Research in Nursing & Health, 1991, 14, 397-403

Effects of a Skin Refrigerant/ Anesthetic and Age on the Pain Responses of Infants Receiving Immunizations Virginia Egbert Maikler

Although pain management is an important nursing responsibility, few interventions for minimizing distress of infants during painful procedures have been studied. Sixty infants, 2 through 6 months of age, were studied during a routine DPT immunization in order to examine (a) the effectiveness of cooling the skin in reducing distress behaviors and (b) the relationship between age and distress behavior. Infants were randomly assigned to experimental or control groups. Prior to the injection, the sites were sprayed for 2 to 3 s with either a skin refrigerant/anesthetic or compressed air. Behaviors were video- and audio-taped during and 60 s post injection, and coded for (a) facial expression, (b) cry, and (c) body movements. MANOVA revealed fewer distress behaviors following refrigerant spray and more complex, varied behavioral responses for older infants. The findings provide further evidence that infants perceive pain and that nursing interventions for pain reduction should be tested and extended to the very young.

In the past two decades, there has been an increase in the study of pain, primarily in adults. Until recently, however, there has been little research reported with neonates and infants. Pain is generally considered a subjective experience, “whatever the experiencing person says it is” (McCaEery, 1979, p. 11). Unfortunately, the infant is unable to communicate this perception through language, making the study of pain in this population a conceptual and methodological challenge. For many years it was thought that infants were incapable of experiencing pain. This misconception was based on an early study by McGraw (1941), who examined the responses of children, birth to 4 years of age, to safety-pin pricks. She found that some neonates within a few hours of birth did not respond, and concluded that the neonate was incapable of experiencing pain. For nearly

20 years no one replicated or validated these results. In 1960, Levy attempted to assess at what age an infant remembers previous pain experiences, using the immunization experience. None of the infants under 6 months of age responded negatively when brought into the mom in which they had previously received an injection. Only 1% of infants 6 months of age responded negatively. This was interpreted to mean that even if young infants did experience pain, they would not remember the experience. As a result, interventions to minimize pain have been, and continue to be, denied to many infants. There has, however, been a growing recognition and consensus that infants do experience pain (Anand & Hickey, 1987). Pain responses following circumcision, heel lances, pin pricks, and ear pinch have been documented in numerous studies (Dale, 1986; Franck, 1986; Grunau &Craig, 1987; John-

Virginia Egbert Maikler, RN, PhD, is an assistant professor in the College of Nursing at Rush University, Chicago. This research was supported in part by the National Center for Nursing Research, Grant No. 1 F31 NR06191, and the Alpha Lambda Chapter of Sigma Theta Tau International. The author thanks Felissa Cohen, RN, PhD; JoAnn Eland, RN, PhD; Agatha Gallo, RN, PhD; Rathe Karrer, PhD; Minu Patel, MS; and Anna Tichy, RN, PhD; for their helpful suggestions during the implementation of this study. This article was received on August 10, 1990, was revised, and accepted for publication July 5, 1991. Requests for reprints can be addressed to Dr. Virginia Maikler, College of Nursing, Rush University, Schweppe Sprague Hall, Room 301, 1743 W. Harrison, Chicago, IL 60612. 0 1991 John Wiley & Sons. Inc. CCC 0160-6891/91/060397-07 $04.00

