Evaluation of the averaged parasympathetic tone activity and its dynamic variation to assess intraoperative nociception in relation to hemodynamic changes in dogs

Abstract

This study aimed to determine the performance of the averaged parasympathetic tone activity (PTAm) and its dynamic variation (ΔPTA) to assess intraoperative nociception in relation to heart rate (HR) and direct mean arterial pressure (MAP) in dogs undergoing laparoscopic ovariectomy. This prospective, observational, clinical study included 32 bitches. The PTAm, HR, MAP, and bispectral index (BIS) were assessed before (pre-stimulus), as well as 1 min and 2 min after, four surgical stimuli: insufflation, introduction of trocars, and removal of the left and right ovaries. A two-way ANOVA was performed to compare PTAm, HR, MAP, and BIS data across surgical stimuli. A ≥ 20% drop in PTAm or a ≥ 20% increase in HR and/or MAP regarding the pre-stimulus values was considered a PTAm-drop and/or a hemodynamic response, respectively. The performance of PTAm pre-stimulus, PTAm 1 min, and ΔPTA in predicting the hemodynamic response was assessed by calculation of the area under the receiver operating characteristic (ROC) curve. At insufflation, PTAm decreased after 1 (p = 0.010) and 2 (p = 0.045)min, and ΔPTA was different (p = 0.005) between dogs that presented hemodynamic response and dogs that did not. At PTAm-drop, MAP increased after 1 min (p = 0.001) and 2 min (p = 0.001) with respect to pre-stimulus value, whereas HR and BIS did not change. ROC curves showed a threshold value of PTAm pre-stimulus ≤51 to detect hemodynamic response (sensitivity 69%, specificity 52%). The PTAm and ΔPTA only assessed intraoperative nociception during insufflation. The PTAm pre-stimulus association to the hemodynamic response in anaesthetized dogs showed poor sensitivity and no specificity.

Keywords:

• Analgesia
• blood pressure
• dog
• heart rate
• nociception
• parasympathetic tone activity
• PTA monitor

1. Introduction

Nociception assessment is one of the most challenging aspects of anesthesia, and its intraoperative monitoring is commonly based on changes in heart rate (HR), blood pressure (BP), and/or respiratory rate (RR) (Gruenewald and Ilies 2013). Cardiorespiratory variables can be modified by diverse circumstances such as cardiovascular or central nervous system illness, anesthetic depth, drugs administered, or surgical procedure (Ruíz-López et al. 2020). Thus, nociception assessment based on clinical variables could be inaccurate in human (Gruenewald et al. 2013) and in veterinary (Mansour et al. 2017) medicine. Hence, it will be helpful to find a more accurate method to assess the analgesia–nociception balance (Gruenewald and Dempfle 2017).

Different monitors have been used in human and veterinary medicine to evaluate the nociception–antinociception balance (Ruíz-López et al. 2020). Nociception monitoring has been developed for human medicine on the basis of electroencephalography (i.e. bispectral index, spectral entropy, qNOX index), electromyography (i.e. nociceptive withdrawal reflex), and autonomous nervous system (ANS) activity (i.e. pupillometry, surgical plethysmographic index, CARDEAN index, skin conductance, analgesia nociception index) (Gruenewald et al. 2013; De Jonckheere et al. 2015). The Analgesia Nociception Index (ANI) is a 0–100 non-invasive index that evaluates the autonomic nervous system (ANS) activity on the basis of the registration of the human heart rate variability (HRV), which has shown promising results in improving nociceptive assessment compared to HR or BP, allowing for a more accurate antinociceptive drug administration (Boselli et al. 2015). The dynamic variation of ANI (ΔANI) (a calculated value) (Boselli et al. 2016) showed a better performance than ANI in predicting hemodynamic changes.

The parasympathetic tone activity (PTA) index is homologous to ANI. The PTA index is obtained using a noninvasive device that analyzes the electrocardiogram (ECG) signal of dogs, cats, and horses. It is calculated from the analysis of the HRV, which is based on beat-to-beat oscillations of the R–R interval due to the influence of the ANS (Mansour et al. 2017). Considering that respiration is related to cardiac vagal changes (Smith et al. 2013), and that respiratory sinus arrhythmia affects the R–R interval (Jeanne et al. 2012), stable breathing is needed for the correct performance of the monitor. The PTA monitor displays two PTA indices updated every second, the immediate PTA (PTAi) as result of the last 54 s and the averaged PTA (PTAm) as a result of the last 176 s. The value of PTA indices ranges from 100 to 0, where 100 indicates the highest parasympathetic modulation and 0 indicates the lowest parasympathetic modulation that correlates with a maximal nociception. The PTA monitor also displays another parameter based on the standard deviation of R–R intervals, so-called energy (Ruíz-López et al. 2022). The quality of the measurement is constantly provided by the monitor.

