Maternal factors regulating preterm infants ’ responses to pain and stress while in maternal kangaroo care 1

posição canguru”, presented to Escola de Enfermagem de Ribeirão Preto, Universidade de São Paulo, WHO Collaborating Centre for Nursing Research Development, Brazil. Supported by FAPESP (2007/05051-1) and CNPq (200113/2008-9). 2 PhD, Adjunct Professor, Faculdade de Enfermagem, Universidade Federal de Goiás, Brasil. 3 PhD, Assistant Professor, University of British Columbia, School of Nursing, Canada. 4 Doctoral Student, Escola de Enfermagem de Ribeirão Preto, Universidade de São Paulo, WHO Collaborating Centre for Nursing Research Development, Brazil. 5 PhD, Adjunct Professor, Departamento de Enfermagem, Universidade Federal de Pernambuco, Brazil. 6 PhD, Associate Professor, Escola de Enfermagem de Ribeirão Preto, Universidade de São Paulo, WHO Collaborating Centre for Nursing Research Development, Brazil. 7 PhD, Full Professor, Escola de Enfermagem de Ribeirão Preto, Universidade de São Paulo, WHO Collaborating Centre for Nursing Research Development, Brazil.


Introduction
We no longer believe that newborns are incapable of feeling pain.Scientific advancements in recent years have contributed to the understanding of anatomical and physiological mechanisms from the time of intrauterine life, as well as contributing to our understanding of the effects of acute pain in preterm and sick newborns.We now acknowledge the existence of continuous pain, in addition to acute and postoperative pain.Scientific advancements also include the development of specific indicators to evaluate pain and determine the effectiveness of neonatal pain treatment (1) .Such advancements, however, have not substantially reduced suffering or costs related to pain, which is considered to be a worldwide public health problem that should be treated as a priority for healthcare services (2) .
A growing number of children, who are now able to survive with the treatment provided in neonatal units of high technological density, are exposed to a diversity of painful and stressful procedures.An average of 12 painful procedures are performed a day, while the heel puncture is the most frequent (3) .The exposure of neonates to pain and stress results in behavioral changes (e.g.facial actions, crying) (4) and immediate biophysiological changes (e.g.increased Heart Rate (HR) and cortisol levels) (5) .Repeated procedures may result in the frequent activation of stress system, making newborns to spend energy without time to recover (6) .
Pain treatment is an essential human right and evidence for its treatment is already available.The kangaroo position stands out among non-pharmacological interventions effective in reducing neonatal pain.It is recommended to alleviate pain from a single heel puncture in full-term and stable preterm newborns (7) .
There are various components interacting in the kangaroo position, such as containment, heart beat sound, rhythmic breathing movements, maternal warmth, the prone position (8) .Therefore, the mother seems to play an important role in the analgesic mechanism of the kangaroo position, though most mothers addressed in studies evaluating the effectiveness of this intervention were instructed not to touch or talk to the infant (7) .
It is believed there are hidden regulating maternal factors that allow mothers to regulate the infant's specific and independent systems, such as behavior, autonomic state, thermoregulation, emotion, and stress (9) .Hence, for the mother to comfort the child in a stressful situation, she should first self-regulate, establishing effective interactive communication, acting as an important external factor to regulate the newborn's biobehavior and the development of self-regulating skills, in addition to the establishment of effective bonds throughout the infant's first year of life (10) .
Considering the proximity of the mother and child in the kangaroo position, we examine what factors promote or hinder the mother's ability to self-regulate and, consequently, to regulate the newborn during the kangaroo position.We highlight the importance of providing integral and humanized care to newborns (11) with the active participation of the mother and family to prevent acute neonatal pain.This study's objective was to investigate the association between maternal factors (behavior, depression and/or anxiety, and stress) and the response of newborns to pain and stress when undergoing heel puncture for the neonatal screening test while held in the kangaroo position.We estimated a sample of 49 participants based on alpha=0.05, the number of predictors=7, effect size=0.35 and β=0.80.Newborns with a gestational age ≤ 36 weeks and six days, Apgar≥7 at the 5 th minute, scheduled for a neonatal screening test, and clinically stable (HR and oxygen saturation within normal levels before data collection) were included in the study.Newborns were not included in the study when they were under assisted ventilatory care, with intraventricular hemorrhaging of grade III or IV or subsequent leukomalacia, were diagnosed with congenital disease of the nervous system, malformation or neurological and renal damage, using corticosteroids or drugs that interfere in the response to nociception, or those with a twin sibling included in the study.With regard to mothers, those with the cognitive ability to understand the instruments were included in the study, while those with a current diagnosis of psychiatric disease other than anxiety and depression or those using corticosteroids or drugs that interfere in nociception or with renal disease were excluded.

