Journal of Critical Care 30 (2015) 1049–1054
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Symptom burden and health-related quality of life among intensive care unit survivors in Argentina: A prospective cohort study☆,☆☆,★ Andrea V. Das Neves, MD a, Daniela N. Vasquez, MD b, Cecilia I. Loudet, MD a, Dante Intile, MD b, María Gabriela Sáenz, MD a, Cecilia Marchena, MD a, Ana L. Gonzalez, MD a, Joaquin Moreira, MD c, Rosa Reina, MD a, Elisa Estenssoro, MD a,⁎ a b c
Servicio de Terapia Intensiva, Hospital Interzonal General de Agudos San Martin de La Plata 1 y 70, 1900 La Plata, Buenos Aires, Argentina Servicio de Terapia Intensiva, Sanatorio Anchorena, Tomas de Anchorena, C1425ELP, 1872 Ciudad Autónoma de Buenos Aires, Argentina Instituto del Diagnostico, 62 n° 370, 1900 La Plata, Buenos Aires, Argentina
a r t i c l e
i n f o
Keywords: HRQOL Critical care EuroQol-5D Mechanical ventilation duration Weakness Traumatic brain injury sequelae
a b s t r a c t Purpose: Our goal was to describe the evolution of selected physical and psychologic symptoms and identify the determinants of health-related quality of life (HRQOL) after intensive care unit (ICU) discharge. Methods: The study is a prospective cohort of consecutive adult patients admitted to a mixed ICU in a universityaffiliated hospital, mechanically ventilated for more than 48 hours. During ICU stay, epidemiological data and events probably associated to worsening outcomes were recorded. After discharge, patients were interviewed at 1, 3, 6, and 12 months. Health-related quality of life was assessed with EuroQoL Questionnaire–5 Dimensions, which includes the EQ-index and EQ–Visual Analogue Scale. Results: One hundred twelve patients were followed up, aged 33 [24-49] years, 68% male, 76% previously healthy, and cranial trauma was the main diagnosis. Physical and psychologic symptoms and moderate/severe problems according to the EQ index progressively decreased after discharge, yet were still highly prevalent after 1 year. EQ index improved from 0.22 [0.01-0.69] to 0.52 [0.08-0.81], 0.66 [0.17-0.79], and 0.68 [0.26-0.86] (P b .001, for all vs month 1). EQ–Visual Analogue Scale remained stable, within acceptable values. Independent determinants of EQ-index were time, duration of mechanical ventilation, shock, weakness, and return to study/work. Conclusions: Determinants of HRQOL after ICU discharge were both related to late sequelae of critical illness and to some events occurring in the ICU. Notwithstanding the high symptom burden, patients still perceived their HRQOL as good. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Population aging, technological advances, and awareness of the benefits of critical care medicine have produced a yearly escalation of patients admitted to the intensive care unit (ICU) and surviving the experience [1]. Nevertheless, after discharge, patients and their families face a long, ☆ The authors have no conflict of interest to declare. ☆☆ Disclosures: Some results from this study were presented in abstract format in the following 2 international conferences: -
-
Das Neves AV, Vasquez DN, Intile D, Moreira J, Loudet C, Sáenz MG, et al: Quality of life in critical illness survivors in Argentina. Evaluation with Euroqol 5D: a prospective cohort study. Am J Respir Crit Care Med 2012;185:A2288. Das Neves AV, Vasquez DN, Loudet C, Moreira J, Intile D, Sáenz MG, et al: Lack of awareness of severity of illness is a common factor after ICU. Intensive Care Med 2013;39: Suppl2;170. A663.
★ This work was supported in part by the Grant Carrillo-Oñativia 2010, National Committee Research in Health, Argentine National Ministry of Health, Ciudad Autónoma de Buenos Aires, Argentina. ⁎ Corresponding author. Calle 42 N° 577, 1900 La Plata, Buenos Aires, Argentina. Tel.: +54 2214220507. E-mail address: [email protected] (E. Estenssoro). http://dx.doi.org/10.1016/j.jcrc.2015.05.021 0883-9441/© 2015 Elsevier Inc. All rights reserved.
difficult time during which the physical, psychologic, and neurocognitive sequels become evident and greatly affect their quality of life. These adverse consequences of critical care were recently described as the post-intensive care syndrome [2], conditions that also involve nonmedical issues, such as economic and employment instability [3-6]. The impact of the burden of these diseases on health-related quality of life (HRQOL) has been studied in patients with acute respiratory distress syndrome (ARDS) [7,8] and in general populations of critically ill patients [9,10]. Information about these problems from Latin America is, however, scarce or nonexistent [11]. On these grounds, our goal was to evaluate the HRQOL and its independent determinants, in patients admitted to a medical-surgical ICU in Argentina at 1, 3, 6, and 12 months after discharge. Secondary objectives were to describe the prevalence and evolution of physical and psychologic problems.
2. Methods This study was conducted in a 14-bed ICU located in a universityaffiliated hospital, between May 1, 2010, and July 1, 2012. We included
1050
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054
consecutive patients 15 years or older who stayed more than 48 hours on mechanical ventilation (MV), except for those who refused to participate, were prisoners, or mentally disabled. Readmissions during the study protocol were considered for follow-up only after the first admission. We recorded age, sex, severity of acute illness and of underlying diseases (Acute Physiology and Chronic Health Evaluation II and Charlson scores, respectively), diagnosis, category of ICU admission (medical, unscheduled/scheduled surgery, and trauma), organ failures (SOFA [Sequential Organ Failure Assessment] score), admission Glasgow score, hospital mortality, duration of MV and ICU, and hospital length of stay. We also registered events and conditions potentially associated to worsening long-term outcomes, such as ARDS, shock (defined as vasopressor requirement), acute kidney injury (defined as the duplication of serum creatinine or a urine output b0.5 mL/kg per hour during 12 hours) [12], Glasgow score, requirement of decompressive craniectomy and, at discharge, Glasgow score, anemia, tracheostomy, jejunostomy/gastrostomy, dysphagia (defined as any abnormality in swallowing as detected by “video deglutition” examination, which was performed in all patients) and “urinary incontinence” (defined as the unintentional loss of urine). Follow-up protocol: before ICU discharge, patients or next of kin signed written consent forms agreeing to participate in the study, which consisted of 4 face-to-face or telephone interviews with either the patients or a relative, 1, 3, 6, and 12 months after hospital discharge. At 1 month, they were invited to a personal interview at a specially designed ICU office. Should the patient not come, he/she was subsequently contacted by up to 3 phone calls. In the latter case, we conducted telephone interviews with the patient or, if unavailable/unable to respond, with the next of kin. The procedure was repeated at each of the prescheduled visits. All interviews were performed by the same 3 researchers (ADN, DNV, and MGS). At each interview, patients were asked about symptoms defined as relevant by the Guidelines of Rehabilitation after Critical Illness of the National Institute for Health and Clinical Excellence (United Kingdom) [6], also if they were receiving any kind of rehabilitation, and if they had returned to their previous activities. A clinical examination was conducted. In addition, the following questionnaires were completed:
values at the different time points were compared with month 1, with t test for paired measurements or Wilcoxon signed rank test for continuous data and McNemar test for categorical data. A P value b .05 was considered as significant. We performed a longitudinal analysis to evaluate the effect of time in the EQ-index, our main outcome measure for HRQOL, and to adjust for important covariates. We fitted our data to a marginal model by means of generalized estimating equations because EQindex exhibited a nonparametric distribution [18]. Generalized estimating equation method was used to account for correlations between repeated measurements within subjects over time. An unstructured correlation matrix was selected. Variables that presented a P value b .10 were entered into a multivariate analysis to generate a best model with predictive factors. Sample size was estimated according to EQ-index, the main outcome measure, for observational studies. We selected a mean value of 0.59, which represents 2 SD below the expected value of 0.91 ± 0.16, for subjects between ages 35 and 44 years [14,19], including the mean of value of patients' age in the ICU (43 ± 17 years) [20]. For a desired total width of the interval of 0.06, we calculated the standardized width for a 95% confidence level, which determined a sample size of 97 subjects [21]. We anticipated a 15% of dropouts, so the sample size was increased to 112 patients. All analyses were performed with STATA 11 software (Stata Corporation, College Station, TX). The Institutional Review Board of the Hospital San Martin de La Plata approved this protocol (no. 20091130TI). 3. Results Of the 208 eligible patients (MV N 48 hours), 86 died in the ICU, 10 fulfilled exclusion criteria, 112 were followed up over time, and 57 (51%) finished the study (Fig. 1). In the 249 interviews, 68% (169/249) of the questionnaires were answered face to face, 22% (56/248) by
- Health-related quality of life was assessed with the EuroQoL Questionnaire–5 Dimensions (EuroQol-5D) Quality of Life [13], a standardized instrument that measures health status in 5 dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. For each of these conditions, the possible answers are level 1, no problems; level 2, some/moderate problems; and level 3, severe/extreme problems [14]. The different weights of the 5 dimensions are summarized in the EQ-index as an estimated number from − 0.005 to 1, which reflects the global health state. Negative values describe a health status perceived as worse than death. The EQ-index was considered the main outcome variable. - The EuroQol-5D also includes the EQ–Visual Analogue Scale (EQ-VAS), a standard vertical 20-cm visual analogue scale similar to a thermometer, in which patients record their global health status [14]. The end points are labeled “best imaginable health state” and “worst imaginable health state (100-0 points, respectively).” This information is considered a quantitative, self-reported measure of health. We used the Argentinian validation of the EuroQol-5D [13]. - Davidson Trauma Scale, which assesses the risk of posttraumatic stress disorder (PTSD). A cutoff point greater than or equal to 40 suggests its diagnosis [15,16]. To further evaluate physical functioning, a 6-minute walk test was performed [17]. Statistical analysis: descriptive statistics used were means ± SD, medians and interquartile ranges (IQR, 0.25-0.75) for continuous variables, and percentages for categorical data. After testing for normality,
Fig. 1. Flow-chart of the study.
