery Psychosis, suicide
Psychosis, suicide, head injury, mental disability Head injury, other injury
Dementia
Neurologic deficits, neurosurgery Psychotic illness or PTSD. confusion Psychiatric history, severe head injury As above Head injury, neurosurgery Neurologic conditions, mental disability Dementia, confusion, overdose Dementia, persistent confusion in ICU Psychosis, cognitive impairment
ICU: Intensive Care Unit; LoS: length of stay; MV=mechanical ventilation; h=hours; IQR: interquartile range; RCT: randomised controlled trial; SD: standard deviation; SAPS: Simplified Acute Physiology Score. If cells are empty, data were not reported in studies. Some studies reported characteristics of the sample at baseline; others reported characteristics of the sample at follow-up.
950
MINERVA ANESTESIOLOGICA
August 2013
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
WADE
Design of study
Age (years)
Sex (% men)
Length of stay in ICU (days)
Apache II score (or similar)
Prospective cohort Prospective cohort
69 (median) 58 (median) 18-87 (range) 52 (median) 39-65 (IQR) 57 (mean) 17-85 (range) 57 (median) 17-82 (range)
60% 56 %
5 (median) 6 (median) 1-51 (range) 10 (median) 5-13 (IQR) 12 (median) 2-101 (range) 8 (median) 1-60 (range)
14 (median) 18 (median) 4-38 (range) 25 (median) 20-31 (IQR)
59 (mean) 16 (SD) 61 (median) 17-86 (range) 47 (mean)
57%
13 (mean) 18 (SD) 7 (median) 2-76 (range)
16 (mean) 5 (SD) 16 (median) 3-36 (range) 12 (mean, maximum score obtained) 11 (SD)
42 (mean) 17 (SD) 63 (mean) 13 (SD) 57 (median) 19-90 (range) 57 (mean) 14 (SD)
76%
52 (mean) 20 (SD) 6 (mean) 6 (SD)
20 (mean) 7 (SD) 18 (median) 12 (IQR)
17 (mean) 17 (SD)
15 (mean) 6 (SD)
Prospective cohort Cross-sectional Case-series prospective cohort RCT (control group) Prospective cohort Cross- sectional Prospective cohort Retrospective cohort Prospective cohort Retrospective cohort Prospective cohort
47% 66% 44 %
68%
52% 47% 43%
17 (median) 4-28 (range)
Non-randomised trial: Pre-and post- controls
Pre- 68 (mean) 14 (SD) post- 62 (mean) 16 (SD)
52% 54%
Pre 21 (mean) 16 (SD) SAPS II score Pre 44 (mean) 14 Post 13 (mean) 18 (SD) (SD) Post 40 (mean) 15(SD)
Cross-sectional cohort
59 (mean) 44 -71 (IQR)
58%
8 (median) 5-13 (IQR)
Nested substudy of RCT (control group) RCT (control group)
68 (median) IQR (56-76)
45%
SAPS II score 37 (median) 30-46 (IQR) 28 (median) 21-33 (IQR)
59 (mean), 16 (SD)
62%
13 (mean) 12 (SD)
19 (mean) 7 (SD)
Cross-sectional cohort
48 (mean) 16 (SD)
63%
12 (mean) 10-14 (CI)
SAPS II 37 (mean) 35-39 (CI)
Prospective cohort Prospective cohort
48 (mean) 16 (SD) 60 (mean) 17-84 (range)
63% 63.%
12 (mean) 10-14 (CI) 7 (median) 0-63 (range)
SAPS II 37 (mean) 35-39 (CI) 19 (mean) 6-34 (range)
11(median) 7(median) 2 sites 9(mean) 10(SD) 8 (median) 85 (range)
16 (median) 14 (median) 2 sites 15(mean) 6 (SD) 22 (median) 7 (range)
57h (median) 62h (IQR)
13 (median) 2-40(range)
RCT (both groups, light v deep sedation) Case series cohort
56 (median)
45%
Cross-sectional cohort Prospective cohort
59 (mean) 17 (SD) 57 (mean) 17(SD)
66% 52%
Prospective cohort
63 (mean) 19 (SD)
68%
Prospective cohort
54 (mean) 45, 63(IQR)
52%
Vol. 79 - No. 8
72.3%
MINERVA ANESTESIOLOGICA
951
WADE
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
Table IV.—Assessment of risk of bias (quality assessment) of post-ICU PTSD studies. Ratings for risk of bias assessment
Follow-up rate*
4. Analysis of association
1. Prevalence ‡ rating (max=6)
2. Risk factor – outcome analysis rating (max=2)
1997-2007 studies Capuzzo (2005) 24
N assessed for PTSD
3. Outcome assessment
First author and year
2. Sample size†
Quality scores
1. Representativeness of sample
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA
63
75%
1
1
0
n/a
2
n/a 1
78
70%
2
1
2
1
5
Girard (2007) 26
43
24%
2
0
1
2
3
2
Griffiths (2006) 27
108
67%
1
1
1
n/a
3
n/a
Jones (2001) 28
30
unclear
0
0
1
0
1
0
Jones (2003) 29
44 Controls
77%
2
0
1
0
3
0
Jones (2007) 30
238
78%
1
2
2
2
5
2
Nickel (2004) 31
41
n/a
1
0
2
0
3
0
Rattray (2005) 32
60 at 6 m 80 at 12 m
55% 73%
2
1
1
n/a
4
n/a
Cuthbertson
(2004)25
Richter (2006) 33
37
n/a
1
0
2
0
3
0
Samuelson (2007) 7
226
72%
2
2
2
2
6
2
Scragg (2001)34
80
n/a
1
1
1
0
3
0
51 at 3 m 45 at 9 m
100% 88%
1
0
1
0
2
0
36 (sum of 2 control groups)
35% 41%
1
0
1
1
2
1
Sukantarat
(2007)35
2008-2012 studies Garrouste-Orgeas (2012) 20 Granja (2008) 36
299
n/a
2
2
1
1
5
1
Jackson (2010) 37
32 (3m) 25 (12m)
38% 29%
1
0
1
n/a
2
n/a
Jones (2010) 38
160 (control group)
91%
2
2
2
n/a
6
n/a
Myhren (2009) 38
255
n/a
1
2
1
1
4
1
(2010) 39
194
76%
2
2
1
2
5
2
Rattray (2010) 40
42
41%
1
0
0
n/a
1
n/a
Treggiari (2009) 41
129
94%
2
2
1
2
5
2
Twigg (2008) 42
44
79%
1
0
2
n/a
3
n/a
Myhren
Van der Schaaf (2009) 43
238
n/a
2
2
1
n/a
5
n/a
Wade (2012) 6
100
64%
2
2
2
2
6
2
Wallen (2008) 44
100
88%
1
1
2
2
4
2
Weinert (2008) 45
149 at 2 m 80 at 6 m
54% 29%
2
1
2
1
5
1
NB Risk of bias assessment ratings were calculated according to criteria relevant for this review. They do not reflect the quality of other aspects of the studies, which may have other primary objectives. *Follow-up rate: % of those enrolled who completed PTSD follow-up (deaths are included in loss to follow-up). No rate given if study cross-sectional. †Quality scores: 0 (poor); 1(adequate); 2(good). Sample size scores; 0 (small): 30-59, 1 (medium): 60-150, 2 (large): >150 (or power calculation done) ‡ Prevalence rating was calculated by adding together quality scores 1-3. Risk factor-outcome analysis rating is simply based on quality score 4.
