[

Original Research Critical Care

]

Delirium and Circadian Rhythm of Melatonin During Weaning From Mechanical Ventilation An Ancillary Study of a Weaning Trial Armand Mekontso Dessap, MD, PhD; Ferran Roche-Campo, MD; Jean-Marie Launay, PharmD, PhD; Anais Charles-Nelson, MSc; Sandrine Katsahian, MD, PhD; Christian Brun-Buisson, MD; and Laurent Brochard, MD

Delirium is frequent in patients in the ICU, but its association with the outcome of weaning from mechanical ventilation has not been assessed. Circadian rhythm alteration may favor delirium. In the current study, we assessed the impact of delirium during weaning and associated alterations in the circadian rhythm of melatonin excretion.

BACKGROUND:

This was a substudy of 70 participants of the B-type Natriuretic Peptide for the Fluid Management of Weaning trial, comparing two fluid management strategies during weaning. Patients with or without delirium (as assessed using the Confusion Assessment Method for the ICU) were compared in terms of baseline characteristics and outcomes and the circadian rhythm of melatonin excretion using the 24-h excretion of its urinary metabolite 6-sulfatoxymelatonin (aMT6s).

METHODS:

Among the 70 patients included, 43 (61.4%) experienced delirium at the initiation of weaning. Delirium at the initiation of weaning was associated with more alcohol consumption, a greater severity of illness, and medication use before weaning (including neuromuscular blockade, antibiotics, sedatives, and narcotics). Delirium at the initiation of weaning was associated with more respiratory and neurologic complications and a reduced probability of successful extubation (Cox multivariate model hazard ratio of successful extubation 5 0.54; 95% CI, 0.30-0.95; P 5 .03). Delirium was also associated with a significant reduction in peak, mean, amplitude, and total values of aMT6s urinary excretion during the first 24 h of weaning (general linear model F statistic 5 5.81, P 5 .019).

RESULTS:

Delirium is frequent at the initiation of ventilator weaning. It is associated with a prolongation of weaning and an alteration in the circadian rhythm of melatonin excretion.

CONCLUSIONS:

TRIAL REGISTRY:

ClinicalTrials.gov; No.: NCT00473148; URL: www.clinicaltrials.gov CHEST 2015; 148(5):1231-1241

Manuscript received March 3, 2015; revision accepted June 15, 2015; originally published Online First July 9, 2015. ABBREVIATIONS: aMT6s 5 6-sulfatoxymelatonin; BMW 5 B-type Natriuretic Peptide for the Fluid Management of Weaning; BNP 5 B-type natriuretic peptide; CAM-ICU 5 Confusion Assessment Method for the ICU; HR 5 hazard ratio; RASS 5 Richmond Agitation-Sedation Scale AFFILIATIONS: From AP-HP (Drs Mekontso Dessap, Roche-Campo, and Brun-Buisson), Hôpitaux Universitaires Henri Mondor, DHU A-TVB, Service de Réanimation Médicale, Créteil, France; Université Paris Est Créteil (Drs Mekontso Dessap and Brun-Buisson), Faculté de Médecine de Créteil, IMRB, Groupe de recherche clinique CARMAS, Créteil, France; Hospital Verge de la Cinta (Dr Roche-Campo), Servei de Medicina Intensiva, Tortosa, Tarragona, Spain; AP-HP (Dr Launay), Hôpital Lariboisière, Service de Biochimie, Paris, France; INSERM

journal.publications.chestnet.org

(Dr Launay), Unité U942, Paris, France; AP-HP (Ms Charles-Nelson and Dr Katsahian), Hôpital Européen Georges Pompidou, Unité d’Épidémiologie et de Recherche Clinique, Paris, France; Sorbonne Universités (Ms Charles-Nelson and Dr Katsahian), UPMC INSERM UMRS1138, Centre de Recherche des Cordeliers, Paris, France; Keenan Research Centre and Critical Care Department (Dr Brochard), St. Michael’s Hospital, Toronto, ON, Canada; and Interdepartmental Division of Critical Care Medicine (Dr Brochard), University of Toronto, Toronto, ON, Canada. FUNDING/SUPPORT: This project was funded by the French Ministry of Health research program (Programme Hospitalier de Recherche Clinique [Contract No. 05104]). CORRESPONDENCE TO: Armand Mekontso Dessap, MD, PhD, Service de Réanimation Médicale, Centre Hospitalo-Universitaire Henri Mondor,

1231

Delirium is a common form of organ dysfunction in the ICU, with some studies showing that . 80% of noncomatose patients develop delirium at some point during the ICU stay.1 Patients who develop delirium have higher mortality and fewer days free of the ventilator than do those who never develop delirium.1 Delirium may prolong mechanical ventilation by interfering with weaning.2 Indeed, weaning contributes at least 40% of the total duration of mechanical ventilation, and successful weaning requires the preservation of brain function.3 However, delirium has received little attention during weaning from mechanical ventilation. Delirium may be associated with an alteration of sleep and circadian rhythm in patients in the ICU.4 Numerous risk factors for delirium, including sensory impairment, severe illness, sepsis, metabolic disorders, CNS injury, substance abuse and withdrawal, and oversedation, affect the sleep-wake cycle.5 Sleep deprivation, which is known to lead to many of the clinical and phys-