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054 Table 1 General description and evolution of patients. Data are expressed as n (%), mean ± SD, or median IQR [25%-75%] No. of patients
112
Age (y) Male sex Charlson 0 (no comorbidities) APACHE II SOFA first 24 h Transferred from: Emergency department Ward Operating room Other institution Admission diagnostic categories: Trauma TBI Medical Emergency surgery Elective surgery ARDS Shock AKI Dialysis a Severe TBI Decompressive craniectomyb Duration of MV (d) LOS ICU LOS hospital
33 [24-49] 76/112 (68) 85/112 (76) 15 ± 6 6±3
56/112 (50%) 46/112 (41%) 32/112 (29%) 15/112 (13%) 9/112 (8%) 64/112 (57%) 85/112 (76%) 37/112 (33%) 11/37 (30%) 21/46 (46%) 18/46 (39%) 15 [9-37] 21 [11-43] 47 [28-79]
Characteristics at discharge: GCS at ICU discharge Tracheostomy at ICU discharge Tube feeding/jejunostomy/gastrostomy at ICU discharge Anemia at ICU discharge Transferred to rehabilitation center
14 ± 1 11/112 (10%) 20/112 (18%) 81/105 (77%) 16/112 (14%)
At 1 y Readmission to the ICU Mortality at 1 ya
12/112 (11%) 7/112 (6%)
66/112 (59%) 11/112 (10%) 25/112 (22%) 10/112 (9%)
APACHE indicates Acute Physiology and Chronic Health Evaluation; SOFA, Sequential Organ Failure Assessment; LOS, length of stay; AKI: acute kidney injury; GCS, Glasgow coma score. a Refers to patients with acute kidney injury. b Refers to patients with TBI.
phone interview with the patient, and 10% (24/248) with the proxy; 20 patients could never be contacted. Population characteristics, events, relevant treatments, and clinical state at ICU discharge are shown in Table 1. Briefly, ours was a very
1051
young, previously healthy, predominantly male population, reflecting the most common single diagnosis, which was severe traumatic brain injury (TBI) (46% of patients). Organ failures, prolonged duration of MV, and extended ICU and hospital stays were very frequent. Long-term mortality and readmission rates were low. During follow-up (Table 2), most physical problems displayed a significant trend to decrease, yet the prevalence of weakness, fatigue, and dyspnea was still relevant at 12 months. Neurologic symptoms such as paresis, swallowing difficulties, and urinary incontinence were still frequent 1 year after discharge. Posttraumatic stress disorder showed a nonsignificant tendency to decline. The utilization of physical rehabilitation decreased significantly over time. Fig. 2A illustrates the behavior of the 5 dimensions of EuroQol-5D. Of note, it is clear that the percentage of patients with moderate/severe problems diminishes over time, with the remarkable exception of anxiety and depression. In Fig. 2B, the evolution of the EQ-index is compared with that of EQ-VAS (self-perceived health status). Whereas the EQ-index improved significantly over time, EQ-VAS stayed within acceptable values from the very beginning. This dissociation occurred in all patients, whether they had TBI (Fig. 2C and D, respectively). With respect to the independent determinants of HRQOL, as measured by the EQ-index, it was strongly and negatively affected by the duration of MV, the presence of shock, and of weakness after 1 year, whereas the effect of time and the return to work/study was positively associated with an improved outcome (Table 3). 4. Discussion In this study, we analyze the evolution of a young, mostly previously healthy population, up to 1 year postdischarge from the ICU. Patients exhibited a high, persistent burden of disease that severely affected their HRQOL. Although most physical symptoms showed significant improvement over the first 6 months, they were still prevalent at 1 year. This was the case for weakness, a near-universally reported problem [7,22]. Although weakness, together with continued fatigue and breathlessness, might have some subjective component, these muscular problems were also reflected in the low predicted value for the 6-minute walk test. What is more, 40% of patients still showed moderate-to-severe mobility problems after 1 year as assessed with the EuroQol-5D and in line with the 24% to 58% described in studies using the same tool [23-26].
Table 2 Physical and psychologic consequences Time interval No. of patients Weakness Fatigue Breathlessness Paresis/paralysis Distance walked in 6 min (% of predicted) Swallowing difficulties Urinary incontinence Pressure sores Tracheostomy Tracheal stenosis Physical therapy DTS (N40)d Return to study/worke EQ-index EQ-VAS
1 month
3 months
6 months
12 months
76 53 (70%) 35 (46%) 11 (14%) 33 (45%) 21 (28%) 15 (20%) 57 (75%) 9 (12%) 8 (11%) 38 (50%) 36% 12% 0.22[0.01-0.69] 68[45-91]
59 33 (56%) 24 (41%) 14 (24%) 22 (36%) 68% 12 (20%)a 7 (12%) 22 (37%)a 4 (7%) 7 (12%) 30 (50%) 29% 31%a 0.52[0.08-0.81]c 68[41-95]
56 23 (40%) b 22 (39%) 8 (14%) 22 (39%) 73% 9 (18%) 6 (11%)a 13 (23%) b 4 (7%) 4 (7%) 19 (33%) 13% 26%a 0.66[0.17-0.79]c 73[53-93]
57 24 (41%)a 25 (42%) 12 (20%) 17 (21%)a 72% 10 (17%) 9 (16%) 4 (7%) c 2 (4%) 3 (5%) 15 (27%) c 20% 31%a 0.68[0.26-0.86]c 74[23-95]
Variables were selected in keeping with the Guidelines of Rehabilitation after Critical Illness of the National Institute for Health and Clinical Excellence [7]. a P b .05 vs month 1. b P b .01 vs month 1. c P b .001 vs month 1. d DTS, Davidson Trauma Scale. A cutoff point more than 40 suggests PTSD. e Refers to the 51% of patients (57/112) who either studied or had employment before ICU admission.
1052
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054 Table 3 Determinants of EQ-index EQ-index
Coef.
