Secondary Logo

Journal Logo

Featured Articles: Original Clinical Research Report

Guiding Opioid Administration by 3 Different Analgesia Nociception Monitoring Indices During General Anesthesia Alters Intraoperative Sufentanil Consumption and Stress Hormone Release: A Randomized Controlled Pilot Study

Funcke, Sandra MD*; Pinnschmidt, Hans O. PhD; Wesseler, Stefan*; Brinkmann, Charlotte*; Beyer, Burkhard MD; Jazbutyte, Virginija PhD§; Behem, Christoph R. MD*; Trepte, Constantin MD*; Nitzschke, Rainer MD*

Author Information
doi: 10.1213/ANE.0000000000004388


See Article, p 1261


  • Question: Does titration of an intraoperative opioid by the 3 different analgesia monitoring indices (Surgical Pleth Index [SPI], Pupillary Pain Index [PPI], and Nociception Level [NoL]) or by clinical signs in the control group lead to different amounts of sufentanil administered during radical retropubic prostatectomy?
  • Findings: The type of nociception monitoring affected sufentanil consumption, and lower levels of opioids were associated with higher levels of stress hormones.
  • Meaning: The monitor readings and target ranges did not seem to represent the same nociception–antinociception state, and lower doses of sufentanil were associated with more intraoperative stress as reflected by the elevated adrenocorticotropic hormone (ACTH) and cortisol concentrations.

Currently, optimal opioid titration during general anesthesia is still challenging, and the effects of opioid over- and underdosing are not well determined. In recent years, different monitoring devices estimating the effect of analgesia during unconsciousness have become available.1–3 Most devices generate an analgesic index from physiological variables determined by different techniques. Previous studies demonstrated that these monitoring devices were able to detect standardized painful stimuli and thereby display alterations in patients’ nociception–antinociception balance.4–9 Several studies have investigated the usability of such devices in the clinical setting,10–17 but no studies have systematically compared the effect of intraoperative analgesia guided by these devices simultaneously. In this pilot study, we primarily tested the hypothesis that titration of opioids using different analgesia monitoring devices or guidance by clinical signs leads to different intraoperative amounts of opioid consumption. The authors further aimed to investigate whether this opioid guidance correlated with an altered release of stress hormones or differences in the follow-up period. Our study should hereby contribute to the question of whether the monitors are already optimized for routine use during general anesthesia for surgery. In this trial, the authors compared opioid guidance by the Surgical Pleth Index (SPI) derived by photoplethysmography, the Pupillary Pain Index (PPI) determined by video pupillometry, the multiparameter index Nociception Level (NoL), and a control group in which only clinical signs of stress were considered.


This pilot study was a prospective interventional clinical trial that randomized patients to 3 different arms of analgesia monitoring or a control group during general anesthesia in the operating theater. The study was approved by the regional ethics review board of the Medical Council of Hamburg, Germany (Ref: PV5586 on July 25, 2017) and registered before patient enrollment at (Identifier “NCT03303651”; principal investigator: R.N.; date of registration: August 28, 2017). The signed written informed consent form addressing study participation, data collection, blood sample analyses, and 2 questionnaires for follow-up was obtained from all subjects ≥1 day before the surgery. The manuscript adheres to the applicable CONsolidated Standards Of Reporting Trials (CONSORT) guidelines.


The present authors included American Society of Anesthesiologists (ASA) class II to III patients scheduled for radical retropubic prostatectomy (RRP) in a high-volume center with >2500 RRPs annually. Thus, only male patients ≥18 years old participated. The exclusion criteria were chronic pain, β -blocker and digitalis therapy, eye disease, pacemaker therapy, high degrees of cardiac arrhythmias (eg, atrial fibrillation), and preoperative medication with steroids. The authors randomly assigned the study patients into 1 of the 4 treatment arms by a sealed envelope system. The randomization sequence was generated by a study nurse who was not involved in data collection before patient recruitment via a computer-generated list in blocks of 12 using the RAND function in Microsoft Excel (Microsoft Corporation, Redmond, WA). Allocation was concealed with sequentially numbered sealed opaque envelopes. The principal investigator evaluated eligibility, a physician of the study staff obtained informed consent, and the principal investigator enrolled the participants by opening the respective concealed envelope containing the patient allocation to the treatment arm on the day of surgery immediately before anesthesia induction.


