American Journal of Respiratory and Critical Care Medicine

Acute respiratory distress syndrome (ARDS) has a high mortality and is associated with significant morbidity. Prior outcome studies have focused predominant on short-term outcomes (6–12 months). We assessed longitudinal neurocognitive, emotional, and quality of life in ARDS survivors at hospital discharge, and 1 and 2 years after hospital discharge using neuropsychologic tests and emotional and quality-of-life questionnaires. Neurocognitive sequelae occurred in 73% (54 of 74) of ARDS survivors at hospital discharge, 46% (30 of 66) at 1 year, and 47% (29 of 62) at 2 years. ARDS survivors report moderate to severe depression (16% and 23%) and anxiety (24% and 23%) at 1 and 2 years, respectively. The ARDS survivors had decreased quality of life, with the physical domains improving at 1 year, with no additional change at 2 years. Role emotional, pain, and general health did not change from hospital discharge to 2 years. Mental health improved during the first year and declined at 2 years. ARDS results in significant neurocognitive and emotional morbidity and decreased quality of life that persists at least 2 years after hospital discharge. ARDS can cause significant long-term, brain-related morbidity manifest by neurocognitive impairments and decreased quality of life.

Acute respiratory distress syndrome (ARDS) is a common cause of mortality and morbidity, affecting an estimated 150,000 people per year in the United States (1); however, recent evidence suggests the incidence actually may be higher than first reported (2). Compared with 20 years ago, mortality has decreased from 80% to 30% of ARDS patients (35) resulting in approximately 100,000 people who survive ARDS each year in the United States (6). ARDS occurs in response to a variety of physiologic insults, including sepsis, trauma, pneumonia, massive transfusion, and other medical or surgical conditions. Treatment of ARDS requires aggressive supportive care, including positive pressure ventilation (5) and increased oxygen concentrations, which include the risks of barotrauma, oxygen toxicity, and nosocomial infection.

ARDS may be a consequence of multiple organ system dysfunction, including the central nervous system (7, 8). Patients who survive ARDS are at risk for physical, emotional, and neurocognitive deficits (912) and diminished quality of life (1015) 6–12 months after hospital discharge. Approximately 33% of general medical intensive care unit (ICU) survivors, some with ARDS, had cognitive impairments (16) 6 months after hospital discharge. In 1999, we reported that a majority of ARDS survivors had neurocognitive impairments, including impaired memory, attention, concentration, mental processing speed, and global intellectual decline 1 year after discharge (10). Since our report of cognitive impairments in ARDS survivors, others have made similar observations (9, 12, 16, 17). The prevalence of neurocognitive impairments varies from 25% (12) to 78% in patients with more severe ARDS (10). Neurocognitive impairments are a major determinant in return to work, work productivity, and life satisfaction after ARDS (12). Even mild cognitive impairments may result in clinically significant problems with driving, money management, and the ability to perform activities of daily living (1821).

Survivors of ARDS have decreased quality of life in the first year after hospital discharge, especially in the physical domains (e.g., physical functioning, bodily pain, role physical) (10, 14, 15). Decreased quality of life has been associated with pulmonary symptoms (15, 21), abnormal pulmonary function (22, 23), muscle wasting and weakness (24), neurocognitive impairments (12), depression (11, 15, 23), anxiety (11), and posttraumatic stress disorder (12).

Neurocognitive and emotional function, and quality of life are important because many patients with ARDS survive and exhibit significant morbidity. Prior ARDS outcome studies have focused predominately on short-term outcomes (6–12 months). There are no prospective neurocognitive and emotional outcome studies of ARDS patients beyond 12 months. We previously reported 1-year outcome data in ARDS survivors (11); however, 2-year outcome data have not been reported. The purpose of this longitudinal outcome study was to characterize neurocognitive and emotional function and quality of life 1 year after hospital discharge in a prospectively identified cohort of ARDS survivors. Preliminary data of this study were previously published in abstract form (25).

Patient Selection

For this study, patients with ARDS were selected from a randomized clinical trial of higher tidal volume versus lower tidal volume ventilation management (22). Between February 1994 and December 1999, patients were invited to participate in a 2-year outcome study to assess neurocognitive and emotional function and quality of life. A detailed description of the ventilation strategy is available (22). This 2-year outcome study was approved by the LDS Hospital Institutional Review Board and conformed to institutional and federal guidelines for the protection of human subjects. Written informed consent was obtained from subject before hospital discharge, after treatment of ARDS.

A total of 120 consecutive patients with ARDS were evaluated for this study; 42 patients died, 3 survivors were excluded (2 with cognitive disability, 1 with Alzheimer's disease), and 1 patient declined the study. Of the 74 ARDS survivors enrolled in the study, 3 died in the first year after hospital discharge from either pulmonary fibrosis/cor pulmonale, liver failure, or diabetic complications. Five survivors declined to return for 1-year follow-up (e.g., busy schedules, not interested), resulting in 66 survivors who completed the 1-year evaluation. Two ARDS survivors died in the second year (bowel obstruction, cardiac failure) and 2 declined to return for the 2-year follow-up, resulting in 62 survivors who completed the 2-year evaluation.

The inclusion criteria for the ventilation study from which our patients were selected were: tracheal intubation, PaO2/FiO2 ⩽ 150 mm Hg, pulmonary wedge pressure ⩽ 18 mm Hg (when available), no clinical evidence of congestive heart failure, diffuse infiltrates in three of four quadrants on chest radiographs, age ⩾ 16 years, and presence of an ARDS risk factor (e.g., aspiration, multiple trauma, pancreatitis, pneumonia, sepsis). Patients were excluded (n = 193) if they had disease states that were deemed irreversible (e.g., liver failure, malignancy, patients with acquired immune deficiency syndrome), traumatic brain injury, prior neurologic disease, prior cognitive disability, or if they were enrolled in another ARDS study (e.g., ARDSNet studies). Patient demographics and medical data were collected prospectively as part of routine clinical care, including length of stay; Acute Physiologic and Chronic Health Evaluation II (i.e., APACHE II) scores (26); LDS Hospital multiple organ failure score (27); Charlson comorbidity index (a weighted index that takes into account the number and the seriousness of comorbid diseases) (28); laboratory values; ventilator data; days receiving sedative, narcotic, and paralytic medications; and outcome data.

