| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Departments of *Anesthesia and Surgery, Institut Gustave Roussy, Villejuif, France
Address correspondence and reprint requests to Dr. V. Billard, Département d’Anesthésie, Institut Gustave Roussy, 39 rue Camille Desmoulins, 94805 Villejuif, France. Address e-mail to billard{at}igr.fr
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
IMPLICATIONS: Given as a target-controlled infusion for thyroid surgery, sufentanil 0.3 ng/mL for intubation and 0.2 ng/mL during surgery, combined with propofol 4 µg/mL (corresponding to a maintenance infusion rate of approximately 7–10 mg · kg-1 · h-1), is recommended to ensure both optimal intraoperative stability and fast recovery.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Propofol-opioid combinations for total IV anesthesia (TIVA) have been extensively studied (5–7). However, most of the published data were extrapolated to sufentanil from clinical studies using other opioids, assuming a fixed potency ratio (e.g., sufentanil/alfentanil = 630) (4). Some studies focused on LOC, or response to intubation and skin incision. Whereas propofol-opioid combinations have been extensively studied for intraabdominal surgery (4), interactions between these drugs during TIVA for thyroid surgery have not been described. Because thyroid surgery involves a very sensitive area, drug requirements are often unpredictable. Also, there is increased risk for postoperative nausea and vomiting after thyroid surgery, and the antiemetic effect of propofol could be an advantage.
In this study, our aim was to assess the clinical use of propofol-sufentanil TIVA for thyroid surgery, and to quantify the influence of the sufentanil concentration on propofol requirements, stability of anesthesia, and recovery.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Patients were randomly assigned to 1 of 3 groups, to receive a fixed sufentanil target concentration (STC) of 0.1, 0.2, or 0.3 ng/mL. Randomization was done by blocks of six patients using PIGAS software and stratified to have a similar sex ratio in the three groups. The study was open, i.e., the investigator was aware of the STC.
Protocol
One hour preoperatively, all patients received hydroxyzine 50–100 mg orally. In the operating room, standard monitoring included electrocardiography, pulse oxymetry, and noninvasive blood pressure measurement. The level of hypnosis was monitored by using the electroencephalogram bispectral index® (BISTM), using an A2000TM monitor, software version 3.3 (Aspect Medical Systems, Newton, MA).
Sufentanil (diluted to 1 µg/mL) was administered IV as an effect-site target-controlled infusion (TCI) using Stanpump software (Dr. S. L. Shafer, Stanford University, CA1) controlling a Graseby 3400TM syringe pump. The Gepts pharmacokinetic model (three-compartment model, not adjusted for weight, age, or sex) was used (8).
The sufentanil TCI was started after 3 min of preoxygenation, at the randomized STC. Then, a propofol TCI was started with an initial target plasma concentration of 4 µg/mL, using a DiprifusorTM (Astra Zeneca, London, UK) based on a three-compartment pharmacokinetic model adjusted to weight (9). If LOC was not obtained after 3 min, the target propofol concentration was increased by steps of 1–2 µg/mL every 2 min. After LOC, atracurium (0.5 mg/kg) was given IV to facilitate orotracheal intubation, performed 2 to 5 min later. Then, the lungs of the patients were ventilated with a 50:50 mixture of air and oxygen and ventilation was adjusted to maintain PETCO2 between 30 and 35 mm Hg.
The STC was maintained constant until thyroid removal, then stopped during immediate histologic examination. If the histology showed a benign pathology, skin closure was completed with propofol only. In case of malignancy, sufentanil TCI was started again at the initial STC to complete appropriate lymph node dissection.
Adequate anesthesia was defined by all the following conditions being met: 1. mean arterial blood pressure (MBP) between 70% and 130% of a reference value defined as the mean of three measurements—preoperative visit, time of preanesthetic medication, and the preinduction value, 2. BISTM between 40 and 60, and 3. no somatic responses (movements or swallowing) to surgical stimulation.