952
MINERVA ANESTESIOLOGICA
August 2013
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
WADE
Table V.—Prevalence of post-ICU PTSD.
First author and year
Time from ICU discharge to outcome assessment
N assessed for PTSD
PTSD measure
(1997-2007 studies, arranged in order of prevalence ratings) Samuelson 226 IES-R 2 m or (2007) 7 later 3m 238 PTSS-14, PDS Jones (2007) 30 Cuthbertson (2004) 25
3m
78
DTS
Rattray (2005) 32
6m 12 m
60 80
IES
Richter (2006) 33
35 m
37
Semi-structured psychiatric interview.
Scragg (2001) 34
Variable: 3-21 m
80
IES
Nickel (2004) 31
Variable: 3-15 m
41
PTSS-10. SCID
Griffiths (2006) 27
3m
108
Girard (2007) 26
6m
43
Trauma screening checklist PTSS-10
Jones (2003) 29 Capuzzo (2005) 24
6m
44 (control group) 63
Sukantarat (2007) 35
3m 9m
3m
51
Vol. 79 - No. 8
3m
ICU memory tool IES IES
45
2m 30 Jones (2001) 28 (2008-2012 studies, in order of prevalence ratings) 3m 160 Jones (2010)16 (control group) Wade (2012)6
IES
100
IES
PDS PDS
Results as reported
Interpretation of results
8≥30% on IES-R 8% borderline PTSD (95%CI: 4.8, 12) 9% 9% diagnosis of PTSD (95%CI: 5.5, 12.9) had PTSD using PDS 22% >27 on DTS 22% possible PTSD (95%CI: 12.8, 31.2) 12% likely PTSD 12% > 40 on DTS (95%CI: 0.8, 19.2) 27≥35% on IES 27% likely at 6m (95%CI : 17.7, 36.3) 24% likely at 12m 24≥35% on IES (95%CI: 15, 33) 32% (5/6 criteria) 32% sub-syndromal PTSD (95%CI: 17,47) 19% full diagnosis of PTSD 19% (6 criteria) (95%CI: 4, 34) 30%> unknown cut-off 30% possible PTSD (95%CI: 20, 40) 16% borderline PTSD 15.6%>30 on IES (95%CI: 7.7, 23.6) 17%≥35 on PTSS-10 17% likely PTSD (95%CI: 5.5, 28.5) 10% full diagnosis of PTSD 9.76% with SCID (95%CI: 0.7,18.9) 52% “PTSD” 52% possible PTSD (95%CI: 42.6, 61.4) on TSC 25≥27% on PTSS-10 25% possible PTSD (95%CI: 12,37.9) 14% likely PTSD 14≥35% on PTSS-10 (95%CI: 3.6, 24.4) 48% some impact 48>19% on IES (95%CI: 33.2, 62.8) 0% PTSD 0% on IES subscales
Prevalence rating (lowest risk =6)
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA
6 5 5
4
3
3
3
3 3
3 2
35>26% on IES (95%CI: 21.1, 48.9) 62>26% on IES (95%CI: 47.8, 76.2) 23>19% on IES (95%CI: 7.9, 38.1)
35% possible PTSD 62% possible PTSD
2
23% some impact
1
13% (95%CI: 7.8, 18.2) Identified by PDS 27% (95%CI: 18.3, 36) PDS severity score >18
13% diagnosis PTSD
6
27% likely PTSD
6
MINERVA ANESTESIOLOGICA
953
WADE
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
Table V.—Continues previousPTSD. page. V.—Prevalence from of post-ICU
First author and year
Time from ICU discharge to outcome assessment
N assessed for PTSD
PTSD measure
Results as reported
Interpretation of results
Prevalence rating (lowest risk =6)
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA
Treggiari (2009)41
4 weeks
129 (whole cohort)
IES-R for 25% of patients PCL for 75% of patients
9-10% (95%CI: 4.1, 13.9) met symptom criteria for a presumptive diagnosis
9-10% diagnosis PTSD
5
Granja (2008)36
6m
299
PTSS-14
18% (95%CI: 13.75, 22.25) with PTSS-14 score>49
18% high risk for PTSD
5
Myhren (2010)39
12 m
194
IES
27% (95%CI: 20.8, 33.2) with IES score >35
27% likely PTSD
5
van der Schaaf (2009)43
12 m
238
IES
18% (95%CI: 13.3, 22.7) with IES score >35
18% likely PTSD
5
Weinert (2008)45
2m 6m
149 80
PDS
17% (95%CI: 11, 23) 15% (95%CI: 7.2, 22.8) met diagnostic criteria
17% diagnosis 15% diagnosis
5
Wallen (2008)44
1m
100
IES-R
13% (95%CI: 6.4, 19.6) with IES-R score>33
13% likely PTSD
4
Myhren (2009)38
4-6 weeks
255
IES
27% (95%CI: 21.6, 32.4) with IES score>35
27% likely PTSD
4
Twigg (2008)42
3m
44
PDS, IES, PTSS-14
16% six criteria (95%CI: 5, 27) 27% five criteria PDS (95%CI: 13.9, 40.1)
16% diagnosis PTSD 27% likely PTSD
3
Garrouste-Orgeas (2012)20
12 m
36 (sum of two control groups)
IES-R
65% pre-group (95%CI: 50.4, 80.6) ≥22 as cut-off 74% post-group (95%CI: 60.3, 88.3) ≥22 as cut-off
65% and 75% of patients with some PTSD symptoms
2
Jackson (2010)37
3m 12 m
32 (control group) 25
PTSS-10
10% >35 (95%CI: -0.4, 20.4) 24% >35 (95%CI: 7.6, 40.4)
10% likely PTSD 24% likely PTSD
2
Rattray (2010)40
2m 6m
42 42
IES
36% >35 (95%CI: 21.5, 50.5) 24% >35 (95%CI: 11.1, 36.9)
36% likely PTSD 24% likely PTSD
1
PTSD measures used: DTS: Davidson Trauma Scale;47 IES: Impact of Events Scale;23 IES-R: Impact of Events Scale-revised (IES-R);22 PCL: PTSD Checklist;48 PDS: Posttraumatic Diagnostic Scale;46 PTSS-10: Post-traumatic Stress Syndrome 10-Questions Inventory;49 PTSS-14: UK Post-traumatic Stress Syndrome 14-Questions Inventory;42 SCID: Structured Clinical Interview for DSM-IV Axis I Disorders;51 TSQ: Trauma Screening Questionnaire.50
954
MINERVA ANESTESIOLOGICA
August 2013
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