SE
z
Time .013 .006 2.08 Days of MV −.006 .002 −3.41 Shock −.214 .080 −2.67 Return to study/work at month 12 .290 .103 2.82 Weakness −.224 .107 −2.08
PNz
95% Confidence interval
0.037 .0007 .0243 0.001 −.011 −.0027 0.008 −.371 −.057 0.005 .088 .491 0.038 −.435 −.012
Coef: coefficient. Because prospective analyses involved multiple evaluations of EQ-index and covariates for each subject over time, we used a repeated-measures analysis that incorporated all periods of follow-up data simultaneously.
Fig. 2. Evaluation of the patients with the different EuroQol instruments. A, Evolution of the 5 components of the EuroQol-5D over time. The percentage of patients with moderate and severe problems diminishes over time, and this decrease is significant at 12 months in each domain. In spite of this, approximately 50% of patients still had important problems with performing usual activities or experienced pain and discomfort. On the contrary, anxiety/depression remained stable throughout the study period. ⁎P b .05 vs month 1, ⁎⁎P b .01 vs month 1; and ⁎⁎⁎P b .001 vs month 1. B, Comparison of the evolution of EQindex and EQ-VAS over the study period in the entire population. Although EQ index experienced a significant improvement over the study period with respect to month 1 value, EQ-VAS did not change and remained within normal values for the Argentinian population. A lack of perception of the present disease and sequelae is evident. C and D, The comparison of EQ-index and EQ-VAS between TBI and non-TBI patients. The dissociation between estimated (EQ-index) and self-perceived (EQ-VAS) health-related quality of life during the first 6 months after hospital discharge is evident in both groups. This implies that mechanisms other than structural brain damage might be operative for this distortion. Data are showed as mean and SD. ⁎P b .05 vs month 1. ⁎⁎P b .01 vs month 1 and ⁎⁎⁎P b .001 vs month 1.
Moderate and severe pain and discomfort were the most frequent symptoms after discharge (80%), decreasing to 48%, a still unacceptably high figure. Ability to self-care and to return to usual activities was significantly improved after 1 year, similar to the studies mentioned. Severe neurologic sequelae, such as incontinence and swallowing difficulties, remain unchanged over time. Remarkably, psychologic symptoms were very frequent, a finding shared with other studies [27-29]. Posttraumatic stress disorder prevalence was high; psychiatric consultation was accepted only in 8 patients (38%); this reluctance has been reported [30]. Posttraumatic stress disorder, however, tended to decrease with time, whereas anxiety/depression remained stable throughout the study period and were present in onethird of patients, similar to the reported prevalence for ICU survivors (23%-48%) [31]. Hypoxia, inflammation, hypoglycemia, older age, and benzodiazepine use, all common in the ICU, have been associated to psychologic symptoms in survivors [28,32]; less is known about the impact of pre-ICU symptoms, level of education, and endogenous patient-related factors. With respect with the return to previous activities, at 1 year, only 30% of patients had returned to study/work, which is lower than the 50% generally reported [32]. Possible causes were the low rate of physical rehabilitation delivered or the unavailability of neurocognitive rehabilitation, a poorly developed activity in Argentina. Many of the patients with TBI who had undergone decompressive craniectomy had not yet received cranioplasty after 1 year, which precluded the return to manual labor and workshops, their usual major activity. As regards the independent determinants of HRQOL, as measured by the EQ-index adjusted by time, we, along with other authors [33,34], have found that HRQOL was associated with few intra-ICU events; in our study, only shock and duration of MV were independent predictors. Shock is the inability of cardiovascular system to satisfy oxygen needs; the resultant tissue hypoxia is a leading cause of multiorgan failure and death. Shock has also been associated with prolonged MV and ICU stay [35,36]. The length of MV might be considered a global indicator of severity and persistence of disease, also having other implications such as prolonged use of sedative and analgesics, which might contribute to adverse cognitive outcomes. Mechanical ventilation duration is also related to ICU length of stay [37]. This is the first study that directly relates shock or duration of MV with a long-term (1-year) outcome—although other investigators have described length of ICU and hospital stays as predictors. With respect to post-ICU conditions independently related with HRQOL, persistence of weakness showed a negative association, whereas a psychosocial issue such as ability to return to previous activities after 1 year and the effect of time per se was positively associated. Other patient characteristics like age (probably because our patients were young), sex, or severity of illness on admission did not impact on HRQOL in this cohort. Not unexpectedly, in the months after ICU discharge, the EQ-index showed a significant improvement. Instead, the EQ-VAS, which
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054
measures self-perceived health status, was already high just 1 month after discharge and remained elevated without significant changes over the entire study period, with similar values to those of the general population median of 75.35 (74.94-75.75) [13]. Thus, patients actually felt better than what was evidenced by the more thoroughly detailed EQ-index. Deterioration of EQ-index with preservation of the EQ-VAS has been described in a general ICU population at 1 time point after discharge [25]. In older ICU survivors, a similar situation of altered SF-12 (Short Form-12 questionnaire) physical with preserved mental component scores has been reported [38], but elderly patients usually adapt to physical limitations and perceive their HRQOL as good [10]. The reason for this discordant behavior between EQ-index and EQ-VAS is not clear. Patients might have developed anosognosia, defined as the lack of awareness of different deficits, ascribed to focal brain lesions or to diffuse brain damage [39]. In our population, anosognosia could have occurred not only in patients with brain injury (severe TBI) but in those without this disorder (Fig. 2C and D, respectively). Our patients had recurrent episodes of shock, ARDS, and required prolonged MV, conditions associated with hypoxia, hypotension, inflammation, high need of benzodiazepine infusions to promote patient adaptation to the ventilator, and other possible mechanisms of injury, which might have contributed to generate or aggravate a global cerebral insult. In addition, the development of psychologic mechanisms for coping with disease cannot be discarded. Denial and recalibration of standards for quality of life have both been described in patients with different diseases [40,41]. All these mechanisms might be operative in patients surviving the ICU experience. A limitation of this study is that it was performed in a single center, so results obtained in this particularly young, previously healthy population might not be representative of all ICU survivors. Baseline evaluation of HRQOL was not performed; nevertheless, this is a rather inexact estimation because recall bias might affect patients' or relatives' evaluation of previous HRQOL [28,42]. We did not measure total dose of benzodiazepine required, which might have been related to psychologic sequelae. In addition, during follow-up, some loss of information occurred: rates of attendance were between 68% and 51% at each time point. Most studies, however, exhibit similar rates of nonresponse. In a systematic review of 53 studies of post-ICU HRQOL, 24 had a response rate between 50% and 79% [10]. Reasons for not attending interviews might vary from feeling too well, to developing symptoms of PTSD such as avoidance and memory recall. Other patients might be still too injured to attend or might have died. In our group, 83% could be contacted, yet some never showed up for interviews. Still, this study has significant strengths: the prospective evaluation of consecutive patients and the personal character of the interview in most cases. Health-related quality of life was repeatedly measured over time, which evidenced the pattern of recovery. Its determinants were identified by longitudinal data analysis, accounting for the effect of time. Finally, to our knowledge, this is the first study evaluating HRQOL in post-ICU patients in South America. We found that rehabilitation and psychologic consultation were low, which underscores the need for creating special care programs to address patients' requirements.
5. Conclusions In this group of patients, HRQOL after ICU discharge was negatively affected by ICU events, such as the presence of shock and the length of MV, and also by important sequelae of critical illness as persistent weakness; whereas the ongoing effect of time and the return to previous activities, a surrogate of global improvement and welfare showed a positively association. Most interestingly, despite the high burden of symptoms, patients exhibited a positive perception of their HRQOL.
