Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Obstet Gynecol Neonatal Nurs. Author manuscript; available in PMC 2014 January 1.
Published in final edited form as:
PMCID: PMC3565562

Endotracheal Suctioning in Preterm Infants Using Four-Handed versus Routine Care

Sharon Cone, PhD, RNC, NNP, Rita H. Pickler, PhD, RN, PNP-BC, FAAN, Mary Jo Grap, PhD, RN, FAAN, Jacqueline McGrath, PhD, RN, FNAP, FAAN, and Paul M. Wiley, DDS, MS, Med



To evaluate the effect of four-handed care on preterm infants’ physiologic and behavioral responses to and recovery from endotracheal suctioning versus routine endotracheal (ETT) suctioning.


Randomized crossover design with infants as their own controls.


Single-family-room newborn intensive care unit in an academic health center.


Ten intubated infants on conventional ventilation with inline suctioning who were fewer than 37 weeks gestation at birth, and less than one week of age.


Each infant was observed twice on a single day. One observation involved routine ETT suctioning and one involved four-handed care. Physiologic and behavioral response data were collected.


No differences were noted when comparing baseline heart rate (HR) or oxygen saturation (SpO2) data to those obtained during and after suctioning while in the routine care condition. In the four-handed care condition, mean SpO2 increased from preobservation 95.49 to during observation saturation 97.75 (p = .001). Salivary cortisol levels did not differ between groups at baseline or postsuctioning. No significant difference in behavior state was observed between the two conditions. More stress and defense behaviors occurred postsuctioning when infants received routine care as opposed to four-handed care (p = .001) and more self-regulatory behaviors were exhibited by infants during (p = .019) and after suctioning (p = .016) when receiving four-handed care. No statistical difference was found in the number of monitor call-backs postsuctioning.


Four-handed care during suctioning was associated with a decrease in stress and defense behaviors and an increase in self-regulatory behaviors.

Keywords: Endotracheal suctioning, preterm infants, developmental care

Neonatal intensive care units (NICUs) have experienced profound advances in technology and treatment modalities over the last two decades. With the advent of surfactant therapy, high-frequency ventilation, and advances in the use of central lines for long-term hyperalimentation, nearly 80% of the infants born at 26 weeks gestation are surviving despite prolonged hospitalizations (Mathews & MacDorman, 2010; Tatad & Frayer, 2003). Preterm infants in the NICU are prone to a high degree of stress from life-sustaining interventions and routine care practices (Altimier, 2007). Stress-producing behaviors have been studied in relation to repositioning, suctioning practices, heel-sticks, and even routine bathing (Evans, Vogelpohl, Bourguignon, & Morcott, 1997; Hill, Engle, Jorgensen, Kralik, & Whitman, 2005; Peters, 1998). Preterm infants are often unable to buffer negative stimuli due to physiologic immaturity (Altimier). Altering NICU caregiving practices to reduce iatrogenic complications remains a significant challenge for infants cared for in the NICU.

In recent years, much attention has been focused on providing developmentally supportive care for preterm infants who are hospitalized in the NICU as a way to modulate the stress experienced by the vulnerable preterm infant. Developmental care is predicated on individualized care planning, family involvement, communication of an infant’s behavioral competencies, vulnerabilities, thresholds, and calm, nurturing environment (Lawhorn & Als, 2010). A number of studies have been conducted testing a variety of developmentally supportive interventions including kangaroo care or skin-to-skin holding, touch, massage, and swaddling. Many studies are conducted based on the belief that providing developmentally supportive care during procedures may serve to protect the infant from the stress of the procedure and help the infant to remain behaviorally organized without detrimental consequence. Providing supportive, nurturing care during procedures such as bathing, heel-sticks, circumcision, and suctioning has been addressed to in the literature (Axelin, Salantera, & Lehtonen, 2006; Ferm, Graves, & L’Huiller, 2002; Huang, Tung, Kuo, & Chang, 2004; Stevens et al., 1999; Taquino & Blackburn, 1994; Ward-Larson, Horn, & Gosnell, 2004); these interventions have been shown to be helpful in decreasing crying and arousal, diminishing pain and stress, and promoting sleep.

Four-handed care supports and assists the clinician and the patient during procedures and care-giving activities. It focuses on the dynamic interplay and vulnerabilities of patients and caregivers in the environment and restructures critical care giving activities to improve outcomes for patients and staff. Despite the fact that two-person suctioning has been recommended by the National Association of Neonatal Nurses (Altimier, Brown, & Tedschi, 2006), it is not a consistent practice in all NICUs. Previous studies have been done to assess the degree by which nonpharmacological measures such as facilitated tucking, containment, and swaddling affect pain scores in infants (Corff, Seidemann, Venkataraman, Lutes, & Yates, 1995; Hill et al., 2005; Ward-Larson et al., 2004), but researchers have not resorted to four-handed care to support the infant behaviorally while also assisting the care provider during a stressful procedure such as endotracheal suctioning.

Preterm and term infants admitted to the NICU may require mechanical ventilation. Those infants who are mechanically ventilated will also require endotracheal intubation and repeated suctioning to remove excessive secretions and to reduce the potential for an obstructed airway. Endotracheal tube (ETT) suctioning, although necessary, is a noxious stimulus, and biobehavioral changes can be observed in the infant during this procedure. These vulnerable infants are at risk for a number of suction-related complications, including hypoxemia, bradycardia, tachycardia, atelectasis, pneumonia, fluctuations in blood pressure and intracranial pressure, localized trauma to the airway, pneumothoraces, tube blockage, and extubation (Alpan, Glick, Pelig, Amit, & Eyal, 1984; Broadsky, Reidy, & Stanievich, 1987; Durand, Sangha, Cabal, Hopenbrouwers, & Hodgman, 1989; Kaiser, Grauss, & Williams, 2008; Kohlhauser et al., 2000; Mosca et al., 1997; Perlman & Volpe, 1983; Shorten, 1991; Simbruner et al., 1981; Storm, 1980).