Assessment of Hypoxemia and Hypotension

Continuous oxygen saturation data to assess hypoxemia were automatically collected using the Ohmeda (Louisville, KY) Biox 3700 and 3740 devices; data were recorded by a computer. Continuous pulse oximetry measurements were assessed during ventilatory support (10). The saturation readings were sampled every 2 minutes, and the median value for each 15-minute period was recorded (29) and categorized if < 90% saturation. The duration of desaturation events was calculated by adding consecutive measurements (each measurement represents a 15-minute interval) that were lower than the saturation threshold.

Continuous mean blood pressure data were automatically collected through the GE-Marquette (Milwaukee, WI) bedside physiologic monitoring system connected to a computer during ventilatory support. Continuous blood pressure was measured from the arterial catheters and were sampled every 2 minutes, and the median value for each 15-minute period was recorded (29). The duration of hypotension events was calculated by adding consecutive measurements (each measurement represents a 15-minute interval) that were more than 60 and less than 50 mm Hg.

Neurocognitive and Emotional Function

Standardized neurocognitive tests were administered at hospital discharge and at 1 and 2 years and assessed general intelligence, attention/concentration, memory, mental processing speed, executive function, and visuospatial abilities. The 1- and 2-year tests were administered in a private office at LDS Hospital or at patients' homes if they were unable to travel to the hospital. There were no reports or indications of severe anxiety for any patients during any test session. The tests included Wechsler Adult Intelligence Test-Revised (30), Wechsler Memory Scale-Revised (31), Rey Auditory-Verbal Learning Test, Rey-Osterrieth Complex Figure Test (copy, immediate recall, and 30-minute delay recall) (32), Trail Making Test Parts A and B (33), and Verbal Fluency test (34). Emotional function was assessed using the Beck Depression Inventory (35) and Beck Anxiety Inventory (36); scores of 0–9 indicate minimal, 10–16 mild, 17–29 moderate, and 30–63 severe depression or anxiety. To assess the possible effects of ARDS on general intellectual functioning, the Oklahoma Premorbid Intelligence Estimation method (OPIE) (37) was used to assess premorbid intellectual function. The OPIE estimate takes current performance on the Wechsler Adult Intelligence Test-Revised into consideration (vocabulary or picture completion subtests) and educational background and current job status. The scores of the OPIE were compared with the measured Full Scale Intelligence Quotient (IQ) at hospital discharge, 1, and 2 years (e.g., Wechsler Adult Intelligence Test-Revised).

Neurocognitive sequelae was defined as scores on 2 or more neuropsychologic tests that are greater than 1.5 SD or one test score that is greater than 2 SD below the normative population mean. Our definition of neurocognitive sequelae is similar to previous studies (16, 38, 39). To assess the magnitude of neurocognitive impairment, an overall impairment score was derived using age, sex, and education corrected T scores (mean = 50, SD = 10) (40, 41) by assigning a numeric value of 1 for each SD the T scores were more than 1 SD below the mean and summing the total number of SDs below the mean across all (n = 25) tests. T scores less than or equal to 40 received a score of 0, scores greater than 1 SD (3039) received a score of 1, scores greater than 2 SD (2029) received a score of 2, scores greater than 3 SD (1019) received a score of 3, and scores greater than 4 SD (0–9) received a score of 4, with a range of 0 to 100 (e.g., 4 × 25 = 100). The overall impairment score was used as the primary outcome.

Quality of Life

The Medical Outcome Study 36-Item Short Form Health Survey (SF-36) (42, 43) was administered at hospital discharge, 1 year, and 2 years after hospital discharge to assess health-related quality of life. The eight domains of the SF-36 (physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health) are clustered to form two higher order domains—the physical and mental health scores. Each domain is scored from 0 to 100, with higher scores indicating better quality of life. The SF-36 has been used in a variety of patient populations, and norms for age and sex are available (43, 44).

Statistical Analysis

Descriptive statistics were performed for demographic, medications, medical, neurocognitive, emotional, and SF-36 data. Neurocognitive sequelae are reported as percentage of patients with sequelae at hospital discharge and 1 and 2 years. Paired t tests were used to compare the OPIE premorbid estimated IQ with patients Full Scale IQ at hospital discharge and at 1 and 2 years.

Primary outcome.

The primary outcome was the magnitude of neurocognitive impairment, which was assessed using the overall impairment score. Repeated measures analysis of variance (RMANOVA) was performed to test if the overall impairment score significantly changed over time. A significant change in overall impairment was concluded if p < 0.05. Pairwise comparisons between time points were also performed using a Bonferroni adjusted significance level 0.02 (0.05/3).

Secondary outcomes.

Neurocognitive test scores at hospital discharge and 1 and 2 years after hospital discharge were analyzed while adjusting for tidal volume and gender using a doubly multivariate RMANOVA. Individual neurocognitive test scores were analyzed using univariate RMANOVA only if the overall battery was significant at the 0.05 two-sided significance level. Adjustment for tidal volume and sex would be included in further analysis only if those factors were significant in the overall test battery. Using the method of Fisher's least significant difference for individual tests that were significant only at the 0.05 two-sided significance level from the univariate RMANOVA, pairwise comparisons between hospital discharge and 2-year test scores and 1-year and 2-year test scores were performed.

Pearson's correlations between ICU length of stay, Acute Physiologic and Chronic Health Evaluation II scores, duration of intubation, oximetry SaO2 less than 90% and mean blood pressure less than 60 and less than 50 mm Hg, and days on sedative, narcotic, and paralytic medications with neurocognitive scores were performed. Correlations were considered significant at the 0.05 two-sided significance level. Pearson's correlations between 1-year Beck Depression Inventory and Beck Anxiety Inventory scores and neurocognitive scores were performed. Correlations were considered significant at the 0.05 two-sided significance level.