398

RESEARCH IN NURSING & HEALTH

ston & Strada, 1986; Mills, 1989; Owens & Todt, 1984). These responses include an intense cry, a distinct facial expression, and a variety of body movements, such as startling, thrashing, and withdrawing . Further, only a few investigators have described how the distress behavior changes as the infant grows and develops (Craig, McMahon, Morison, & Zaskow, 1984; Izard, 1983; McGraw, 1941). Very young infants demonstrate “reflexive withdrawal” followed by dramatic, diffuse behavior consisting of intense crying and body rigidity. This behavior changes with age, with older infants and toddlers orienting more to the source of pain, crying less intensely, and attempting to flee from the pain source. The facial expression of pain can be observed in the young infant, while the expression of anger is not apparent until late infancy and toddler years. Although these studies provide some guidance for measuring pain responses of infants, additional research is needed to establish developmentally-appropriateobservation measures. Several investigators have tested interventions designed to decrease pain for the neonate. For circumcisions, penile blocks (Holve et al., 1983; Maxwell, Yaster, Wetzel, & Niebyl, 1987; Stang, Gunnar, Snellman, Conden, & Kestenbaum, 1988; Williamson & Williamson, 1983) and topical lidocaine (Mudge & Younger, 1989) reduced the intensity of distress behaviors. During heelstick for blood sampling, Harpin and Rutter (1983) found that a mechanical device thought to be less painful than a manual stick resulted in less palmer sweating in the neonates. Field and Goldson ( 1984) offered pacifiers to neonates receiving heelsticks and reported significantly less distress behavior. Conversely, Beaver (1987) found an increase in distress behavior during heelstick when stroking the premature infant’s opposite leg, an intervention thought to interfere with transmission along pain fibers. The sample size was small ( N = 8) and consisted of premature infants for whom the study of pain is just beginning. While these few studies have greatly contributed to our knowledge of the positive effects of interventions with the neonate, to date there are no reports of interventions with infants. Infants, however, routinely are subjected to repeated painful injections during immunization. This undoubtedly contributes to the fear children express of needle injections, medical care, and hospitals (Broome & Hellier, 1987). Although no one pain theory is universally accepted, the gate control theory of pain information transmission, introduced by Melzack and Wall in 1965, and restated by Wall (1978), has the greatest research support. They suggest that the interplay

between nerve fibers in the spinal cord can inhibit or facilitate the transmissions of pain information. For example, stimulating fibers that carry temperature information would inhibit pain information reaching the brain. Although conducted a decade prior to the publication of the gate control theory, Travel1 (1955) attempted to reduce the pain of injection for adults by cooling the skin at the injection site. Patients reported less pain when either ice cubes or a coolant spray was applied. Similarly, Eland (1981) tested the effects of a refrigerant/anesthetic spray applied to the skin of preschoolers prior to injection. The mean amount of expressed pain was less in those who received the refrigeranuanesthetic than those who received air. Given the seemingly beneficial effect of cooling the skin for adults and older children, the critical need to investigate interventions with infants, and the continuing gaps in our knowledge of the impact age and development have on pain response, the present study was conducted. It was hypothesized that (a) the application of a refrigerantlanesthetic to the skin would decrease the frequency and intensity of distress behaviors associated with a painful injection, and (b) distress behavior would increase in the variety of movement and decrease in the intensity of cry with increasing age.

METHOD

Sample Sixty infants, ages 6 to 30 weeks, with normal birth histories and no chronic health problems, were recruited from a pediatric ambulatory care department of a health maintenance organization, and from the office of a pediatrician in private practice. All were scheduled to receive a routine diphtheria-pertussis-tetanus (DPT) immunization and had no previous injections except routine immunizations. Eighteen percent of the infants were from the private practice setting and 82% were from the ambulatory care setting. Twenty-seven of the infants were male (45%); 33 were female (55%). The average age was 14.9 weeks. The sample was composed of 58% blacks, 17% whites, 17% hispanic, and 8% other ethnic groups.

Measures Facial expression. The facial expression of pain was coded using the Maximally Discriminative Facial Movement Coding System (MAX) (Izard,