Although ETT suctioning is one of the most commonly performed nursing procedures with infants who are ventilated (Barker & Rutter, 1995; Cignqacco et al., 2007), it has not been rigorously or extensively studied in the term or preterm newborn (Clifton-Koeppel, 2006). Endotracheal tube suctioning is usually performed by a sole (independent) caregiver, typically a nurse or a respiratory therapist (Gardner & Shirland, 2009) even though two-person suctioning (four-handed care) has been recommended as standard of care by professional organizations (Altimier et al., 2006). Although some NICUs have moved to a model of ETT suctioning involving two people, the practice is not a widely accepted, and no published data on the clinical feasibility or effectiveness of this practice exist (Cignqacco et al., 2010). Therefore, the purpose of this study was to compare the effect of four-handed care on preterm infants’ physiologic responses (oxygenation, heart rate, and stress) and behavioral responses (state, stress and defense behaviors, self-regulatory behaviors) to and recovery (time physiologic vital signs return to baseline) from endotracheal suctioning versus routine endotracheal suctioning performed by a sole caregiver.


Design and Sample

A crossover design was used with infants serving as their own controls. The data collection protocol was a randomized order of assignment to four-handed care or routine suctioning practices. The study was conducted in a single-family-room-designed NICU at an academic health center and was approved by the Institutional Review Board.

A convenience sample of 10 infants less than 37 weeks gestation at birth, less than one week old, intubated, on conventional ventilation, and receiving inline suctioning were recruited with parental informed consent. Infants less than 37 weeks gestation are regarded as preterm, frequently require mechanical ventilation, and are often unable to buffer negative stimuli. Infants on high-frequency ventilators such as oscillators or jet ventilators were excluded as those ventilators are used to manage specific pathologies. Only infants receiving inline suctioning were included to control for the affects that may be contributed by open-ended suctioning. Infants were excluded if they had chromosomal or genetic abnormities, congenital heart disease, dysmorphic syndrome, or were receiving paralytics, analgesics, or sedating medications because these conditions were felt to potentially alter the physiologic and behavioral responses of infants. All infants received both observations.

Physiologic Responses

Heart rate (HR) and oxygen saturation (SpO2) were used to measure physiologic response. In the preterm infant, the degree of stability in HR and SpO2 are indicative of wellness and indicates how well the infant tolerates the suctioning procedure. Accelerations or decelerations of HR may indicate unstable cardiac activity and decreased oxygen saturations are caused by changes in respiration (rate, pauses, apneic episodes). Heart rate was measured by electrical signals (electrocardiogram [ECG]) transmitted to patch electrodes that were placed on the infant’s trunk. The ECG signal is sampled at a rate of 1,000 samples per second. Heart rate data were collected using the Criticare Systems Scholar III monitor, which also documented alarm conditions (e.g., apnea, bradycardia). A neonatal oximetry sensor was attached to the infant’s foot to record SpO2 (Model 504-US, Criticare Systems). The pulse oximeter provides two analog output signals corresponding to the percent oxygen saturation level and pulse waveform. Each of these signals is sampled by the data acquisition card at a rate of 1,000 samples per second/channel, and the data are stored. Time series analysis is used to examine changes in SpO2 during the measurement period. These data were linked through a serial connection to the BioPac Data Acquisition System (MP150), which digitizes the data for easy storage and off-line analysis, and then to a notebook computer, where the numerical values of each parameter were logged, time-stamped, and recorded at regular 1-second intervals. Heart rate and SpO2 were reported in time sequenced vignettes: presuctioning observation lasting 10 minutes, during suctioning observation lasting from the initiation of suctioning to the completion of suctioning, and postsuctioning observation lasting an additional 10 minutes. The mean, the standard deviation from the mean, and two standard deviations above and below mean heart rate, and SpO2 were calculated in each of the reported time sequences. The 10-minute period was chosen to gain baseline data prior to the suctioning intervention and to capture potential residual effects of suctioning on HR and SpO2 post suctioning intervention.

Stress Response

Stress was measured with salivary cortisol, which is highly correlated with plasma cortisol in preterm infants. Normative values for preterm infants have not been well established; however, in a study of 48 newborns with a mean gestational age of 30.6 weeks +/− 1.9 weeks and birth weights of 1085 grams +/− 252.7, salivary cortisol levels ranged from below detection to 3.6 μg/dL (Calixto, Martinez, Jorge, Moreira, & Martinelli, 2002). Reference ranges for serum cortisol in well preterm infants range from 3.99 μg/dL–26.97 μg/dL in infants born at 24 weeks gestation to 2.39 μg/dL–16.1 μg/dL in infants born at 29 weeks gestation (Heckman, Wudy, Haack, & Pohandt, 1999). In preterm infants, cortisol levels rise with painful procedures (Herrington, Olomu, & Geller, 2004) and have been found to decrease with comforting procedures such as massage (Acolet et al., 1993) and skin-to-skin care (Morelius, Theodorsson, & Nelson, 2005). Cortisol is elevated in response to stress and has been used as an indicator of clinical intervention efficacy in preterm infants (Elverson & Wilson, 2005). Salivary cortisol has become relatively easy to collect in preterm infants using the filter paper method (Neu, Goldstein, Gao, & Laundenslager, 2007). In this study, pre- and postsuctioning salivary cortisol levels were obtained to assess infants’ stress. Pre-suctioning cortisol levels were obtained after an undisturbed behavioral state score was obtained. Postsalivary cortisol levels were obtained 30 minutes after suctioning. Cortisol samples were processed in an accredited lab using standard procedures.