Materials and Methods Patients This study was performed in one of the nine participating centers (Henri Mondor University Hospital, Creteil, France) of the B-type Natriuretic Peptide for the Fluid Management of Weaning (BMW) trial,9 as an ancillary study (planed a priori). We explored delirium and melatonin excretion in all the 70 consecutive participants enrolled in this study. The BMW study was a randomized controlled trial comparing a biomarker-guided depletive fluid management strategy with usual care during ventilator weaning. A detailed description of the BMW study design has been published previously.9 Inclusion criteria were those allowing early initiation of ventilator weaning: mechanical ventilation through an endotracheal tube for at least 24 h, oxygen saturation ⱖ 90% with Fio2 ⱕ 50%, and positive end-expiratory pressure ⱕ 8 cm H2O; hemodynamic stability during the previous 12 h; sedation stopped or decreased over the previous 48 h; stable neurologic status with Ramsay score ⱕ 510; and a body temperature between 36.0°C and 39.0°C. Permanent noninclusion criteria were pregnancy or lactation, age , 18 years, known allergy to furosemide or sulfonamides, tracheostomy on inclusion, hepatic encephalopathy, cerebral edema, acute hydrocephalus, myasthenia gravis, acute idiopathic polyradiculoneuropathy, decision to withdraw life support, and prolonged cardiac arrest with a poor neurologic prognosis. Temporary noninclusion criteria were extubation scheduled on the same day, persistent acute right ventricular failure, renal insufficiency (defined as any of the following: need for renal replacement therapy, plasma urea . 25 mmol/L, plasma creatinine . 180 mmol/L or creatinine clearance , 30 mL/min, or . 25% increase in plasma creatinine over the previous 24 h), injection of iodinated contrast agent in the previous 6 h, blood sodium level . 150 mEq/L, blood potassium level , 3.5 mEq/L, or metabolic alkalosis with arterial pH . 7.50. When inclusion was delayed because of a temporary noninclusion criterion, 51, avenue du Mal de Lattre de Tassigny 94 010 Créteil Cedex, France; e-mail: [email protected] © 2015 AMERICAN COLLEGE OF CHEST PHYSICIANS. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.15-0525

1232 Original Research

iologic manifestations also found in delirium, is frequent in patients in the ICU6 and may be associated with delirium.4 To assess the circadian rhythm, melatonin metabolites can be measured. Melatonin is a major secretory product of the pineal gland, which plays an important physiologic role in sleep and circadian rhythm regulation.7 After metabolism and degradation of melatonin in the liver, the excretion of melatonin takes place in the kidneys as 6-sulfatoxymelatonin (aMT6s), which can be measured directly.7 Urine concentrations of aMT6s have been shown to correlate well with simultaneously sampled plasma levels of melatonin in healthy individuals.8 We performed the current study to examine the relationship between delirium and difficulties in weaning from mechanical ventilation. We also compared the circadian rhythm of melatonin excretion between patients with and without delirium.

enrollment was performed after correction of the abnormal value. The protocol was approved by our institution’s local ethics committee (Comité de Protection des Personnes Ile-de-France IX, approval number 06-035), and informed consent was signed by the patient or a close relative. The main result of the BMW trial was to show that a B-type natriuretic peptide (BNP)-guided depletive fluid management strategy decreased the duration of weaning without increasing adverse events.9 Delirium Assessment In all patients, neurologic status was assessed at the initiation of weaning (inclusion), then daily (up to successful weaning or 3 days after inclusion) and before extubation by an investigator not involved in the patient’s care (F. R.-C.). Patient status was defined as normal, delirious, or comatose using the Richmond Agitation-Sedation Scale (RASS)11 for arousal and the Confusion Assessment Method for the ICU (CAMICU) for delirium,12 as described previously.1 Briefly, the CAM-ICU assessment was positive if patients demonstrated an acute change or fluctuation in the course of their mental status (as determined by abnormalities or fluctuations in the RASS score), in addition to inattention and either disorganized thinking or an altered level of consciousness. By definition, patients were delirious if they responded to verbal stimulation with eye opening (RASS scores of 23 to 14) and had positive CAM-ICU assessments. Patients were defined as comatose if they responded to physical/painful stimulation with movement only but had no eye opening (RASS score, 24) or if they had no response to verbal or physical stimulation (RASS score, 25). We classified patients into the motoric subtypes based on the criteria used by Peterson et al13 (hypoactive, hyperactive, and mixed). Patients were defined as normal if they were not delirious or comatose. Circadian Rhythm of Melatonin Excretion Urine was collected through the indwelling urine catheter in 3-h intervals during the first 24 h of weaning in all included patients. Samples of 5 mL of urine were obtained from each urine portion by nurses during routine care and were stored at 280°C for later analysis. aMT6s was determined from diluted (1/250) urine samples by a radioimmunologic assay according to the manufacturer’s instructions (Stockgrand Ltd). The intra- and interassay coefficients of variation were 6.6% and 9.3%, respectively. The amount of aMT6s excretion was calculated as the product of aMT6s concentration by urine volume for each 3-h interval.

[

148#5 CHEST NOVEMBER 2015

]

Patient Management To standardize the weaning process, all patients were ventilated using a computer-driven automated weaning system (Evita Smart Care System, Dräger).14 Patients were assigned to one of two groups (BNP-guided depletive fluid management or usual care based on clinical evaluation). In the control group, the physicians were blinded to the BNP assay results, and all treatments, including diuretics, were administered according to usual care. In the interventional group, fluid intake was restricted, and furosemide was administered9 when the BNP level was ⱖ 200 pg/mL.15 In both groups, the automated weaning system gradually decreased the pressure support level during the weaning period while maintaining the patient within a zone of respiratory comfort, as described previously.14 When the patient was deemed “ready for separation” by the system, extubation was performed as soon as possible, provided the patient met the other criteria required for extubation.9 Assist-control ventilation was resumed during ventilation in case of respiratory worsening. The diagnosis of ventilator-associated pneumonia was based on usual criteria.16,17 Ventilator-associated complication was defined as the presence of at least one of the following two signs of worsening oxygenation: minimum daily Fio2 values increase ⱖ 20% for ⱖ 2 calendar days or minimum daily positive end-expiratory pressure values increase by ⱖ 3 cm H2O for ⱖ 2 calendar days.18 There was no systematic regimentation of sleep hygiene in the unit during the study period. The only three drugs used for sedation/analgesia were midazolam, fentanyl, and propofol. All patients were followed up until discharge from the hospital or until day 60 after inclusion, whichever occurred first. End Points The primary end point was the cumulative rate of successful extubation in patients with or without delirium. Secondary end points included the rate of complications during weaning, weaning, ventilator and hospitalization durations, and mortality in patients with or without delirium. We also assessed the circadian rhythm of melatonin excretion in patients with and without delirium.

