Anxiety is a frequent patient concern during the perioperative period. Apart from compromising patients' well-being, it may affect induction of anesthesia. As suggested by clinical experience, the more anxious the patient, the higher the hypnotic dose required for induction of anesthesia (propofol being the hypnotic drug most frequently investigated).1,2 However, this assumption has been recently challenged.3
Some studies that investigated the effects of perioperative anxiety recorded the propofol dose required to obtain loss of consciousness (LOC) or a defined level of sedation1,2 but did not consider the hemodynamic effects of anxiety. Anxiety and acute mental stress are associated with increased heart rate (HR) and cardiac output.4–8 Because hemodynamic status during induction of anesthesia may modify propofol requirements for LOC,9–12 we hypothesized that HR before induction of anesthesia may affect propofol dose requirements for LOC and that anxiety before induction of anesthesia may influence HR. Although intuitive, this latter relationship has not been previously demonstrated in the immediate preinduction setting.
We therefore designed this study to address, first, the effect of perioperative HR on the propofol dose required for LOC and, second, the effect of perioperative anxiety on HR.
The study was conducted between June 2007 and February 2008. It was approved by the Ethics Committee of Cochin Hospital, Assistance Publique-Hôpitaux de Paris. Written informed consent was obtained from each patient.
Consecutive female patients, 18-65 yr old, ASA physical status I or II, undergoing scheduled gynecological surgery were approached to participate in the study. Exclusion criteria were as follows: pregnancy; neurologic or psychiatric disease; chronic medication with β-blockers, anxiolytics, antidepressants, or opioids; documented alcohol or drug abuse; or inability to complete the anxiety forms.
Assessment of Anxiety
Anxiety was assessed with the Spielberger's State-Trait Anxiety Inventory (STAI) form.13 This is a validated tool for self-reporting anxiety, already used during the perioperative period.1–3,14 It comprises 2 sets of 20 statements.
The first set relates to the immediate situation, the state anxiety. It includes statements such as “I feel calm” or “I am worried.” For each statement, the patient is required to select 1 of 4 responses: not at all, somewhat, moderately so, or very much so. To the best of our knowledge, there is no published cutoff value of the STAI-state score discriminating the anxious from nonanxious patient.
The second set of statements is intended to reflect underlying long-term or trait anxiety. Examples include “I feel pleasant” or “I am a steady person.” Again, the subject is required to select a single response: almost never, sometimes, often, or all the time.
A score for each set, ranging between 20 and 80, may then be calculated by an investigator using a scoring key. Higher scores indicate higher anxiety. The score for the STAI-trait set does not change as the context changes, contrary to the score for the STAI-state. It takes about 10 min to complete both forms.
The patients were asked to complete the 2 STAI forms (state and trait) on the morning of surgery, while waiting in their room in the surgical ward and the STAI-state form in the operating room, immediately before induction of anesthesia.
Assessment of HR and Induction of Anesthesia
No preanesthetic medication was given on the morning of surgery. On arrival in the operating room, a 20-gauge IV catheter was inserted in the left hand. A 3-way stopcock was connected to the catheter and devoted to propofol infusion. Intravenous lidocaine to decrease pain caused by propofol infusion was forbidden. For perioperative warming, a forced-air blanket (Bair Hugger®, Arizant, Eden Prairie, MN) was used because it may affect STAI results.14 Patients were monitored with noninvasive arterial blood pressure, 5-lead electrocardiogram, and pulse oximetry.
Once the patient had completed the STAI-state form in the operating room, HR and systolic blood pressure (SBP) were measured. HR and SBP values used for further analysis were the mean of 3 consecutive measurements.
Anesthesia was induced with a continuous 200-mL/h 1% propofol infusion (AstraZeneca, Rueil-Malmaison, France) with the Base Primea® pump (Fresenius-Vial, Brezins, France) until LOC, defined as loss of verbal contact. Loss of verbal contact was tested every 15 s from the start of propofol infusion by asking the patient, in a normal voice but without tactile stimulation, to say her name. Until LOC, the airway was managed with a facemask but without jaw thrust to avoid patient stimulation. The same investigator identified when LOC occurred and inserted a mark in the Rugloop® (Demed Company, Temse, Belgium) recording. The investigator was blinded to the STAI form results. The software running the pump allowed for real-time calculation of propofol effect-site concentrations (Ce) according to the pharmacokinetic model of Schnider et al.15 This model considers patient's age, sex, and lean body mass. The propofol dose and Ce were continuously recorded until LOC and stored on a personal computer with Rugloop software (http://www.demed.be). After LOC, the continuous infusion was switched to a Ce target-controlled infusion, and the study was terminated.
Rugloop files were analyzed a posteriori. LOC was identified through the mark inserted in the operating room. Predicted Ce and propofol dose at the time of LOC were retrieved.
