Application of SPI-guided analgesia in laparoscopic gynecologic surgery: a randomized controlled trial evaluating the remifentanil-sparing effect and predictive value of time-weighted SPI

This study was designed as a composite study: the primary component was a prospective, randomized, controlled efficacy trial, and the secondary component was a validation analysis of a predefined exploratory index based on the data from the trial. The study protocol was approved by the Ethics Committee of the Affiliated People’s Hospital of Fujian University of Traditional Chinese Medicine (Approval No. 2024-037-02, Approval Date: July 10, 2024), and registration was completed in the China Clinical Trial Registry (Registration No. ChiCTR2400088444). All participants were enrolled in this study after signing a written informed consent, and the flowchart of the study protocol is shown in Fig.  1 . Fig. 1 The flow diagram of study protocol. NRS numeric rating scale, PACU post-anesthesia monitoring treatment room The flow diagram of study protocol. NRS numeric rating scale, PACU post-anesthesia monitoring treatment room Consecutive patients who undergo elective laparoscopic gynecological surgery (including hysterectomy, adnexal surgery, etc.) between October 2024 and February 2026 were included. Inclusion criteria: Aged 18 to 75 years, American Society of Anesthesiologists (ASA) classification I-II. EXCLUSION CRITERIA: Severe cardiovascular, hepatic, and renal insufficiency; long-term use of drugs affecting autonomic function, such as opioids, α-2 agonists, or β-blockers; history of chronic pain or long-term analgesic use; psychiatric or neurologic disorders; allergy to the study medications; and pregnant or lactating women. SAMPLE SIZE CALCULATION: This study was a randomized controlled trial with a superiority design. The primary outcome of this study was the total intraoperative remifentanil dose, normalized to body weight and infusion duration, expressed as µg/kg/h. The sample size calculation was based on the subject group’s pre-trial and published data from similar studies [ 9 , 10 ]. The mean and standard deviation of intraoperative remifentanil dosage in the conventional analgesia group were expected to be 420 ± 90 µg/h. Assuming that the SPI-directed group reduces remifentanil dosage by approximately 20% (i.e., approximately 84 µg/h), a clinical superiority cutoff value of δ = 50 µg/h was set (equivalent to approximately 12% of the mean value in the control group). With a test level of α = 0.05 (bilateral) and a test efficacy of 1-β = 0.80, the required sample size for each group was estimated according to the following formula:. It was calculated that a minimum of 27 patients were required in each group. Considering a case dropout or incomplete data rate of approximately 15%, the final plan was to include a total sample size of 66 cases, which were randomly assigned in a 1:1 ratio to the SPI-directed group versus the conventional analgesia group (33 cases in each group). Sample size estimation was done using PASS 15.0 software. A computer-generated randomization sequence was used for group assignment in this study. The random allocation sequences were generated and kept sealed by an independent researcher and blinded before induction of anesthesia. Given that the SPI monitoring interface was visible intraoperatively, the anesthesia administrator could not be blinded; however, the postoperative pain assessor, data recorder, and statistical analyst were blinded to the grouping information. A uniform general intravenous anesthesia protocol was used for all patients. Anesthesia induction: target-controlled infusion of propofol (target concentration 4 µg/ml), sufentanil 0.4 µg/kg, and rocuronium bromide 0.6 mg/kg was used. Anesthesia maintenance: target-controlled infusion of propofol (1–4 µg/ml) and remifentanil (1–8 ng/ml) was used, and target concentrations were adjusted according to the Bispectral Index (BIS) monitoring values (target range 40–60) to adjust the target concentration.In the event of BIS values  60 persisting beyond 60 s, the propofol target was increased by 0.5 µg/mL. No other analgesic drugs were used intraoperatively except as specified in the study protocol. Flurbiprofen ester 50 mg was administered intravenously 30 min before the end of surgery to prevent postoperative pain [ 7 ]. To avoid interference of vasoactive drugs on autonomic function and SPI values, patients with intraoperative use of vasoactive drugs were excluded from this study.Surgical procedures were performed using a German Karl Storz TC200 laparoscopic system (Manufacturer: Karl Storz