SPI, also called the Surgical Plethysmographic Index or Surgical Stress Index, is a simple, non-invasive monitoring tool used to assess nociception during anesthesia by analyzing the waveform and heartbeat of photoplethysmography in pulse oximeters. SPI is calculated using the pulse photoplethysmographic amplitude (PPGA) and heartbeat interval (HBI), which are measurable in the photoplethysmography [9] as follows in this equation,

SPI values range from 0 (no stress) to 100 (maximum stress level), and in previous studies, an SPI range of 20–50 usually indicates adequate analgesia during surgery under general anesthesia [10–13]. It is also recommended to keep SPI values below 50 and avoid rapid increases (more than 10-point increase over a short time) as another criterion to maintain adequate analgesia [14].

The effectiveness of SPI as a tool for monitoring and evaluating nociception and administering analgesics in adults was demonstrated in several studies [10–13,15]. Notably, they have shown that SPI-guided analgesia offers several clinical benefits in comparison to analgesia based on conventional hemodynamic parameters during surgery under general anesthesia [10,12,13,16,17]. In contrast, limitations have been reported for various clinical situations and confounding factors [18–23], one of which may be age. Thus, limitations presented in the studies of pediatric patients must be considered carefully [19,24,25].

Although the number of studies related to SPI in children is limited, efforts to reduce the incidence of postoperative pain and complications by appropriately evaluating nociception during surgery and performing appropriate analgesia have been attempted in various forms. The results have been reported in several ways, also [19,24–26]. Harju et al. observed changes in intraoperative SPI in two groups receiving ultrasound-guided ilioinguinal and iliohypogastric nerve block with saline or ropivacaine before the beginning of surgery after anesthesia induction in patients under 24 months old undergoing inguinal hernia surgery. As a result, an increase in SPI was observed in both groups after endotracheal intubation, whereas an increase in SPI during surgical incision was observed only in the saline-injected group. Harju et al. [26] reported that the reactivity of SPI to surgical stimuli was blunted in the ropivacaine-injected group, suggesting that SPI monitoring may be useful for determining the analgesic effect of the nerve block in pediatric patients undergoing regional analgesia, but also reported that the reactivity of the SPI was rather small, and there was a limitation of marked inter-individual variability in SPI reactions to nociception. More importantly, several other pediatric studies about SPI have shown that children have less confidence in SPI than adults, and thus children should receive intraoperative analgesia based on SPI levels that differ from those of adults. Supporting this, Park et al. [19] found that the cardiovascular structure and function of children—especially the vascular distensibility—is different from adults and that the actual nociception levels in children are not reflected in SPI values and the pediatric SPI value tends to be lower during surgery. A possible explanation is that ultrasonographic examinations of vascular structures showed lower vascular wall stress and higher distensibility in children than in adolescents [27], and basal catecholamine concentrations and resting muscle sympathetic nerve activity are also lower in children than in adults [28]. Due to these characteristics, the stimulation of vascular smooth muscle and vascular contractile force that occur during the activation of the sympathetic nervous system may be less in children. Two factors determine SPI values, HBI and PPGA, of which PPGA depends on the vascular wall distensibility and intravascular pulse pressure [29] and has double the effect on SPI values compared to HBI (Equation 1). Because children have lower vascular contractility and higher vascular distensibility than adults, activation of the sympathetic nervous system due to nociception is unlikely to significantly reduce PPGA as in adults, which can result in underestimation of SPI values.

The notion that there is less reduction in PPGA in children than in adults due to the unique characteristics of children’s vascular response at the time of nociception is supported by comparing the SPI data of previous studies. Compared with the results of two studies that reported changes in SPI, heart rate (HR), and PPGA during endotracheal intubation after administration of 2 μg/kg of fentanyl for anesthesia induction [30,31], the change in SPI values (the amount of increase) after endotracheal intubation was relatively higher in the adult study [30], while the change in HR was similar. Therefore, it can be inferred that the amount of change in PPGA, that is, the amount of decrease in PPGA, would have been higher in adults. Indeed, the median difference for PPGA before and after intubation reported in the pediatric study [31] was small (–0.51). Fig. 2 schematically shows the data from these two studies and the changes which occur in the photoplethysmography in adults and children at the time of nociception, the difference between them, and how the SPI values are changed.