Studies about the intraoperative use of the PTA index in veterinary species are limited. The PTA index and its dynamic variation over 1 min (ΔPTA) predicted noninvasive cardiovascular changes in anaesthetized dogs undergoing different surgeries (Mansour et al. 2017), even when dogs were sedated with different protocols (Mansour et al. 2020). Also, the PTA index seems to be useful to detect nociceptive stimuli during castration in male dogs (Gavet et al. 2022). Furthermore, PTAm was able to detect medium- and high-intensity electrical nociceptive stimuli in experimentally anaesthetized dogs (Aguado et al. 2020). The PTA index was useful to optimize antinociceptive drug delivery, since it was associated with the recognition of antinociception levels between treatments in swine (Leitão et al. 2019). In horses, the PTA index seems to be influenced by the health condition of the animal (Mansour et al. 2021). A study showed that fentanyl constant rate infusion increased PTAm values in anesthetized horses (Dmitrović et al. 2021). However, PTAm seems to be a poor indicator of sympathetic activation after surgical incision in anesthetized horses (Ruíz-López et al. 2022). Moreover, in the same study, it was demonstrated that the PTAm value can be influenced by ketamine but not by dobutamine or morphine (Ruíz-López et al. 2022).

In view of the importance of assessing intraoperative nociception and the limited published studies using this monitor in dogs during clinical surgical conditions, the aim of this study was to determine the performance of PTAm and ΔPTA in assessing intraoperative nociception in relation to HR and direct mean arterial pressure (MAP) in bitches undergoing laparoscopic ovariectomy. It was hypothesized that changes in the PTAm or ΔPTA may coincide with or even anticipate changes in HR or MAP after clinical surgical stimuli.

2. Materials and methods

This observational, prospective, clinical study was performed at the Veterinary Teaching Hospital, University of Córdoba and approved by the ethical committee for animal welfare of the Teaching Hospital of the University of Córdoba (CEBAHCV32/2016). All procedures were conducted in compliance with the ethical principles of good practice in animal experimentation and with previous informed consent of the owners.

2.1. Animals

Thirty-two client-owned adult female dogs undergoing laparoscopic ovariectomy were enrolled in this study. Health status was assessed by means of physical examination, electrocardiography, hemogram, and serum biochemical analyses. Only ASA I or II patients without previous treatments were included. Brachycephalic breeds were excluded, due to their high vasovagal tonus index (Doxey and Boswood 2004), since a high vasovagal tonus index means a higher parasympathetic component on the HRV.

2.2. Anesthesia and ventilation

Food, but not water, was withheld for 8 h prior to surgery. Dogs were premedicated with 4 µg/kg dexmedetomidine (Dexdomitor, Ecuphar, Barcelona, Spain) and 4 mg/kg pethidine (Dolantina, Kern Pharma SL, Barcelona, Spain) intramuscularly (IM). Prior to the anesthesia induction phase, an oxygen flow rate of 4 L/min was delivered via a face mask and an adult-size circle breathing circuit (Adult patient circuit, Datex Ohmeda Division Instrumentarium AB, Bromma, Sweden). A 20-gauge cannula (VasoVet, B Braun GmbH, Melsungen, Germany) was placed in the cephalic vein, and the venous access was used for administering lactated Ringer’s solution (Lactato-RingerVet, B Braun SA, Barcelona, Spain) (5 mL/kg/h) via a peristaltic pump (NIKI V4 Volumetric Infusion Pump, Everest Tecnovet, Barcelona, Spain). Anesthesia was induced with propofol (Propofol Lipuro, B Braun SA, Barcelona, Spain) intravenously (IV) administered to effect, and orotracheal intubation was performed when palpebral and swallowing reflexes disappeared. Following placement of a cuffed orotracheal tube, the flow rate was reduced to 50 mL/kg/min of 50% oxygen and 50% air. Anesthesia was maintained with isoflurane (IsoVet, B Braun SA, Barcelona, Spain) using end-tidal isoflurane (EtIso) 1.3–1.8%. The same anesthetist treated all animals. The concentration of inhalant anesthetic was adjusted to maintain the animal in a surgical plane of anesthesia: absent palpebral reflex, jaw relaxation, rostroventral rotation of the eyeballs, and stable respiration. If a light anesthesia depth (presence of palpebral reflex and jaw tension and/or presence of spontaneous ventilation) was assessed, EtIso was incremented by 0.2%. Dogs were positioned in dorsal recumbency after anesthesia induction, and volume-controlled mechanical ventilation was set initially at a respiratory rate (RR) of 12 breaths/min, tidal volume of 10 mL/kg, maximum peak pressure of 15 cmH₂O, plateau pressure of 10 cmH₂O, PEEP of 4 cmH₂O, and I:E ratio of 1:2 to maintain an end-tidal carbon dioxide pressure (EtCO2) between 4.6 and 6 kPa (35 and 45 mmHg).

