Behavioral pain scales, vital signs, and pupilometry to pain assessment in the critically ill patient: A cross sectional study

Behavioral pain scales, vital signs, and pupilometry to pain assessment in the critically ill patient: A cross sectional study ☆ Yolanda L´ opez-De-Audícana-Jimenez-De-Aberasturi a,b,c,*,1 , Ana Vallejo-De-La-Cueva b,c,2 , Naiara Parraza-Diez d,e,3 a University of the Basque Country UPV/EHU, Vitoria-Gasteiz School of Nursing, Jos´ e Atxotegi, Vitoria-Gasteiz 01009, Spain b Bioaraba Health Research Institute, Jose Atxotegi, Vitoria-Gasteiz 01009, Spain c Osakidetza Basque Health Service, Araba University Hospital, Jose Atxotegi, Vitoria-Gasteiz 01009, Spain d Epidemiology and Public Health Group, Bioaraba Health Research Institute, Vitoria-Gasteiz, Spain e Research Network on Chronic Diseases, Primary Care, and Health Promotion RICAPPS, Madrid, Spain ARTICLE INFO Keywords: Critical illness Nociception Pain Measurement Reflex, Pupillary ABSTRACT Background: Detecting pain in sedated critically ill patients requires utmost attention. Aim: To assess the pain in mechanically ventilated critically ill patients with the Behavioral Pain Scale (BPS), Behavioral Indicators Pain Scale (ESCID), the pupillary dilation response (PDR), and vital signs. Design: Cross-sectional study Methods: The study was conducted between March and December 2019, involving patients with a baseline BPS of 3, ESCID of 0, and RASS between −1 and −4. Patients with mobility limitations or altered pupillary reflexes were excluded. We measured before and after non-painful stimulation (NP) followed by 10–20–30–40 mA stimuli and endotracheal aspirate (ETA). The primary outcome was the pain measured with BPS and ESCID scales and PDR with AlgiScan®pupilometer defined as BPS≥4, ESCID≥1, and PDR≥11,5 %. We performed a descriptive study and analyzed the agreement between pain detection methods. Results: Thirty-one patients were included, and 183 measurements were recorded. The scales showed minimal changes. Approximately 30 % of patients reported pain at a 30 mA stimulus, increasing to over 70 % after ETA. The PDR ranged from 2 % to 6-33% on the ETA, even in pain-free patients, with pain incidence between 70 % and 100 % for the 40 mA and ETA stimuli. Vital signs did not show relevant changes. The PDR had over 90 % agreement with scales for no pain. For higher-intensity stimuli, agreement ranged from 60 % to 80 %. Disagreement occurred when there was no pain by scales (BPS<4; ESCID<1) and pain with PDR (PDR≥11.5). Conclusions: Pain behavioral scores and vital signs were low in critically ill patients. PDR detected a nociceptive pain response in no-pain patients. 1. Introduction Pain is a frequent symptom in medical and surgical critical care units and is still remembered at discharge [1]. Multiple causes of pain make critically ill patients particularly vulnerable [2]. International guidelines emphasize the need for adequate pain assessment, monitoring, and analgesia adjustment to improve patient comfort, and adjusted use of sedative drugs [3,4]. Validated behavioral scales, such as the Behavioral Pain Scale (BPS) and Pain Behavior Scale (ESCID), are recommended for assessing pain in non-communicative patients. However, medication and the severity of the underlying disease may diminish the patient’s behavioral response to pain, particularly in sedated and mechanically ventilated patients [5, 6]. ☆ Trial registration Phase 1 of the project PUPIPAIN ClinicalTrials.gov Identifier: NCT04078113 * Corresponding author at: University of the Basque Country UPV/EHU,Vitoria-Gasteiz School of Nursing, Jos´ e Atxotegi, Vitoria-Gasteiz 01009, Spain. E-mail addresses: [email protected],[email protected] (Y. L´ opez-De-Audícana-Jimenez-De-Aberasturi), delacueva_79@ hotmail.com (A. Vallejo-De-La-Cueva), [email protected] (N. Parraza-Diez). 1 ORCID iD: https://orcid.org/0000-0001-5703-2706 2 ORCID iD:https://orcid.org/0000-0003-4094-6506 3 ORCID iD:https://orcid.org/0000-0002-5902-0351 Contents lists available at ScienceDirect Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro https://doi.org/10.1016/j.clineuro.2024.108644 Received 27 September 2024; Received in revised form 16 November 2024; Accepted 16 November 2024 Clinical Neurology and Neurosurgery 247 (2024) 108644 Available online 18 November 2024 0303-8467/© 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/bync-nd/4.0/ ). Ongoing efforts are essential to enhance pain assessment and management precision tools for patients with a limited ability to express pain behaviorally. Standardized assessment and recording of pain in noncommunicative patients is insufficient. Less than half of intensive care professionals evaluate the presence of pain [7,8]. Subjective evaluation based on clinical experience remains a common practice for the detection of pain in these patients [9,10]. It is important to note that vital signs are unreliable