- Email: [email protected]
Finger position for chest compressions in cardiac arrest in infants
Resuscitation 44 (2000) 43 – 46 www.elsevier.com/locate/resuscitation
Finger position for chest compressions in cardiac arrest in infants Fiona Clements a,*, John McGowan b a b
Royal Hospital for Sick Children, Yorkhill, Glasgow G3 8SJ, UK Southern General Hospital, 1345 Go6an Rd., Glasgow G51 4TF, UK
Received 26 July 1999; received in revised form 1 December 1999; accepted 7 December 1999
Abstract Objecti6e: To determine whether the recommended method of locating finger position for chest compression in infant cardiac arrest can cause pressure on the abdomen or xiphisternum. Design: The length from the inter-nipple line to the xiphisternum was calculated in 30 infants. These lengths were compared with the finger position achieved by 30 adults, using the recommended method, on templates of infant chests. Results: The mean infant lower sternal length was 2.3 cm (95% CI 1.6). The mean distance covered by the adults fingers was 4.4 cm (95% CI 0.9). Conclusion: If any infant in this study had chest compressions performed by any of the adults, using the recommended method, pressure would be exerted on the xiphisternum or abdomen. We suggest changing the method of locating finger position, to one using sternal anatomy. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: BLS; Cardiac arrest; Infant; Chest compression; Paediatric resuscitation
1. Introduction The current recommendations of both the European Resuscitation Council and the American Heart Association for determining the finger or hand position during external chest compressions, are the same [1,2]. In adults, the rescuer’s hands are placed two finger breadths above the lower sternal edge. In children older than 1 year, one hand is positioned on the lower half of the sternum. However, in infants, the position is located by placing two fingers on the sternum, one finger’s breadth below the inter-nipple line. The rationale for selecting this method of locating finger position in infants is unclear from the literature. It has been observed that compression of the xiphoid process or abdomen, during chest compressions can rupture the liver, Thaler reported two cases [3,4]. The European Resuscitation Council does not recommend abdominal thrusts in a choking infant, as this may rupture the abdomi* Corresponding author. Tel.: +44-141-2010602.
nal viscera [1]. We have investigated whether the described method of determining finger position could cause pressure on the abdomen or the xiphoid process.
2. Materials and methods Thirty infants, under the age of 1 year had the length of their sternum, from the sternal notch to the start of the xiphisternum, and the distance from their sternal notch to the inter-nipple line, measured (Fig. 1). In addition, the infants sex, date of birth, age when measured, length, weight and head circumference were recorded. It proved easier to calculate the length of the sternum from the inter-nipple line to the xiphisternum by subtraction, than to measure it directly, as the lower edge of the sternal notch is an easily located landmark. Infants were excluded from the study if they were born pre-term or if they had thoracic, sternal or spinal deformities, including surgery. Ethical
0300-9572/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 0 - 9 5 7 2 ( 9 9 ) 0 0 1 6 5 - 3
F. Clements, J. McGowan / Resuscitation 44 (2000) 43–46
44
3. Results The results are summarised in Tables 1 and 2 and Fig. 2.
4. Discussion
Fig. 1. Lengths measured on the infants chests. Length x, lower edge of sternal notch to inter-nipple line. Length y, lower edge of sternal notch to xiphisternum.
approval had been granted and parental consent was obtained before each measurement. Thirty adults were asked to place their fingers on a paper template, using the guideline recommended above. The only markings on the template were two nipples and a horizontal mid-line. The position of the lower edge of their lower finger was marked on the template. The distance from the inter-nipple line to the edge of the lower finger on the template was measured, using the same tape measure used to measure the infants. The adults sex and height were also recorded. A z-test for unmatched samples was used to compare the mean lengths.
It has been shown that the heart lies under the lower third of the infant sternum [5,6]. More importantly, it has been observed that better arterial pressures are generated, in infant cardiac arrest, when chest compressions are performed on the lower third than on the middle third of the sternum [7]. The method described by the European Resuscitation Council and the American Heart Association to determine finger position, relies on the anatomical relationship between the nipples and the lower sternum. It has been noted, however, that in infants, the position of the nipples does not predict the position of the heart [6]. From Tables 1 and 2 and Fig. 1, it can be seen that the length covered by three fingers of an adult exceeds the distance from the nipple-line to the xiphisternum. No infant in this series had a lower sternal length of 4 cm; all were shorter than this. The shortest length covered by the adults fingers was 4 cm. If these adults used the guideline for finger position, they would be pressing on the xiphisternum or upper abdomen of all of these infants.
