Biochemical markers for the circadian rhythm were studied in patients treated at the ICU (intensive care unit) of two regional hospitals. A normal rhythm is characterized by a relatively higher melatonin and a lower cortisol excretion at night. Disturbances affect sleep, mood and cognitive performance. All urine excreted between 07:00 and 22:00 hours (day) and between 22:00 and 07:00 hours (night) was collected and sampled throughout the entire ICU period (median, 10 days) in 16 patients for the excretion of 6-SMT (6-sulphatoxymelatonin), which is a metabolite of melatonin, and free cortisol. The overall excretion of 6-SMT was slightly lower and the cortisol excretion higher than reported for healthy reference populations. Mechanical ventilation was associated with a markedly lower 6-SMT excretion (median, 198 ng/h) compared with periods without such help (555 ng/h; P<0.0001), whereas infusion of adrenergic drugs increased the 6-SMT excretion (P<0.01). Five patients (31%) showed a virtually absent melatonin excretion for 24 h or more. The diurnal rhythms were consistently or periodically disturbed in 65% and 75% of the patients. These alterations cannot be explained by excessive exposure to light at night. In conclusion, there was hyposecretion of melatonin during mechanical ventilation, an overall high cortisol excretion and a disturbed diurnal rhythm of both of these hormones in most patients treated in two ICU departments.

INTRODUCTION

Sleep-wake disturbances are frequently recognized as a problem for patients in the ICU (intensive care unit). Neurophysiological methods demonstrate sleep fragmentation and suppression of Stage 3, Stage 4 and REM (rapid eye movement) sleep [14]. Therapeutic procedures, personal care and nurse–patient communication may cause such disturbances [1] and the illness is probably another major cause [2,5].

More recently, the pattern of melatonin secretion has been associated with sleeping disturbances in the ICU. Both melatonin and cortisol are biological markers of the circadian rhythm, the former showing a peak at night and the latter at noon. Shilo et al. [6] measured the urinary concentration of a melatonin metabolite, 6-SMT (6-sulphatoxymelatonin), during 24 h in 14 ICU patients. They found a disturbed melatonin excretion pattern in all patients, and most of them showed no nocturnal rise. Mundigler et al. [7] reported a high 6-SMT excretion without nocturnal peak in ICU patients with sepsis, whereas the nocturnal peak was preserved after transthoracic oesophagectomy [8]. However, no previous studies have followed the biochemical markers of the circadian cycle for more than a few days in ICU patients.

In the present study, we measured melatonin and cortisol excretion throughout the entire ICU stay (median, 10 days) in 16 patients. The purpose was to identify factors that correlate with the rate of excretion of these hormones and their diurnal variation. Disturbances of melatonin secretion promote fatigue and, in turn, poor co-operation with the nursing staff during mobilization.

METHODS

Between November 2000 and October 2001, the circadian rhythm was assessed in 16 patients (seven females and nine males) aged between 41 and 88 (median, 71) years throughout a period of intensive care at two Swedish hospitals, South Hospital in Stockholm and the Regional Hospital in Gällivare. After Ethics Committee approval, inclusion in the study was effected within 48 h (usually <24 h) after admission to the ICU. The patient gave his/her informed consent before inclusion or, if possible, on a later occasion. The criterion for inclusion was an expected stay at the ICU of at least 2 days. Patients with a markedly impaired kidney function (serum creatinine >250 μmol/l) were excluded. The severity of illness was assessed using the APACHE II (Acute Physiology And Chronic Health Evaluation 2) scoring system (Table 1).

Table 1
Demographic data and APACHE II score on admission to the ICU

F, female; M, male.