indicators for assessing pain [11]. However, healthcare professionals sometimes consider the variability of vital signs as a factor in evaluating and making therapeutic decisions regarding a patient’s pain situation. This situation highlights the daily limitations and difficulty in detecting pain. Pupillary dilation is a sensitive indicator of pain in sedated patients in a surgical environment [12,13]. The study will explore whether autonomic responses, measured through video pupillometry, can help detect pain in sedated critically ill patients. It will assess reactions to non-painful stimuli, calibrated electrical nociceptive stimuli (10, 20, 30, and 40 mA), and endotracheal aspiration (ETA) stimuli using the BPS, ESCID, pupillary reflex (PDR), and vital signs. Additionally, the study analyzes the behavior of the tools in patients categorized by pain according to the BPS. 2. Methods 2.1. Study design It was a cross-sectional study conducted between March 2019 and December 2019. Patient recruitment and intervention protocol measurements were carried out during the first week of ICU admission. 2.2. Patients Patients were selected in the order of admission to the unit. Patients with the following criteria were included: over eighteen years of age, on mechanical ventilation, with a baseline BPS of three and a RASS between one and four, and unable to communicate verbally. Exclusion criteria included patients with limited behavioral expression of pain (in treatment with muscle relaxants and, neuromuscular diseases), patients with disorders of the pupillary response (ophthalmological pathologies, involvement of the third cranial nerve, Glasgow less than six, intracranial hypertension), and patients in treatment with epinephrine, clonidine, dexmedetomidine, tramadol, ketamine, adrenaline, Ca antagonists, or antiemetics. 2.3. Protocol and measurements All patients were given a protocol of stimuli, starting with a nonpainful stimulation (NP) followed by 10, 20, 30, and 40 mA stimuli (Fig. 1). The non-painful stimulus used was the rubbing of a gauze pad on the forearm. After them, a potentially painful procedure, such as endotracheal suctioning (ETA) was performed [1,2]. Baseline values and maximum post-stimulus variations in all indicators were recorded. Five minutes were waited between each nociceptive stimulus until the patient recovered the baseline values. The patient’s baseline pain was assessed before the protocol and all patients began it with a BPS value of three. The protocol was conducted in a controlled environment. In the hour preceding the bstimulation, we avoided painful interventions and silenced to prevent and minimize distractions. The analgosedation regimen and the rest of the medications remained constant before, during, and after the stimuli. Three researchers measured behavioral scales and performed pupillometry simultaneously. A collaborating unit member recorded vital signs from the bedside monitor. All measurements performed by the investigators were blinded to each other. The nurse in charge of the patient performed the NP and ETA stimuli. 2.4. Variables 2.4.1. BPS and ESCID scales The BPS scale is a validated scale with psychometric properties to assess pain in sedated patients [14] whose score ranges from 3 to 12 while the ESCID, a scale recommended by the Spanish Society of Intensive Care Medicine and Coronary Units [6], from 0 to 10. Scores of 1–3 indicate mild to moderate pain, 4–6 indicate medium to severe pain, and above 6 indicate severe pain. Our unit’s analgesia protocol included the validated BPS and ESCID scales. Values of BPS greater than or equal to four and ESCID greater than or equal to one were considered pain [6,15,16]. 2.4.2. Vital signs Vital signs, hemodynamic parameters (heart rate-HR, an invasive systolic blood pressure-SBP, diastolic blood pressure-DBP, and medium blood pressure-MBP), respiratory parameters (respiratory rate-RR and pulse oximetry saturation-02 Sat) were recorded with continuous multimodally monitors. Variations higher than 10 % were considered indicative of pain. 2.4.3. Pupil dilation response The pupillary dilation reflex is a physiological autonomic response to stimuli. In this study, PDR was quantified as the percentage change from the initial size of the pupil, measured objectively using video pupillometry. For this purpose, the AlgiScan®portable video pupillometer (ID Company, Marseille, France) was used. The study employed both Dilation Reflex Mode (DRP) for continuous measurements during the NP and ETA stimulus, and Tetanus Mode for assessing responses to calibrated electrical stimuli (10–40 mA). These calibrated nociceptive stimuli were applied through an electrode placed on a clean, uninjured area of skin over the ulnar nerve at the wrist. Measurements were made at the optimal impedance level, as indicated by the pupilometer and were taken with a 5 minutes wait time between each. Pupillometry was conducted on one randomly selected eye, while the other was closed. We used a protective silicone screen on the study eye. Measurements were made once the pupil had stabilized in darkness. A PDR value greater than 11.5 % was considered pain, as indicated by previous studies [17]. 