Propofol, fluid administration, and other drugs were given as indicated to achieve those criteria according to a decision matrix described in Table 1 (10). Additional atracurium was administered when movements disturbed the surgeon despite adequate BISTM and MBP values.
|
Postoperative Care
Pain was assessed at regular intervals for 2 h by using a verbal pain score then a visual analog scale ranging from 0 to 10. Postoperative pain was controlled by using propacetamol 2 g IV, administered at the beginning of skin closure, supplemented in the recovery room by morphine 3 mg every 5 min until the visual analog scale score was below 3. When postoperative nausea or vomiting (PONV) occurred, either metoclopramide (10–20 mg) or ondansetron (4–8 mg) was given. After discharge from the recovery room, all patients received oral paracetamol, mephenesine, and maxicaine to control throat and neck pain. If pain relief was not achieved, subcutaneous morphine (0.5–1 mg every 4–6 h) and oral paracetamol (1 g every 6 h) were given.
Data Recording and Analysis
MBP, heart rate (HR), BISTM values, target propofol concentration, and predicted effect-site propofol and sufentanil concentrations were recorded every 5 min during anesthesia, before and after intubation, and before each change in a target concentration. Motor responses and additional treatments (for cardiovascular support, pain, or PONV) were noted.
Quality of anesthesia was assessed during separate periods: from induction to incision, from incision to thyroid removal, and during closure, by the following criteria:
Hemodynamic and BISTM response to intubation were specifically analyzed.
Propofol effect-site concentrations necessary to achieve adequate anesthesia was analyzed: 1) before intubation, after LOC, 2) before incision, and 3) at thyroid removal—during the stimulating surgical period.
Total doses of propofol and sufentanil were calculated per patient, and normalized as mass unit/h.
During recovery, three end-points were distinguished: 1)adequate spontaneous ventilation, 2) positive response to a verbal command, and 3) extubation (assuming both other end-points were achieved).
For each end-point, the time from the end of anesthesia (propofol target concentration set to 0) and the predicted effect-site residual propofol and sufentanil concentrations were analyzed.
Twenty-four hours postoperatively, analgesic consumption and the incidence of nausea and vomiting were recorded and each patient was interviewed for global satisfaction, any recall of intraoperative events, pain score, and possible side effects.
The number of patients in each group was calculated based on an expected reduction of propofol requirement of 60% (2). With an interindividual variability of 50%, 15 patients per group should be necessary to achieve a power of 80%. Because the small number of patients in each group did not provide normal distribution of the variables, all results were expressed as median and range. Patients characteristics (age, weight) as well as all quantitative variables were compared among the three STC groups by a using the nonparametric Kruskal-Wallis test (11). Qualitative variables were compared by using an exact Fisher test. A P value < 0.05 was considered as statistically significant.
Pharmacodynamic Modeling
To interpolate the influence of sufentanil on the propofol concentrations necessary for anesthesia and recovery, predicted concentrations were fitted to the interaction model described by Vuyk et al. (12). This model assumed that propofol alone, sufentanil alone, or both together are all able to provide adequate anesthesia. The contribution of each drug to the effect depends on its concentration (Cpro or Csuf), expressed as a fraction of the concentration provid-ing the same effect when used alone (Cproalone or Csufalone) and also depends on a possible interaction expressed by the product of concentrations weighted by a proportional interaction coefficient e:
|
|
Solving this equation allows the propofol concentration to be expressed as a function of sufentanil concentration:
|
|
If the interaction between drugs is purely additive (no synergy), the interaction coefficient e = 0, and the equation becomes:
|
|
The predicted concentrations before intubation, before incision, at thyroid removal, and during recovery were fitted to both additive and nonadditive models by a simple least squares method using the solver function of Microsoft Excel version 7.0 for Windows 95. As the nonadditive model estimated one more variable ("e") than the additive model, the residual sum of squares of both fitted curves were compared by using a 
2 test with one degree of freedom. Then, the correlation coefficient was calculated for each fit. As 45 pairs of data points were analyzed, we considered a correlation between propofol and sufentanil concentrations to be significant when r > 0.30 (44 degrees of freedom, P < 0.05).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Quality and Stability of Anesthesia
From Induction to Incision.