All patients received preoperative care after local standards in all groups. Patients had been fasting for ≥6 hours except for a carbohydrate supplement with 150 mL clear apple juice. No premedication was given before surgery. On arrival at the operation theater, the study staff started convective air warming to minimize perioperative hypothermia and obtained the first blood samples when establishing an intravenous line. The routine monitoring included electrocardiography, noninvasive blood pressure measurement, pulse oximetry, central temperature measurement, and capnography. For the induction of general anesthesia, the patients received a standardized sufentanil bolus of 30 µg followed by a propofol bolus of 2 mg·kg−1 body weight and rocuronium 0.5 mg·kg−1 for tracheal intubation. Anesthesia was maintained with sevoflurane 1.7–2.0 vol% (minimal alveolar concentration [MAC], 1.0 MAC). Patients received a continuous low-dose norepinephrine administration compensating for the vasodilatation associated with general anesthesia to maintain the mean arterial pressure above 65 mm Hg as a clinical standard procedure. Additionally, repetition of muscle relaxation was conducted and guided strictly by acceleromyography (Infinity Trident NMT SmartPod; Draeger, Luebeck, Germany) to induce deep muscle relaxation throughout the surgical preparation. Further application of sufentanil was either guided by the respective analgesic-monitoring device as described in the study protocol or was administered at the discretion of the attending anesthesiologist using clinical signs only (control group). At the end of surgery, all patients received 1 g of metamizol and 150 µg of clonidine to prevent shivering.

Study Protocol

Patients were randomly assigned to 1 of the 4 treatment arms: the control group or to 1 of the 3 analgesia monitoring systems.

The first analgesia monitoring system was the CARESCAPE B650 patient monitor (General Electric Healthcare, Helsinki, Finland), which calculates the SPI from the normalized heart rate and pulse wave amplitude derived from finger plethysmography.1 This numerical index ranges between 0 (low sympathetic tone) and 100 (high sympathetic tone). Treatments that interfere with the SPI are those that influence the heart rate (eg, atropine, pacemaker) or the hemodynamic status (eg, changes in intravascular volume, posture changes). An SPI score of 50 has been validated as a threshold value, and a target range of 20–50 has been used in previous studies.8,10,11,18

The second monitoring system was the AlgiScan device (IDMed, Marseille, France), which reports the PPI and is derived from video pupillometry. It includes a camera measuring the extent of pupillary reflex dilation (PRD) after an electric nociceptive stimulation.6 The system automatically increases the intensity of the electric stimulation (10–60 mA) and displays the degree of PRD as an index. The numerical PPI ranges from 0 to 10; a low value indicates deep analgesia, and a high value indicates insufficient or light analgesia.6,19 To date, there have been no studies investigating a PPI threshold for nociception during surgery. Disturbance in measuring the PPI has to be expected in cases of eye diseases or limited access to the patient’s eye, and the measurement is only intermittent. A PPI score of 2 or 3 is supposed to represent an optimal level of analgesia during surgery according to the manufacturer.

The third analgesia monitoring device was the PMD200 instrument (Medasense Biometrics Ltd, Ramat-Gan, Israel), which provides the NoL derived from a finger probe. It continuously calculates the NoL with a multiparametric approach based on pulse rate, pulse rate variability, pulse wave amplitude, skin conductance level, skin conductance fluctuations, skin temperature, and finger motion.2 The composite algorithm of the device analyzes the data, and the numerical index NoL is presented on a scale from 0 (no pain) to 100 (extreme pain).7 Clinical conditions interfering with the correct acquisition of the plethysmography signal or skin conductance, severe arrhythmias (eg, atrial fibrillation), and rapid changes in intravascular volume or in posture influence the NoL value. A NoL score between 10 and 25 as the target range to guide analgesic administration has been recommended by the manufacturer and has been used in a previous published randomized controlled trial.17

These scores were reassessed in intervals of 5 minutes, and a bolus of 5 µg sufentanil was administered in intervals of 5 minutes if the scores were calculated above defined threshold values (SPI >50 and NoL >25 for >60 seconds, PPI >3 in intermittent measurements every 5 minutes) independently from heart rate and blood pressure. If the score persisted above the threshold, another 5 µg of sufentanil was given every 5 minutes without a defined maximum dose. In the control group, opioid administration was guided according to the standard clinical practice of the attending anesthesiologist considering physiological parameters (eg, heart rate, blood pressure, lacrimation, and sweating).


All data were obtained in a prospective manner, and patients were blinded to the group assignment. The staff could not be blinded during the intraoperative period. Patients were treated by anesthesiologists who were not part of the study team but were instructed to strictly follow the study protocol. During the postoperative period in the postanesthetic care unit (PACU), the staff was blinded to the group assignment, and patients received analgesics after a standardized protocol.

The primary outcome parameter was the intraoperative amount of sufentanil administered. Sufentanil administration was documented during surgery in intervals of 5 minutes by a study assistant who was not involved in the patient’s treatment. The secondary outcome variables were time from discontinuation of the anesthetic to extubation, duration of time in the PACU until fit-for-discharge, maximum numeric rating scale (NRS) score in the PACU, total amount of piritramide administered in the PACU, and the stress hormone serum levels measured before, during, and after surgery.

Blood samples were taken for analysis of adrenocorticotropic hormone (ACTH) and cortisol. In parallel, the study staff collected blood samples for total plasma protein to adapt plasma ACTH and serum cortisol measurements for blood loss and hemodilutional effects. Blood samples were taken on arrival of the patient in the operation theater (“Baseline”), at the end of surgery (“Skin closure”), at arrival at the PACU (“After extubation”), and before discharge to the ward (“End of PACU”).