Table 1 Infant sternal lengths, from inter-nipple line to xiphisternum Sample (weeks)
Mean age (weeks) (+S.D.)
n
0–4 4–12 12–26 26–52 All ages
4.1 6.4 21.1 46.4 21.6
7 2.6 5 1.7 9 2.4 9 2.1 30 2.3
days (3.3) (2.3) (6.7) (4.2) (18.7)
Mean length (cm)
95% confidence interval
Range (cm)
1.4 1.6 1.6 1.2 1.6
2–3.5 1–3 1–3.5 1.5–3 1–3.5
Table 2 Measurements from adults Measurement
n
Mean length
Length, from internipple line, to lower edge of adult fingers Adult heights
30 4.4 cm 30 1.67 m
95% confidence interval
Range (cm)
0.9 0.2
4.0–5.5 1.47–1.9
F. Clements, J. McGowan / Resuscitation 44 (2000) 43–46
45
Fig. 2. Line A, length from infants inter-nipple line to xiphoid. Line B: length from inter-nipple line to lower edge of adults fingers (mean values with 95% CI). z test for unmatched samples, z=11.85; P=0.01.
The predictive value of a zero numerator has been described elsewhere [8]. To summarise this, the fact that no infant in this study had a lower sternal length that would accommodate the length of three adult fingers does not mean that no infant will have a long enough sternum. The probability of this happening is 3/n%. Using this ‘rule of three’, in only a maximum of 10% of infants under 1 year would the recommended method place the rescuers fingers wholly on the bony sternum. The lengths, weights and head circumferences of the infant sample were plotted on standard growth curves. Only four were smaller than predicted, and removing their data did not alter the mean length or standard deviation, so this was not a group of under-sized babies. The mean height of the adults used in this study was 1.67 m (5%6¦), so they were not an excessively tall group.
Given these findings, it is surprising that injuries following cardiopulmonary resuscitation are not more commonly reported. Health-care professionals would be likely to move their fingers fully onto the sternum during chest compressions. Lay rescuers are likely to stick rigidly to the system they have been taught. If injuries have been noted, it may have been assumed that the rescuer’s fingers had been wrongly positioned. In the light of the data presented here, we would recommend that the method used to determine finger position during cardiac arrests in infants, be changed to a method that uses sternal anatomy, rather than the position of the nipples. Running a finger along the lower costal margin to locate the end of the bony sternum, leaving the finger on that edge and placing two fingers of the other hand up from it, may be a better method. It has the added advantage of being similar to the methods used to locate the hand position for chest
46
F. Clements, J. McGowan / Resuscitation 44 (2000) 43–46
compressions in older children and adults, and may assist with teaching and retention of CPR skills, particularly in lay people.
Acknowledgements We thank Alex McMahon of the Robertson Centre for Biostatistics for statistical advice. We also thank wee Shona for being the first baby we annoyed in this study.
References [1] European Resuscitation Council. Paediatric basic life support. Resuscitation 1998;37:97–100
.
[2] Emergency Cardiac Care Committee and Subcommittees of the American Heart Association. Guidelines for cardiopulmonary resuscitation and emergency cardiac care. J Am Med Assoc 1992;268:171 – 302 [3] Thaler MM, Krause VW. Serious trauma in children after external cardiac massage. New Engl J Med 1962;267:500. [4] Thaler MM, Stobie GHC. An improved technique of external cardiac compression in infants and young children. New Engl J Med 1963;269:606 – 10. [5] Phillips GWL, Zideman DA. Relation of infant heart to sternum: its significance in cardiopulmonary resuscitation. Lancet 1986;I:1024 – 5. [6] Finholt DA, Kettrick RG, Wagner HR, Swedlow DB. The heart is under the lower third of the sternum. AJDC 1986;140:646 – 9. [7] Orlowski JP. Optimal position for external cardiac massage in children. Crit Care Med 1984;12:224. [8] Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. J Am Med Assoc 1983;249:1743– 5.
