The need of functional assessment after anaesthesia is increasing because ‘fast-track anaesthesia' and patient-orientated perioperative management are expanding. There is a need for patients to recover fast, not only for medical reasons but also in their functional and cognitive abilities in order to become independent again as soon as possible. In this context, the assessment of psychomotor function in the intermediate postoperative period also becomes important.
Deterioration of psychomotor abilities within the early hours following surgery and anaesthesia has been demonstrated in several studies and quantified by changes in electrophysiological parameters [1-3]. In the intermediate recovery period, results are conflicting. Kubitz and colleagues demonstrated recovery of psychomotor functions within 24 h in patients older than 80 yr scheduled for cataract surgery . In contrast, Heath and colleagues demonstrated reduced memory performance in female patients, 24 h after day case dilatation and curettage . Tzabar and colleagues reported a greater number of cognitive lapses in younger patients when the patients used a cognitive failures questionnaire, which allows self-assessment. They did not use objective testing .
It is known from long-term outcome studies that cognitive failure may persist, especially in elderly patients . In a subsequent study, age was identified as a risk factor for late postoperative cognitive dysfunction (POCD). Inpatient surgery was another factor . Rasmussen and colleagues showed that there was no difference in the incidence of POCD between patients receiving either general or regional anaesthesia after 3 months . Late cognitive decrement is of concern as it implies the possibility of permanent cerebral dysfunction. In these long-term outcome studies a large test battery, which takes about 45 min to complete each session, was used.
The Short Performance Test (Syndrom Kurztest (SKT)), which only takes about 10-15 min per session, has been shown to be especially sensitive for slight psychomotor impairment in neurological patients [9-11]. With respect to anaesthetics, the SKT has been used in volunteer studies to document reduced psychomotor function following the application of ketamine or midazolam/flumazenil [12,13].
The aim of this pilot study was to document the time course of psychomotor recovery by using the SKT in a typical clinical setting during a period of 24 h after elective surgery. The hypothesis was that psychomotor function is diminished after propofol/remifentanil anaesthesia. Owing to the character of this pilot study, we did not include a healthy control group, because it was not intended to establish a causal relation and to quantify our findings in comparison with the average population. It is already known that the cause of cognitive failure following anaesthesia and surgery is multifactorial.
After approval of the local Ethics Committee, 37 patients American Society of Anesthesiologists (ASA) Grade I-III scheduled for elective surgery under general anaesthesia were included in the study. All patients were native German speakers, aged 18 yr or older and gave written informed consent. Exclusion criteria were neurological diseases, consumption of medication affecting the central nervous system, inability to comply with the protocol or to follow procedures, cardiac surgery or neurosurgery. Patients with severe visual, auditory or neuromuscular diseases, drug dependence, alcoholism, a known intolerance of the used drugs and pregnancy were also excluded.
Every patient was premedicated with midazolam 0.1 mg kg−1 orally 45 min before the intended start of the procedure. Anaesthesia was induced with remifentanil 0.5 μg kg−1 min−1 as a continuous infusion. One minute later, a target-controlled infusion with propofol was commenced with an estimated plasma concentration of 3 μg mL−1. The patients received rocuronium 0.6 mg kg−1 before endotracheal intubation was performed. Propofol and remifentanil were used for maintenance of anaesthesia. Intraoperative capnography was used to ensure normocapnia.
Fifteen minutes before the end of surgery, the patients received novaminsulfon 2 g and piritramide 7.5 mg intravenously (i.v.). Postoperative pain was treated with piritramide 3-5 mg repeated on patient demand. The patients were allowed to receive piritramide up to 10 min before the testing started. In case the patients needed piritramide immediately before the testing was scheduled, the test was delayed for at least 10 min till the patient felt ready for the testing. The single doses and the total consumption of piritramide were documented.