No.SexAgeDiagnosisAPACHE II scoreDays in ICUSurvival
62 Septic shock after gastrointestinal surgery 22 32 No 
86 SIRS after acute aortic aneurysm surgery 24 47 Yes 
87 Sepsis+abdominal pain and previous myocardial infarction 23 30 Yes 
58 Multiple rib fractures, pneumothorax and subdural haematoma 26 10 Yes 
68 Sepsis after ileus operation 17 11 Yes 
76 Re-operation for suture rupture after gastrectomy 17 10 Yes 
41 Guillan–Barrés paralysis 34 Yes 
44 Pneumonia 12 14 Yes 
70 Aortic aneurysm and chronic obstructive lung disease Yes 
10 46 Transferred from another hospital 2 weeks after multi-trauma Yes 
11 78 Lung and kidney insufficiency after aortic aneurysm surgery. Previous dialysis 17 Yes 
12 74 Spontaneous bladder and colonic perforation, sutured surgically 18 12 No 
13 79 Myocardial infarction and cardiogenic shock 24 No 
14 83 Cerebrovascular lesion with aspiration pneumonia 21 13 No 
15 61 Pneumonia and chronic obstructive pulmonary disease 15 Yes 
16 88 Lung embolus after surgery for acute fracture of the femur 25 Yes 
No.SexAgeDiagnosisAPACHE II scoreDays in ICUSurvival
62 Septic shock after gastrointestinal surgery 22 32 No 
86 SIRS after acute aortic aneurysm surgery 24 47 Yes 
87 Sepsis+abdominal pain and previous myocardial infarction 23 30 Yes 
58 Multiple rib fractures, pneumothorax and subdural haematoma 26 10 Yes 
68 Sepsis after ileus operation 17 11 Yes 
76 Re-operation for suture rupture after gastrectomy 17 10 Yes 
41 Guillan–Barrés paralysis 34 Yes 
44 Pneumonia 12 14 Yes 
70 Aortic aneurysm and chronic obstructive lung disease Yes 
10 46 Transferred from another hospital 2 weeks after multi-trauma Yes 
11 78 Lung and kidney insufficiency after aortic aneurysm surgery. Previous dialysis 17 Yes 
12 74 Spontaneous bladder and colonic perforation, sutured surgically 18 12 No 
13 79 Myocardial infarction and cardiogenic shock 24 No 
14 83 Cerebrovascular lesion with aspiration pneumonia 21 13 No 
15 61 Pneumonia and chronic obstructive pulmonary disease 15 Yes 
16 88 Lung embolus after surgery for acute fracture of the femur 25 Yes 

The illumination in the patient's room was monitored in the first seven patients by a Universal Photometer model S2 (Hagner, Solna, Sweden). Continuous measurements were made at a frequency of 0.5/min for up to 1 week. Photometer data were stored in a SatelLite U universal logger 4232 (Mitec Instrument AB, Säffle, Sweden) and then analysed on a standard PC. The light detector was placed as close to the upper part of the patient's bed as possible, but so as not to interfere with the daily care of the patient. The position and the angle of the detector were adjusted to mimic the illumination at the patient's eyes. The resulting curves were analysed to ensure that a day and night illumination pattern had been maintained for the patients studied.

Melatonin secretion is normally increased at bedtime and remains high until early morning. In contrast, cortisol secretion falls. Both hormones express a circadian rhythm which is driven by an endogenous oscillator situated in the hypothalamus [9]. To discriminate between the day/night phases of melatonin and cortisol excretion, the patients had an indwelling urinary bladder catheter from which all urine was collected between 07:00 and 22:00 hours (day) and between 22:00 and 07:00 hours (night) during the entire stay in the ICU (see Table 1). The catheter was wrapped throughout its length with black plastic to protect the urine from light. The volume was measured (median excretion rate, 106 ml/h) and two samples were frozen to −18°C for later analysis.

In the first sample, the concentration of a metabolite of melatonin, 6-SMT, was measured by RIA using an 125I-labelled tracer and antibody against sheep 6-SMT [10,11] (Stockgrand Ltd, Guildford, Surrey, U.K.). The urinary excretion of 6-SMT correlates closely with the melatonin concentration in plasma [12,13], even in the presence of a low creatinine clearance [14], and represents the average plasma level during the period of urine collection. A reference population excreted 400 ng/h 6-SMT between 07:00 and 22:00 hours, and 1500 ng/h between 22:00 and 07:00 hours [6], which is similar to patients (age, 56 years) with chronic disease who are free from acute illness [7]. Age-matched controls yielded a slightly lower reference value, 350 ng/h, for an entire 24 h period [14].

The second urine sample was analysed by an immunoassay in which the sampled cortisol competes with horseradish peroxidase-labelled cortisol for a limited number of binding sites [15] (Nordic Biosite, Täby, Sweden). The product of urine volume and cortisol concentration measured in this way provides a good approximation of the metabolically active cortisol fraction in plasma. According the manufacturer of the kit, the normal range for the cortisol excretion is 0.5–8 μg/h.

Both assays had a relative specificity of 100% for the measured substrate. The coefficient of variation was 3.5% or less, and all samples were analysed in duplicate, the mean value being used in the calculations.

The results of the 6-SMT and cortisol analyses were expressed as the median and interquartile range for the excreted amounts/h of day or night. Statistics were used to test for differences in the rates depending on the use of mechanical ventilation [assisted/controlled ventilation or CPAP (continuous positive airway pressure)], administration of certain drugs (adrenergic agents, propofol, benzodiazepines, opiates and cortisol) and the development of fever. Univariate analysis was performed using the Mann–Whitney or Kruskal–Wallis tests. Multivariate analyses comprised stepwise multiple regression and three-way ANOVA and were performed using the log-transformed 6-SMT and cortisol excretion rates. P<0.05 was considered significant.