Behavioral Responses

Regulation of the state system and behavioral responses exhibited by a newborn demonstrate the newborn’s underlying neurologic and behavioral competence and represent a significant developmental process. The state system identifies variations in sleep, wakefulness, and crying. Behavioral responses address an infant’s consolability and adaptability and are indicative of a newborn’s ability to respond to environmental conditions.

To assess the effect of four-handed care on behavioral state the Anderson Behavioral State Scale (ABSS) was evaluated before, during, and after observations. The ABSS was designed with consideration for the linear relationships between the states, heart rate, and energy consumption. The ABSS is particularly useful for assessing states in preterm infants because it breaks down the typical five or six states into 12 measures of state (four sleep states, five awake states, and three crying states), allowing a more sensitive indication of infant state that more closely captures the behavioral states exhibited by preterm infants. For scoring purposes the infant is observed for 30 seconds and the highest behavioral state noted in that time period is the score given. An undisturbed ABSS was obtained prior to handling the infant at the start of both observation conditions, again just prior to the initiation of suctioning, and at five other times occurring over a 10-minute time span, 2 minutes apart.

In addition to recognition of behavioral state, behavioral cues consist of autonomic and motor behaviors that are deemed essential for providing developmentally supportive care to preterm and critically ill newborns. Developmentally supportive care focuses on reducing the stressors associated with intensive care. Stress and defense behaviors (SDB) are thought to occur when the balance between too much or not enough stimulation exceeds the infant’s capacities (Harrison, Roane, & Weaver, 2004). Stress and defense behaviors are important to answer the study question because they provide an indication of how well the infant is tolerating the suctioning intervention. Behaviors that have been recognized as signs of stress in the preterm and critically ill newborn include hiccoughs, facial grimace, eye clinch, gaze aversion, tongue protrusion, finger splay, struggling movements, crying, whining, fussing, a cry face, spitting up or vomiting, halt hand or a hunger posture (Harrison et al.). These behaviors were considered defensive in nature when demonstrated by the infant before, during, and after the suctioning interventions were recorded and counted. Self-regulatory and approach behaviors (SRB) are indicative that an infant is ready for increased interaction with the environment and are also an indicator of well-being (White-Traut, Nelson, Silverstri, Cunningham, & Patel, 1997; White-Traut et al., 2004; White-Traut et al., 1999). Self-regulatory and approach behaviors were also deemed important to answering the question because they too provide an indication of how well the infant is tolerating the suctioning intervention. Behaviors that have been recognized as evidence of self-regulation and encourage engagement in the developing infant included subtle engagement cues: eye widening, facial brightening, raising of eyebrows and face opening, hand opening, fingers lightly flexed, searching eye movements, grasping and holding on, and efforts and successes at tucking trunk and limbs together (White-Traut et al., 1997; White-Traut et al., 2004; White-Traut et al., 1999). Potent engagement cues included facial gaze or focus, mutual gaze, and smooth cyclic movements of extremities. Counts of these infant behaviors of engagement were recorded during the three time sequences as described above, before, during, and after the suctioning interventions (Figure 1).

Figure 1
Counts of Self-Regulatory Behaviors Before, During, and After Suctioning in Routine vs. Four-Handed Care Condition.


Recovery was defined as the time it took for physiologic measures (HR and SpO2) to return within 10% of baseline using vital signs obtained from the monitor. Additionally, the percent of time HR and SpO2 fell outside of a normal range was calculated as it became apparent that some infants only momentarily returned to within 10% of baseline, thus not really recovering to a state of physiologic stability. Recovery was redefined to reflect a calculation of the percent of time the HR and SpO2 fell two standard deviations above and below the mean during and after suctioning when compared to the baseline presuctioning calculated mean. Recovery was further defined as the absence of call-back monitor alarms. The number and the type of monitor alarms (low SpO2, high SpO2, low HR, high HR) were recorded and notation was made of the type of event and if self-recovery or intervention was required. Recovery was evaluated ten minutes after suctioning and infants were monitored for an additional 10 minutes for monitor alarms or call-backs.


Demographic data about the infants included birth weight and gestation, race, ethnicity, sex, 1- and 5-minute APGAR scores. Infant severity of illness was measured using both the Score for Neonatal Acute Physiology (SNAPPE-II) and the Neonatal Medical Index (NMI). The SNAPPE-II was obtained to measure severity of illness computed on the first day of life. Scores on the SNAPPE-II range from 0 to 162 and are calculated from nine variables: temperature, mean blood pressure, PaO2/FiO2, pH, multiple seizures, urine output, birth weight, being small for gestational age, and the presence of an AP-GAR score at 5 minutes of life (Richardson, Gray, McCormick, Workman, & Goldman, 1993). Higher scores are indicative of increased risk for mortality and a lower score is indicative of better probable outcome. The NMI classifies infants by summarizing their medical conditions. Classifications range from 1 for infants born weighing more than 1000 g and without major complications, to a score of 5 for infants born weighing fewer than 1000 g and with very serious complications (Korner et al., 1987). A higher score is indicative of better outcomes and a lower score is indicative of increased instability.


Study personnel screened all patients in the NICU daily for eligibility. If a patient met inclusion criteria, the study was explained to the parent and consent was obtained. Each infant was observed twice on a single day during the first week of life. One observation involved routine suctioning care and one observation involved four-handed care (the intervention). Treatment order was randomly assigned. All observations coincided with routine care schedules and only medically indicated or scheduled suctioning opportunities were used for observations. Staff providing direct care on the study day was the same for both observations.

