- Email: [email protected]
EMLA® cream and oral glucose for immunization pain in 3-month-old infants
Pain 104 (2003) 381–388 www.elsevier.com/locate/pain
EMLAw cream and oral glucose for immunization pain in 3-month-old infants Viveca Lindha,*, Urban Wiklundb, Hans K. Blomquista, Stellan Ha˚kanssona a
Department of Clinical Sciences, Pediatrics, University Hospital, Umea˚ University, S-901 85 Umea˚, Sweden b Department of Biomedical Engineering and Informatics, University Hospital, S-901 85 Umea˚, Sweden Received 26 March 2002; received in revised form 21 January 2003; accepted 27 January 2003
Abstract The objective of this study is to determine whether use of lidocaine – prilocaine 5% cream (EMLAw) and oral glucose decreases pain associated with diphteria – pertussis – tetanus (DPT) immunization in 3-month-old infants. Design: randomized, double-blind, controlled trial in outpatient paediatric practice in northern Sweden. EMLA or placebo was applied to the infant’s lateral region of the right thigh and covered with an occlusive dressing 1 h before the immunization. In addition, 1 ml of glucose (300 mg/ml) or placebo (water) was instilled on the baby’s tongue within 2 min before the DPT-injection. Forty-five infants received EMLA and glucose and 45 infants placebo cream and water. ECG was recorded and stored in a computer and the procedure was videotaped. The parents and the nurse assessed the infants’ pain on a visual analogue scale (VAS) after the immunization. Heart rate and heart rate variability pre- and post-injection were calculated. From the videotapes, the modified behavioural pain scale (MBPS) was used to assess pain scores during baseline and after immunization. The latency of the first cry and total crying time were measured. The parents and the nurse scored the infants’ pain on the VAS significantly lower in the treatment group than in the placebo group. The infants’ responses to the immunization measured as the difference in MBPS scores pre- and post-injection were significantly lower in the EMLA –glucose group compared with the placebo group. More infants cried after the immunization in the placebo group compared with the EMLA – glucose group and the latency of the first cry after the injection was shorter in the placebo group. A biphasic transient heart rate response with a marked deceleration followed by a subsequent acceleration was seen more frequently in the placebo group compared to the EMLA – glucose group. EMLA and glucose alleviate immunization pain in 3-month-old infants. q 2003 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Lidocaine–prilocaine; Topical; Heart rate variability; Modified behavioural pain scale; Crying
1. Introduction With the intention of inflicting as little procedural pain as possible, other routes of administration than the subcutaneous (s.c.) and intra-muscular (i.m.) injections are preferred in paediatric care. For the reason of best possible immune response to vaccine antigens, however, it is at present not possible to avoid the s.c. and i.m. routes in childhood immunizations. In order to meet stricter requirements for gentle handling, it is important to evaluate various pain-reducing strategies (Jacobson et al., 2001). Topical lidocaine –prilocaine patch (EMLAw) has no adverse effect on the antibody response of immunizations (Dohlwitz et al., 1998; Halperin et al., 2000). In adults, the * Corresponding author. Tel.: þ 46-90-785-2216; fax: þ 46-90-12-37-28. E-mail address: [email protected] (V. Lindh).
lidocaine – prilocaine cream decreases the pain from needle puncture and i.m. infiltration by approximately 50% (Himelstein et al., 1996; Taddio et al., 1992). EMLA also reduces injection pain in infants receiving their first i.m. DPT (diphtheria – pertussis –tetanus) immunization (Uhari, 1993; Taddio et al., 1994) and in 12-month-old infants undergoing s.c. MMR (measles – mumps – rubella) immunization (Halperin et al., 2000). Sweet-tasting solutions alleviate pain in newborn infants undergoing noxious procedures (Eriksson et al., 1999; Skogsdal et al., 1997; Stevens and Ohlsson, 2000). Glucose has at least the same analgesic effect as sucrose (Carbajal et al., 1999; Guala et al., 2001). The analgesic effect continues beyond the newborn period and shortens crying time after immunizations in 2-, 4- and 6-month-old infants (Barr et al., 1995; Lewindon et al., 1998; Ramenghi et al.,
0304-3959/03/$30.00 q 2003 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0304-3959(03)00046-0
382
V. Lindh et al. / Pain 104 (2003) 381–388
2002). The analgesic effect of saccharide combined with EMLA has shown to reduce pain during circumcision more effective than saccharide alone (Mohan et al., 1998). Hitherto this combination has not been tested for vaccination pain. Behavioural measures such as crying, the latency period until the first cry, the visual analogue scale (VAS), and the modified behavioural pain scale (MBPS) have been used to assess immunization pain in infants (Uhari, 1993; Taddio et al., 1994; Barr et al., 1995; Lewindon et al., 1998; Finley and McGrath, 1998; Halperin et al., 2000). An increased heart rate (HR) response to vaccination has also been described (Johnston and Strada, 1986). In our previous studies, the autonomic modulation of HR in response to heel lancing and venipuncture was analysed (Lindh et al., 1997, 1999, 2000), and was found to be a useful tool for objective assessment of pain. Spectral analysis of heart rate variability (HRV) was used to determine the frequency content of the fluctuating HR. The power of the high-frequency component (above 0.15 Hz) mirrors parasympathetic activity, i.e. vagal tone and reflects instant responses to changes in autonomic balance. The slower fluctuations are mediated by both sympathetic and parasympathetic activity (Task Force, 1996). The aim of this study was to examine whether topical application of EMLA combined with oral glucose 300 mg/ml decreases the pain associated with i.m. DPT immunization in three-month-old infants when assessed by HRV and behavioural scores.
2. Method 2.1. Materials The investigator or the nurse in charge provided verbal and written information to the parents before the first DPT vaccination and verbal parental consent was obtained. A total of 90 infants, 3 months of age, who were receiving their regular DPT vaccination were recruited for the study. These infants were born at term, were of normal birth weight, were otherwise healthy and required only routine post-natal care on the maternity ward.
envelopes labelled with the subject’s study number. The list of random numbers was not opened until all the analyses were completed. 2.3. Procedure The investigator saw the family at the Well Baby Clinic when the baby was scheduled for the first DPT vaccination. One gram of EMLAw or placebo was applied to the lateral region of the right thigh. The cream was covered with an occlusive dressing (Tegadermw, 3 M) for 1 h. Three electrocardiogram (ECG) electrodes (Blue Sensor, Medicotestw) were attached to the baby’s chest. The parents were instructed to feed the baby within a period of 1 h before the immunization. The procedure was performed in a separate, quiet room. All vaccination procedures were performed by the same investigator (V.L.). The mother ðn ¼ 84Þ or father ðn ¼ 6Þ held the baby during the vaccination and was told to comfort the baby as usual. The procedure was videotaped, including sound recordings, and the ECG signal was sampled and stored in a computer (BioBench, National Instruments, Austin, TX). The recordings were divided into three sequences: (a) a 3-min baseline period when the baby was held in the parent’s lap; (b) when the investigator (V.L.) instilled 1 ml of glucose (300 mg/ml) or water on the baby’s tongue with a syringe, and wiped the lateral aspect of the baby’s thigh with alcohol and (c) the i.m. administration of 0.5 ml of Infanrixw (SmithKline Beecham Pharma) using a 2.5 cm 23-gauge needle. The video and ECG recordings were continued until the baby had settled down after the injection. In the group of infants receiving EMLA –glucose, there were 25 boys and 20 girls ðn ¼ 45Þ; and in the placebo – water group, there were 27 boys and 18 girls ðn ¼ 45Þ: Immediately after the procedure, the parents were asked to rate the infant’s experience of pain during the vaccination on a VAS-scale where zero was designated “no pain” and 10 the “worst possible pain”. For the purpose of obtaining two independent VAS-ratings, a rating of which the parents were not aware was done by the investigator before the parents gave their opinion. The study was approved by the Ethics Committee of the Medical Faculty of Umea˚ University and was performed in accordance with the Declaration of Helsinki.
