OBJECTIVE: Pediatric consensus-driven cardiopulmonary resuscitation guidelines target chest compression (CC) depths of one third to one half anterior-posterior (AP) chest depth. Estimates for this target as assessed by computed tomography (CT) measurements of internal and external AP chest dimensions could direct future pediatric cardiopulmonary resuscitation guidelines.
METHODS: A total of 280 consecutive chest CT scans in permuted blocks of 20 for each of 14 age divisions between 0 and 8 years were reconstructed and analyzed. External and internal AP depths were measured at midsternum, and residual chest depth was calculated at simulated one-third and one-half AP compressions.
RESULTS: After a simulated compression calculation, one-half external AP depth CC would result in residual internal depth of <10 mm for 94% (263 of 280) of children 3 months to 8 years. For a one-third external AP CC, only 0.4% (1 of 280) of children 3 months to 8 years had a calculated residual internal chest depth <10 mm.
CONCLUSIONS: By using CT reconstruction estimates of chest dimensions across the developmental spectrum from 0 to 8 years of age, we demonstrated that a simulated CC targeting approximately one-third external AP chest depth seems radiographically appropriate for children aged 3 months to 8 years, whereas simulated CC targeting approximately one-half external AP chest depth seems radiographically to be too deep, resulting in residual internal chest depth of <10 mm for most patients of this age.
Survival rates for children who experience cardiac arrest are historically low. Donoghue et al1 reported only 4% neurologically intact survival overall out-of-hospital. In-hospital, ∼17% of children survive and 13% of adults survive neurologically intact.2 Survival to return of spontaneous circulation, hospital admission, and hospital discharge has been associated with quality of chest compressions (CCs) in adults, in particular adequate depth of compression.3 Current adult CC guidelines specify that a CC depth target of 38 to 51 mm corresponds to ∼16% to 21% relative depth of the adult chest.4 Real-time directive and corrective feedback can improve adherence to these guidelines and resuscitation outcomes.5 The American Heart Association (AHA), European Resuscitation Council (ERC), and International Liaison Committee on Resuscitation (ILCOR) guidelines and treatment recommendations were created by consensus, and the relative depth of CC was estimated grossly and without evidence of internal and external chest depths. Specific evidence for target depth of CC parameters in children are not known or published; thus, therapeutic targets for pediatric CCs are based on extrapolation and extension from adults, animal models, and consensus interpretation of the literature. AHA pediatric cardiopulmonary resuscitation (CPR) guidelines 2000 stated that CCs provide circulation as a result of changes in intrathoracic pressures and/or direct compression of the heart. In Guidelines 2000, the recommended target of compression depth for children is approximately one third to one half the depth of the chest (1.0–1.5 in).6 In Guidelines 2005, no new research data that support specific CC depth targets were presented. The previous recommendation of a target depth of 1.0 to 1.5 in has been removed. For children 1 to 8 years of age, the AHA recommends “approximately one third to one half the anterior-posterior diameter of the chest,”4 the ERC recommends “approximately one third of the depth of the chest,”7 and ILCOR recommends “approximately one third the anterior-posterior diameter of the chest.”8
The purpose of this study was to characterize the internal and external chest depths of children of various ages between the ages of 0 to 8 years by using chest computed tomography (CT) technology. These data can then be used to understand further the maximum absolute and maximum relative depths of CCs that can be performed on children and to provide evidence-based guidance for CPR feedback devices in the future. The evidence from this study could emphasize the lack of data to support current pediatric CC guidelines and begin to question the safety and efficacy of these current guidelines. Our secondary objective was to calculate and estimate residual internal chest depths if simulated external CCs of one-half, one-third, and one-fifth external antero-posterior (AP) chest depth were delivered. We hypothesized that CT reconstruction estimates of chest dimensions across the developmental spectrum from 0 to 8 years would demonstrate that simulated CC targeting approximately one-third external AP CC depth seems appropriate for children 3 months to 8 years.