SE & Co. KG; Tuttlingen, Germany). The system was equipped with a continuous-flow insufflation system (Karl Storz); a standard electronic insufflator was not used. Pneumoperitoneum pressure was standardized at 12–14 mmHg. SPI-oriented group: SPI was continuously monitored and displayed using a GE Datex-Ohmeda S/5™ anesthesia workstation, with a set intervention threshold of SPI > 50. When the SPI value was consistently higher than this threshold for more than 15 s, remifentanil 0.5 µg/kg was given intravenously (2–5 min to complete the injection). Conventional analgesia group: The SPI monitoring function on the anesthesia machine was turned on, but the screen display was masked. Adjustment of analgesic medication was based on conventional vital signs: remifentanil 0.5 µg/kg was administered intravenously when heart rate or mean arterial pressure increased by more than 20% from the baseline value prior to induction of anesthesia (2–5 min to completion). The probe was placed on the index finger of the non-blood-pressure-measuring side. SPI values were continuously monitored and recorded using the GE Healthcare S/5™ Anesthesia Workstation (GE Healthcare, Helsinki, Finland). The probe was placed on the finger of the non-blood-pressure-monitoring side, and the signal was acquired by a finger-cuff pulse oximetry probe. The SPI value was calculated based on the normalized photoplethysmographic volumetric pulse wave amplitude (PPGA) versus heartbeat intervals (HBI). The formula for calculating the SPI is as follows [8]: SPI = 100 - (0.7 × PPGA norm + 0.3 × HBI norm ). Where PPGA norm is the normalized photoplethysmographic volumetric pulse wave amplitude, and HBI norm is the normalized heartbeat interval. SPI takes values in the range of 0-100. In this study, SPI data were recorded every 5 min due to the default data export settings of the anesthesia information system; this low sampling frequency is a major limitation of the study, as discussed in the Limitations section. A critical limitation of the study design is that SPI data were recorded at 5-minute intervals due to the default export settings of the anesthesia information system. SPI is a highly dynamic signal that responds within seconds to surgical stimuli; therefore, this low sampling frequency likely underestimates the true cumulative nociceptive burden and attenuates the apparent predictive performance of Tb-TW-SPI. The primary outcome of this study was the total intraoperative remifentanil dose (µg/h), automatically recorded by the anesthesia information system. Secondary outcomes included: (1) postoperative pain scores in PACU; (2) perioperative adverse events; and (3) an exploratory analysis evaluating the predictive performance of the novel Tb-TW-SPI metric for moderate-to-severe postoperative pain. To comprehensively evaluate noxious stimulus load, this study introduced and calculated the Threshold-based Time-Weighted SPI (Tb-TW-SPI). This metric aims to quantify the cumulative stimulus load exceeding a preset SPI threshold during surgery, comprehensively reflecting the total exposure to high-intensity noxious stimuli. The threshold was set at SPI > 43, based on our prior meta-analysis findings, representing a commonly used clinical cutoff value distinguishing adequate analgesia from noxious stimulus responses [ 1 ]. Tb-TW-SPI Calculation Method: Tb-TW-SPI quantifies cumulative load by calculating the area under the curve where SPI values exceed the threshold (43). Specifically, it first extracts regularly spaced SPI sequences from anesthesia records to generate a threshold-exceeding sequence: SPI_excess(t i ) = max{SPI(t i )−43,0}. Subsequently, the SPI_excess sequence is integrated using the trapezoidal rule. The final unit of this metric is “SPI units × minutes.” Detailed calculation steps and examples are provided in the Supplementary Materials. We acknowledge the conceptual inconsistency between the therapeutic intervention threshold (SPI > 50) and the analytical threshold (SPI > 43). To assess the robustness of our findings, we performed a sensitivity analysis using the intervention-consistent threshold (SPI > 50) to re-calculate Tb-TW-SPI. The results are presented in Supplementary Materials 3. Postoperative Pain Outcome Variables: Postoperative pain outcomes were used to evaluate the predictive efficacy of Tb-TW-SPI, defined as ① Primary pain variable (continuous): The highest NRS score (0–10) self-reported by the patient during the PACU period. ② Secondary pain variable (binary): Based on the primary variable, “moderate-to-severe postoperative pain” was defined as an NRS score ≥ 4. Intention-to-treat