RESULTS

The 16 patients had an APACHE II score of between 3 and 26 (median, 18) on admission and spent 3–47 (median, 10) days in the ICU. Four patients died (Table 1).

The ICU environment had normal day/night cycles of light, except for a few occasional short exposures to light at night in connection with emergency treatments. The peak illumination at noon varied between 250 and 500 lux and the illumination at night was about 50 lux (Figure 1).

Representative recording of the illumination in the area of the patient's head during the first 6 days in patient 1

Figure 1
Representative recording of the illumination in the area of the patient's head during the first 6 days in patient 1
Figure 1
Representative recording of the illumination in the area of the patient's head during the first 6 days in patient 1

The overall median 6-SMT excretion was 252 ng/h (interquartile range, 120–615) with marked variations depending on clinical state and medical treatment. Five patients (31%) had a 6-SMT excretion during at least one 24 h period of less than 40 ng/h, or only 10% of the normal excretion in the daytime.

Collection periods with mechanical ventilation were associated with a lower 6-SMT excretion [198 (78–456) ng/h] than during collection periods when such assistance was not given [555 ng/h (305–1066); P<0.0001]. This difference was independent of whether assisted/controlled ventilation or CPAP was used and persisted throughout the ICU stay (Table 2 and Figure 2).

Table 2
Univariate analysis of factors affecting to the excretion of 6-SMT and cortisol during intensive care

Values are medians (25th–75th percentiles). The Mann–Whitney (two groups) or Kruskal–Wallis (three groups) test was used in the statistical analysis.

Factor6-SMT (ng/h)Cortisol (μg/h)P
Mechanical ventilation    
 No (n=87) 554 (305–1066)  <0.0001 
 CPAP (n=112) 195 (101–402)   
 Assisted/controlled (n=133) 203 (69–492)   
Adrenergic therapy    
 No (n=284) 238 (123–527)  <0.01 
 Yes (n=58) 549 (94–1656)   
Fever    
 <38°C (n=159) 304 (69–492)  <0.02 
 38–39°C (n=113) 238 (112–793)   
 >39°C (n=61) 165 (36–315)   
Cortisone therapy    
 No (n=321)  21 (11–39) <0.0001 
 Yes (n=22)  167 (72–262)  
Propofol therapy    
 No (n=226)  28 (14–54) <0.01 
 Yes (n=116)  17 (10–35)  
Factor6-SMT (ng/h)Cortisol (μg/h)P
Mechanical ventilation    
 No (n=87) 554 (305–1066)  <0.0001 
 CPAP (n=112) 195 (101–402)   
 Assisted/controlled (n=133) 203 (69–492)   
Adrenergic therapy    
 No (n=284) 238 (123–527)  <0.01 
 Yes (n=58) 549 (94–1656)   
Fever    
 <38°C (n=159) 304 (69–492)  <0.02 
 38–39°C (n=113) 238 (112–793)   
 >39°C (n=61) 165 (36–315)   
Cortisone therapy    
 No (n=321)  21 (11–39) <0.0001 
 Yes (n=22)  167 (72–262)  
Propofol therapy    
 No (n=226)  28 (14–54) <0.01 
 Yes (n=116)  17 (10–35)  

Urinary excretion of 6-SMT for different periods of the ICU stay depending on the mode of ventilation

Figure 2
Urinary excretion of 6-SMT for different periods of the ICU stay depending on the mode of ventilation

Time periods with adrenergic therapy were excluded.

Figure 2
Urinary excretion of 6-SMT for different periods of the ICU stay depending on the mode of ventilation

Time periods with adrenergic therapy were excluded.

6-SMT excretion was also lower during fever, but higher when adrenergic drugs were infused (Table 2 and Figure 3). The cortisol excretion, which amounted to 22 (11–42) μg/h, was higher during cortisone and lower during propofol treatment (Table 2 and Figure 4). Similar results were obtained by multivariate analyses for the entire period of care (Table 3) and also for the first 10 days only (results not shown). Day/night differences were not statistically significant in any of these analyses.