Presuctioning Care for Routine Care and Four-Handed Care Observations

Endotracheal suctioning procedures in neonates were reviewed with staff prior to suctioning observations (Table 1). On the day of observation a Nelcor pulse oximetry probe was placed on the infant to minimize disrupting the infant’s behavior state immediately before observation. Just prior to suctioning, behavior state was measured using the ABSS, and a salivary cortisol sample was taken. The infant’s monitor leads were then transferred to the computer data acquisition system and physiologic data were recorded minimally for 10 minutes. For the routine care and the four-handed care observation, behavior state was measured with the ABSS as described above. During both observations, physiologic data (HR, SpO2) were recorded continuously as described above. The infant was then reconnected to the hospital bedside monitor and observed for call-backs for an additional 10 minutes.

Table 1
Major Variables and Sampling Times

Routine Care

In the research setting, no specific suctioning protocols dictated how supportive care was provided during ETT suctioning. Typically a sole operator (clinician) suctioned the infant within a cluster of other caregiving activities such as obtaining vital signs, diaper changes, and was often followed by a gavage feeding. For the study, routine suctioning was the only care procedure that took place during the observation period.

Four-Handed Care

After presuctioning care was completed, the research assistant (RA) warmed her hands. The bedside nurse and the RA mutually agreed that the four-handed care procedure could begin. The RA opened the incubator porthole doors and placed her hands in the incubator. She assessed the infant’s attempts at self-regulation (observed for signs of stability, and self-regulation cues) and provided supportive care if the infant was disorganized by touching the infant gently, supporting positioning with hands to promote flexion, or assisting the infant in achieving a calm and regulated state by allowing finger grasp or plantar grasp to aid in motor regulation. When the RA and the bedside caregiver mutually agreed that they were ready to hyper-oxygenate prior to suctioning, the 100% O2 Cal/2 minute button was enabled, thus increasing the percentage of inspired oxygen to 100%, a common practice in this research setting. The RA was instructed to be aware of the needs of the bedside caregiver during suctioning and to provide assistance if needed to promote optimal performance of the procedure. The RA remained with the infant until an assessment of self-regulation had been achieved, as noted by stabilization of SpO2, heart rate and behavior, not to exceed 10 minutes at which point the RA removed her hands from the incubator. Documentation of the duration of the suctioning episode, the number of times the catheter was passed, and saline use was completed.

Postsuctioning Care for Routine Care and Four-Handed Care Observations

After 10 minutes had elapsed, the infant was reconnected to the hospital bedside monitor and observed for call-backs. Once the 10-minute recovery period had passed, further care deemed necessary was provided to the infant (i.e., diaper change, feeding, repositioning). A second salivary cortisol level was obtained 30 minutes after the completion of the suctioning procedure in the routine and the four-handed care conditions.

Data Analysis

Analysis of demographic data, including distribution, frequency, and measures of central tendency was completed for gender, race, ethnicity, birth weight, gestational age, APGAR scores (1-minute and 5-minute), SNAPPE-II, and NMI. Paired t tests were performed to determine differences in the duration of suctioning episodes and number of catheter passes between groups receiving routine care versus four-handed care.

The aim of the study was to compare the effect of four-handed care on preterm infants’ physiologic responses (oxygenation, heart rate, and stress) and behavioral responses (state, stress and defense behaviors, self-regulatory behaviors) to and recovery (return to physiologic baseline, call-backs) from endotracheal suctioning versus routine endotracheal suctioning performed by a sole caregiver. For analysis of the differences within and across groups on physiologic response, paired samples t tests and ANOVA were used. Dependent paired t tests were used to analyze stress (cortisol) responses. Spearman correlation was used to analyze the behavioral state scores where paired samples t tests and repeated measures analysis of variance (RMANOVA) were used for analysis of behavioral responses. A paired samples t test was run to compare recovery response.



The sample included 10 preterm infants, all non-Hispanic. Four (40%) of the participants were female, and seven (70%) were African American (Table 2;). Using a Wilcoxon signed-rank test, duration of suctioning did not differ between groups (routine = 73 seconds; four-handed = 75 seconds; Z = −.351, p = .726). The number of catheter passes was not different between groups (routine = 1.7; four-handed = 1.8; Z = −.577, p = .564.

Table 2
Demographics for Study Participants (N = 10)

Physiological Responses

Oxygenation and Heart Rate

No significant differences were found in HR (p = .09) or SpO2 (p = .329) during routine care among the three time periods (before, during, or after suctioning). No significant differences were found in HR (p = .297) or SpO2 (p = .068) during four-handed care among same three time periods. In the four-handed care condition, a significant increase of SpO2 from baseline to during suctioning period (p = 0.001) was observed (Table 3).

Table 3
Paired Samples t Tests for Heart Rate and Oxygen Saturation during Routine Care and Four-handed Care

Stress Responses

Salivary cortisol did not differ between groups at baseline routine (1.50 μg/dL ± .70) and four-handed care (1.39 μg/dL± 1.47); t(9) = .793, p = .448. Postsuctioning salivary cortisol also did not differ (routine care = 1.57 μg/dL ± 1.87; four-handed = 1.46 μg/dL ± 1.43); t(9) = .479, p = .664.

Behavioral Responses

Behavior State

No significant differences in behavior state were observed between the two conditions at any time point. No relationships in the ABSS scores were observed during the undisturbed, presuctioning, and 4-minute interval. At 2, 6, and 8 minutes a small to moderate correlation as assessed by Spearman’s was found (see Table 4).

