Animal-assisted Therapy in Pediatric Surgery: Post-operative Benefits

Interest in animal-assisted therapy has been fueled by studies supporting the many health benefits in the adult and pediatric age. The purpose of this study was to study the impact of an AAT program on cardiovascular, neurological and endocrinological responses to stress and pain in the immediate post-operative period following pediatric surgery.

Forty immunocompetent children undergoing surgical procedures (including orchidopexy, inguinal or umbilical hernia repair, circumcision, varicocele treatment) were sequentially enrolled.

This was a randomised open-label, controlled, pilot study. The different arms consisted of an experimental group with a AAT session after a surgical procedure and a control group with standard care after surgery.

The study variables were determined in each patient independently of the assigned group, before and after the experimental intervention, by a researcher unblinded to the patient's group.

The outcomes of the study were to define the impact of AAT on neurological, cardiovascular and endocrinological signs, in response to stress and pain in children undergoing surgical procedures.

As outcome measures the investigators considered:

• for neurological impact, the difference in the prevalence of beta (>14 Hz) electroencephalogram (EEG) activity between intervention and control group (20-23);
• for autonomic impact, the difference in blood pressure (BP) between intervention and control group;
• for cardiac impact, the difference in heart rate (HR) between intervention and control group;
• for respiratory impact, the difference in oxygen saturation (SpO2) between intervention and control group;
• for cerebral oxygenation, the difference in prefrontal oxygenation (HbO2) with near-infrared spectroscopy (NIRS) between intervention and control group;
• for endocrinological impact, the difference in salivary cortisol levels, between intervention and control group. The Wong-Baker Faces pain scale (FPS) was used to measure the child's self-reported pain.

At admission, auxological examination of the children included measurement of height, weight and body mass index. Height measurement was performed with patients in an upright position, without shoes, with their heels together, arms extended down the sides of the body and head positioned parallel to the floor. Weight was measured with the children barefooted and wearing light clothes, standing upright in the centre of the scale platform with their arms extended down the sides of the body. BMI was calculated as body weight in kilograms divided by body height squared in meters.

EEG activity, cerebral prefrontal oxygenation, heart rate, blood pressure, oxygen saturation, salivary cortisol levels, faces pain scale were considered as indicators of neurological, cardiovascular and endocrinological response to stress and pain.

Data collection was performed in the following phases:

• for all parameters, post-operatively baseline (T1), two hours after surgery at re-admission to the Unit. Pre dog-intervention in the AAT-group;
• for all parameters, in the twenty minutes following T1 (T2). During the dog-intervention in the AAT-group. For the STAND-group, the child was asked to sit quietly;
• only for salivary cortisol levels, between 11 pm and midnight (T3, the time when cortisol is normally at its lowest)

Vitals signs

HR, BP, SpO2, HbO2 were monitored (Dräger Primus ®) and recorded as follows:

• T1 (every 5 minutes, for 10 minutes. Mean values were used for the statistical analysis);
• T2 (every 5 minutes, for 20 minutes. Mean values were used for the statistical analysis).

Endocrinological parameters Salivary cortisol levels were measured at T1, after T2 and T3. The saliva samples were collected using a standardized salivette and frozen at -20° until analysis in the laboratory. After thawing, salivary fluids were centrifugated to precipitate mucins and cortisol was assayed in the supernatant with a solid-phase radioimmunoassay, wherein 125-I labeled cortisol competes for a fixed time with cortisol in the biological sample (The Coat-A-Count Cortisol, Siemens, Los Angeles, CA).

Pain response The faces pain scale was used to measure the child's self-reported pain at T1 and at the end of T2. This scale consisted of 6 cartoon faces with varying expressions ranging from very happy to very sad. The child rated the pain intensity on a scale, with point 0 being no pain and point 10 being the worst pain.

Analgesic treatment in the first 12 hours post-intervention was recorded.

Electroencephalogram Regarding the AAT-group patients, an EEG recording was obtained when the child was awake, 2 hours after surgery and before and during AAT intervention. In the STAND-group patients, and EEG recording was made 2 hours after surgery.

Any change in the normal physiological structure of the EEG, correlating the effect of the anaesthesia, post-operative stress of AAT intervention was recorded and monitored.

In both cases, the protocol included an EEG recording tasting about 20 minutes, while awake, with open and closed eyes.

Parental permission was obtained through a written and oral informed consent. Written assent by the patient was also obtained in children eight years of age and older before enrolment.

Statistical analysis A desired sample size of 20 patients per group for a total of 40 subjects was determined based on 80% power alpha=0.10 (pilot study), using a 2-tailed test, to detect an effect size of 0.8 for the primary endpoint measured on a continuous scale, and an absolute difference in prevalence of the primary endpoint of 31 to 34% for the expected prevalence in the control group of 5 and 10% respectively.

Continuous variables were described as the mean and standard deviation (SD) or median and quartiles and categorical variables as counts and percentages. The prevalence of EEG beta activity was compared with the Fisher exact test. To test the effects of AAT on vital signs and on endocrinological parameters over time in the two groups, regression models for repeated measures were used, including a main effect for group and time, as well as their interaction. Distribution of pain scores at the end of the session in the two groups was compared with the Fisher exact test. A 2-sided p-value<0.05 was considered statistically significant. All statistical analyses were performed using Stata 13.1 (StataCorp, College Station, Texas, USA).