2.2. Randomization 2.4. Analyses As infants were enrolled, they were consecutively assigned a subject number that served as the link to the random allocation to either the EMLA –glucose or placebo –water group. The randomization and packaging of the EMLA –glucose and placebo – water were performed by the hospital pharmacy with a computer-generated list of random numbers in balanced blocks of six. Accordingly, the baby received either EMLA and glucose or placebo and water. The EMLA patch and glucose solution or the placebo patch and water (packaged in syringes) were contained in
2.4.1. ECG The ECG was sampled at 1000 Hz and R-waves were detected as previously described (Lindh et al., 1997). All recordings were inspected and corrected for detection errors. Artefacts caused by signal saturation due to baseline wander or arrhythmic beats were replaced by interpolation. However, all segments where three or more consecutive beats were interpolated were marked and not considered for further analysis. The beat-to-beat HR was calculated and
V. Lindh et al. / Pain 104 (2003) 381–388
converted to a time series by cubic spline interpolation and resampling at 4.8 Hz. Three segments were selected for statistical analysis: (a) 45 s from the middle of the baseline recording; (b) the last 45 s before the injection and (c) 45 s starting immediately after the injection. HRmean was calculated for each segment. Spectral analysis was assessed on detrended data using autoregressive modelling of order 30. The detrending was accomplished by using a digital filter with a cut-off frequency at approximately 0.02 Hz as previously described (Wiklund et al., 1997). The total spectral power (Ptot) and the power of the low-frequency (PLF, 0.02– 0.15 Hz) and high-frequency (PHF, 0.15 – 2.0 Hz) components were calculated and log-transformed. In the context of pain, power of the high-frequency component, i.e. vagal tone, has shown to decrease in term newborns during heel lancing and circumcision (Porter et al., 1988; Gunnar et al., 1995; Lindh et al., 1999). The HR response during the first 45 s after injection was also evaluated utilizing Poincare´ graphs, i.e., plots of each R – R interval against the subsequent value (Woo et al., 1992). The geometry of this plot visualizes the nature of variability and has been shown to correspond to sympathovagal balance and different clinical conditions (Otzenberger et al., 1998; Brennan et al., 2001). A Poincare´ -plot of an HR recording in a healthy, satisfied term newborn will show a high variability and the swarm of points will appear spread in the upper right-hand corner of the plot. If the baby suddenly is subjected to a painful procedure, HR will increase and the variability will decrease. The swarm of points will gather and appear in the lower left-hand corner of the plot (Lindh, 2002). In this study, the geometric appearance of the Poincare´ -plot was quantified by calculating the area of the smallest convex polygon that enclosed all points in the plot (Datta et al., 2000; Kawa et al., 1998). The minimum and maximum values of the R – R interval during the post-injection phase were determined. 2.4.2. Video recordings The facial expression, cry and body movements of the infants were assessed by means of the MBPS (Taddio et al., 1994). Two experienced paediatric nurses, working independently of each other and blinded to the study treatment, rated three sequences: (a) baseline; (b) 0 –10 s after the injection and (c) 11– 20 s after the injection. For each infant, differences between MBPS scores after the injection and the score at baseline were calculated. Whether or not a cry was present was noted, and latency of the first cry after the injection and total crying time were measured. 2.4.3. Statistical analyses Statistical analyses of the HRV data from the three sequences were calculated using analysis of variance (ANOVA) for repeated measurements. Otherwise the ttest for independent samples was used for intergroup comparisons and the x 2-test for the frequency of infants
383
who cried after the injection in the placebo and treatment groups, respectively. Cohen’s kappa (Norusis, 1997) was used to calculate the agreement between raters of the MBPS scores. The level of two-sided statistical significance was defined as a P-value , 0.05. Based upon the increase in HR during venipuncture in newborn infants (Lindh et al., 2000), it was calculated that 45 infants in each group represented 95% statistical power (a difference between the groups of HRmean 14 beats/min SD 18). The SPSS software (SPSS, Chicago, IL) was used for statistical analysis.
3. Results 3.1. Excluded recordings One video (and audio) recording in the EMLA – glucose group was excluded because the camera accidentally stopped before the study procedure was completed. In one video recording in the placebo group and two in the treatment group, there were moments when the view of the infant was blocked, which made a complete assessment of MBPS impossible. These were excluded from the calculations of the difference between MBPS scores before and after the injection. One ECG recording was not stored on the computer and two were excluded because of frequent artefacts, two in the EMLA –glucose group and one in the placebo group. 3.2. HR and HRV HR, total variability and the power of the low-frequency component during the sequence after the injection were significantly increased in both groups compared to baseline (see Fig. 1). The power of the high-frequency component was somewhat increased in the placebo group and decreased in the EMLA – glucose group. No statistically significant differences between groups were seen, i.e. no significant effect of the interaction between the group and time factors was found (analysis of variance, repeated measurements). A biphasic transient HR response was significantly more frequent in the placebo group during the sequence after the injection, as compared with the EMLA – glucose group. These responses were characterized by a marked deceleration followed by a subsequent acceleration. The decelerations, i.e. the maximal R –R intervals, were significantly larger in the placebo group ðP ¼ 0:03Þ; and the accelerations, i.e. the minimal R – R intervals, were significantly shorter ðP ¼ 0:03Þ than in the EMLA – glucose group. The maximal R –R interval corresponding to the longest interval between heartbeats was 1.9 s in the placebo group and 0.6 s in the treatment group. A larger polygon area (. 0.017 s2) in the Poincare´ -plot occurred more frequently in the placebo group (10/44) than in the EMLA – glucose group (1/43) (x 2test P ¼ 0.007, see Fig. 2). The power of the high-frequency component after
384
V. Lindh et al. / Pain 104 (2003) 381–388
Fig. 1. Mean values and standard errors of the mean for (A) HR, (B) total variability, (C) power of the low-frequency component and (D) power of the highfrequency component during rest, the sequence when glucose or water was administered, and after the immunization. *P , 0:05 indicates comparisons between base line versus vaccination.
vaccination, as compared with baseline values, showed three distinctly different patterns: (a) a low variability as a consequence of a rapidly increasing HR; (b) a relatively unchanged variability in infants where no marked response to the immunization occurred or (c) an increased variability due to the biphasic HR response. The infants in the placebo group responded significantly more often with either low or increased high-frequency power as compared with the EMLA –glucose group (x 2-test P ¼ 0.006, see Table 1). 3.3. Crying time and behavioural pain measurements More infants cried after the vaccination in the placebo group than in the EMLA – glucose group, (x 2-test P ¼ 0:001; see Table 2). Among the infants who cried, the total Table 1 Number of infants in the EMLA–glucose and placebo–water groups with low, unchanged or high power of the high-frequency component after the immunization HF component
EMLA–glucose
Placebo–water
Total
Low (,3.0 log mHz2) Unchanged High (.3.8 log mHz2)
12 30 1
15 19 10
27 49 11
crying time after the injection was shorter in the EMLA – glucose group, although it did not reach statistical significance. The infants in the placebo group started to cry sooner after the injection than the infants in the treatment group P , 0:001: Total crying time and latency of cry are given in Table 3. The parents and the investigator rated the infants’ pain on the VAS significantly lower in the treatment group than in the placebo group (VAS-parents P ¼ 0:001; VAS-nurse P , 0:001Þ: The agreement between the two MBPS raters, calculated with Cohen’s kappa, was 1.0 for baseline and 0.5 for the assessments after injection, which means that the agreement exceeded chance levels. In cases where the ratings diverged, Table 2 Number of infants who cried or did not cry after the immunization: comparison between the EMLA–glucose and placebo–water groups EMLA–glucose Placebo–water Total Infants (n) who cried after the 32 injection Infants (n) who did not cry 12 after the injection Total 44
44
76
1
13
45
89
V. Lindh et al. / Pain 104 (2003) 381–388
385
Fig. 2. HR recordings in three infants during the complete study procedure, followed by Poincare´ -plots for the sequences after the vaccination with the corresponding enclosing polygon outlined: (A) illustrates an unchanged HR after the vaccination, (B) an increased HR response and (C) a biphasic HR response.