1053
References [1] Needham DM, Bronskill SE, Calinawan JR, BMath CA, Sibbald WJ, Pronovost PJ, et al. Projected incidence of mechanical ventilation in Ontario to 2026: preparing for the aging baby boomers. Crit Care Med 2005;33:574–9. [2] Needham DM, Davidson J, Cohen H, Hopkins R, Weinert C, Wunsch H, et al. Improving long-term outcomes after discharge from intensive care unit: report from a stakeholders' conference. Crit Care Med 2012;40:502–9. [3] Jones Ch, Skirrow P, Griffiths RD, Humphris G, Ingleby S, Eddleston J, et al. Rehabilitation after critical illness: a randomized, controlled trial. Crit Care Med 2003;31: 2456–61. [4] Garrouste-Orgeas M, Soufir L, Tabah A, Schwebel C, Vesin A, Adrie C, et al. A multifaceted program for improving quality of care in intensive care units: IATROREF study. Crit Care Med 2012;40:468–76. [5] Elliott D, McKinley S, Alison J, Aitken LM, King M, Leslie GD, et al. Health-related quality of life and physical recovery after a critical illness: a multi-centre randomised controlled trial of a home-based physical rehabilitation program. Crit Care 2011;15:R142. [6] Rehabilitation after critical illness. National Institute for Clinical Excellence; 2009[Available at: www.nice.org.uk/CG83. Accessed March 29, 2010]. [7] Herridge MS, Cheung AM, Tansey CM, Matte-Martyn A, Diaz-Granados N, Al-Saidi F, et al. One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med 2003;348:683–93. [8] Dowdy DW, Eid MP, Sedrakyan A, Mendez-Tellez PA, Pronovost PJ, Herridge MS, et al. Quality of life in adult survivors of critical illness: a systematic review of the literature. Intensive Care Med 2005;31:611–20. [9] Cuthbertson BH, Elders A, Hall S, Taylor J, MacLennan G, Mackirdy F, et al. Mortality and quality of life in the five years after severe sepsis. Crit Care 2013; 172:R70. [10] Oeyen SG, Vandijck DM, Benoit DD, Annemans L, Decruyenaere J, et al. Quality of life after intensive care: a systematic review of the literature. Crit Care Med 2010;38: 2386–400. [11] Griffiths J, Hatch RA, Bishop J, Morgan K, Jenkinson C, Cuthbertson B, et al. An exploration of social and economic outcome and associated health-related quality of life after critical illness in general intensive care unit survivors: a 12-month follow-up study. Crit Care 2013;17:R100. [12] Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, et al. RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care 2006;10:R73. [13] Augustovski FA, Irazola VE, Velazquez, Gibbons L, Craig B, et al. Argentine valuation of the EQ-5D health states. Value Health 2009;4:5875–96. [14] The EuroQol Group. EuroQol—a new facility for the measurement of health-related quality of life. Health Policy 1990;16:199–208. [15] Davidson JR, Book SW, Colket JT. Assessment of a new self-report scale for PTSD. Psychol Med 1999;27:1531–60. [16] Bobes-García J, Portilla MP, Bascarán MD, Sáiz PA, Bousoño M, editors. Banco de Instrumentos básicos para la práctica de la psiquiatría clínica. Barcelona: Editorial Ars Medica; 2004. p. 225. [17] ATS statement. Guidelines for the six-minute walk test. Am J Respir Crit Care Med 2002;166:111–7. [18] Fitzmaurice GM, Laird NM, Ware JH, editors. Applied longitudinal analysis. New Jersey, NY: John Wiley & Sons; 2004. p. 19–44 [291-21]. [19] Cox CE, Martinu T, Sathy SJ, Clay A, Chia J, Gray A, et al. Expectations and outcomes of prolonged mechanical ventilation. Crit Care Med 2009;11:2888–94. [20] Estenssoro E, Reina R, Canales HS, Saenz MG, Gonzalez FE, Aprea MM, et al. The distinct clinical profile of chronically critically ill patients: a cohort study. Crit Care 2006;10:R89. [21] Browner WS, Newman TB, Hulley SB. Estimating sample size and power: applications and examples. In: Hulley SB, Cummings SR, Browner WS, editors. Designing clinical research. Philadelphia, PA: Lippincott Williams & Wilkins; 2007. [22] Brower R. Consequences of bed rest. Crit Care Med 2009;37:S422–8. [23] García Lizana F, Peres Bota D, De Cubber M, Vincent JL. Long-term outcome in ICU patients: what about quality of life? Intensive Care Med 2003;29:1286–93. [24] Granja C, Teixeira-Pinto A, Costa-Pereira A. Quality of life after intensive care—evaluation with EQ-5D questionnaire. Intensive Care Med 2002;28: 898–907. [25] Badia X, Diaz-Prieto A, Gorriz MT, Herdman H, Torrado H, Farrero E, et al. Using EuroQol-5D to measure changes in quality of life 12 months after discharge from an intensive care unit. Intensive Care Med 2001;27:1901–7. [26] Linko R, Suojaranta-Ylinen R, Karlsson S, Ruokonen E, Varpula T, Pettilä V, et al. Oneyear mortality, quality of life and predicted life-time cost-utility in critically ill patients with acute respiratory failure. Crit Care 2010;14:R60. [27] Davydow DS, Gifford JM, Desai SV, Sanjay V, Desai, Bienvenu O, et al. Depression in general intensive care unit survivors: a systematic review. Intensive Care Med 2009; 35:796–809. [28] Bienvenu O, Colantuoni E, Mendez-Tellez P, Dinglas V, Shanholtz C, Husain N, et al. Depressive symptoms and impaired physical function after acute lung injury: a 2-year longitudinal study. Am J Respir Crit Care Med 2012;185:517–24. [29] Hopkins R, Key CW, Suchyta MR, Weaver L, Orme J. Risk factors for depression and anxiety in survivors of acute respiratory distress syndrome. Gen Hosp Psychiatry 2010;32:147–55. [30] Cuthbertson BH, Hull A, Strachan M, Scott J. Post-traumatic stress disorder after critical illness requiring general intensive care. Intensive Care Med 2004;30:450–5. [31] Desai S, Law T, Needham DM. Long-term complications of critical care. Crit Care Med 2011;3:371–9.
1054
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054
[32] Myhren H, Ekeberg Ø, Stokland O. Health-related quality of life and return to work after critical illness in general intensive care unit patients: a 1-year follow-up study. Crit Care Med 2010;38:1554–61. [33] Granja C, Amaro A, Dias C, Costa Pereira A. Outcome of ICU survivors: a comprehensive review. The role of patient-reported outcome studies. Acta Anaesthesiol Scand 2012;56:1092–103. [34] Orwelius L, Nordlund A, Nordlund P, Simonsson E, Bäckman C, Samuelsson A, et al. Preexisting disease: the most important factor for health related quality of life long-term after critical illness: a prospective, longitudinal, multicentre trial. Crit Care 2010;14:R67. [35] Estenssoro E, González F, Laffaire E, Canales H, Saenz MG, Reina R, et al. Shock on admission day is the best predictor of prolonged mechanical ventilation in the ICU. Chest 2005;127(2):598–603. [36] Laupland KB, Kirkpatrick AW, Kortbeek JB, Zuege DJ. Long-term mortality outcome associated with prolonged admission to the ICU. Chest 2006;129:954–9.
[37] Hopkins RO, Jackson J. Critical illness long-term neurocognitive function after critical illness. Chest 2006;130:869–78. [38] Vest MT, Murphy TE, Araujo K, Pisani M. Disability in activities of daily living, depression, and quality of life among older medical ICU survivors: a prospective cohort study. Health Qual Life Outcomes 2011;9:9. [39] Jehkonen M, Laihosalo M, Kettunen J. Anosognosia after stroke: assessment, occurrence, subtypes and impact on functional outcome reviewed. Acta Neurol Scand 2006;114:293–306. [40] Rayson D. On denying denial. J Clin Oncol 2013;31:4371–2. [41] Neuman HB, Park J, Fuzesi S, Temple L. Rectal cancer patients' quality of life with a temporary stoma: shifting perspectives. Dis Colon Rectum 2012;55:1117–24. [42] Dinglas VD, Gifford JM, Husain N, Colantuoni E, Needham D. Quality of life before intensive care using EQ-5D: patient versus proxy responses. Crit Care Med 2013;41: 9–14.