Method
Of the 217 recruited mother-child dyads, 148 (68.2%) did not participate in the study due to various reasons, such as being transferred to another unit and difficult access to mothers.Data were collected from 69 mother-child dyads, though 27 (39.1%)were excluded from the analysis either due to the mother's unavailability to attend the interview, insufficient volume of saliva, or failure during filming, among other reasons.Therefore, 42 mother-child dyads composed the final sample and a post-hoc β = 0.77 was obtained.
The studied neonatal variables included facial actions, sleep and wake states, crying, HR and salivary cortisol levels while the maternal variables included behavior, salivary cortisol levels, and depression and/or anxiety.The period in which each variable was measured is presented in Figure 1.
The mothers' and infants' behavioral data were filmed by two digital SONY Handycam model DVD92 camcorders.Two trained individuals coded, second-bysecond, the neonatal behavioral variables (facial actions, sleep and wake states) in addition to the maternal variable (behavior), using digital images through the Magix Movie Editor 15 Pro Plus.Three facial actions (eyebrow bulge, squeezed eyes, and naso-labial furrow) of newborns and five sleep-wake states (deep sleep, active sleep, quiet wake, active wake, and crying) (13) were coded through the Neonatal Facial Coding System (NFCS) (12) .The 23 categories of mothers' behavior were coded through the Maternal Mood and Behavior during her Infant Pain Coding System (MMBIPCS) and later grouped into 'typical' and 'typically depressed' (14) .
Inter-rater reliability was computed through the Kappa coefficient with the Generalized Sequential Querier (GSEQ), version 5.0, and the following was obtained: k=90%, k=84% and k=84%, respectively.A psychologist assessed the mothers' anxiety and depression levels using the Beck Depression Inventory (BDI) and the Beck Anxiety Inventory (BAI) (15) .The general introductory part of the Structured Clinical Interview DSM-IV-SCID-CV (16) was used to describe the psychiatric history of mothers.Based on the scores obtained in the BDI and BAI and the duration of anxiety in mothers during the procedure, newborns were divided into two groups: 'exposed' (BDI≥18, BAI≥20 and MMBIPCS>30s) (n=17) and 'non-exposed' (BDI<18, BAI<20 and MMBIPCS≤30s) (n=25) to maternal depression and/or anxiety.Nursing assistants, nurses and the psychologist, trained prior to data collection, collaborated in the study.
After the mothers and/or legal guardians signed free and informed consent forms, the mothers received identified Salivettes ® tubes with written instructions to collect two samples of saliva, one at the night prior to the exam, and the other on the morning of the exam day.Another saliva sample was collected from both the newborn and the mother prior to the neonatal screening test while the infant was in a supine position, either in the incubator or crib, and the mother was seated close to her child.For the next 10 minutes, the newborn's HR was monitored and his/her face was filmed while in the incubator or crib (baseline phase).Afterward, the newborn was positioned in maternal kangaroo care for 15 minutes (17) .A nurse collected blood for the neonatal screening test in accordance with the unit's protocol using a Unistik ® lancet (procedure phase) while the mother's behavior and interaction with the newborn were also filmed.After the exam, the newborn was kept in the kangaroo position for another ten minutes (recovery phase).The video recording and cardiac monitoring ceased and the newborn was transferred to the incubator or crib where s/he remained without any manipulation for another ten minutes when an additional sample of saliva after the procedure was collected from both the mother and child.A psychologist applied the BDI, BAI and the SCID-CV within one to seven days after the procedure (Figure 1).