Outcome assessment of PTSD studies In the 1997-2007 group of studies, outcomes were assessed at varying time points from 2 to 35 months after ICU discharge (Table V). From 2008-2012 outcomes were assessed at between 1 and 12 months. Several PTSD measures were used in the 26 studies, including self-report questionnaires such as the Posttraumatic Diagnostic Scale (PDS),46 the Impact of Events Scale (IES),23 the Impact of Events Scale-revised (IESR),22 the Davidson Trauma Scale (DTS),47 and the PTSD Checklist (PCL);48 screening instruments such as the Post-traumatic Stress Syndrome 10-Questions Inventory (PTSS-10),49 the UK Post-traumatic Stress Syndrome 14-Questions Inventory,42 and the Trauma Screening Questionnaire (TSQ),50 and a clinical interview - the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID).51 PTSD prevalence estimates (1997-2007) The range of PTSD prevalence estimates found was 0 to 62% across 13 studies (Table V). If only studies with a lower risk of bias (scoring 4-6) were included in the assessment (N.=4), the range of prevalence rate estimates was narrower, at 8% to 27%. Median prevalence of the four low-risk studies was 13%. PTSD prevalence estimates (2008-2012) The range of PTSD prevalence estimates was 9% to 75% across 13 studies. When only studies with lower risk of bias (4-6) were included (N.=9), the range of estimates was 9% to 27%. Median prevalence of the nine low-risk studies was 18%. Risk factors for PTSD There were eight studies with a low or medium risk of bias (1-2) for risk factor-outcome analysis in 2008-2012 versus four from 19972007 (Table VI). Pooling data from the two periods, Table VII shows that five studies identified age (usually younger age) as a risk factor, whereas seven studies did not. Most studies (7
Vol. 79 - No. 8
WADE
vs. 3) found that gender was not a risk factor for PTSD. Age and sex were less likely to be identified as risk factors for PTSD in the later studies than the earlier ones. Psychiatric history was found to be a risk factor in five studies (including one where results were unclear) but not in three. Three studies investigated an association between time since discharge from ICU and PTSD, and found no effect. The clinical risk factors most consistently detected were aspects of sedation and mechanical ventilation (Table VI). From 1997 to 2007 studies, lorazepam dose,26 administration of midazolam 6 and duration of sedation 30 were associated with PTSD in three studies. Another study found no effect of duration of sedation,33 and one study found no effect of total midazolam or total propofol dose.26 From 2008 to 2012 studies, days of sedation and use of benzodiazepines were associated with PTSD in one study.6 Light vs deep sedation 41 and degree of sedation 45 had no effect in two other studies. Mechanical ventilation was more frequently identified as a risk factor between 2008-12 than 1997-2007, with small effect sizes. One 2008 study found a nonlinear association between wakefulness during ventilation and PTSD (with lower PTSD in the least and most awake).45 Other clinical risk factors were identified in one study each: physical restraint,30 administration of inotropes/vasopressors; administration of antipsychotics; Therapeutic Intervention Scoring System (TISS) score; number of organs supported, days of cardiovascular support and number of psychoactive drug groups given.6 Diagnostic groupings, illness severity scores such as Apache II and LoS in the ICU were not identified as risk factors for PTSD in either period (Table VI). Acute psychological risk factors were frequently identified as risk factors for PTSD, with 28 associations found across both periods (Table VII). In studies focusing on memory and delirium, factors measured were traumatic memories (association found in 1 study),26 pain memories (association in 2 studies),38, 39 factual memories (2 associations),38, 39 delusional/delirious memories (association in 6; no association in 2), intrusive memories (association in 2 studies),6, 36 amnesia in ICU or pre-ICU (asso-
MINERVA ANESTESIOLOGICA
955
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA WADE
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
ciation in 2 studies,6, 36 none in 1 study),45 and delirium (association in 1 study;6 no association in another).26 Mood, stress, fear, agitation, panic, loss of control and inability to express wishes in the ICU, along with pessimism, peritraumatic dissociation and illness perceptions were other psychological risk factors tested (in 6 studies). All had positive associations with PTSD, with the exception of peritraumatic disassociation.20 Discussion In this systematic review, 13 eligible studies of post-ICU PTSD were retrieved from 19972007 and 13 studies from 2008-2012. This suggests an increasing interest in post-critical care PTSD in the last four years. Compared to 19972007, the 2008-12 period included more multisite studies, more medium to large studies, less restrictive inclusion criteria, more RCTs and a larger number of patients overall. More of the later studies were of high quality (i.e. had a lower risk of bias) than earlier studies. There was a remarkable concordance about the prevalence of PTSD, based on the higher quality studies from both periods (Table V). The range of estimates for PTSD from 1997-2007 was 8-27% with 13% as the median prevalence. From 2008-2012 studies, PTSD was estimated between 9-27% with a median prevalence of 18%. Variation in results may be related to differences in ICU populations, different measures used with different diagnostic thresholds and different lengths of follow-up. It should be noted that eight out of 26 studies used the (unrevised) IES 23 as a measure of PTSD. The IES is a good measure of distress related to life events but is not a measure of PTSD, as it includes only two of the three clusters of PTSD symptoms.4 Others used screening tools that cannot confirm the presence of PTSD, such as the PTSS-10 49 or TSQ.50 In the 2008 period, more studies (four vs one) used the PDS,46 a questionnaire that conforms to current diagnostic criteria for PTSD.4 A further three (vs. one from 1997-2007), used the IES-R,22 which includes the three clusters of PTSD symptoms. In the early period, it was problematic that