Results Delirium and Weaning

A total of 70 patients were included in the study. At the initiation of weaning, 43 patients (61.4%) were delirious (14 with hypoactive subtype and 29 with mixed or hyperactive subtype), 24 (34.3%) had a normal mental status, and three (4.3%) were comatose (as assessed by the RASS). These three patients and one additional patient with normal mental status at initiation of weaning later developed delirium; thus, the overall prevalence of delirium during the entire weaning was 67.1% (47 patients). Figure 1 shows the number of patients in each of these three categories over the first 3 days of weaning and before extubation. Of the 43 initially delirious patients, seven died and 36 had survived at day 60; all but one survivor were weaned successfully. Of the 24 initially nondelirious patients, two died and 22 had survived at day 60; all survivors were weaned successfully. Table 1 shows the baseline characteristics of patients at the initiation of weaning. As compared with patients with a normal mental status, those with delirium at the initiation of weaning had higher Sequential Organ Failure Assessment scores at ICU admission

journal.publications.chestnet.org

Statistical Analysis Data were analyzed using SPSS Base 13 (IBM Corporation) and R 2.15.2 (The R Foundation for Statistical Computing). Categorical variables were expressed as percentages, and continuous data were expressed as median (25th-75th percentiles) unless otherwise specified. We used the x2 or Fisher exact test to compare categorical variables between groups and the Student t test or Mann-Whitney U test to compare continuous variables, as appropriate. Successful extubation was defined as the patient being alive and without reintubation 72 h after extubation.9 Tracheotomy weaning was defined as spontaneous breathing during 72 h without connection to the ventilator for ⱖ 3 h/d. The effect of delirium on the cumulative probability of successful extubation was assessed using the Kaplan-Meier method and log-rank statistics. To evaluate factors associated with complications during weaning or successful extubation by multivariate analysis, all patients’ characteristics at baseline, before weaning, and at the initiation of weaning were assessed, and significant or marginally significant univariate risk factors (P , .10) were examined using backward stepwise logistic regression or Cox proportional hazard regression, respectively. Coefficients were computed by the method of maximum likelihood. We also used a competing risk model (cumulative incidence function of the Gray model)19,20 to estimate the effect of delirium on weaning outcome, while considering death as a competing event. The strength of the association between each variable and the outcome was assessed using the subhazard ratio associated with the cumulative incidence function estimated using the cmprsk package developed by Gray in the R software (http://biowww.dfci.harvard. edu/~gray/cmprsk_2.1-4.tar.gz). The global effect of delirium and time on aMT6s urinary excretion and their potential interaction were assessed by two-way analysis of variance for repeated measures (general linear model), providing multivariate analyses for the repeated-measures data. Peak time, peak value, mean value, and amplitude (peak value minus the mean value) of aMT6s urinary excretion were compared between patients with delirium and those without at the initiation of weaning. Two-sided P values , .05 were considered significant.

and a greater prevalence of chronic alcohol consumption, shock, neuromuscular blockade, antibiotic treatment, and cumulative doses of sedative and narcotic drugs before weaning. Tables 2 and 3 show the complications and main outcomes of weaning, respectively. As compared with patients with normal mental status, those with delirium at the initiation of weaning exhibited more complications during weaning (40 [93%] vs 15 [63%], P 5 .002), including respiratory worsening requiring a return to assist-control ventilation, ventilator-associated pneumonia or complication, continuous sedation or analgesia because of clinical worsening, and need for fluid loading or catecholamine infusion. The cumulative doses of sedative, narcotic, and neuroleptic medications used during weaning were also higher in patients with delirium as compared with patients with a normal mental status at the initiation of weaning. The greater prevalence of complications in delirious patients persisted after multivariate analysis by logistic regression (OR 5 5.95; 95% CI, 1.26-28.13; P 5 .02 (e-Table 1). By Kaplan-Meier analysis, the probability of successful extubation was significantly lower with delirium at the 1233

normal mental status (Table 4). Interestingly, the mean value for aMT6s in delirious patients was below our laboratory normal value but was in the normal range for nondelirious patients. The urinary excretion profile of aMT6s did not differ according to delirium subtype (F statistic 5 0.57, P 5 .455).

Discussion

Figure 1 – Mental status of patients during weaning from mechanical ventilation. Patient status was defined as normal (white bars), delirious (gray bars), or comatose (black bars) using the Richmond AgitationSedation Scale for arousal and the Confusion Assessment Method for the ICU for delirium. Patients depicted are those ventilated with a computerdriven automated weaning system; the remainder are either extubated at that time (n 5 26 at d 2 and n 5 41 d 3) with or without subsequent reintubation, switched back to assist-control ventilation because of clinical worsening (n 5 1 at d 3 and n 5 1 at extubation d), or dead (n 5 5 at extubation d).