2.5. Statistical analysis This study is part of PUPIPAIN project (ClinicalTrials.gov Identifier: NCT04078113), for which a sample size of at least 50 pain-response 40mA 10mA 20mA 30mA NP ETA NP ETA Fig. 1. Study protocol design. NP: Non painful estimulus; mA: Miliamperes; ETA: Endotracheal aspiration; 5 minutes was maintained between stimulations. Y. L´ opez-De-Audícana-Jimenez-De-Aberasturi et al. Clinical Neurology and Neurosurgery 247 (2024) 108644 2 measures according to BPS was established [17]. Patients’demographic characteristics were described using medians and interquartile range or frequency analysis as appropriate. We conducted a descriptive and graphical study of the scores of BPS, ESCID, PDR, and vital signs. The pain score for the different tools according to the BPS pain classification was calculated. The agreement between the BPS and the other pain assessment tools was also evaluated using the kappa index. The analyses were repeated to study the pain of the different tools according to PDR. Statistical analysis was performed with IBP SPSS Statistics, version 23 for Windows. A statistical significance level of p=0.05 was considered in the study. The data from the study are available in a private repository of Basque Health System as required by Spanish law [Real Decreto 1090/ 2015]. 2.6. Ethical considerations The Clinical Research Ethics Committee of Basque Country approved this study (CEIC-X; internal code: PI2017100). All patients were hemodynamically stable, receiving continuous intravenous analgesics and sedatives. Relatives or legal representatives were informed and provided written consent for participation. 3. Results 3.1. Demographic and clinical characteristics of participants Thirty-one patients were included, and 183 measurements were recorded. Fig. 2 shows the flowchart of the recruited patients and the general characteristics of the patients are presented in Table 1. The mean age of the 31 included patients was 62.9 (17.1) years. They had a Charlson comorbidity index of 3.59 (0.55), an APACHE II score of 21 (16.5–23.5), and 7.4 % died during ICU admission [18]. All patients received an analgesic treatment regimen and had a moderate level of sedation with a BIS of 63 (19.46) and an RASS of −3.39 (0.76). 3.2. Variation in scales, vital signs and PDR scores, and incidence of pain after stimulations Fig. 3 graphically illustrates the tools’behavior in response to the stimuli. We have included the results in supplementary Table SDC 1 for further details. ASSESSED FOR ELIGIBILITY (n= 50) ENROLLED (n= 32) Excluded pacients (n=18): 1. Family refusal (n=6) 2. Not mee�ng inclusion criteria (RASS>-1) (n=6) 3. No family references (n=4) 4. Technical problem (n=2) Pa�ent excluded for safety (n=1) FINAL SAMPLE (n= 31) GLOBAL MEASUREMENTS (n= 183) Measurements not carried out (n=3): 1. Protocol adjustment (n=2) 2. Incomplete protocol (n=1) Fig. 2. Flowchart of recruited patients. Table 1 General characteristics of the patients (n=31). Characteristics Values Age (yr) mean (SD) 62,35 (17,08) Sex ratio Men, Women, n (%) 19/12 (61.3/ 38.7) Body mass index (kg/m 2 ), median (25–75th percentile) 27 (24−30) Acute Physiology And Chronic Health Evaluation II, median (25–75th) percentile) 21 (16.5–23.5) Charlson comorbility Index, mean (SD) 3.59 (0.55) ICU mortality, n (%) 4 (14.8) After-ICU mortality, n (%) 2 (7.4) Glasgow, median (25–75th percentile) 15 (15−15) Bispectral index (BIS), mean (SD) 63 (19.46) RASS, mean (SD) −3.39 (0.76) Admission Values Acute respiratory failure n (%) 8 (25.8) Postoperative n (%) 10 (32.2) Sepsis n (%) 9 (29.1) Multiple trauma n (%) 4 (12.9) Sedative, analgesic and vasoactive drugs during the procedure Values Propofol, n (%) 9 (29.03) Dose, (mg/kg/h) mean (SD) 1.49 (0.88) Midazolam, n (%) 16 (51.6) Dose (mg/kg/h), mean (SD) 0.09 (0.05) Fentanyl, n (%) 20 (64.5) Dose (µg/kg/h), mean (SD) 1.14 (0.74) Remifentanil, n (%) 8 (25.8) Dose (µg/kg/h), mean (SD) 7.05 (4.12) Morphine, n (%) 3 (9.6) Dose (mg/kg/h), mean (SD) 1.00 (0.44) Norepinephrine, n (%) 22 (71) Dose (µg/kg/min), mean (SD 0.2 (0.13) Dobutamine, n (%) 3 (9.7) Dose (µg/kg/min), mean (SD) 4.10 (0.97) Y. L´ opez-De-Audícana-Jimenez-De-Aberasturi et al. Clinical Neurology and Neurosurgery 247 (2024) 108644 3 No significant changes were noted in median scores for BPS and ESCID after 10–40 mA stimuli, achieving scores of 3 and 0, respectively. Less than 10 % of the patients showed pain (BPS≥4 or ESCID≥1) for the lower-intensity stimuli, but this increased with intensity: nearly 30 % for 30 mA, over 40 % for 40 mA, and over 70 % for ETA. Similarly, vital signs show a variation of no more than 3 %, with less than 25 % of patients experiencing pain (Var.