Table 4
Spearman Correlation Results for Anderson Behavioral State Scale (ABSS) during Routine and Four-Handed Care Conditions

Stress and Defense Behaviors

Differences were found in the number of stress and defense behaviors exhibited across suctioning for the routine condition, F(2,18) = 6.274, p = .009, with a Bonferroni post-hoc analysis revealing significant differences before and during suctioning, F(2) - 6.274, p = .017. For the four-handed care condition, no differences in the number of stress and defense behaviors exhibited before, during and after suctioning were found, F(2,18) = 1.275, p = .304 (Table 5). No differences in the number of stress and defense behaviors were observed between routine versus four-handed care conditions before (p = .357) or during suctioning (p = .168) (Figure 2). More stress and defense behaviors were seen during routine care than four-handed care in the postsuctioning period (p = .001) (Table 6).

Figure 2
Counts of Stress and Defense Behaviors Before, During, and After Suctioning in Routine vs. Four-Handed Care Condition.
Table 5
Repeated Measures ANOVA for Stress and Defense Behaviors (SDB) & Self-Regulatory Behaviors (SRB) Before, During, and After Suctioning
Table 6
Paired Samples t Tests for Stress and Defense Behaviors (SDB) and Self-Regulatory Behaviors (SRB) Before, During, and After Suctioning for Routine and Four-Handed Care

Self-Regulatory Behaviors

No differences in the number of self-regulatory behaviors were exhibited before, during, or after suctioning in either the routine care condition (p = .149) or under the four-handed care condition (p = .165) (Table 4). No differences were seen in self-regulatory behaviors under the two conditions before suctioning (p = .257), however, there was a difference in the number of self-regulatory behaviors exhibited during suctioning (p = .019) and after suctioning (p = .016) (Table 2). Infants exhibited more self-regulatory behaviors during four-handed care as compared to routine care condition at both these measurement period (Table 5).


A significant difference in the rate at which the HR returned to baseline was observed when comparing the routine care condition (M = 0, SD = 0) to the four-handed care condition (M = 8.6, SD = 11.03), t(9) = −2.47, p = .036. During four-handed care 50% were noted to have a HR outside of baseline. No significant difference in the rate at which the SpO2 returned to baseline postsuctioning was observed when comparing the routine care condition (M = 4.40, SD = 13.91) to the four-handed care condition (M = 0, SD = 0), t(9) = 1.00, p = .343. Of note differences appreciated in the routine care condition were due to one infant outlier. No differences in monitor call-backs occurred when comparing the routine condition (M = .20, SD = .632) to the four-handed care condition (M = .40, SD = .699), t(9) = −.612, p = .555.


Endotracheal suctioning is regarded as a noxious routine care procedure that can produce stress for the vulnerable preterm infant. With the increased focus on providing developmentally supportive care in NICUs, implementation of non-pharmacologic strategies are sought to potentially reduce negative outcomes and aid infants’ attempts at self-regulation. Although previous studies have shown that nonpharmacological measures such as facilitated tucking, containment, nesting, and swaddling reduce pain scores in infants, researchers have not explored the provision of four-handed care, an intervention that serves to support the infant behaviorally during a stressful procedure such as endotracheal suctioning.

Ten preterm infants acting as their own controls were enrolled in this study with 50% at 24 weeks gestation at birth. Enrollment was slowed in this study by a recent shift in neonatal care practice where even very preterm infants were intubated in the delivery room for the administration of surfactant and then extubated to continuous positive airway pressure (CPAP). Thus, fewer than the expected number of infants met inclusion criteria. All infants remained enrolled during both suctioning observations. The order of suctioning was randomly assigned, and both suctioning interventions took place on the same day and were approximately 4 hours apart. The duration of the suctioning events did not differ between groups, nor did the number of required catheter passes. These results may have been affected by the use of in-line suctioning, which is routinely performed as it is unobtrusive suctioning technique as compared to the open-ended suctioning techniques historically used in the NICU.

No significant differences were observed in HR or SpO2 during the routine care condition across time periods or for the four-handed care condition across time periods. However, in the four-handed care condition, a significant increase of SpO2 was observed from baseline to suctioning. In this small pilot study, four-handed care provided by the research assistant did not appear to alter basic physiologic functions. However, the study setting uses the therapeutic positioning strategy of nesting commonly seen in NICUs. These practices may have influenced study findings as infants in the routine care condition were maintained with the positioning devices that the nurse caring for the infant had established. Infants who appeared well nested displayed little physiologic or behavioral responses during the suctioning event. Because infants who were nested or swaddled were in those conditions in both observations, it is not clear what effect these other interventions may have had on the study outcomes.

In previous studies of noxious procedures in preterm infants, researchers demonstrated HR changes over the course of a procedure (Granau, Oberlander, Holsti, & Whitfield, 1998; Stevens, Johnston, Petryshen, & Taddio, 1996). In this study, no appreciable difference in HR was noted during routine or four-handed care. This finding could be attributed to several variables, including small sample size, the lack of control for nesting, or the less noxious nature of in-line suctioning as opposed to traditional open-ended suctioning methods. Stressful events such as suctioning have also been shown to influence the metabolic rate by increasing the demand for oxygen (Clifton-Koeppel, 2006), and a number of researchers have indicated that suctioning can have a serious effect on infant behavior and physiologic responses. Although hypoxemia was not noted in this study, oxygen levels were increased in a like fashion in both groups by manually enabling a 100%O2/Cal/2 minute button on the ventilator in preparation for suctioning. In the future, researchers may consider increasing oxygenation to a lesser degree when providing preoxygenation. This practice may still achieve the desired result but may decrease the risk of hyperoxygenation, which is known to be potentially harmful to the developing visual system of the preterm infant.