General Schema of Study Design and Subject Enrollment
This study was a retrospective study approved by the institutional review board at the Children's Hospital of Philadelphia. All pediatric chest CT scans previously performed by the radiology department at the Children's Hospital of Philadelphia were considered eligible except for patients with severe scoliosis or obvious major chest wall deformities (pectus excavatum/ carinatum, chest hypoplasia). Twenty consecutive retrospective chest CT scans that were available were reviewed and analyzed for each age group: <3 months, 3 months to <6 months, 6 months to <9 months, 9 months to <12 months, 12 months to <15 months, 15 months to <18 months, 18 months to <21 months, 21 months to <2 years, 2 years to <3 years, 3 years to <4 years, 4 years to <5 years, 5 years to <6 years, 6 years to <7 years, and 7 years to <8 years. By using CT reconstruction, individual chest depths were measured and residual internal depth resulting from several simulated CC depths was calculated. Measurements were assessed by using objective measurement scales and were supervised and confirmed by an attending radiologist expert in pediatric chest CT.
We next examined the percentages of patients who would have <10 mm of residual chest depth during a simulated one-half external AP chest depth compression. Our investigative team prospectively selected 10-mm residual chest depth as a cutoff by consensus, because we believed that <10 mm residual chest depth would still have the potential to injure the intrathoracic structures and may not actually be achievable because of the presence of the thickness of the myocardium.
Sample Size and Power
From pilot studies of children, the coefficient of variation for measurements of chest depth have been reported to be ∼10% to 13%.9 Given this known variability, a sample of 20 patients in each age group was chosen to limit the calculated 95% confidence interval for measurements in each group to <0.5 cm in infants and up and <2 cm in adult-size children with 80% power. To capture changes in measurements across development in detail, we divided the age range of 0 to 8 years into 14 age groups, thereby giving a total number of 280 CT scans to review.
Internal and external chest depth measurements were performed as demonstrated in Fig 1. By using an axial slice at the midsternal level below the level of the intermammary line and in the typical location where CCs are delivered, we calculated the external chest depth by measuring a line drawn perpendicularly from the skin anteriorly to the skin posteriorly. In addition, we calculated the internal depth available by measuring a line drawn perpendicularly from the posterior midsternum to the anterior vertebral body. The depth of a proposed CC was calculated for one half (50%), one third (33%), and one fifth (20%) the measured external chest depth. Residual internal chest depth after simulated CC was calculated as the internal chest depth from chest CT scan minus the depth of the proposed CC (assuming that the full depth of the CC would be transmitted to the internal structures).
Graphic representation of the range of internal and external chest depth measured across age groups was displayed by using box and whisker plots. We performed multiple regression to assess the association between changes in measurements with age and gender. We calculated the proportion of patients with measured residual chest depths of <0-mm and <10-mm residual chest depth across age groups for both the calculated compression depths and adult recommendations (38–51 mm). A value of 0 mm would indicate that no remaining internal chest depth would available during the simulated CC, and a negative value would suggest that it would be impossible to compress to that particular target depth because of there not being enough available internal chest depth to accommodate the CC. A positive value would reflect the remaining available internal chest depth after a simulated CC.
Figure 2 displays the internal and external chest depths by age. The predicted internal AP depth was found to be y (mm) = 51.7 + 0.38 * age (months) + 3.4 * male (yes = 1, no = 0) (R2 = 0.69, P < .001).
Figure 3 displays residual internal chest depth during simulated one-third and one-half AP external chest depth compressions. As with the measured internal chest depths, the residual internal chest depth measurements available for compression also increased with age (P < .001 for both one-half and one-third external chest depth compressions).
Figure 4 displays the percentage of patients in each age group with <0 mm and <10 mm of residual internal chest depth during a one-half AP external chest depth compression. For one-half external AP CC, 94% (263 of 280) of children aged 3 months to 8 years had a calculated residual internal chest depth <10 mm. For one-third external AP CC, only 0.4% (1 of 280) of children aged 3 months to 8 years had a calculated residual internal chest depth <10 mm.