For ACTH analyses, the ethylenediaminetetraacetic acid (EDTA) blood samples were immediately cooled in ice water. All blood samples were then transported to the central laboratory for ACTH (IMMULITE 2000XPi system; Siemens Healthineers Germany, Marburg, Germany) and serum cortisol (Cobas e 411 analyzer; Roche Germany, Mannheim, Germany) level measurements, both according to the manufacturers’ recommendations.

On arrival at the PACU, the patient’s level of analgesia was measured using the NRS scale (0–10). NRS ≤3 was considered to indicate no or mild pain, and NRS >3 was considered to indicate moderate-to-intense pain.20 The NRS score was reassessed every 15 minutes, and in the case of NRS >3, patients received a bolus of 3.75 mg piritramide. Patients reached fit-for-discharge criteria when they had no relevant pain (NRS <3) for >15 minutes, the patient was fully awake, vital signs were within normal ranges, and the cumulative urine output was >400 mL and clear. In addition to the data on postoperative pain and consumption of analgesics, the occurrence of postoperative nausea and vomiting (PONV), shivering, or other anesthesia-related events was documented.

On postoperative day 2, patients completed quality of recovery (QoR)21 and persistence of pain questionnaires. Furthermore, patients received an envelope with a second short questionnaire to assess prolonged persistence of pain and overall satisfaction with anesthetic care. The study staff asked the patients to voluntarily return the questionnaire by mail on postoperative day 21.

Statistical Analysis

Statistical analyses were performed using SPSS 25.0 (IBM SPSS Statistics, Inc, Armonk, NY) and Prism 8 (GraphPad Software, La Jolla, CA). Continuous baseline variables are presented as the mean (standard deviation [SD]), and categorical variables are presented as category counts and percentages. Standardized differences were computed to indicate variation in baseline data among groups. Because most of the dependent variables showed skewed data distributions (assessed via histograms) or variance inhomogeneity among groups, nonparametric testing by Kruskal–Wallis tests was applied for global group comparisons of variables. If the Kruskal–Wallis test results were significant, pairwise group comparisons were performed by Mann–Whitney U tests, and the differences in medians were reported as Hodges–Lehmann estimators with 99% confidence intervals (CIs).

Time courses of hormone levels were analyzed via linear mixed-effect models (SPSS routine GENLINMIXED) with random intercepts for patients, assuming a variance components covariance structure. More information on repeated-measures designs can be found in a prior tutorial on these topics.22 The intervention group, time point (treated as a categorical variable), time slot (onset of surgery morning versus noon, to account for a circadian rhythm), and the interaction between the intervention group and time point were considered fixed effects. The authors adjusted for log2-transformed current total plasma protein concentrations at the various time points to account for eventual hormone dilution effects due to blood loss and crystalloid fluid therapy.

Areas under the curve (AUCs) were computed for hormone levels over time and analyzed by linear models using the adjusted (to body weight and duration of surgery) amount of sufentanil and treatment group as well as their interaction term as independent variables. The interaction term was not significant and was thus removed from the final model. ACTH data were natural logarithm (ln)-transformed before these analyses because they were right skewed (InACTH). Model computations were adjusted for plasma protein levels. Linear models were also used to explore associations between sufentanil and stress hormone levels at the end of surgery, using hormone levels as dependent variables and group and the adjusted amount of sufentanil as independent variables.

All tests were 2-tailed with α set to .05. The significance criterion for the primary end point was Bonferroni adjusted to account for 4 secondary end points. Hence, P values <.01 were considered significant. Data are presented with differences in means/medians, including the CI set to 99%, to account for 6 pairwise group comparisons.

Sample Size Justification

The sample size calculation of this pilot study was performed on the basis of pretest data due to a lack of adequate published data. The authors collected deidentified pretest data for sample size calculation systematically in 15 patients using the 3 monitoring devices according to the manufacturers’ recommendation and user manuals. All 3 monitoring devices are approved by the appropriate authorities for monitoring analgesia nociception balance during general anesthesia in Europe and are available for purchase. The devices have not yet been used in daily clinical routine throughout the institution but are used irregularly for specific procedures. Therefore, the authors collected pretest data for estimating sufentanil administration in the 3 intervention groups of the pilot study. The mean total sufentanil consumption in the intervention groups SPI, PPI, and NoL was 100, 36.9, and 48.3 µg, respectively, in these pretests, compared to 88.8 µg of the clinical standard practice in retrospective chart reviews. As calculated from these data by analysis of variance (ANOVA) using the software package PASS 2008 version 08.0.6 (NCSS LLC, Kaysville, UT), a total sample size of 44 patients (11/group) achieves 84% power to detect significant differences among the 4 treatment group means (100, 36.9, 48.3, and 88.8 µg) with a 1% significance level, assuming the SD within a group to be 40 µg (the highest within-group SD in the pretests). The 1% significance level Bonferroni correction adjusts for 4 secondary end points in addition to the primary end point. The study staff scheduled 48 patients (12/group) to account for a 10% drop-out rate.