initiation of weaning (P 5 .02 for log-rank test) (Fig 2), and this difference persisted after multivariate analysis by Cox regression (hazard ratio [HR] 5 0.54; 95% CI, 0.30-0.95) (e-Table 2). The probability of successful extubation was still significantly reduced with delirium at the initiation of weaning while adjusting for death as a competing event, both in univariate and multivariate analysis (sub-HR 5 0.52; 95% CI, 0.29-0.93; P 5 .03; and sub-HR 5 0.55; 95% CI, 0.30-0.99; P 5 .048, respectively) (e-Fig 1, e-Table 3). These results were not impacted by the BMW randomization group, which yielded nonsignificant results (P . .10) in this patient sample by univariate analysis for logistic regression, Cox regression, and competing risk. Patients with and without delirium did not differ in terms of time to discharge and mortality in ICU or hospital but differed in terms of ventilator-free days at all time points (Table 3). aMT6s Urinary Excretion

aMT6s urinary excretion during the first 24 h following inclusion is displayed in Figure 3. By general linear model, both time interval (F statistic 5 11.68, P , .0001) and delirium (F statistic 5 5.81, P 5 .019) showed a significant correlation with aMT6s urinary excretion, with a significant interaction (F statistic 5 2.65, P 5 .019). No significant difference was noted concerning the peak time, but the peak value, mean value, amplitude, and total 24-h excretion of aMT6s were significantly reduced in patients with delirium as compared with those with 1234 Original Research

To our knowledge, this is the first report to assess the effect of delirium on the outcome of weaning from mechanical ventilation. A majority of patients were delirious at weaning initiation, and delirium was associated with more neurologic and respiratory complications and a prolongation of weaning, with less ventilator-free days up to day 60. Circadian rhythm, as assessed by aMT6s urinary excretion, was more altered in delirious patients as compared with those with a normal mental status. Delirium During Weaning

The global prevalence of delirium during weaning was high (67.1%) in our study. This finding is consistent with previous reports showing that delirium may occur in up to 80% of patients during the course of critical illness.21 A recent study showed that delirium occurs predominantly during the weaning phase in patients receiving mechanical ventilation.22 Delirium at the initiation of weaning was associated with more complications (mainly respiratory and neurologic) during weaning and with a prolongation of weaning in univariate and multivariate analyses in our study. These findings confirm previous reports on the prolongation of ventilator use in delirious patients1,23,24 and suggest that delirium is specifically associated with weaning difficulties. Weaning is a complex and stressful25 process that requires patients’ cooperation. Other mental diseases, like depression, were shown to be associated with complicated weaning.26 In our study, patients with delirium at the initiation of weaning had notorious risk factors for delirium, including a greater severity of illness, treatment with sedative and narcotic drugs,21 and alcoholism.27 Differentiating alcohol withdrawal syndrome from other causes of delirium may not be easy.28 The greater prevalence of antibiotic and neuromuscular blockade use among patients with delirium may be related to their greater severity of illness, although a direct effect of some antibiotics on mental status cannot be ruled out.29 Circadian Rhythm of Melatonin Excretion

Melatonin is produced by the pineal gland and is involved in the maintenance of circadian rhythm and

[

148#5 CHEST NOVEMBER 2015

]

TABLE 1

] Characteristics of Patients at ICU Admission and Initiation of Weaning From Mechanical Ventilation Mental Status at Initiation of Weaninga Normal (n 5 24) 63 (46-76)

60 (48-75)

.814

Male sex

15 (62.5)

29 (67.4)

.683

SAPS II at ICU admission

39 (32-50)

49 (36-57)

.095

SOFA score at ICU admission

5.5 (4.0-7.8)

Delirious (n 5 43)

P Value

Characteristic Age, y

8.0 (6.0-11.0)

MacCabe and Jackson class No fatal disease

10 (41.7)

28 (65.1)

Rapidly fatal disease

12 (50.0)

11 (25.6)

Ultimately fatal disease Chronic alcohol consumption

.010 .123

2 (8.3)

4 (9.3)

1 (4.2)

11 (25.6)

.044

8 (33.3)

19 (44.2)

.385

Cardiopulmonary disease LVD

10 (41.7)

21 (48.8)

.572

Coronary heart disease

Hypertension

3 (12.5)

10 (23.3)

.350

COPD

3 (12.5)

13 (30.2)

.103

Sleep apnea

2 (8.3)

5 (11.6)

.673

Smoking

9 (37.5)

24 (55.8)

.151

Reason for intubation

.579

Coma

3 (12.5)

9 (20.9)

Septic shock

2 (8.3)

6 (14.0)

Cardiogenic pulmonary edema

5 (20.8)

12 (27.9)

Pneumonia

9 (37.5)

9 (20.9)

Cardiac arrest

1 (4.2)

3 (7.0)

Surgery

4 (16.7)

4 (9.3)

Events between ICU admission and weaning Shockb

13 (54.2)

36 (83.7)

.009

Ventilator-associated pneumonia

3 (12.5)

11 (25.6)

.207

ARDSb

7 (29.2)

18 (41.9)

.303

Use of neuromuscular blockers

1 (4.2)

11 (25.6)

.044

Steroid treatment Antibiotic treatment Duration of invasive mechanical ventilation, d

7 (29.2)

18 (41.9)

.303

19 (79.2)

42 (97.7)

.011

2.8 (1.6-6.9)

4.1 (2.6-7.4)

.133

120 (21-385)

.010

Total dose of CNS drugs administered before weaning, mg Midazolam Median (IQR) Mean (SD)

25 (0-164) 106 (169)

282 (473)

Median (IQR)

1.0 (0.0-2.4)

5.3 (2.4-13.2)

Mean (SD)

3.3 (6.3)

…

Fentanyl

11.1 (15.1)

.0001 …

Propofol Median (IQR) Mean (SD) No. of days with continuous sedation

0 (0-0) 489 (1,250) 1.0 (1.0-2.8)

0 (0-2,450) 2,765 (6,736) 3.0 (2.0-6.3)

.043 … .005 (Continued)

journal.publications.chestnet.org

1235

TABLE 1

] (continued) Mental Status at Initiation of Weaninga

Characteristic

Normal (n 5 24)

Delirious (n 5 43)