Other authors have examined the effects of various developmentally supportive interventions during procedures. For example, in 2004, Ward-Larson et al. (2004) focused on the relief of pain during endotracheal suctioning using facilitated tucking, a positioning strategy in 40 preterm infants with gestational ages of 23 to 32 weeks. Facilitated tucking has been described by several researchers as a nonpharmacologic comfort measure for pain in preterm neonates and is defined as the gentle containment of the infant’s arms and legs in flexed midline position close to the infant’s truck with the infant in a side-lying or supine position (Corff et al., 1995; Hill et al., 2005). Ward-Larson et al. described facilitated tucking as the caregiver hand swaddling the infant by placing a hand on the infant’s head and feet while providing flexion and containment. Four-handed care differs from facilitated tucking in that four-handed care is focused on the needs of the infant and the caregiver (operator), and the goal is to provide anticipatory support for both parties during a procedure. Four-handed care does not necessarily involve repositioning of the infant. Ward-Larson et al. reported significant differences in pain relief as measured by the Premature Infant Pain Profile (PIPP) between those infants receiving tucking and those infants who did not, findings that are consistent with the hypothesis of the current study that infants who received four-handed care would demonstrate more physiologic stability and “recover” faster than those receiving routine care. This hypothesis however, was not supported in the current study. A second study of 32 preterm infants (Huang et al., 2004) with similar gestational ages employing facilitated tucking and swaddling was also conducted. The researchers found that although both interventions alleviated the pain response as measured by the PIPP, swaddling had a significantly greater effect on recovery of HR to baseline.

In this study, four-handed care was not different from the usual care routine. In both conditions, study infants were nested, and this procedure was not controlled for. Thus, it is possible that the practice of nesting affected the outcomes and that nesting may be sufficient as a developmentally supportive intervention, although we did not test that assumption. However, further evidence that swaddling might have affected the study outcome can be found in the report of a randomized controlled cross-over study of 15 preterm infants between 27 and 36 weeks gestation where swaddling immediately after heel-stick resulted in a lesser degree of facial activity and HR that returned to baseline more rapidly than in the control condition. Additionally, there was a faster increase and faster stability in oxygen saturation in the swaddled condition (Fearon, Kisilevsky, Hains, Muir, & Tranmer, 1997). Pressler, Turnage-Carrier, and Kenner (2010) suggested that in conducting clinical research it is difficult to create an intervention that follows a strict developmental care protocol. They contended that it is unethical to deny control group newborns interventions that in todays practice could be considered routine care, such as non-nutritive sucking, providing clustered care, skin-to-skin contact, or nesting. In this study, even though infants served as their own controls, the effects of these other potentially supportive interventions were not controlled; thus, their effect on the outcomes is unknown.

Salivary cortisol levels were also measured and found not to differ at baseline or postsuctioning with either routine or four-handed care. Although results were nonsignificant, high correlations were measured for the presalivary cortisol (r = .964) and for postsalivary cortisol (r = .939). This result indicates that although presalivary cortisol was increased at baseline, they also increased at the same rate for postsalivary cortisol regardless of treatment assignment.

During the routine care condition there was a statistically significant difference between the undisturbed ABSS and the ABSS assessed at 10 minutes. Although infants received four-handed care there were differences noted between the undisturbed ABSS and the ABSS at 2-, 4-, 8-, and 10-minute observations. This finding is consistent with the other research findings (Granau et al., 1998; Stevens et al., 1996) where preterm infants demonstrated sleep and activity changes over the course of noxious procedures. Other researchers suggested that nonpharmacologic interventions like four-handed care can also activate the attention of neonates, distracting them and thus modulating their pain response (Bellieni et al., 2001). This finding may explain why in this study when infants received the four-handed care intervention they were in a higher level of attention or alertness. Caution however must be exercised because of the small sample size. This activation of alertness could also substantiate why when infants received four-handed care demonstrated less stress and defense behaviors and demonstrated more self-regulatory and approach behaviors. However, there were no differences in monitor call-backs between conditions, a finding that may also have been affected by the routine use of supportive nesting.

Implications for Future Research

The aim of this study was to explore the provision of four-handed care as a means to behaviorally support infants; the extent to which the professional caregiver experienced that support was not examined. Additionally, as Lawhorn and Als (2010) contended, a caregiver providing support could be a parent, and this type of intervention was not examined. In this study, the four-handed care intervention was provided by trained researchers who were very experienced in neonatal care. All procedures were well outlined to ensure that those infants in the routine care condition and those infants in the four-handed care condition adhered to routine protocol. These areas need future study.

Although in this pilot study, the sample size was small and 50% of the infants were at extremely early gestational ages, the number of stress and defense behaviors exhibited after suctioning was found to significantly increase from baseline in the routine care condition and the numbers of self-regulatory behaviors were found to increase in the four-handed care condition. These findings, though positive, need further investigation before recommendations for the adoption of four-handed care into routine care can be made.


Funded in part by National Institute of Nursing Research, National Institutes of Health, R01 NR005182.


The authors report no conflict of interest or relevant financial relationships.

Contributor Information

Sharon Cone, Nurse Manager of the Newborn Intensive Care Unit at the Children’s Hospital of Richmond—Virginia Commonwealth University Health System and the ECMO Coordinator for VCU Health System, Richmond, VA.

Rita H. Pickler, Professor and nurse scientist at Cincinnati Children’s Hospital Medical Center, Cincinnati, OH.

Mary Jo Grap, Interim Associate Dean for Research and Nursing Alumni Endowed Professor in the Adult Health and Nursing Systems Department, Virginia Commonwealth University, School of Nursing, Richmond, VA.