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Symptom burden and health-related quality of life among intensive care unit survivors in Argentina: A prospective cohort study☆,☆☆,★ Andrea V. Das Neves, MD a, Daniela N. Vasquez, MD b, Cecilia I. Loudet, MD a, Dante Intile, MD b, María Gabriela Sáenz, MD a, Cecilia Marchena, MD a, Ana L. Gonzalez, MD a, Joaquin Moreira, MD c, Rosa Reina, MD a, Elisa Estenssoro, MD a,⁎ a b c
Servicio de Terapia Intensiva, Hospital Interzonal General de Agudos San Martin de La Plata 1 y 70, 1900 La Plata, Buenos Aires, Argentina Servicio de Terapia Intensiva, Sanatorio Anchorena, Tomas de Anchorena, C1425ELP, 1872 Ciudad Autónoma de Buenos Aires, Argentina Instituto del Diagnostico, 62 n° 370, 1900 La Plata, Buenos Aires, Argentina
a r t i c l e
i n f o
Keywords: HRQOL Critical care EuroQol-5D Mechanical ventilation duration Weakness Traumatic brain injury sequelae
a b s t r a c t Purpose: Our goal was to describe the evolution of selected physical and psychologic symptoms and identify the determinants of health-related quality of life (HRQOL) after intensive care unit (ICU) discharge. Methods: The study is a prospective cohort of consecutive adult patients admitted to a mixed ICU in a universityaffiliated hospital, mechanically ventilated for more than 48 hours. During ICU stay, epidemiological data and events probably associated to worsening outcomes were recorded. After discharge, patients were interviewed at 1, 3, 6, and 12 months. Health-related quality of life was assessed with EuroQoL Questionnaire–5 Dimensions, which includes the EQ-index and EQ–Visual Analogue Scale. Results: One hundred twelve patients were followed up, aged 33 [24-49] years, 68% male, 76% previously healthy, and cranial trauma was the main diagnosis. Physical and psychologic symptoms and moderate/severe problems according to the EQ index progressively decreased after discharge, yet were still highly prevalent after 1 year. EQ index improved from 0.22 [0.01-0.69] to 0.52 [0.08-0.81], 0.66 [0.17-0.79], and 0.68 [0.26-0.86] (P b .001, for all vs month 1). EQ–Visual Analogue Scale remained stable, within acceptable values. Independent determinants of EQ-index were time, duration of mechanical ventilation, shock, weakness, and return to study/work. Conclusions: Determinants of HRQOL after ICU discharge were both related to late sequelae of critical illness and to some events occurring in the ICU. Notwithstanding the high symptom burden, patients still perceived their HRQOL as good. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Population aging, technological advances, and awareness of the benefits of critical care medicine have produced a yearly escalation of patients admitted to the intensive care unit (ICU) and surviving the experience [1]. Nevertheless, after discharge, patients and their families face a long, ☆ The authors have no conflict of interest to declare. ☆☆ Disclosures: Some results from this study were presented in abstract format in the following 2 international conferences: -
-
Das Neves AV, Vasquez DN, Intile D, Moreira J, Loudet C, Sáenz MG, et al: Quality of life in critical illness survivors in Argentina. Evaluation with Euroqol 5D: a prospective cohort study. Am J Respir Crit Care Med 2012;185:A2288. Das Neves AV, Vasquez DN, Loudet C, Moreira J, Intile D, Sáenz MG, et al: Lack of awareness of severity of illness is a common factor after ICU. Intensive Care Med 2013;39: Suppl2;170. A663.
★ This work was supported in part by the Grant Carrillo-Oñativia 2010, National Committee Research in Health, Argentine National Ministry of Health, Ciudad Autónoma de Buenos Aires, Argentina. ⁎ Corresponding author. Calle 42 N° 577, 1900 La Plata, Buenos Aires, Argentina. Tel.: +54 2214220507. E-mail address: [email protected] (E. Estenssoro). http://dx.doi.org/10.1016/j.jcrc.2015.05.021 0883-9441/© 2015 Elsevier Inc. All rights reserved.
difficult time during which the physical, psychologic, and neurocognitive sequels become evident and greatly affect their quality of life. These adverse consequences of critical care were recently described as the post-intensive care syndrome [2], conditions that also involve nonmedical issues, such as economic and employment instability [3-6]. The impact of the burden of these diseases on health-related quality of life (HRQOL) has been studied in patients with acute respiratory distress syndrome (ARDS) [7,8] and in general populations of critically ill patients [9,10]. Information about these problems from Latin America is, however, scarce or nonexistent [11]. On these grounds, our goal was to evaluate the HRQOL and its independent determinants, in patients admitted to a medical-surgical ICU in Argentina at 1, 3, 6, and 12 months after discharge. Secondary objectives were to describe the prevalence and evolution of physical and psychologic problems.
2. Methods This study was conducted in a 14-bed ICU located in a universityaffiliated hospital, between May 1, 2010, and July 1, 2012. We included
1050
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054
consecutive patients 15 years or older who stayed more than 48 hours on mechanical ventilation (MV), except for those who refused to participate, were prisoners, or mentally disabled. Readmissions during the study protocol were considered for follow-up only after the first admission. We recorded age, sex, severity of acute illness and of underlying diseases (Acute Physiology and Chronic Health Evaluation II and Charlson scores, respectively), diagnosis, category of ICU admission (medical, unscheduled/scheduled surgery, and trauma), organ failures (SOFA [Sequential Organ Failure Assessment] score), admission Glasgow score, hospital mortality, duration of MV and ICU, and hospital length of stay. We also registered events and conditions potentially associated to worsening long-term outcomes, such as ARDS, shock (defined as vasopressor requirement), acute kidney injury (defined as the duplication of serum creatinine or a urine output b0.5 mL/kg per hour during 12 hours) [12], Glasgow score, requirement of decompressive craniectomy and, at discharge, Glasgow score, anemia, tracheostomy, jejunostomy/gastrostomy, dysphagia (defined as any abnormality in swallowing as detected by “video deglutition” examination, which was performed in all patients) and “urinary incontinence” (defined as the unintentional loss of urine). Follow-up protocol: before ICU discharge, patients or next of kin signed written consent forms agreeing to participate in the study, which consisted of 4 face-to-face or telephone interviews with either the patients or a relative, 1, 3, 6, and 12 months after hospital discharge. At 1 month, they were invited to a personal interview at a specially designed ICU office. Should the patient not come, he/she was subsequently contacted by up to 3 phone calls. In the latter case, we conducted telephone interviews with the patient or, if unavailable/unable to respond, with the next of kin. The procedure was repeated at each of the prescheduled visits. All interviews were performed by the same 3 researchers (ADN, DNV, and MGS). At each interview, patients were asked about symptoms defined as relevant by the Guidelines of Rehabilitation after Critical Illness of the National Institute for Health and Clinical Excellence (United Kingdom) [6], also if they were receiving any kind of rehabilitation, and if they had returned to their previous activities. A clinical examination was conducted. In addition, the following questionnaires were completed:
values at the different time points were compared with month 1, with t test for paired measurements or Wilcoxon signed rank test for continuous data and McNemar test for categorical data. A P value b .05 was considered as significant. We performed a longitudinal analysis to evaluate the effect of time in the EQ-index, our main outcome measure for HRQOL, and to adjust for important covariates. We fitted our data to a marginal model by means of generalized estimating equations because EQindex exhibited a nonparametric distribution [18]. Generalized estimating equation method was used to account for correlations between repeated measurements within subjects over time. An unstructured correlation matrix was selected. Variables that presented a P value b .10 were entered into a multivariate analysis to generate a best model with predictive factors. Sample size was estimated according to EQ-index, the main outcome measure, for observational studies. We selected a mean value of 0.59, which represents 2 SD below the expected value of 0.91 ± 0.16, for subjects between ages 35 and 44 years [14,19], including the mean of value of patients' age in the ICU (43 ± 17 years) [20]. For a desired total width of the interval of 0.06, we calculated the standardized width for a 95% confidence level, which determined a sample size of 97 subjects [21]. We anticipated a 15% of dropouts, so the sample size was increased to 112 patients. All analyses were performed with STATA 11 software (Stata Corporation, College Station, TX). The Institutional Review Board of the Hospital San Martin de La Plata approved this protocol (no. 20091130TI). 3. Results Of the 208 eligible patients (MV N 48 hours), 86 died in the ICU, 10 fulfilled exclusion criteria, 112 were followed up over time, and 57 (51%) finished the study (Fig. 1). In the 249 interviews, 68% (169/249) of the questionnaires were answered face to face, 22% (56/248) by
- Health-related quality of life was assessed with the EuroQoL Questionnaire–5 Dimensions (EuroQol-5D) Quality of Life [13], a standardized instrument that measures health status in 5 dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. For each of these conditions, the possible answers are level 1, no problems; level 2, some/moderate problems; and level 3, severe/extreme problems [14]. The different weights of the 5 dimensions are summarized in the EQ-index as an estimated number from − 0.005 to 1, which reflects the global health state. Negative values describe a health status perceived as worse than death. The EQ-index was considered the main outcome variable. - The EuroQol-5D also includes the EQ–Visual Analogue Scale (EQ-VAS), a standard vertical 20-cm visual analogue scale similar to a thermometer, in which patients record their global health status [14]. The end points are labeled “best imaginable health state” and “worst imaginable health state (100-0 points, respectively).” This information is considered a quantitative, self-reported measure of health. We used the Argentinian validation of the EuroQol-5D [13]. - Davidson Trauma Scale, which assesses the risk of posttraumatic stress disorder (PTSD). A cutoff point greater than or equal to 40 suggests its diagnosis [15,16]. To further evaluate physical functioning, a 6-minute walk test was performed [17]. Statistical analysis: descriptive statistics used were means ± SD, medians and interquartile ranges (IQR, 0.25-0.75) for continuous variables, and percentages for categorical data. After testing for normality,
Fig. 1. Flow-chart of the study.