analysis was used. Continuous variables: those conforming to normal distribution were expressed as mean ± standard deviation, and the independent samples t-test was used for comparison between groups; those not normally distributed were expressed as median (interquartile range), and the Mann-Whitney U test was used for comparison between groups. Categorical variables: Expressed as frequencies (percentages), comparisons between groups were made using the chi-square test or Fisher’s exact probability method. (All statistical analyses were two-sided tests, and differences were considered statistically significant at P  < 0.05.) The aim of this exploratory analysis was to assess the predictive value of time-weighted SPI for postoperative pain. Rationale for sample selection: Only data from patients in the conventional analgesia group ( n  = 30) were included in the analysis. The reason: intraoperative analgesia in the SPI-oriented group was based on real-time SPI values for active interventions, and their SPI sequences had become an integral part of the treatment and did not reflect the injurious stimulus load in the unintervened state. In contrast, the SPI data of the conventional analgesia group remained in a state of natural variability throughout the operation and did not receive goal-directed treatment, making it more suitable for use in constructing and validating postoperative pain prediction models. This sample size ( n  = 30) meets the minimum requirements for exploratory correlation analysis with ROC curve analysis. Correlation analysis: Pearson correlation coefficients wegenerate a threshold-exceedingre used to assess the strength of the linear association between the Tb-TW-SPI and NRS pain scores within the PACU. Predictive efficacy assessment and comparison: Plot the ROC curve for Tb-TW-SPI in predicting moderate to severe postoperative pain. Calculate the AUC and its 95% confidence interval as a composite indicator of prediction accuracy. Determine the optimal diagnostic cut-off value for Tb-TW-SPI prediction of moderate-to-severe pain by maximizing the Jordon’s index (sensitivity + specificity − 1) and report the corresponding sensitivity and specificity. Calculate both the mean intraoperative SPI and the AUC of the maximum intraoperative SPI to assess the relative superiority of Tb-TW-SPI. A two-by-two comparison of the AUC of the three metrics was performed using the DeLong test. (All analyses were performed using R version 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria). The pROC package was used for ROC curve generation, and the DeLong test was employed for pairwise comparisons of AUC values.)

A total of 66 cases met the inclusion criteria during the study period, and 61 patients completed the randomization group, and all entered the final analysis (Fig.  1 : Consolidated Standards of Reporting Trials flow chart). Patients in both groups were well-matched on baseline data such as age, body mass index (BMI), ASA classification, duration of surgery, Baseline MAP/HR and Intraoperative MAP/HR with no statistical differences (Table  1 ).Baseline MAP and HR were defined as the values measured after a 5-minute rest in the supine position prior to induction of anesthesia. Intraoperative mean MAP and HR were calculated as the average of all recorded values from the completion of endotracheal intubation to the end of surgical closure. Table 1 Comparison of baseline characteristics of patients in two groups SPI-oriented group ( n  = 31) Conventional analgesia group ( n  = 30) Age (year) 44.62 ± 12.94 38.47 ± 11.45 BMI (kg/m 2 ) 23.0 (4.0) 23.5 (3.0) ASA classification (I/II)ᵃ 4/27 3/27 Duration of surgery (h) 1.90 (1.41) 1.83 (0.80) Baseline MAP (mmHg) 95.10 ± 11.83 90.33 ± 8.53 Baseline HR (bpm) 75.48 ± 8.9 78.07 ± 13.46 Intraoperative MAP (mmHg) 88.07 ± 8.91 86.68 ± 8.03 Intraoperative HR (bpm) 62.25 ± 8.73 63.93 ± 9.77 Blood Loss(ml) 20 (20–65) 20 (20–50) Nasopharyngeal temperature(℃) 36.41 ± 0.38 36.53 ± 0.29 Conservative surgery, n (%) Myomectomy 11(35.48%) 14(46.67%) Ovarian/Adnexal/Tubal surgery 6(19.35%) 8(26.67%) Hysterectomy 10(32.26%) 8(26.67%) Adhesiolysis/Endometriosis/Others 4(12.90%) 0(0%) Data are presented as mean ± SD, median (IQR), or numbers ᵃASA, American Society of Anesthesiologists BMI, Body Mass Index IQR, interquartile range; SD, standard deviation HR,Heart Rate MAP,Mean Arterial Pressure Nasopharyngeal temperature,Nasopharyngeal temperature denotes the lowest intraoperative value recorded Comparison of baseline characteristics of patients in two groups Data are presented as mean ± SD, median (IQR), or numbers