≥10 %) even at the highest intensities. PDR showed a 2 % and 3 % variation for no-pain (NP) and 10 mA stimuli. This increased from 6 % to 34 % with higher intensities (20 mA to ETA). Less than 7 % showed pain (PDR≥11,5 %) with NP or 10 mA; however, 30 % reported pain with 20–30 mA stimuli, 70 % with 40 mA, 0 5 10 15 20 25 30 35 40 0 10 20 30 40 50 60 70 80 90 100 NP 10mA 20mA 30mA 40mA ETA score of the different tools represented by lines srabybdetneserperniaphtiwstneitap% BPS ESCID PDR BPS score ESCID score PDR score 0 5 10 15 20 25 30 35 40 0 10 20 30 40 50 60 70 80 90 100 NP 10mA 20mA 30mA 40mA ETA score of the different tools represented by lines % patients with pain represented by bars SBP DBP MBP HR RF O2 Sat HR score RF score SBP score DBP score MBP score O2 Sat score Fig. 3. Pain outcome with different tools and scores. BPS: Behavioral Pain Scale; DBP: Diastolic blood pressure; ESCID: Behavioral Indicators of Pain Scale; ETA: Endotracheal aspiration; HR: Heart rate; IQR: Interquartile range; MBP: Mean arterial blood pressure; N: patient number; NP: Non-painful; PDR: Pupillary dilation reflex; RF: Respiratory rate; SBP: Systolic blood pressure; O2 SAT: Oxygen saturation. Table 2 Tool scores according to BPS. Medians and interquartile range. NP 10 mA 20 mA 30 mA 40 mA ETA ESCID        BPS<4 BPS≥4 0 (0−0) – 0 (0−0) – 0 (0−0) 2 (1-x) 0 (0−0) 2 (1–2.25) 0 (0−0) 2 (1−3) 0 (0−0) 4 (2.25–4.75) PDR        BPS<4 BPS≥4 2 (1−5) – 3 (0−7) – 6 (3.5–11.5) 14.5 (12-xx) 7 (3.5–10.5) 19.5 (6.2–44.2) 15 (8−18) 20 (12.5–44.5) 33 (18−53) 37.5 (25–50.7) HR        BPS<4 BPS≥4 0 (−1.4–1.1) – 1 (−1−2) – 0 (−1.7–1.8) −1,2 (−1.2-xx) 0 (−0.5–3.6) 0.5 (−1.2–3) 1.2 (−1.2–2.1) 0 (0−3) 1.6 (0–7.5) 1.6 (−1.5–8.3) RF        BPS<4 BPS≥4 0 (0−0) – 0 (0−0) – 0 (0−0) −2.6 (−5.2-xx) 0 (0−0) 0 (0−0) 0 (0−0) 0 (0–12.9) 0 (0–8.3) 0 (0–15.1) SBP        BPS<4 BPS≥4 0 (−1.1–1.7) – 0 (−0.7–4.4) – 0 (−0.8–1.7) −1.3 (−3.7-xx) 0.8 (−0.3–2.8) 0.3 (0–4.6) 0.8 (−1.6–3.1) 0 (−1.3–5.2) 1.4 (−0.8–7.) 3 (−0.5–7.7) DBP        BPS<4 BPS≥4−1.3(−3.8–1.7) – 1.4 (−1.6–2.6) – 0 (−1.7–1.8) 0.8 (−1.7-xx) 1.5 (−0.7–3.3) 1.3 (−0.4–5.1) 4.4 (−2.7–5.5) 0 (−2.7–4.4) 1.2 (−7.3–5) 1.6 (−3.6–10.4) MBP        BPS<4 BPS≥4 0 (−2.6–2.7) – 1.1 (−1.3–2.3) – 0 (−0.5–2) −0.5 (−2.5-xx) 0 (−1.7–3.5) 0.92 (−0.2–6) 1.1 (−2.7–3.5) 0 (−2.1–3.7) 2.4 (−2.6–8.1) 1.2 (−1.3–6.9) O2 Sat        BPS<4 BPS≥4 0 (0−0) – 0 (0−0) – 0 (0−0) 0 (0−0) 0 (0−0) 0 (−1.02–0) 0 (0−0) 0 (0−0) 0 (0−0) 0 (−1−0) DBP: Diastolic blood pressure; ESCID: Behavioral Indicators of Pain Scale; ETA: Endotracheal aspiration; HR: Heart rate; IQR: Interquartile range; MBP: Mean arterial blood pressure; NP: Non-painful; PDR: Pupillary dilation reflex; RF: Respiratory rate; SBP: Systolic blood pressure; O2 SAT: Oxygen saturation; –: No data. Y. L´ opez-De-Audícana-Jimenez-De-Aberasturi et al. Clinical Neurology and Neurosurgery 247 (2024) 108644 4 and 100 % with ETA. 3.3. Variation in scales, vital signs, and PDR scores categorized by BPS as pain, no pain patient Table 2 presents the tool scores for patients previously classified with and without pain according to BPS. In the non-pain stimuli, the ESCID scored 0 for all pain-free patients, while the PDR recorded 2 %, and vital signs showed no more than 1 % variation. The PDR increased with stimulus intensity, rising from 2 % to 6–33 % in the ETA. In pain patients, according to the BPS, the ESCID showed scores of 2 in the 20–40 mA stimuli and 4 in the ETA. PDR scores increased with stimulus intensity from 14.5 % at 20 mA to 37.5 % in the ETA, and all vital signs showed variations of more than 1 % except respiratory rate, which did not change. All the tools had a lower score in cases where the BPS did not identify pain compared to those where it did. 3.4. Agreement between BPS scale with vital signs and PDR as pain assessment tools Table 3 shows the agreement levels for classifying patients’pain using the BPS scale compared to other available pain assessment tools. The ESCID behavioral scale had a more than 90 % concordance with the BPS for all intensity stimuli. The kappa index was above 0.815. Discrepancies were found in the ETA, with false positives and negatives occurring with the ESCID. Vital signs had greater than 90 % agreement with the BPS at the NP and 10 mA stimulus, but this dropped to 40 % at higher intensity levels. The kappa index obtained for each stimulus was no higher than 0.1. The discrepancies were observed with BPS≥4 and vital sign variations <10 %. The PDR had over 90 % agreement with the BPS in the NP and 10 mA stimuli. For higher-intensity stimuli, agreement ranged from 60 % to 80 %. The Kappa index for each stimulation ranged from 0.3 to 0.5. Disagreements occurred mainly in cases when the BPS indicated no pain (BPS<4) but the PDR indicated pain (PDR≥11.5). 