Results are presented as median (interquartile range; range). The associations between STAI-trait, STAI-state in the ward and in the operating room, HR and SBP versus dose, and Ce at the time of LOC were studied with the Spearman test. A P value <0.01 was considered statistically significant because 5 associations were tested for dose and Ce. Associations with a P value <0.05 were entered into a multiple regression analysis. Analysis was conducted with StatView software (SAS Institute, Cary, NC).
For comparison of STAI-state measured in the ward and in the operating room, we used a Wilcoxon test, and concordance between the 2 values was measured with the concordance correlation coefficient. A P value <0.05 was considered statistically significant. Analysis was conducted with StatView software (SAS Institute).
For calculation of the sample size, we used a method for studies involving linear regression because all the variables tested were continuous quantitative variables. Our main hypothesis was to demonstrate a significant relationship between HR (the x axis or independent variable) in the operating room and propofol dose for LOC (the y axis or dependent variable). With the method described by Dupont and Plummer,16 calculation requires knowing the sd of the x variable (HR) and the y variable (propofol dose), which was obtained for the first 20 patients included in the study (78 ± 15 per minute and 116 ± 25 mg, respectively), and to select an α and β risk (0.05 and 0.20, respectively). In addition, you must define what the expected regression coefficient between the x and y variables may be. We chose a 0.5 value according to Kazama,17 who studied the determinants of propofol induction dose requirements expressed in milligram with a methodology similar to ours. He obtained correlation coefficients between 0.4 and 0.6. Therefore, we selected the intermediate value of 0.5. With the aforementioned values, 34 patients had to be included (PS-Power and sample size calculation software).16
Forty-five patients were enrolled. Their characteristics are presented in Table 1. One patient was not able to complete the STAI-state form in the operating room because of a panic attack.
The time that elapsed between the 2 STAI-state forms was 3 h (1-5 h; range: 0.5-22 h). No difference was observed between the score obtained for the STAI-state assessment in the ward and in the operating room (Table 1). Upon LOC, the propofol dose was 107 mg (97-128 mg; range: 79-183 mg), and the predicted Ce was 4.8 μg/mL (4.4-5.3 μg/mL; range: 3.5-6.7 μg/mL).
On the morning of surgery, no significant correlation was observed between STAI-state (situational anxiety) and propofol dose for LOC and between STAI-trait (proneness to anxiety) and propofol dose (Table 2). Similar results were observed with Ce.
In the operating room, no significant correlation was observed between STAI-state (situational anxiety) and propofol dose required for LOC (Table 2). Similar results were observed with Ce. A significant relationship was observed between HR and propofol dose at LOC but not between SBP and propofol dose (Table 2). Similar results were observed with Ce. In multiple regression, analyzing the effect of STAI-state in the operating room and HR in the operating room on propofol dose and Ce at LOC, only HR was significant (P = 0.0011 for propofol dose and P = 0.0021 for propofol Ce) (Fig. 1). In the operating room, a significant relationship was observed between STAI-state and HR but not between STAI-state and SBP (Fig. 2).
Of the variables we investigated, HR recorded just before starting the propofol infusion was the only predictor of propofol dose or Ce for LOC in multiple regression analysis. This is in agreement with the results of Morley et al.,3 who also demonstrated a significant relationship between preinduction HR and the propofol dose required for loss of verbal response or to obtain a bispectral index (BIS) value of 50.
