Thorax: first published as 10.1136/thx.39.8.624 on 1 August 1984. Downloaded from Thorax 1984;39:624-627 Changes in rib cage geometry during childhood P OPENSHAW, S EDWARDS, P HELMS From Guy's Hospital, St Bartholomew's Hospital, and Brompton Hospital, London ABSTRACT Age related changes in rib cage geometry were found from measurements made on chest radiographs from 38 individuals aged from 1 month to 31 years and on computed tomogra- phy (CT) scans in another 28 individuals, aged from 3 months to 18 years. Chest radiographs were taken for minor respiratory symptoms or fever and only films showing no abnormality were used. CT scans were obtained in children undergoing staging for solid tumours in whom no intrathoracic deposits were found. In infants and very young children the ribs were found to be more horizontal and the sternal clavicular heads and diaphragmatic domes higher than in older children and young adults. Most of these changes were observed in the first two years of life, with something close to the adult pattern by the age of 2 years. Similarly cross sectional chest shape changed from the rounded infantile form to the more ovoid adult form by the same age. The configuration of the ribs observed in infancy and early childhood reduces the potential for thoracic expansion and may contribute to the frequency of respiratory problems found in the very young. copyright. Respiratory problems are frequent in childhood and Methods most significant illnesses in preschool children are referrable to the respiratory tract.' The reasons for We examined radiographs in 38 subjects aged 1 this age distribution are likely to be the immaturity month to 31 years and computed tomography (CT) of the host defences in the lungs and the lesser abil- scans of the thorax in 28 children and adolescents http://thorax.bmj.com/ ity of the respiratory pump to cope with the aged 3 months to 18 years. increased respiratory loads associated with airflow obstruction. The respiratory pump (rib cage, acces- sory muscles, diaphragm, and abdominal muscles) is CHEST RADIOGRAPHS thought to be less efficient in the very young child Chest radiographs were taken from the paediatric than in the mature adult because of the instability of accident and emergency department at Guy's Hospi- the thoracic cage and the lower efficiency of the tal and from adults having routine films taken before diaphragm.2 Not only is the rib cage more pliable in employment at the Brompton Hospital. Radio- the very young3 but the ribs also appear to be more graphs which were poorly orientated, as assessed by on September 26, 2021 by guest. Protected horizontal than in the adult.4-6 The pliable rib cage the position of the sternal heads of the clavicles in allows the chest wall to move inwards during strong relation to the spine, were excluded, as were films diaphragmatic contraction, while the more horizon- from any subject with a history of wheezing, asthma, tal lie of the ribs is likely to limit the potential for or recent pneumonia. All of the children underwent thoracic expansion by rib cage movement in the radiographic examination because of minor cephalad direction. respiratory symptoms or fever, and only those films Although it is known that rib cage geometry showing no abnormality were accepted for the changes from early infancy to adulthood, the timing study. Clear celluloid film was placed over each of this change has not be described. radiograph and the following points were marked: the most lateral points of ribs 1-10; the necks of ribs 1-10; the centres of each vertebral body; the dome of each diaphragm; the position of the clavicles. For each film a "slope index" was calculated for each rib Address for reprint requests: Dr P Openshaw, Department of pair. The maximum potential width of each rib pair Respiratory Medicine, Hammersmith Hospital, London W12 OHS. was calculated by adding the distance between the Accepted 17 April 1984 necks and most lateral points of the ribs to the dis- 624 Thorax: first published as 10.1136/thx.39.8.624 on 1 August 1984. Downloaded from Changes in rib cage geometry during childhood 625 tance between the necks. The separation of the most From the CT scans the thoracic index7-11 was lateral points of each rib pair was then measured derived by dividing the anteroposterior diameter by directly and expressed as a ratio of the maximal the lateral diameter at three thoracic levels- potential width. Once a slope index had been calcu- namely, the manubriosternal junction, the dia- lated for each rib pair an average was taken for