2.3. Monitoring

A 22-gauge cannula (VasoVet, B Braun GmbH, Melsungen, Germany) was placed in the dorsal pedal artery and connected to a pressure system filled with heparinized saline (1 UI/mL) solution to display the systolic arterial pressure (SAP), diastolic arterial pressure (DAP), and MAP on the monitor. The zero-reference level of the pressure transducer was set at the manubrium of the sternum. Esophageal temperature (T) was obtained by a probe positioned at the thoracic portion of the esophagus and maintained within 37 °C and 38 °C by an electric blanket (Thermo Control Professional Heating Mat, Veticare BV, Vianen, Netherlands). Adhesive surface electrodes were attached to the pads from a lead II ECG monitor, and a side-stream capnograph was used for determining EtIso, end-tidal carbon dioxide (EtCO2), and RR (Datex Ohmeda Multiparameter Monitor, GE Healthcare, Helsinki, Finland).

After induction, the hair of the forehead was clipped and the skin degreased to position the disposable BIS-XP Sensor (Campagnol et al. 2007), which was connected to the bispectral monitor (BIS Pediatric XP, Aspect Medical System Inc., Massachusetts, USA). The bispectral monitor comprises the bispectral index (BIS), the signal quality index (SQI), the electromyography (EMG), and the suppression ratio (SR). The BIS was not used to assess anesthesia depth and administer isoflurane accordingly. It was used to retrospectively evaluate if anesthetic depth was similar across the four surgical stimuli evaluated.

For PTAm (PhysioDoloris; Mdoloris Medical Systems, Loos, France) (software 2.2.0.0) measurement, flattened crocodile clips were attached to the skin using a three-lead system to a lead II ECG after premedication. Yellow and red electrodes were respectively placed at the level of the olecranon of the left and right thoracic limbs, while the black electrode was attached over the patellar ligaments of the right pelvic limb (Mansour et al. 2017).

2.4. Study design

The four surgical stimuli applied were pneumoperitoneum insufflation (insufflation), introduction of trocars (trocar), and removal (manipulation and incision) of the left (left ovary) and right (right ovary) ovaries. Parameters were evaluated before each of them started and after each of them finished as detailed in the next paragraph.

The following parameters were monitored every 5 min during anesthesia, as well as before (pre-stimulus) and exactly 1 (1 min) and 2 min (2 min) after each surgical stimulus: PTAm, quality signal of PTA, BIS, SQI, EMG, SR, HR, SAP, MAP, DAP, T, EtIso, EtCO2, and RR. If SQI was under 90, the BIS values were rejected. Data of PTAm were only registered when the quality signal was good, the energy index was between 0.05 and 2.5, and HR from the PTA monitor coincided with the HR displayed on the multiparameter monitor. If one of the three considerations was not accomplished, data were rejected.

Times between surgical stimuli were adjusted to register the studied parameters before and at 1 and 2 min after stimulus (Figure 1).

Figure 1. Surgical stimuli applied and times for registration of averaged PTA (PTAm), heart rate (HR), mean arterial pressure (MAP), and bispectral index (BIS). The surgical stimuli were pneumoperitoneum insufflation, introduction of trocars, removal of the left ovary, and removal of the right ovary. The variables were registered before the stimulus (pre-stimulus) and at 1 (1 min) and 2 min (2 min) after the stimulus.

Fentanyl 0.002 mg/kg IV (Fentanest, Kern Pharma SL, Barcelona, Spain) was administered as rescue analgesia when the animal breathed spontaneously and HR and/or MAP increased more than 20% with respect to the pre-stimulus value for more than 3 min. If fentanyl was given, no further data from the animal were registered.

Times from premedication to first nociceptive stimulus and total time of anesthesia were recorded.

2.5. Statistical analysis

A sample size of 32 animals was determined (G*Power, v.3.1.9.2., Dusseldorf, Germany) considering a significance of 0.05 and a power of 0.8 to identify a decrease of 20% from a PTAm value of 50 and a standard deviation of 19 with an effect size of 0.5. The value of 50 was chosen since the manufacturer indicated that values under 50 are related to nociception.