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054 Table 1 General description and evolution of patients. Data are expressed as n (%), mean ± SD, or median IQR [25%-75%] No. of patients
112
Age (y) Male sex Charlson 0 (no comorbidities) APACHE II SOFA first 24 h Transferred from: Emergency department Ward Operating room Other institution Admission diagnostic categories: Trauma TBI Medical Emergency surgery Elective surgery ARDS Shock AKI Dialysis a Severe TBI Decompressive craniectomyb Duration of MV (d) LOS ICU LOS hospital
33 [24-49] 76/112 (68) 85/112 (76) 15 ± 6 6±3
56/112 (50%) 46/112 (41%) 32/112 (29%) 15/112 (13%) 9/112 (8%) 64/112 (57%) 85/112 (76%) 37/112 (33%) 11/37 (30%) 21/46 (46%) 18/46 (39%) 15 [9-37] 21 [11-43] 47 [28-79]
Characteristics at discharge: GCS at ICU discharge Tracheostomy at ICU discharge Tube feeding/jejunostomy/gastrostomy at ICU discharge Anemia at ICU discharge Transferred to rehabilitation center
14 ± 1 11/112 (10%) 20/112 (18%) 81/105 (77%) 16/112 (14%)
At 1 y Readmission to the ICU Mortality at 1 ya
12/112 (11%) 7/112 (6%)
66/112 (59%) 11/112 (10%) 25/112 (22%) 10/112 (9%)
APACHE indicates Acute Physiology and Chronic Health Evaluation; SOFA, Sequential Organ Failure Assessment; LOS, length of stay; AKI: acute kidney injury; GCS, Glasgow coma score. a Refers to patients with acute kidney injury. b Refers to patients with TBI.
phone interview with the patient, and 10% (24/248) with the proxy; 20 patients could never be contacted. Population characteristics, events, relevant treatments, and clinical state at ICU discharge are shown in Table 1. Briefly, ours was a very
1051
young, previously healthy, predominantly male population, reflecting the most common single diagnosis, which was severe traumatic brain injury (TBI) (46% of patients). Organ failures, prolonged duration of MV, and extended ICU and hospital stays were very frequent. Long-term mortality and readmission rates were low. During follow-up (Table 2), most physical problems displayed a significant trend to decrease, yet the prevalence of weakness, fatigue, and dyspnea was still relevant at 12 months. Neurologic symptoms such as paresis, swallowing difficulties, and urinary incontinence were still frequent 1 year after discharge. Posttraumatic stress disorder showed a nonsignificant tendency to decline. The utilization of physical rehabilitation decreased significantly over time. Fig. 2A illustrates the behavior of the 5 dimensions of EuroQol-5D. Of note, it is clear that the percentage of patients with moderate/severe problems diminishes over time, with the remarkable exception of anxiety and depression. In Fig. 2B, the evolution of the EQ-index is compared with that of EQ-VAS (self-perceived health status). Whereas the EQ-index improved significantly over time, EQ-VAS stayed within acceptable values from the very beginning. This dissociation occurred in all patients, whether they had TBI (Fig. 2C and D, respectively). With respect to the independent determinants of HRQOL, as measured by the EQ-index, it was strongly and negatively affected by the duration of MV, the presence of shock, and of weakness after 1 year, whereas the effect of time and the return to work/study was positively associated with an improved outcome (Table 3). 4. Discussion In this study, we analyze the evolution of a young, mostly previously healthy population, up to 1 year postdischarge from the ICU. Patients exhibited a high, persistent burden of disease that severely affected their HRQOL. Although most physical symptoms showed significant improvement over the first 6 months, they were still prevalent at 1 year. This was the case for weakness, a near-universally reported problem [7,22]. Although weakness, together with continued fatigue and breathlessness, might have some subjective component, these muscular problems were also reflected in the low predicted value for the 6-minute walk test. What is more, 40% of patients still showed moderate-to-severe mobility problems after 1 year as assessed with the EuroQol-5D and in line with the 24% to 58% described in studies using the same tool [23-26].
Table 2 Physical and psychologic consequences Time interval No. of patients Weakness Fatigue Breathlessness Paresis/paralysis Distance walked in 6 min (% of predicted) Swallowing difficulties Urinary incontinence Pressure sores Tracheostomy Tracheal stenosis Physical therapy DTS (N40)d Return to study/worke EQ-index EQ-VAS
1 month
3 months
6 months
12 months
76 53 (70%) 35 (46%) 11 (14%) 33 (45%) 21 (28%) 15 (20%) 57 (75%) 9 (12%) 8 (11%) 38 (50%) 36% 12% 0.22[0.01-0.69] 68[45-91]
59 33 (56%) 24 (41%) 14 (24%) 22 (36%) 68% 12 (20%)a 7 (12%) 22 (37%)a 4 (7%) 7 (12%) 30 (50%) 29% 31%a 0.52[0.08-0.81]c 68[41-95]
56 23 (40%) b 22 (39%) 8 (14%) 22 (39%) 73% 9 (18%) 6 (11%)a 13 (23%) b 4 (7%) 4 (7%) 19 (33%) 13% 26%a 0.66[0.17-0.79]c 73[53-93]
57 24 (41%)a 25 (42%) 12 (20%) 17 (21%)a 72% 10 (17%) 9 (16%) 4 (7%) c 2 (4%) 3 (5%) 15 (27%) c 20% 31%a 0.68[0.26-0.86]c 74[23-95]
Variables were selected in keeping with the Guidelines of Rehabilitation after Critical Illness of the National Institute for Health and Clinical Excellence [7]. a P b .05 vs month 1. b P b .01 vs month 1. c P b .001 vs month 1. d DTS, Davidson Trauma Scale. A cutoff point more than 40 suggests PTSD. e Refers to the 51% of patients (57/112) who either studied or had employment before ICU admission.
1052
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054 Table 3 Determinants of EQ-index EQ-index
Coef.