Urinary excretion of 6-SMT in patients 1 and 2 who both remained for more than 1 month in the ICU

Figure 3
Urinary excretion of 6-SMT in patients 1 and 2 who both remained for more than 1 month in the ICU
Figure 3
Urinary excretion of 6-SMT in patients 1 and 2 who both remained for more than 1 month in the ICU

Urinary excretion of 6-SMT (left-hand panel) and cortisol (right-hand panel) in three ICU patients

Figure 4
Urinary excretion of 6-SMT (left-hand panel) and cortisol (right-hand panel) in three ICU patients

Top, very low 6-SMT excretion and disturbance of the day/night variations during mechanical ventilation. Middle, normal 6-SMT excretion, but disturbed rhythm, during mechanical ventilation. Cortisol excretion is initially very high, which is probably due to preceding surgery, but the rhythm is preserved. Bottom, disturbed circadian rhythm of both 6-SMT and cortisol.

Figure 4
Urinary excretion of 6-SMT (left-hand panel) and cortisol (right-hand panel) in three ICU patients

Top, very low 6-SMT excretion and disturbance of the day/night variations during mechanical ventilation. Middle, normal 6-SMT excretion, but disturbed rhythm, during mechanical ventilation. Cortisol excretion is initially very high, which is probably due to preceding surgery, but the rhythm is preserved. Bottom, disturbed circadian rhythm of both 6-SMT and cortisol.

Table 3
Multivariate analysis of factors that significantly affect the excretion rates of 6-SMT and cortisol during intensive care

Stepwise multiple regression was used to calculate F values. The F value indicates the strength of the factor.

ParameterFactorF valueDirection of effect
6-SMT Mechanical ventilation 66 Decrease 
 Benzodiazepine treatment 18 Increase 
 Adrenergic treatment 10 Increase 
Cortisol Cortisone treatment 39 Increase 
 Propofol treatment Decrease 
ParameterFactorF valueDirection of effect
6-SMT Mechanical ventilation 66 Decrease 
 Benzodiazepine treatment 18 Increase 
 Adrenergic treatment 10 Increase 
Cortisol Cortisone treatment 39 Increase 
 Propofol treatment Decrease 

A sedative in the form of a benzodiazepine, propofol or opiate was given during one-third of the collection periods when the patient breathed spontaneously, whereas one such drug was administered during all periods with CPAP and two drugs during assisted/controlled ventilation. Further analyses of 257 collection periods, which were free from the confounding influence of adrenergic drugs and cortisone treatment, showed that 6-SMT excretion was higher when benzodiazepines, but not propofol or opioids, were administered (P<0.01, three-way ANOVA). However, the differences in excretion for the three kinds of sedation were fairly small (<100 ng/h). Benzodiazepines were also associated with a higher 6-SMT excretion in a subanalysis involving only the day/night periods with assisted/controlled ventilation (P<0.002).

6-SMT excretion showed a consistently disturbed day/night rhythm in seven patients, partially disturbed (usually during mechanical ventilation) in five patients, and a normal rhythm in four patients (Table 4 and Figure 3). For the cortisol excretion, five patients were judged to have a consistently disturbed day/night rhythm, five a partially disturbed, and six had a normal rhythm (Table 5 and Figure 4). The degree of the disturbance did not correlate with the age of the patient or with the APACHE II score.

Table 4
Urinary excretion of a metabolite of melatonin, 6-SMT, in 16 patients treated in the ICU

Values are the medians (interquartile range). Data are the product of concentration and urine volume divided by the duration of urine collection. Day, between 07:00 hours and 22:00 hours; Night, between 22:00 hours and 07:00 hours.