the mean values of the two ratings for each sequence were used. The infants’ responses to the immunization, measured as the difference in MBPS scorings pre- and post-injection, were significantly lower in the EMLA –glucose group than in the placebo group (difference between baseline and 0 – 10 s after the injection, P , 0:001; difference between baseline and 11 –20 s, P ¼ 0:004Þ: Means and standard
deviations for the VAS, the MBPS, duration of time until crying, and total crying time are shown in Table 3.
4. Discussion The main result of this study was that the combined use of EMLA and glucose alleviates the vaccination pain
386
V. Lindh et al. / Pain 104 (2003) 381–388
Table 3 VAS and MBPS scores, latency of cry and total crying time after the vaccination: comparison between the EMLA–glucose and placebo–water groups
VAS parents VAS nurse MBPS baseline MBPS 0–10 s after injection MBPS 11–20 s after injection Diff baseline–MBPS 0–10 s after injection Diff baseline–MBPS 11–20 s after injection Latency of cry after injection (s) Total crying time (s)
EMLA–glucose
Placebo– water
3 ^ 2* 2 ^ 2* 1.9 ^ 0.4 5.5 ^ 2.0* 5.4 ^ 2.4* 3.6 ^ 2.1* 3.5 ^ 2.3* 6.4 ^ 3.2* 41 ^ 44
5^2 5^2 1.8 ^ 0.4 7.7 ^ 1.7 6.8 ^ 2.2 5.9 ^ 1.7 5 ^ 2.3 3.8 ^ 2.3 50 ^ 52
*P , 0:05:
in 3-month-old infants. The behavioural pain responses clearly showed a significant reduction of the pain response in the treatment group as compared with the placebo group. The mean values of the HR and HRV data per se did not discriminate the pain responses in the EMLA – glucose group from the placebo group. However, a marked biphasic HR response was significantly more common in the placebo group. The response to the injection was easily observed and the behavioural parameters clearly indicated a decreased pain response in the treatment group. Fewer infants cried in the EMLA – glucose group and among those who cried the latency of the first cry was prolonged as compared with the infants receiving placebo – water. However, the pain was not completely alleviated. There was an increase in MBPS scores after the injection as compared to baseline and 36/44 infants cried even if EMLA – glucose was applied. The VAS ratings made by the investigator and the parents correlated well to the MBPS scores obtained from video analyses. The biphasic or increasing HR responses after the injection made it difficult to draw conclusions from the mean values of the data. Consequently, the data were best described in terms of the different response patterns. This was also true for the power of the high-frequency component. The range was wider in the placebo group, implying that the injection without analgesia influenced vagal tone more than it did in the EMLA – glucose group. There were no clear correlations between the VAS scorings and the mean values of the HRV data due to these different response patterns. However, nine of the 11 infants with the most pronounced biphasic HR response were scored above VAS 5 by the parents. The observed biphasic HR response immediately after immunization has been described by Johnston and Strada (1986), and heel lancing in term newborns caused a similar but weaker response (Lindh et al., 1999). This response pattern is associated with the “fear paralysis” – reflex, characterized by a sympathetic inhibition together with a vagal bradycardia. It can be
triggered by a sudden noise, pain or by restraining the infant (Lagercrantz et al., 1995). A high pre-injection mean HR (146 beats/min) was seen in the infants in this study. HR during sleep is approximately 120 beats/min at this age (Harrington et al., 2001). In a previous study (Lindh et al., 2000), the baseline mean HR in newborns before metabolic screening in the maternity ward was 122 beats/min. The newborns were lying in their mothers’ arms in the parents’ room on the ward. The environment was calm and the investigator was able to accomplish the venipuncture by gently holding the baby’s hand. Few infants cried during this procedure and mean HR in the group that received no analgesic was 144 beats/min after venipuncture. The procedure surrounding the immunization in the present study probably provided more external stimuli to the baby. A 3-month-old baby perceives the difference between being at home and being at the clinic, and some parents are tense when it comes to the first vaccination. Along with preceding activities such as weighing and examinations, an environment may be created where the baby does not feel safe and calm. This might be a hindrance to pain management, and optimizing environmental conditions represents a challenge. Hypotonic – hyporesponsive episodes (HHE) are described as adverse events after immunizations. Some of these reported events, occurring within several minutes after the injection, are probably episodes of vasovagal syncope (DuVernoy and Braun, 2000). The experienced pain during and after the vaccination, taken together with potential fear and tension before the injection, may cause the vasovagal response. This seems to be an additional reason to prevent the pain of injections. The two-group design of this study does not make it possible to differentiate whether the analgesic effect was due to EMLA or glucose. A four-group design, adding one glucose – placebo group and one water – EMLA group, would have given an answer to that question. There were several reasons why the two-group design was chosen. During the planning of the study, there were studies already
V. Lindh et al. / Pain 104 (2003) 381–388
performed on either EMLA or sweet-tasting solutions for vaccination pain and the combined use had proven more effective than separate use for circumcision pain. Our intention was to use “the best proven treatment” and evaluate the effect during the vaccination procedure. The ethical issue of choosing a study design with a nontreatment group was discussed in the light of the prevailing practice which still is to perform vaccinations without analgesia. The infants in the placebo group were not subjected to extra harm as compared with infants not participating in the study (WMA, 2002). Furthermore, one of the methodological intentions was to visualize the HRV pattern during the placebo procedure and the vaccination where analgesics were used. A limitation of the behavioural data is that the VASrater could not be fully blinded to the vasoconstrictive effect of EMLA causing paleness on the application site in some infants. However observer bias is completely avoided in the analyses of the HR and HRV data. Methods to avoid or alleviate vaccination pain have not been extensively developed and evaluated. The parents’ VAS ratings may well be used as an easily obtained behavioural measure when assessing vaccination pain in future studies. However, behavioural and physiological responses do not always occur in parallel and they present different dimensions of the pain response. A robust behavioural response might only tell us that the baby is healthy and alert enough to display the discomfort as compared with a sick or drowsy baby showing a modest response to the same stimulus (Barr, 1998). Infants born preterm are shown to be less behaviourally reactive and to have higher HR during a heel prick as compared with newly born term infants (Johnston and Stevens, 1996). The ability to mount a response to pain or other challenges in terms of HR and HRV might also be reduced in already distressed or immature infants (Lindh, 2002). In addition, there are individual differences where some infants tend to show a high behavioural and low physiologic reactivity or the opposite reaction (Morison et al., 2001). There is a need for precise measures in order to further evaluate the clinical significance of these characteristics. Our present aim is to assess the infants’ autonomous responses “bedside” by means of a real time analysis of HRV data. The magnitude of the biphasic HR response, measured as an area of a polygon, might be one of the indicators of the infants’ pain response that can be instantly evaluated. In summary, our results show that the infants receiving pre-treatment with EMLA and glucose had significantly lower pain scores, fewer infants cried, the crying started later, and the biphasic HR responses were less pronounced compared to the placebo group. EMLA and glucose could be used with advantage to alleviate vaccination pain in 3month-old infants.