In summary, Park et al. [19] reported that for children undergoing tonsillectomy under general anesthesia, intraoperative fentanyl requirement was lower for the SPI-guided analgesia group than for hemodynamic parameters-based analgesia, but postoperative pain and emergence agitation scores and rescue fentanyl dose were higher in the SPI-guided analgesia group, and thus SPI values in children are less likely to accurately reflect the nociception-antinociception balance due to their unique cardiovascular structure and physiology. It was concluded that re-establishment of the target SPI range or the development of new formulas for children is necessary for proper intraoperative analgesia in children undergoing general anesthesia.

Ledowski et al. [24] also found that SPI cut-off values of 50, the most commonly used cut-off value in SPI-related studies, do not have clinically appropriate sensitivity or specificity to predict the presence of acute pain in the recovery room, and suggested that it may be more appropriate to adjust to a cutoff of 40 or lower (which is lower than the existing cut-off value of 50) to provide intraoperative analgesia that avoids moderate and severe postoperative pain in children. Ledowski et al. [24] reaffirmed that, in children, intraoperative analgesia during general anesthesia should be performed based on SPI values that differ from those in adults.

Song et al. [25] also reported there was no significant difference in SPI changes during cranial pin fixation among the groups received sufentanil infusion at three different rates (0.2, 0.5, and 0.8 μg/kg/h) in pediatric patients aged 2–12 years who underwent cranial pinning under general anesthesia, suggesting that SPI may not be sensitive enough to monitor a level of nociception and analgesia in children.

A systematic review using meta-analysis of six randomized controlled trials, examining the efficacy and safety of SPIguided analgesia and conventional analgesia based on hemodynamic parameters (blood pressure and HR) and during surgery under general anesthesia, showed that SPI-based analgesia reduced intraoperative opioid requirement and shortened extubation time, and there was no difference in the degree of postoperative pain and the incidence of perioperative complications [32]. Based on these findings, the systematic review suggested that nociception monitoring and analgesic administration using SPI are more useful for providing proper analgesia under general anesthesia in various clinical situations. However, considering that the number of randomized controlled studies included in this study is few, and only one of the included studies is for children [19], and the results of other studies related to SPI in children are conflicting [24–26], further studies are needed to determine the efficacy of SPI for intraoperative analgesia, especially in children.

HRV refers to the degree of change (variability) in time and frequency analysis of the interval between successive heartbeats, which reflects the activity and balance of the autonomic nervous system, i.e., the interactions between the sympathetic and parasympathetic nervous systems. HRV monitoring devices calculate and measure HRV parameters such as standard deviations of normal RR intervals on standard electrocardiogram (time-domain analysis) or high-frequency (HF), low-frequency (LF), very low-frequency (VLF) power, and LF/HF ratio (frequency domain [power spectral density] analysis) quickly, easily, and non-invasively [33,34]. The LF/HF ratio is a quantitative measure of the overall balance between the sympathetic and parasympathetic nervous system and, when the ratio is high, indicates an increase in sympathetic activity or inhibition of parasympathetic activity.

HRV can be measured in both conscious patients and those under sedation or anesthesia [35–37], but similar to SPI, it can be affected by several physiological or psychological conditions such as age (typically decreasing with age) [38], drugs [39], psychological problems [40], comorbidities [41,42], depth of sedation or anesthesia [43,44], and surgical stimuli [45]. Several clinical studies examining the correlation between the intensity of noxious stimuli such as surgical stimuli and corresponding HRV in anesthetized adult patients suggested that HRV could be used as an objective pain assessment tool [46,47]. In contrast, studies for HRV in unanesthetized adult patients showed conflicting results [48,49].