However, our results do not exclude an effect of anxiety on propofol dose or Ce but rather support an indirect effect. We observed a strong relationship between STAI-state in the operating room and HR, i.e., the more anxious the patient, the more rapid the HR. Although intuitive, this relationship has not been demonstrated in the immediate preinduction setting. Anxiety is associated with an increase in both HR and cardiac output.4–7 Similarly, acute mental stress, as would be expected in the operating room, is also responsible for increased HR and cardiac output.8 Anxiety and stress-induced adrenaline release may account for this hemodynamic pattern as suggested by Fell et al.18 However, anxiety is only one of the determinants of HR; anemia, dehydration, or other perioperative factors that we did not control for in this study may influence it. Conflicting results have been reported in the 3 published studies concerning the effects of anxiety on propofol dose for producing a defined end point.1–3 Maranets and Kain1 reported that increased STAI-trait but not STAI-state on the day of surgery was associated with higher propofol dose administered as a bolus to achieve a BIS value between 40 and 60. Similarly, Hong et al.2 reported that increased anxiety assessed by a visual analogic scale but not by the STAI scale on the day of surgery was associated with higher propofol dose administered with a target-controlled infusion to reach a predetermined level of sedation. On the contrary, Morley et al.3 did not observe any influence of the STAI-trait and STAI-state measured in the surgical ward and propofol dose administered at a 40 mg · kg−1 · h−1 infusion rate to achieve LOC or a BIS value of 50. In this study, we did not observe a relationship between state and trait anxiety measured in the ward and in the operating room and propofol dose or Ce for LOC and suggest that the effect of anxiety on propofol dose is probably an indirect one, related to HR changes. However, we cannot exclude a type 1 error, i.e., an underpowered study, because the calculation of the sample size relied only on the demonstration of a significant relationship between HR in the operating room and propofol dose and did not consider the multiple comparisons we made. This hypothesis is suggested by the trend toward statistical significance we observed between STAI-state in the ward and in the operating room and propofol dose, with P values between 0.01 and 0.05. Apart from an effect of HR, the depth of anesthesia used as an end point may be responsible for the discrepancies observed among studies because the hemodynamic effect of anxiety, and probably cardiac output, may disappear as the anesthesia level deepens, reducing the dose discrepancies in anxiety states. In this study, we chose LOC as our clinical end point because it is the commonly accepted clinical end point used in many clinical investigations of the effects of hypnotic drugs.12,15,19 Furthermore, studying loss of responsiveness to a noxious stimulus such as laryngoscopy or tracheal intubation, which is of more clinical interest to the anesthesiologist, would have entailed the coadministration of opioids and made analysis of the results more complex. Apart from a pharmacologic mechanism, selection of the end point and propofol infusion rate may induce a measurement error. The error of dose can be relatively high if a bolus of propofol is given because LOC is assessed clinically at intervals of several seconds. We chose to administer a slow continuous infusion to minimize error. If a target BIS value is chosen as the end point for depth of anesthesia, this introduces several other potential sources of error. First, BIS is calculated from the raw electroencephalogram with a lag time of 30-45 s.20 The BIS value upon LOC probably underestimates the true level of depth of anesthesia and may explain why BIS values at LOC are high in many studies. Second, if a low BIS value is the end point as in the studies by Maranets and Kain1 and Morley et al.,3 this is associated with another potential source of error because the relationship between propofol dose and BIS value may not be linear. Taking into consideration the aforementioned sources of error, we estimate that the method associated with the lowest risk of error is the one we used: continuous propofol infusion and LOC estimation every 15 s.
Lack of arterial propofol concentration21 and of cardiac output measurements in our study makes it difficult to evaluate mechanisms responsible for the increased propofol dose required when HR, and probably anxiety, increase. In other words, it is not possible from the current data to determine whether the increased requirements are related to a pharmacokinetic or a pharmacodynamic change. Previous studies have demonstrated that hemodynamic status, especially cardiac output, modifies propofol requirements.9–12 Because there is a good relationship between HR and cardiac output in healthy patients but not between SBP and cardiac output, our results suggest that increased propofol requirements for LOC when HR increases are related to increased cardiac output.22 This assumption may be further supported by the effect of preanesthetic medication with β-blockers on propofol requirements.23
The model by Schnider et al.15 used to predict propofol Ce concentration from infused propofol dose considers patient age, weight, sex, and height. Because the model does not consider the hemodynamic status, i.e., HR, values obtained in this study should be accepted with caution. Although we studied healthy patients and tried to decrease the variability of the pharmacokinetic model24,25 and the inaccuracy of estimated propofol Ce concentration during the first minute of infusion26 with a slow infusion rate for inducing anesthesia (200 mL/min), we cannot be certain of the accuracy of the predicted Ce in this study.
This study was an explanatory study and did not intend to draw practical considerations, i.e., dose or target selection for inducing anesthesia. However, the wide variability of both propofol dose (range: 79-183 mg) and predicted Ce (3.5-6.7 μg/mL) highlights that the best dose or concentration choice for induction of anesthesia should be based on the individual titration of dose or concentration using a clinical end point for titrating. This titration is of particular importance during induction because the hemodynamic effect of anxiety, and probably cardiac output, may fade as the anesthesia level deepens.
Because all the variables tested in this study were continuous quantitative variables (propofol dose, propofol Ce, anxiety scores, HR, and SBP), we decided to use a regression analysis. We could have used 2 groups of patients for our analysis but that would imply separating the patients into anxious versus nonanxious with an arbitrary cutoff value that could not be defined from the published data.
In conclusion, increased perioperative HR is responsible for increased propofol dose required for LOC. Perioperative situational anxiety accounts for increased HR and propofol requirement.