The study staff screened 80 patients scheduled for RRP in the operation theater where the study was conducted between October 16 and December 5, 2017. A CONSORT flow diagram (Figure 1: CONSORT diagram) shows the details of assessment and exclusion. After obtaining written informed consent, 48 patients were randomized to the 4 groups (12 each). All of them were adult, male Europeans, and Table 1 gives an overview of the patient characteristics.

Table 1. - Baseline Characteristics
Ctrl. (n = 12) SPI (n = 12) PPI (n = 12) NoL (n = 12) Standardized Difference
Biometric data
 Age (y) 61 (6) 64 (7) 62 (8) 64 (5) −0.48 −0.16 −0.41 0.23 0.01 −0.20
 Height (cm) 180 (6) 177 (7) 182 (7) 179 (8) 0.19 −0.21 0.63 −0.34 0.34 0.74
 Weight (kg) 91 (14) 84 (10) 90 (9) 86 (14) 0.34 0.06 0.59 −0.35 0.21 0.67
 BMI (kg·cm−2) 27.8 (3.6) 26.8 (2.2) 27.4 (3.3) 26.8 (3.0) 0.33 0.14 0.36 −0.19 0.00 0.21
 ASA class
  II 10 (83.3%) 11 (92.7%) 9 (75.0%) 9 (75.0%)
  III 2 (16.7%) 1 (8.3%) 3 (25.0%) 3 (25.0%) −0.21 −0.21 0.25 0.00 0.46 0.46
  Metabolic 4 (33.3%) 2 (16.7%) 2 (16.7%) 2 (16.7%) 0.39 0.39 0.39 0.00 0.00 0.00
  Anticoagulants 1 (8.3%) 2 (16.7%) 1 (8.3%) 4 (33.3%) −0.65 0.00 −0.25 0.65 0.39 −0.25
  Cardiovasculara 2 (16.7%) 5 (41.7%) 5 (41.7%) 2 (16.7%) 0.00 −0.57 −0.57 −0.57 −0.57 0.00
  Others 2 (16.7%) 5 (41.7%) 3 (25.0%) 6 (50.0%) −0.76 −0.21 −0.57 0.53 0.17 −0.36
  None 6 (50.0%) 3 (25.0%) 5 (41.7%) 1 (8.3%) 1.03 0.17 0.53 −0.83 −0.46 0.36
NRS = 0 before surgery 10 (83.3%) 11 (91.7%) 12 (100.0%) 9 (90.0%)b −0.20 −0.63 −0.25 −0.47 −0.06 0.43
Baseline characteristics of the 4 treatment groups: SPI, PPI, NoL, and Ctrl., displayed as means (SD) respectively counts (%). Standardized difference = difference in means or proportions divided by SD; imbalance defined as absolute value >0.20 or < −0.2, respectively (small effect size).
Abbreviations: ACE, angiotensin-converting-enzyme; ASA, American Society of Anesthesiologists; AT1, angiotensin II type 1; BMI, body mass index; Ctrl., control; NoL, Nociception Level; NRS, numeric rating scale; PPI, Pupillary Pain Index; SD, standard deviation; SPI, Surgical Pleth Index.
aCardiovascular: antihypertensive (ACE inhibitors, calcium channel blockers, diuretics, AT1 receptor antagonists) and antiarrhythmic medication, except β-blocker therapy (exclusion criterion).
bTwo missing values.

Figure 1.:
Study flow chart (CONSORT diagram). CONSORT indicates CONsolidated Standards Of Reporting Trials; PPI, Pupillary Pain Index; SPI, Surgical Pleth Index.

The type of analgesia nociception monitoring affected the total amount of sufentanil administered (Figure 2: sufentanil consumption). For further statistical analysis, the authors adjusted the sufentanil consumption to the body weight and duration of surgery (Table 2) with the following medians (quartiles) (μg·kg−1·minute−1·10–3): control = 5.6 (4.4–6.4), SPI = 7.2 (4.8–8.4), PPI = 2.0 (1.8–2.9), NoL = 3.8 (3.3–5.1); Pgroup < .001. The results of the pairwise comparison analyzed by the Mann–Whitney U test revealed group differences (reported as the Hodges–Lehmann estimator [99% CI], Pvalues) of PPI versus SPI (−5.1 [−6.6 to −1.3], P< .001); NoL versus PPI (1.7 [0.6–3.4], P< .001), and control versus PPI (3.4 [2.0–4.6], P< .001), but not for NoL versus SPI (−3.0 [−5.2 to 0.2], P= .024), control versus SPI (−1.6 [−3.7 to 1.7], P= .128), and control versus NoL (1.6 [−0.2 to 3.3], P= .017).