ribs phragmatic dome, and midway between these two 7-10. reference levels. The thoracic index used was the In infants and very young children radiographs mean of that found at these three levels. were taken with the film behind the chest whereas older children and adults were positioned with the Results film in front of the chest. A further difference was that films taken in very young uncooperative chil- A similar pattern of changing rib cage geometry dren were taken in the upright posture with the arms emerged from both chest radiograph and CT scan partly supported. To observe how these different measurements. For diaphragm level, sternal head techniques could affect our measurements we com- level, rib slope index, and thoracic index the most pared films taken posteroanterior and anteropos- rapid period of change appeared to be from birth to terior in 22 children and adults above the age of 5 the age of about 2 years (figs 1 and 2). The mean years and films taken with arm support in the erect position of the left diaphragmatic dome was found posture and without arm support in the supine pos- to be at the level of the 8th thoracic vertebra at ture in infants of less than 6 months. birth, descending to thoracic vertebra 11 in the young adult. For subjects of all ages the right dia- CT SCANS phragmatic dome was found to be a half vertebral CT scans were obtained from children who were space higher than the left. The medial ends of the being examined to stage lymphomas and other solid clavicles also changed their projection from between tumours. Only scans from those children in whom cervical vertebra 7 and thoracic vertebra 1 at birth no deposits were found in the thoracic cavity were to the body of cervical vertebra 3 in the adult sub- used in this analysis. jects. The rib slope index decreased from birth to adulthood, a change which was complete by 10 years copyright. of age. The cross sectional thoracic index also C 5, changed over this period, with a more rounded shape in early infancy and the establishment of the * *SZ C 7 Sternal ends of http://thorax.bmj.com/ 1.0 0 clavicles 0 0 0 0 0 T 2- 0 .0 a 0.00 0 . 0 . Tr I/ le0 .0 0 4-ll. A 0 -6 0 a x 0*9- 0o 00 0 0 .0 0 C 'g T 6 0 AA on September 26, 2021 by guest. Protected vs 0 A A 0 I- A 0 00 A 0 > T 8. 0 v U) 0 A Left dome of A S 0 0 A 0 0 v diaphragm 0.8 - A S OI v 00 0 T10 0 0 0 00 * supine 0 0 0* 0.0 o erect anteroposterior T12- a : A erect posteroanterior 9 9 -11- I 0t i lb 3b I I I I -- Age (years) 0.1 1 3 10 30 Age (years) Fig 1 Vertebral level ofprojection ofsternal clavicular heads and the dome ofthe left hemidiaphragm based on Fig 2 Relationship between the mean slope index for rib chest radiographs, showing the descent with increasing age pairs 7-10 and age based on chest radiographs. A slope (measurements taken from the centre ofeach vertebral index ofI indicates a horizontal lie ofthe ribs and lower body). indices a downward slope. Thorax: first published as 10.1136/thx.39.8.624 on 1 August 1984. Downloaded from 626 Openshaw, Edwards, Helms 0.9* Thoracic configuration Infant Child/Adult 0 8- 0 Sternum 0 7- 0 0 *- B ,........a . i.r 0* *0 .X..... ................ ............... * . ....... 0-6- @0 ........ 02 ....... Abdomen ........ s * 0 @0 0. \...... 0*5- 0 T- ...... .... - Thoracic cross section I I 01 1 3 10 30 Spine Age (years) Rib Fig 3 Relationship ofthe thoracic index (anteroposteriorllateral thoracic diameter) to age based on k Stemuml computed tomography scans. move ovoid adult pattern by the age of 2-3 years Fig 4 Observed changes in configuration and cross (fig 3). sectional shape ofthe thorax from infancy to early No significant differences in slope iindex were childhood. Upper panel: infantile and adult rib cage found between anteroposterior and postceroanterior configurations. Lower panel: how rib growth at films in 22 age matched subjects above tihe age of 5 costochondral junctions and posterior rib angles could years (fig 2) (p > 0-1, two tailed Mann-Witney U explain the observed changes in cross sectional shape ofthe thorax. test). In 14 infants less than 6 months oif age there copyright. was no significant difference between slccpe indices opposite to those actually observed. Finally, the determined from supine and from erect films (fig 2; change in clavicular and diaphragmatic levels would p > 0-1, two tailed Mann-Witney U test) be expected if the young child systematically adopted a lordotic posture when held erect; the simi- Discussion lar features seen with the supine views (fig 2), how- ever, suggest that this is unlikely.