387
References Barr RG. Reflections on measuring pain in infants: dissociation in responsive systems and “honest signalling”. Arch Dis Child Fetal Neonatal Ed 1998;79:F152–6. Barr RG, Young SN, Wright JH, Cassidy KL, Hendricks L, Bedard Y, Yaremko J, Leduc D, Treherne S. “Sucrose analgesia” and diphtheria – tetanus–pertussis immunizations at 2 and 4 months. J Dev Behav Pediatr 1995;16:220 –5. Brennan M, Palaniswami M, Kamen P. Do existing measures of Poincare´ plot geometry reflect nonlinear features of heart rate variability. IEEE Trans Biomed Eng 2001;48:1342–7. Carbajal R, Chauvet X, Couderc S, Olivier-Martin M. Randomised trial of analgesic effects of sucrose, glucose, and pacifiers in term neonates. Br Med J 1999;319:1393– 7. Datta A, Pal S, Pal NR. A connectionist model for convex-hull of a planar set. Neural Networks 2000;13:377–84. Dohlwitz A, Hellenberg L, Svedmyr J, Tober L, Wigertz A. No negative influence of EMLA application prior to BCG vaccination. Acta Paediatr 1998;87:480–1. DuVernoy TS, Braun MM. Hypotonic –hyporesponsive episodes reported to the Vaccine Adverse Event Reporting System (VAERS), 1996– 1998. Pediatrics 2000;106:E52. Eriksson M, Gradin M, Schollin J. Oral glucose and venipuncture reduce blood sampling pain in newborns. Early Hum Dev 1999;55:211–8. Finley A, McGrath P. Measurement of pain in infants and children. Prog Pain Res Manage 1998;10:83–102. Guala A, Pastore G, Liverani ME, Giroletti G, Gulino E, Meriggi AL, Licardi G, Garipoli V. Glucose or sucrose as an analgesic for newborns: a randomised controlled blind trial. Minerva Pediatr 2001;53:271–4. Gunnar MR, Porter FL, Wolf CM, Rigatuso J, Larson MC. Neonatal stress reactivity: predictions to later emotional temperament. Child Dev 1995; 66:1–13. Halperin SA, McGrath P, Smith B, Houston T. Lidocaine–prilocaine patch decreases the pain associated with the subcutaneous administration of measles–mumps–rubella vaccine but does not adversely affect the antibody response. J Pediatr 2000;136:789–94. Harrington C, Kirjavainen T, Teng A, Sullivan CE. Cardiovascular responses to three simple, provocative tests of autonomic activity in sleeping infants. J Appl Physiol 2001;91:561–8. Himelstein BP, Cnaan A, Blackall CS, Zhao H, Cavalieri G, Cohen DE. Topical application of lidocaine–prilocaine (EMLA) cream reduces the pain of intramuscular infiltration of saline solution. J Pediatr 1996;129: 718– 21. Jacobson RM, Swan A, Adegbenro A, Ludington SL, Wollan PC, Poland GA. Making vaccines more acceptable – methods to prevent and minimize pain and other common adverse events associated with vaccines. Vaccine 2001;19:2418 –27. Johnston CC, Stevens BJ. Experience in a neonatal intensive care unit affects pain response. Pediatrics 1996;98:925– 30. Johnston CC, Strada ME. Acute pain response in infants: a multidimensional description. Pain 1986;24:373–82. Kawa A, Stahlhut M, Berezin A, Bock E, Berezin V. A simple procedure for morphometric analysis of processes and growth cones of neurons in culture using parameters derived from the contour and convex hull of the object. J Neurosci Methods 1998;79:53–64. Lagercrantz H, Edner A, Runold M, Yuan S-Z. Development of autonomic control in infants. In: Rognum TO, editor. Sudden infant death syndrome. New trends in the nineties. Oslo: Scandinavium University Press, 1995. p. 214 –7. Lewindon PJ, Harkness L, Lewindon N. Randomised controlled trial of sucrose by mouth for the relief of infant crying after immunisation. Arch Dis Child 1998;78:453– 6. Lindh V. Procedural pain and distress in infants. Alleviation of acute pain assessed by heart rate variability and behavioural measures. Medical
388
V. Lindh et al. / Pain 104 (2003) 381–388
Dissertation, Umea˚ , Sweden: Department of Clinical Sciences, Pediatrics, Umea˚ University, 2002. Lindh V, Wiklund U, Sandman PO, Hakansson S. Assessment of acute pain in preterm infants by evaluation of facial expression and frequency domain analysis of heart rate variability. Early Hum Dev 1997;48: 131– 42. Lindh V, Wiklund U, Hakansson S. Heel lancing in term new-born infants: an evaluation of pain by frequency domain analysis of heart rate variability. Pain 1999;80:143 –8. Lindh V, Wiklund U, Hakansson S. Assessment of the effect of EMLA during venipuncture in the newborn by analysis of heart rate variability. Pain 2000;86:247 –54. Mohan CG, Risucci DA, Casimir M, Gulrajani-LaCorte M. Comparison of analgesics in ameliorating the pain of circumcision. J Perinatol 1998; 18:13–19. Morison SJ, Grunau RE, Oberlander TF, Whitfield MF. Relations between behavioral and cardiac autonomic reactivity to acute pain in preterm neonates. Clin J Pain 2001;17:350–8. Norusis MJ. SPSS 7.5 guide to data analysis. Engelwood Cliffs, NJ: Prentice-Hall, 1997. Otzenberger H, Gronfier C, Simon C, Charloux A, Ehrhart J, Piquard F, Brandenberger G. Dynamic heart rate variability: a tool for exploring sympathovagal balance continuously during sleep in men. Am J Physiol 1998;275:H946 –50. Porter FL, Porges SW, Marshall RE. Newborn pain cries and vagal tone: parallel changes in response to circumcision. Child Dev 1988;59: 495– 505. Ramenghi LA, Webb AV, Shevlin PM, Green M, Evans DJ, Levene MI. Intra-oral administration of sweet-tasting substances and infants’ crying
response to immunization: a randomized, placebo-controlled trial. Biol Neonate 2002;81:163–9. Skogsdal Y, Eriksson M, Schollin J. Analgesia in newborns given oral glucose. Acta Paediatr 1997;86:217–20. Stevens B, Ohlsson A. Sucrose for analgesia in newborn infants undergoing painful procedures (Cochrane review). The Cochrane Library, Issue 1. Oxford: Update Software, 2000. Taddio A, Nulman I, Reid E, Shaw J, Koren BS. Effect of lidocaine– prilocaine cream (EMLA) on pain of intramuscular fluzone injection. Can J Hosp Pharm 1992;45:227 –30. Taddio A, Nulman I, Goldbach M, Ipp M, Koren G. Use of lidocaine– prilocaine cream for vaccination pain in infants. J Pediatr 1994;124: 643 –8. Task Force of the European Society of Cardiology and North American Society of Pacing and Electrophysiology, Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Circulation 1996;93:1043–65. Uhari M. A eutectic mixture of lidocaine and prilocaine for alleviating vaccination pain in infants. Pediatrics 1993;92:719–21. Wiklund U, Akay M, Niklasson U. Short-term analysis of heart-rate variability by adapted wavelet transforms. IEEE Eng Med Biol Mag 1997;16:113–8. (see also p. 138). WMA. The Declaration of Helsinki. Ethical principles for medical research involving human subjects. Note of clarification on paragraph 29. World Medical Association, General Assembly Washington 2002; 2002. Woo MA, Stevenson WG, Moser DK, Trelease RB, Harper RM. Patterns of beat-to-beat heart rate variability in advanced heart failure. Am Heart J 1992;123:704–10.