1. Maranets I, Kain Z. Preoperative anxiety and intraoperative anesthetic requirements. Anesth Analg 1999;89:1346–51
2. Hong JY, Kang IS, Koong MK, Yoon HJ, Jee YS, Park JW, Park MH. Preoperative anxiety and propofol requirement in conscious sedation for ovum retrieval. J Korean Med Sci 2003;18:863–8
3. Morley A, Papageorgiou C, Marinaki A, Cooper D, Lewis C. The effect of preoperative anxiety on induction of anaesthesia with propofol. Anaesthesia 2008;63:467–73
4. Piccirillo G, Elvira S, Bucca C, Viola E, Cacciafesta M, Marigliano V. Abnormal passive head-up tilt test in subjects with symptoms of anxiety power spectral analysis study of heart rate and blood pressure. Int J Cardiol 1997;60:121–31
5. Tenenbaum G, Milgram RM. Trait and state anxiety in Israeli student athletes. J Clin Psychol 1978;34:691–3
6. Bergamaschi M, Longoni AM. Cardiovascular events in anxiety: experimental studies in the conscious dog. Am Heart J 1973;86:385–94
7. Guyton AC, Hall JE. Cardiac output, venous return, and their regulation. In: Guyton AC, Hall JE, eds. Textbook of medical physiology. 10th ed. Philadelphia: WB Saunders, 2000:210–22
8. Gregg ME, James JE, Matyas TA, Thorsteinsson EB. Hemodynamic profile of stress-induced anticipation and recovery. Int J Psychophysiol 1999;34:147–62
9. Upton R, Ludbroock G, grant C, Martinez A. Cardiac output is a determinant of the initial concentrations of propofol after short-infusion administration. Anesth Analg 1999;89:545–52
10. Johnson KB, Egan TD, Kern SE, White JL, McJames SW, Syroid N, Whiddon D, Church T. The influence of hemorrhagic shock on propofol. A pharmacokinetic and pharmacodynamic analysis. Anesthesiology 2003;99:409–20
11. Myburgh JA, Upton RN, grant C, Martinez A. Epinephrine, norepinephrine and dopamine infusions decrease propofol concentrations during continuous propofol infusion in an ovine model. Intensive Care Med 2001;27:276–82
12. Adachi YU, Watanabe K, Higuchi H, Satoh T. The determinants of propofol induction of anesthesia dose. Anesth Analg 2001;92:656–61
13. Spielberger CD. Manual for the State-Trait Anxiety Inventory (STAI form T). Palo Alto, CA: Consulting Psychologists Press, 1983
14. Kimberger O, Illievich U, Lenhardt R. The effect of skin surface warming on pre-operative anxiety in neurosurgery patients. Anaesthesia 2007;62:140–5
15. Schnider TW, Minto CF, Shafer SL. The influence of age on propofol pharmacodynamics. Anesthesiology 1999;90:1502–16
16. Dupont WD, Plummer WD. Power and sample size calculations for studies involving linear regression. Control Clin Trials 1998;19:589–601
17. Kazama T. Relation between initial blood distribution volume and propofol induction dose requirement. Anesthesiology 2001;94:205–10
18. Fell D, Derbyshire D, Maile C, Larsson I, Ellis R, Achola K, Smith G. Measurement of plasma catecholamine concentrations. An assessment of anxiety. Br J Anaesth 1985;57:770–4
19. Mongardon N, Servin F, Perrin M, Bedairia E, Retout S, Yazbeck Y, Faucher P, Montravers P, Desmonts JM, Guglielminotti J. Predicted propofol effect-site concentration for induction and emergence of anesthesia during early pregnancy. Anesth Analg 2009;109:90–5
20. Pilge S, Zanner R, Schneider G, Blum J, Kreuzer M, Kochs EF. Time delay of index calculation: analysis of cerebral state, bispectral, and narcotrend indices. Anesthesiology 2006;104:488–94
21. Coetzee JF, Glen JB, Wium CA, Boshoff L. Pharmacokinetic model selection for target controlled infusions of propofol. Assessment of three parameter sets. Anesthesiology 1995;82:1328–45
22. Weibel ER, Taylor CR, Hoppeler H. The concept of symmorphosis: a testable hypothesis of structure-function relationship. Proc Natl Acad Sci USA 1991;88:10357–61
23. Ghosh I, Bithal PK, Dash HH, Chaturvedi A, Prabhakar H. Both clonidine and metoprolol modify anesthetic depth indicators and reduce intraoperative propofol requirement. J Anesth 2008;22:131–4
24. Doufas AG, Bakhshandeh M, Bjorksten AR, Shafer SL, Sessler DI. Induction speed is not a determinant of propofol pharmacodynamics. Anesthesiology 2004;101:1112–21
25. Hu C, Horstman DJ, Shafer SL. Variability of target-controlled infusion is less than the variability after bolus injection. Anesthesiology 2005;102:639–45
© 2010 International Anesthesia Research Society
26. Avram MJ, Krejcie TC. Using front-end kinetics to optimize target-controlled drug infusions. Anesthesiology 2003;99:1078–86