The statistical analysis was performed using IBM Statistics SPSS v25 (IBM^® SPSS^® Statistics for Windows, version 25.0, IBM Co., Armonk, NY, USA). Normality of the data distribution was assessed using a Shapiro–Wilk test. The PTAm, HR, MAP, and BIS parameters were analyzed using a two-way repeated-measures ANOVA with a Bonferroni post hoc test. Two within-subject effects were considered in the analysis of dependent variables in each patient: surgical stimulus (insufflation, trocar, left ovary, and right ovary) and timepoint (before the surgical stimulus (pre-stimulus), 1 min after the surgical stimulus (1 min), and 2 min after the surgical stimulus (2 min)).

For the statistical analysis, PTAm-drop was defined as a ≥ 20% decrease in PTAm 1 min or 2 min after stimulus with respect to the pre-stimulus value of any surgical stimulus. A hemodynamic response was considered if HR and/or MAP increased by ≥20% with respect to the pre-stimulus value, 1 min or 2 min after any surgical stimulus. When PTAm-drop or hemodynamic response was determined, a one-way repeated-measures ANOVA with a Bonferroni test was performed for PTAm, HR, MAP, and BIS.

The ΔPTA at 1 min was calculated at each surgical stimulus using the following equation: (PTAm 1 min – PTAm pre-stimulus)/([PTAm 1 min + PTAm pre-stimulus]/2) × 100, as previously described (Mansour et al. 2017; 2020).

The relationship between PTAm pre-stimulus, PTAm 1 min, and ΔPTA and hemodynamic response was assessed by calculation of the area (AUC) under the receiver operating characteristic (ROC) curve using pooled data from surgical stimuli. Data were expressed as the mean ± standard deviation. A p-value <0.05 was considered statistically significant.

3. Results

Thirty-two bitches from different breeds undergoing laparoscopic ovariectomy aged 2.9 ± 2.3 years and weighing 17.0 ± 6.6 kg were included in this study. Time from premedication to the first nociceptive stimulus was 50 ± 19 min, and the surgery length was 37 ± 10 min. Fentanyl boluses were required once in eight animals. Fentanyl was administered after data from the four surgical stimuli were registered; therefore, no data were lost in this study.

The variations of PTAm, HR, MAP, and BIS during the four surgical stimuli are shown in Table 1. A significant decrease was observed for PTAm (p = 0.007) only 1 min (−16: 95% CI 3–29; p = 0.010) and 2 min (−14: 95% CI 0.5–27; p = 0.045) after pneumoperitoneum insufflation. The percentages of statistically significant decreases for PTAm during pneumoperitoneum insufflation were 23% at 1 min and 21% at 2 min. 

Table 1. Mean ± standard deviation for averaged parasympathetic tone activity (PTAm), heart rate (HR), mean arterial pressure (MAP), and bispectral index (BIS).

Variables	Surgical stimuli	Pre-Stimulus	1 min	2 min
PTAm (dimensionless)	Insufflation	67 ± 21	51 ± 21*	53 ± 21*
	Trocar	54 ± 22	48 ± 19	46 ± 22
	Left ovary	49 ± 22	54 ± 21	49 ± 21
	Right ovary	56 ± 21	58 ± 21	54 ± 21
HR (beats/min)
	Insufflation	86 ± 17	82 ± 17	83 ± 14
	Trocar	83 ± 14	91 ± 12*^,†	83 ± 15
	Left ovary	83 ± 16	91 ± 13*^,†	87 ± 14
	Right ovary	81 ± 15	87 ± 12*	85 ± 13
MAP (mmHg)
	Insufflation	67 ± 12	78 ± 12*	76 ± 11*
	Trocar	77 ± 11	86 ± 14*	88 ± 13*
	Left ovary	86 ± 14	95 ± 14*	96 ± 15*
	Right ovary	89 ± 13	91 ± 13	91 ± 13
BIS (dimensionless)
	Insufflation	56 ± 11	54 ± 11	55 ± 10
	Trocar	56 ± 12	56 ± 12	53 ± 12
	Left ovary	53 ± 11	55 ± 13	54 ± 12
	Right ovary	54 ± 12	53 ± 12	51 ± 12

Mean ± standard deviation for averaged parasympathetic tone activity (PTAm) (dimensionless), heart rate (HR) (beats/min), mean arterial pressure (MAP) (mmHg), and bispectral index (BIS) (dimensionless), before (pre-stimulus) and 1 and 2 min after four surgical stimuli: pneumoperitoneum insufflation (insufflation), introduction of trocars (trocar), and removal (manipulation and incision) of the left (left ovary) and right (right ovary) ovaries. * Statistically different from pre-stimulus (p < 0.05); † statistically different from 2 min (p < 0.05).