6-SMT excretion (ng/h)
No.Day (ng/h)Night (ng/h)Evaluation of the time course
188 (131–239) 280 (149–341) Rhythm lost during mechanical ventilation 
685 (384–1071) 850 (392–1423) Rhythm lost during ventilation but boosted by 20-fold increased during noradrenaline infusion 
434 (248–646) 655 (288–1651) Rhythm lost during mechanical ventilation, regained during CPAP 
265 (213–491) 313 (192–648) Gradually higher values upon recovery 
65 (30–170) 21 (11–41) No diurnal rhythm, higher excretion after extubation 
2090 (987–2484) 1419 (1021–3783) Higher levels and normal rhythm after extubation 
10 (7–29) 10 (8–29) Very low levels and no diurnal rhythm 
151 (72–211) 223 (74–254) Low levels but highest in the middle of ICU stay 
66 (28–221) 104 (103–247) Diurnal rhythm preserved 
10 764 (613–862) 554 (475–614) Normal values daytime but no increase at night 
11 193 (144–256) 104 (54–116) Reversed pattern (values higher daytime) 
12 358 (184–483) 212 (128–342) Normal values daytime but no increase at night 
13 7351 7121 No increase at night-time 
14 215 (150–3189) 229 (129–2970) No diurnal rhythm, 10-fold increase after adrenaline 
15 172 (76–199) 447 (263–547) Diurnal rhythm preserved 
16 6541 15821 Diurnal rhythm preserved 
6-SMT excretion (ng/h)
No.Day (ng/h)Night (ng/h)Evaluation of the time course
188 (131–239) 280 (149–341) Rhythm lost during mechanical ventilation 
685 (384–1071) 850 (392–1423) Rhythm lost during ventilation but boosted by 20-fold increased during noradrenaline infusion 
434 (248–646) 655 (288–1651) Rhythm lost during mechanical ventilation, regained during CPAP 
265 (213–491) 313 (192–648) Gradually higher values upon recovery 
65 (30–170) 21 (11–41) No diurnal rhythm, higher excretion after extubation 
2090 (987–2484) 1419 (1021–3783) Higher levels and normal rhythm after extubation 
10 (7–29) 10 (8–29) Very low levels and no diurnal rhythm 
151 (72–211) 223 (74–254) Low levels but highest in the middle of ICU stay 
66 (28–221) 104 (103–247) Diurnal rhythm preserved 
10 764 (613–862) 554 (475–614) Normal values daytime but no increase at night 
11 193 (144–256) 104 (54–116) Reversed pattern (values higher daytime) 
12 358 (184–483) 212 (128–342) Normal values daytime but no increase at night 
13 7351 7121 No increase at night-time 
14 215 (150–3189) 229 (129–2970) No diurnal rhythm, 10-fold increase after adrenaline 
15 172 (76–199) 447 (263–547) Diurnal rhythm preserved 
16 6541 15821 Diurnal rhythm preserved 
1

Only two values in each group.

Table 5
Urinary excretion of cortisol in 16 patients treated in the ICU

Values are the medians (interquartile range).

Cortisol excretion (μg/h)
No.DayNightEvaluation of time course
132 (21–266) 30 (16–121) Treated with cortisol during the first 10 days 
13 (8–19) 8 (5–23) Rhythm lost during ventilation but boosted by 10-fold increase during noradrenaline infusion 
85 (26–133) 84 (28–126) Loss of diurnal rhythm 
17 (11–21) 19 (16–54) Loss of diurnal rhythm 
35 (28–65) 33 (24–40) No diurnal rhythm, increase after extubation 
37 (33–80) 28 (21–39) Diurnal rhythm preserved 
28 (16–36) 13 (7–33) Diurnal rhythm preserved 
16 (9–35) 13 (9–29) Low levels and loss of diurnal rhythm during mechanical ventilation 
23 (19–30) 30 (26–38) Loss of diurnal rhythm 
10 54 (35–271) 32 (24–44) Diurnal rhythm preserved 
11 15 (14–15) 10 (9–11) Diurnal rhythm preserved 
12 48 (28–95) 25 (20–56) Low values and loss of diurnal rhythm from 5 days before death 
13 5.71 6.21 Possible loss of diurnal rhythm 
14 21 (17–30) 17 (14–24) Haphazard variation in whether day or night values are the highest 
15 0.20 (0.19–0.21) 0.33 (0.28–0.73) Very low levels and lack of diurnal rhythm 
16 151 111 Diurnal rhythm preserved 
Cortisol excretion (μg/h)
No.DayNightEvaluation of time course
132 (21–266) 30 (16–121) Treated with cortisol during the first 10 days 
13 (8–19) 8 (5–23) Rhythm lost during ventilation but boosted by 10-fold increase during noradrenaline infusion 
85 (26–133) 84 (28–126) Loss of diurnal rhythm 
17 (11–21) 19 (16–54) Loss of diurnal rhythm 
35 (28–65) 33 (24–40) No diurnal rhythm, increase after extubation 
37 (33–80) 28 (21–39) Diurnal rhythm preserved 
28 (16–36) 13 (7–33) Diurnal rhythm preserved 
16 (9–35) 13 (9–29) Low levels and loss of diurnal rhythm during mechanical ventilation 
23 (19–30) 30 (26–38) Loss of diurnal rhythm 
10 54 (35–271) 32 (24–44) Diurnal rhythm preserved 
11 15 (14–15) 10 (9–11) Diurnal rhythm preserved 
12 48 (28–95) 25 (20–56) Low values and loss of diurnal rhythm from 5 days before death 
13 5.71 6.21 Possible loss of diurnal rhythm 
14 21 (17–30) 17 (14–24) Haphazard variation in whether day or night values are the highest 
15 0.20 (0.19–0.21) 0.33 (0.28–0.73) Very low levels and lack of diurnal rhythm 
16 151 111 Diurnal rhythm preserved 
1

Only two values in each group.

