Secondary Logo

Journal Logo

Original Article

Using the Sustained Attention to Response Test to detect cognitive dysfunction after day case surgery

Thompson, J. P.; Rushman, S. C.; Fox, A. J.; Lloyd, A. J.; Atcheson, R. A.

Author Information
European Journal of Anaesthesiology: August 2002 - Volume 19 - Issue 8 - p 585-588



Cognitive and psychomotor dysfunction may occur after anaesthesia and surgery, especially in the elderly and in those with pre-existing disease [1]. Postoperative cognitive or psychomotor dysfunction is also related to the drugs used during anaesthesia, anxiety and other psychological factors, and the duration of bed rest and a hospital stay [2-4]. Impaired psychomotor and cognitive function is more common after major surgery, but may also be detected 24-48 h after day case surgery performed under general anaesthesia [2-5]. This might lead to problems relating to lack of supervision in the postoperative period, particularly in elderly patients [1,6].

Methods of measuring recovery and assessing psychomotor and cognitive function are time consuming, complex and difficult to carry out readily at the bedside [1]. Over 35 different tests have been used in varying combinations in different studies [7]. The majority derive from psychomotor tasks devised to assess attention, reaction time and memory in normal individuals [8]. The Sustained Attention to Response Test (SART) is a computerized test of sustained attention that has been correlated with measures of cognitive dysfunction in both volunteers and patients after traumatic head injury [9,10]. Sustained attention is defined as 'the ability to self-sustain mindful, conscious processing of stimuli whose repetitive, non-arousing qualities would otherwise lead to habituation and distraction to other stimuli' [9]. Drugs and alcohol, as well as brain injury, affect sustained attention [11], and our hypothesis was that sustained attention may be useful as an indicator of postoperative cognitive dysfunction. SART has been reported as being brief and simple to administer [9,10]. We therefore performed a pilot study both to assess the feasibility of using SART in a hospital setting and to evaluate possible postoperative deficits in attention in patients undergoing day case surgery under local and general anaesthesia.


SART consists of 225 single digits generated and displayed in random order and size on a computer screen over 4.3 min. The subjects are told to respond as accurately and as quickly as possible to the appearance of a digit on the screen by pressing the computer's mouse button, unless it is the number 3. The ability to withhold the response to a '3' requires sustained attention. The number of times a subject presses the mouse button in response to a 3 appearing on the screen measures their ability to attend to the test. These errors of commission (incorrect responses) are predicted to be sensitive to cognitive dysfunction [9,10]. The test records reaction times before and after responses, together with any errors in responding. The speed with which a subject responds to a number appearing on the screen tends to increase with successive responses until an incorrect response is made. At this point, the subject response slows and concentration returns and improves the number of subsequent correct responses, associated with a slowing of the reaction time. This automatic drift in concentration is seen in normal subjects and in those with traumatic head injury, but the slowing after an incorrect response is affected by the ability to recognize that there has been a mistake. Patients who have sustained a head injury are unable to respond to and correct their mistakes, so that their reaction times before and after incorrect responses are similar [9]. This speeding of responses before an error is predictive of the occurrence of errors during SART, and both reaction times and the number of errors correlate with self-reported cognitive errors during daily life in volunteers [9].

After local Ethics Committee approval and informed consent, we recruited 35 patients undergoing a variety of different day case surgical procedures under general anaesthesia (GA group). A short demonstration of the test was given by one of the investigators. Each patient performed the test preoperatively and it was then repeated 2 h after anaesthesia, when all patients were judged fit for discharge from hospital according to conventional clinical criteria, i.e. they were pain-free, independently mobile, with no nausea or vomiting. We also studied a parallel group of 25 patients before and after operation undergoing day case surgery under local anaesthesia (LA group) to assess other psychological effects of day case surgery per se. Patients were not randomized and no effort was made to standardize their anaesthetic technique since we were not trying to compare specific methods of anaesthesia; no patient received premedication.