The HR increased significantly 1 min after introduction of trocars compared to the pre-stimulus value (+8 beats per minute (bpm): 95% CI 2–14 bpm; p = 0.007) and decreased 2 min after introduction of trocars compared to 1 min (−8 bpm: 95% CI 1–14 bpm; p = 0.018). After the removal of the left ovary, HR significantly increased at 1 min (+8 bpm: 95% CI 2–13 bpm; p = 0.004) compared to the pre-stimulus value but decreased at 2 min (−4 bpm: 95% CI 1–8 bpm; p = 0.015) compared to the 1 min value. A significant elevation of HR 1 min after removal of the right ovary was detected with respect to the pre-stimulus value (+6 bpm: 95% CI 1–10 bpm; p = 0.009). The percentages of statistically significant increments for HR during introduction of trocars was 10% at 1 min, during removal of left ovary was 10% at 1 min and during removal of right ovary was 7% at 1 min.

Significant differences were observed for MAP (p = 0.013), which increased significantly with respect to the pre-stimulus value 1 min (+11 mmHg: 95% CI 4–17 mmHg; p = 0.001) and 2 min (+9 mmHg: 95% CI 3–14 mmHg; p = 0.003) after pneumoperitoneum insufflation. A similar trend was observed after the introduction of trocars (1 min: +9 mmHg: 95% CI 5–14 mmHg; p = 0.001; 2 min: +11 mmHg: 95% CI 6–16 mmHg; p = 0.001) and after removal of the left ovary (1 min: +9 mmHg: 95% CI 3–13 mmHg; p = 0.001; 2 min: +10 mmHg: 95% CI 4–15 mmHg; p = 0.001). The percentages of statistically significant increments for MAP during insufflation were 16% at 1 min and 13% at 2 min, during introduction of trocars were 12% at 1 min and 14% at 2 min, and during removal of left ovary were 10% at 1 min and 12% at 2 min.

There were no changes in BIS (p = 0.088).

The variations of PTAm, HR, MAP, and BIS during PTAm-drop are shown in Table 2. Thirty-seven PTAm-drops were detected throughout the study, with a significant reduction in PTAm values after 1 min (−29: 95% CI 22–37; p = 0.001) and 2 min (−27: 95% CI 19–35; p = 0.001) with respect to the pre-stimulus value. During PTAm-drops, the MAP increased significantly after 1 min (+11 mmHg: 95% CI 7–17 mmHg; p = 0.001) and 2 min (+10 mmHg: 95% CI 6–14 mmHg; p = 0.001) compared to the pre-stimulus value. However, HR (p = 0.192) and BIS (p = 0.245) were not significantly different from their pre-stimulus values.

Table 2. Mean ± standard deviation for averaged parasympathetic tone activity (PTAm), heart rate (HR), mean arterial pressure (MAP), and bispectral index (BIS) during PTAm-drops.

PTAm-Drop
Variables	Pre-Stimulus	1 min	2 min
PTAm (dimensionless)	64 ± 22	35 ± 14*	37 ± 17*
HR (beats/min)	85 ± 14	88 ± 14	84 ± 13
MAP (mmHg)	78 ± 15	89 ± 14*	88 ± 13*
BIS (dimensionless)	56 ± 8	55 ± 8	54 ± 10

Mean ± standard deviation for averaged parasympathetic tone activity (PTAm) (dimensionless), heart rate (HR) (beats/min), mean arterial pressure (MAP) (mmHg), and bispectral index (BIS) (dimensionless), before (pre-stimulus) and 1 and 2 min after any of the surgical stimuli when PTAm decreased by ≥20% with respect to the pre-stimulus value (PTAm-drop). * Statistically different from pre-stimulus (p < 0.05).

Thirty-four hemodynamic responses were detected during surgical stimuli. The values of PTAm before stimuli (PTAm pre-stim) and PTAm 1 min after surgical stimuli (PTAm 1 min) were not different between dogs that presented a hemodynamic response and animals that did not. The ΔPTA was significantly different between dogs that presented hemodynamic response and dogs that did not after pneumoperitoneum insufflation (−40: 95% CI 12–68; p = 0.005) but not after the other surgical stimuli or during PTAm-drops (p = 0.104) (Figure 2).

Figure 2. Dynamic variations of PTA at 1 min (ΔPTA) in dogs that showed a hemodynamic response (hemodynamic response) and in dogs that did not showed a hemodynamic response (no hemodynamic response) at four surgical stimuli: pneumoperitoneum insufflation (insufflation), introduction of trocars (trocar), removal of the left ovary (left ovary), removal of the right ovary (right ovary) and when PTAm decreased ≥20% after any of the surgical stimuli regarding the pre-stimulus value (PTAm-drop). ^*Statistically different from the no hemodynamic response group (p < 0.05).