Finger position for chest compressions in cardiac arrest in infants Fiona Clements a,*, John McGowan b a b
Royal Hospital for Sick Children, Yorkhill, Glasgow G3 8SJ, UK Southern General Hospital, 1345 Go6an Rd., Glasgow G51 4TF, UK
Received 26 July 1999; received in revised form 1 December 1999; accepted 7 December 1999
Abstract Objecti6e: To determine whether the recommended method of locating finger position for chest compression in infant cardiac arrest can cause pressure on the abdomen or xiphisternum. Design: The length from the inter-nipple line to the xiphisternum was calculated in 30 infants. These lengths were compared with the finger position achieved by 30 adults, using the recommended method, on templates of infant chests. Results: The mean infant lower sternal length was 2.3 cm (95% CI 1.6). The mean distance covered by the adults fingers was 4.4 cm (95% CI 0.9). Conclusion: If any infant in this study had chest compressions performed by any of the adults, using the recommended method, pressure would be exerted on the xiphisternum or abdomen. We suggest changing the method of locating finger position, to one using sternal anatomy. © 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: BLS; Cardiac arrest; Infant; Chest compression; Paediatric resuscitation
1. Introduction The current recommendations of both the European Resuscitation Council and the American Heart Association for determining the finger or hand position during external chest compressions, are the same [1,2]. In adults, the rescuer’s hands are placed two finger breadths above the lower sternal edge. In children older than 1 year, one hand is positioned on the lower half of the sternum. However, in infants, the position is located by placing two fingers on the sternum, one finger’s breadth below the inter-nipple line. The rationale for selecting this method of locating finger position in infants is unclear from the literature. It has been observed that compression of the xiphoid process or abdomen, during chest compressions can rupture the liver, Thaler reported two cases [3,4]. The European Resuscitation Council does not recommend abdominal thrusts in a choking infant, as this may rupture the abdomi* Corresponding author. Tel.: +44-141-2010602.
nal viscera [1]. We have investigated whether the described method of determining finger position could cause pressure on the abdomen or the xiphoid process.
2. Materials and methods Thirty infants, under the age of 1 year had the length of their sternum, from the sternal notch to the start of the xiphisternum, and the distance from their sternal notch to the inter-nipple line, measured (Fig. 1). In addition, the infants sex, date of birth, age when measured, length, weight and head circumference were recorded. It proved easier to calculate the length of the sternum from the inter-nipple line to the xiphisternum by subtraction, than to measure it directly, as the lower edge of the sternal notch is an easily located landmark. Infants were excluded from the study if they were born pre-term or if they had thoracic, sternal or spinal deformities, including surgery. Ethical
0300-9572/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 0 - 9 5 7 2 ( 9 9 ) 0 0 1 6 5 - 3
F. Clements, J. McGowan / Resuscitation 44 (2000) 43–46
44
3. Results The results are summarised in Tables 1 and 2 and Fig. 2.
4. Discussion
Fig. 1. Lengths measured on the infants chests. Length x, lower edge of sternal notch to inter-nipple line. Length y, lower edge of sternal notch to xiphisternum.
approval had been granted and parental consent was obtained before each measurement. Thirty adults were asked to place their fingers on a paper template, using the guideline recommended above. The only markings on the template were two nipples and a horizontal mid-line. The position of the lower edge of their lower finger was marked on the template. The distance from the inter-nipple line to the edge of the lower finger on the template was measured, using the same tape measure used to measure the infants. The adults sex and height were also recorded. A z-test for unmatched samples was used to compare the mean lengths.
It has been shown that the heart lies under the lower third of the infant sternum [5,6]. More importantly, it has been observed that better arterial pressures are generated, in infant cardiac arrest, when chest compressions are performed on the lower third than on the middle third of the sternum [7]. The method described by the European Resuscitation Council and the American Heart Association to determine finger position, relies on the anatomical relationship between the nipples and the lower sternum. It has been noted, however, that in infants, the position of the nipples does not predict the position of the heart [6]. From Tables 1 and 2 and Fig. 1, it can be seen that the length covered by three fingers of an adult exceeds the distance from the nipple-line to the xiphisternum. No infant in this series had a lower sternal length of 4 cm; all were shorter than this. The shortest length covered by the adults fingers was 4 cm. If these adults used the guideline for finger position, they would be pressing on the xiphisternum or upper abdomen of all of these infants.