The SKT is a timed test for assessing cognitive impairment of memory, motor skills and attention. Originally the SKT was introduced in Germany; however, the test has been translated and validated in English and other languages [10,14,15]. It has been shown in neurological patients to mirror excellently the course of early and moderate severe dementia and to be sensitive to therapy in these stages. It can also be applied to patients suffering from other organic or non-organic diseases affecting the brain or to quantify the effects of drugs acting on the central nervous system. It was designed with special emphasis on an attractive game-like character to motivate patients and thereby reduce changes in motivation, influencing the test consistency. The SKT consists of nine subtests. For every task, a maximum performance time of 60 s is allowed; therefore, the test battery is completed in about 10 min. There exist five parallel versions to reduce learning effects. We used the SKT because it did not seem practical to apply a large test battery of about 45 min per session in the early recovery period after the operation.
The following is a brief description of the nine SKT subtests. For more detailed information, see the studies by Erzigkeit and colleagues [9,10]. Subtest 1 (ST 1) requires 12 picture objects to be named. The time needed to name the shown objects correctly is recorded. Patients are instructed to keep the items in mind. Language fluency and cognitive speed performance are measured by this subtest. In subtest 2 (ST 2) the patient is asked to recall the 12 objects that were shown in ST 1. The number of missing items is recorded. This test measures immediate recall. Then there is a brief learning phase of 5 s, in which the same 12 objects are shown again. The patient is asked to keep them in mind for later recall. In subtests 3-5 (ST 3, ST 4, ST 5) the patient has to work with a magnetic board. First the patient has to read two-digit numbers (ST 3), then re-organize the numbers in ascending order by moving them from the upper part of the board to the lower part (ST 4) and finally replace the numerals to their original position (ST 5). For each test, the time needed is recorded. Motor skills, cognitive flexibility and speed of processing are measured. In subtest 6 (ST 6), three different symbols are presented in random order on a board. The patient has to count loudly one of the symbols, thereby neglecting the two distracter symbols. The time needed is recorded. Language fluency and cognitive speed performance are measured. Subtest 7 (ST 7) is an interference task. Two letters are presented on a board in random order. The patient has to read as fast as possible, but giving the opposite letter; e.g. if A and B are shown, the patient has to read B if A is presented and vice versa. If the patient makes a mistake, he is interrupted by the observer and has to correct his reading. The time needed is recorded. Cognitive rigidity is assessed by this subtest. In subtest 8 (ST 8) the patient is asked to recall the 12 objects that were shown in ST 1 at the beginning of the test. The number of missing items is recorded. The test measures the delayed recall. In subtest 9 (ST 9) the patient has to identify the 12 objects that were shown in ST 1, out of 48 objects presented on a big board. The number of missing items is recorded. This subtest quantifies the ability of recognition.
All patients were given a standardized introduction and explanation to the test. Patients were informed that they had to work as fast and as accurately as possible. The observer measured the time needed to complete each subtest with a stopwatch. If a patient was not able to complete a subtest in 60 s, the subtest was terminated and 60 s was documented as the result for that subtest. We defined ‘cognitive impairment' as a shorthand for reduced cognitive performance compared with baseline implicitly referring to the definition of psychomotor functioning that underlies the SKT.
The complete SKT was performed on the day before surgery (T0), 90 min after extubation (T90) and on the first day after anaesthesia (T1), approximately 24 h after surgery. In addition, we used ST 3, ST 4 and ST 5 to measure psychomotor skills 10 and 30 min immediately after extubation (T10, T30). Measurements at baseline and at the first postoperative day were carried out in a quiet room. The other sessions were performed in the postanaesthesia care unit (PACU). To minimize any learning effect we used different versions.
At extubation as well as 10 and 90 min after extubation, blood samples for propofol analysis were withdrawn into EDTA tubes and refrigerated until centrifugation within a few hours. Plasma was then transferred into storage tubes with no extra contents and frozen at −40°C until analysis. The propofol concentration in plasma was determined by high-performance liquid chromatography with a photodiode array detector (HPLC-DAD). For protein precipitation, 200 μl acetonitrile was added to 200 μL serum. After 5 min of vortexing and centrifugation, 50 μL of the supernatant was injected for HPLC. The detection wavelength was 219 nm with a bandwidth of 10 nm. The method was calibrated and validated according to the guidelines of the Society of Toxicological and Forensic Chemistry. The limits of detection and of quantification were 0.05 and 0.15 μg mL−1, respectively. Details about the used instruments and HPLC conditions were described elsewhere .