EMLAw cream and oral glucose for immunization pain in 3-month-old infants Viveca Lindha,*, Urban Wiklundb, Hans K. Blomquista, Stellan Ha˚kanssona a
Department of Clinical Sciences, Pediatrics, University Hospital, Umea˚ University, S-901 85 Umea˚, Sweden b Department of Biomedical Engineering and Informatics, University Hospital, S-901 85 Umea˚, Sweden Received 26 March 2002; received in revised form 21 January 2003; accepted 27 January 2003
Abstract The objective of this study is to determine whether use of lidocaine – prilocaine 5% cream (EMLAw) and oral glucose decreases pain associated with diphteria – pertussis – tetanus (DPT) immunization in 3-month-old infants. Design: randomized, double-blind, controlled trial in outpatient paediatric practice in northern Sweden. EMLA or placebo was applied to the infant’s lateral region of the right thigh and covered with an occlusive dressing 1 h before the immunization. In addition, 1 ml of glucose (300 mg/ml) or placebo (water) was instilled on the baby’s tongue within 2 min before the DPT-injection. Forty-five infants received EMLA and glucose and 45 infants placebo cream and water. ECG was recorded and stored in a computer and the procedure was videotaped. The parents and the nurse assessed the infants’ pain on a visual analogue scale (VAS) after the immunization. Heart rate and heart rate variability pre- and post-injection were calculated. From the videotapes, the modified behavioural pain scale (MBPS) was used to assess pain scores during baseline and after immunization. The latency of the first cry and total crying time were measured. The parents and the nurse scored the infants’ pain on the VAS significantly lower in the treatment group than in the placebo group. The infants’ responses to the immunization measured as the difference in MBPS scores pre- and post-injection were significantly lower in the EMLA –glucose group compared with the placebo group. More infants cried after the immunization in the placebo group compared with the EMLA – glucose group and the latency of the first cry after the injection was shorter in the placebo group. A biphasic transient heart rate response with a marked deceleration followed by a subsequent acceleration was seen more frequently in the placebo group compared to the EMLA – glucose group. EMLA and glucose alleviate immunization pain in 3-month-old infants. q 2003 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Lidocaine–prilocaine; Topical; Heart rate variability; Modified behavioural pain scale; Crying
1. Introduction With the intention of inflicting as little procedural pain as possible, other routes of administration than the subcutaneous (s.c.) and intra-muscular (i.m.) injections are preferred in paediatric care. For the reason of best possible immune response to vaccine antigens, however, it is at present not possible to avoid the s.c. and i.m. routes in childhood immunizations. In order to meet stricter requirements for gentle handling, it is important to evaluate various pain-reducing strategies (Jacobson et al., 2001). Topical lidocaine –prilocaine patch (EMLAw) has no adverse effect on the antibody response of immunizations (Dohlwitz et al., 1998; Halperin et al., 2000). In adults, the * Corresponding author. Tel.: þ 46-90-785-2216; fax: þ 46-90-12-37-28. E-mail address: [email protected] (V. Lindh).
lidocaine – prilocaine cream decreases the pain from needle puncture and i.m. infiltration by approximately 50% (Himelstein et al., 1996; Taddio et al., 1992). EMLA also reduces injection pain in infants receiving their first i.m. DPT (diphtheria – pertussis –tetanus) immunization (Uhari, 1993; Taddio et al., 1994) and in 12-month-old infants undergoing s.c. MMR (measles – mumps – rubella) immunization (Halperin et al., 2000). Sweet-tasting solutions alleviate pain in newborn infants undergoing noxious procedures (Eriksson et al., 1999; Skogsdal et al., 1997; Stevens and Ohlsson, 2000). Glucose has at least the same analgesic effect as sucrose (Carbajal et al., 1999; Guala et al., 2001). The analgesic effect continues beyond the newborn period and shortens crying time after immunizations in 2-, 4- and 6-month-old infants (Barr et al., 1995; Lewindon et al., 1998; Ramenghi et al.,
0304-3959/03/$30.00 q 2003 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0304-3959(03)00046-0
382
V. Lindh et al. / Pain 104 (2003) 381–388
2002). The analgesic effect of saccharide combined with EMLA has shown to reduce pain during circumcision more effective than saccharide alone (Mohan et al., 1998). Hitherto this combination has not been tested for vaccination pain. Behavioural measures such as crying, the latency period until the first cry, the visual analogue scale (VAS), and the modified behavioural pain scale (MBPS) have been used to assess immunization pain in infants (Uhari, 1993; Taddio et al., 1994; Barr et al., 1995; Lewindon et al., 1998; Finley and McGrath, 1998; Halperin et al., 2000). An increased heart rate (HR) response to vaccination has also been described (Johnston and Strada, 1986). In our previous studies, the autonomic modulation of HR in response to heel lancing and venipuncture was analysed (Lindh et al., 1997, 1999, 2000), and was found to be a useful tool for objective assessment of pain. Spectral analysis of heart rate variability (HRV) was used to determine the frequency content of the fluctuating HR. The power of the high-frequency component (above 0.15 Hz) mirrors parasympathetic activity, i.e. vagal tone and reflects instant responses to changes in autonomic balance. The slower fluctuations are mediated by both sympathetic and parasympathetic activity (Task Force, 1996). The aim of this study was to examine whether topical application of EMLA combined with oral glucose 300 mg/ml decreases the pain associated with i.m. DPT immunization in three-month-old infants when assessed by HRV and behavioural scores.
2. Method 2.1. Materials The investigator or the nurse in charge provided verbal and written information to the parents before the first DPT vaccination and verbal parental consent was obtained. A total of 90 infants, 3 months of age, who were receiving their regular DPT vaccination were recruited for the study. These infants were born at term, were of normal birth weight, were otherwise healthy and required only routine post-natal care on the maternity ward.
envelopes labelled with the subject’s study number. The list of random numbers was not opened until all the analyses were completed. 2.3. Procedure The investigator saw the family at the Well Baby Clinic when the baby was scheduled for the first DPT vaccination. One gram of EMLAw or placebo was applied to the lateral region of the right thigh. The cream was covered with an occlusive dressing (Tegadermw, 3 M) for 1 h. Three electrocardiogram (ECG) electrodes (Blue Sensor, Medicotestw) were attached to the baby’s chest. The parents were instructed to feed the baby within a period of 1 h before the immunization. The procedure was performed in a separate, quiet room. All vaccination procedures were performed by the same investigator (V.L.). The mother ðn ¼ 84Þ or father ðn ¼ 6Þ held the baby during the vaccination and was told to comfort the baby as usual. The procedure was videotaped, including sound recordings, and the ECG signal was sampled and stored in a computer (BioBench, National Instruments, Austin, TX). The recordings were divided into three sequences: (a) a 3-min baseline period when the baby was held in the parent’s lap; (b) when the investigator (V.L.) instilled 1 ml of glucose (300 mg/ml) or water on the baby’s tongue with a syringe, and wiped the lateral aspect of the baby’s thigh with alcohol and (c) the i.m. administration of 0.5 ml of Infanrixw (SmithKline Beecham Pharma) using a 2.5 cm 23-gauge needle. The video and ECG recordings were continued until the baby had settled down after the injection. In the group of infants receiving EMLA –glucose, there were 25 boys and 20 girls ðn ¼ 45Þ; and in the placebo – water group, there were 27 boys and 18 girls ðn ¼ 45Þ: Immediately after the procedure, the parents were asked to rate the infant’s experience of pain during the vaccination on a VAS-scale where zero was designated “no pain” and 10 the “worst possible pain”. For the purpose of obtaining two independent VAS-ratings, a rating of which the parents were not aware was done by the investigator before the parents gave their opinion. The study was approved by the Ethics Committee of the Medical Faculty of Umea˚ University and was performed in accordance with the Declaration of Helsinki.
