Hemodynamics and tissue oxygenation during balanced anesthesia with a high antinociceptive contribution: an observational study

One gram of Paracetamol was administered orally prior to admission to the OR. Intravenous access was secured using an 18G IV catheter in a left large forearm vein, and a continuous background infusion of crystalloids was started at 90 ml hr^-1. Anesthesia was induced with remifentanil (1 μg kg^-1), propofol (1–3 mg kg^-1), and cisatracurium (0.1 mg kg^-1).

In patients with preexisting cardiovascular comorbidities such as hypertension, reported peripheral or coronary artery disease in whom post-induction hypotension—defined as a 20% decrease [15, 16] in baseline mean arterial pressure (MAP)—can be expected, a norepinephrine (10 μg) single bolus was co-administered prophylactically.

After tracheal intubation, the lungs were ventilated in a volume controlled mode with a tidal volume of 8 ml kg^-1 lean body weight with a mixture of O₂/air (F_iO₂ 0.4) and 5 cm H₂O PEEP. The respiratory rate was adjusted to keep an end-tidal CO₂ pressure between 4.5 and 5.5 kPa.

Anesthesia was maintained with propofol (4 mg kg^-1 hr^-1) and remifentanil (0.25 μg kg^-1 min^-1) and adequate doses of hypnotics and analgesics to tolerate laryngoscopy [2, 17] and to keep BIS between 40 and 60. If necessary, these doses could be increased in 30% steps. In addition, norepinephrine was infused as required to keep MAP above 80% of individual baseline (preoperative ambulatory pressure) [15, 16].

The level of respiratory induced variation in the pulse oximetry plethysmographic waveform amplitude (∆POP) [18] was assessed 2 min after intubation; if it was found to be above 10%, hypovolemia was assumed and a fluid bolus of 500 ml colloid solution (Voluven®, Fresenius, Bad Homburg, Germany) was administered.

Administration of propofol, remifentanil, and norepinephrine was stopped simultaneously when the surgical procedure was finished. The trachea was extubated under continuous suction only after the patients were consciously responding and after they deeply inhaled upon verbal command. The extubation time, defined as the time between cessation of the syringe pumps and the moment of extubation, was recorded together with the incidence of any coughing reflex upon extubation. Any patient movement during the procedure was recorded. Two hours after recovery from anesthesia, patients were explicitly asked for any (adverse) experience in the peri-anesthetic period.

In addition to standard monitoring, a bispectral index (BIS) sensor (Aspect Medical Systems, Norwood, MA, USA) was placed on the forehead [19, 20]. The non-invasive Inspectra® device (Model 650, Hutchinson Technology, Hutchinson, KA, USA) measures the peripheral StO₂ based on near-infrared spectrometry [9]. This device emits near-infrared light (600–800 nm) to the tissue, which then undergoes changes in its spectrum, mostly depending on the degree of microvascular hemoglobin saturation. The probe can be placed on different anatomical locations; but it was placed on the left thenar eminence because of the limited fat-layer with little inter-individual variation [9]. The anesthetist was blinded for StO₂ readings. The Nexfin® monitor (model 1) is a non-invasive advanced hemodynamic monitoring device with a working principle that is based on the volume clamp method using a photoplethysmograph and an inflatable cuff placed around the intermediate phalanx [21]. This device provides continuous calculation of arterial blood pressure, systemic vascular resistance, stroke volume, cardiac output (CO), and heart rate in a beat-to-beat fashion [10‐14].

The Philips MP70 anesthesia monitor (Philips, Eindhoven, Netherlands) was connected to a PC, and all patient data were recorded at a 1-s interval using Rugloop II data-manager software (Demed, Temse, Belgium). The time and dose of all administered drugs were recorded in the Rugloop system, together with any noteworthy events or patient responses. The following variables were recorded: SpO₂, StO₂, heart rate (HR), MAP, CO, Cardiac Index (CI), and BIS. The electronic data were imported into Microsoft Excel 2010® (Microsoft, Redmond, WA, USA). After a graphical representation, a visual inspection of the data plots was performed to correct obvious atypical values caused by artifacts.

All statistics were performed using Microsoft Excel 2010® and PASW Statistics 18 (SPSS Inc, Chicago, IL, USA). Data were expressed as mean (SD), median (range), or number of patients (%). Normality of continuous variables was assessed using the Kolmogorov-Smirnov test. Continuous repetitive data before induction, and at 20 and 60 min thereafter were compared using the paired t test. Statistical significance level was set at p <0.05.

Figure 1A,B,C,D,E shows the time course of both the individual and mean values (in bold) of the investigated variables for the first 90 min. All data were synchronized at the start of induction of anesthesia. Figure 2 shows the mean values of all the investigated variables with a time interval of 1 min.