956
studies using the same measure used different cut-points denoting different meanings (such as likely PTSD, possible PTSD, probable PTSD, borderline PTSD and diagnosed PTSD). For example in studies using the IES,23 some used a cut-point of 19, some of 26 and others of 30 or 35. Clearly this would result in different prevalence rates being calculated. There was more standardisation with regards to cut-points in the later studies. For example all 2008-2012 studies using the IES used a cut-point of >35. When studies were arranged in order of time to follow-up (table not shown) no pattern could be discerned. Of note, in spite of the recommendation of earlier systematic reviews, few longitudinal studies with PTSD assessment at multiple time-points were carried out. Therefore there is still no clarity about the likely trajectory of PTSD in ICU survivors over time. On examination, the most important source of variation for PTSD results was in the quality of studies. Therefore, table 5 showing PTSD prevalence and table 6 showing risk factors for PTSD were ranked in order of quality scores within the two time periods. The likely interpretation of the 2008-2012 data (based on a larger number of high quality studies) is that 9% -15% of ICU survivors would fulfil all diagnostic criteria for PTSD, while 13% to 27% are highly likely to have PTSD or meet most diagnostic criteria. The median prevalence suggests that around one in five critical care patients develop high levels of PTSD symptoms after intensive care. Interventions to reduce PTSD prevalence would clearly be desirable 52 and in the UK, National Institute of Health and Clinical Excellence guidelines recommend that support and rehabilitation be made available to prevent future psychological morbidity.53 However, to design effective preventative interventions, a clear picture of important risk factors is needed.18 Across the 26 studies, there was evidence for and against the importance of socio-demographic risk factors such as younger age and female sex, but fewer recent studies found associations with those factors (Table VI). No studies in the earlier, and very few in the later period, investigated the role of ethnicity or socio-economic cir-
MINERVA ANESTESIOLOGICA
August 2013
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
cumstances in post-ICU PTSD.6, 38, 39 The importance of previous psychiatric history remains uncertain as very few 2008-12 studies tested it as a risk factor, although the earlier systematic reviews had strongly recommended that it should be included in future studies. Similar to 1997-2007 studies, the 2008-2012 studies found that clinical factors such as diagnostic group, illness severity scores such as the Apache II and length of stay in the ICU were not risk factors for PTSD. Of note, one 2012 study 6 found associations with both TISS score and number of organs supported and PTSD. These variables may be a better reflection of illness severity during an admission than Apache II scores measured within 24 hours of admission. In fact TISS score is a measure of number and type of interventions received; it is possible therefore that increasing, more invasive interventions rather than illness severity are a risk factor for PTSD. Other clinical factors such as sedation and mechanical ventilation continued to be investigated (Table VI), as recommended by earlier reviews. Between 1997 and 2012, 7 studies investigated a link between sedation and PTSD. Between 1997-2007 3 high-quality studies found positive associations between PTSD and lorazepam dose,26 administration of midazolam,7 and duration of sedation and opiates;30 while one lowquality, small study (N.=37) found no association with duration of sedation.33 One of these studies (also small, N.=43) 26 also found no positive association between other sedatives (apart from lorazepam) and PTSD. From 2008-2012, only one high-quality study found positive associations with both duration of sedation and administration of benzodiazepines, and PTSD.6 However another high-quality study found no association between depth of sedation (heavy vs light) and PTSD 41 while a medium-quality study 45 did not find an association with sedation intensity score (average sedative exposure per hour) and PTSD. Although evidence is not definitive, these results suggest that receiving benzodiazepines for sedation, or being sedated for longer, are risk factors for PTSD. However the depth of sedation or average amount of sedative drug received in a given time do not appear to increase the risk.