Similar to analysis of neurocognitive outcome, the SF-36 data were analyzed using a doubly multivariate RMANOVA to determine if the domain scores of the SF-36 significantly changed over time. Individual domain scores were analyzed using univariate RMANOVA if the overall analysis was significant at the 0.05 two-sided significance level. For individual domain scores that were significant from the univariate RMANOVA at the 0.05 two-sided significance level, pairwise comparisons between hospital discharge and 2-year domain scores and 1-year and 2-year domain scores were performed.

Descriptive statistics and medical data are presented in Table 1

TABLE 1. Medical data for the ards survivors (n = 74)


Medical Variables

Mean ± SD

Range
Number of ARDS risk factors 3.1 ± 1.3 0–11
Intubation duration, d 28 ± 19 3–94
ICU length of stay, d 34 ± 20 5–98
Hospital length of stay, d 39 ± 22 7–122
Charlson comorbidity index  1 ± 1.2 0–4
Study enrollment
 APACHE II score18.1 ± 6.6 5–41
 MOF score 7.2 ± 3.5 0–18
 PaO2/FIO2107 ± 3254–151
 FIO2, % 68 ± 1540–100
 PaO2, mm Hg 69 ± 1347–121
Total ICU stay
 Mean MOF score 6.5 ± 2.2 0–18
 Mean PaO2/FIO2187 ± 6554–202
 Mean FIO2, mm Hg 52 ± 1035–74
 Hours oximetry SaO2 < 90% (n = 72)106 ± 128 0–656
 Hours MBP < 60 mm Hg (n = 44) 2.1 ± 3.6 0–22
 Hours MBP < 50 mm Hg (n = 9)0.08 ± 0.23 0–1.3
 Days receiving sedatives 20 ± 15 3–85
 Days receiving narcotics 16 ± 16 0–80
 Days receiving paralytics
  5 ± 6
 0–38

Definition of abbreviations: APACHE II = Acute Physiology and Chronic Health Evaluation (27); ARDS = Acute Respiratory Distress Syndrome; ICU = Intensive Care Unit; MBP = mean blood pressure; MOF = multiple organ failure (28).

ARDS risk factors are aspiration, multiple trauma, pancreatitis, pneumonia, sepsis, systemic inflammatory response, shock, major surgery, chronic alcohol use, multiple fractures, near drowning, or amniotic fluid embolism.

LDS Hospital multiple organ failure score (28) is obtained by assigning a score of 1 for each organ system in failure including renal, cardiovascular, coagulation, hepatic, central nervous, gastrointestinal, and immunologic and summing the total number of organs that failed at least once after ARDS onset. Definitions of organ failure are as follows: renal = creatinine ⩾ 3.5 mg/dl, dialysis, or urine outcome < 25 ml/h for 24 h; cardiac = cardiac index < 1.8 L/min/m2, systolic blood pressure < 90 mm Hg with pressor drugs, or arterial-venous oxygen difference > 7.0 ml O2/dl; coagulation = platelet count < 60,000/μl or prothrombin time or partial thromboplastin time > 1.5 × control; hepatic = bilirubin > 5 mg/dl or prothrombin time or partial thromboplastin time > 1.5 × control; central nervous = Glasgow coma score < 8 for 3 d; gastrointestinal = pancreatitis causing shock, ruptured necrotic/ischemic viscus, or gastrointestinal hemorrhage requiring more than 2 U packed red blood cells; and immunologic = prednisone or equivalent of 50 mg/d for at least 7 d or ⩾ 20 mg/d for at least 1 mo, or other immunosuppressive agent, or known acquired immune deficiency syndrome (28).

. There were 41 females and 33 males, with a mean age of 46 ± 16 years (range 16–81 years), median age (interquartile range) of 46 (3557) years, and mean education level of 13 ± 2.3 years (range 9–22 years). Excluding patients that died, the follow-up rate was 93% at 1 year and 90% at 2 years. Thirty-one patients (42%) were referred for rehabilitation before hospital discharge: 23 for physical weakness or deconditioning, 4 for limb fractures, and 4 for limb amputations. Nine of these patients (12%) were subsequently identified on the rehabilitation unit as having cognitive impairments and received cognitive rehabilitation. At 2 years, 34% (21 of 62) were working or were full-time students, 34% (21 of 62) were receiving disability payments started after hospital discharge for ARDS, and 32% (20 of 62) were not working or retired. For ARDS survivors who completed the 1-year outcome, 70% (46 of 66) had neurocognitive sequelae at hospital discharge, 46% (30 of 66) at 1 year, and 47% (29 of 62) 2 years after hospital discharge.

At 2 years, intellectual function as measured by the Wechsler Adult Intelligence Test-Revised resulted in a mean Full Scale IQ of 99.7 ± 11.9, suggesting average intellectual functioning in the ARDS survivors. The OPIE premorbid IQ estimate had a mean of 99.3 ± 10.2. The OPIE premorbid IQ was significantly lower than the patients with ARDS measured using the Full Scale IQ at hospital discharge (89.7 versus 99.3; t = 9.75, p ⩽ 0.001); however, there was no difference between the OPIE and the patients measured Full Scale IQ at 1 year (p = 0.2) or 2 years (p = 0.72). Figure 1

shows the intelligence and memory scores for the ARDS patients.

The Beck Depression Inventory and Beck Anxiety Inventory mean scores at 2 years were 10.4 ± 10.6 (range 0–51) and 10.0 ± 10.1 (range 0–45), respectively, indicating minimal depression and anxiety. However, 23% (n = 14 patients) had scores greater than 16 on the Beck Depression and Beck Anxiety Inventories, indicating moderate to severe symptoms of depression and anxiety.