3.5. Variation in the scales, and vital signs scores in patients categorized by PDR as pain, no pain patient Table SDC 2 in the supplementary material presents the results of pain assessments classified by the PDR tool. The BPS and ESCID showed scores of three and zero in patients who did not show pain according to the PDR, even with the 30 and 40 mA stimuli. In contrast, patients experiencing pain (PDR ≥11.5 %) scored four and three on these scales. All patients reported pain during ETA. For vital signs, patients without pain showed variations from −3–1.5 %, while those with PDR ≥11.5 % had variations ranging from Table 3 Agreement between BPS and different pain assessment tools. NP 10 mA 20 mA 30 mA 40 mA ETA ESCID Agreement N (%) 31 (100) 30 (96.8) 31 (100) 30 (96.7) 27 (96.4) 29 (93.6) Disagreement N (%) BPS<4; ESCID≥1 BPS≥4; ESCID<1 – – – 1 (3.2) 1 (3.2) – – – – 1 (3.2) – 1 (3.2) 1 (3.6) 1 (3.6) – 2(6,4) 1 (3.2) 1 (3.2) Kappa (p) – – 1.00** 0.924** 0.929** 0.815** PDR Agreement N (%) 29 (93.5) 29 (93.5) 24 (77.5) 25 (80.7) 18 (64.3) 24 (77.4) Disagreement N (%) BPS<4; PDR≥11.5 BPS≥4; PDR<11.5 2 (6.5) 2 (6.5) – 2 (6.5) 2 (6.5) – 7 (22.6) 7 (22.6) – 6 (19.4) 3 (9.7) 3 (9.7) 10 (35.7) 8 (28.6) 2 (7.1) 7 (22.6) 7 (22.6) – Kappa (p) – – 0.29* 0.56* 0.301  HR Agreement N (%) 31 (100) 29 (93.5) 26 (83.9) 18 (58.1) 16 (57.1) 13 (42.0) Disagreement N (%) BPS<4; Var>10 % BPS≥4; Var<10 % – – – 2 (6.5) 2 (6.5) – 5 (16.1) −3 (9.7) 2 (6.5) 13 (41.9) −3 (9.7) 10 (32.3) 12 (42.9) 1 (3.6) 11 (39.3) 18 (58.0) 1 (3.2) 17 (54.8) Kappa (p) – – −0.084 −0.175 0.092 0.082 RF Agreement N (%) 28 (90.3) 28 (90.3) 27 (87.1) 22 (70.9) 16 (57.1) 13 (42.0) Disagreement N (%) BPS<4; Var>10 % BPS≥4; Var<10 % 2 (6.5) 3 (9.7) – 2 (6.5) 3 (9.7) – 3 (9.7) 2 (6.5) 2 (6.5) 9 (29.1) – 9 (29.0) 12 (42.9) 2 (7.1) 10 (35.7) 18 (58.0) 1 (3.2) 17 (54.8) Kappa (p) – – −0.069 0.131 0.102 0.082 SBP Agreement N (%) 29 (93.5) 31 (100) 29 (93.5) 20 (64.5) 16 (57.1) 12 (38.7) Disagreement N (%) BPS<4; Var>10 % BPS≥4; Var<10 % 2 (6.5) 2 (6.5) – – – – 2 (6.5) – 2 (6.5) 11 (35.5) 1 (3.2) 10 (32.3) 12 (42.9) – 12 (42.9) 19 (31.3) 2 (6.5) 17 (54.8) Kappa (p) – – – −0.062 0.082 0.003 DBP Agreement N (%) 28 (90.3) 30 (96.8) 26 (83.9) 22 (70.9) 14 (50.0) 14 (45.2) Disagreement N (%) BPS<4; Var>10 % BPS≥4; Var<10 % 2 (6.5) 3 (9.7) – 1 (3.2) 1 (3.2) – 5 816.1) 3 (9.7) 2 (6.5) 9 (29.1) – 9 (29.0) 14 (50.0) 3 (10.7) 11 (39.3) 17 (54.8) 1 (3.2) 16 (51.6) Kappa (p) – – −0.084 0.131 −0.048 0.108 MBP Agreement N (%) 29 (93.5) 31 (100) 27 (87.1) 21 (67.7) 17 (60.7) 12 (38.7) Disagreement N (%) BPS<4; Var>10 % BPS≥4; Var<10 % 1 (3.2) 2 (6.5) – – – – 3 (9.7) 2 (6.5) 2 (6.5) 10 (32.3) – 10 (32.3) 11 (39.3) – 11 (39.3) 19 (31.3) 1 (3.2) 18 (58.1) Kappa (p) – – −0.069 – 0.163 0.058 O2 Sat Agreement N (%) 31 (100) 31 (100) 29 (93.5) 21 (67.7) 15 (53.6) 7 (22.6) Disagreement N (%) BPS<4; Var>10 % BPS≥4; Var<10 % – – – – – – 2 (6.5) – 2 (6.5) 10 (32.3) – 10 (32.3) 13 (46.4) – 13 (46.4) 24 (77.4) – 24 (77.4) Kappa (p) – – – – – – DBP: Diastolic blood pressure; ESCID: Behavioral Indicators of Pain Scale; ETA: Endotracheal aspiration; HR: Heart rate; IQR: Interquartile range; MBP: Mean arterial blood pressure; N: patient number; NP: Non-painful; PDR: Pupillary dilation reflex; RF: Respiratory rate; SBP: Systolic blood pressure; O2 SAT: Oxygen saturation; –: No data. Pain scores: ESCID ≥1, PDR ≥11,5 %, Vital signs variation before and after stimuli ≥10 %. ** p<0.001; *p<0.05 Y. L´ opez-De-Audícana-Jimenez-De-Aberasturi et al. Clinical Neurology and Neurosurgery 247 (2024) 108644 5 −2.5–3.5 %. 3.6. Agreement between PDR with scales and vital signs as pain assessment tools Table SDC 3 outlines the concordance between the PDR and other pain assessment tools. The BPS and ESCID had more than 90 % concordance with the PDR for the NP and 10 mA stimuli. Agreement decreased with stimulus intensity and remained above 77 % for AET. The kappa index with the scales ranged from 0.2 to 0.6, with significant discrepancies noted for stimuli above 20 mA in patients with PDR≥11.5 %. Vital signs had 80–90 % agreement with PDR with NP and 10 mA stimuli, decreasing to 60–30 % with higher stimuli. The kappa indices were around 0.1, with the most significant discrepancies observed in patients with PDR≥11.5 %. 4. Discussion This study aimed to describe the behavior of clinical assessment tools for pain and autonomic responses as the vital signs and pupillary reactivity to nociceptive stimuli. According to international recommendations, the adequate assessment and treatment of pain is a priority area of attention in sedated patients. It recommends the use of indirect pain assessment methods such as behavioral scales [4,9,14]. However, assessing pain in critically ventilated patients using behavioral scales involves complexities that need careful consideration [5,6]. Applying behavioral scales to evaluate pain should be more common than it is. The implementation of scales is low, potentially exacerbated by the mistaken belief that immobile, sedated