LOC was achieved with the initial target propofol plasma concentration (4 µg/mL in plasma) in 30 patients (67%) independently of the STC group, with a median propofol effect-site concentration of 3.5 µg/mL. From 1 to 4 propofol target increases were necessary in the other 15 patients. Intubation was performed without changing the propofol target after LOC in all patients. At the time of intubation, sufentanil effect-site concentration was stable in all patients and identical to the target. Both pre- and postintubation MBP and HR values decreased slightly with increasing STC (Table 2). A major hypertensive response (MBP >130 mm Hg) was observed only in the 0.1 ng/mL STC group. Tachycardia >120 bpm occurred with the two smallest STC groups. Because of neuromuscular blockade, no motor response to intubation was observed.
|
During Surgery, Until Thyroid Removal.
Only 6 patients (respectively 1, 3, and 2) remained within the desired BISTM and MBP range for the whole period without any propofol target adjustment. For most patients (39/45), a median of 2 propofol adjustments (range, 1–15) was necessary to correct inadequate anesthesia, independently of the STC.
Hemodynamic stability differed significantly with the STC (Table 3). Hypertension requiring treatment, and tachycardia >120 bpm, were significantly more frequent in the small STC group, whereas hypotension and ephedrine treatment (6-mg bolus, repeated if necessary after the next MBP measurement) were more frequent in the large STC group. The patients given intermediate STC (0.2 ng/mL) had fewer hemodynamic side effects than the other two groups.
|
During Histologic Analysis, Local Hemostasis, and Closure.
MBP and BISTM remained within the desired range without propofol adjustment for 28 patients, whatever the STC. Sixteen other patients required 1 to 3 propofol adjustments, and 1 patient in the smallest STC group required 12 propofol adjustments because of high BISTM or movement. For this patient, skin closure lasted >60 min and the propofol infusion had been reduced too soon, triggering a motor response without awareness.
Influence of STC on Propofol Requirements
At intubation, the propofol effect-site concentration depended mainly on the propofol target previously needed to produce LOC (Fig. 1, top). It decreased slightly but not significantly with increasing STC.
|
At thyroid removal, again, the propofol concentration decreased significantly with increasing STC (P < 0.001, Fig. 1, bottom). The median propofol target at the end of thyroid removal was respectively 5.0, 4.0, and 2.5 µg/mL.
Total propofol consumption decreased significantly by 10% with increasing STC from 0.1 to 0.2 ng/mL, and by 36% when STC increased from 0.1 to 0.3 ng/mL (Table 4).
|
The times to respond to verbal commands and to extubation of the trachea did not differ significantly among groups. Nevertheless, the patient who was apneic for 39 min in the large STC group could only be extubated 71 min after the end of the propofol TCI.
Effect-Site Concentration at the Different Recovery End Points.
Residual sufentanil effect-site concentration when resuming spontaneous breathing increased significantly (P < 0.001) with increasing intraoperative STC (Fig. 2). The median concentration was 0.05 ng/mL in the small STC group, 0.10 ng/mL in the intermediate group, and 0.13 ng/mL in the large group. This increase was associated with a significant decrease (P < 0.01) of the propofol effect-site concentration at the same end-point (median values 2.7 versus 1.9 versus 1.3 µg/mL, respectively) and the relationship between sufentanil and propofol predicted concentrations at spontaneous breathing followed the usual profile of pharmacodynamic interactions (Fig. 2). Similar findings were observed at the recovery of response to verbal commands and extubation.
|
During the next 22 h, 32 patients required no morphine, 4 patients required 5–10 mg, and the morphine consumption was not recorded for the remaining 9 patients. Paracetamol consumption ranged from 0 to 8 g/24 h, with a median value of 4 g/24 h in all groups. The incidence of PONV was 33% in the recovery room and 27% in the next 22 h, with no difference among the groups.