DISCUSSION

The overall excretion of 6-SMT in ICU patients was slightly lower than in reference populations [6,7,14]. The increase expected to occur at night was often absent, indicating loss of diurnal rhythm. Several of the ICU patients showed periodical presence of both the normal excretion rate and expected day/night variation, but the diurnal rhythm was most typically disturbed when the overall excretion was low. In contrast, the excretion of cortisol over 24 h was more than 5 times higher than the average for a healthy population. The lower values at night, which is part of the normal sleep/wakefulness pattern, were usually absent.

The excretion rates of both 6-SMT and cortisol differed depending on treatment factors and medical condition (Table 2). Adrenergic drugs clearly increased 6-SMT excretion and, as expected, cortisone treatment increased urinary cortisol levels. Importantly, 6-SMT excretion was decreased by more than 50% during mechanical ventilation, sometimes even to below the level of detection. The difference from those who breathed spontaneously was independent of the time period of the ICU stay and whether assisted/controlled ventilation or CPAP had been applied. These patients showed both hyposecretion of melatonin and a disrupted diurnal rhythm.

The choice of sedative seemed to have little influence on the 6-SMT level, although these drugs were used more often during mechanical ventilation. If anything, benzodiazepines (mostly midazolam) was associated with a slightly higher 6-SMT excretion. Similar covariance was seen with body temperature. Exogenous melatonin lowers the body temperature [16], but the body reacts to fever by increasing melatonin secretion, which is why high levels can be found in patients with sepsis [7]. In our study, however, fever was associated with low 6-SMT levels. Since the correlation between melatonin and body temperature disappeared in the multivariate analyses, fever was probably more likely to develop in those who were ventilated. A few patients showed an increase in 6-SMT excretion when they developed fever, but this pattern was inconsistent. Furthermore, the relatively high between-patient variation was probably also boosted by factors that were not anticipated by us, or were difficult to evaluate statistically. For example, a ‘trauma response’ seemed to increase the early cortisol excretion in those who were transferred to the ICU after surgery, such as patient 6 (Figure 4).

Apart from in infants, who became breathless and blue during sleep [17], there is little previous evidence that melatonin levels are low during critical illness. The mechanism remains to be elucidated. The strongest known inhibitor of melatonin secretion is bright light, which also helps to synchronize the circadian rhythm [18,19], but our patients were subjected to a normal day/night rhythm of light [20]. More likely causes include the underlying disease and the mixture of treatments associated with mechanical ventilation. The stressful situation, as evidenced by the high cortisol excretion, as well as environmental noise, may have contributed [21].

These patients may have suffered from hyposecretion of melatonin. In animal models, such hyposecretion impairs mitochondrial oxidative phosphorylation [22] and the capacity to survive endotoxinaemia [23]. Melatonin has antioxidant actions which prevent ischaemia-induced renal damage [24,25]. In humans, case reports suggest that damage to the pineal gland results in drowsiness, which can be reversed with exogenous melatonin [26,27]. Supplementation with melatonin may also help to co-ordinate and enhance the secretion of growth hormone and prolactin [27].

Disturbances of the circadian rhythm and sleep in ICU patients have been well demonstrated by electro-physiological methods [14]. In the present study, it was not an issue as to whether urine was collected during the daytime or at night, as it should if the biological rhythms were intact. Disruption of the circadian rhythm is known to affect sleep, mood and cognitive performance [2830]. Although a precise assessment of this rhythm would ideally include phase-angle shifts [7], the frequent hyposecretion of melatonin and the disturbed melatonin/cortisol patterns suggest that melatonin could be a useful medication in the ICU, in particular during mechanical ventilation. Shilo et al. [31] improved sleep/wakefulness patterns by providing exogenous melatonin to eight patients with respiratory failure. The benefits of such therapy could be a less sleepy patient, who co-operates better with the nursing staff during mobilization. The consumption of sedatives might decrease, since melatonin is a hypnotic agent [16,30].

It is concluded that hyposecretion of melatonin is very common during mechanical ventilation in the ICU, suggesting a need for exogenous supplementation.