Data were analysed using the SART software previously described [9,10] on a 486 MHz laptop computer. Statistical analysis was performed using general linear model analysis of variance for repeated measures, with SPSS® for Windows® software (v.9.0; SPSS Inc., Chicago, IL, USA).


Mean (±SD) age was 47.8 (±15.2) yr in the LA group and 31.6 (±10.8) yr in the GA group. The types of surgery under GA included short gynaecological, general surgery and dental cases. LA cases included short general and body surface surgical cases under local analgesia without sedative drugs. All patients in the GA group were ASA I-II and underwent balanced GA using propofol, isoflurane, non-steroidal anti-inflammatory drugs (NSAIDs) and low-dose opioids (fentanyl, alfentanil). The mean (±SD) duration of surgery was 40.5 (±16.8) and 43.2 (±22.3) min in the LA and GA groups, respectively. The gender distribution was similar (12 females, 13 males in the LA group; 17 females, 18 males in the GA group).

The results of the SART are shown in Table 1. The numbers of correct responses were similar in both groups, both before and after operation. However, patients in the GA group made more mistakes, but had faster reaction times preoperatively, for both correct and incorrect responses than those undergoing procedures under LA (P < 0.05 between groups). Both groups showed a quickening in reaction time before and a slowing of reaction times after making an incorrect response, both before and after surgery. The number of incorrect responses increased after surgery in both groups (P < 0.05), although there was no significant interaction between groups. However, reaction times for correct responses were significantly shorter on repeating the SART after surgery under LA (P < 0.05 within group). In contrast, postoperative reaction times in the GA group were similar to preoperative values, despite an increased number of incorrect responses.

Table 1
Table 1:
Results of SART.


The number of incorrect responses to SART increased in patients 2 h after minor surgery, which was associated with a decrease in reaction times before and an increase in reaction time after an incorrect response. This pattern is characteristic of a deficit in sustained attention in healthy volunteers [9] and was similar whether surgery was performed under GA or LA. The number of correct responses remained unchanged in both groups, implying that the capability to perform the SART was unaffected by anaesthesia or surgery. Reaction times in the GA group were similar before and after operation, suggesting that gross psychomotor responses were unchanged. It is known that different aspects of cognitive function recover at different rates, even after minor or day case surgery [2-4]. We found an increase in incorrect responses (or errors of commission) without changes in reaction times, which suggests a deficit in sustained attention after operation. This is potentially of greater concern than a global cognitive deficit since the patient or their carer may not be aware of it and activities such as driving, operating machinery and cooking will be affected [11].

The study was designed as a pilot study to assess the feasibility of using SART in hospital patients undergoing surgery. Robertson and colleagues validated SART in healthy volunteers and in patients who had suffered traumatic brain injury and found that all the subjects studied made errors that seemed to be a result of a drift in automatic responses. More importantly, SART scores correlated well with other measures of sustained attention and with self-reported slips in attention in everyday life in both volunteers and patients after traumatic head injury [9]. These slips in everyday attention have been reported after GA and have been evaluated by complicated cognitive tests that may be protracted and difficult to analyse [1]. In contrast, we found SART to be simple, rapid, easy to administer, acceptable to the patients (even those with limited computer skills) and versatile enough to be used in hospital and at the bedside by anaesthetists with no formal psychological training. Results from the SART correlate with more time-consuming and complicated measures of cognitive dysfunction [9,10]. We did not standardize anaesthetic techniques so our data cannot be related to any specific drug or type of procedure, but we observed effects in patients undergoing day case surgery under GA and LA. The recorded absolute differences were small (<15%) but similar in magnitude to previous studies [9,10], and the fact that there seems to be a difference in day case patients (when little difference would be expected) suggests that it is worthy of further study.