INFANT PAIN / MAIKLER

1983). The MAX was derived from differential emotions theory which provides definitions of fundamental emotions, the anatomy of facial expressive muscles, and cross-culturally standardized photographs of faces expressing these emotions. Content, criterion, and construct validity are discussed in the accompanying manual. For example, in one study, there was a high level of agreement ( r = .88) among untrained observers regarding what emotions were indicated by the infants’ facial expressions. In another study, interrater reliability of .81 was reported (Izard, 1983). Similarly, Johnston and Strada (1986) reported an interrater reliability of .88 in a descriptive study of infants in pain. In this study, a research assistant coded the videotaped segments, observing movements in three anatomical regions for the presence or absence of clearly defined appearances of the pain-distress expression. The brow region was observed for (a) sharply lowered and drawn together brows with bulges between them, (b) vertical furrows on forehead, and (c) a broadened, bulging nasal root. The eyeslnoselcheek region was observed for (a) tightly closed eyes with a horizontal furrow of the upper lid, (b) raised cheeks with an increase in tissue mass, and (c) a deepening of the nasolabial fold. The mouth region was observed for an open, squarish, angular mouth. The facial expression was coded as pain-distress only if the described appearance was observed simultaneously in the three anatomic areas. The 60-s segment then was measured on two dimensions: (a) the time from the needle touching the skin to the pain-distress expression (latency to pain expression), and (b) the duration of the pain-distress expression. Interrater reliability was .90 during training and .98 during the study. Cry. The criteria for cry were developed from characteristics reported in the literature (Fisichelli, Karelitz, & Haber, 1969; Murray, Amundson, & Hollien, 1976; Porter, Miller, & Marshall, 1986; Zeskind & Field, 1985). Initial content validity was established through a panel of faculty and/ or researchers in the Speech and Hearing Science Department at a large midwestem university. They provided consultation during criteria development and tested the criteria with the training tapes. For the study, the coder noted the time displayed in the lower comer of the videotaped segment at the beginning and end of each cry burst. The type of each burst was coded as (a) intense, defined as urgent, arousing, high-pitched, piercing, loud, screaming, or (b) prorest, defined as less arousing, rhythmical, lower pitched, muscial. Additional sounds were noted, such as cooing, “mumming,”

399

or calling. Cry behavior then was measured on four dimensions: (a) the time from the needle touching the skin to the first sound (latency to cry), (b) the duration of crying, (c) the duration of intense crying, and (d) the duration of protest crying. Interrater reliability was .87 during training and .95 during the study. Body movement. The criteria for body movement were developed from behavior reported in the literature (Craig et al., 1984; Dale, 1986; Johnston & Strada, 1986). Initial content validity was established through a panel of faculty from the Developmental Psychology Department and the Maternal Child Nursing Department of a large midwestem university. They provided consultation during criteria development and tested the criteria with the training tapes. For the study, videotaped segments were coded for movement of the extremities and torso, and coded extension, flexion, waving, or kicking. Two total body descriptors were noted: (a) startle, a sudden total body jerk, and (b) symmetry, the majority of movements of both arms or both legs occurring at the same time and with the same motion. Movement then was measured on four dimensions: (a) the time from the needle stick to the onset of movement of any extremity (latency to movement), (b) the total number of movements, (c) the presence of a startle response with the needle insertion, and (d) the symmetry of movement. Interrater reliability was .89 during training and .95 during the study. Procedure Assignment to study groups was determined by numbering envelopes from 1 through 60,selecting a sequence of 60 1s and 2s from a table of random numbers, writing this number on the inner flap, and sealing the envelope. A coin toss determined that “1” was the experimental group and “2” the control group. Medical records of infants scheduled for routine health maintenance visits were reviewed daily against inclusion criteria for the sample. Upon arrival at the office, parents of eligible infants received written and verbal information about the study. If parents agreed to their infant’s participation, the consent was signed. While the physician examined the infant, the investigator randomly selected an envelope, assigning its code number to the infant. The number on the envelope was videorecorded in order to match the video segment with the code number. The envelope then was opened to determine which spray, refrigeranthesthetic or air,the infant would receive.

400

RESEARCH IN NURSING 8 HEALTH

A separate room was used for the study. It contained an infant seat placed on an examining table. A Sony CCD-V5 video camera with tripod stood three feet from the table. After the physician’s examination, the infant, accompanied by the parent, was brought to the room and placed in the infant seat facing the camera. The area of injection was swabbed with alcohol and allowed to dry. The experimental group had the skin sprayed for 2 to 3 s with FrigidermB (dichlorotetrafluorethane), a nonflammable refrigerantlanesthetic solution not considered toxic if absorbed (Poison Control Center, Rush Presbyterian St. Luke’s Medical Center, Chicago, IL , personal communication, August 10, 1987). Cooling begins immediately on contact with the skin. Although Eland (1981) sprayed the site for 3 to 5 s, three infants in the pilot study exhibited distress behaviors with this procedure. Thus, the spray time for this study was reduced to 2 to 3 s. If continued, freezing results after 20 to 30 s of spraying. Hanke and O’Brian (1987) found no microscopic skin damage for spray times of 45 s, but several minutes can result in freeze bums. In this study, the spray time of 2 to 3 s resulted in a brief frost on the skin; no redness remained beyond 1 min. The control group received compressed air (Dust-off@) to the area. The injection was given by the investigator. The infant’s responses were video and audiotaped from the application of the spray until 60 s after the injection. Neither investigator nor parents interacted with the infants during the taping. During the pilot a longer taping was attempted, but parents expressed increasing anxiety and a strong desire to pick up their infant as time increased beyond 60 s. Several other variables related to the injection were held constant: (a) all infants were awake and, except two, quiet prior to the procedure, (b) both spray cans were held at the same distance from the skin, (c) spray times were 2 to 3 s, (d) the volume of DFT solution was 0.5 cc in a 1-cc disposable syringe with a 5/8-in. 25-gauge needle attached, (e) the investigator administered all injections, (f) all injections were administered in the left anterior thigh, (g) the skin was pulled taut and quickly penetrated with the syringe at a 90” angle, (h) the solution was instilled over 5 s, and (i) the needle was quickly withdrawn. Coding. A research assistant coded the audiovideotaped segments after completing a training program. Prior to coding, the tapes were edited by the investigator to exclude the application of the spray to prevent the research assistant, who was not present during data collection, from identifying group assignment. Initial reliability of coding was attained by two coders using training