SE
z
Time .013 .006 2.08 Days of MV −.006 .002 −3.41 Shock −.214 .080 −2.67 Return to study/work at month 12 .290 .103 2.82 Weakness −.224 .107 −2.08
PNz
95% Confidence interval
0.037 .0007 .0243 0.001 −.011 −.0027 0.008 −.371 −.057 0.005 .088 .491 0.038 −.435 −.012
Coef: coefficient. Because prospective analyses involved multiple evaluations of EQ-index and covariates for each subject over time, we used a repeated-measures analysis that incorporated all periods of follow-up data simultaneously.
Fig. 2. Evaluation of the patients with the different EuroQol instruments. A, Evolution of the 5 components of the EuroQol-5D over time. The percentage of patients with moderate and severe problems diminishes over time, and this decrease is significant at 12 months in each domain. In spite of this, approximately 50% of patients still had important problems with performing usual activities or experienced pain and discomfort. On the contrary, anxiety/depression remained stable throughout the study period. ⁎P b .05 vs month 1, ⁎⁎P b .01 vs month 1; and ⁎⁎⁎P b .001 vs month 1. B, Comparison of the evolution of EQindex and EQ-VAS over the study period in the entire population. Although EQ index experienced a significant improvement over the study period with respect to month 1 value, EQ-VAS did not change and remained within normal values for the Argentinian population. A lack of perception of the present disease and sequelae is evident. C and D, The comparison of EQ-index and EQ-VAS between TBI and non-TBI patients. The dissociation between estimated (EQ-index) and self-perceived (EQ-VAS) health-related quality of life during the first 6 months after hospital discharge is evident in both groups. This implies that mechanisms other than structural brain damage might be operative for this distortion. Data are showed as mean and SD. ⁎P b .05 vs month 1. ⁎⁎P b .01 vs month 1 and ⁎⁎⁎P b .001 vs month 1.
Moderate and severe pain and discomfort were the most frequent symptoms after discharge (80%), decreasing to 48%, a still unacceptably high figure. Ability to self-care and to return to usual activities was significantly improved after 1 year, similar to the studies mentioned. Severe neurologic sequelae, such as incontinence and swallowing difficulties, remain unchanged over time. Remarkably, psychologic symptoms were very frequent, a finding shared with other studies [27-29]. Posttraumatic stress disorder prevalence was high; psychiatric consultation was accepted only in 8 patients (38%); this reluctance has been reported [30]. Posttraumatic stress disorder, however, tended to decrease with time, whereas anxiety/depression remained stable throughout the study period and were present in onethird of patients, similar to the reported prevalence for ICU survivors (23%-48%) [31]. Hypoxia, inflammation, hypoglycemia, older age, and benzodiazepine use, all common in the ICU, have been associated to psychologic symptoms in survivors [28,32]; less is known about the impact of pre-ICU symptoms, level of education, and endogenous patient-related factors. With respect with the return to previous activities, at 1 year, only 30% of patients had returned to study/work, which is lower than the 50% generally reported [32]. Possible causes were the low rate of physical rehabilitation delivered or the unavailability of neurocognitive rehabilitation, a poorly developed activity in Argentina. Many of the patients with TBI who had undergone decompressive craniectomy had not yet received cranioplasty after 1 year, which precluded the return to manual labor and workshops, their usual major activity. As regards the independent determinants of HRQOL, as measured by the EQ-index adjusted by time, we, along with other authors [33,34], have found that HRQOL was associated with few intra-ICU events; in our study, only shock and duration of MV were independent predictors. Shock is the inability of cardiovascular system to satisfy oxygen needs; the resultant tissue hypoxia is a leading cause of multiorgan failure and death. Shock has also been associated with prolonged MV and ICU stay [35,36]. The length of MV might be considered a global indicator of severity and persistence of disease, also having other implications such as prolonged use of sedative and analgesics, which might contribute to adverse cognitive outcomes. Mechanical ventilation duration is also related to ICU length of stay [37]. This is the first study that directly relates shock or duration of MV with a long-term (1-year) outcome—although other investigators have described length of ICU and hospital stays as predictors. With respect to post-ICU conditions independently related with HRQOL, persistence of weakness showed a negative association, whereas a psychosocial issue such as ability to return to previous activities after 1 year and the effect of time per se was positively associated. Other patient characteristics like age (probably because our patients were young), sex, or severity of illness on admission did not impact on HRQOL in this cohort. Not unexpectedly, in the months after ICU discharge, the EQ-index showed a significant improvement. Instead, the EQ-VAS, which
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054
measures self-perceived health status, was already high just 1 month after discharge and remained elevated without significant changes over the entire study period, with similar values to those of the general population median of 75.35 (74.94-75.75) [13]. Thus, patients actually felt better than what was evidenced by the more thoroughly detailed EQ-index. Deterioration of EQ-index with preservation of the EQ-VAS has been described in a general ICU population at 1 time point after discharge [25]. In older ICU survivors, a similar situation of altered SF-12 (Short Form-12 questionnaire) physical with preserved mental component scores has been reported [38], but elderly patients usually adapt to physical limitations and perceive their HRQOL as good [10]. The reason for this discordant behavior between EQ-index and EQ-VAS is not clear. Patients might have developed anosognosia, defined as the lack of awareness of different deficits, ascribed to focal brain lesions or to diffuse brain damage [39]. In our population, anosognosia could have occurred not only in patients with brain injury (severe TBI) but in those without this disorder (Fig. 2C and D, respectively). Our patients had recurrent episodes of shock, ARDS, and required prolonged MV, conditions associated with hypoxia, hypotension, inflammation, high need of benzodiazepine infusions to promote patient adaptation to the ventilator, and other possible mechanisms of injury, which might have contributed to generate or aggravate a global cerebral insult. In addition, the development of psychologic mechanisms for coping with disease cannot be discarded. Denial and recalibration of standards for quality of life have both been described in patients with different diseases [40,41]. All these mechanisms might be operative in patients surviving the ICU experience. A limitation of this study is that it was performed in a single center, so results obtained in this particularly young, previously healthy population might not be representative of all ICU survivors. Baseline evaluation of HRQOL was not performed; nevertheless, this is a rather inexact estimation because recall bias might affect patients' or relatives' evaluation of previous HRQOL [28,42]. We did not measure total dose of benzodiazepine required, which might have been related to psychologic sequelae. In addition, during follow-up, some loss of information occurred: rates of attendance were between 68% and 51% at each time point. Most studies, however, exhibit similar rates of nonresponse. In a systematic review of 53 studies of post-ICU HRQOL, 24 had a response rate between 50% and 79% [10]. Reasons for not attending interviews might vary from feeling too well, to developing symptoms of PTSD such as avoidance and memory recall. Other patients might be still too injured to attend or might have died. In our group, 83% could be contacted, yet some never showed up for interviews. Still, this study has significant strengths: the prospective evaluation of consecutive patients and the personal character of the interview in most cases. Health-related quality of life was repeatedly measured over time, which evidenced the pattern of recovery. Its determinants were identified by longitudinal data analysis, accounting for the effect of time. Finally, to our knowledge, this is the first study evaluating HRQOL in post-ICU patients in South America. We found that rehabilitation and psychologic consultation were low, which underscores the need for creating special care programs to address patients' requirements.
5. Conclusions In this group of patients, HRQOL after ICU discharge was negatively affected by ICU events, such as the presence of shock and the length of MV, and also by important sequelae of critical illness as persistent weakness; whereas the ongoing effect of time and the return to previous activities, a surrogate of global improvement and welfare showed a positively association. Most interestingly, despite the high burden of symptoms, patients exhibited a positive perception of their HRQOL.