Results
The    infants after the procedure and during recovery (18) .This study's results add to such evidence beyond the prenatal period to the population of preterm newborns.

Discussion
There was also a positive association between the mothers' salivary cortisol levels and the newborns' pain scores (NFCS).This is the first time this relationship is reported in the literature.Such a relationship, however, requires further investigation through statistical analysis that would allow adjusting for potential confounding variables.A more refined analysis of this variable was not possible in this study due to restrictions imposed by the statistical assumption of homoscedasticity.
The mothers' levels of salivary cortisol from the night before, along with the newborns' GA, were predictors of neonatal HR.The relationship between the autonomic nervous system and the hypothalamicpituitary-adrenal (HPA) axis is very complex.The inverse relationship between the cortisol level and HR variability may be explained by a system of negative feedback in which appropriate HPA axis regulation is influenced, in part, by the modulation of the autonomic nervous system (19) .Since the kangaroo position involves very close contact between mother and child, the inverse relationship between the newborns' HR and the mothers' salivary cortisol levels suggests there is mediation of the kangaroo position in relation to the newborns' autonomic nervous system and HPA axis.The close contact also encourages the self-regulation of maternal stress, which contributes to the effectiveness of the kangaroo position in alleviating neonatal pain and stress.Even though these assumptions cannot be appropriately confirmed due to the lack of a control group not using the kangaroo position, we observed that the mothers' stress contributed to their newborns' responses to pain and stress caused by the neonatal screening test.
In contrast with the mothers' salivary cortisol levels, the mothers' behavior and depression and/or anxiety did not affect the newborns' responses to pain and stress.
Even though a positive association was found between the maternal behavior and the responses of infants to pain, such an association was weak in most cases.Finally, maternal kangaroo care may have softened the negative effect of depression and anxiety in the response and regulation of newborns.The prevalence of post delivery depression in mothers using the mother-child kangaroo method during the hospitalization of infants in a neonatal unit decreased, from 37.3% at admission, to 16.9% at discharge (p<0.001) (25).Nevertheless, a causal relationship between the kangaroo position and a lack of negative effects in the exposure of newborns to their mothers' depression and/or anxiety was not established in this study due to the non-existence of a control group (lacking kangaroo position).

Conclusion
This descriptive exploratory study was conducted in the Neonatal Intermediary Care Unit (NICU) of a university hospital in the interior of São Paulo, Brazil from July 2008 to January 2010.The study was approved by the hospital's Ethics Research Committee (protocol No. 7999/2007).