P Value

4.0 (2.3-5.0)

5.0 (3.0-6.0)

.069

Clinical characteristics at initiation of weaning SOFA score at initiation of weaning Temperature at initiation of weaning, °C

36.8 (36.5-37.6)

37.1 (36.7-37.6)

Randomization arm

.342 .676

Usual-care arm

11 (45.8)

22 (51.2)

B-type natriuretic peptide arm

13 (54.2)

21 (48.8)

12 (10-16)

12 (10-15)

Ventilation settings at initiation of weaning Pressure support level, cm H2O PEEP level, cm H2O PaO2/FIO2 ratio, mm Hg PaCO2, mm Hg pH

5 (5-6)

5 (5-6)

228 (183-290)

198 (164-260)

40 (35-50)

39 (34-45)

.597 .478 .128 .413

7.46 (7.39-7.48)

7.45 (7.40-7.47)

.351

140 (136-145)

141 (137-143)

.953

Blood chemistry at initiation of weaning Blood sodium, mEq/L Plasma creatinine, mmol/L Blood glucose, mg/dL

82 (63-97) 119 (101-148)

74 (56-102)

.476

131 (115-149)

.252

Data are presented as No. (%) or median (IQR) unless indicated otherwise. IQR 5 interquartile range; LVD 5 left ventricular systolic dysfunction; PEEP 5 positive end-expiratory pressure; SAPS 5 Simplified Acute Physiology Score; SOFA 5 Sequential Organ Failure Assessment. aThree patients were comatose at initiation of weaning according to the Richmond Agitation-Sedation Scale and were excluded from analysis. bAt admission or later during the ICU stay.

the sleep-wake cycle.30 The melatonin rhythm (as evidenced by its profile in plasma, or its major metabolite, aMT6s in urine) is currently the most commonly used index of circadian pacemaker in humans.31 Melatonin levels in human plasma usually begin to increase between 6:00 pm and 8:00 pm and peak between midnight and 5:00 am,31 being followed by a rapid decrease. Interestingly, the 24-h profile of aMT6s excretion exhibited a phase delay in both groups of patients, with and without delirium, which may not be explained fully by the time interval for hepatic processing and urinary excretion.32 This result is in accordance with previous findings in patients in the ICU33 and with a study showing a temporal disorganization of circadian rhythmicity in patients who are mechanically ventilated.34 Patients in the ICU have limited natural and phasic light exposure, which is the main external signal for maintaining circadian rhythm. In addition, sepsis, systemic inflammatory response, hormone interactions, medications (including opioids and benzodiazepines), acuity of illness, burn, and mechanical ventilation have been shown to alter melatonin excretion.35-38 Many of these factors were prevalent in the patients with delirium in our study, who also had a more severe alteration of

1236 Original Research

circadian rhythm of melatonin excretion (significantly reduced peak value, mean value, amplitude, and total 24-h excretion of aMT6s) than did those without delirium. Delirium after cardiac surgery has been associated with decreased levels of tryptophan, the melatonin precursor.39 Alteration in melatonin secretion may contribute to delirium in patients in the ICU through various mechanisms including impaired sleep.40 Sleep deprivation and delirium share many features: (1) They affect the majority of patients in the ICU,1 (2) they have overlapping risk factors (eg, benzodiazepine and opiate use)41 and common hallmarks (eg, inattention, fluctuating mental status, and cognitive dysfunction),42 and (3) they affect the same regions of the CNS (including the prefrontal and parietal cortexes) and involve common neurohormonal changes in the brain (including decreased acetylcholine or increased dopamine levels).4 Sleep deprivation has been associated with mental status change in critically ill patients43 and with occurrence of delirium after surgery.44 Poor sleep quality was associated with delirium and subsequent failure of noninvasive ventilation in patients with acute hypercapnic respiratory failure.45

[

148#5 CHEST NOVEMBER 2015

]

TABLE 2

] Complications Recorded During Weaning From Mechanical Ventilation Mental Status at Initiation of Weaninga

Complication

P Value

Normal (n 5 24)

Delirious (n 5 43)

Respiratory worsening requiring a return to assist-control ventilation

5 (20.8)

21 (48.8)

.024

Ventilator-associated pneumonia

2 (8.3)

13 (30.2)

.039

Ventilator-associated complication

1 (4.2)

11 (25.6)

.044

Respiratory function

Need for noninvasive ventilation after extubation Reintubation within 72 h after extubation Tracheostomy

11 of 23 (47.8)

16 of 38 (42.1)

3 of 23 (13.0)

6 of 38 (15.8)

1 (4.2)

.663 . .99

5 (11.6)

.408

Cardiovascular function Need for fluid loading

7 (29.2)

23 (53.5)

.055

Need for catecholamine infusion

9 (37.5)

21 (48.8)

.371

Need for continuous sedation because of clinical worsening

2 (8.3)

30 (69.8)

, .001

Need for continuous analgesia because of clinical worsening

1 (4.2)

24 (55.8)

, .001

Median (IQR)

0 (0-0)

0 (0-72)

Mean (SD)

1 (4)

Neurologic function

Total dose of CNS drugs during weaning, mg Midazolam

, .001

182 (478)

Fentanyl

, .001

Median (IQR)

0.0 (0.0-0.0)

0.8 (0.0-9.8)

Mean (SD)

0.0 (0.0)

6.9 (12.7)

Propofol Median (IQR) Mean (SD)

, .001 0 (0-0) 23 (110)

240 (0-1,450) 2,655 (6,742)

Levomepromazin

, .001

Median (IQR)

0 (0-0)

Mean (SD)

0 (0)

0 (0-400) 190 (276)

Data are given as No. (%) unless otherwise indicated. See Table 1 for expansion of abbreviation. Three patients were comatose at initiation of weaning according to the Richmond Agitation-Sedation Scale and were excluded from analysis.