Table 1 Infant sternal lengths, from inter-nipple line to xiphisternum Sample (weeks)
Mean age (weeks) (+S.D.)
n
0–4 4–12 12–26 26–52 All ages
4.1 6.4 21.1 46.4 21.6
7 2.6 5 1.7 9 2.4 9 2.1 30 2.3
days (3.3) (2.3) (6.7) (4.2) (18.7)
Mean length (cm)
95% confidence interval
Range (cm)
1.4 1.6 1.6 1.2 1.6
2–3.5 1–3 1–3.5 1.5–3 1–3.5
Table 2 Measurements from adults Measurement
n
Mean length
Length, from internipple line, to lower edge of adult fingers Adult heights
30 4.4 cm 30 1.67 m
95% confidence interval
Range (cm)
0.9 0.2
4.0–5.5 1.47–1.9
F. Clements, J. McGowan / Resuscitation 44 (2000) 43–46
45
Fig. 2. Line A, length from infants inter-nipple line to xiphoid. Line B: length from inter-nipple line to lower edge of adults fingers (mean values with 95% CI). z test for unmatched samples, z=11.85; P=0.01.
The predictive value of a zero numerator has been described elsewhere [8]. To summarise this, the fact that no infant in this study had a lower sternal length that would accommodate the length of three adult fingers does not mean that no infant will have a long enough sternum. The probability of this happening is 3/n%. Using this ‘rule of three’, in only a maximum of 10% of infants under 1 year would the recommended method place the rescuers fingers wholly on the bony sternum. The lengths, weights and head circumferences of the infant sample were plotted on standard growth curves. Only four were smaller than predicted, and removing their data did not alter the mean length or standard deviation, so this was not a group of under-sized babies. The mean height of the adults used in this study was 1.67 m (5%6¦), so they were not an excessively tall group.
Given these findings, it is surprising that injuries following cardiopulmonary resuscitation are not more commonly reported. Health-care professionals would be likely to move their fingers fully onto the sternum during chest compressions. Lay rescuers are likely to stick rigidly to the system they have been taught. If injuries have been noted, it may have been assumed that the rescuer’s fingers had been wrongly positioned. In the light of the data presented here, we would recommend that the method used to determine finger position during cardiac arrests in infants, be changed to a method that uses sternal anatomy, rather than the position of the nipples. Running a finger along the lower costal margin to locate the end of the bony sternum, leaving the finger on that edge and placing two fingers of the other hand up from it, may be a better method. It has the added advantage of being similar to the methods used to locate the hand position for chest
46
F. Clements, J. McGowan / Resuscitation 44 (2000) 43–46
compressions in older children and adults, and may assist with teaching and retention of CPR skills, particularly in lay people.
Acknowledgements We thank Alex McMahon of the Robertson Centre for Biostatistics for statistical advice. We also thank wee Shona for being the first baby we annoyed in this study.
References [1] European Resuscitation Council. Paediatric basic life support. Resuscitation 1998;37:97–100
.
[2] Emergency Cardiac Care Committee and Subcommittees of the American Heart Association. Guidelines for cardiopulmonary resuscitation and emergency cardiac care. J Am Med Assoc 1992;268:171 – 302 [3] Thaler MM, Krause VW. Serious trauma in children after external cardiac massage. New Engl J Med 1962;267:500. [4] Thaler MM, Stobie GHC. An improved technique of external cardiac compression in infants and young children. New Engl J Med 1963;269:606 – 10. [5] Phillips GWL, Zideman DA. Relation of infant heart to sternum: its significance in cardiopulmonary resuscitation. Lancet 1986;I:1024 – 5. [6] Finholt DA, Kettrick RG, Wagner HR, Swedlow DB. The heart is under the lower third of the sternum. AJDC 1986;140:646 – 9. [7] Orlowski JP. Optimal position for external cardiac massage in children. Crit Care Med 1984;12:224. [8] Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. J Am Med Assoc 1983;249:1743– 5.