2.2. Randomization 2.4. Analyses As infants were enrolled, they were consecutively assigned a subject number that served as the link to the random allocation to either the EMLA –glucose or placebo –water group. The randomization and packaging of the EMLA –glucose and placebo – water were performed by the hospital pharmacy with a computer-generated list of random numbers in balanced blocks of six. Accordingly, the baby received either EMLA and glucose or placebo and water. The EMLA patch and glucose solution or the placebo patch and water (packaged in syringes) were contained in
2.4.1. ECG The ECG was sampled at 1000 Hz and R-waves were detected as previously described (Lindh et al., 1997). All recordings were inspected and corrected for detection errors. Artefacts caused by signal saturation due to baseline wander or arrhythmic beats were replaced by interpolation. However, all segments where three or more consecutive beats were interpolated were marked and not considered for further analysis. The beat-to-beat HR was calculated and
V. Lindh et al. / Pain 104 (2003) 381–388
converted to a time series by cubic spline interpolation and resampling at 4.8 Hz. Three segments were selected for statistical analysis: (a) 45 s from the middle of the baseline recording; (b) the last 45 s before the injection and (c) 45 s starting immediately after the injection. HRmean was calculated for each segment. Spectral analysis was assessed on detrended data using autoregressive modelling of order 30. The detrending was accomplished by using a digital filter with a cut-off frequency at approximately 0.02 Hz as previously described (Wiklund et al., 1997). The total spectral power (Ptot) and the power of the low-frequency (PLF, 0.02– 0.15 Hz) and high-frequency (PHF, 0.15 – 2.0 Hz) components were calculated and log-transformed. In the context of pain, power of the high-frequency component, i.e. vagal tone, has shown to decrease in term newborns during heel lancing and circumcision (Porter et al., 1988; Gunnar et al., 1995; Lindh et al., 1999). The HR response during the first 45 s after injection was also evaluated utilizing Poincare´ graphs, i.e., plots of each R – R interval against the subsequent value (Woo et al., 1992). The geometry of this plot visualizes the nature of variability and has been shown to correspond to sympathovagal balance and different clinical conditions (Otzenberger et al., 1998; Brennan et al., 2001). A Poincare´ -plot of an HR recording in a healthy, satisfied term newborn will show a high variability and the swarm of points will appear spread in the upper right-hand corner of the plot. If the baby suddenly is subjected to a painful procedure, HR will increase and the variability will decrease. The swarm of points will gather and appear in the lower left-hand corner of the plot (Lindh, 2002). In this study, the geometric appearance of the Poincare´ -plot was quantified by calculating the area of the smallest convex polygon that enclosed all points in the plot (Datta et al., 2000; Kawa et al., 1998). The minimum and maximum values of the R – R interval during the post-injection phase were determined. 2.4.2. Video recordings The facial expression, cry and body movements of the infants were assessed by means of the MBPS (Taddio et al., 1994). Two experienced paediatric nurses, working independently of each other and blinded to the study treatment, rated three sequences: (a) baseline; (b) 0 –10 s after the injection and (c) 11– 20 s after the injection. For each infant, differences between MBPS scores after the injection and the score at baseline were calculated. Whether or not a cry was present was noted, and latency of the first cry after the injection and total crying time were measured. 2.4.3. Statistical analyses Statistical analyses of the HRV data from the three sequences were calculated using analysis of variance (ANOVA) for repeated measurements. Otherwise the ttest for independent samples was used for intergroup comparisons and the x 2-test for the frequency of infants
383
who cried after the injection in the placebo and treatment groups, respectively. Cohen’s kappa (Norusis, 1997) was used to calculate the agreement between raters of the MBPS scores. The level of two-sided statistical significance was defined as a P-value , 0.05. Based upon the increase in HR during venipuncture in newborn infants (Lindh et al., 2000), it was calculated that 45 infants in each group represented 95% statistical power (a difference between the groups of HRmean 14 beats/min SD 18). The SPSS software (SPSS, Chicago, IL) was used for statistical analysis.
3. Results 3.1. Excluded recordings One video (and audio) recording in the EMLA – glucose group was excluded because the camera accidentally stopped before the study procedure was completed. In one video recording in the placebo group and two in the treatment group, there were moments when the view of the infant was blocked, which made a complete assessment of MBPS impossible. These were excluded from the calculations of the difference between MBPS scores before and after the injection. One ECG recording was not stored on the computer and two were excluded because of frequent artefacts, two in the EMLA –glucose group and one in the placebo group. 3.2. HR and HRV HR, total variability and the power of the low-frequency component during the sequence after the injection were significantly increased in both groups compared to baseline (see Fig. 1). The power of the high-frequency component was somewhat increased in the placebo group and decreased in the EMLA – glucose group. No statistically significant differences between groups were seen, i.e. no significant effect of the interaction between the group and time factors was found (analysis of variance, repeated measurements). A biphasic transient HR response was significantly more frequent in the placebo group during the sequence after the injection, as compared with the EMLA – glucose group. These responses were characterized by a marked deceleration followed by a subsequent acceleration. The decelerations, i.e. the maximal R –R intervals, were significantly larger in the placebo group ðP ¼ 0:03Þ; and the accelerations, i.e. the minimal R – R intervals, were significantly shorter ðP ¼ 0:03Þ than in the EMLA – glucose group. The maximal R –R interval corresponding to the longest interval between heartbeats was 1.9 s in the placebo group and 0.6 s in the treatment group. A larger polygon area (. 0.017 s2) in the Poincare´ -plot occurred more frequently in the placebo group (10/44) than in the EMLA – glucose group (1/43) (x 2test P ¼ 0.007, see Fig. 2). The power of the high-frequency component after
384
V. Lindh et al. / Pain 104 (2003) 381–388
Fig. 1. Mean values and standard errors of the mean for (A) HR, (B) total variability, (C) power of the low-frequency component and (D) power of the highfrequency component during rest, the sequence when glucose or water was administered, and after the immunization. *P , 0:05 indicates comparisons between base line versus vaccination.
vaccination, as compared with baseline values, showed three distinctly different patterns: (a) a low variability as a consequence of a rapidly increasing HR; (b) a relatively unchanged variability in infants where no marked response to the immunization occurred or (c) an increased variability due to the biphasic HR response. The infants in the placebo group responded significantly more often with either low or increased high-frequency power as compared with the EMLA –glucose group (x 2-test P ¼ 0.006, see Table 1). 3.3. Crying time and behavioural pain measurements More infants cried after the vaccination in the placebo group than in the EMLA – glucose group, (x 2-test P ¼ 0:001; see Table 2). Among the infants who cried, the total Table 1 Number of infants in the EMLA–glucose and placebo–water groups with low, unchanged or high power of the high-frequency component after the immunization HF component
EMLA–glucose
Placebo–water
Total
Low (,3.0 log mHz2) Unchanged High (.3.8 log mHz2)
12 30 1
15 19 10
27 49 11
crying time after the injection was shorter in the EMLA – glucose group, although it did not reach statistical significance. The infants in the placebo group started to cry sooner after the injection than the infants in the treatment group P , 0:001: Total crying time and latency of cry are given in Table 3. The parents and the investigator rated the infants’ pain on the VAS significantly lower in the treatment group than in the placebo group (VAS-parents P ¼ 0:001; VAS-nurse P , 0:001Þ: The agreement between the two MBPS raters, calculated with Cohen’s kappa, was 1.0 for baseline and 0.5 for the assessments after injection, which means that the agreement exceeded chance levels. In cases where the ratings diverged, Table 2 Number of infants who cried or did not cry after the immunization: comparison between the EMLA–glucose and placebo–water groups EMLA–glucose Placebo–water Total Infants (n) who cried after the 32 injection Infants (n) who did not cry 12 after the injection Total 44
44
76
1
13
45
89
V. Lindh et al. / Pain 104 (2003) 381–388
385
Fig. 2. HR recordings in three infants during the complete study procedure, followed by Poincare´ -plots for the sequences after the vaccination with the corresponding enclosing polygon outlined: (A) illustrates an unchanged HR after the vaccination, (B) an increased HR response and (C) a biphasic HR response.
the mean values of the two ratings for each sequence were used. The infants’ responses to the immunization, measured as the difference in MBPS scorings pre- and post-injection, were significantly lower in the EMLA –glucose group than in the placebo group (difference between baseline and 0 – 10 s after the injection, P , 0:001; difference between baseline and 11 –20 s, P ¼ 0:004Þ: Means and standard
deviations for the VAS, the MBPS, duration of time until crying, and total crying time are shown in Table 3.