MAP decreased significantly from 109 (16) to 83 (14) mmHg (p <0.05) after induction and remained stable afterwards. No inadvertent severe hypertension following prophylactic norepinephrine administration was observed at induction (Figure 1A). Heart rate also decreased upon induction of anesthesia from 73 (12) to 54 (8) bpm (p <0.05) but remained stable afterwards (56 (9) bpm at 60 min; non-significant vs. 20 min). After an initial drop in cardiac index from 3.0 (0.7) L min^-1 m^-2 at induction to 2.1 (0.4) L min^-1 m^-2 at 20 min thereafter (p <0.05), it gradually increased again to 2.4 (0.4) L min m^-2 (p <0.05) at 60 min. StO₂ increased significantly within minutes after induction of anesthesia from 83 (6)% before induction to 86 (4)% at 20 min after induction of anesthesia (p <0.05) to remain stable thereafter (87 (4)% at 60 min after induction of anesthesia). The mean (SD) BIS value dropped from 95 (4) before induction to 44 (9) after 20 min (p <0.05) and remained stable afterwards (mean BIS 44 (10) at 60 min).

Figure 3 shows the changes in CI and StO₂ from pre-anesthesia level to 20 min after induction in a scatterplot. Accordingly, a majority of patients have an increased StO₂ despite decreased CI. There was no significant correlation between CI and StO₂ values.

Complete akinesia was realized in all patients in the period between induction of anesthesia and the end of surgery, while no additional relaxant was administered after intubation and 90% of the cases lasted longer than 45 min. No neuromuscular reversal agents were needed.

The median (range) extubation time was 311 (72–600) s with an interquartile range of 253–386 s. None of the patients exhibited a coughing reflex upon extubation and none of the patients reported awareness or unpleasant experiences in the peri-anesthetic period upon explicit questioning 2 h after extubation.

Firstly, the current study was neither designed as a randomized controlled trial nor has there been a control group. The study was nonetheless aimed to elucidate the effects of a balanced high-antinociceptive general anesthesia on tissue oxygen saturation in a clinical setting, for which an observational study would be the most appropriate.

Secondly, the current study was performed solely in patients undergoing ophthalmic surgery and therefore we cannot draw firm conclusions on the question whether this anesthetic regimen would be applicable during other types of surgery. Nevertheless, we speculate that in other surgical procedures associated with intermittent painful stimuli but requiring absolute akinesia (e.g., rigid laryngoscopy) and subsequently requiring rapid recovery with a fully responsive and cooperative patient, the investigated high-antinociceptive balanced anesthesia might be of particular use. Further studies should address this issue.

While invasive measurements are conventionally preferred to continuously assess hemodynamic variables, we used the non-invasive Nexfin® device, for which accuracy and precision of MAP measurements were recently shown, to meet the Association for the Advancement of Medical Instrumentation criteria in patients undergoing thoracic surgery [24] and also, were shown to be not inferior to noninvasive cuff manometric measurement of MAP in hemodynamically stable patients under general anesthesia [10]. This device also measures CO with a percentage error (PE) between 23% and 26% compared to thermodilution derived CO measurements as reported for patients who are awake before and after coronary artery bypass surgery [25]. Another study [14], investigating the influence of phenylephrine infusion on Nexfin® CO measurements, found a high concordance (94%) between phenylephrine-induced changes in CO measured by Nexfin® and by esophageal Doppler. The PE found in this study was however relatively high with a PE of 33% before and 42.5% after phenylephrine infusion, suggesting a reduced accuracy after vasopressor administration. Yet, Nexfin® CO accuracy still requires further investigation in patients under general anesthesia and in situations such as hemodynamic instability or during infusion of (other) vasoactive drugs, e.g., norepinephrine. Nevertheless, this device allows continuous and non-invasive monitoring of blood pressure and flow without adding additional risk to the patient. Additionally, recent technologic advances have also allowed the ability to calculate dynamic preload variables, such as pulse pressure and stroke volume variation, in an automated continuous noninvasive fashion [18], yet with a reported [26] higher accuracy in assessing fluid responsiveness than clinical ∆POP assessment, as was performed in the current study.

Finally, the site and technology to determine tissue oxygenation may have some limitations. In particular, myoglobin can have relatively high oxygen saturation even in case of tissue hypoperfusion and decreased oxygenation, and it could be a major contributor to the readings obtained. As a consequence, there is a debate on which site should be preferred to accurately measure StO₂[27]. However, the systematic increase in StO₂ relative to baseline we observed in the current study convincingly demonstrates a favorable mixed effect of the anesthesia on global StO₂.