Of the 37 patients questioned on the day after surgery, 18 were very satisfied with their anesthesia, 14 were satisfied, 5 were moderately satisfied, and none was unhappy, without any difference related to STC.
Propofol-Sufentanil Interaction Modeling.
In the range of concentrations studied, no statistically significant improvement was observed when using the interaction model versus the simple additive model, either for the three intraoperative end-points, or during recovery. The fitted curves for both models are shown for one end-point in Figure 1 (bottom) as an example.
The correlation between STC and propofol predicted concentration was not statistically significant among groups before intubation (r = 0.16 with the additive model versus 0.18 with the interaction model) but was statistically significant before incision (r = 0.46 whatever the model, P < 0.01) and at thyroid removal (r = 0.63 with the additive model and 0.64 with interaction model, P < 0.01).
In other words, in the range of concentrations studied, increasing the STC linearly decreased propofol requirement, without a demonstrated synergistic interaction.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
LOC was obtained with a target propofol concentration of 4 µg/mL in 30 of 45 patients (67%). This result is consistent with previous studies (13) and with the logistic model published by Schraag et al. (14). According to this model, the theoretical probability for LOC with propofol 3.5 µg/mL ranged between 44% (combined with 0.1 ng/mL sufentanil) and 76% (with 0.3 ng/mL). Achieving 95% probability of LOC would theoretically require an initial propofol target concentration >5 µg/mL (14) or 6 µg/mL (2). Such targets may be chosen for patients requiring a rapid induction, and in whom overdosage does not introduce a major cardiovascular risk. In other situations, an initial target of 4 µg/mL, followed by up-titration until LOC, can ensure 100% success without hemodynamic side effects, as shown in our study.
The lack of a statistically significant influence of STC on the propofol effect-site concentration required for LOC may be attributable to an insufficient number of patients relative to the variability, but may support the fact that the influence of opioids on sedation is minor (15). This weak influence of opioids on sedation in the absence of noxious stimuli has also been reported when using BISTM monitoring to assess the level of sedation or anesthesia (16–18).
Response to Intubation
Orotracheal intubation was performed in all patients with a stable sufentanil effect-site concentration identical to the target concentration because the equilibration time between plasma and effect-site concentrations are minimized when using TCI (19). Sufentanil concentrations at intubation were small in our study compared with those recommended by Vuyk et al. (5) based on alfentanil data. According to these authors, propofol 3.5 µg/mL combined with sufentanil 0.3 ng/mL should block the response to intubation in only 50% of the patients. This discrepancy may be explained by the demanding criteria chosen by them as a blocked response: all responses were considered (motor, hemodynamic, and sweating) with a narrow tolerance to changes (BP increase <15 mm Hg, HR <90 bpm). In our study, neuromuscular blockade both facilitated intubation and inhibited motor response, and the MBP and HR changes were considered without any threshold. Despite the lack of a statistically significant difference among groups, some patients in the smallest STC groups had clinically unacceptable hemodynamic responses, whereas no such responses were observed with an STC of 0.3 ng/mL. Our results suggest that a sufentanil concentration of 0.3 ng/mL (corresponding to an initial bolus dose of 12 µg) could be recommended for intubation in paralyzed patients, but may be insufficient when not using a muscle relaxant.
Optimal Combination for Thyroid Surgery
Thyroidectomy is a peripheral surgery but is located on very sensitive areas, neck and trachea. There are no data in the literature quantifying the level of stimulation caused by thyroid surgery compared with other "minor" (skin, breast, etc.) or "major" surgery (abdominal or thoracic).
Minor surgery could be performed with propofol concentrations >3 µg/mL combined with meperidine, whereas major surgery required propofol target >4 µg/mL (20). When combined with sufentanil with an STC of 0.1 ng/mL, abdominal surgery required at least 5 µg/mL propofol, decreasing to 3 µg/mL propofol when combined with an STC of 0.3 ng/mL (5). These requirements are similar to our results, suggesting that thyroid surgery is closer to visceral than to peripheral surgery regarding drug requirements.