Figure 3 shows the ROC curves for PTAm pre-stimulus, PTAm 1 min, and ΔPTA. The AUC values of PTAm pre-stimulus (AUC = 0.609: 95% CI 0.499–0.718; p = 0.048) and ΔPTA (AUC = 0.323: 95% CI 0.209–0.437; p = 0.002) were statistically significant. However, only PTAm pre-stimulus presented an AUC value higher than 0.5. A PTAm pre-stimulus ≤51 correlated with a hemodynamic response with a sensitivity of 69% and a specificity of 52%.

Figure 3. Comparison ROC curves (n = 147). Performance of PTAm previous to the surgical stimuli (PTAm pre-stimulus), PTAm 1 min after the surgical stimuli (PTAm 1 min) and dynamic variation of PTAm (ΔPTA) to predict hemodynamic response. Area under the curve (AUC). ^*Statistically different (p < 0.05).

4. Discussion

The PTAm decreased significantly after insufflation only. During the PTAm-drops detected throughout the procedure, the MAP increased significantly, but the HR did not change. A threshold value of PTAm pre-stimulus ≤51 was associated to a hemodynamic response with poor sensitivity and no specificity (69% and 52%, respectively). In this study, a significant increase in the MAP and/or HR values after surgical stimuli was considered an intraoperative nociceptive response. The MAP increased after insufflation, trocar, and left ovary. In contrast, the HR did not increase after insufflation but increased after the other stimuli.

To discard whether changes in HR, MAP, or PTAm could be influenced by changes to the anesthesia depth, the BIS (March and Muir 2005; Morgaz et al. 2009) was monitored during the procedure. The BIS was only used to retrospectively evaluate if anesthetic depth was similar across the four surgical evaluated stimuli. Anesthetic depth was not adjusted according to BIS values. The BIS data were analyzed, and, since no changes were observed throughout the procedure, the changes in HR, MAP, or PTAm were considered to be due to nociceptive stimuli. Values of BIS ˂65 have been used previously as an indication of surgical anesthesia. We acknowledge some limitations to the use of BIS in dogs due to its low specificity and sensitivity (Bleijenberg et al. 2011).

Sudden increments in HR and/or MAP intraoperatively are clinically interpreted as a nociceptive response (Lardone et al. 2017). Some studies considered a 20% increment in the HR and/or the MAP as a nociceptive hemodynamic response to evaluate ANI (Jeanne et al. 2014; Boselli et al. 2016) and PTA (Mansour et al. 2017; Aguado et al. 2020; Ruíz-López et al. 2022) monitor performance. A validated monitor to assess nociception would be useful for the anesthetist, since hemodynamic variables can be influenced by the drugs used, the animal condition, and the anesthesia depth (Mansour et al. 2017; Ruíz-López et al. 2020), as occurs in humans (Gruenewald and Ilies 2013). Jeanne et al. (2012) found the ANI to be more sensitive to intraoperative stimuli than HR or BP during laparoscopic surgery in humans.

In our study MAP changed significantly after the first three stimuli, but not during the last one. It could be explained because there was no enough time for MAP to decrease to the baseline value after the stimulus of the left ovary. Significant changes on MAP were more frequent and were maintained at 1 and 2 min compared to HR, what could mean than MAP is more sensitive to nociceptive stimuli than HR. Determining this was no the aim of our study.

The first published study to describe the performance of the PTA index to determine nociception in dogs (Mansour et al. 2017) showed that the PTA index changed earlier than hemodynamic variables. Same authors also found that PTA index changed before hemodynamic variables when morphine or morphine plus acepromazine was given, but not when the animals were premedicated with morphine plus medetomidine (Mansour et al. 2020). Under our study conditions, considering the ≥20% intraoperative increments in HR and/or MAP as a nociceptive hemodynamic response, the PTAm did not change significantly between dogs that presented a hemodynamic response and dogs that did not.

In our study, the PTAm decreased significantly only 1 and 2 min after insufflation, accompanied by a significant increase in the MAP. It could be thought that there was not enough time for PTAm to return to baseline values between stimuli and posterior nociceptive stimuli were neglected. In a recent study, PTAm was able to detect medium-intensity nociceptive stimuli that did not elicit cardiovascular changes in experimentally anaesthetized dogs. However nociceptive stimuli of higher intensities produced a faster cardiovascular response compared to the PTAm response (Aguado et al. 2020). The PTAm decreased at insufflation and did not vary significantly at the next pre-defined surgical stimuli, suggesting that it is more interesting to consider a fixed threshold value for PTAm than a percentage reduction, as Kommula et al. (2019) did.

