Awareness during general anaesthesia can be responsible for the recollection of some intraoperative events . This recollection can lead to psychological sequelae for the patient and medicolegal concerns for the anaesthetist . The type of memory involved in this situation is called ‘explicit memory’.
Another type of memory, ‘implicit memory’, has also been described during general anaesthesia . This recollection occurs when a subject remembers information without being aware of it. It is still unclear whether this memory can be processed by unconscious patients or whether it is caused by short periods of awareness . In a previous experiment, implicit or explicit memory could not be found when consciousness was lost using propofol as sole anaesthetic agent . However, propofol is often administered with an opiate, e.g. remifentanil, during total intravenous anaesthesia. This combination markedly decreases the concentration of propofol necessary to obtain loss of consciousness . This low hypnotic concentration could result in a higher risk of implicit memory during anaesthesia.
The bispectral index (BIS, Aspect Medical Systems, Newton, MA, USA) has been used to detect intraoperative explicit or implicit memory with varied results. Indeed, Andrade and colleagues  did not find any implicit or explicit memory with a mean BIS value of 85 ± 4 while Lubke and colleagues  showed implicit memory occurring with a BIS value between 40 and 60.
We therefore decided to study explicit and implicit memory during a low-propofol-high-remifentanil anaesthesia technique using BIS monitoring.
With approval of the local ethics committee and after written informed consent, 10 American Society of Anesthesiologists (ASA) I-II patients, younger than 50-yr old and undergoing minor surgery, were enrolled.
All patients had French as their native language, with no hearing issues, memory impairment or past history of drug abuse. Due to the possible impairment of implicit and explicit memory by midazolam , no patient received any benzodiazepine on the day of surgery or the day before.
The patients of our experimental group were anaesthetised with a combined target-controlled infusion of propofol and remifentanil using the ‘Infusion Toolbox’ software with the pharmacokinetic model of Marsh  for propofol and the model of Minto  for remifentanil. BIS was continuously recorded during the whole study period (BIS A 2000, Aspect Medical Systems, Newton, MA, USA, with BIS Sensor, Aspect Medical Systems, Newton, MA, USA).
First, a remifentanil was started with a target-calculated plasma concentration of 8 ng mL−1. When the calculated effect-site concentration was reached, propofol was started with an initial target plasma concentration of 0.5 μg mL−1. This plasma concentration was then progressively increased by steps of 0.2 μg mL−1 every 3 min until the patient was unresponsive to verbal command. The calculated effect-site concentration of propofol was then maintained at this level. Ventilation was assisted with a face mask and 100% oxygen. Before surgery began, a list of 20 French common words was played via headphones to each patient and repeated three times in a different sequence (duration about 3 min). Then the patient's unconsciousness (patient's unresponsiveness to a verbal command) was checked again. Thereafter, the target plasma concentrations of propofol and remifentanil were adjusted on clinical grounds and a neuromuscular blocker was administered as required for surgery.
Between 2 and 4 h after recovery, three memory tests (see below) were performed via a written questionnaire.
A group of 20 patients from a previous study  was used as a control group. These patients answered the three memory tests after propofol anaesthesia without having previously heard the list of words.
Three different memory tests, as described by Light and Singh  and adapted to French , were used to study implicit and explicit memory under anaesthesia. These were the word stem completion test, the free recall test and the forced-choice recognition test.
The word stem completion test investigated both implicit and explicit memory. The patients were given a list of 40 stems of three letters which are the stems of each target word of two lists (the target list of 20 words heard by the patients when anaesthetised and a control list of 20 words that they did not hear). The patients had to complete in writing these stems with a word they remembered having heard during anaesthesia or, if they did not remember it, with the first word that came to mind.
The second test was the free recall test, which investigated exclusively explicit memory. The subjects had to write down the words they remembered.
The third test was the forced-choice recognition test, which investigated both explicit and implicit memory. Each word of the target list (heard by the subject) was paired with a word of the control list (not heard by the subject). The subjects had to choose the word they remembered in each pair, or if they did not, the more familiar one.
The Patient characteristics data of the two groups and the results of the three memory tests were analysed using the Mann-Whitney non-parametric test. A P-value <0.05 was considered significant.
Ten patients were enrolled in the experimental group: two men and eight women. Mean ± SD age was 36 ± 11 yr, interval between recovery and memory tests was 174 ± 47 min and calculated effect-site concentration of propofol at loss of consciousness was 1.3 ± 0.6 μg mL−1. There were no significant differences from the results of the control group except for the effect-site concentration of propofol at loss of consciousness which was significantly lower (P < 0.05) in the experimental group. Median (range) BIS value of the patients of the experimental group during word presentation was 93 (80-98) (Table 1).
Results of the memory tests are shown in Table 2. The patients in the experimental group scored 1.0 ± 1.2 words (over 20) in the word stem completion test of the target list, whereas patients of the control group scored 1.2 ± 0.9 words (P > 0.05).
There was no statistical difference between the two groups in the results of the word stem completion test for the control list: 3.8 ± 1.8 and 3.4 ± 0.9 words for the experimental and control groups respectively (P > 0.05).
No patient in either group remembered any word in the free recall test.
The results of the forced-choice recognition test were 7.0 ± 2.5 words for the experimental group and 8.5 ± 1.6 words for the control group (P > 0.05).