In many surgical situations, a balanced high-antinociceptive general anesthesia is indicated, and a combination of propofol/remifentanil is appreciated. Based on the pharmacodynamic principle of equivalent anesthetic depth at different ratios of hypnotic/opiate, the chosen combination of propofol/remifentanil induced a sufficient level of anesthesia—reflected in normal BIS values between 40–60—although with a relatively higher contribution of remifentanil [2], resulting in a higher tolerance for sudden noxious stimuli. The favorable pharmacokinetic profile of this remifentanil/propofol combination resulted in a short median time to extubation—defined as the time interval between stopping the syringe pumps and the moment of extubation—of 311 s with low variability, together with optimal recurrence conditions: the fast elimination of propofol with still adequate analgesia at extubation results at the moment of endotracheal extubation in a complete suppression of potentially deleterious coughing reflexes and appropriate cooperation in all patients.

Anesthesia was induced with propofol, 1–3 mg kg^-1, and remifentanil, 1 μg kg^-1: a combination that can acutely decrease vascular tonus, heart rate, and cardiac output with resulting relative hypotension [28], which was also observed in the current study (Figure 2).

Firstly, in order to preserve cardiac preload, 13/40 patients were administered fluid as the ∆POP value was >10%, indicating fluid responsiveness, i.e., cardiac preload dependency.

Secondly, the decrease in CI following anesthesia induction is largely attributable to the (remifentanil-induced) decrease in heart rate (Figure 2), while stroke volume itself was relatively maintained probably because of both increased cardiac filling time and anesthesia-induced peripheral vasodilation and reduced cardiac afterload. The decrease in blood pressure after induction of anesthesia was anticipated with a single bolus of 10 μg norepinephrine at induction and a concomitantly started background infusion of norepinephrine. Most importantly, the administration of norepinephrine at the reported dose, in combination with goal-directed fluid optimization, was able to preserve a MAP at 80% of individual baseline values with a benign decrease in CO and without exerting adverse effects on StO₂.

A major concern of combining potent hypnotics/analgetics with a vasopressor is a negative end-effect on cardiac output and ultimately on tissue oxygenation. Therefore, a rational dosing is essential to obtain the desired intubation conditions and surgical stability, with optimal preservation of hemodynamic homeostasis. The stable BIS value within the recommended range reflects an adequate and reliable depth of hypnosis and demonstrates that a strong antinociceptive effect can be safely obtained if combined with an adequately adapted administration of propofol. Secondary, while the slight increase in tissue oxygenation is reassuring, a significant decrease in CI between awake and post-induction state demonstrates the potent hemodynamic effects of induction of even a balanced anesthesia. This decrease in CI is mainly based on a decrease in systemic resistance and heart rate and a preserved stroke volume. The resulting decrease in cardiac oxygen consumption, combined with a preserved peripheral tissue oxygenation, suggests that a CI decrease to this extent can be considered acceptable for induction of anesthesia. The subsequent increase in CI and stable MAP and StO₂ demonstrates that the moderate continuous doses of vasopressor are sufficient and appropriate to sustain hemodynamic stability during guaranteed optimal surgical conditions. However, the potent effects of the vasopressors must be reckoned with and must be considered in respect to other cardiovascular comorbidities. In addition, a more subtle administration of hypnotics/analgetics, such as based on TCI models, may be more appropriate in vulnerable patients.

While this desired MAP may appear relatively high for some clinicians, several reports [15, 16] acknowledge the utmost importance of a preserved MAP up this value to prevent adverse outcome. Importantly, care must be taken that potent vasopressors like norepinephrine are administered with respect to appropriate safety measures, such as the preferable use of a dedicated line and avoidance of inadvertent boluses.

For patients under general anesthesia, StO₂ values around 80% are considered within the normal range [29]. The increase in StO₂ to 86(4)% in our study was most probably caused by peripheral vasodilation due to the combination of propofol and remifentanil. This finding is in contrast to a previous study in which no differences in StO₂ were found before and after induction of anesthesia [29]. The use of sufentanil in this previous study may account for this difference. Nevertheless, StO₂ is believed to be a flow-dependent variable [9]. Furthermore, it has been shown to be independent of MAP and remained almost constant at different levels of MAP [30]. It has however not yet been investigated whether thenar StO₂ values also correlate with oxygen saturation of tissues of vital importance (e.g., heart, brain, kidney, liver). Subsequently, although the data show favorable values of thenar StO₂, indicating adequate peripheral tissue oxygenation (at least in the range of MAP and CI values observed), ‘macro’ hemodynamic values such as CI, heart rate, and MAP are of great importance for assuring patient safety.