The PTAm value changes quickly. So, it would be more appropriate to compare it with data provided by direct BP measurement rather than data obtained from oscillometric arterial BP monitoring. This may account for the differences found with respect to the study by Mansour et al. (2017), in which HR and SAP were determined 5 min after nociceptive stimulus, or by Mansour et al. (2020), where mean arterial pressure was collected noninvasively every 5 min, which might have increased the ability of the PTA index to anticipate a hemodynamic response, showing good performance.

In our study, changes in PTAm and ΔPTA were accompanied by changes in MAP after the insufflation stimulus. It could mean that baseline values were not recovered. Similarly, PTAm-drops were accompanied by an increase in MAP. However, the HR did not show the same behavior among the surgical stimuli. In the Mansour et al. (2017) study, the HR increased 1 min after the incision, as well as 1 and 5 min after the PTA value decreased by at least 20% (9%, 10%, and 18% respectively). The 20% cutoff considered in our study could justify the different results obtained with respect to Mansour et al. (2017). In the same way, Mansour et al. (2020) did not find changes greater than 20% in the HR at any of the timepoints evaluated in their study, regardless of the premedication (morphine, morphine plus medetomidine, or morphine plus acepromazine). However, they found significant changes in HR lower than 20% in all groups. In the study of Aguado et al. (2020), the high-intensity electrical stimulation increased MAP in all dogs, but HR in only half on them.

Specificity and sensitivity must be considered when evaluating the PTAm assessment. Boselli et al. (2015) reported that an ANI ≤55 predicted an increase in HR and/or SAP by >20% within the next 5 min with a sensitivity and specificity of 88% and 83%, respectively. Recently, it was found that the correlations of BP and HR with ANI are good in neurosurgical patients undergoing elective supratentorial craniotomy; however, the r-values did not support the results (r = 0.258, p < 0.0001 for BP and ANI and r = 0.280, p < 0.0001 for HR and ANI) (Kommula et al. 2019).

The PTAm is influenced by instant variations; thus, Boselli et al. (2016) and Mansour et al. (2017) showed that dynamic variations of ΔANI or ΔPTA may provide better reliability. In dogs, Mansour et al. (2017) showed that the ROC curve of the PTA index 1 min after predefined times determined a threshold value <46 for the prediction of hemodynamic events with 60% sensitivity and 63% specificity. However, the ΔPTA seems to be more effective since a ΔPTA < −18% was shown to correctly predict the hemodynamic response within 5 min, with a sensitivity and specificity of 76% and 72%, respectively. Recently, Mansour et al. (2020) found a ΔPTA threshold value of −20% if the premedication was morphine (sensitivity 71.7%, specificity 71.4%), a threshold value of −7.3% if the premedication was morphine plus acepromazine (sensitivity 84.8%, specificity 57.5%), and a threshold value of −6.1% if the premedication was morphine plus medetomidine (sensitivity 81.6%, specificity 65.1%).

In our study, the ΔPTA showed an AUC of 0.323. A threshold value of PTAm pre-stimulus ≤51 showed a poor sensitivity (69%) with no specificity (52%) in predicting a hemodynamic response. The variability in ΔPTA between studies could be explained by the different breeds (Doxey and Boswood 2004), anesthetic protocols used (Mansour et al. 2020), and nociceptive stimuli applied due to the different surgeries performed, as well as because of the differences in hemodynamic data registration times and the methods used to do so. However, the ΔPTA seems more suitable for retrospective studies since it needs to be calculated on the basis of the PTA values provided by the monitor. This might be considered a limitation for clinical purposes since retrospective information does not help the anesthetist to maintain an adequate intraoperative nociception–antinociception balance.

Even if the PTA software was developed on the basis of the HRV of small dogs (<10 kg) and big dogs (>10 kg), the variability between breeds is still huge. Therefore, this might complicate getting a precise performance of the monitor. There was no breed restriction, except for brachycephalic dogs, in the present study. Nonetheless, autonomic nervous system variability may be diverse across breeds (Doxey and Boswood 2004) and could be considered a limitation; however, the purpose of the study was to evaluate the PTA monitor performance under clinical conditions and not only for a specific breed. More studies in animals are needed before determining the utility of the PTA monitor.