Primary Outcome

The primary outcome, magnitude of neurocognitive impairment, was assessed using the mean overall impairment scores. The mean overall impairment scores at hospital discharge were 13.3 ± 8.7 (range 0–42), at 1 year 5.7 ± 5.8 (range 0–26), and 2 years 5.9 ± 6.3 (range 0–27). The RMANOVA showed significant differences in the overall impairment score over time [F(2,51) = 31.96, p < 0.001]. Pairwise comparisons between time points showed significant improvement in neurocognitive impairment between hospital discharge and 1-year follow-up (difference = 7.2, p < 0.05) and hospital discharge and 2-year follow-up (difference = 7.4, p < 0.05). There was no difference between the 1- and 2-year overall impairment scores (difference = 0.17).

Secondary Outcomes

Results of the RMANOVA for neurocognitive scores are shown in Table 2

TABLE 2. Rmanova results for neurocognitive outcome


Multivariate Results

F

p Value
Between-subjects effects
 Tidal volume 0.9 0.6
 Sex 1.7 0.09
 Tidal volume × sex 0.5 0.95
Within-subjects effects
 Time 2.7< 0.001
 Time × tidal volume 1.1 0.4
 Time × sex 0.9 0.7
 Time × tidal volume × sex 1.1 0.3
Univariate Results for Time
F
p
Neurocognitive test
WAIS-R
 FSIQ47.7< 0.001
 VIQ20.1< 0.001
 PIQ73.9< 0.001
 Information 1.5 0.23
 Digit span 5.5 0.007
 Vocabulary 1.8 0.17
 Arithmetic10.1< 0.001
 Comprehension 1.0 0.37
 Similarities13.4< 0.001
 Picture completion16.5< 0.001
 Picture arrangement29.1< 0.001
 Block design52.2< 0.001
 Object assembly22.6< 0.001
 Digit symbol61.6< 0.001
Trails Part A18.9< 0.001
Trails Part B35.9< 0.001
Complex figure 30-sec delay recall11.8< 0.001
RAVL Trial 112.3< 0.001
RAVL Trial 5 8.3 0.001
RAVL delay recall21.1< 0.001
WMS-R
 Verbal memory index 4.8 0.01
 Visual memory index 9.5< 0.001
 General memory index 4.4 0.02
 Attention/concentration index 3.8 0.03
 Delay recall index17.8< 0.001
Hospital vs. 2 yr1 vs. 2 yr
Post Hoc Comparisons for Time
p
p
Neurocognitive test
WAIS-R
 FSIQ< 0.0010.03
 VIQ< 0.0010.77
 PIQ< 0.0010.01
 Digit span 0.020.99
 Arithmetic 0.0010.99
 Similarities< 0.0010.99
 Picture completion< 0.0010.05
 Picture arrangement< 0.0010.09
 Block design< 0.0010.95
 Object assembly< 0.0010.99
 Digit symbol< 0.0010.68
Trails Part A< 0.0010.99
Trails Part B< 0.0010.99
Complex figure 30-sec delay recall< 0.0010.04
RAVL Trial 1< 0.0010.99
RAVL Trial 5 0.0040.99
RAVL delay recall< 0.0010.59
WMS-R
 Verbal memory index 0.0080.99
 Visual memory index 0.0010.26
 General memory index 0.040.99
 Attention/concentration index 0.190.99
 Delay recall index
< 0.001
0.23

Definition of abbreviations: FSIQ = Full Scale Intelligence Quotient; PIQ = Performance Intelligence Quotient; RAVL = Rey auditory verbal learning test; VIQ = Verbal Intelligence Quotient; WAIS-R = Wechsler Adult Intelligence Scale-Revised; WMS-R = Wechsler Memory Scale Revised.

. Effects from tidal volume and gender were not significant. The neurocognitive scores were significantly different over time [F(50,150) = 2.7, p < 0.001]. Only three of the neurocognitive test scores differed between 1 and 2 years (Table 2).

The duration of hypoxemia measured by the duration of oximetry SaO2 < 90%, correlated with digit span (r = −0.247, p = 0.03), Rey Auditory Verbal Learning Trial 5 (r = −0.197, p = 0.04), and Trail Making Test Part B (r = 0.269, p = 0.01) at hospital discharge, which were in the expected direction. There were no significant correlations between duration of hypoxemia with any neurocognitive scores at 1- and 2-year follow-up. There were no significant correlations between neurocognitive scores and ICU length of stay, Acute Physiologic and Chronic Health Evaluation II scores, duration of intubation, or the number of days the patients received sedative, narcotic, or paralytic medications.

Of the 74 subjects, 60% (44 of 74) had at least one 15-minute period with a mean blood pressure less than 60 mm Hg and 12% (9 of 74) less than 50 mm Hg. There were no significant correlations with the mean blood pressure less than 60 mm Hg and any neuropsychologic test score. The mean blood pressure less than 50 mm Hg correlated with memory scores at hospital discharge: Rey-Osterrieth Complex Figure delay recall (r = .375, p = 0.001); Delay Recall Index of the Wechsler Memory Scale-Revised (r = 0.219, p = 0.04); at 1 year with Rey Auditory Verbal Learning Trial 5 (r = 0.258, p = 0.04), and Rey Auditory Verbal Learning Test delay recall (r = 0.320, p = 0.009). There were no significant correlations between neuropsychologic test scores and mean blood pressure less than 50 mm Hg at 2 years. The Beck Depression Inventory correlated with object assembly (r = −0.268, p = 0.04) and the Beck Anxiety Inventory correlated with object assembly (r = −0.354, p = 0.006) at 2 years.

Quality of Life

The SF-36 data for hospital discharge and at 1 and 2 years after hospital discharge compared with normal population data are shown in Figure 2

. Results of the RMANOVA for SF-36 data are shown in Table 3

TABLE 3. Rmanova results for the SF-36


Multivariate Results

F

p
Between-subjects effects
 Time 8.9< 0.001
Univariate Results
F
p
SF-36 domains
 Physical functioning46.4< 0.001
 Social functioning79.1< 0.001
 Role physical34.3< 0.001
 Vitality25.9< 0.001
 Role emotional 2.1 0.13
 Pain 1.7 0.2
 General health 0.70.5
 Mental health 5.3 0.006
Hospital vs. 2-yr1-yr vs. 2-yr
Post Hoc Comparisons
p
p
Physical functioning< 0.001 0.99
Social functioning< 0.001 0.42
Role physical< 0.001 0.68
Vitality< 0.001 0.98
Mental health
0.99
0.02
. The SF-36 scores were significantly differed over time [F(2,102) = 8.9, p < 0.001]. The SF-36 domains showed three distinct patterns in our ARDS survivors. Pattern 1 for physical functioning, social functioning, role physical, and vitality domains were low at hospital discharge, improved during the first year, with no additional improvement at 2 years. Pattern 2 for role emotional, pain, and general health domains were stable from hospital discharge through 2-year follow-up. Pattern 3 for the mental health domain showed significant improvement during the first year, but returned to the hospital discharge level at 2 years.