Table 2. - Outcome Parameters
Ctrl. (n = 12) SPI (n = 12) PPI (n = 12) NoL (n = 12) P
Primary end point
 Sufentanil consumption (μg·kg−1·minute−1·10−3) 5.6 (4.4–6.4) 7.2 (4.8–8.4) 2.0 (1.8–2.9) 3.8 (3.3–5.1) <.001a
Secondary end points
 Time from end of narcotics to extubation (min) 13 (12–18) 16 (12–18) 12 (10–18) 12 (9–14) .289
 Duration of time in PACU until fit-for-discharge (min) 124 (96–148) 122 (89–143) 138 (127–163) 133 (122–146) .147
 Maximum NRS in PACU 4.5 (4–5) 5 (4–5.5) 5 (4.5–6) 4.5 (3.5–6) .497
 Total amount of piritramid in PACU (mg) 7.5 (7.5–11.3) 7.5 (3.8–13.1) 11.3 (5.6–15) 7.5 (1.9–11.3) .367
Outcome parameters of the 4 study groups. Values are displayed as the median (quartiles) with P values (group comparison by Kruskal–Wallis test). NRS from 0 = no pain to 10 = worst pain.
Abbreviations: Ctrl., control; NoL, Nociception Level; NRS, Numeric Rating Scale; PACU, postanesthetic care unit; PPI, Pupillary Pain Index; SPI, Surgical Pleth Index.
aPairwise group comparison of sufentanil consumption adjusted to body weight and duration of surgery analyzed by Mann–Whitney U test: Ctrl. versus NoL P = .017, Ctrl. versus PPI P < .001, Ctrl. versus SPI P = .128, NoL versus PPI P < .001, NoL versus SPI P = .024, and PPI versus SPI P< .001.

Figure 2.:
Sufentanil consumption. Total sufentanil dose for the introduction and maintenance of anesthesia, applied to each patient, as the primary outcome variable in the study groups. The data are presented as medians, ranges, and interquartile ranges (25th–75th percentile) without adjustment for the patient’s body weight and duration of anesthesia. The 4 groups, namely, Ctrl. (guided by clinical signs), SPI, PPI, and NoL, were compared by Kruskal–Wallis (P group < .001) and Mann–Whitney U tests (PPI versus Ctrl.: P < .001, NoL versus Ctrl.: P = .001, SPI versus Ctrl.: P = .977, SPI versus NoL: P = .006, PPI versus NoL: P < .001, SPI versus PPI: P < .001). Ctrl. indicates control; NoL, Nociception Level; PPI, Pupillary Pain Index; SPI, Surgical Pleth Index.

The time course of ACTH and cortisol in the 4 groups was different (Figure 3: release of stress hormones; Supplemental Digital Content 1, Table 1, Mixed model analysis indicated the main effects of treatment group, time and an interaction effect of group and time (all P< .001; Supplemental Digital Content 2, Table 2, ACTH and cortisol concentrations were lowest in the SPI patients and highest in PPI patients at the end of surgery. The AUC analysis (Figure 3) indicated differences between cumulative lnACTH levels (ng·liter−1·minute) of NoL versus PPI (−1.079 [−1.95 to −0.208], P =.001) and PPI versus SPI (1.192 [0.317–2.068], P < .001), but no difference for control versus NoL (0.605 [−0.233 to 1.443], P = .052), control versus PPI (−0.474 [−1.339 to 0.391], P =.136), control versus SPI (0.718 [−0.133 to 1.569], P = .024), and NoL versus SPI (0.113 [−0.725 to 0.952], P= .709) (differences in means [99% CI], P values). AUC analysis of cortisol levels (µg·liter−1·minute) showed differences between PPI versus SPI (46,710 [21,145–72,274], P< .001), NoL versus SPI (27,645 [3163–52,162], P= .003), and control versus SPI (31,824 [6974–56,675], P< .001), but no difference for NoL versus PPI (−19,065 [−44,502 to 6373], P =.044), control versus NoL (4179 [−20,283; 28,642], P= .638), and control versus PPI (−14,885 [−40,143 to 10,373], P =.109) (differences in means [99% CI], P values).

Figure 3.:
Release of stress hormones. Plasma levels of (A) ACTH and (B) serum cortisol at 4 different time points during the day of surgery (“Baseline” before induction of general anesthesia, “Skin closure” at the end of surgery, “After extubation” on arrival at the PACU, and “End of PACU” before discharge to the ward). The Ctrl. group (guided by clinical signs) is compared to titration by the SPI, PPI, or NoL. Data are presented as mixed-model–estimated marginal means and 95% CI adjusted for plasma protein = 64 g/L. ACTH was log-transformed before mixed-model analysis; the data shown are based on back transformation. Post hoc pairwise comparison of ACTH levels by contrasts: *Ctrl. versus SPI P < .001 and Ctrl. versus PPI P < .001 at “Skin closure.” Post hoc pairwise comparisons of cortisol levels: †Ctrl. versus SPI P < 0.001 at skin closure. §Ctrl. versus SPI P < 0.001 at after extubation. AUC analyses (AUC estimated marginal mean [95% CI]) revealed cumulative blood levels of ACTH (ng·liter−1·minute) for SPI 16,499 (10,759–25,301), PPI 54,360 (35,067–84,269), NoL 18,477 (12,245–27,880), and Ctrl. 33,842 (22,154–51,697), and cumulative blood levels of cortisol (µg·liter−1·minute) for SPI 51,751 (39,267–64,235), PPI 98,461(85,660–111,261), NoL 79,396 (67,384–91,408), and Ctrl. 83,575 (71,204–95,947). ACTH indicates adrenocorticotropic hormone; AUC, area under the curve; CI, confidence interval; Ctrl., control; NoL, Nociception Level; PACU, postanesthesia care unit; PPI, Pupillary Pain Index; SPI, Surgical Pleth Index.