Jacqueline McGrath, Professor in the School of Nursing, University of Connecticut, Storrs, CT and coordinator of nursing research, Connecticut Children’s Medical Center, Hartford, CT.

Paul M. Wiley, Assistant professor in the School of Dentistry, Virginia Commonwealth University, Richmond, VA.


  • Acolet D, Modi N, Giannokoulopoulos X, Bond C, Weg W, Clow A, Glover V. Changes in plasma cortisol and cate-choloamine concentrations in response to massage in preterm infants. Archives of Diseases in Childhood. 1993;68:29–31. [PMC free article] [PubMed]
  • Alpan G, Glick B, Peleg O, Amit Y, Eyal F. Pneumothorax due to endotracheal tube suction. American Journal of Perinatology. 1984;1:3345–3348. [PubMed]
  • Altimier L. Neonatal intensive care units (NICU) environment. In: Kenner C, Lott JW, editors. Comprehensive neonatal care: A physiologic perspective. St. Louis, MO: W.B. Saunders; 2007. pp. 480–490.
  • Altimier L, Brown B, Tedeschi L. National Association of Neonatal Nurses guidelines for neonatal nursing policies, procedures, competencies and clinical pathways. Glenview, IL: National Association of Neonatal Nurses; 2006.
  • Axelin A, Salantera S, Lehtonen L. Facilitated tucking by parents’ in pain management of preterm infants—A randomized crossover trial. Early Human Development. 2006;82:241–247. [PubMed]
  • Barker DP, Rutter N. Exposure to invasive procedures in neonatal intensive care unit admissions. Archives of Disease in Childhood. 1995;72:F47–F48. [PMC free article] [PubMed]
  • Bellieni CV, Buonocore G, Nenci A, Franci N, Cordelli DM, Bagnoli F. Sensoiral saturation: An effective analgesic tool for heel-prick in preterm infants. Biology of the Neonate. 2001;80:15–18. [PubMed]
  • Broadsky L, Reidy M, Stanievich JF. The effects of suctioning techniques on the distal tracheal mucosa in intubated low birth weight infants. International Journal of Pediatric Otorhinolarygology. 1987;14:1–14. [PubMed]
  • Calixto C, Martinez FE, Jorge SM, Moreira AC, Martinelli CE. Correlation between plasma and salivary cortisol levels in preterm infants. Journal of Pediatrics. 2002;140:116–118. [PubMed]
  • Cignqacco E, Axelin A, Stoffel L, Sellam G, Anand KJS, Engberg S. Facilitated tucking as a non-pharmacological intervention for neonatal pain relief: Is it clinically feasible? ACTA Paediatrica. 2010;99:1763–1765. [PubMed]
  • Cignqacco E, Hamers JPH, Stoffel L, van Lingen RA, Schutz N, Muller R, Nelle M. Routine procedures in NICUs: Factors influencing pain assessment and ranking by pain intensity. Swiss Medical Weekly. 2007;138:484–491. [PubMed]
  • Clifton-Koeppel R. Endotracheal tube suctioning in the newborn: A review of the literature. Newborn and Infant Nursing Reviews. 2006;6:94–99.
  • Corff KE, Seidemann R, Venkataraman PS, Lutes L, Yates B. Facilitated tucking: A nonpharmacologic comfort measure in preterm neonates. Journal of Obstetric, Gynecologic & Neonatal Nursing. 1995;24:143–147. [PubMed]
  • Durand M, Sangha B, Cabal LA, Hopenbrouwers T, Hodgman JE. Cardiopulmonary and intracranial pressure changes related to endotracheal suctioning in preterm infants. Critical Care Medicine. 1989;17:506–510. [PubMed]
  • Elverson CA, Wilson ME. Cortisol: Circadian rhythm and response to a stressor. Newborn and Infant Nursing Reviews. 2005;5:159–169.
  • Evans JC, Vogelpohl DG, Bourguignon CM, Morcott CS. Pain behaviors in LBW infants accompany some “non-painful” caregiving procedures. Neonatal Network. 1997;16:33–40. [PubMed]
  • Fearon I, Kisilevsky BS, Mains S, Muir DW, Tranmer J. Swaddling after heal lance: Age specific effects on behavioral recovery in preterm infants. Journal of Developmental & Behavioral Pediatrics. 1997;18:222–232. [PubMed]
  • Ferm D, Graves C, L’Huillier M. Swaddled bathing in the newborn intensive care unit. Newborn and Infant Nursing Reviews. 2002;2:3–4.
  • Gardner DL, Shirland L. Evidence-based guideline for suctioning the intubated neonate and infant. Neonatal Network. 2009;28:281–302. [PubMed]
  • Granau RE, Oberlander T, Holsti L, Whitfield MF. Bedside application of the neonatal facial coding system in pain assessment of premature infants. Pain. 1998;76:277–286. [PubMed]
  • Harrison LL, Roane C, Weaver M. The relationship between physiological and behavioral measures of stress in preterm infants. Journal of Obstetric, Gynecological & Neonatal Nursing. 2004;33:236–245. [PubMed]
  • Heckman M, Wudy SA, Haack D, Pohlandt F. Reference range for serum cortisol in well preterm infants. Archives of Disease in Childhood, Fetal & Neonatal Edition. 1999;81:F171–F174. [PMC free article] [PubMed]
  • Herrington CJ, Olomu IN, Geller SM. Salivary cortisol as indicators of pain in preterm infants. Clinical Nursing Research. 2004;13:53–68. [PubMed]
  • Hill S, Engle S, Jorgensen J, Kralik A, Whitman K. Effects of facilitated tucking during routine care of infants born preterm. Pediatric Physical Therapy. 2005;17:158–163. [PubMed]