Our data extend previous findings because we assessed 2-year neurocognitive outcome in a prospectively identified cohort of ARDS survivors. The premorbid estimated IQ of patients with ARDS was significantly lower than their measured IQ at hospital discharge; however, the patients' IQ improved to their premorbid level by 1 year with no additional improvement at 2 years. Seventy percent of our ARDS survivors had neurocognitive sequelae at hospital discharge, 45% at 1 year, and 47% at 2 years. There was no improvement in neurocognitive sequelae from 1 to 2 years. Other studies report neurocognitive impairment prevalence rates of 20 to 35% 1 year after ARDS (9, 12, 16). The neurocognitive impairments at 2 years are similar to those in other ARDS survivors (9, 12, 17) and in medical ICU survivors (16). The 2-year neurocognitive scores of the ARDS survivors with neurocognitive sequelae (∼ 50% of survivors) fell below the 6th percentile of the normal distribution of cognitive function. These ARDS survivors found activities that require executive function, memory, attention, or quick mental processing speed to be very difficult or impossible. The neurocognitive impairments in our patients are similar to those observed in medical ICU survivors (16) after carbon monoxide poisoning (39) and several years after elective coronary artery bypass surgery (38).

It is difficult to compare our results with previous findings (9, 12, 16, 17) because our study differs with respect to the definition of sequelae, neuropsychologic tests administered, time to follow-up, patient population, study design (prospective versus retrospective), and ARDS etiology and severity. Differences in the neurocognitive tests may have also contributed to the observed differences. For example, one study used a self-reported questionnaire of memory (9), and another administered only a brief memory and attention test instead of a comprehensive neuropsychologic test battery (12).

Although 42% of the ARDS patients received rehabilitation therapy, most of the ARDS patients were not evaluated for cognitive impairments, with only 12% identified as having post-ICU cognitive impairments by the clinical rehabilitation team and subsequently received cognitive rehabilitation therapy. The cognitive impairments in the patients with ARDS appear to be underrecognized by ICU and rehabilitation providers. Studies suggest that many physicians fail to recognize (or assess) cognitive impairment in 35 to 90% of patients in non-ICU clinical practice settings (45, 46). In addition, cognitive impairments are rarely evaluated in ICU patients (16) and may be overlooked in one of every two cases (47). Education regarding cognitive sequelae after ARDS is needed to enhance referral of patients to rehabilitation, not only for physical debilitation and weakness, but also for cognitive impairments.

Only two patients had no episodes of hypoxemia, despite attentiveness of respiratory therapists, nurses, and critical care physicians to all patients. The duration of hypoxemia at hospital discharge modestly correlated with attention, verbal memory, and executive function. The correlations between the duration of hypoxemia and neuropsychologic outcome are less robust than we previously reported (10). The duration of hypoxemia did not correlate with any neurocognitive score at 2 years. The duration of hypotension modestly correlated with memory impairments at hospital discharge and 1 year, but not at 2 years. These findings suggest that the effects of hypoxemia and hypotension play a role in neurocognitive sequelae at discharge and 1 year but not at 2 years, which may be due to improvement in cognitive sequelae in some patients over time. We did not find relationships between ICU length of stay, Acute Physiologic and Chronic Health Evaluation II scores, and duration of intubation, tidal volume, gender, or days receiving sedative, narcotic, or paralytic medications with neurocognitive function. Others have reported that the cumulative dose of some sedatives (48) and delirium (4951) may contribute to neurocognitive and affective sequelae in critically ill patients.

One potential mechanism of neurocognitive impairments is hyperglycemia, which exacerbates hypoxic (52) and ischemic brain injury (53, 54). Hyperglycemia is associated with poor neurologic outcome after head injury (55, 56) and stroke (53). Other potential mechanisms for the neurocognitive impairments include elevated cytokines and inflammation from sepsis (57). For example, patients with elevated cytokines from chronic fatigue syndrome have impaired attention, concentration, and memory (58). A study that induced cytokine activation in healthy volunteers found elevated cytokine levels were associated with impaired memory (59). Inflammation has also been implicated in the pathogenesis of cognitive impairments associated with Alzheimer's disease (60). The etiology of the neurocognitive impairments is undoubtedly multifactorial and the subject of ongoing discussion. Future investigations are required to further delineate etiologic determinants and identify interventions to improve or prevent neurocognitive impairments.

Our ARDS patients reported depression and anxiety that persisted up to 2 years after hospital discharge. The depression and anxiety observed in our ARDS survivors is similar to that previously reported (22, 61, 62). Critically ill patients who survive their ICU stay frequently report depression and anxiety (9, 13, 15, 22, 24, 63). Approximately 23% of our patients with ARDS reported moderate to severe symptoms of depression and anxiety at 2 years. However, the Beck Depression Inventory and Beck Anxiety Inventory correlated only with one neurocognitive test, object assembly, at 2 years. Because only 1 test of 25 correlated with depression and anxiety, this suggests that symptoms of depression and anxiety did not influence neurocognitive performance. Similarly, at 1 year, there was no association between depression and anxiety and neurocognitive outcome (11).

The determinants of depression and anxiety associated with ARDS and critical illness is unknown. Potential mechanisms that may contribute to affective morbidity include organ dysfunction, medications (64), pain, altered sensory inputs, sleep deprivation, elevated cytokines (65), stress related activation of the hypothalamic pituitary axis (66), cerebral atrophy (67), hypoxemia (68), neurotransmitter aberrations from brain injury (69), elevated catecholamine levels from administration of epinephrine or norepinephrine in the ICU, and stress from critical illness and ICU treatment (7072).