ᵃASA, American Society of Anesthesiologists BMI, Body Mass Index IQR, interquartile range; SD, standard deviation HR,Heart Rate MAP,Mean Arterial Pressure Nasopharyngeal temperature,Nasopharyngeal temperature denotes the lowest intraoperative value recorded To assess potential confounding physiological influences on the Surgical Pleth Index (SPI), key intraoperative variables were compared between groups. The depth of anesthesia, as measured by the Bispectral Index (BIS), was maintained within the target range (40–60). Furthermore, there were no statistically significant differences in intraoperative mean arterial pressure (MAP) or heart rate (HR) between the two groups ( P  > 0.05 for both), suggesting comparable hemodynamic profiles. Surgical factors, including pneumoperitoneum pressure, were standardized to 12–14 mmHg and the degree of Trendelenburg positioning was maintained at approximately 30 degrees for all patients as per institutional protocol for laparoscopic gynecologic surgery, without intergroup variation. MAIN OUTCOME: Intraoperative remifentanil dosage was significantly lower in the SPI-directed group than in the conventional analgesia group [5.84(5.02,6.62)vs. 6.96༈5.81,8.19༉ µg/kg/h; P  = 0.016, Wilcoxon rank-sum test] (Table  2 ). Secondary outcome: the difference in NRS pain scores in the PACU between the two groups was not statistically significant [SPI-oriented group: median 2 (IQR 1.5) vs. conventional analgesia group: median 3 (IQR 2), P  > 0.05] (Fig.  2 ). Table 2 Comparison of intraoperative conditions and outcome indicators between the two groups of patients SPI oriented group ( n  = 31) Conventional analgesia group ( n  = 30) Intraoperative propofol(mg/kg/h) 7.117 (6.072,8.068) 7.116 (6.311,8.374) Intraoperative remifentanil(µg/kg/h) * 5.84(5.02,6.62) 6.96(5.81,8.19) Intraoperative hypertension (yes/no) a 14/17 13/17 Intraoperative tachycardia (yes/no) a 2/29 7/23 NRS score in PACUᵇ 2 (1.5) 3 (2) Mean intraoperative SPI 39.85 (13.15) 41.81 (12.12) Maximum intraoperative SPI * 77 (14) 83 (13.5) Tb-TW-SPIᶜ 675 (858.75) 932.5 (946.25) Data are presented as mean (SD), median (IQR), or numbers a Hypertension was defined as MAP >20% above baseline for ≥60 sec; Tachycard ia as HR >100 bpm for ≥60 sec ᵇ NRS, Numeric Rating Scale (0-10); PACU, post-anesthesia care unit ᶜ Tb-TW-SPI, Threshold-based Time-Weighted Surgical Pleth Index * P < 0.05 indicates statistical significance between groups SPI, Surgical Pleth Index.IQR, interquartile range; SD, standard deviation Comparison of intraoperative conditions and outcome indicators between the two groups of patients Data are presented as mean (SD), median (IQR), or numbers a Hypertension was defined as MAP >20% above baseline for ≥60 sec; Tachycard ia as HR >100 bpm for ≥60 sec ᵇ NRS, Numeric Rating Scale (0-10); PACU, post-anesthesia care unit ᶜ Tb-TW-SPI, Threshold-based Time-Weighted Surgical Pleth Index * P < 0.05 indicates statistical significance between groups SPI, Surgical Pleth Index.IQR, interquartile range; SD, standard deviation Fig. 2 Box plots showing the distribution of NRS pain scores in PACU for the SPI-guided group and the conventional analgesia group. The left panel represents the conventional analgesia group, and the right panel represents the SPI-guided group. The boxes represent the interquartile range [3 (2) vs. 2 (1.5), P  > 0.05], the horizontal lines within the boxes indicate the median, and the whiskers extend to the minimum. There was no statistically significant difference between the two groups (Mann-Whitney U test, P  > 0.05) Box plots showing the distribution of NRS pain scores in PACU for the SPI-guided group and the conventional analgesia group. The left panel represents the conventional analgesia group, and the right panel represents the SPI-guided group. The boxes represent the interquartile range [3 (2) vs. 2 (1.5), P  > 0.05], the horizontal lines within the boxes indicate the median, and the whiskers extend to the minimum. There was no statistically significant difference between the two groups (Mann-Whitney U test, P  > 0.05) Spearman rank correlation analysis revealed a weak-to-moderate positive trend between Tb-TW-SPI and peak NRS scores in the PACU, which did not reach conventional statistical significance ( r  = 0.356, P  = 0.053). However, when using moderate-to-severe pain (NRS ≥ 4) as the grouping variable, Tb-TW-SPI values were significantly higher in patients who experienced pain within the conventional analgesia group than in those who did not [Mann-Whitney U test: W = 140, median (IQR): 1327.50 (982.50, 2265.00) vs. 727.50 (407.50, 1177.50); P  = 