There are conflicting results about implicit memory during general anaesthesia. It is still unclear whether this implicit learning may occur when subjects are unconscious or whether short periods of awareness are responsible for this memory . We have showed in a previous study  that implicit recollection was eliminated when young ASA I/II patients had lost consciousness when propofol was used alone in the absence of a surgical stimulus. However, when propofol is combined with an opiate, loss of consciousness may appear at a much lower concentration of propofol because of drug synergy . This situation is often encountered in clinical practice since the use of remifentanil has become widespread. This low concentration of hypnotic may increase the risk of implicit recollection during anaesthesia. Indeed, several studies using nitrous oxide as the sole hypnotic agent have shown implicit learning to occur during general anaesthesia [11-13] while another experiment using isoflurane less than 0.3% end tidal also revealed implicit memory . This brought us to investigate implicit learning in unconscious patients anaesthetised with a high concentration of remifentanil and the lowest concentration of propofol associated with loss of consciousness.
The two groups answered the word stem completion test of the control list with no significant difference. As no patient heard the words of this control list, it can be assumed that the two groups are comparable.
The patients of the experimental group never found out significantly more words than the patients of the control group in the three tests. As the patients in the latter group had never heard the tape before, it can be assumed that the patients of the former group had neither implicit nor explicit learning while unconscious. This suggests that implicit memory cannot be processed when consciousness is lost and that implicit learning only occurs because of short periods of awareness.
These results were obtained with a surprisingly high median (range) BIS value of 93 (80-98) during word presentation. Indeed, Liu and colleagues  demonstrated explicit memory with a BIS value of 87, and Lubke and colleagues  found implicit memory with BIS value between 40 and 60. So it was expected that implicit memory may be found with such a high mean BIS value. This negative result suggests that BIS may not be a reliable monitor of implicit memory when a high concentration of remifentanil is used. These high BIS values may have been caused by an interference of the electromyogram with BIS. Indeed, Renna and colleagues  showed that BIS can be biased by the electromyogram during high-dose fentanyl induction without neuromuscular blocking agent.
In our experiment, memory was investigated when patients were not exposed to a noxious surgical stimulus and it has been suggested that catecholamine released by anaesthetised patients in response to a painful surgical stimulus could be responsible for implicit learning . Indeed, it has been proven that epinephrine enhances learning in anaesthetised rats  although this remains to be proven in human studies. However, if analgesia is sufficient, i.e. a high remifentanil concentration, no catecholamines will be released in response to a noxious stimulus and implicit memory will not be enhanced. Other studies with a larger number of patients should be conducted to confirm the results of this study.
In conclusion, we found that in a group of young ASA I/II patients, in the absence of a noxious stimulus, no explicit or implicit memory was found when the calculated concentration of propofol (combined with a high concentration of remifentanil) was maintained at the level associated with loss of consciousness. Moreover, there was no implicit or explicit memory with a high BIS value suggesting that BIS is not a reliable monitor for intraoperative memory when a high concentration of remifentanil is used in the absence of a neuromuscular block.
1. Bailey AR, Jones JG. Patient's memories of events during general anaesthesia. Anaesthesia
2. Gonheim MM, Block RI. Learning and memory
during general anesthesia: an update. Anesthesiology
3. Lequeux PY, Cantraine F, Levarlet M, Barvais L. Absence of explicit
and implicit memory
in unconscious patients using a TCI of propofol
. Acta Anaesthesiol Scand
4. Vuyk J. Pharmacokinetic and pharmacodynamic interactions between opioids and propofol
. J Clin Anesth
1997; 9 (Suppl 6)
5. Andrade J, Englert L, Harper C, Edwards ND. Comparing the effects of stimulation and propofol
infusion rate on implicit
and explicit memory
formation. Br J Anaesth
6. Lubke GH, Kerssens C, Phaf H, Sebel PS. Dependence of explicit
and implicit memory
on hypnotic state in trauma patients. Anesthesiology
7. Hirshman E, Passannante A, Henzler A. The effect of midazolam on implicit memory
tests. Brain Cognition
8. Marsh B, White M, Morton N, Kenny GN. Pharmacokinetic model driven infusion of propofol
in children. Br J Anaesth
9. Minto CF, Schnider TW, Shafer SL. Pharmacokinetics and pharmacodynamics of remifentanil
. II. Model application. Anesthesiology
10. Light LL, Singh A. Implicit
and explicit memory
in young and older adults. J Exp Psychol Learn Mem Cogn
11. Ghoneim MM, Block RI, Dhanaraj VJ, Todd MM, Choi WW, Brown CK. Auditory evoked responses and learning and awareness during general anesthesia. Acta Anaesthesiol Scand
12. Jelicic M, De Roode A, Bovill JG, et al
. Unconscious learning during anaesthesia. Anaesthesia
13. Block RI, Ghoneim MM, Sum Ping ST, et al
. Human learning during general anaesthesia and surgery. Br J Anaesth
14. Lubke GH, Kerssens C, Gershon RY, Sebel PS. Memory
formation during general anesthesia for emergency Cesarean sections. Anesthesiology
15. Liu J, Singh H, White PF. Electroencephalographic bispectral index correlates with intraoperative
recall and depth of propofol
-induced sedation. Anesth Analg
16. Renna M, Wigmore T, Mofeez A, Gillbe C. Biasing effect of the electromyogram on BIS: a controlled study during high-dose fentanyl induction. J Clin Monitor Comput
17. Stapleton CL, Andrade J. An investigation of learning during propofol
sedation and anesthesia using the process dissociation procedure. Anesthesiology
18. Darolia MK, Yadava A, Malhotra S. Effect of epinephrine on learning under anaesthesia. J Indian Acad Appl Psychol