Figure 4 shows a scatterplot of the total sufentanil consumption versus the plasma levels of ACTH (A) and serum cortisol (B) at the end of surgery. The groups with more sufentanil administration were associated with lower levels of stress hormones. Levels of ACTH and cortisol at the end of surgery differed among treatment groups (P= .002), while in the presence of the treatment group in the model, the administered dose of sufentanil did not show an association (P= .232) with hormone levels.

Figure 4.:
Scatterplots of total sufentanil consumption versus stress hormone levels ([A] ACTH and (B) serum cortisol) at the end of surgery analyzed by monitoring systems. ACTH indicates adrenocorticotropic hormone; Ctrl., control; NoL, Nociception Level; PPI, Pupillary Pain Index; SPI, Surgical Pleth Index.

The results from the nonparametric testing of the postoperative secondary end points are shown in Table 2. Kruskal–Wallis tests revealed no differences between the study groups in this period. Further secondary analyses on intra- and postoperative parameters are shown in Supplemental Digital Content 3, Table 3,, but no evidence for a difference between the study groups was observed. None of the patients reported recall of intraoperative events or awareness.


This study showed that the choice of analgesia monitoring systems led to different intraoperative amounts of opioid administered in male patients undergoing RRP. Patients received less sufentanil in the PPI group than in the control, SPI, and NoL groups. The lower levels of opioids in the PPI group were associated with higher intraoperative levels of cortisol and ACTH. Although titration by SPI did not reduce intraoperative opioids when compared to the NoL and the control groups, it was associated with lower levels of cortisol. During the follow-up period, no evidence for a difference was observed in postoperative pain, analgesic requirements, or recovery times (time to extubation and time in PACU).

Previous studies have investigated the effect of analgesia management by measuring the nociception–antinociception balance compared to a control group using clinical signs and have yielded different results with regard to opioid consumption, intraoperative adverse events, and outcome parameters. Several studies showed that remifentanil administration guided by SPI reduced the total amount infused, decreased the rate of hemodynamic alterations, and resulted in fewer incidents of intraoperative movement.10,11,15 Nevertheless, recovery time, postoperative analgesia, and satisfaction during the PACU stay and the day after surgery were not influenced. In contrast, other studies noted no change in the amount of opioid administered under SPI guidance,13,18 similar to our findings. One might contend that administering the proper amount of opioid at the appropriate time is more beneficial compared to phases of over- and underdosing in patients in whom opioids are administered based on clinical signs.

Nociception from surgical stress leads to the release of stress hormones during general anesthesia.23 ACTH and cortisol have been proven to correlate well with the severity of surgical stress and to vary with different analgesic levels,24–27 which are in line with our findings. In addition, the perioperative levels of ACTH and cortisol found in our study are comparable to those noted in previous studies investigating stress hormone release during open abdominal surgery.23,24 Interestingly, the same total amount of opioid when guided by SPI was associated with lower ACTH and cortisol levels at the end of surgery compared to those of the control. Hence, it comes into question whether titrating analgesia to the individual need of each patient at different time points by the SPI may have improved the intraoperative maintenance of the nociception–antinociception balance. At this time, there are no studies proving that the extent of the perioperative increase in stress hormone levels is associated with adverse events or a deterioration of the patient outcome.

The effect of PPI-guided opioid dosing has been investigated with remifentanil, but not with sufentanil. In this study, the use of PPI led to a reduction in sufentanil consumption but was associated with higher increases in ACTH and cortisol at the end of surgery. Previous studies have also found a reduction in remifentanil when guided by PPI, compared to the control.12,14 Interestingly, intraoperative PPI guidance of remifentanil administration was associated with a reduction in postoperative pain levels and analgesic requirements in the first 12 hours after surgery without influencing recovery times. In our study, there was no evidence that intraoperative guidance of sufentanil administration using PPI affected patient recovery.

The NoL is a newer multiparametric index that includes different physiological parameters.2 The NoL has been validated as an instrument for detecting nociceptive stimulation.5,7,28 In our study, the NoL did not reduce opioid consumption or alter the release of stress hormones compared to that in the control group. This finding is in contrast to a prior study using the NoL to titrate analgesic medication that revealed a reduction in opioid consumption and fewer unwanted events compared to the control group.17

There are several limitations to our study. First, only male patients were included by choosing to investigate patients undergoing prostatectomy. Prostatectomy is a highly standardized procedure, and the surgical trauma induced can be expected to be quite comparable between groups in a high-volume center. Another important point is that opioid administration in the control group was subjective and influenced by the local guidelines and standards of the respective institution. Furthermore, there is some uncertainty about the validity of the threshold values chosen, especially for the PPI. In addition, the PPI score is only available with intermittent measurements, and the extinction of the pupillary pain reflex already begins at moderate dosages of opioids that are commonly used for general anesthesia.8 Therefore, nociception could possibly be missed by the intermittent PPI measurements. In contrast, SPI and NoL are displayed continuously, and both measures calculate the nociception–antinociception balance from hemodynamically influenced parameters. Finally, this investigation is a pilot study that was designed to reveal differences in opioid administration because there have been no previous studies comparing the clinical application of different analgesia monitoring systems. Thus, the sample size calculation was based on pretest data on intraoperative sufentanil administration. Hence, the study may be underpowered for detecting differences in the follow-up period, and further studies with more patients will be necessary.

