The incidence and significance of systolic cardiac murmurs in infants.

In the course of carrying out routine physical examinations on newborn babies, certain infants were found who presented loud systolic cardiac murmurs in the absence of other abnormal physical signs. The significance of such a finding was doubtful, but the murmur in many cases was so loud as to suggest the possibility of congenital heart disease. An investigation was therefore carried out to assess the incidence and the significance of systolic cardiac murmurs occurring in the newborn. Published records of such murmurs are few. Taussig (1947) has stated that 'a murmur audible in the newborn period is always of doubtful significance', but 'the persistence of a murmur for a period of weeks is usually indicative of a congenital malformation of the heart'. Siemsen (1938) examined newborn infants for the presence of cardiac murmurs at least twice a week during their hospital stay and found a wide variation in the incidence of murmurs. He placed the incidence as high as 25% of 105 infants examined, but the percentage with murmurs increased with the number of examinations per patient. These murmurs were systolic, mostly soft and blowing, and usually loudest over the apex. Lyon, Rauh and Stirling (1940) place the incidence of murmurs at 1 90% in a series of 7,673 newborn infants examined during the first week of life. The incidence appeared to be uninfluenced by such factors as sex of the infant, month of birth, birth weight and the occurrence of syphilis in the mother. They were able to follow the course of 92 infants with systolic murmurs in many cases up to I year of age. Four infants died, and two of these who came to necropsy were found to have congenital heart disease. Seventy-one infants were entirely normal, 14 had persistent murmurs, two had inconstant murmurs, and one had extrasystoles. Cox (1948) has assessed the frequency of systolic murmurs as 5% of 630 'ostensibly healthy infants', but these were not all restricted to the newborn period. In just under half of the cases presenting murmurs the murmur persisted. Method The material for this investigation consisted of all newborn babies examined in the Simpson Memorial Maternity Pavilion, Edinburgh, over a five-month period from April to August, 1950. The total number of babies examined was 1,133. All examinations at birth and subsequent follow-up were carried out by one observer. Each baby was subjected to a general clinical examination as soon as possible after birth. In particular the precordium was palpated for thrills and the heart auscultated at the four valvular areas. A chest piece 3 cm. in diameter, with diaphragm, was used for the auscultation. When a murmur was detected it was classified according to its intensity using the classification of systolic murmurs adopted by Levine and Harvey (1949). The classification was as follows:Grade I is the faintest murmur that can be distinctly heard (not likely to be audible during the first few seconds of auscultation). Grade II is a soft murmur that is heard immediately on auscultation. Grade III is a murmur louder than Grade II. Grade IV is a murmur louder than Grade III. Grades V and VI are the loudest of all and were not heard in this series. When any real doubt existed as to the presence of a Grade I murmur a negative finding was recorded and the baby examined on another occasion. No diastolic murmurs were detected in the series. Repeated auscultation of the heart was carried out as often as could be arranged during the first 10 days of life, and in over three-quarters of the cases auscultation was carried out on two or more separate days within this period. Radiographic and electrocardiographic examinations were carried out in the majority of babies with murmurs. The variable results of such examinations were difficult to assess, and seldom proved of real value in making a diagnosis of congenital heart disease in the neonatal period. A group of 20 infants with systolic murmurs was