We recognize that the study has limitations. Consistent with the broad aims of the study, the allocation of patients to the LA or GA groups was not randomized and, therefore, direct comparisons between the groups have been largely avoided. Patients in the LA group were older and had slower reaction times, but made less incorrect responses both before and after surgery. However, the pattern of responses was similar in both groups: the number of incorrect responses increased after surgery (although this was associated with a decrease in response time in the LA group). Furthermore, although no patients received premedication and the GA or LA that most patients received was similar, the precise anaesthetic regimen was not controlled. Conclusions about possible explanations for these results must be limited, but the overall similarity in the pattern of responses between the groups suggests that psychological or other factors (anxiety, psychological stress, pain) may be more important than pharmacological factors (type of anaesthesia) after minor day case surgery. This is consistent with previous data showing little difference between other aspects of postoperative cognitive dysfunction in patients undergoing minor surgery under LA or GA [12].

Further studies should assess the duration of this reduction in sustained attention, in patients undergoing minor and more extensive or prolonged surgery, should investigate the effect of specific anaesthetic types, and might usefully compare SART with other measures of postoperative cognitive dysfunction. A further assessment of the test would be to compare patients undergoing surgery under GA or LA with age-matched controls not undergoing surgery. This might allow for the effects of preoperative anxiety or other psychological factors, the possible learning effects of repeating the test within a hospital environment and the pharmacological effects of residual GA.

In summary, SART was simple to administer and may be a potentially useful tool for the assessment of sustained attention in the perioperative period by investigators without formal psychological training. These preliminary data suggest a deficit in sustained attention in patients after minor day case surgery under GA or LA, despite patients being apparently fit for discharge. Further studies are required to clarify the clinical significance of these findings.


The authors thank Ian Robertson, Professor of Psychology, Trinity College, Dublin, who supplied SART and provided helpful comments and advice on the project.


1. Kortilla KT. Post-anaesthetic psychomotor and cognitive function. Eur J Anaesthesiol 1995; 12(Suppl 10): 43-46.
2. Pollard BJ, Bryan A, Bennett D, et al. Recovery after oral surgery with halothane, enflurane, isoflurane or propofol anaesthesia. Br J Anaesth 1994; 72: 559-566.
3. Nightingale JJ, Lewis IH. Recovery from day-case anaesthesia: comparison of total i.v. anaesthesia using propofol with an inhalational technique. Br J Anaesth 1992; 68: 356-359.
4. Heath PJ, Ogg TW, Gilks WR. Recovery after day-case anaesthesia. Anaesthesia 1990; 45: 911-915.
5. Kubitz J, Epple J, Bach A, et al. Psychomotor recovery in very old patients after total intravenous or balanced anaesthesia for cataract surgery. Br J Anaesth 2001; 86: 203-208.
6. Tzabar Y, Asbury AJ, Millar K. Cognitive failures after general anaesthesia for day-case surgery. Br J Anaesth 1996; 76: 194-197.
7. Dodds C, Allison J. Postoperative cognitive deficit in the elderly surgical patient. Br J Anaesth 1998; 81: 449-462.
8. Broadbent DE, Cooper PF, Fitzgerald P, Parkes KR. The cognitive failures questionnaire (CFQ) and its correlates. Br J Psychol 1982; 21: 1-16.
9. Robertson IH, Manly T, Andrade J, Baddeley BT, Yiend J. 'Oops!': performance correlates of everyday attentional failures in traumatic brain injured and normal subjects. Neuropsychologia 1997; 35: 747-758.
10. Manly T, Robertson IH, Galloway M, Hawkins K. The absent mind: further investigations of sustained attention to response. Neuropsychologia 1999; 39: 661-670.
11. Robertson IH, Ward T, Ridgeway V, Nimmo-Smith I. The structure of normal human attention: the test of everyday attention. J Int Neuropsychol Soc 1996; 2: 525-534.
12. Campbell DNC, Lim M, Kerr Muir M, et al. A prospective randomised study of local versus general anaesthesia for cataract surgery. Anaesthesia 1993; 48: 422-428.


© 2002 European Society of Anaesthesiology