0.0417], suggesting that this cumulative index can effectively discriminate between different clinical states regarding the occurrence of postoperative moderate-to-severe pain. (Figure  3 a and b). In this exploratory analysis (Fig.  4 ), Tb-TW-SPI showed modest discriminative ability (AUC = 0.74), though the 95% confidence interval was wide ((95% CI: 0.52–0.96) and the p-value (0.037) should be interpreted cautiously given the small sample size ( n  = 30) and low event count ( n  = 9). These findings are hypothesis-generating rather than confirmatory.As a methodological comparison, the AUC of the mean intraoperative SPI was 0.70, and the AUC of the maximum intraoperative SPI was 0.63. Using the Youden index, the optimal cut-off value of Tb-TW-SPI for predicting moderate-to-severe pain in this sample was 1210, yielding a sensitivity of 67% and a specificity of 76% (true positive = 6, false positive = 5). Given the limited number of events ( n  = 9) and the small sample size ( n  = 30), this cut-off should be interpreted as a descriptive, data-driven estimate rather than a validated clinical threshold. It is provided primarily to characterize the present sample and to inform sample size calculations for future confirmatory studies. It should be noted, however, that the absence of a statistically significant difference in this exploratory analysis does not negate the conceptual value of cumulative nociceptive load assessment. The relatively sparse sampling interval (5 min) mandated by the anesthesia information system may have underestimated the true area under the curve for Tb-TW-SPI, thereby attenuating its potential advantage over simpler metrics. Higher-frequency data acquisition would be required to fully evaluate the incremental value of this time-weighted approach. Fig. 3 ( a ) Scatter plot illustrating the correlation between time-weighted SPI (Tb-TW-SPI) and the highest postoperative NRS score recorded in the PACU. A positive linear correlation was observed, but it did not reach statistical significance (Spearman r  = 0.356, P  = 0.053). The solid line represents the linear regression fit. ( b ) Box plot of Tb-TW-SPI values in patients with (NRS ≥ 4) and without (NRS < 4) moderate-to-severe postoperative pain. Median (IQR): 1327.50 (982.50–2265.00) vs. 727.50 (407.50–1177.50). Mann-Whitney U test: W = 140, P  = 0.0417 ( a ) Scatter plot illustrating the correlation between time-weighted SPI (Tb-TW-SPI) and the highest postoperative NRS score recorded in the PACU. A positive linear correlation was observed, but it did not reach statistical significance (Spearman r  = 0.356, P  = 0.053). The solid line represents the linear regression fit. ( b ) Box plot of Tb-TW-SPI values in patients with (NRS ≥ 4) and without (NRS < 4) moderate-to-severe postoperative pain. Median (IQR): 1327.50 (982.50–2265.00) vs. 727.50 (407.50–1177.50). Mann-Whitney U test: W = 140, P  = 0.0417 Fig. 4 Receiver operating characteristic (ROC) curves comparing the predictive performance of three SPI-derived metrics for moderate-to-severe postoperative pain (NRS ≥ 4). The area under the curve (AUC) was 0.74 for Tb-TW-SPI (95% CI: 0.523–0.959), 0.70 for mean SPI(95% CI: 0.459–0.948), and 0.63 for maximum SPI(95% CI: 0.415–0.839). The diagonal grey line indicates the reference line (AUC = 0.5).Pairwise comparisons of AUC using DeLong’s test revealed no statistically significant differences between the metrics (all P  > 0.05) Receiver operating characteristic (ROC) curves comparing the predictive performance of three SPI-derived metrics for moderate-to-severe postoperative pain (NRS ≥ 4). The area under the curve (AUC) was 0.74 for Tb-TW-SPI (95% CI: 0.523–0.959), 0.70 for mean SPI(95% CI: 0.459–0.948), and 0.63 for maximum SPI(95% CI: 0.415–0.839). The diagonal grey line indicates the reference line (AUC = 0.5).Pairwise comparisons of AUC using DeLong’s test revealed no statistically significant differences between the metrics (all P  > 0.05)