  • Huang CM, Tung WS, Kuo LL, Chang YJ. Comparison of pain responses of premature infants to the heelstick between containment and swaddling. Journal of Nursing Research. 2004;12:31–39. [PubMed]
  • Kaiser JR, Gauss CH, Williams DK. Tracheal suctioning is associated with prolonged disturbances of cerebral hemodynamics in very low birth weight infants. Journal of Perinatology. 2008;28:34–41. [PubMed]
  • Kohlhauser C, Bernert G, Hermon M, Popow C, Siedl R, Pollak A. Effects of endotracheal suctioning in high-frequency oscillatory and conventionally ventilated low birth weight neonates on cerebral haemodynamics observed by near infrared spectroscopy. Pediatric Pulmonology. 2000;29:270–275. [PubMed]
  • Korner AF, Fraemer HC, Reade EP, Forrest T, Dimiceli S, Thorn VA. A methodological approach to developing an assessment procedure for testing neurobehavioral maturity of preterm infants. Child Development. 1987;58:1478–1487. [PubMed]
  • Lawhorn G, Als H. Theoretical perspective for developmentally supportive care. In: Kenner C, McGrath JM, editors. Developmental care of newborns and infants: A guide for health professionals. Glenview, IL: National Association of Neonatal Nurses; 2010. pp. 19–41.
  • Mathews TJ, MacDorman MF. Infant mortality statistics from the 2006 period linked birth/infant death data set (National Vital Statistics Report) 12. Vol. 57. Hyattsville, MD: National Center for Health Statistics; 2010. pp. 1–31. [PubMed]
  • Morelius E, Theodorsson E, Nelson Salivary cortisol and mood and pain profiles during skin-to-skin care for an unselected group of mothers and infants in neonatal intensive care. Pediatrics. 2005;116:1105–1113. [PubMed]
  • Mosca FA, Colnaghi M, Lattanzio M, Bray M, Pugliese S, Fumagalli M. Closed versus open ended endotracheal suctioning in preterm infants: Effects on cerebral oxygenation and blood volume. Biology of the Neonate. 1997;72:9–14. [PubMed]
  • Neu M, Goldstein M, Gao D, Laundenslager ML. Salivary cortisol in preterm infants: Validation of a simple method for collecting saliva for cortisol determination. Early Human Development. 2007;83:47–54. [PubMed]
  • Perlman JM, Volpe JJ. Suctioning in the preterm infant: Effects on cerebral blood flow velocity, intracranial pressure and arterial blood pressure. Pediatrics. 1983;72:329–342. [PubMed]
  • Peters KL. Bathing premature infants: Physiological and behavioral consequences. American Journal of Critical Care. 1998;7:90–100. [PubMed]
  • Pressler JL, Turnage-Carrier CS, Kenner CA. Developmental care: State of the science. In: Kenner C, Lott JW, editors. Comprehensive neonatal care: A physiologic perspective. St. Louis, MO: W.B. Saunders; 2010. pp. 1–17.
  • Richardson DK, Gray JE, McCormick MC, Workman K, Goldman DA. Score for neonatal acute physiology: A physiologic severity of illness for the neonatal intensive care unit. Pediatrics. 1993;91:617–623. [PubMed]
  • Shorten DR, Byrne PJ, Jones RL. Infant responses to saline instillations and endotracheal suctioning. Journal of Obstetric, Gynecologic, & Neonatal Nursing. 1991;20:464–469. [PubMed]
  • Simbruner G, Coradello H, Fodor M, Havelec L, Lubec G, Pollak A. Effect of tracheal suction on oxygenation, circulation, and lung mechanics in newborn infants. Archives of Disease in Childhood. 1981;56:326–330. [PMC free article] [PubMed]
  • Stevens BJ, Johnston CC, Franck L, Petryshen P, Jack A, Foster G. The efficacy of developmentally sensitive interventions and sucrose for relieving procedural pain in VLBW neonate. Nursing Research. 1999;98:35–43. [PubMed]
  • Stevens B, Johnston CC, Petryshen P, Taddio A. The premature infant pain profile. Clinical Journal of Pain. 1996;12:13–22. [PubMed]
  • Storm W. Transient bacteremia following endotracheal suctioning in ventilated newborns. Pediatrics. 1980;65:487–490. [PubMed]
  • Taquino L, Blackburn S. The effects of containment during suctioning and heelstick on physiologic and behavioral response of preterm infants. Neonatal Network. 1994;13:55. [PubMed]
  • Tatad AM, Frayer WW. Trends in the NICU: A review of 25 years of experience. American Journal of Perinatology. 2003;20:441–446. [PubMed]
  • Ward-Larson C, Horn RA, Gosnell F. The efficacy of facilitated tucking for relieving procedural pain of endotracheal suctioning in very low birth weight infants. The American Journal of Maternal Child Nursing. 2004;29:151–158. [PubMed]
  • White-Traut RC, Nelson MN, Silverstri JM, Cunningham N, Patel M. Responses of preterm infant to unimodal and multimodal sensory intervention. Pediatric Nursing. 1997;23:169–175. [PubMed]
  • White-Traut RC, Nelson MN, Silverstri JM, Patel M, Berbaum M, Gu GC, Rey PM. Developmental patterns of physiologic response to multisensory intervention in extreme premature and high-risk infants. Journal of Obstetric, Gynecologic & Neonatal Nursing. 2004;33:266–275. [PubMed]
  • White-Traut RC, Nelson MN, Silverstri JM, Patel M, Vasan U, Han BK, Bradford L. Developmental intervention for preterm infants diagnosed with periventricular leukomalacia. Research in Nursing & Health. 1999;22:131–143. [PubMed]