The most important result is the use of different agents according to the expected length of analgesia and sedation. In the American guidelines [
18] for short-term sedation only propofol is recommended, and for long-term sedation midazolam and lorazepam are recommended. In our survey the most commonly used agent for sedation up to 24 hours and during weaning from ventilation was propofol. Midazolam was used mainly for sedation longer than 72 hours. Lorazepam was not used by any department. This is due mainly to handling (2 mg ampoule) and costs, because lorazepam is one of the more expensive agents in Germany for sedation.
Whereas the American guidelines recommend fentanyl, hydromorphone and morphine for analgesia in all phases [
18], in our survey fentanyl and sufentanil were used most often for analgesia for up to 24 hours, between 24 and 72 hours and for weaning from ventilation. In addition, NSAIDs were used preferentially in short-term sedation (less than 24 hours). For longer than 72 hours, fentanyl was preferred for analgesia. Whereas in the British survey [
15] from the year 2000 alfentanil was very often used, this opioid did not have a role in German ICUs. In the Danish survey the preferred drugs for analgesia were morphine (94%), fentanyl (76%) and sufentanil (43%) [
16]. Our survey shows that morphine did not have a major role in analgesia and sedation in German ICUs. Specifically in Germany, piritramide is a frequently used agent for postoperative analgesia. The reason may be that in Germany some anesthesiologists claim to achieve a lower incidence of nausea and vomiting with piritramid than with morphine [
22].
Noticeable was the widespread use of central neuroaxial techniques in analgesia for up to 24 hours. Brodner and colleagues [
23] and Beattie and colleagues [
24] showed that the perioperative use of epidural analgesia leads to a shortened length of stay in the ICU and also a decrease in cardiac events.
Clonidine as an adjuvant for sedation was used in our survey in a high percentage in all phases, whereas haloperidol, which is recommended in the American guidelines, was not a selectable option in our questionnaire. Most often clonidine was used in the phases longer than 72 hours and during weaning from ventilation. A reasonable use (with regard to time of ventilation and ICU stay) of this agent during weaning was shown by Walz and colleagues [
25]. Bohrer and colleagues showed [
26] that with clonidine the requirements for opioids and sedation may be reduced.
Ketamine (S) was preferred as an adjuvant in the phases of sedation longer than 24 hours. There have been few studies for long-term sedation with ketamine, as Ostermann and colleagues [
11] showed in their review. One of the reasons for the use of ketamine is the lower negative influence on bowel motility than with opioids [
27].
In our survey 43% of the hospitals stated that they had have established an oral policy for analgesia and sedation. A procedure in writing was used in 21%. In the survey by Murdoch and colleagues [
15] 43% of the British ICUs stated that they had procedures in writing for analgesia and sedation, and 51% had a defined oral policy. In addition, in the 1987 survey of British ICUs by Bion and colleagues [
28], 40% stated that they had established a formal procedure. In other surveys it was shown that with the use of standard operating procedures a decrease in the durations of sedation and ventilation, and with this a reduction of costs, is possible [
29,
30]. Mascia and colleagues [
31] showed that the use of written standard operating procedures decreases the duration of ventilation, the length of stay in the ICU and the overall hospital stay.
A sedation scale for the monitoring of analgesia and sedation was used by 31% of the hospitals questioned; 8% stated that they used the Ramsay sedation scale [
21] for monitoring sedation. In the survey by Soliman and colleagues [
17], 49% of the German hospitals answered that they used the Ramsay sedation scale [
19]. In English hospitals in the survey by Murdoch and colleagues [
15], 60% were using a sedation scale. In the survey of Danish ICUs [
16] from the year 1996/1997, 16% of the hospitals answered that they were using a sedation scale.
More recent studies showed that close monitoring with the help of a scoring system can lead to a decrease in the length of ICU stay and in the length of hospital stay [
32].
Nearly all hospitals in our survey stated that they paid attention to cost in their choice of medication. However, the survey showed that there were no significant differences in the use of medications between the hospitals that answered yes and those that answered no. Murdoch and colleagues [
15] came to the same conclusion in their survey of English ICUs. More expensive agents may be useful with regard to overall costs because the length of stay in the ICU may be reduced, as was shown by Barrientos-Vega and colleagues [
33] and Dahaba and colleagues [
34].
Questioned on whether the expected length of sedation had a role in selecting the medication, 92% of the hospitals agreed. The analysis showed that for short-term sedation agents were also used that had a long context-sensitive half-time (fentanyl 45%, midazolam 40%) [
35]. Nearly all ICUs tried to maintain a day–night rhythm, although only few studies exist [
36,
37] that have shown advantages of it for the patients. The Danish survey [
16] yielded almost the same results.
In our survey the use of neuromuscular blocking agents had almost disappeared, confirming the results of the European [
17], British [
15] and Danish [
16] surveys of the routine use of neuromuscular blocking agents in intensive care medicine.
The incidence of withdrawal in long-term sedation is 60–80% [
38,
39]. In our survey, values between 20% and 25% were stated, which is explained by the fact that all patients, even short-term patients, were included.
In our survey the return rate was 84%. Christensen and colleagues [
16] in Denmark and Murdoch and colleagues [
15] in the UK achieved similar return rates (92.5% and 79%, respectively). In a pan-European questionnaire about the practice of analgesia and sedation by Soliman and colleagues [
17] the return rate was 20%.
One of the problems of this survey was the limitation to ICUs run by the department of anesthesiology. We do not have data on whether the patients were mainly postoperative and trauma patients, or whether the ICUs also routinely took care of patients from the department of internal medicine. Hack and colleagues showed in their survey [
40] that the most of the interdisciplinary ICUs in general hospitals in Germany are run by the department of anesthesiology [
40].