The present study initially achieved the predefined dual objectives and obtained findings with both clinical reference value and methodological significance. While the absolute reduction in remifentanil use was 1.12 µg/kg/h (from 5.84 to 6.96 µg/kg/h). Within the framework of enhanced recovery after surgery, a reduction in intraoperative opioid use—even a modest one—may help lower the incidence of opioid-related adverse events. These events include postoperative nausea and vomiting. Importantly, this reduction was achieved without compromising analgesic efficacy, as reflected by the NRS scores recorded in this study. First, this study confirmed that SPI-based oriented analgesic strategies can safely and effectively save opioid dosage in laparoscopic gynecologic surgery. The results of the RCT showed that intraoperative remifentanil dosage was significantly lower in the SPI-directed group (intervention threshold > 50) compared with the conventional analgesia group 1.38 µg/kg/h reduction; from 7.46 to 6.08 µg/kg/h, P  = 0.016) and did not lead to increased pain in the early postoperative period (the difference in NRS scores in the PACU was not statistically significant). This result is consistent with previous studies [ 9 , 10 ] and provides new high-quality evidence for the effectiveness of SPI in the gynecologic laparoscopic surgery population. While this opioid-sparing effect did not translate into a statistically significant improvement in immediate PACU pain scores, it is consistent with the ERAS paradigm of minimizing perioperative opioid exposure. The absence of worsening pain indicates that SPI guidance maintained a comparable analgesic ceiling while titrating to a lower opioid requirement. However, the discrepancy with the findings of Park et al. [ 11 ] suggests that the selection of SPI target thresholds may be surgery-type dependent. The present study used SPI > 50 as the intervention threshold, achieving a good balance between opioid savings and pain control. This suggests that future clinical applications should be individualized by combining the type of surgery, anesthesia regimen, and patient characteristics, rather than mechanically applying a single threshold. Second, the innovative “time-weighted SPI” index proposed in this study provides a new perspective for understanding the relationship between injurious stimuli and postoperative pain and may explain the heterogeneity among previous studies. Exploratory analyses showed that the correlation between Tb-TW-SPI and postoperative pain ( r  = 0.37, P  = 0.044) and its predictive efficacy (AUC = 0.74) were numerically higher AUC than maximum SPI (AUC = 0.63) but differed less from the average intraoperative SPI (AUC = 0.70). The conceptual advantage of Tb-TW-SPI is that it quantifies the cumulative burden of nociceptive-anti-nociceptive imbalance by integrating the intensity and duration of suprathreshold (SPI > 43) SPI exposure. Importantly, this metric does not claim to measure pure nociceptive input; rather, it reflects nociceptive activity that was insufficiently suppressed by the chosen analgesic strategy. Elevated Tb-TW-SPI values may therefore arise from delayed analgesic titration, pharmacodynamic latency of remifentanil, or variability in clinician response, in addition to intense surgical stimulation. This composite nature may paradoxically enhance its predictive utility for postoperative pain, as it captures real-world imperfections in intraoperative analgesia delivery. Although Tb-TW-SPI did not statistically outperform mean SPI (DeLong test, P  > 0.05), its value lies in quantifying cumulative imbalance. However, given the low-resolution (5-min) sampling, the current comparison likely underestimates the true advantage of a time-weighted approach. The 5-min interval may have attenuated the differences between Tb-TW-SPI and simpler metrics, as both suffer from the same undersampling. Future studies using high-frequency data (e.g., 5-second intervals) are required to determine whether a time-weighted metric offers incremental predictive value over static thresholds. Limitations regarding the exploratory ROC analysis. Although the Tb-TW-SPI showed a moderate AUC of 0.74, the 95% confidence interval was wide (0.52–0.96), reflecting substantial estimation uncertainty due to the small sample size ( n  = 30) and low event count (moderate-to-severe pain, n  = 9). The proposed cut-off value of 1210 is therefore preliminary and should not be used for clinical decision-making. Its primary utility lies in providing an effect size reference for designing larger, adequately powered validation studies. Third, the idea of Tb-TW-SPI provides methodological insights to reconcile the controversy in the literature regarding the “optimal” static SPI threshold. Previous studies have reported multiple “optimal” predictive cutoff values ranging from 30 to 44 [ 12 – 15 ], and this inconsistency has caused confusion in clinical applications. Instead of continuing the research path of finding a single optimal static threshold, the present study fixes the commonly used clinical threshold (SPI = 43) and focuses instead on the cumulative amount of stimulation that exceeds the threshold. This cumulative load-based assessment may provide a more stable and comparable