In addition, numerous studies have investigated the relationship between established behavioral indicators of pain and HRV at the evaluation of acute pain in pediatric patients, including preterm infants, newborn (neonate) infants, infants, and children in various clinical settings [50–52]. Consequently, HRV usually responded well to pain in most pediatric age groups, but the findings in infants were inconsistent [53]. On the other hand, physiological indicators (nociception monitoring devices), including HRV, are the only tools that can measure nociception in anesthetized younger children, but the results reported so far have limitations due to the nature of HRV, which is susceptible to several factors [54,55].

Therefore, several real-time algorithms and indices using them have been developed to correct such confounding factors of HRV so that changes in HRV correspond well to the intensity of nociception. These algorithms and indexes include analgesia nociception index, cardiorespiratory coherence, and newborn infant parasympathetic evaluation (NIPE).

ANI (MetroDoloris Medical Systems, France) combines electrocardiogram and respiratory rate with high-frequency adjustment (0.15–0.4 Hz) in the frequency domain analysis of HRV and shows parasympathetic activity in the numerical range of 0 (maximal pain) to 100 (no pain) (as opposed to SPI). Previous studies on ANI regarded ANI ≥ 50 as appropriate analgesia and predicted that ANS responsiveness was caused by nociceptive stimulation when ANI < 30 [56,57].

The efficacy of ANI in intraoperative nociception monitoring and analgesia has been well documented through studies performed in adults [56,58,59]. Recent studies in pediatric patients also reported that ANI was more sensitive to detecting surgical stimuli during surgery and useful for monitoring intraoperative analgesia than other hemodynamic parameters [60–62].

Considering that ANI is shown to reflect not only intraoperative pain but also postoperative pain, that is, the degree of pain in conscious patients, it may have advantages over other pain monitoring devices [63,64]. Nevertheless, there are still studies that reported the opposite results [57,65], much like the controversial issue of its reliability in conscious patients in the correlation between HRV and pain intensity [48,49].

Cardiorespiratory coherence, a non-invasive nociception monitoring device, measures a degree of ANS activity by analyzing the linear combination intensity between heart rate and respiration during general anesthesia and assesses the degree of pain in a range from 0 (low coherence, strong nociception) to 1 (high coherence, no nociception). It includes real-time cardiorespiratory coherence; cardiorespiratory coherence algorithm, or wavelet transform cardiorespiratory coherence algorithm. Although there are very few studies evaluating the efficacy of cardiorespiratory coherence to date, most of them have been conducted in children, and they reported notable outcomes that cardiorespiratory coherence is more sensitive and superior to hemodynamic parameters in detecting nociception, antinociception, and movement in pediatric patients undergoing general anesthesia [66–68].

ANI was developed for HRV analysis in children over 2 years old and adults. Newborns, infants, and children under 2 years of age require a modified approach to HRV analysis because they have a lower HRV due to an immaturity of the ANS and high basal heart rate. The NIPE index (MetroDoloris Medical Systems), a modified form of ANI nociception monitoring, was developed for use in newborns (including premature infants) and children under 2 years of age [69]. The basic principle of NIPE is to analyze the parasympathetic activity of the ANS in real-time using HRV analysis. HRV signals above 0.15 Hz allow for automated HRV analysis of data indicative of parasympathetic nervous system activity via a high pass filter and exhibiting physiological respiratory sinus arrhythmia [69]. This automated analysis is quantified by the NIPE index, which ranges between 0–100 to reflect relative parasympathetic activity, with higher values indicating higher levels of parasympathetic activity. A NIPE index of less than 50 in newborns and infants under anesthesia usually suggests a presence of stress or nociception and insufficient analgesia [70].