As expected, increasing STC significantly decreased propofol requirements, as well as the interpatient variability in the propofol requirements (Fig. 1, bottom). Within the range of sufentanil concentrations studied, synergism could not be demonstrated, whereas most of the published studies reported a ceiling effect on propofol-opioids interactions (4,6,7,21). This difference may be attributable to a too large or too narrow range of opioid concentrations studied. It may also be more marked when considering responses to very painful stimuli such as during abdominal (4) or orthopedic surgery (21).
Although adequate anesthesia could be obtained with all STC combined with propofol, the incidence of "inadequate anesthesia," as defined by Glen (22) differed markedly with STC.
Whereas BISTM stability was similar for all STC, hemodynamic stability from incision until thyroid removal showed a clear benefit of 0.2 ng/mL STC compared with both other concentrations studied (Table 3).
Frequent motor responses to surgical stimulation in patients receiving sufentanil 0.1 ng/mL (4/15 patients) suggests that this concentration cannot be recommend. None of the patients who moved had any recall of intraoperative events, which was consistent with adequate BISTM values (23) suggesting that movement may have other causes than awareness, involving subcortical or medullar structures. Motor response during surgery may have been underestimated because of a partial neuromuscular blockade. Unfortunately, neuromuscular blockade was not monitored in our study. However, surgery started 16 to 48 minutes after the induction of anesthesia and lasted 54–332 minutes, and was similar among STC groups, and all movements were observed during final neck dissection. However, a partial neuromuscular recovery (<50% block) could be expected from 45 minutes after a 0.5 mg/kg atracurium bolus (simulation performed retrospectively with Stanpump software). In other words, all patients were expected to be able to move during the surgery period, and similarly among groups.
Recovery
In our study design, extubation was performed only when patients were breathing and responding to commands, whereas in clinical practice spontaneous breathing may be considered sufficient. A time to spontaneous ventilation of about 15 minutes was clinically acceptable in our patients, because infusions were stopped at the end of skin closure and wound dressing took approximately 10 to 15 minutes. Recovery within acceptable times was achieved in all patients in the two smallest STC groups (Table 4). Conversely both spontaneous ventilation and response to commands were significantly delayed in the STC 0.3 ng/mL group, and one patient had unacceptable recovery times (71 minutes from the end of surgery to extubation, after a 2-hour procedure!) suggesting that maintaining this sufentanil target is not suitable when rapid recovery is desirable.
Rapid recovery is usually desirable after short surgical procedures. Sufentanil has a shorter context sensitive half-time than fentanyl or alfentanil for infusions lasting <6–8 hours (24). Our results were consistent with this feature but emphasize that this "shortest" recovery time may exceed half an hour. Remifentanil, because of its ultra-short context sensitive half-time, which is virtually independent of the duration of infusion, might be a better choice of opioid when a rapid recovery is indicated (4,25). However, sufentanil is an acceptable option as long as STC remains 
0.2 ng/mL.
We conclude that TIVA combining propofol and sufentanil in a TCI mode is suitable for thyroid surgery. An STC of at least 0.3 ng/mL should be chosen for intubation combined with a target propofol concentration of 4 µg/mL or larger, if necessary, to achieve LOC. During surgery, an STC of 0.2 ng/mL seemed optimal to ensure hemodynamic stability, lack of movement, and fast recovery, combined with a propofol target concentration of 4 µg/mL.
|
Acknowledgments |
|---|
We acknowledge Mrs. Joret (Janssen-Cilag), Dr. Laplanche, Mrs. Mercier, Mrs. Delicourt, and Dr. Treich (Institut Gustave Roussy) for their help on data management.
|
Footnotes |
|---|
1 Software graciously available at http://anesthesia.stanford. edu/pkpd. 
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Godet, M. Reina, M. Raux, J. Amour, V. De Castro, and P. Coriat Anaesthesia for carotid endarterectomy: comparison of hypnotic- and opioid-based techniques{dagger} Br. J. Anaesth., March 1, 2004; 92(3): 329 - 334. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