Abbreviations

     
  • APACHE II

    Acute Physiology And Chronic Health Evaluation 2

  •  
  • CPAP

    continuous positive airway pressure

  •  
  • ICU

    intensive care unit

  •  
  • 6-SMT

    6-sulphatoxymelatonin

We thank Dr Johan Wikner, who participated in the planning of the study, the staff at the ICU departments at Stockholm South Hospital and Gällivare Regional Hospital for collecting the urine samples, and Ms Ewa Thuresson, who served as co-ordinator. The analyses of 6-SMT were performed by Ms Judie English of Stockgrand Ltd at the University of Guildford, Surrey, U.K., and the analyses of cortisol by Ms Monica Nordlund at the Department of Clinical Research at Stockholm South Hospital. Funding for this study was obtained from the Departments of Anaesthesia at the Stockholm South and Gällivare Hospitals, and from the Gällivare County.

References

References
1
Hilton
 
A.
 
Quantity and quality of patient's sleep and sleep-disturbing factors in a respiratory intensive care unit
J. Adv. Nursing
1976
, vol. 
1
 (pg. 
453
-
468
)
2
Aurell
 
J.
Elmqvist
 
D.
 
Sleep in the surgical intensive care unit: continuous polygraphic recording of sleep in nine patients receiving postoperative care
Br. Med. J.
1985
, vol. 
290
 (pg. 
1029
-
1032
)
3
Richards
 
K. C.
Bairnsfather
 
L.
 
A description of night sleep patterns in the critical care unit
Heart Lung
1988
, vol. 
17
 (pg. 
35
-
42
)
4
Freedman
 
N. S.
Gazendam
 
J.
Levan
 
L.
Pack
 
A. I.
Schwab
 
R. J.
 
Abnormal sleep/wake cycles and the effect of environmental noise on sleep disruption in the intensive care unit
Am. J. Respir. Crit. Care Med.
2001
, vol. 
163
 (pg. 
451
-
457
)
5
Fontaine
 
D. K.
 
Measurement of nocturnal sleep patterns in trauma patients
Heart Lung
1989
, vol. 
18
 (pg. 
402
-
410
)
6
Shilo
 
L.
Dagan
 
Y.
Smorjik
 
Y.
, et al 
Patients in the intensive care unit suffer from severe lack of sleep associated with loss of normal melatonin secretion pattern
Am. J. Med. Sci.
1999
, vol. 
317
 (pg. 
278
-
281
)
7
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
, vol. 
30
 (pg. 
536
-
540
)
8
Goto
 
K.
Miyake
 
A.
Mizobuchi
 
S.
Ohe
 
K.
Morita
 
K.
 
Melatonin excretion pattern in critically ill patients requiring mechanical ventilation
Anesth. Analg.
2001
, vol. 
94
 pg. 
S-109
 
9
Moore-Ede
 
M. C.
Czeisler
 
C. A.
Richardson
 
G. S.
 
Circadian timekeeping in health and disease. Part 2. Clinical implications of circadian rhythmicity
N. Engl. J. Med.
1983
, vol. 
309
 (pg. 
530
-
536
)
10
Arendt
 
J.
Bojkowski
 
C.
Franey
 
C.
Wright
 
J.
Marks
 
V.
 
Immunoassay of 6-hydroxymelatonin sulphate in human plasma and urine: abolition of the urinary 24-hour rhythm with atenolol
J. Clin. Endo. Metab.
1985
, vol. 
60
 (pg. 
1166
-
1173
)
11
Aldhous
 
M. E.
Arendt
 
J.
 
Radioimmunoassay for 6-sulphatoxymelatonin in urine using an iodinated tracer
Ann. Clin. Biochem.
1988
, vol. 
25
 (pg. 
298
-
303
)
12
Markey
 
S. D. P.
Higa
 
S.
Shih
 
M.
Danfordt
 
D. N.
Tamarkin
 
L.
 
The correlation between human plasma melatonin levels and urinary 6-hydroxymelatonin excretion
Clin. Chim. Acta
1985
, vol. 
150
 (pg. 
221
-
225
)
13
Bojkowski
 
C. J.
Arendt
 
J.
Shih
 
M. C.
Markey
 
S. P.
 
Melatonin secretion in humans assessed by measuring its metabolite, 6-sulphatoxymelatonin
Clin. Chem.
1987
, vol. 
33
 (pg. 
1343
-
1348
)
14
Baskett
 
J. J.
Cockrem
 
J. F.
Antunovich
 
T. A.
 
Sulphatoxymelatonin excretion in older people: relationship to plasma melatonin and renal function
J. Pineal Res.
1998
, vol. 
24
 (pg. 
58
-
61
)
15
Arakawa
 
H.
Maeda
 
M.
Tsuji
 
A.
 