4. Discussion The main result of this study was that the combined use of EMLA and glucose alleviates the vaccination pain
386
V. Lindh et al. / Pain 104 (2003) 381–388
Table 3 VAS and MBPS scores, latency of cry and total crying time after the vaccination: comparison between the EMLA–glucose and placebo–water groups
VAS parents VAS nurse MBPS baseline MBPS 0–10 s after injection MBPS 11–20 s after injection Diff baseline–MBPS 0–10 s after injection Diff baseline–MBPS 11–20 s after injection Latency of cry after injection (s) Total crying time (s)
EMLA–glucose
Placebo– water
3 ^ 2* 2 ^ 2* 1.9 ^ 0.4 5.5 ^ 2.0* 5.4 ^ 2.4* 3.6 ^ 2.1* 3.5 ^ 2.3* 6.4 ^ 3.2* 41 ^ 44
5^2 5^2 1.8 ^ 0.4 7.7 ^ 1.7 6.8 ^ 2.2 5.9 ^ 1.7 5 ^ 2.3 3.8 ^ 2.3 50 ^ 52
*P , 0:05:
in 3-month-old infants. The behavioural pain responses clearly showed a significant reduction of the pain response in the treatment group as compared with the placebo group. The mean values of the HR and HRV data per se did not discriminate the pain responses in the EMLA – glucose group from the placebo group. However, a marked biphasic HR response was significantly more common in the placebo group. The response to the injection was easily observed and the behavioural parameters clearly indicated a decreased pain response in the treatment group. Fewer infants cried in the EMLA – glucose group and among those who cried the latency of the first cry was prolonged as compared with the infants receiving placebo – water. However, the pain was not completely alleviated. There was an increase in MBPS scores after the injection as compared to baseline and 36/44 infants cried even if EMLA – glucose was applied. The VAS ratings made by the investigator and the parents correlated well to the MBPS scores obtained from video analyses. The biphasic or increasing HR responses after the injection made it difficult to draw conclusions from the mean values of the data. Consequently, the data were best described in terms of the different response patterns. This was also true for the power of the high-frequency component. The range was wider in the placebo group, implying that the injection without analgesia influenced vagal tone more than it did in the EMLA – glucose group. There were no clear correlations between the VAS scorings and the mean values of the HRV data due to these different response patterns. However, nine of the 11 infants with the most pronounced biphasic HR response were scored above VAS 5 by the parents. The observed biphasic HR response immediately after immunization has been described by Johnston and Strada (1986), and heel lancing in term newborns caused a similar but weaker response (Lindh et al., 1999). This response pattern is associated with the “fear paralysis” – reflex, characterized by a sympathetic inhibition together with a vagal bradycardia. It can be
triggered by a sudden noise, pain or by restraining the infant (Lagercrantz et al., 1995). A high pre-injection mean HR (146 beats/min) was seen in the infants in this study. HR during sleep is approximately 120 beats/min at this age (Harrington et al., 2001). In a previous study (Lindh et al., 2000), the baseline mean HR in newborns before metabolic screening in the maternity ward was 122 beats/min. The newborns were lying in their mothers’ arms in the parents’ room on the ward. The environment was calm and the investigator was able to accomplish the venipuncture by gently holding the baby’s hand. Few infants cried during this procedure and mean HR in the group that received no analgesic was 144 beats/min after venipuncture. The procedure surrounding the immunization in the present study probably provided more external stimuli to the baby. A 3-month-old baby perceives the difference between being at home and being at the clinic, and some parents are tense when it comes to the first vaccination. Along with preceding activities such as weighing and examinations, an environment may be created where the baby does not feel safe and calm. This might be a hindrance to pain management, and optimizing environmental conditions represents a challenge. Hypotonic – hyporesponsive episodes (HHE) are described as adverse events after immunizations. Some of these reported events, occurring within several minutes after the injection, are probably episodes of vasovagal syncope (DuVernoy and Braun, 2000). The experienced pain during and after the vaccination, taken together with potential fear and tension before the injection, may cause the vasovagal response. This seems to be an additional reason to prevent the pain of injections. The two-group design of this study does not make it possible to differentiate whether the analgesic effect was due to EMLA or glucose. A four-group design, adding one glucose – placebo group and one water – EMLA group, would have given an answer to that question. There were several reasons why the two-group design was chosen. During the planning of the study, there were studies already
V. Lindh et al. / Pain 104 (2003) 381–388
performed on either EMLA or sweet-tasting solutions for vaccination pain and the combined use had proven more effective than separate use for circumcision pain. Our intention was to use “the best proven treatment” and evaluate the effect during the vaccination procedure. The ethical issue of choosing a study design with a nontreatment group was discussed in the light of the prevailing practice which still is to perform vaccinations without analgesia. The infants in the placebo group were not subjected to extra harm as compared with infants not participating in the study (WMA, 2002). Furthermore, one of the methodological intentions was to visualize the HRV pattern during the placebo procedure and the vaccination where analgesics were used. A limitation of the behavioural data is that the VASrater could not be fully blinded to the vasoconstrictive effect of EMLA causing paleness on the application site in some infants. However observer bias is completely avoided in the analyses of the HR and HRV data. Methods to avoid or alleviate vaccination pain have not been extensively developed and evaluated. The parents’ VAS ratings may well be used as an easily obtained behavioural measure when assessing vaccination pain in future studies. However, behavioural and physiological responses do not always occur in parallel and they present different dimensions of the pain response. A robust behavioural response might only tell us that the baby is healthy and alert enough to display the discomfort as compared with a sick or drowsy baby showing a modest response to the same stimulus (Barr, 1998). Infants born preterm are shown to be less behaviourally reactive and to have higher HR during a heel prick as compared with newly born term infants (Johnston and Stevens, 1996). The ability to mount a response to pain or other challenges in terms of HR and HRV might also be reduced in already distressed or immature infants (Lindh, 2002). In addition, there are individual differences where some infants tend to show a high behavioural and low physiologic reactivity or the opposite reaction (Morison et al., 2001). There is a need for precise measures in order to further evaluate the clinical significance of these characteristics. Our present aim is to assess the infants’ autonomous responses “bedside” by means of a real time analysis of HRV data. The magnitude of the biphasic HR response, measured as an area of a polygon, might be one of the indicators of the infants’ pain response that can be instantly evaluated. In summary, our results show that the infants receiving pre-treatment with EMLA and glucose had significantly lower pain scores, fewer infants cried, the crying started later, and the biphasic HR responses were less pronounced compared to the placebo group. EMLA and glucose could be used with advantage to alleviate vaccination pain in 3month-old infants.