Non-invasive blood pressure, heart rate and oxygen saturation were monitored in the PACU. The patients self-assessed pain, fatigue and nausea using a labelled 100 mm visual analogue scale (VAS) in the PACU. In a purposely designed protocol, we documented patients' characteristics, drugs administered during surgery and the VAS and SKT scores. All data were stored on computer disk and transferred to our main database afterwards.
This was a pilot study because the SKT has not been used in the perioperative setting before. Therefore, effect size estimates were not available to calculate test power or to determine the required sample size given a desired power of 80%. We recruited as many patients as possible with the resources available. With respect to the central limit theorem, we expected a group of patients with n > 30 sufficient for interpreting results. For all calculations, the raw data of the SKT were used. The distribution of data was tested using the Kolmogorov-Smirnoff test. Since not all data were normally distributed, non-parametric statistics were applied to analyse differences among different measurements (Friedman test, Wilcoxon signed rank sum test). In order to avoid alpha inflation, we first ran a global omnibus test (Friedman test), integrating all nine SKT subtests, before analysing the subtest separately. For all significant findings, explained variance (η2) based on a parametric analysis is reported as a proxy of the effect size. For non-significant effects, the observed power is reported based on the empirical estimate of the effect size with the present sample. Non-parametric correlations were calculated for the propofol plasma concentration and the relative changes in psychomotor performance at T10 and T90. A value of P < 0.05 was considered significant. Retrospectively, we defined one age subgroup, for descriptive reasons only, because in many studies POCD is predominantly considered as a problem for elderly patients . Patients aged between 18 and 60 yr were analysed separately. The same analysis as for the whole collective was applied to the subgroup.
Data from 36 patients (24 males, 12 females) were analysed. The mean (standard deviation (SD)) age was 53.1 (13.6) yr, height was 169.6 (8.9) cm and weight was 76.2 (16.2) kg. Mean (SD) extubation time was 9.2 (6.4) min and duration of anaesthesia was 102.2 (52.1) min. All patients except one completed the SKT in the PACU. One patient refused to participate further in the study, because he was not able to concentrate any more after being told of his serious diagnosis immediately after surgery. Therefore, his data were excluded from further analysis. Seven patients did not take part in the study at the first postoperative day because they were discharged before the scheduled time.
The patients performed the preoperative testing well. In comparison with the age-related scores given in the SKT manual, all patients except one reached normal scores for psychometric performance .
The omnibus repeated measurement test considering the summarized score of all nine subtests is highly significant (P ≤ 0.001), indicating substantial changes in psychomotor performance over the observational period (η2 = 68%; Fig. 1). Follow-up testing revealed markedly reduced performance 90 min after surgery for all SKT subtests (P ≤ 0.007 vs. baseline T0; Table 1; 12% < η2 < 73%). However, there was a continuous improvement in the performance comparing T10, T30 and T90 as indicated by the test results of ST 3, ST 4 and ST 5. The summarized score of ST 3, ST 4 and ST 5 was as follows: T0: 37 (16.5); T10: 90 (64); T30: 69 (40); T90: 52 (20); T1: 39 (13); P ≤ 0.001. Data are given as median (interquartile range). Details for the subtests over time are listed in Table 1. Detailed results of ST 5 are depicted in Figure 2.
The impairment of psychomotor function was still present in the three memory tasks ST 2, ST 8 and ST 9 on the first postoperative day (P ≤ 0.005; T1 vs. T0; 27% < η2 < 48%; power for the non-significant effects: between 7% and 31%). However, in eight subtests the results were significantly better than 90 min after extubation (P ≤ 0.005; T1 vs. T90; ST 9: P = 0.06).