Vol. 79 - No. 8
WADE
Exactly why sedation factors should be linked to longer-term PTSD is not yet obvious. Certainly there has long been concern, and indeed controversy, about reports that extended periods of benzodiazepine usage in the community, particularly among older patients, may be linked to poor long-term neuropsychiatric outcomes including cognitive decline, psychotic symptoms, anxiety and depression.54, 55 These effects have been attributed to benzodiazepine-associated brain damage,56 or deficiencies in serotonin, noradrenaline and dopamine caused by benzodiazepines.54 Other studies have found that benzodiazepine use in the ICU is associated with a greater likelihood of delirium.30, 57 This is possibly due to the effects of these drugs on levels of neurotransmitters such as dopamine or acetylcholine, particularly in critically ill patients with deranged physiology, whose ability to excrete drugs may also be impaired. It may be that patients suffering delirium, especially if hallucinations and paranoid delusions are among their delirious symptoms, are more likely to develop PTSD. But although a number of studies have shown that patients who have memories of paranoid delusions shortly after leaving the ICU are more likely to develop PTSD, an association between acute delirious symptoms and PTSD was only found in one of the 26 studies reviewed here.6 Another likely explanatory mechanism may be that benzodiazepines are known to cause amnesia as well as sedation.58 Patients who have long periods of amnesia for real events that occurred in the ICU, while remembering terrifying delusional memories, may be more prone to develop PTSD. While these explanations remain in the realms of hypothesis, the clinical implications of the studies that measured sedation as a risk factor, are that reducing the use of benzodiazepines and length of time a patient is sedated where possible, could reduce their risk of ICU-related PTSD in future. Use of mechanical ventilation and duration of MV were found to be risk factors in 3 out of 4 later studies (2008-2012), but no associations with days of ventilation were found in 3 out of 4 earlier studies (1997-2007). It is known that mechanical ventilation, particularly if prolonged, can be a highly stressful, invasive and
MINERVA ANESTESIOLOGICA
957
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA WADE
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
Table VI.—Risk factors for post-ICU PTSD. 1stauthor/year
Statistical analysis of risk factors
1997-2007 studies in order of quality rating Multivariable Girard 2007 26 logistic regression
Age/other sociodemographics
Sex
Severity of illness
Age Y (younger people at greater risk)
Y (females at greater risk)
N
Jones 2007 27
Structural equation modelling
Samuelson 2007 7
Multivariable logistic regression
Age Y younger
Y female
N
Cuthbertson 2004 25
Age Y younger
N
N
Age N
N
Nickel 2004 31
Univariable (Spearman’s correlations) Multivariable linear regression T-test, ANOVA Mann Whitney test
Richter 2006 33
T-test, Mann Whitney test
Age N
N
Age N
N
Scragg 2001 34 Jones 2001 28
Jones 2003 29
N
ANOVA
Sukantarat 200735 Univariable (Spearman’s correlations) 2008-2012 studies in order of quality Multivariable logistic regression Myhren 2010 39
Treggiari 2009 41 Wade 2012 6
T-test Multivariable linear regression
Wallen 2008 44
Multivariable logistic regression
Garrouste-Orgeas, 2012 20 Granja 2008 36
Kruskal-Wallis, chi square Multivariable logistic regression
Myhren 2009 38 (4-6 weeks)
Multivariable linear regression
Weinert 2008 45 (2m and 6m)
T-test, ANOVA, correlations
Age N Education level, low Y Unemployed Y
N
N
Age N Socio-economic N Ethnicity N
N
N
Age Y (younger) Age N
N
Y
Y female N
N N
N
N
Age N
Age Y (older) Unemployed Y
Y: association found, N: no association found; MV: mechanical ventilation Risk factor analysis rating: 2= lowest risk of bias. 1= moderate risk of bias 0: highest risk of bias * Among psychological factors, FMs= factual memories, DMs: delusional/delirious memories, IMs: intrusive memories
958
MINERVA ANESTESIOLOGICA
August 2013
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
Prior psychological history
Y
Y
N Y ?unclear N ?unclear
Clinical/other factors Total lorazepam dose Y Midazolam N Propofol N Days of MV N Prolonged sedation Y Physical restraint Y Use of midazolam Y Days of MV N
Days MV Y Time since discharge N
Duration MV N Duration of sedation N Time to follow-up N
Time of outcome assessment N
Y
N
Light vs deep sedation N Days of sedation Y Benzodiazepines Y Inotropes/ vasopressors Y Antipsychotics Y TISS score Y Number organs Y Days MV Y Days CV support Y Number drug groups Y Receiving endotracheal MV Y
MV received Y Duration of MV N
Wakefulness in MV N (non-linear; lower PTSD in most and least awake) Degree of sedation N
Vol. 79 - No. 8
Psychological/ neuro-cognitive factors*
Diagnosis groups
Days of delirium in ICU N Measured at 6m: Traumatic memories Y Measured at 1-2 weeks: DMs Y Agitation in ICU Y Measured at 5 days: ICU fear Y ICU stress Y DMs N Amnesia N
N
Measured at 2 weeks: DMs Y
N
N
N
WADE
LoS in ICU
Risk factor analysis rating
N
2
N
2 2
N
1
N
0 0 0
N
Measured at 2 weeks: DMs Y
0 0 0
Measured at 4-6w: Pessimism Y Pain memories Y Lack of control Y FMs Y DMs N
N
N
2
Measured in ICU: Mood Y IM Y Illness perception Y Stress Y Delirium Y Amnesia in ICU Y
N
N
2 2
N
N
2
N
N
1
N
N
1
N
N
1
Measured at 3m: Peritraumatic dissociation N Measured at 6m: DMs Y Panic Y IM Y Amnesia for pre-ICU hospital stay Y Measured at 4-6w: Pessimism Y Pain memories Y Lack of control Y Inability express needs Y DMs Y FMs Y Measured at 2m: ICU amnesia N DMs Y
MINERVA ANESTESIOLOGICA
1
959
WADE
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
Y (3) N (2) Y (2) N (2) Y (5) N (4)
Y (1) N (3) Y (3) N (1) Y (4) N (4)
Other psychological factors ‡
Y (0) N (5) Y (0) N (6) Y (0) N (11)
Types of memory and delirious symptoms
Y (0) N (6) Y (1) N (5) Y (1) N (11)
MV
Y (0) N (4) Y (0) N (6) Y (0) N (10)
Sedation
Y (4) N (2) Y (1) N (1) Y (5) N (3)
Days in ICU
Y (0) N (0) Y (3) N (2) Y (3) N (2)
Illness severity
Y (2) N (2) Y (1) N (5) Y (3) N (7)
Psychiatric history
Y (3) N (3) Y (2) N (4) Y (5) N (7)
Diagnosis in ICU
1997-2007 (n=10) 2008-2012 (n=8) Total
Sex
Studies
Socio-economic position or ethnicity
Table VII.—Summary table of post-ICU PTSD risk factors.