patients are unlikely to experience pain [3]. The potential subjectivity of scales makes it difficult to reach a consensus on pain diagnosis among professionals [7]. There is a clinical concern that patients may be underdiagnosed. Research indicates that in critically ill patients sedated and on mechanical ventilation, behavioral scale scores, and vital signs showed minimal changes in response to nociceptive stimuli, with no significant variations even in patients experiencing pain. These results aligned with previous multicenter studies, in which tool variations were also minimal [6,16]. However, there was a high incidence of pain during ATE, with values higher than four points on the BPS scale in 70 % of the patients. Pupillary size variation showed changes from the lowest intensity stimulus. It showed increasing variations with increasing stimulus intensity and reached values above 37 % in patients with pain, according to BPS. A nociceptive response was noted at a pain rate, PDR≥11,5 %, from a 20 mA stimulus. These findings corroborated those obtained in studies performed in deeply anesthetized surgical patients, which also evidenced a progressive increase in PDR with the intensity of calibrated electrical stimuli in a surgical approach [19] and higher PDR values in areas without sensory blockage [13,20]. Although cautious in its interpretation, it is essential to note that pain has been reported in 100 % of the studied patients, according to PDR. However, it should also be noted that, in this study, the PDR after AET reached scores of 33 % in patients who did not show pain according to the scale. This nociceptive response was close to that reached by patients who did show pain according to BPS and ESCID. Paulus [16] and Sabourdin [19] identified an increase in the PDR with increasing stimulus intensity in critically ill patients with RASS-5. They also identified higher scores in patients who demonstrated pain. Still, the authors did not present data from the analysis of the pupillary response by pain/no pain groups according to the BPS scale in their works. No studies have been found that have helped us discuss this situation. The validity of the indicators to detect pain versus BPS was variable. A high degree of agreement of behavioral scales to identify pain was confirmed [6]. The agreement of behavioral scales with vital signs was weaker, and there was no correlation between behavioral scales or PDR, especially in patients with pain. Vital signs are often used to suggest potential pain, especially in patients with behavioral limitations. Current evidence does not support using vital signs for this purpose [9,14, 21]. Patients taking drugs that slow the heart rate, patients on cardiotonics, vasopressors, opioids, or those experiencing significant physiologic stress may not show changes in vital signs related to painful stimuli. Despite this, given the existing clinical need for objective tools to identify pain, systems are being developed to integrate some vital signs as indicators of pain [22,23]. In the case of the PDR, agreement with the pain scales was good for low-intensity stimuli. However, the degree of agreement decreased after high-intensity stimuli, revealing some patients who appeared calm but had elevated nociceptive responses (PDR ≥11.5 %). Given the impossibility of patients to verbalize pain and evidence of its presence, it could only evidence elevated nociception in patients without signs of pain. However, this fact may lead us to consider under-diagnosed pain patients. We have not yet found previous studies to compare the results. Our findings could have significant clinical implications, particularly for critically ill patients on mechanical ventilation with limited mobility and an inability to communicate their pain. The incidence of postintervention pain in the units remains very high and an objective tool to monitor pain and guide individualized treatments remains needed. Pupillary reflex dilation would be a reliable physiological indicator of noxious stimulation independent of other physiological variables. Some studies propose the pupillometry as an instrument to assess autonomic responses to stimuli to adjust the patient’s need for analgesia avoiding situations of overor under-analgesia in the surgical field [13,20,24]. Li [25] advanced that pupillary size variation could be an optimal indicator of pain in critically ill patients unable to demonstrate a behavioral response to pain. PDR has obtained sensitivity and specificity for detecting pain close to 80 % in critically ill-sedated patients [15−17, 26]. This study introduces a new line of research. With all, this study would have several limitations. Firstly, the patients’behavioral response to pain was low. It would have affected the recording the BPS and the classification of pain according to the scale. On the other hand, pupillary reactivity is an autonomous reflex in stressful situations and not only in pain. Even though all patients studied showed a PDR after the stimuli, the presence of opioids, would decrease its amplitude [27]. Due to this, stressor stimuli were limited, and opioid doses remained constant during the protocol to minimize potential bias in the measurements. Further studies are needed to validate these findings and evaluate the technique’s applicability in clinical practice for mechanically ventilated patients. 