tapes developed by the investigator. During the study, 10%of the segments were randomly selected for assessment of reliability. As reported earlier, interrater reliability exceeded the .85 level, established as the minimal acceptable score using a percent-agreement method.

RESULTS

Distress Response Means and standard deviations of measures are presented in Table 1. It was hypothesized that the application of the skin refrigeranthesthetic would result in decreased distress response by the infant. A multivariate analysis of variance (MANOVA) was used to consider all measures of facial expression, cry, and body movements simultaneously (Table 2). Results revealed a significant difference between the treatment group and the control group. There was no interaction between age and group. A follow-up univariate analysis of variance (ANOVA) was conducted to determine on which measures there were significant effects (Table 3). The infants in the experimental group startled less when the needle was inserted and took longer to begin to cry than those in the control group. Their movements were less symmetrical. There were no differences for the remaining variables.

Response by Age It was hypothesized that the infant’s age also might influence the response to a painful stimulus. Infants were divided into two age groups (under 16 weeks and over 16 weeks). MANOVA revealed a significant main effect for age (Table 2). The followup ANOVAs showed that the duration of the pain expression was longer for the younger infants compared to the older infants (Table 4). The younger infants exhibited longer periods of “intense” crying and more Symmetrical movement. The older infants had longer periods of “protest” crying.

DISCUSSION The findings from this study with infants are consistent with those in neonates, demonstrating that efforts to reduce pain result in diminishing the distress response exhibited by the infant. Although all infants demonstrated distress behavior, those who received the local application of refrigeranu

'Seconds. b'fes/no response (1 = yes, 2 = no). Number.

Pain expression durationa Cry Latency to crya Cry durationa Intense crying durationa Protest crying durationa Movement Startle with needle insertionb Latency to movementa Total number of movementsC Symmetry of movementb

Facial expression Latency to pain expressiona

Behavior

2.12 ? 46.50 ? 28.79 2 19.86 2 0.74 5.46 15.90 13.57 0.50 2.23 20.71 0.50

1.66 2 56.13 2 43.19 2 12.31 2 1.38 ? 1.81 t 26.13 2 1.38 t

1.24 18.56 20.40 20.13 0.50 6.34 25.24 0.25

2.39 2 43.38 2 23.38 t 25.00 2 1.63 ? 3.25 2 38.50 t 1.94 ?

0.79 17.40 23.38 18.43 0.36 3.83 20.28 0.47

2.35 2 50.86 2 38.36 t 12.50 2 1.86 2 3.21 2 28.07 ? 1.71 2

1.36 2 0.50 1.79 2 1.67 35.00 t 14.96 1.93 2 0.27

0.80 18.04 19.46 16.65

2.42 2 0.94 50.43 t 7.83

16 Weeks or > (n = 14)

2.00 ? 0.96 55.00 t- 6.50

(n = 16)

Control (n = 30)

2.64 t- 1.08 51.43 2 12.83

anesthetic and age on the pain responses of infants receiving immunizations.

Although pain management is an important nursing responsibility, few interventions for minimizing distress of infants during painful procedures have b...
568KB Sizes 0 Downloads 0 Views