a

Clinical Implications

It cannot be concluded from our study that better management of delirium independent of underlying disease management will necessarily translate into more rapid removal of ventilator support. However, a better control of delirium, through an optimized use of sedative agents, early mobilization, or other measures may help hasten the weaning process.46 In addition, the potential role of melatonin in the development of delirium encourages testing the effects of melatonin supplementation to improve sleep quality and reduce delirium in patients in the ICU. Studies of melatonin supplementation in acutely ill patients are scarce, with heteroge-

journal.publications.chestnet.org

neous methodologies47 but encouraging results48; future well-designed trials in those who are mechanically ventilated are warranted. Strengths and Limitations

The strengths of our study include (1) the standardization of the weaning process (which was automated using a closed loop system); (2) the early assessment of delirium, starting at the initiation of weaning; and (3) the blind assessment of aMT6s urinary excretion. However, our study has limitations. First, we did not explore sleep in the patients, and aMT6s excretion was assessed only during 24 h. Second, the ancillary nature and

1237

TABLE 3

] Main Outcomes of Weaning From Mechanical Ventilation Mental Status at Initiation of Weaninga

Outcome

Normal (n 5 24)

Delirious (n 5 43)

Time to first extubation, h

P Value .086

Median (IQR)

27.5 (21.7-62.0)

Mean (SD)

43.0 (29.6)

49.7 (21.4-137.6) 111.7 (176.6)

Time to successful extubation, h

.086

Median (IQR)

27.4 (21.1-72.2)

Mean (SD)

56.4 (59.3)

51.0 (21.9-143.5) 164.8 (255.3)

Ventilator-free days from randomization to d 28

.044

Median (IQR)

26.6 (23.7-27.1)

23.1 (0.8-27.1)

Mean (SD)

23.5 (7.6)

17.7 (11.5)

Median (IQR)

58.6 (55.7-59.1)

54.4 (25.2-58.9)

Mean (SD)

52.9 (16.4)

41.7 (23.8)

Ventilator-free days from randomization to d 60

.027

Time from randomization to ICU discharge,b d

.314

Median (IQR)

7.0 (4.8-11.0)

Mean (SD)

9.2 (6.5)

9.0 (5.0-16.5) 18.2 (20.3)

Time from inclusion to hospital discharge,b d

.350

Median (IQR)

24.0 (16.5-38.8)

29.5 (18.3-59.5)

Mean (SD)

29.1 (17.4)

34.3 (19.0)

ICU mortality,c No. (%)

2 (8.3)

6 (14.0)

.701

Hospital mortality, No. (%)

2 (8.3)

7 (16.3)

.472

D 60 mortality,c No. (%)

2 (8.3)

7 (16.3)

.472

c

See Table 1 for expansion of abbreviation. Three patients were comatose at initiation of weaning according to the Richmond Agitation-Sedation Scale and were excluded from analysis. bPatients still in ICU or in hospital at last follow-up (d 60) were attributed a length of stay of 60 d in ICU or hospital, respectively; patients deceased in ICU or hospital were excluded from analysis. cMortality analyses are unadjusted for differences in severity or mortality risk. a

monocentric design of this study and the specific inclusion and exclusion criteria used in the BMW trial may have limited the generalizability of our findings; how-

Figure 2 – Probability of successful extubation within 60 d after inclusion (Kaplan-Meier model). *Three patients were comatose at initiation of weaning according to the Richmond Agitation-Sedation Scale and were excluded from analysis.

1238 Original Research

ever, we included a mixed population of medical and surgical patients with various reasons for intubation. Third, patients were placed in single-bed or two-bed rooms depending on availability, and ICU architectural

Figure 3 – Twenty-four-hour profiles of urinary aMT6s excretion in patients with or without delirium at the initiation of weaning. Values are means, and error bars represent SDs. aMT6s 5 6-sulfatoxymelatonin.

[

148#5 CHEST NOVEMBER 2015

]

TABLE 4

] 6-Sulfatoxymelatonin Excretion on the First Day of Weaning From Mechanical Ventilation Mental Status at Initiation of Weaninga

Parameter

Normal (n 5 24)

Delirious (n 5 43)

Peak time, h:min

P Value .663

Median (IQR)

8:00 (5:00-11:00)

8:00 (5:00-11:00)

Mean (SD)

9:00 (4:49)

8:25 (5:25)

Peak value, ng/3 h

.007

Median (IQR)

6,755 (4,891-12,841)

5,550 (3,250-7,980)

Mean (SD)

9,539 (7,051)

6,020 (3,389)

Amplitude,b ng/3 h

.007

Median (IQR)

4,366 (2,952-7,645)

3,177 (1,710-5,337)

Mean (SD)

6,152 (5,145)

3,620 (2,305)

Mean value, ng/3 h

.020

Median (IQR)

2,526 (1,777-4,990)

Mean (SD)

3,387 (2,106)

2360 (1,455-3,416) 2,400 (1,289)

Total 24-h excretion, ng

.023

Median (IQR)

20,212 (13,207-39,920)

18,880 (11,642-27,325)

Mean (SD)

26,941 (16,984)

19,177 (10,337)

Normal mean (SD) values obtained in our laboratory in 12 healthy individuals were as follows: peak time (h:min): 4:25 (2:05); peak value: 6,750 (1,284) ng/3 h; amplitude: 3,804 (1,148) ng/3 h; mean value: 2,946 (1,071) ng/3 h; and total 24-h excretion: 23,571 (9,187) ng. Mean values of 6-sulfatoxymelatonin excretion were significantly lower in delirious patients as compared with healthy individuals (P , .01). See Table 1 for expansion of abbreviation. aThree patients were comatose at initiation of weaning according to the Richmond Agitation-Sedation Scale and were excluded from analysis. bAmplitude denotes peak value minus mean value.

design has been shown recently to affect delirium prevalence.49 Last, because the vast majority of delirious patients had been affected since the initiation of weaning, we could not assess the effect of circadian rhythm alteration on subsequent mental status change during weaning.