Age
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA
Y (5)* N (3) Y (12) † N (2) Y (17) N (5)
Y (2) N (0) Y (9) N (1) Y (11) ‡ N (1)
Y: significant effect found for factor (x) number of times N: no significant effect found for factor (x) number of times *traumatic memories, delirious memories or amnesia †pain memories, factual memories, delusional memories, intrusive memories, delirium, amnesia in ICU and pre-ICU ‡Stress, agitation, fear, panic, mood, loss of control, inability to express needs in ICU. Also pessimism, peritraumatic dissociation and illness perceptions
terrifying procedure. It is associated with lifethreatening illness such as respiratory failure and the inability to breathe independently, factors that would be expected to cause anxiety. Therefore it would not be unusual if the experience of MV were to trigger an anxiety disorder such as PTSD. Nevertheless in three of the four studies which detected the association, it was found to have a small effect size in a univariable analysis. This suggests that MV is not a strong risk factor for PTSD, and its effect may be explained by the related factor of sedation, as the two clinical factors tend to occur together. Judged by the number of associations found across studies,28 psychological reactions to ICU would be the most important group of risk factors for PTSD. However there are problems with interpreting these associations. From 19972007, a number of studies showed that traumatic or delusional memories were associated with PTSD. These were measured at times between 2 weeks and 6 months after ICU, and could therefore be symptoms of disorder (acute stress disorder or PTSD) rather than risk factors for a disorder. The picture is even less clear in the 2008 to 2012 studies, as many types of memory (delusional/delirious memories, factual memories, pain memories, intrusive memories, ICU amnesia, pre-ICU amnesia) were investigated at different time points in one or two studies each.
960
Emotional reactions measured in or immediately post-ICU were found to predict PTSD in two high quality studies, one from each time period.6, 7 In one study acute emotional reactions confounded most other risk factors.6 Loss of psychological control and the personality trait of pessimism were also associated with PTSD in two later high quality studies of the same cohort of patients.38, 39 It is likely that acute psychological or neuro-cognitive factors such as the impact of delusions and delirium, and emotional stress in the ICU, are important risk factors for PTSD, but still more research is needed to confirm these findings The experience of hallucinations and delusions, along with the threat to life from critical illness, and the stressful and invasive procedures it necessitates, is likely to lead to extreme reactions such as anxiety, panic attacks and low mood in critical care patients. Without psychological support to deal with these symptoms, the patient may continue to be traumatized after transfer to a medical or surgical ward. As relatives may also suffer from PTSD,13 there may be conflict within families after ICU, causing further stress during a patient’s recovery period at home. These experiences may predispose patients to develop mental health disorders such as PTSD after the ICU. Strengths of this review are that it was based on a prespecified protocol and used systematic and
MINERVA ANESTESIOLOGICA
August 2013
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
explicit methods to identify, select and appraise studies. Assessment of risk of bias (quality assessment) was carried out. No studies of diagnostic sub-groups of ICU patients were included, and therefore estimates of prevalence were not inflated by patient groups with exceptionally high rates. The review included studies of mechanically ventilated ICU patients. As ventilated patients have many underlying conditions, they are broadly representative of the ICU population. A limitation of the review was that no statistical aggregation of results was possible due to heterogeneity of methods and results. Therefore we have presented medians and ranges of estimates of PTSD. We could not carry out a meta-analysis, and therefore could not test quantitatively for publication bias, which is a common limitation of systematic reviews. We did not include unpublished results or results published only in abstracts or conference papers. This review also carries the risk of bias due to possible selective reporting of associations within studies. It is likely there was under-reporting of null associations. Conclusions Evidence from this systematic review suggests that at least one in five patients may develop PTSD after intensive care. The most consistent risk factors identified were clinical (benzodiazepine use and duration of sedation) and acute psychological (relating to stress, delirium and memory problems associated with ICU). Although more studies and higher quality studies have been published since 2008 than before, the evidence base for prevalence and risk factors for PTSD after ICU can still not be considered definitive. In the meantime, clinical implications of the evidence to date are that the use of benzodiazepines and duration of sedation should be limited if clinically feasible, and psychological support should be provided both during and following ICU admissions to mitigate the effect of delusions, delirium and emotional stress in ICU, and reduce longer-term PTSD. Additionally, there is limited evidence that reducing the amount of invasive ICU medical interventions, if clinically feasible, might also reduce the traumatic stress suffered by ICU patients.
Vol. 79 - No. 8
WADE
Key messages —— Up to 27% of ICU survivors suffer from PTSD up to a year after leaving the unit, with likely serious consequences for their health and well-being. —— The most important known clinical risk factors for post-ICU PTSD are use of benzodiazepines and duration of sedation in the ICU. —— Extreme stress reactions, delirious symptoms and memory problems in the ICU are risk factors for PTSD, but have not been investigated in a consistent way. —— The quality and quantity of evidence about prevalence of and risk factors for postICU PTSD have improved over time, but cannot yet be considered definitive.