dimension that helps to reconcile the heterogeneity between previous studies. This realization is consistent with the recent view that there are limitations in the predictive power of single monitoring metrics [ 16 ], and the integration of more complex dynamic metrics or multimodal information may be needed in the future. The strengths of this study include (1) the use of a prospective RCT design with a high grade of evidence for the primary endpoint; (2) the exploratory analysis was based on data from a control group that did not receive goal-directed therapy, which can more truly reflect the natural associations between variables; and (3) the principle of calculating the Tb-TW-SPI metrics is clear and clinically actionable. Limitations include (1) the single-center study was limited to the gynecologic surgery population, which limits the extrapolation of the conclusions; (2) the most significant limitation of this study is the 5-minute SPI sampling interval. This low resolution almost certainly misses transient nociceptive events (e.g., trocar insertion, pneumoperitoneum, uterine traction), resulting in a conservative and potentially biased estimate of Tb-TW-SPI. Therefore, the reported AUC of 0.74 and the proposed cut-off of 1210 should not be used clinically. Future studies must adopt high-frequency SPI sampling (e.g., every 5 s) to properly evaluate the time-weighted approach; (3)the interpretation of Tb-TW-SPI is inherently composite: as correctly pointed out by the reviewer, elevated SPI values may reflect not only surgical stimulation but also delayed analgesic titration, pharmacodynamic latency of remifentanil, or inter-clinician variability in response. Our current data do not allow us to disentangle these components. Therefore, Tb-TW-SPI should be understood as a metric of cumulative nociceptive-anti-nociceptive imbalance under real-world conditions, not as a pure measure of nociceptive input. Future studies using protocol-driven, automated analgesic delivery (e.g., closed-loop systems) could help separate these factors.(4) the sample size of the exploratory analysis was small ( n  = 30), and the predictive efficacy of the Tb-TW-SPI (AUC = 0.74) and the optimal cut-off value (1210) need to be validated in an independent large-scale cohort; (5) there is a conceptual inconsistency between the therapeutic intervention threshold (SPI > 50) and the primary analytical threshold for Tb-TW-SPI (SPI > 43). We explicitly acknowledge this as a limitation of the original study design. To address this concern, we performed a sensitivity analysis using the intervention-consistent threshold (SPI > 50) to re-calculate Tb-TW-SPI. The results, presented in Supplementary Table S1 , confirmed that the time-weighted metric retains moderate predictive utility (AUC = 0.744) even when the calculation threshold is aligned with the intervention threshold. This supports the robustness of the time-weighted concept beyond the specific threshold chosen.(6) it has not been compared with the data of other control groups who did not receive goal-directed treatment; and (7) it has not been compared with other studies. validation; and (8) no direct comparisons were made with other emerging injurious surveillance metrics (e.g., Nociception Level Index [NOL], Pupillary Pain Index [PPI]); (9) Future studies should record SVV, PPV, fluid administration, and temperature to minimize physiological confounding of SPI-derived metrics; (8)although BIS was used to guide propofol titration (target 40–60) in all patients, the BIS values were not recorded for analysis. However, intraoperative propofol consumption did not differ between groups (Table  2 ), suggesting comparable anesthetic depth. Looking forward, the Tb-TW-SPI index is expected to promote the shift of perioperative analgesia to a precise and individualized management model based on the assessment of “harm load.” Future research directions include (1) validating the predictive efficacy and generalizability of Tb-TW-SPI in multi-center and different types of surgeries; (2) exploring the integration of Tb-TW-SPI as a core parameter into the intelligent analgesic closed-loop control system; (3) conducting comparative effectiveness studies to evaluate the clinical benefits and cost-effectiveness of Tb-TW-SPI compared with the static SPI thresholding strategy and other commercially available monitoring metrics; and promoting the development of perioperative analgesia in a more accurate and individualized way. benefit and cost-effectiveness, and promote the change of perioperative analgesia from empirical medication to precision management.