We examined a subgroup of patients younger than 60 yr (n = 25; age 47±11 yr). The omnibus repeated measurement test considering the summarized score of all subtests indicated a significant difference among the three measurements (P ≤ 0.001). Except in ST 7, the post hoc comparison revealed significant differences in each subtest over time (P ≤ 0.02; for ST 7 P = 0.18). At T90 the patients' performances were significantly impaired for all subtests except for ST 7 (P ≤ 0.04 vs. baseline T0). In addition, except for ST 7 the patients demonstrated better results on the first postoperative day than at T90 (P ≤ 0.03). However, at T1 there was a reduced psychomotor performance left in four subtests, which compromised ST 4 in addition to the three memory tasks ST 2, ST 8 and ST 9 (P ≤ 0.03). As an example, detailed results for the summarized score and for ST 5 are shown in Figures 1 and 2.
Plasma concentration of propofol after extubation was 1.3 ± 1.1 μg mL−1. Data for T10 and T90 are given in Table 2. There were no significant correlations between propofol serum concentrations and relative changes in the SKT results at T10 and T90.
Data are given in Table 2 for mental scales and postoperative pain treatment. On average, the patients asked for additional pain treatment once during their stay in the PACU (interquartile range: 2). The patients received piritramide 4 mg (7 mg) i.v. in the PACU. One patient had an epidural catheter and six patients used a patient-controlled (PCA) device for analgesia (standard in our hospital). From the PCA patients, the doses of bolus application of opioids given before the patients used the PCA device were included in the calculation. The PCA device did not allow exact calculation, as long as the whole syringe was not used. This did not happen in the PACU; therefore, the PCA dose could not be included in the analyses.
With the SKT we demonstrated a pronounced decline in psychomotor performance in the early postoperative period following propofol/remifentanil-based anaesthesia. In part, psychomotor decrement was reversed at the first postoperative day, especially in the tasks measuring attention . In the memory subtests (ST 2, ST 8 and ST 9), some of the patients did not reach their preoperative level even 24 h after surgery; therefore, the results for the group were still significantly lower than the baseline values. Interpreting our results, it has to be considered that seven patients were not evaluated during the first postoperative day, because they were already discharged. This limits generalizing our findings to other patient populations, especially to ambulatory patients. Our results were similar when patients older than 60 yr were excluded for analysing retrospectively the subgroup of younger patients. This indicates that psychomotor decrement within 24 h after anaesthesia is not a problem of only elderly patients.
In two volunteer studies the SKT was used to document effects of anaesthetics on psychomotor functions. The complete SKT battery was used by Schlager and colleagues, who showed that psychometric abilities were significantly reduced after the oral application of midazolam 15 mg . The volunteers regained their baseline levels after midazolam was antagonized with flumazenil. Engelhardt and colleagues  used ST 3, ST 4 and ST 5 to investigate psychometric effects after infusion with (s)-ketamine and racemic ketamine. The mean times to reach baseline test performance were significantly shorter in the volunteers who received (s)-ketamine.
So far, there is no gold standard to measure psychometric functions in the early and intermediate postoperative period. The SKT is sensitive in different patient populations to discrete changes in psychomotor performances [10,11]. Norms exist for four age groups and three levels of premorbid intelligence . In the present study, we worked with the raw data in order to use the precise information the original test offers. Norm data were calculated to determine psychomotor performance in the preoperative state only. However, comparison with reference data, as described in the SKT manual, remains limited, because we did not know the intelligence quotient of our patients (intelligence quotient was guessed as normal). We had the intention for a within-subject comparison design. It was not the objective of the study to compare the patients with the average population.
Kim and colleagues  confirmed the presence of two primary factors of attention and memory in the SKT by factor analyses. The test-retest reliability of these two primary factors was estimated to be 0.75 and 0.93, respectively. Interestingly, in the prolonged recovery period 24 h after anaesthesia, memory functions measured with SKT ST 2, ST 8 and ST 9 are selectively affected, while the scores in the other tests, measuring attention, had regained baseline levels. Underlying mechanisms remain unclear, but obviously brain structures involved in memory performance are more sensitive during the postoperative course than ones activated during attention tasks.