NIPE does an excellent job reflecting a balance of nociception and antinociception. NIPE decreases in the presence of nociception in anesthetized infants and newborns, while increases in cases of loss of nociception or administration of analgesics [70,71]. In addition, Valencia-Ramos et al. [72] reported that NIPE monitoring reflected a change in the degree of comfort during nebulization for conscious infants with bronchitis in the intensive care unit. NIPE was suggested as a comfort monitoring system for infants. NIPE may also reflect psychological conditions such as psychological stability or discomfort and stress in conscious infants and newborns. Taken together, even if the evidence is still lacking, NIPE monitoring of nociception and analgesia, or discomfort and comfort in newborns and infants under two years of age under anesthesia or awareness may be more effective than other nociception monitoring modalities. However, Cremillieux et al. [73] reported that the NIPE index during painful procedures of premature infants in the neonatal intensive care unit did not reliably reflect acute pain. Thus, further research is required to study these issues.

A SCA measures an increase in stress reflected in changes in the activity of the sympathetic nervous system. When the sympathetic nerves are activated, the plantar sweat glands in the palms and soles are filled with sweat. When the sweat reaches the skin, the skin resistance decreases, and the skin conduction increases, and when the sweat is reabsorbed, the skin conduction decreases again. When the skin nerves are activated by stimulation, both amplitude and frequency of efferent skin nerve bursts increase. Therefore, the increase in the number of skin conductance fluctuations (NSCF) and amplitude of skin conductance fluctuations in SCA measurements may be interpreted as increased activity of the sympathetic nervous system [74,75].

Unlike other nociception monitoring tools, the SCA index is not affected by circulatory changes, cardiac activity, vasoactive drugs, or neuromuscular blockade. Therefore, SCA is more sensitive and specifically associated with pain and noxious stimuli [76]. The SCA index responds quickly (in seconds) and allows continuous and objective monitoring with a wide range of indications in various clinical situations. It also has higher sensitivity and specificity in assessing pain than other monitoring devices currently available. SCA has been extensively studied, particularly in pediatric patients, for its efficacy and is known to be useful in assessing pain and analgesia. For premature infants, SCA is more sensitive and specific than behavioral-state observations in evaluating heel stick pain during blood sampling, tactile stimulation stress, and high decibel sound stimuli [77–79]. Both in healthy newborns and artificially ventilated infants, an increased NSCF on SCA monitoring correlated well with occurrences of pain and discomfort [80,81].

Several studies evaluating the efficacy of SCA concerning general anesthesia in adult patients have shown that the SCA index is useful in monitoring perioperative stress with increasing values during nociceptive stimulation, such as endotracheal intubation and tetanic stimulation, and decreasing values during analgesic infusion [82,83]. Regarding postoperative pain, several studies have reported that the SCA index is well correlated with the numeric rating scale assessed by the recovery room [84,85].

Few studies of SCA have been conducted on children under general anesthesia. Sabourdin et al. [61] reported that the responses of SCA to remifentanil at different infusion rates in children under general anesthesia was less sensitive than those of ANI. Therefore, despite many advantages of SCA, the evidence is still lacking in its efficacy for analgesia in pediatric patients undergoing general anesthesia. Thus further studies are required on these topics.

Pupillometry is based on the evaluation of the pupillary reflex dilatation induced by nociceptive stimulation. Pupillometry is a non-invasive monitoring technique which measures dynamic pupillary diameter by an infrared camera. Pupillary diameter increases in response to nociceptive stimulation.

The degree of pupillary dilatation in patients with anesthesia using propofol or inhalational anesthetics is associated with the intensity of nociceptive stimuli during operation. The pupillary reflex dilatation by nociception in pediatric patients was maintained during deep sedation with ketamine, and the degree of pupillary dilatation was related to the intensity of the nociceptive stimuli [86]. Therefore, pupillometry can be a useful nociception monitoring tool in pediatric patients sedated with ketamine. Additionally, it has been suggested that pupillometry is a useful nociception monitoring device by other studies in children and adolescents undergoing general anesthesia, which showed that nociceptive stimuli during anesthesia cause pupillary reflex dilatation, and administration of opioid analgesics or deep anesthesia (hypnosis) decreases the pupil diameter [87–89]. However, it must be considered that other factors such as severe anxiety [90] or drugs [91,92] also influence the pupil diameter and reflex.