Isoflurane was used for anesthesia maintenance, which could have influenced the study results considering that the ANI monitor seemed to perform better in detecting moderate nociceptive stimuli in humans under propofol/fentanyl anesthesia than under halogenated agents (Boselli et al. 2013). It is possible that inhaled anesthetics are associated with higher sympathetic activity (Boselli and Jeanne 2014). Alpha-2 agonists may influence the sympatho-vagal balance (Khan et al. 1999); however, their use is routine in healthy clinical cases to decrease stress prior to surgery, and the aim was to evaluate the monitor in a clinical context. Nevertheless, a standardized anesthetic protocol was used in all dogs, and a low dose of dexmedetomidine was used before the surgery.

The type of the nociceptive stimulus should be considered since differences between stimuli might influence results (Valverde et al. 2003). In the present study, nociceptive stimuli were always similar since only laparoscopic ovariectomies were included, which could partially explain the differences with previous results reported in dogs (Mansour et al. 2017) or in humans (Boselli et al. 2016), including different types of surgeries. Standardized electrical nociceptive stimuli have been used in human research for the assessment of the ANI during anesthesia (Luginbühl et al. 2010; Gruenewald et al. 2013), and they were recently used in dogs (Gavet et al. 2022). However, most ANI performance results are from clinical studies that used different stimuli (Boselli et al. 2016; Ruíz-López et al. 2022). In the same way, most of the veterinary studies were performed in clinical circumstances (Mansour et al. 2017; 2020; Dmitrović et al. 2021; Mansour et al. 2021; Ruíz-López et al. 2022). No differences in intraoperative nociception level were found between laparoscopic and open hepatic resection in humans (Hashimoto et al. 2015). It could be suggested that the laparoscopic surgery performed in our study produced a repetitive surgical intraoperative stimulus. The surgery was performed by the same surgeon that standardized the four surgical stimuli applied during the procedure. Pneumoperitoneum insufflation of 15 mmHg showed no changes in heart rate in the Trendelenburg position and only a slight increment in the mean aortic pressure (Park et al. 2015). During our study, pneumoperitoneum pressure was set to 10 mmHg for the procedure. Pneumoperitoneum increases sympathetic nervous system in humans (Barczyński and Herman 2002). It could be explained by three theories: increase in intra-abdominal pressure causes decrease in venous return and cardiac output, hypercarbia stimulates sympathetic nervous system, or distension of the abdominal muscles might produce nociception (Sato et al. 2000). The latest authors found the last option the most suitable explanation for the cardiac sympathetic activation.

There are several possible limitations to the present study.

Alpha-2 agonists could decrease the performance of the PTA monitor (Mansour et al. 2020), dexmedetomidine was used as premedication because the aim of this study was to evaluate the monitor in clinical conditions.

The point of time at which the PTAm changed before the HR and/or MAP changed was not observed. Equilibration times between stimuli could have facilitated the return of PTAm to its baseline values, and it might allow to observe new changes during the following stimuli. However, this was a clinical study and the aim was to evaluate the monitor under clinical conditions.

A different assessment at intervals of time instead of looking at screenshots at 1 and 2 min after surgical stimuli could have led to different results. However, considering that PTAm is the result of the last 176 s, the data registered at 1 and 2 min after surgical stimuli were the result of changes that occurred over a period of time. Aguado et al. (2020) found that the PTAm response to a medium-intensity stimulus occurs at 66 ± 11 s and the response to a high-intensity stimulus occurs at 60 ± 22 s. Therefore, it would be expected to register changes in PTAm at 1 and 2 min after surgical stimuli. Aguado et al. (2020) considered values second by second what allowed to get all the information, but what would not be practical for clinical applicability.

The PTAi is the result of the last 54 s, and we cannot dismiss that it could have detected earlier changes. So, it can be another limitation of this study. Nevertheless, this would lead to confusion in the consideration of shorter drops or punctual drops observed with PTAi. The manufacturer describes the PTAm as the anesthesiologist value, since PTAi can vary due to the surgeon’s punctual manipulations.

The nature of the surgical stimuli could be considered as a limitation, while its repeatability should be consider a strength.

5. Conclusions

In the present study, the PTAm and ΔPTA only decreased during insufflation that was the first stimulus. Throughout the procedure, PTAm drops were related to an increase in MAP but not in HR. A PTAm pre-stimulus ≤51 is associated to a hemodynamic response in anaesthetized dogs undergoing laparoscopic ovariectomy with a poor sensitivity and no specificity. Further research should be performed to assess the usefulness of the PTA monitor in dogs of different breeds and under different clinical conditions.

Ethical approval

The study was approved by the Ethical Committee of Animal Welfare of the Veterinary Teaching Hospital, University of Córdoba (CEBAHCV32/2016). All procedures were conducted in compliance with the ethical principles of good practice in animal experimentation and with previous informed consent from the owners.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research received no external funding.