In conclusion, our pilot study directly compared opioid titration by different techniques of analgesic monitoring and traditional guidance using clinical signs. We showed that the use of different nociception monitors leads to the intraoperative administration of different amounts of sufentanil. Nociception–antinociception balance is differently reflected by the analgesia monitoring systems, and the monitored readings and the target ranges do not represent the same nociception–antinociception state. This finding is in contrast with the results from the well-validated monitoring systems on the depth of hypnosis because the assessment of the depth of hypnosis does not greatly differ between the various electroencephalographic monitoring systems.29 In the present study, the type of analgesia monitoring was associated with alterations in the neuroendocrine stress response. Lower intraoperative doses of opioid in the PPI group were associated with an increase in the release of stress hormones. SPI-guided analgesia was associated with lower cortisol and ACTH, a desired situation, despite the absence of reduction in opioid consumption. To date, there is no gold standard for intraoperative monitoring of nociception. The current study suggests that without further studies investigating the clinical applicability, including the definition of threshold values and ranges, these monitors are not yet optimized for use in the daily clinical routine.


The authors thank Kai Bremer, CRNA, for technical assistance and all colleagues from the Department of Anesthesiology and the Martini-Klinik staff in the operation theater for their organizational support to conduct this study during the daily clinical routine of the University Medical Center Hamburg-Eppendorf, Hamburg, Germany.


Name: Sandra Funcke, MD.

Contribution: This author helped design the protocol, prepare the submission for scientific and ethical approval, enroll the participants, collect and analyze the data, and write and edit the manuscript.

Name: Hans O. Pinnschmidt, PhD.

Contribution: This author helped design the study, analyze the statistical data, and write and edit the manuscript.

Name: Stefan Wesseler.

Contribution: This author helped enroll the participants, collect and analyze the data, and edit the manuscript.

Name: Charlotte Brinkmann.

Contribution: This author helped enroll the participants, collect the data, and edit the manuscript.

Name: Burkhard Beyer, MD.

Contribution: This author helped design the protocol, collect the data, and edit the manuscript.

Name: Virginija Jazbutyte, PhD.

Contribution: This author helped collect the data and edit the manuscript.

Name: Christoph R. Behem, MD.

Contribution: This author helped collect the data and edit the manuscript.

Name: Constantin Trepte, MD.

Contribution: This author helped collect the data and edit the manuscript.

Name: Rainer Nitzschke, MD.

Contribution: This author helped design the protocol, prepare the submission for scientific and ethical approval, enroll the participants, collect and analyze the data, and write and edit the manuscript.

This manuscript was handled by: Honorio T. Benzon, MD.