Chemiluminescence enzyme immunoassay of cortisol using peroxidase as label
Ann. Biochem.
1979
, vol. 
97
 (pg. 
248
-
254
)
16
Dollins
 
A. B.
Zhdanova
 
I. V.
Wurtman
 
R. J.
Lynch
 
H. J.
Deng
 
M. H.
 
Effect of inducing nocturnal serum melatonin concentrations in daytime on sleep, mood, body temperature, and performance
Proc. Natl. Acad. Sci. U.S.A.
1994
, vol. 
91
 (pg. 
1824
-
1828
)
17
Sivan
 
Y.
Landon
 
M.
Kuint
 
J.
Zisapel
 
N.
 
Low melatonin production in infants with a life-threatening event
Dev. Med. Child Neurol.
2000
, vol. 
42
 (pg. 
487
-
491
)
18
Boulos
 
Z.
Macchi
 
M. M.
Sturchler
 
M. P.
, et al 
Light visor treatment for jet lag after westward travel across six time zones
Aviat. Space Environ. Med.
2002
, vol. 
73
 (pg. 
953
-
963
)
19
Eastman
 
C.
Martin
 
S.
 
How to use light and dark to produce circadian adaptation to night shift work
Ann. Med.
1999
, vol. 
31
 (pg. 
87
-
98
)
20
Arendt
 
J.
Bojkowski
 
C.
Folkard
 
S.
, et al 
Some effects of melatonin and the control of its secretion in humans
Ciba Found. Symp.
1985
, vol. 
117
 (pg. 
266
-
283
)
21
Weibel
 
L.
Maccari
 
S.
Van Reeth
 
O.
 
Circadian clock functioning is linked to acute stress reactivity in rats
J. Biol. Rhythms
2002
, vol. 
17
 (pg. 
438
-
446
)
22
Martin
 
M.
Macias
 
M.
Leon
 
J.
Escames
 
G.
Khaldy
 
H.
Acuna-Castroviejo
 
D.
 
Melatonin increases the activity of the oxidative phosphorylation enzymes and the production of ATP in rat brain and liver mitochondria
Int. J. Biochem. Cell. Biol.
2002
, vol. 
34
 (pg. 
348
-
357
)
23
Maestroni
 
G. J.
 
Melatonin as a therapeutic agent in experimental endotoxic shock
J. Pineal Res.
1996
, vol. 
20
 (pg. 
84
-
91
)
24
Tunez
 
I.
del Carmen Munoz
 
M.
Feijoo
 
M.
Valdelvira
 
M. E.
Rafael Munoz-Casteneda
 
J.
Montilla
 
P.
 
Melatonin effect on renal oxidative stress under constant light exposure
Cell Biochem. Funct.
2003
, vol. 
21
 (pg. 
35
-
40
)
25
Kunduzova
 
O. R.
Escourrou
 
G.
Seguelas
 
M. H.
, et al 
Prevention of apoptotic and necrotic cell death, caspase-3-activation, and renal dysfunction by melatonin after ischemia/reperfusion
FASEB J.
2003
, vol. 
17
 (pg. 
872
-
874
)
26
Lehmann
 
E. D.
Cockerell
 
O. C.
Rudge
 
P.
 
Somnolence associated with melatonin deficiency after pinealectomy
Lancet
1996
, vol. 
347
 pg. 
323
 
27
Petterborg
 
L. J.
Thalén
 
B.-E.
Kjellman
 
B. F.
Wetterberg
 
L.
 
Effect of melatonin replacement on serum hormone rhythms in a patient lacking endogenous melatonin
Brain Res. Bull.
1991
, vol. 
27
 (pg. 
181
-
185
)
28
Herxheimer
 
A.
Petrie
 
K. J.
 
Melatonin for the prevention and treatment of jet lag
Cochrane Database Syst. Rev.
2002
pg. 
CD001520
 
29
Takahashi
 
T.
Sasaki
 
M.
Itoh
 
H.
, et al 
Melatonin alleviates jet lag symptoms caused by an 11-hour eastward flight
Psych. Clin. Neurosci.
2002
, vol. 
56
 (pg. 
301
-
302
)
30
Cardinali
 
D. P.
Brusco
 
L. I.
Lloret
 
S. P.
Furio
 
A. M.
 
Melatonin in sleep disorders and jet-lag
Neuroendocrinol. Lett.
2002
, vol. 
23
 
Suppl 1
(pg. 
9
-
13
)
31
Shilo
 
L.
Dagan
 
Y.
Smorjik
 
Y.
, et al 
Effect of melatonin on sleep quality of COPD intensive care patients: a pilot study
Chronobiol. Int.
2000
, vol. 
17
 (pg. 
71
-
76
)

Author notes

1

Present address: Department of Anesthesiology, University of Texas Medical Branch, Galveston, Texas, U.S.A.