References 1. Desai SV, Law TJ, Needham DM. Long-term complications of critical care. Crit Care Med 2011;39:371-9. 2. Perrins J, King N, Collings J. Assessment of long-term psychological well-being following intensive care. Intensive Crit Care Nurs 1998;14:108-16. 3. Schelling G, Stoll C, Haller M, Briegel J, Manert W, Hummel T et al. Health-related quality of life and posttraumatic stress disorder in survivors of the acute respiratory distress syndrome. Crit Care Med 1998;26:651-9. 4. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders. 4th edition. Washington DC: Author; 1994. 5. Dedert EA, Calhoun PS, Watkins LL, Sherwood A, Beckham JC. Posttraumatic stress disorder, cardiovascular, and metabolic disease: a review of the evidence. Ann Behav Med 2010;39:61-78. 6. Wade DM, Howell DC, Weinman JA, Hardy RJ, Mythen MG, Brewin CR et al. Investigating risk factors for psychological morbidity three months after intensive care, a prospective cohort study. Crit Care 2012;16:R192. 7. Samuelson KA, Lundberg D, Fridlund B. Stressful memories and psychological distress in adult mechanically ventilated intensive care patients - a 2-month follow-up study. Acta Anaesthesiol Scand 2007;51:671-8. 8. Ely EW, Siegel MD, Inouye SK. Delirium in the intensive care unit: an under-recognized syndrome of organ dysfunction. Semin Respir Crit Care Med 2001;22:115-26. 9. Jackson JC, Hart RP, Gordon SM, Hopkins RO, Girard TD, Ely EW. Post-traumatic stress disorder and post-traumatic stress symptoms following critical illness in medical intensive care unit patients: assessing the magnitude of the problem. Crit Care 2007;11:R27. 10. Griffiths J, Fortune G, Barber V, Young JD. The prevalence of post traumatic stress disorder in survivors of ICU treatment: a systematic review. Intens Care Med 2007;33:150618. 11. Davydow DS, Gifford JM, Desai SV, Needham DM, Bienvenu OJ. Posttraumatic stress disorder in general intensive
MINERVA ANESTESIOLOGICA
961
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA WADE
IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
care unit survivors: a systematic review. Gen Hosp Psych 2008;30:421-34. 12. MikkelsenME, Christie JD, Lanken PN, Biester RC, Thompson BT, Bellamy SL et al. The adult respiratory distress syndrome cognitive outcomes study. Am. J Respir Crit Care Med 2012;185:1307-15. 13. Davidson JE, Jones C, Bienvenu OJ. Family response to critical illness: Postintensive care syndrome-family. Crit Care Med 2012;40:618-24. 14. Mealer M, Jones J, Newman J, McFann KK, Rothbaum B, Moss M. The presence of resilience is associated with a healthier psychological profile in intensive care unit (ICU) nurses: results of a national survey. Int J Nurs Stud 2012;49:292-9. 15. Egerod I, Storli SL, Akerman E. Intensive care patient diaries in Scandinavia: a comparative study of emergence and evolution. Nurs Inq 2011;18:235-46. 16. Jones C, Backman C, Capuzzo M, Egerod I, Flaatten H, Granja C et al. Intensive care diaries reduce new onset post traumatic stress disorder following critical illness: a randomised, controlled trial. Crit Care 2010;14:R168. 17. Peris A, Bonizzoli M, Iozzelli D, Migliaccio ML, Zagli G, Bacchereti A et al. Early intra-intensive care unit psychological intervention promotes recovery from post traumatic stress disorders, anxiety and depression symptoms in critically ill patients. Crit Care 2011;15:R41. 18. Michie S, Abraham C, Eccles MP, Francis JJ, Hardeman W, Johnston M. Strengthening evaluation and implementation by specifying components of behaviour change interventions: a study protocol. Implement Sci 2011;6:10. 19. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 2009;62:1006-12. 20. Garrouste-Orgeas M, Coquet I, Perier A, Timsit JF, Pochard F, Lancrin F et al. Impact of an intensive care unit diary on psychological distress in patients and relatives. Crit Care Med 2012;40:2033-40. 21. SIGN. SIGN 50: A guideline developers handbook. Edinburgh: Scottish Intercollegiate Guidelines Network;2004. 22. Weiss DS, Marmar CR. The impact of event scale-revised. In: Wilson JP & Keane TM (editors). Assessing psychological trauma and PTSD. New York: Guilford Press;1997. p. 399-411 23. Horowitz M, Wilner N, Alvarez W. Impact of Event Scale: a measure of subjective stress. Psychosom Med 1979;41:20918. 24. Capuzzo M, Valpondi V, Cingolani E, Gianstefani G, De LS, Grassi L et al. Post-traumatic stress disorder-related symptoms after intensive care. Minerva Anestesiol 2005;71:167-79. 25. Cuthbertson BH, Hull A, Strachan M, Scott J. Post-traumatic stress disorder after critical illness requiring general intensive care. Intens Care Med 2004;30:450-5. 26. Girard TD, Shintani AK, Jackson JC, Gordon SM, Pun BT, Henderson MS et al. Risk factors for post-traumatic stress disorder symptoms following critical illness requiring mechanical ventilation: A prospective cohort study. Crit Care 2007;11:R28 27. Griffiths J, Gager M, Alder N, Fawcett D, Waldmann C, Quinlan J. A self-report-based study of the incidence and associations of sexual dysfunction in survivors of intensive care treatment. Intens Care Med 2006;32:445-51. 28. Jones C, Griffiths RD, Humphris G, Skirrow PM. Memory, delusions, and the development of acute posttraumatic stress disorder-related symptoms after intensive care. Crit Care Med 2001;29:573-80. 29. Jones C, Skirrow P, Griffiths RD, Humphris GH, Ingleby S, Eddleston J et al. Rehabilitation after critical illness: a randomized, controlled trial. Crit Care Med 2003;31:2456-61. 30. Jones C, Backman C, Capuzzo M, Flaatten H, Rylander C, Griffiths RD. Precipitants of post-traumatic stress disorder
962