    1. Huiku M, Uutela K, van Gils M, et al. Assessment of surgical stress during general anaesthesia. Br J Anaesth. 2007;98:447–455.
    2. Ben-Israel N, Kliger M, Zuckerman G, Katz Y, Edry R. Monitoring the nociception level: a multi-parameter approach. J Clin Monit Comput. 2013;27:659–668.
    3. Barvais L, Engelman E, Eba JM, Coussaert E, Cantraine F, Kenny GN. Effect site concentrations of remifentanil and pupil response to noxious stimulation. Br J Anaesth. 2003;91:347–352.
    4. Gruenewald M, Ilies C. Monitoring the nociception-anti-nociception balance. Best Pract Res Clin Anaesthesiol. 2013;27:235–247.
    5. Martini CH, Boon M, Broens SJ, et al. Ability of the nociception level, a multiparameter composite of autonomic signals, to detect noxious stimuli during propofol-remifentanil anesthesia. Anesthesiology. 2015;123:524–534.
    6. Guglielminotti J, Grillot N, Paule M, et al. Prediction of movement to surgical stimulation by the pupillary dilatation reflex amplitude evoked by a standardized noxious test. Anesthesiology. 2015;122:985–993.
    7. Edry R, Recea V, Dikust Y, Sessler DI. Preliminary intraoperative validation of the nociception level index: a noninvasive nociception monitor. Anesthesiology. 2016;125:193–203.
    8. Funcke S, Sauerlaender S, Pinnschmidt HO, et al. Validation of innovative techniques for monitoring nociception during general anesthesia: a clinical study using tetanic and intracutaneous electrical stimulation. Anesthesiology. 2017;127:272–283.
    9. Wildemeersch D, Baeten M, Peeters N, Saldien V, Vercauteren M, Hans G. Pupillary dilation reflex and pupillary pain index evaluation during general anaesthesia: a pilot study. Rom J Anaesth Intensive Care. 2018;25:19–23.
    10. Chen X, Thee C, Gruenewald M, et al. Comparison of surgical stress index-guided analgesia with standard clinical practice during routine general anesthesia: a pilot study. Anesthesiology. 2010;112:1175–1183.
    11. Bergmann I, Göhner A, Crozier TA, et al. Surgical Pleth index-guided remifentanil administration reduces remifentanil and propofol consumption and shortens recovery times in outpatient anaesthesia. Br J Anaesth. 2013;110:622–628.
    12. Abad Torrent A, Rodríguez Bustamante V, Carrasco Fons N, Roca Tutusaus FJ, Blanco Vargas D, González García C. The use of pupillometry as monitoring of intraoperative analgesia in the consumption of analgesics during the first 12 hours after surgery. Rev Esp Anestesiol Reanim. 2016;63:253–260.
    13. Colombo R, Raimondi F, Rech R, et al. Surgical pleth index guided analgesia blunts the intraoperative sympathetic response to laparoscopic cholecystectomy. Minerva Anestesiol. 2015;81:837–845.
    14. Sabourdin N, Barrois J, Louvet N, et al. Pupillometry-guided intraoperative remifentanil administration versus standard practice influences opioid use: a randomized study. Anesthesiology. 2017;127:284–292.
    15. Rogobete AF, Sandesc D, Cradigati CA, et al. Implications of entropy and surgical pleth index-guided general anaesthesia on clinical outcomes in critically ill polytrauma patients. A prospective observational non-randomized single centre study. J Clin Monit Comput. 2018;32:771–778.
    16. Defresne A, Barvais L, Clement F, Bonhomme V. Standardised noxious stimulation-guided individual adjustment of remifentanil target-controlled infusion to prevent haemodynamic responses to laryngoscopy and surgical incision: a randomised controlled trial. Eur J Anaesthesiol. 2018;35:173–183.
    17. Meijer FS, Martini CH, Broens S, et al. Nociception-guided versus standard care during remifentanil-propofol anesthesia: a randomized controlled trial. Anesthesiology. 2019;130:745–755.
    18. Gruenewald M, Willms S, Broch O, Kott M, Steinfath M, Bein B. Sufentanil administration guided by surgical pleth index vs standard practice during sevoflurane anaesthesia: a randomized controlled pilot study. Br J Anaesth. 2014;112:898–905.
    19. Wildemeersch D, Peeters N, Saldien V, Vercauteren M, Hans G. Pain assessment by pupil dilation reflex in response to noxious stimulation in anaesthetized adults. Acta Anaesthesiol Scand. 2018 April 19 [Epub ahead of print].
    20. Hjermstad MJ, Fayers PM, Haugen DF, et al.; European Palliative Care Research Collaborative (EPCRC). Studies comparing numerical rating scales, verbal rating scales, and visual analogue scales for assessment of pain intensity in adults: a systematic literature review. J Pain Symptom Manage. 2011;41:1073–1093.
    21. Myles PS, Hunt JO, Nightingale CE, et al. Development and psychometric testing of a quality of recovery score after general anesthesia and surgery in adults. Anesth Analg. 1999;88:83–90.
    22. Schober P, Vetter TR. Repeated measures designs and analysis of longitudinal data: if at first you do not succeed-try, try again. Anesth Analg. 2018;127:569–575.
    23. Furuya K, Shimizu R, Hirabayashi Y, Ishii R, Fukuda H. Stress hormone responses to major intra-abdominal surgery during and immediately after sevoflurane-nitrous oxide anaesthesia in elderly patients. Can J Anaesth. 1993;40:435–439.
    24. Schricker T, Carli F, Schreiber M, et al. Propofol/sufentanil anesthesia suppresses the metabolic and endocrine response during, not after, lower abdominal surgery. Anesth Analg. 2000;90:450–455.
    25. Prete A, Yan Q, Al-Tarrah K, et al. The cortisol stress response induced by surgery: a systematic review and meta-analysis. Clin Endocrinol (Oxf). 2018;89:554–567.
    26. Ledowski T, Bein B, Hanss R, et al. Neuroendocrine stress response and heart rate variability: a comparison of total intravenous versus balanced anesthesia. Anesth Analg. 2005;101:1700–1705.
    27. Chen X, Thee C, Gruenewald M, et al. Correlation of surgical pleth index with stress hormones during propofol-remifentanil anaesthesia. Scientific World Journal. 2012;2012:879158.
    28. Stöckle PA, Julien M, Issa R, et al. Validation of the PMD100 and its NOL index to detect nociception at different infusion regimen of remifentanil in patients under general anesthesia. Minerva Anestesiol. 2018;84:1160–1168.
    29. Avidan MS, Mashour GA. Prevention of intraoperative awareness with explicit recall: making sense of the evidence. Anesthesiology. 2013;118:449–456.

    Supplemental Digital Content

    Copyright © 2019 International Anesthesia Research Society