387
References Barr RG. Reflections on measuring pain in infants: dissociation in responsive systems and “honest signalling”. Arch Dis Child Fetal Neonatal Ed 1998;79:F152–6. Barr RG, Young SN, Wright JH, Cassidy KL, Hendricks L, Bedard Y, Yaremko J, Leduc D, Treherne S. “Sucrose analgesia” and diphtheria – tetanus–pertussis immunizations at 2 and 4 months. J Dev Behav Pediatr 1995;16:220 –5. Brennan M, Palaniswami M, Kamen P. Do existing measures of Poincare´ plot geometry reflect nonlinear features of heart rate variability. IEEE Trans Biomed Eng 2001;48:1342–7. Carbajal R, Chauvet X, Couderc S, Olivier-Martin M. Randomised trial of analgesic effects of sucrose, glucose, and pacifiers in term neonates. Br Med J 1999;319:1393– 7. Datta A, Pal S, Pal NR. A connectionist model for convex-hull of a planar set. Neural Networks 2000;13:377–84. Dohlwitz A, Hellenberg L, Svedmyr J, Tober L, Wigertz A. No negative influence of EMLA application prior to BCG vaccination. Acta Paediatr 1998;87:480–1. DuVernoy TS, Braun MM. Hypotonic –hyporesponsive episodes reported to the Vaccine Adverse Event Reporting System (VAERS), 1996– 1998. Pediatrics 2000;106:E52. Eriksson M, Gradin M, Schollin J. Oral glucose and venipuncture reduce blood sampling pain in newborns. Early Hum Dev 1999;55:211–8. Finley A, McGrath P. Measurement of pain in infants and children. Prog Pain Res Manage 1998;10:83–102. Guala A, Pastore G, Liverani ME, Giroletti G, Gulino E, Meriggi AL, Licardi G, Garipoli V. Glucose or sucrose as an analgesic for newborns: a randomised controlled blind trial. Minerva Pediatr 2001;53:271–4. Gunnar MR, Porter FL, Wolf CM, Rigatuso J, Larson MC. Neonatal stress reactivity: predictions to later emotional temperament. Child Dev 1995; 66:1–13. Halperin SA, McGrath P, Smith B, Houston T. Lidocaine–prilocaine patch decreases the pain associated with the subcutaneous administration of measles–mumps–rubella vaccine but does not adversely affect the antibody response. J Pediatr 2000;136:789–94. Harrington C, Kirjavainen T, Teng A, Sullivan CE. Cardiovascular responses to three simple, provocative tests of autonomic activity in sleeping infants. J Appl Physiol 2001;91:561–8. Himelstein BP, Cnaan A, Blackall CS, Zhao H, Cavalieri G, Cohen DE. Topical application of lidocaine–prilocaine (EMLA) cream reduces the pain of intramuscular infiltration of saline solution. J Pediatr 1996;129: 718– 21. Jacobson RM, Swan A, Adegbenro A, Ludington SL, Wollan PC, Poland GA. Making vaccines more acceptable – methods to prevent and minimize pain and other common adverse events associated with vaccines. Vaccine 2001;19:2418 –27. Johnston CC, Stevens BJ. Experience in a neonatal intensive care unit affects pain response. Pediatrics 1996;98:925– 30. Johnston CC, Strada ME. Acute pain response in infants: a multidimensional description. Pain 1986;24:373–82. Kawa A, Stahlhut M, Berezin A, Bock E, Berezin V. A simple procedure for morphometric analysis of processes and growth cones of neurons in culture using parameters derived from the contour and convex hull of the object. J Neurosci Methods 1998;79:53–64. Lagercrantz H, Edner A, Runold M, Yuan S-Z. Development of autonomic control in infants. In: Rognum TO, editor. Sudden infant death syndrome. New trends in the nineties. Oslo: Scandinavium University Press, 1995. p. 214 –7. Lewindon PJ, Harkness L, Lewindon N. Randomised controlled trial of sucrose by mouth for the relief of infant crying after immunisation. Arch Dis Child 1998;78:453– 6. Lindh V. Procedural pain and distress in infants. Alleviation of acute pain assessed by heart rate variability and behavioural measures. Medical
388
V. Lindh et al. / Pain 104 (2003) 381–388
Dissertation, Umea˚ , Sweden: Department of Clinical Sciences, Pediatrics, Umea˚ University, 2002. Lindh V, Wiklund U, Sandman PO, Hakansson S. Assessment of acute pain in preterm infants by evaluation of facial expression and frequency domain analysis of heart rate variability. Early Hum Dev 1997;48: 131– 42. Lindh V, Wiklund U, Hakansson S. Heel lancing in term new-born infants: an evaluation of pain by frequency domain analysis of heart rate variability. Pain 1999;80:143 –8. Lindh V, Wiklund U, Hakansson S. Assessment of the effect of EMLA during venipuncture in the newborn by analysis of heart rate variability. Pain 2000;86:247 –54. Mohan CG, Risucci DA, Casimir M, Gulrajani-LaCorte M. Comparison of analgesics in ameliorating the pain of circumcision. J Perinatol 1998; 18:13–19. Morison SJ, Grunau RE, Oberlander TF, Whitfield MF. Relations between behavioral and cardiac autonomic reactivity to acute pain in preterm neonates. Clin J Pain 2001;17:350–8. Norusis MJ. SPSS 7.5 guide to data analysis. Engelwood Cliffs, NJ: Prentice-Hall, 1997. Otzenberger H, Gronfier C, Simon C, Charloux A, Ehrhart J, Piquard F, Brandenberger G. Dynamic heart rate variability: a tool for exploring sympathovagal balance continuously during sleep in men. Am J Physiol 1998;275:H946 –50. Porter FL, Porges SW, Marshall RE. Newborn pain cries and vagal tone: parallel changes in response to circumcision. Child Dev 1988;59: 495– 505. Ramenghi LA, Webb AV, Shevlin PM, Green M, Evans DJ, Levene MI. Intra-oral administration of sweet-tasting substances and infants’ crying
response to immunization: a randomized, placebo-controlled trial. Biol Neonate 2002;81:163–9. Skogsdal Y, Eriksson M, Schollin J. Analgesia in newborns given oral glucose. Acta Paediatr 1997;86:217–20. Stevens B, Ohlsson A. Sucrose for analgesia in newborn infants undergoing painful procedures (Cochrane review). The Cochrane Library, Issue 1. Oxford: Update Software, 2000. Taddio A, Nulman I, Reid E, Shaw J, Koren BS. Effect of lidocaine– prilocaine cream (EMLA) on pain of intramuscular fluzone injection. Can J Hosp Pharm 1992;45:227 –30. Taddio A, Nulman I, Goldbach M, Ipp M, Koren G. Use of lidocaine– prilocaine cream for vaccination pain in infants. J Pediatr 1994;124: 643 –8. Task Force of the European Society of Cardiology and North American Society of Pacing and Electrophysiology, Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Circulation 1996;93:1043–65. Uhari M. A eutectic mixture of lidocaine and prilocaine for alleviating vaccination pain in infants. Pediatrics 1993;92:719–21. Wiklund U, Akay M, Niklasson U. Short-term analysis of heart-rate variability by adapted wavelet transforms. IEEE Eng Med Biol Mag 1997;16:113–8. (see also p. 138). WMA. The Declaration of Helsinki. Ethical principles for medical research involving human subjects. Note of clarification on paragraph 29. World Medical Association, General Assembly Washington 2002; 2002. Woo MA, Stevenson WG, Moser DK, Trelease RB, Harper RM. Patterns of beat-to-beat heart rate variability in advanced heart failure. Am Heart J 1992;123:704–10.