Figure 5 displays residual internal chest depth during a simulated one-fifth (20%) chest depth compression. Figure 6 displays the percentage of pediatric patients with >10 mm of residual internal chest depth during the current adult recommended 38-mm (1.5 in) and 51-mm (2 in) simulated CC.
This is one of the first studies to attempt to evaluate and challenge objectively the current pediatric CC depth recommendations. The current AHA, ERC, and ILCOR recommendations are extrapolated from adult data and are consensus based. This study suggests that the current recommendations of one-third to one-half external AP chest depth are not ideal and may not be attainable or safe for all children. With a one-half chest depth compression, 25% of patients in the 3- to 12-month group would theoretically have no residual internal depth. This suggests that during a one-half AP CC, there would be no room for the intrathoracic structures, including the heart and great vessels. This may actually be impossible to achieve but if achievable could potentially be harmful to the structures being compressed. This is also the case for 21% of 1- to 3-year-olds and 8% of 3- to 8-year-old children.
The percentages of patients who would have <10 mm of residual chest depth during a simulated one-half external AP chest depth compression are displayed in Fig 4. These results suggest that during a one-half external AP CC, 98% of 3- to 12-month-olds would have <10 mm of residual chest depth; 96% of 1- to 3-year-olds and 88% of 3- to 8-year-olds would also have <10 mm of residual chest depth during a one-half AP CC. As mentioned in the methods, we prospectively selected 10-mm residual chest depth as a cutoff by consensus, because we believed that <10-mm residual chest depth would still have the potential to injure the intrathoracic structures and may actually not be achievable because of the presence of the thickness of the myocardium. We believed that this was a very conservative estimate of “overcompression.” Because 10 mm of residual chest depth still may not be enough to accommodate the intrathoracic structures, these results provide more impressive data suggesting that a one-half external AP chest depth compression may not be achievable or safe for young children.
For completeness, we further calculated and compared our findings with current adult CC depth recommendations. Pickard et al10 found that the external compression depth target for adults, 38 to 51 mm, corresponds to ∼16% to 20% relative depth of compression for male adults and 17% to 21% for female adults. The residual internal chest depth during a simulated 20% external AP CC (Fig 5) suggests that all pediatric patients would have >20 mm of residual internal depth (100% >10 mm, as well). We then measured the residual internal depth during the current adult recommended CC of 38 mm and 51 mm (Fig 6). During a 38-mm simulated CC, 86% of 3- to 12-month-olds, 98% of 1- to 3-year-olds, and 100% of 3- to 8-year-olds would have >10 mm of residual chest depth. During a 51-mm simulated CC, 15% of 3- to 12-month-olds, 55% of 1- to 3-year-olds, and 98% of 3- to 8-year-olds would have >10 mm of residual chest depth. These findings suggest that for 3- to 8-year-olds, the current adult recommendations of 38 to 51 mm might be a safe alternative to the current pediatric recommendations. For the 3-month to 3-year-old, a 51-mm CC seems to be too deep, but a 38-mm CC would potentially be a safe alternative.
We believe that this study will have a positive impact on the current field of pediatric resuscitation by providing radiographic evidence for CC depth recommendations that are approximately one third the external AP depth of the pediatric chest. With the growing field of CPR real-time-assist and feedback technologies, it has become more important than ever to ensure that the current goal CC depths are both safe and attainable in the pediatric population, because these recommendations are anticipated to be programmed into the CPR feedback technologies. Thus, the information obtained from this study begins to question the current CC depth guidelines and provide a basis for additional research into these recommendations.
Target parameters for delivery of CCs should generally apply to the wide range and spectrum of chest wall characteristics of patients across the pediatric spectrum. Thus, therapeutic targets for feedback and quality assurance of pediatric CCs should focus on the delivery of adequately deep and fast compressions, with efforts to promote full release according to current consensus treatment recommendations and guidelines.