The quality of life in our ARDS survivors was significantly lower than the normal population, and was similar to previous reports in other ARDS patients (9, 10, 1215, 73). We found three distinct patterns for the SF-36 domains in our ARDS patients from hospital discharge to 2 years after discharge. The first pattern showed physical function, social function, role physical, and vitality domains were low at hospital discharge, improved at 1 year with no additional improvement at 2 years, similar to neurocognitive function. A second pattern for role emotional, pain, and general health domains showed higher scores at hospital discharge, albeit lower than normal, with no change over time. Of interest is the lack of change in the general health domain from hospital discharge to 2 years. The lack of change is likely the result of the nature of the items included in the general health domain, because they assess patients' perceptions of their general health (e.g., is your health excellent to poor compared with most people, do you get sick more easily than other people, do you expect your health to get worse) rather than compared with their current health or health post-ARDS. That is, they were asked to assess their general health compared with other individuals, rather than to how their health was before ARDS or the post-ARDS effects on their general health. A third pattern for the mental health domain showed improvement during the first year and then declined to the hospital discharge level at 2 years. These results raise the question as to why the mental health domain declines at 2 years. One possible reason for the decline in mental health at 2 years may be depression, which has been associated with decreased life satisfaction (15) and lower quality of life (11).

The strengths of this study include the longitudinal prospective cohort study design, consistent follow-up times, high follow-up rates, automated hypoxemia and hypotension data capture, comprehensive measures of neurocognitive outcome, and a well-defined ARDS population.

The limitations of our study and most prior ARDS outcome studies include the inability to measure premorbid cognitive and emotional function, quality of life, and the lack of an appropriate control group. Because we did not follow a control group of ICU survivors who did not have ARDS, the neurocognitive sequelae we observed may not be specific to ARDS but rather may represent morbidity from critical illness. We also did not follow a normal control group; however, it should be noted that we used demographically (i.e., age, sex, and education) corrected neuropsychologic test scores for statistical analyses, which correct for variables known to effect test performance. The use of the automated moving median technique for oximetry and blood pressure measurements minimizes variability (29), and could miss brief periods of hypoxemia or hypotension. The correlations we found between hypoxemia and hypotension with neurocognitive sequelae may underestimate the magnitude of brief decrements. Our findings may not generalize to other ARDS patients with less severe ARDS, patients with acute lung injury, or patients who are not enrolled in randomized clinical trials because such studies often employ stringent inclusion and exclusion criteria.

ARDS is a life-threatening illness that results in significant long-term morbidity, including neurocognitive impairments, depression, anxiety, and decreased quality of life. ARDS survivors' neurocognitive impairments were more likely to improve than decline at 1 year with little additional improvement at 2 years. These patients with ARDS had decreased quality of life with the physical domains improving during the first year; little change in role emotional, pain, and general health; and improvement, then subsequent decline, in mental health at 2 years. ARDS is common, and survival appears to be improving, yet physicians need to be mindful of adverse brain-related outcomes in many of these patients. As survival improves, attention to proximal determinants and possible interventions to prevent cognitive and emotional morbidity are warranted and should be an emphasis in ARDS and critical care outcomes research.