5. Conclusions The PDR recorded higher modification values after stimuli in this study than the other tools, with strong agreement between the scales. The PDR detected elevated nociceptive responses in patients who did not show behavioral pain responses. Ethics approval and consent to participate The Clinical Research Ethics Committee of Basque Country approved this study. CEIC-E; internal code: PI2017100. Informed consent was obtained from all individual participants included in the study or their legally authorized representatives. The study was covered by Professional Civil Liability insurance in any adverse events. Funding This work was supported by AID in Research Projects from the Basque Government Department of Health with file number 2018111017. Y. L´ opez-De-Audícana-Jimenez-De-Aberasturi et al. Clinical Neurology and Neurosurgery 247 (2024) 108644 6 CRediT authorship contribution statement Naiara Parraza-Diez: Writing –original draft, Software, Methodology, Formal analysis. Ana Vallejo-De-La-Cueva: Writing –original draft, Investigation, Funding acquisition, Conceptualization. Yolanda L´ opez-De-Audícana-Jimenez-De-Aberasturi: Writing –original draft, Supervision, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments To those who formed part of the research team collaborating in the recruitment and data collection phase of the PUPIPAIN project for their availability, dedication and time, as well as to the, head of investigation service of Health Research Institute, BIOARABA, and the heads of service and the nurse staff of the Intensive Care Units of the Araba University Hospital, for the support and facilitating attitude shown throughout the project. Conflicts of interest The authors have no conflicts of interest to report. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.clineuro.2024.108644. References [1] G. Robleda, F. Roche-Campo, L. Membrilla-Martínez, A. Fern´ andez-Lucio, M. Villamor-V´ azquez, A. Merten, et al., Evaluaci´ on del dolor durante la movilizaci´ on y la aspiraci´ on endotraqueal en pacientes críticos, Med. Intensiv. 40 (2) (2016) 96–104, https://doi.org/10.1016/j.medin.2015.03.004. [2] K.A. Puntillo, A. Max, J.F. Timsit, L. Vignoud, G. Chanques, G. Robleda, et al., Determinants of procedural pain intensity in the intensive care unit. The Europain®study, Am. J. Respir. Crit. Care Med. 189 (1) (2014) 39–47, https://doi. org/10.1164/rccm.201306-1174OC. [3] J. Caballero, M. García-S´ anchez, E. Palencia-Herrej´ on, T. Mu˜ noz-Martínez, J. M. G´ omez-García, I. Ceniceros-Rozal´ en, et al., Sobresedaci´ on Zero como herramienta de confort, seguridad y gesti´ on en la unidades de cuidados intensivos, Med. Intensiv. 44 (2020) 239–247, https://doi.org/10.1016/j.medin.2019.09.010. [4] J.W. Devlin, Y. Skrobik, C. G´ elinas, D.M. Needham, A.J.C. Slooter, P. P. Pandharipande, et al., Clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU, Crit. Care Med. 46 (9) (2018) e825–e873, https://doi.org/10.1097/CCM.0000000000003299. [5] C. G´ elinas, C. Johnston, Pain assessment in the critically ill ventilated adult: validation of the Critical-Care Pain Observation Tool and physiologic indicators, Clin. J. Pain. 23 (6) (2007) 497–505, https://doi.org/10.1097/ AJP.0b013e31806a23fb. [6] I. Latorre-Marco, M. Acevedo-Nuevo, M. Solís-Mu˜ noz, L. Hern´ andez-S´ anchez, C. L´ opez-L´ opez, M.M. S´ anchez-S´ anchez, et al., Psychometric validation of the behavioral indicators of pain scale for the assessment of pain in mechanically ventilated and unable to self-report critical care patients, Med Intensiv. 40 (8) (2016) 463–473, https://doi.org/10.1016/j.medin.2016.06.004. [7] S. Arias-Rivera, C. L´ opez-L´ opez, M.J. Frade-Mera, G. Via-Clavero, J.J. RodríguezMond´ ejar, M.M. S´ anchez-S´ anchez, et al., Valoraci´ on de la analgesia, sedaci´ on, contenciones y delirio en los pacientes ingresados en unidades de cuidados intensivos espa˜ nolas. Proyecto ASCyD, Enfermería Intensiv. 31 (1) (2020) 3–18, https://doi.org/10.1016/j.enfi.2018.11.002. [8] M. García-S´ anchez, J. Caballero-L´ opez, I. Ceniceros-Rozal´ en, C. Gim´ enez-Esparza Vich, M.A. Romera-Ortega, C. Pardo-Rey, et al., Pr´ acticas de analgosedaci´ on y delirium en Unidades de Cuidados Intensivos espa˜ nolas: Encuesta 2013-2014, Med Intensiv. 