Our study is a retrospective, an observational, and a single-center study and thus is limited by the patient selection, which includes a variety of children who underwent chest CT scans at the Children's Hospital of Philadelphia. Although we did have strict exclusion criteria for those with chest wall deformities, we did not describe in detail patients who were included. The included patients did have a variety of prescan diagnoses ranging from pneumonia and tumors to congenital abnormalities. This patient selection may not be identical to patients who require CCs and eliminates those with severe thoracic deformity. In our analysis, we also assumed that the external diameter of the chest does not change as a result of the compressions. This could happen if the chest collapses or if the position of the chest in the release phase is modified by the compressions (eg, as a result of leaning). The major limitation to this study is that measurements were performed in 2 dimensions and without any hemodynamic information supporting ideal CC depth. On the basis of this study, additional specific volumetric analysis of the chest cavity and that of the heart both before and during a CC is warranted. With the advancing field of diagnostic radiology, measuring static and dynamic volumes of the chest cavities and heart is becoming more practical. A final limitation is that we did not account for potential soft tissue compressibility during our calculated external CCs when we performed our calculations. We assumed that all of the one-third or one-half external AP chest depth compressions would be translated to the internal chest depth. Some patients may have more or less soft tissue around the sternum and back that would have absorbed some of the CC and could have caused an overestimation of the actual CC depth translated to the internal chest depth. We believe this to be a very small amount of compressibility.
Using CT reconstruction estimates of chest dimensions across the developmental spectrum from 0 to 8 years of age, we demonstrated that simulated CC targeting approximately one-third external AP chest depth seems appropriate for the majority of children aged 3 months to 8 years. Simulated CC targeting approximately one-half external AP chest depth may be too deep, resulting in residual internal chest depth of <10 mm. Current consensus-recommended one-third to one-half external AP chest depth may not be feasible or desirable for all children who are younger than 8 years and might need to be revised to recommend “at least one-third external AP chest depth.” Use of a constant CC depth target of 38 mm would be expected to be adequate for >98% of 1 to 8-year-old children, with >10 mm of residual chest depth. This simulated CC study suggests that additional studies should be conducted to confirm these findings with actual CC and with hemodynamic measures during real CPR.
Dr Nadkarni receives unrestricted grant support from the Laerdal Foundation for Acute Care Medicine; Medical Education Technologies, Inc; Agency for Healthcare Research and Quality; and National Institutes of Health. Dr Nadkarni has served as the Chairman of the Emergency Cardiovascular Care Committee of the American Heart Association and is the current co-chairman of ILCOR.
We thank the Center for Simulation, Advanced Education, and Innovation at the Children's Hospital of Philadelphia; Center for Resuscitation Science at the University of Pennsylvania School of Medicine; and Kristy Arbogast, PhD, Matthew Maltese, MS, Robert A. Berg, MD, Stephanie Tuttle, MBA, Naoki Shimizu, MD, Mark Helfaer, MD, and Akira Nishisaki, MD.
- Accepted February 24, 2009.
- Address correspondence to Matthew Braga, MD, Pediatric Critical Care, Children's Hospital at Dartmouth, Dartmouth-Hitchcock Medical Center, One Medical Center Dr, Lebanon, NH 03756-0001. E-mail:
The views expressed in the manuscript are of the authors and are not meant to represent the position of the American Heart Association or the International Liaison Committee on Resuscitation.
Financial Disclosure: The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on This Subject:
Specific evidence for target chest compression depth parameters in children are not known or published, and thus therapeutic targets for pediatric chest compressions are based on extrapolation and extension from adults, animal models, and consensus interpretation of the literature.
What This Study Adds:
We characterized chest depths of children by using chest computed tomography. These data were used to understand further the maximum absolute and maximum relative depths of chest compressions and could lead to improved guidelines for chest compressions in children.
- ↵International Liaison Committee on Resuscitation. 2005 international consensus on cardiopulmonary resuscitation and emergency cardiovascular care science with treatment recommendations. Circulation.2005;112 (8)[suppl 1]:III-73– III-90
- ↵Pedersen IM, Hermans JJ, Molenbroek JF. Measurement of the Sternum for Better Cardiopulmonary Resuscitation. Abstract presented at the Nordic Ergonomics Society conference; October 1–3, 2007; Lysekil, Sweden
- Copyright © 2009 by the American Academy of Pediatrics