1. Rubenfeld GD, Doyle RL, Matthay MA. Evaluation of definitions of ARDS. Am J Respir Crit Care Med 1995;151:1270–1271.
2. Rubenfeld GD. Epidemiology of acute lung injury. Crit Care Med 2003;31:S276–S284.
3. Abel SJ, Finney SJ, Brett SJ, Keogh BF, Morgan CJ, Evans TW. Reduced mortality in association with the acute respiratory distress syndrome (ARDS). Thorax 1998;53:292–294.
4. Milberg JA, Davis DR, Steinberg KP, Hudson LD. Improved survival of patients with acute respiratory distress syndrome (ARDS): 1983–1993. JAMA 1995;273:306–309.
5. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000;342:1301–1308.
6. Bersten AD, Edibam C, Hunt T, Moran J. Incidence and mortality of acute lung injury and the acute respiratory distress syndrome in three Australian States. Am J Respir Crit Care Med 2002;165:443–448.
7. Bell RC, Coalson JJ, Smith JD, Johanson WG Jr. Multiple organ system failure and infection in adult respiratory distress syndrome. Ann Intern Med 1983;99:293–298.
8. Montgomery AB, Stager MA, Carrico CJ, Hudson LD. Causes of mortality in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1985;132:485–489.
9. Al-Saidi F, McAndrews MP, Cheunt AM, Tansey CM, Matte-Martyn A, Diaz-Granados N, Herridge MS. Neuropsychological sequelae in ARDS survivors. Am J Respir Crit Care Med 2003;167:A737.
10. Hopkins RO, Weaver LK, Pope D, Orme JF, Bigler ED, Larson-Lohr LV. Neuropsychological sequelae and impaired health status in survivors of severe acute respiratory distress syndrome. Am J Respir Crit Care Med 1999;160:50–56.
11. Hopkins RO, Weaver LK, Chan KJ, Orme JF. Quality of life, emotional, and cognitive function following acute respiratory distress syndrome. J Int Neuropsychol Soc (In press)
12. Rothenhausler HB, Ehrentraut S, Stoll C, Schelling G, Kapfhammer HP. The relationship between cognitive performance and employment and health status in long-term survivors of the acute respiratory distress syndrome: results of an exploratory study. Gen Hosp Psychiatry 2001;23:90–96.
13. Angus DC, Musthafa AA, Clermont G, Griffin MF, Linde-Zwirble WT, Dremsizov T, Pinsky MR. Quality-adjusted survival in the first year after the acute respiratory distress syndrome. Am J Respir Crit Care Med 2001;163:1389–1394.
14. Davidson TA, Caldwell ES, Curtis JR, Hudson LD, Steinberg KP. Reduced quality of life in survivors of acute respiratory distress syndrome compared with critically ill control patients. JAMA 1999;281:354–360.
15. Weinert CR, Gross CR, Kangas JR, Cury CL, Marinelli WA. Health-related quality of life after acute lung injury. Am J Respir Crit Care Med 1997;156:1120–1128.
16. Jackson JC, Hart RP, Gordon SM, Shintani A, Truman B, May L, Ely EW. Six-month neuropsychological outcome of medical intensive care unit patients. Crit Care Med 2003;31:1226–1234.
17. Marquis K, Curtis J, Caldwell E, Davidson T, Davis J, Sanchez P, Rosenbaum G, Hudson LD, Steinberg KP. Neuropsychological sequelae in survivors of ARDS compared with critically ill control patients [abstract]. Am J Respir Crit Care Med 2000;161:A383.
18. Albert SM, Michaels K, Padilla M, Pelton G, Bell K, Marder K, Stern Y, Devanand DP. Functional significance of mild cognitive impairment in elderly patients without a dementia diagnosis. Am J Geriatr Psychiatry 1997;7:213–220.
19. Griffith HR, Belue K, Sicola A, Krzywanski S, Zamrini E, Harrell L, Marson DC. Impaired financial abilities in mild cognitive impairment: a direct assessment approach. Neurology 2003;60:449–457.
20. Nygard L. Instrumental activities of daily living: a stepping stone towards Alzheimer's disease diagnosis in subjects with mild cognitive impairment? Acta Neuro Scand 2003;179:S42–S46.
21. Tabert MH, Albert SM, Borukhova-Milov BA, Camacho Y, Pelton G, Liu X, Stern Y, Devanand DP. Functional deficits in patients with mild cognitive impairments: prediction of AD. Neurology 2002;58:758–764.
22. Orme J Jr, Romney JS, Hopkins RO, Pope D, Chan KJ, Thomsen G, Crapo RO, Weaver LK. Pulmonary function and health-related quality of life in survivors of acute respiratory distress syndrome. Am J Respir Crit Care Med 2003;167:690–694.
23. Schelling G, Stoll C, Haller M, Briegel J, Manert W, Hummel T, Lenhart AHM, Polaset J, Meier M, Preuss U, et al. Health-related quality of life and posttraumatic stress disorder in survivors of the acute respiratory distress syndrome. Crit Care Med 1998;26:651–659.
24. Herridge MS, Cheung AM, Tansey CM, Matte-Martyn A, Diaz-Granados N, Al-Saidi F, Cooper AB, Guest CB, Mazer CD, Mehta S, et al. and C. C. C. T. Group. One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med 2003;348:683–693.
25. Hopkins RO, Collingridge D, Weaver LK, Chan K, Orme JF Jr. Neuropsychological sequelae following acute respiratory distress syndrome: two year outcome. J Int Neuropsychol Soc 2003;9:584.
26. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 1985;13:818–829.
27. Suchyta MR, Clemmer T, Orme JF Jr, Morris AH, Elliott CG. Increased survival of ARDS patients with severe hypoxemia (ECMO criteria). Chest 1991;99:951–955.
28. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373–383.
29. Oniki TA, Gardner RM. Computerized detection of arterial oxygen desaturations in an intensive care unit. American Medical Informatics Association; 1994.
30. Wechsler D, editor. Wechsler Adult Intelligence Scale, 3rd ed. San Antonio: The Psychological Corporation; 1997.
31. Wechsler D, editor. Wechsler Memory Scale. 3rd ed. San Antonio: The Psychological Corporation; 1997.
32. Rey A. L'examen clinique en psychologic. Paris: Presse Universitaire de France; 1958.
33. Reitan RM, Wolfson D. The Halstead-Reitan neuropsychological test battery: theory and clinical interpretation. Tucson: Neuropsychology Press; 1985.
34. Naugle RI, Cullum CM, Bigler E. Introduction to clinical neuropsychology: a case book. Austin, TX: Pro-Ed; 1998.
35. Beck AT. Beck Depression Inventory: manual. San Antonio: Psychology Corporation; 1987.
36. Beck A, Steer RA. Beck Anxiety Inventory. San Antonio: The Psychological Corporation; 1993.
37. Scott JG, Krull KR, Williamson DJG, Adams RL, Iverson GL. Oklahoma premorbid intelligence estimation (OPIE): utilization in clinical samples. Clin Neuropsychol 1997;11:146–154.
38. Newman M, Kirchner J, Phillips-Bute B, Gaver V, Grocott H, Jones R, Mark D, Reves J, Blumenthal J; Neurological Outcome Research Group and the Cardiothoracic Anesthesiology Research Endeavors Investigators. Longitudinal assessment of neurocognitive function after coronary-artery bypass surgery. N Engl J Med 2001;344:395–402.