Figure 1 -
Figure 1 -Organization chart of the data collection and measured variables Colored envelops marked the beginning of each phase of data collection during filming.The infants' and mothers' medical records were searched so that their clinical and demographic variables were collected.The number of painful procedures performed in the 24 hours prior to the screening test was obtained through records in the infants' medical files and exams requests.Data were stored in the Statistical Package for the Social Sciences version 17.0 through double entry.The duration (seconds) and percentage of time [(duration of behavior/duration of phase) x 100] of neonatal and maternal behaviors were computed through GSEQ and the mean duration calculated for the NFCS, neonatal HR and maternal and neonatal salivary cortisol levels in each phase of the data collection (baseline, procedure, recovery).The relationship between the newborns' response variables (NFCS score, percentage of time of crying, HR and salivary cortisol levels) and the mothers' explanatory variables (percentage of time of behavior and salivary cortisol levels) were initially examined for each phase through bivariate analysis, with This paper presented an investigation of maternal factors related to newborns' responses to pain and stress when exposed to the neonatal screening test while in the kangaroo position.Various significant associations were found, suggesting the existence of a co-regulation between mother and child in the context of neonatal pain and stress while the infant is held in the kangaroo position.Maternal salivary cortisol levels, which are an indicator of stress, were associated with various neonatal variables (NFCS score, duration of crying, HR, and salivary cortisol levels), though no association was found between the mothers' salivary cortisol levels and their own behavior (typical and typically depressed) or depression and/or anxiety.The relationship between the mothers' salivary cortisol levels and neonatal variables is still seldom studied in the context of neonatal pain.No other study has examined this relationship when the infant experiences such a painful event while held in the kangaroo position.The mothers' levels of salivary cortisol before the procedure explained the variance in the newborns' levels of salivary cortisol after the procedure.Such findings are similar to those of another study addressing the relationship between plasma levels of cortisol and maternal psychosocial stress with the behavioral and physiological responses of full-term newborns to calcaneal heel puncture.The study reports that high levels of plasma cortisol observed in mothers during pregnancy explained high levels of salivary cortisol observed in their Additionally, only two studies used a validated scale to measure neonatal pain, while the maternal behavior was evaluated differently in the different the studies.Both the infants' and the mothers' behavior was coded in intervals of time (e.g. at each five second point) and not second-by-second as we did in this study.Depression and/or anxiety of mothers has been seldom investigated in the context of neonatal pain.Among the factors that may have contributed to the fact that no positive association was found between the mothers' depression and/or anxiety and the responses of their newborns to pain is the reduced number of mothers with depression and/or anxiety in the sample, the exclusive use of self-reporting instruments, and evaluation in the post delivery period.Depression and/ or anxiety experienced during pregnancy may expose newborns to a greater risk of presenting an altered response to pain and stress.Future studies should assess mothers both in the prenatal and post-natal periods, as well as use both self-reporting scales and observation.
Maternal stress, measured through the mothers' levels of salivary cortisol collected during the post delivery period, influenced the response of newborns to pain and stress when subject to the neonatal screening test in the kangaroo position.The newborns' HR variations were explained by the mothers' salivary cortisol levels taken the night before and newborns' age; with a greater contribution of the mothers' cortisol levels to the model.The mothers' cortisol levels observed before the procedure explained the newborns' cortisol levels in response to the painful exam.This study's limitations are the relatively small number of mothers in the group with depression and/or anxiety in addition to the impossibility to establish a causal relationship among variables due to the study's cross-sectional nature.We expect this study to contribute to the acknowledgement that mothers play an important role in the regulation of stress and acute pain for preterm infants while being held in the kangaroo position.This study's findings can also support the development of protocols to implement the kangaroo position for the management of neonatal acute pain, favoring the interaction between mother and newborn and the establishment of more integrated and humanized neonatal care with the active participation of mothers.

Table 1 -
Clinical and demographic characteristics of

Table 2 -
Person's linear correlation matrix (r) and values of p probabilities among the neonatal response variables during the procedure and neonatal and maternal confounding variables.GA=gestational Age; Pain=number of pain events in the 24 hrs prior to the procedure; DS=average percentage of time in baseline deep sleep; AS= average time in baseline active sleep; QW= average percentage of time in baseline quiet wake state; AW= average percentage of time in baseline active wake state; CR= average percentage of time in baseline crying; MK=maternal experience with kangaroo method before the procedure; NFCS= score obtained on the pain scale; Cry= average percentage of time crying; HR=average neonatal HR.Cortisol=levels of salivary cortisol before the puncture.*0.01 < p <0.05; † p<0.01; n.s.= not statistically significant.
Multiple regression analysis was not performed for the variables 'facial actions' and 'crying' because these did not meet the homoscedasticity assumption.Most mothers (61.9%) did not present depression or only mild depression (14.9%).Ten (23.8%) mothers presented moderate depression and none presented severe depression.Similarly, most mothers did not present anxiety (52.4%) or it was mild (26.2%).Nine (21.4%) mothers presented moderate (19.0%) or severe (2.4%)

Table 3 -
Linear regression model for neonatal HR during procedure

Table 4 -
Linear regression model for the newborns' levels of salivary cortisol after the procedure β=standardized weight; t=statistics t; * p<0.05The explanatory model of the neonatal HR during the procedure is presented in Table3.The mothers' levels of salivary cortisol and the newborns' age contributed significantly to the model; the mothers' levels of salivary