Conclusions In conclusion, delirium was common at the initiation of ventilator weaning and was associated with a prolonged

journal.publications.chestnet.org

weaning process. aMT6s urinary excretion showed a greater alteration of circadian rhythm in delirious patients, suggesting a possible role for sleep deprivation in delirium pathophysiology. Further studies are needed to test whether melatonin supplementation may improve the sleep-wake cycle, reduce delirium onset, and accelerate weaning from the ventilator in patients in the ICU.

1239

Acknowledgments Author contributions: A. M. D. had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis, including and especially any adverse effects. A. M. D., F. R.-C., and L. B. contributed to the conception and design of the study; A. M. D. and F. R.-C. contributed to the recruitment of the patients and data collection; A. M. D., F. R.-C., A. C.-N., S. K., C. B.-B., and L. B. contributed to the data analysis and interpretation; A. M. D., F. R.-C., C. B.-B., and L. B. contributed to the drafting of the report; all authors contributed to review and revision of the report; and all have seen and approved the final version. Conflict of interest: L. B. has been a consultant for Dräger, and his research laboratory has received research grants from Covidien Ltd, GE, Dräger, and Vygon SA. None declared (A. M. D., F. R.-C., J.-M. L., A. C.-N., S. K., C. B.-B.). Role of sponsors: Supported and promoted by the French publicly funded hospital clinical research program (Program Hospitalier de Recherche Clinique). Biosite France supplied the BNP assay devices and kits (Triage MeterPlus) for the BMW study. Drager Medical provided the AWS-equipped ventilators for the BMW study. Additional information: The e-Figure and e-Tables can be found in the Supplemental Materials section of the online article.

References 1. Ely EW, Shintani A, Truman B, et al. Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit. JAMA. 2004;291(14):1753-1762. 2. Esteban A, Alía I, Ibañez J, Benito S, Tobin MJ; The Spanish Lung Failure Collaborative Group. Modes of mechanical ventilation and weaning. A national survey of Spanish hospitals. Chest. 1994; 106(4):1188-1193. 3. Boles JM, Bion J, Connors A, et al. Weaning from mechanical ventilation. Eur Respir J. 2007;29(5):1033-1056. 4. Weinhouse GL. Delirium and sleep disturbances in the intensive care unit: can we do better? Curr Opin Anaesthesiol. 2014;27(4):403-408. 5. Maldonado A, Bauer TT, Ferrer M, et al. Capnometric recirculation gas tonometry and weaning from mechanical ventilation. Am J Respir Crit Care Med. 2000;161(1):171-176. 6. Drouot X, Roche-Campo F, Thille AW, et al. A new classification for sleep analysis in critically ill patients. Sleep Med. 2012;13(1):7-14. 7. Mirick DK, Davis S. Melatonin as a biomarker of circadian dysregulation. Cancer Epidemiol Biomarkers Prev. 2008;17(12):3306-3313. 8. Bojkowski CJ, Arendt J, Shih MC, Markey SP. Melatonin secretion in humans assessed by measuring its metab-

1240 Original Research

olite, 6-sulfatoxymelatonin. Clin Chem. 1987;33(8):1343-1348.

tilated medical patients. Crit Care Med. 2005;33(6):1260-1265.

9. Mekontso Dessap A, Roche-Campo F, Kouatchet A, et al. Natriuretic peptidedriven fluid management during ventilator weaning: a randomized controlled trial. Am J Respir Crit Care Med. 2012;186(12):1256-1263.

24. Mehta S, Cook D, Devlin JW, et al; SLEAP Investigators; Canadian Critical Care Trials Group. Prevalence, risk factors, and outcomes of delirium in mechanically ventilated adults. Crit Care Med. 2015;43(3):557-566.

10. Ramsay MA, Savege TM, Simpson BR, Goodwin R. Controlled sedation with alphaxalone-alphadolone. BMJ. 1974;2(5920):656-659.

25. Riggio RE, Singer RD, Hartman K, Sneider R. Psychological issues in the care of critically-ill respirator patients: differential perceptions of patients, relatives, and staff. Psychol Rep. 1982;51(2):363-369.

11. Sessler CN, Gosnell MS, Grap MJ, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med. 2002;166(10):1338-1344.

26. Jubran A, Lawm G, Kelly J, et al. Depressive disorders during weaning from prolonged mechanical ventilation. Intensive Care Med. 2010;36(5):828-835.

12. Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilated patients: validity and reliability of the confusion assessment method for the intensive care unit (CAM-ICU). JAMA. 2001;286(21):2703-2710.

27. Ouimet S, Kavanagh BP, Gottfried SB, Skrobik Y. Incidence, risk factors and consequences of ICU delirium. Intensive Care Med. 2007;33(1):66-73. 28. Sullivan JT, Sykora K, Schneiderman J, Naranjo CA, Sellers EM. Assessment of alcohol withdrawal: the revised clinical institute withdrawal assessment for alcohol scale (CIWA-Ar). Br J Addict. 1989;84(11):1353-1357.

13. Peterson JF, Pun BT, Dittus RS, et al. Delirium and its motoric subtypes: a study of 614 critically ill patients. J Am Geriatr Soc. 2006;54(3):479-484. 14. Lellouche F, Mancebo J, Jolliet P, et al. A multicenter randomized trial of computer-driven protocolized weaning from mechanical ventilation. Am J Respir Crit Care Med. 2006;174(8):894-900.