39. Weaver LK, Hopkins RO, Chan KJ, Churchill S, Elliott CG, Clemmer TP, Orme JF Jr, Thomas FO, Morris AH. Hyperbaric oxygen for acute carbon monoxide poisoning. N Engl J Med 2002;347:1057–1067.
40. Heaton RK, Grant I, Matthews CG. Comprehensive norms for an expanded Halstead-Reitan battery: demographic corrections, research findings, and clinical applications. Odessa, FL: Psychological Assessment Resources Inc; 1991.
41. Heaton, R. K. Comprehensive norms for an expanded Halstead-Reitan battery: a supplement for the WAIS-R. Odessa: Psychological Assessment Resources Inc; 1994.
42. Stewart A, Hays R, Ware J. The MOS short form general health survey: reliability and validity in a patient population. Med Care 1988;26:724–732.
43. Ware JE, Kosinski M, Keller SD. SF-36 physical and mental health summary scales: a user's manual. Boston: The Health Institute, New England Medical Center; 1994.
44. Ware JE. SF-36 Health survey manual and interpretation guide. Boston: The Health Institute, New England Medical Center; 1993.
45. O'Connor DW, Pollitt PA, Hyde JB, Brook CP, Reiss BB, Roth M. Do general practitioners miss dementia in elderly patients? BMJ 1988;297:1107–1110.
46. Callahan CM, Hendrie HC, Tierney WM. Documentation and evaluation of cognitive impairment in elderly primary care patients. Ann Intern Med 1995;122:422–429.
47. Wilkes MM, Navickis RJ. Patient survival after human albumin administration: a meta-analysis of randomized, controlled trials. Ann Intern Med 2001;135:149–164.
48. Starr JL, Whalley LJ. Drug induced dementia. Drug Saf 1994;11:310–317.
49. Dolan MM, Hawkes WG, Zimmerman SI, Morrison RS, Gruber-Baldini AL, Hebel JR, Magaziner J. Delirium on hospital admission in aged hip fracture patients: prediction of mortality and 2-year functional outcomes. J Gerontol A Biol Sci Med Sci 2001;55A:M527–M534.
50. Jackson JC, Gordon SM, Hart RP, Hopkins RO, Ely EW. The association between delirium and cognitive decline: a review of the empirical literature. Neuropsychol Rev 2004;14:87–98.
51. Katz IR, Curyto JJ, TenHave T MJ, Sands L, Kallan JJ. Validating the diagnosis of delirium and evaluating its association with deterioration over a one-year period. Am J Geriatr Psychiatry 2001;9:148–159.
52. Yager JY, Thornhill JA. The effect of age on susceptibility to hypoxic-ischemic brain damage. Neurosci Biobehav Rev 1997;21:167–174.
53. Bruno A, Biller J, Adams HP Jr, Clarke WR, Woolson RF, Williams LS, Hansen MD. Acute blood glucose level and outcome from ischemic stroke: Trial of ORG 10172 in Acute Stroke Treatment (TOAST) Investigators. Neurology 1999;52:280–284.
54. Yip PK, He YY, Hsu CY, Garg N, Marangos P, Hogan EL. Effect of plasma glucose on infarct size in focal cerebral ischemia-reperfusion. Neurology 1991;41:899–905.
55. Lam AM, Winn HR, Cullen BF, Sundling N. Hyperglycemia and neurological outcome in patients with head injury. J Neurosurg 1991;75:545–551.
56. Young B, Ott L, Dempsey R, Haack D, Tibbs P. Relationship between admission hyperglycemia and neurologic outcome of severely brain-injured patients. Ann Surg 1989;210:466–472; discussion 472–473.
57. Sieser A, Schwarz S, Brainin M. Critical illness polyneuropathy: clinical aspects and long-term outcome. Wien Klin Wochenschr 1992;104:294–300.
58. Patarca-Montero R, Antoni M, Fletcher MA, Klimas NG. Cytokine and other immunologic markers in chronic fatigue syndrome and their relation to neuropsychological factors. Appl Neuropsychol 2001;8:51–64.
59. Reichenberg, A, Yirmiya R, Schuld A, Kraus T, Haack M, Morag A, Pollmacher T. Cytokine-associated emotional and cognitive disturbances in humans. Arch Gen Psychiatry 2001;58:445–452.
60. McGeer PL, Akiyama H, Itagaki S, McGeer EG. Immune system response in Alzheimer's disease. Can J Neurol Sci 1989;16:516–527.
61. Rincon HG, Granados M, Unutzer J, Gomez M, Duran R, Badiel M, Salas C, Martinez J, Mejia J, Ordonez C, et al. Prevalence, detection and treatment of anxiety, depression, and delirium in the adult critical care unit. Psychosomatics 2001;42:391–396.
62. Scragg P, Jones A, Fauvel N. Psychological problems following ICU treatment. Anaesthesia 2001;56:9–14.
63. Kapfhammer HP, Rothenhausler HB, Krauseneck T, Stoll C, Schelling G. Posttraumatic stress disorder and health-related quality of life in long-term survivors of acute respiratory distress syndrome. Am J Psychiatry 2004;161:45–52.
64. Nelson BJ, Weinert CR, Bury CL, Marinelli WA, Gross CR. Intensive care unit drug use and subsequent quality of life in acute lung injury patients. Crit Care Med 2000;28:3626–3630.
65. Katz IR. On the inseparability of mental and physical health in aged persons: lessons from depression and medical comorbidity. Am J Geriatr Psychiatry 1996;4:1–16.
66. Gold PW, Licinio J, Wong ML, Chrousos GP. Corticotropin releasing hormone in the pathophysiology of melancholic and atypical depression and in the mechanism of action of antidepressant drugs. Ann N Y Acad Sci 1995;771:716–729.
67. Pearlson GD, Rabins PV, Kim WS, Speedie LJ, Moberg PJ, Burns A, Bascom MJ. Structural brain CT changes and cognitive deficits in elderly depressives with and without reversible dementia (“pseudodementia”). Psychol Med 1989;19:573–584.
68. Katz IR. Is there a hypoxic affective syndrome? Psychosomatics 1982;23:846–853.
69. Frasure-Smith N, Lesperance F, Talajic M. The impact of negative emotions on prognosis following myocardial infarction: is it more than depression? Health Psychol 1995;14:388–398.
70. McCartney JR, Boland RJ. Anxiety and delirium in the intensive care unit. Crit Care Clin 1994;10:673–680.
71. Skodol AE. Anxiety in the medically ill: nosology and principles of differential diagnosis. Semin Clin Neuropsychiatry 1999;4:64–71.
72. Szokol JW, Vender JS. Anxiety, delirium, and pain in the intensive care unit. Crit Care Clin 2001;17:821–842.
73. McHugh LG, Milberg JA, Whitcomb ME, Schoene RB, Maunder RJ, Hudson LD. Recovery of function in survivors of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1994;150:90–94.
Correspondence and requests for reprints should be addressed to Ramona O. Hopkins, Ph.D., Critical Care Medicine, LDS Hospital, Eighth Avenue and C Street, Salt Lake City, UT 84602. Email:

Related

No related items
American Journal of Respiratory and Critical Care Medicine
171
4

Click to see any corrections or updates and to confirm this is the authentic version of record