The results of this study suggest that an SPI-directed analgesia strategy in laparoscopic gynecologic surgery can effectively and safely achieve a moderate reduction in intraoperative remifentanil dosage without increasing the risk of early postoperative pain. In addition, the newly proposed Tb-TW-SPI metric demonstrated exploratory, hypothesis-generating potential for predicting postoperative pain, with an AUC of 0.74. However, Due to the low-frequency SPI sampling (every 5 min), the reported AUC of 0.74 should be considered hypothesis-generating only and requires prospective validation with high-frequency data, and the proposed cut-off (1210) is not ready for clinical use. Large-scale, prospective studies with high-frequency SPI sampling are required before any clinical application can be considered.

Optimizing postoperative pain management is a core aspect of Enhanced Recovery After Surgery (ERAS) [ 1 , 2 ]. Good pain management not only directly affects the quality of patients’ postoperative recovery and early mobility but is also a key factor in reducing postoperative complications and hospitalization time. However, pain, as a subjective feeling, has long lacked objective and precise evaluation indexes, posing a major challenge to the precise implementation and effect assessment of perioperative analgesia. Opioids, as the basis of intraoperative analgesia, and their overuse are closely associated with adverse reactions such as respiratory depression, nausea and vomiting, and postoperative intestinal obstruction [ 3 ]. Therefore, the development of objective and reliable monitoring tools for individualized quantitative assessment of injurious stimuli and precise titration of analgesic drugs has always been an important direction of perioperative medical research [ 4 ]. The Surgical Pleth Index (SPI) is a continuous, noninvasive index derived from the normalized photoplethysmographic pulse wave amplitude and heartbeat intervals, designed to reflect the balance between nociception and anti-nociception during general anesthesia [ 5 ]. Our team’s preliminary systematic evaluation and meta-analysis [ 1 ] included 10 studies involving more than 1000 patients. The results showed that SPI had moderate predictive value for moderate to severe postoperative pain, with a combined sensitivity of 0.74, a combined specificity of 0.65, and a combined AUC of 0.76. However, this meta-analysis revealed significant heterogeneity among studies, and inconsistency in the diagnostic thresholds of the SPI was one of the important sources of heterogeneity. This suggests that clinical strategies relying on a single static threshold (e.g., > 50) have limitations, as they are difficult to adequately reflect the dynamic evolution and cumulative effects of injurious stimuli, reducing the robustness and generalizability of predictions. Physiologically, the intensity of postoperative pain is not only dependent on the instantaneous peak value of the injurious stimulus but may be closely related to the cumulative exposure to the injurious stimulus—i.e., the product of the stimulus intensity and the duration of action [ 6 ]. In order to more accurately quantify the intraoperative injurious stimulus load and to overcome the heterogeneity caused by the inconsistency of diagnostic thresholds in the previous meta-analysis [ 1 ], the present study changed the research strategy: instead of seeking another “optimal” static threshold, we innovatively proposed the “Threshold-based Time-Weighted SPI” (Tb-TW-SPI), based on the commonly used truncation value determined by the previous meta-analysis (SPI = 43) [ 1 , 7 ]. Threshold-based Time-Weighted SPI(Tb-TW-SPI). This metric integrates the intensity and duration of suprathreshold SPI exposure by calculating the weighted cumulative area of SPI values above a set threshold throughout the procedure. It is important to clarify that Tb-TW-SPI does not directly measure the intensity of surgical nociception per se; rather, it quantifies the cumulative burden of nociceptive-anti-nociceptive imbalance—i.e., nociceptive activity that remains insufficiently suppressed by the ongoing analgesic regimen. This composite includes not only surgical stimulation but also factors such as pharmacodynamic latency and the timeliness of analgesic interventions. The metric is expected to provide a more stable and clinically meaningful predictor than a single point-in-time assessment. To test these hypotheses, we conducted a two-phase study: Phase 1 (primary) was a randomized controlled trial to validate the opioid-sparing effect of SPI-guided analgesia; Phase 2 (exploratory) introduced and preliminarily evaluated a novel metric, Tb-TW-SPI, for its potential to predict postoperative pain. Phase I (primary objective): to conduct a randomized controlled trial aimed at obtaining high-quality evidence to validate whether an SPI-directed analgesic strategy can effectively reduce remifentanil dosage in laparoscopic gynecologic surgery. Phase 2 (exploratory aim): To develop and validate a new metric, “time-weighted SPI,” using data generated from this RCT and to assess its potential to overcome the heterogeneity of previous studies and to improve the efficacy of postoperative pain prediction.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2 Supplementary Material 3 Supplementary Material 3 Supplementary Material 4 Supplementary Material 4