29. Lertxundi U, Palacios RH, Gutierrez FC, Echaburu SD, Garcia MG, Gomez CA. Levofloxacin-induced delirium in a patient suffering from schizoaffective disorder and multiple sclerosis. Curr Drug Saf 2013;8(3):199-200.

15. Mekontso-Dessap A, de Prost N, Girou E, et al. B-type natriuretic peptide and weaning from mechanical ventilation. Intensive Care Med. 2006;32(10):1529-1536.

30. Shigeta H, Yasui A, Nimura Y, et al. Postoperative delirium and melatonin levels in elderly patients. Am J Surg. 2001;182(5):449-454.

16. Chastre J, Fagon JY. Ventilatorassociated pneumonia. Am J Respir Crit Care Med. 2002;165(7):867-903.

31. Benloucif S, Burgess HJ, Klerman EB, et al. Measuring melatonin in humans. J Clin Sleep Med. 2008;4(1):66-69.

17. Mekontso Dessap A, Katsahian S, RocheCampo F, et al. Ventilator-associated pneumonia during weaning from mechanical ventilation: role of fluid management. Chest. 2014;146(1):58-65.

32. Crasson M, Kjiri S, Colin A, et al. Serum melatonin and urinary 6-sulfatoxymelatonin in major depression. Psychoneuroendocrinology. 2004;29(1):1-12.

18. Magill SS, Klompas M, Balk R, et al. Developing a new, national approach to surveillance for ventilator-associated events: executive summary. Chest. 2013;144(5):1448-1452.

33. Mundigler G, Delle-Karth G, Koreny M, et al. Impaired circadian rhythm of melatonin secretion in sedated critically ill patients with severe sepsis. Crit Care Med. 2002;30(3):536-540.

19. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999; 94(446):496-509.

34. Gehlbach BK, Chapotot F, Leproult R, et al. Temporal disorganization of circadian rhythmicity and sleep-wake regulation in mechanically ventilated patients receiving continuous intravenous sedation. Sleep. 2012;35(8):1105-1114.

20. Gray RJ. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16(3):1141-1154.

35. Srinivasan V, Pandi-Perumal SR, Spence DW, Kato H, Cardinali DP. Melatonin in septic shock: some recent concepts. J Crit Care. 2010;25(4):656.e1-656.e6.

21. Maldonado JR. Pathoetiological model of delirium: a comprehensive understanding of the neurobiology of delirium and an evidence-based approach to prevention and treatment. Crit Care Clin. 2008;24(4):789-856.

36. Verceles AC, Silhan L, Terrin M, Netzer G, Shanholtz C, Scharf SM. Circadian rhythm disruption in severe sepsis: the effect of ambient light on urinary 6-sulfatoxymelatonin secretion. Intensive Care Med. 2012;38(5):804-810.

22. Leite MA, Osaku EF, Costa CR, et al. Delirium during weaning from mechanical ventilation. Crit Care Res Pract. 2014; 2014:546349.

37. Gögenur I, Middleton B, Kristiansen VB, Skene DJ, Rosenberg J. Disturbances in melatonin and core body temperature circadian rhythms after minimal invasive surgery. Acta Anaesthesiol Scand. 2007;51(8):1099-1106.

23. Micek ST, Anand NJ, Laible BR, Shannon WD, Kollef MH. Delirium as detected by the CAM-ICU predicts restraint use among mechanically ven-

[

148#5 CHEST NOVEMBER 2015

]

38. Pina G, Brun J, Tissot S, Claustrat B. Long-term alteration of daily melatonin, 6-sulfatoxymelatonin, cortisol, and temperature profiles in burn patients: a preliminary report. Chronobiol Int. 2010;27(2):378-392. 39. van der Mast RC, van den Broek WW, Fekkes D, Pepplinkhuizen L, Habbema JD. Is delirium after cardiac surgery related to plasma amino acids and physical condition? J Neuropsychiatry Clin Neurosci. 2000;12(1):57-63. 40. Fitzgerald JM, Adamis D, Trzepacz PT, et al. Delirium: a disturbance of circadian integrity? Med Hypotheses. 2013;81(4): 568-576. 41. Bourne RS, Mills GH. Sleep disruption in critically ill patients–pharmacological considerations. Anaesthesia. 2004;59(4): 374-384.

journal.publications.chestnet.org

42. Harrison Y, Horne JA, Rothwell A. Prefrontal neuropsychological effects of sleep deprivation in young adults–a model for healthy aging? Sleep. 2000; 23(8):1067-1073. 43. Helton MC, Gordon SH, Nunnery SL. The correlation between sleep deprivation and the intensive care unit syndrome. Heart Lung. 1980;9(3):464-468.

46. Barr J, Fraser GL, Puntillo K, et al; American College of Critical Care Medicine. Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med. 2013;41(1):263-306. 47. Bellapart J, Boots R. Potential use of melatonin in sleep and delirium in the critically ill. Br J Anaesth. 2012;108(4):572-580.

44. Yildizeli B, Ozyurtkan MO, Batirel HF, Kuşcu K, Bekiroğlu N, Yüksel M. Factors associated with postoperative delirium after thoracic surgery. Ann Thorac Surg. 2005;79(3):1004-1009.

48. Hatta K, Kishi Y, Wada K, et al; DELIRIA-J Group. Preventive effects of ramelteon on delirium: a randomized placebo-controlled trial. JAMA Psychiatry. 2014;71(4):397-403.

45. Roche Campo F, Drouot X, Thille AW, et al. Poor sleep quality is associated with late noninvasive ventilation failure in patients with acute hypercapnic respiratory failure. Crit Care Med. 2010;38(2):477-485.

49. Caruso P, Guardian L, Tiengo T, Dos Santos LS, Junior PM. ICU architectural design affects the delirium prevalence: a comparison between single-bed and multibed rooms*. Crit Care Med. 2014;42(10):2204-2210.

1241