There is an ongoing discussion among experts in the field of dementia research, about the sensitivity and specificity of the SKT and the Mini-Mental State Examination (MMSE) . The MMSE is a short mental state questionnaire and can be applied in about 5 min. Its sensitivity increases with the grade of dementia, from 20% in the early stages to 87% in the more advanced stages. Validation studies in the perioperative setting are not available despite the MMSE being used to study recovery characteristics comparing different anaesthetic regimens [21,22]. One limitation of the MMSE is a pronounced learning effect, which has been demonstrated in normal adults, even if the test has been presented again after 3 months . By applying different versions of the SKT, we attempted to reduce learning effects in our study. However, we cannot exclude a bias of learning effects in our results. A further limitation interpreting our results is that postoperative testing was done in the PACU. This may be associated with increased variability because of the disturbance and distraction of the patients. However, this holds true for many studies performed in the early postoperative period.
Our results replicate and extend previous studies assessing psychometric dimensions perioperatively. Heath and colleagues demonstrated reduced memory performance in patients, who received propofol, assessed 24 h after day case dilatation and curettage. The authors used the Wechsler logical memory function passages . Biedler and colleagues studied patients scheduled for day case gynaecological laparoscopy after propofol/remifentanil or sevoflurane/fentanyl anaesthesia with a self-composed test-battery . Verbal learning abilities were equal to the preoperative level after 4 h. Reaction times, concentration and speed of information processing remained diminished. The different sensitivity of the memory test battery and the different patient samples (inpatient vs. outpatient surgery, females only) may explain, to some extent, why the patients in the study by Biedler and colleagues did not show any decline in memory function after 4 h.
Our data are in contrast to the findings of Kubitz and colleagues, who did not demonstrate diminished memory function after 30 and 90 min in a patient population older than 80 yr . The octogenarians received either balanced anaesthesia or a total intravenous anaesthesia with propofol/remifentanil for cataract surgery. Again, this opens up the question of test sensitivity. Kubitz and colleagues used a short-term memory test, where the patients had to recall five words that were presented twice 2 min before the testing. However, causes of cognitive dysfunction in the intermediate postoperative period are multifactorial. The effects of the analgesics used for postoperative pain treatment, postoperative pain itself, sleep deprivation and the type of surgery may play a major role in addition to cerebral effects of the anaesthetics. Unfortunately, whether the patients complained about pain or whether they needed any analgesics was not reported by Kubitz and colleagues. Our patients assessed their pain by the VAS and indicated a medium range. The interpretation of the role of pain and analgesics on our results is limited, because the VAS and medication were recorded during the stay in the PACU only. Further studies need to include documentation of the VAS and the used medication for the whole period of psychometric assessment.
Our study opens up the question of whether the impairment of cognitive functions, which we have shown, is of any clinical relevance. Should we inform our patients about the risk of a reduced memory function even 24 h following anaesthesia? We cannot answer this question yet. The SKT is a promising tool to address this question in further studies, especially quantifying the cognitive deficit and comparing different patient groups.
This study was supported by AstraZeneca GmbH from D-22880 Wedel/Germany and by a grant of the Charité - Universitätsmedizin Berlin. The results have been presented in part at the ‘Haupstadtkongress für Anästhesie und Intensivmedizin' in Berlin, Germany, 2004 and at the 16th European Students Conference in Berlin, Germany, 2005.
1. Biedler A, Juckenhöfel S, Feisel C, Wilhelm W, Larsen R. Cognitive impairment in the early postoperative period
and sevoflurane-fentanyl anesthesia. Anaesthesist
2000; 49: 286-290.
2. Kubitz J, Epple J, Bach A, Motsch J, Martin E, Schmidt H. Psychomotor recovery in very old patients after total intravenous or balanced anaesthesia for cataract surgery. Br J Anaesth
2001; 86: 203-208.
3. Rundshagen I, Schnabel K, Schulte am Esch J. Recovery of memory performance after general anaesthesia: clinical and electrophysiologic findings. Br J Anaesth
2002; 88: 362-368.
4. Heath PJ, Ogg TW, Gilks WR. Recovery after day-case anaesthesia. A 24-h comparison of recovery after thiopentone or propofol
1990; 45: 911-915.