1053
References [1] Needham DM, Bronskill SE, Calinawan JR, BMath CA, Sibbald WJ, Pronovost PJ, et al. Projected incidence of mechanical ventilation in Ontario to 2026: preparing for the aging baby boomers. Crit Care Med 2005;33:574–9. [2] Needham DM, Davidson J, Cohen H, Hopkins R, Weinert C, Wunsch H, et al. Improving long-term outcomes after discharge from intensive care unit: report from a stakeholders' conference. Crit Care Med 2012;40:502–9. [3] Jones Ch, Skirrow P, Griffiths RD, Humphris G, Ingleby S, Eddleston J, et al. Rehabilitation after critical illness: a randomized, controlled trial. Crit Care Med 2003;31: 2456–61. [4] Garrouste-Orgeas M, Soufir L, Tabah A, Schwebel C, Vesin A, Adrie C, et al. A multifaceted program for improving quality of care in intensive care units: IATROREF study. Crit Care Med 2012;40:468–76. [5] Elliott D, McKinley S, Alison J, Aitken LM, King M, Leslie GD, et al. Health-related quality of life and physical recovery after a critical illness: a multi-centre randomised controlled trial of a home-based physical rehabilitation program. Crit Care 2011;15:R142. [6] Rehabilitation after critical illness. National Institute for Clinical Excellence; 2009[Available at: www.nice.org.uk/CG83. Accessed March 29, 2010]. [7] Herridge MS, Cheung AM, Tansey CM, Matte-Martyn A, Diaz-Granados N, Al-Saidi F, et al. One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med 2003;348:683–93. [8] Dowdy DW, Eid MP, Sedrakyan A, Mendez-Tellez PA, Pronovost PJ, Herridge MS, et al. Quality of life in adult survivors of critical illness: a systematic review of the literature. Intensive Care Med 2005;31:611–20. [9] Cuthbertson BH, Elders A, Hall S, Taylor J, MacLennan G, Mackirdy F, et al. Mortality and quality of life in the five years after severe sepsis. Crit Care 2013; 172:R70. [10] Oeyen SG, Vandijck DM, Benoit DD, Annemans L, Decruyenaere J, et al. Quality of life after intensive care: a systematic review of the literature. Crit Care Med 2010;38: 2386–400. [11] Griffiths J, Hatch RA, Bishop J, Morgan K, Jenkinson C, Cuthbertson B, et al. An exploration of social and economic outcome and associated health-related quality of life after critical illness in general intensive care unit survivors: a 12-month follow-up study. Crit Care 2013;17:R100. [12] Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, et al. RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care 2006;10:R73. [13] Augustovski FA, Irazola VE, Velazquez, Gibbons L, Craig B, et al. Argentine valuation of the EQ-5D health states. Value Health 2009;4:5875–96. [14] The EuroQol Group. EuroQol—a new facility for the measurement of health-related quality of life. Health Policy 1990;16:199–208. [15] Davidson JR, Book SW, Colket JT. Assessment of a new self-report scale for PTSD. Psychol Med 1999;27:1531–60. [16] Bobes-García J, Portilla MP, Bascarán MD, Sáiz PA, Bousoño M, editors. Banco de Instrumentos básicos para la práctica de la psiquiatría clínica. Barcelona: Editorial Ars Medica; 2004. p. 225. [17] ATS statement. Guidelines for the six-minute walk test. Am J Respir Crit Care Med 2002;166:111–7. [18] Fitzmaurice GM, Laird NM, Ware JH, editors. Applied longitudinal analysis. New Jersey, NY: John Wiley & Sons; 2004. p. 19–44 [291-21]. [19] Cox CE, Martinu T, Sathy SJ, Clay A, Chia J, Gray A, et al. Expectations and outcomes of prolonged mechanical ventilation. Crit Care Med 2009;11:2888–94. [20] Estenssoro E, Reina R, Canales HS, Saenz MG, Gonzalez FE, Aprea MM, et al. The distinct clinical profile of chronically critically ill patients: a cohort study. Crit Care 2006;10:R89. [21] Browner WS, Newman TB, Hulley SB. Estimating sample size and power: applications and examples. In: Hulley SB, Cummings SR, Browner WS, editors. Designing clinical research. Philadelphia, PA: Lippincott Williams & Wilkins; 2007. [22] Brower R. Consequences of bed rest. Crit Care Med 2009;37:S422–8. [23] García Lizana F, Peres Bota D, De Cubber M, Vincent JL. Long-term outcome in ICU patients: what about quality of life? Intensive Care Med 2003;29:1286–93. [24] Granja C, Teixeira-Pinto A, Costa-Pereira A. Quality of life after intensive care—evaluation with EQ-5D questionnaire. Intensive Care Med 2002;28: 898–907. [25] Badia X, Diaz-Prieto A, Gorriz MT, Herdman H, Torrado H, Farrero E, et al. Using EuroQol-5D to measure changes in quality of life 12 months after discharge from an intensive care unit. Intensive Care Med 2001;27:1901–7. [26] Linko R, Suojaranta-Ylinen R, Karlsson S, Ruokonen E, Varpula T, Pettilä V, et al. Oneyear mortality, quality of life and predicted life-time cost-utility in critically ill patients with acute respiratory failure. Crit Care 2010;14:R60. [27] Davydow DS, Gifford JM, Desai SV, Sanjay V, Desai, Bienvenu O, et al. Depression in general intensive care unit survivors: a systematic review. Intensive Care Med 2009; 35:796–809. [28] Bienvenu O, Colantuoni E, Mendez-Tellez P, Dinglas V, Shanholtz C, Husain N, et al. Depressive symptoms and impaired physical function after acute lung injury: a 2-year longitudinal study. Am J Respir Crit Care Med 2012;185:517–24. [29] Hopkins R, Key CW, Suchyta MR, Weaver L, Orme J. Risk factors for depression and anxiety in survivors of acute respiratory distress syndrome. Gen Hosp Psychiatry 2010;32:147–55. [30] Cuthbertson BH, Hull A, Strachan M, Scott J. Post-traumatic stress disorder after critical illness requiring general intensive care. Intensive Care Med 2004;30:450–5. [31] Desai S, Law T, Needham DM. Long-term complications of critical care. Crit Care Med 2011;3:371–9.
1054
A.V. Das Neves et al. / Journal of Critical Care 30 (2015) 1049–1054
[32] Myhren H, Ekeberg Ø, Stokland O. Health-related quality of life and return to work after critical illness in general intensive care unit patients: a 1-year follow-up study. Crit Care Med 2010;38:1554–61. [33] Granja C, Amaro A, Dias C, Costa Pereira A. Outcome of ICU survivors: a comprehensive review. The role of patient-reported outcome studies. Acta Anaesthesiol Scand 2012;56:1092–103. [34] Orwelius L, Nordlund A, Nordlund P, Simonsson E, Bäckman C, Samuelsson A, et al. Preexisting disease: the most important factor for health related quality of life long-term after critical illness: a prospective, longitudinal, multicentre trial. Crit Care 2010;14:R67. [35] Estenssoro E, González F, Laffaire E, Canales H, Saenz MG, Reina R, et al. Shock on admission day is the best predictor of prolonged mechanical ventilation in the ICU. Chest 2005;127(2):598–603. [36] Laupland KB, Kirkpatrick AW, Kortbeek JB, Zuege DJ. Long-term mortality outcome associated with prolonged admission to the ICU. Chest 2006;129:954–9.
[37] Hopkins RO, Jackson J. Critical illness long-term neurocognitive function after critical illness. Chest 2006;130:869–78. [38] Vest MT, Murphy TE, Araujo K, Pisani M. Disability in activities of daily living, depression, and quality of life among older medical ICU survivors: a prospective cohort study. Health Qual Life Outcomes 2011;9:9. [39] Jehkonen M, Laihosalo M, Kettunen J. Anosognosia after stroke: assessment, occurrence, subtypes and impact on functional outcome reviewed. Acta Neurol Scand 2006;114:293–306. [40] Rayson D. On denying denial. J Clin Oncol 2013;31:4371–2. [41] Neuman HB, Park J, Fuzesi S, Temple L. Rectal cancer patients' quality of life with a temporary stoma: shifting perspectives. Dis Colon Rectum 2012;55:1117–24. [42] Dinglas VD, Gifford JM, Husain N, Colantuoni E, Needham D. Quality of life before intensive care using EQ-5D: patient versus proxy responses. Crit Care Med 2013;41: 9–14.