43 (4) (2019) 225–233, https://doi.org/10.1016/j.medin.2018.12.003. [9] K.M. Garrett, Best practices for managing pain, sedation, and delirium in the mechanically ventilated patient, Crit. Care Nurs. Clin. North Am. 28 (4) (2016) 437–450, https://doi.org/10.1016/j.cnc.2016.07.004. [10] A. Gerber, A.L. Thevoz, A.S. Ramelet, Expert clinical reasoning and pain assessment in mechanically ventilated patients: a descriptive study, Aust. Crit. Care 28 (1) (2015) 2–8, https://doi.org/10.1016/j.aucc.2014.06.002. [11] T.S. Shahiri, C. G´ elinas, The validity of vital signs for pain assessment in critically ill adults: a narrative review, Pain. Manag Nurs. 24 (3) (2023) 318–328, https:// doi.org/10.1016/j.pmn.2023.01.004. [12] J. Guglielminotti, N. Grillot, M. Paule, F. Mentr´ e, F. Servin, P. Montravers, et al., Prediction of movement to surgical stimulation by the pupillary dilatation reflex amplitude evoked by a standardized noxious test, Anesthesiology 122 (5) (2015) 985–993, https://doi.org/10.1097/ALN.0000000000000624. [13] S. Isnardon, M. Vinclair, C. Genty, A. Hebrard, P. Albaladejo, J.F. Payen, Pupillometry to detect pain response during general anaesthesia following unilateral popliteal sciatic nerve block: a prospective, observational study, Eur. J. Anaesthesiol. 30 (7) (2013) 429–434, https://doi.org/10.1097/ EJA.0b013e32835f0030. [14] J. Barr, G.L. Fraser, K. Puntillo, E.W. Ely, C. G´ elinas, J.F. Dasta, et al., Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit, Crit. Care Med. 41 (1) (2013) 263–306, https:// doi.org/10.1097/CCM.0b013e3182783b72. [15] A.C. Lukaszewicz, D. Dereu, E. Gayat, D. Payen, The relevance of pupillometry for evaluation of analgesia before noxious procedures in the intensive care unit, Anesth. Analg. 120 (6) (2015) 1297–1300, https://doi.org/10.1213/ ANE.0000000000000609. [16] J. Paulus, A. Roquilly, H. Beloeil, J. Th´ eraud, K. Asehnoune, C. Lejus, Pupillary reflex measurement predicts insufficient analgesia before endotracheal suctioning in critically ill patients, Crit. Care 17 (4) (2013) R161, https://doi.org/10.1186/ cc12840. [17] Y. L´ opez de Audícana-Jimenez de Aberasturi, A. Vallejo-De la Cueva, N. Aretxabala-Cortajarena, A. Quintano-Rodero, C. Rodriguez-Nu˜ nez, P. M. Pelegrin-Gaspar, et al., Pupillary dilation reflex and behavioural pain scale: Study of diagnostic test, Intensive Crit. Care Nurs. 74 (2023) 103332, https://doi. org/10.1016/j.iccn.2022.103332. [18] M. Charlson, T.P. Szatrowski, J. Peterson, J. Gold, Validation of a combined comorbidity index, J. Clin. Epidemiol. 47 (11) (1994) 1245–1251, https://doi.org/ 10.1016/0895-4356(94)90129-5. [19] N. Sabourdin, T. Giral, R. Wolk, N. Louvet, I. Constant, Pupillary reflex dilation in response to incremental nociceptive stimuli in patients receiving intravenous ketamine, J. Clin. Monit. Comput. 32 (5) (2018) 921–928, https://doi.org/ 10.1007/s10877-017-0072-5. [20] B. Duceau, M. Baubillier, G. Bouroche, A. Albi-Feldzer, C. Jayr, Pupillary reflex for evaluation of thoracic paravertebral block: a prospective observational feasibility study, Anesth. Analg. 125 (4) (2017) 1342–1347, https://doi.org/10.1213/ ANE.0000000000002003. [21] K. Herr, P.J. Coyne, M. McCaffery, R. Manworren, S. Merkel, Pain assessment in the patient unable to self-report: position statement with clinical practice recommendations, Pain. Manag Nurs. 12 (4) (2011) 230–250, https://doi.org/ 10.1016/j.pmn.2011.10.002. [22] K. Ruetzler, M. Montalvo, O.M. Rotem, E. Ekrami, J. R¨ ossler, J.A.A. Duran, et al., Generalizability of nociception level as a measure of intraoperative nociceptive stimulation: a retrospective analysis, Acta Anaesthesiol. Scand. 67 (9) (2023) 1187–1193, https://doi.org/10.1111/aas.14286. [23] K. Yoshida, S. Obara, S. Inoue, Analgesia nociception index and high frequency variability index: promising indicators of relative parasympathetic tone, J. Anesth. 37 (1) (2023) 130–137, https://doi.org/10.1007/s00540-022-03126-8. [24] N. Sabourdin, J. Barrois, N. Louvet, A. Rigouzzo, M.L. Guye, C. Dadure, et al., Pupillometry-guided intraoperative remifentanil administration versus standard practice influences opioid use, A Random Study Anesth. 127 (2) (2017) 284–292, https://doi.org/10.1097/ALN.0000000000001705. [25] D. Li, C. Miaskowski, D. Burkhardt, K. Puntillo, Evaluations of physiologic reactivity and reflexive behaviors during noxious procedures in sedated critically ill patients, 472.e9-13, J. Crit. Care 24 (3) (2009), https://doi.org/10.1016/j. jcrc.2008.07.005. [26] M. Vinclair, C. Schilte, F. Roudaud, J. Lavolaine, G. Francony, P. Bouzat, et al., Using pupillary pain index to assess nociception in sedated critically Ill patients, Anesth. Analg. 129 (6) (2019) 1540–1546, https://doi.org/10.1213/ ANE.0000000000004173. [27] L. Barvais, E. Engelman, J.M. Eba, E. Coussaert, F. Cantraine, G.N. Kenny, Effect site concentrations of remifentanil and pupil response to noxious stimulation, Br. J. Anaesth. 91 (3) (2003) 347–352, https://doi.org/10.1093/bja/aeg178. Y. L´ opez-De-Audícana-Jimenez-De-Aberasturi et al. Clinical Neurology and Neurosurgery 247 (2024) 108644 7