following intensive care: a hypothesis generating study of diversity in care. Intens Care Med 2007;33:978-85. 31. Nickel M, Leiberich P, Nickel C, Tritt K, Mitterlehner F, Rother W et al. The occurrence of posttraumatic stress disorder in patients following intensive care treatment: a crosssectional study in a random sample. J Intens Care Med 2004;19:285-90. 32. Rattray JE, Johnston M, W. Predictors of emotional outcomes of intensive care. Anaesthesia 2005;60:1085-92. 33. Richter JC, Waydhas C, Pajonk FG. Incidence of posttraumatic stress disorder after prolonged surgical intensive care unit treatment. Psychosomatics 2006;47:223-30. 34. Scragg P, Jones A, Fauvel N. Psychological problems following ICU treatment. Anaesthesia 2001;56:9-14. 35. Sukantarat K, Greer S, Brett S, Williamson R. Physical and psychological sequelae of critical illness. B J Health Psych 2007;12:1-74. 36. Granja C, Gomes E, Amaro A, Ribeiro O, Jones C, Carneiro A et al. Understanding posttraumatic stress disorderrelated symptoms after critical care: the early illness amnesia hypothesis. Crit Care Med 2008;36:2801-9. 37. Jackson JC, Girard TD, Gordon SM, Thompson JL, Shintani AK, Thomason JW et al. Long-term cognitive and psychological outcomes in the awakening and breathing controlled trial. Am J Resp Crit Care Med 2010;182:183-91. 38. Myhren H, Toien K, Ekeberg O, Karlsson S, Sandvik L, Stokland O. Patients’ memory and psychological distress after ICU stay compared with expectations of the relatives. Intens Care Med 2009;35:2078-86. 39. Myhren H, Ekeberg O, Toien K, Karlsson S, Stokland O. Posttraumatic stress, anxiety and depression symptoms in patients during the first year post intensive care unit discharge. Crit Care 2010;14:R14. 40. Rattray J, Crocker C, Jones M, Connaghan J. Patients’ perceptions of and emotional outcome after intensive care: results from a multicentre study. Nurs in Crit Care 2010;15:86-93. 41. Treggiari MM. Randomized trial of light versus deep sedation on mental health after critical illness. Crit Care Med 2009;37:2526-34. 42. Twigg E, Humphris G, Jones C, Bramwell R, Griffiths RD. Use of a screening questionnaire for post-traumatic stress disorder (PTSD) on a sample of UK ICU patients. Acta Anaesthesiol Scand 2008;52:202-8. 43. Van der Schaaf M, Beelen A, Dongelmans DA, Vroom MB, Nollet F. Functional status after intensive care: a challenge for rehabilitation professionals to improve outcome. J Rehab Med: 2009;41: 360-366. 44. Wallen K, Chaboyer W, Thalib L, Creedy DK. Symptoms of acute posttraumatic stress disorder after intensive care. Am J Crit Care 2008;17:534-44. 45. Weinert CR, Sprenkle M. Post-ICU consequences of patient wakefulness and sedative exposure during mechanical ventilation. Intens Care Med 2008;34:82-90. 46. Foa EB, Cashman L, Jaycox L, Perry, K: The validation of a self-report measure of posttraumatic stress disorder: The Posttraumatic Diagnostic Scale. Psychol Assessment 1997;9:445-51. 47. Davidson JR, Book SW, Colket JT, Tupler LA, Roth S, David D et al. Assessment of a new self-rating scale for post-traumatic stress disorder. Psychol Med, 1997;27:153- 60. 48. Blanchard EB, Jones-Alexander J, Buckley TC, Forneris CA. Psychometric properties of the PTSD checklist (PCL). Behav Res Ther 1996;34:669-73. 49. Stoll C, Kapfhammer HP, Rothenhausler HB, Haller M, Briegel J, Schmidt M et al. Sensitivity and specificity of a screening test to document traumatic experiences and to diagnose post-traumatic stress disorder in ARDS patients after intensive care treatment. Intens Care Med 1999;25:697704.
MINERVA ANESTESIOLOGICA
August 2013
This document is protected by international copyright laws. No additional reproduction is authorized. It is permitted for personal use to download and save only one file and print only one copy of this Article. It is not permitted to make additional copies (either sporadically or systematically, either printed or electronic) of the Article for any purpose. It is not permitted to distribute the electronic copy of the article through online internet and/or intranet file sharing systems, electronic mailing or any other means which may allow access to the Article. The use of all or any part of the Article for any Commercial Use is not permitted. The creation of derivative works from the Article is not permitted. The production of reprints for personal or commercial use is not permitted. It is not permitted to remove, cover, overlay, obscure, block, or change any copyright notices or terms of use which the Publisher may post on the Article. It is not permitted to frame or use framing techniques to enclose any trademark, logo, or other proprietary information of the Publisher.
©
COPYRIGHT 2013 EDIZIONI MINERVA MEDICA IDENTIFYING CLINICAL AND ACUTE PSYCHOLOGICAL RISK FACTORS FOR PTSD AFTER CRITICAL CARE
50. Brewin CR, Fuchkan N, Huntley Z, Scragg P. Diagnostic accuracy of the Trauma Screening Questionnaire after the 2005 London bombings. J Trauma Stress 2010;23:393- 8. 51. Spitzer RL, Williams JB, Gibbon M, First MB. The Structured Clinical Interview for DSM-III-R (SCID). I: History, rationale, and description. Arch Gen Psychiatry 1992;49:624-9. 52. Hatch R, McKechnie S, Griffiths J. Psychological intervention to prevent ICU-related PTSD: who, when and for how long? Crit Care 2011;15:141. 53. Tan T, Brett SJ, Stokes T. Rehabilitation after critical illness: summary of NICE guidance. BMJ 2009;338:b822. 54. Longo LP, Johnson B. Addiction: Part I. Benzodiazepines--
55. 56. 57.
58.
WADE
side effects, abuse risk and alternatives. Am Fam Physician 2001;61:2121-8. Patten SB. Still no evidence that benzodiazepines cause depression. Int J Psychiatr in Clin Prac 2008;12:85-8. Golombok S, Moodley P, Lader M. Cognitive impairment in long-term benzodiazepine users. Psychol Med 1988;18:365-74. Pandharipande P, Shintani A, Peterson J, Pun BT, Wilkinson GR, Dittus RS et al. Lorazepam is an independent risk factor for transitioning to delirium in intensive care unit patients. Anesthesiology 2006;104:21-6. Ghoneim MM. Drugs and human memory (part 1): Clinical, theoretical, and methodologic issues. Anesthesiology 2004;100:987-1002.
Acknowledgments.—We are very grateful for the contribution of Elijah Rhone (ER), who carried out data extraction of the 2008-12 studies, and Kate Cheney the UCL librarian who advised on the search strategy. We would also like to thank Rosalind Raine and John Weinman for their support in supervising the early stages of work on this systematic review. Funding.—DW is funded by, and DH and MM receive a portion of their funding from, the UCLH/UCL National Institute of Health Research Biomedical Research Centre. Rebecca Hardy is supported by the MRC. There are no conflicts of interest. Received on October 8, 2012 – Accepted for publication on March 25, 2013. Corresponding author: Dr. D. Wade, Health Psychologist, The Critical Care Unit, UCLH NHS Foundation Trust, 235 Euston Rd, London NW1 2BU, UK. E-mail: Dorothy.
[email protected]
Vol. 79 - No. 8
MINERVA ANESTESIOLOGICA
963