5. Tzabar Y, Asbury AJ, Millar K. Cognitive failures after general anaesthesia for day case surgery. Br J Anaesth
1996; 76: 194-197.
6. Moller JT, Cluitmans P, Rasmussen LS et al
. Long-term postoperative cognitive dysfunction in the elderly: ISPOCD 1 study. Lancet
1998; 351: 857-861.
7. Canet J, Raeder J, Rasmussen LS et al
. Cognitive dysfunction after minor surgery in the elderly. Acta Anaesthesiol Scand
2003; 47: 1204-1210.
8. Rasmussen LS, Johnson T, Kuipers HM et al
. Does anaesthesia cause postoperative cognitive dysfunction? A randomised study of regional versus general anaesthesia in 438 elderly patients. Acta Anaesthesiol Scand
2003; 47: 260-266.
9. Erzigkeit H. The SKT: a short cognitive performance test as an instrument for the assessment of clinical efficacy of cognition enhancers. In: Bergner M, Reisberg B, eds. Diagnosis and Treatment of Senile Dementia
. Berlin: Springer, 1989: 164-174.
10. Lehfeld H, Erzigkeit H. The SKT - a short cognitive performance test for assessing deficits of memory and attention. Int Psychogeriatr
1997; 9 (Suppl 1): 115-121.
11. Overall JE, Schaltenbrand R. The SKT neurophsychological test battery. J Geriatr Psychiatry Neurol
1992; 5: 220-227.
12. Schlager A, Morawetz RF, Huber C, Furtwaengler W, Luger TJ, Schreithofer D. Residual impairment of cognitive abilities and reaction time caused by midazolam after reversion with flumazenil. Anaesthesist
1991; 40: 33-38.
13. Engelhardt W, Stahl K, Marouche A, Hartung E. Recovery time after (S)-ketamine or ketamine racemate. Recovery time after short anesthesia in volunteers. Anaesthesist
1998; 47: 184-192.
14. Lehfeld H, Rudinger G, Rietz C et al
. Evidence of the cross-cultural stability of the factor structure of the SKT short test for assessing deficits of memory and attention. Int Psychogeriatr
1997; 9 (Suppl 2): 139-153.
15. Fornazzari L, Cumsille F, Quevedo F et al
. Spanish validation of the Syndrom Kurztest (SKT). Alzheimer Dis Assoc Disord
2001; 15: 211-215.
16. Pragst F, Herzler M, Erxleben BT. Systematic toxicological analysis by high-performance liquid chromatography with diode array detection (HPLC-DAD). Clin Chem Lab Med
2004; 42: 1325-1340.
17. Parikh SS, Chung F. Postoperative delirium in the elderly. Anesth Analg
1995; 80: 1223-1232.
18. Erzigkeit H. SKT Manual
. Weinheim, Germany: Beltz Test GmbH, 1989.
19. Kim YS, Nibbelink DW, Overall JE. Factor structure and scoring of the SKT test battery. J Clin Psychol
1993; 49: 61-71.
20. Ihl R, Grass-Kapanke B, Jaenner M, Weyer G. Neurospychometric tests in cross sectional and longitudinal studies - a regression analysis of ADAS - Cog, SKT and MMSE. Pharmacopsychiatry
1999; 32: 248-254.
21. Bekker AY, Berklayd P, Osborn I, Bloom M, Yarmush J, Turndorf H. The recovery of cognitive function after remifentanil
nitrous oxide anesthesia is faster than after an isoflurane nitrous oxide-fentanyl combination in elderly patients. Anesth Analg
2000; 91: 117-122.
22. Chen X, Zhao M, White P et al
. The recovery of cognitive function after general anesthesia in elderly patients: a comparison of desflurane and sevoflurane. Anesth Analg
2001; 93: 1489-1494.
23. Helkala EL, Kivipelto M, Hallikainen M et al
. Usefulness of repeated presentation of Mini-Mental State Examination as a diagnostic procedure - a population-based study. Acta Neurol Scand
2002; 106: 341-346.