[Congenital heart diseases in newborns and infants; early detection, differentiation and accuracy of clinical diagnoses (author's transl)].

In order to facilitate for the general physicians the making of a suitable selection of babies who are in the most urgent need of specialized treatment at cardiac centres, simple methods for diagnosing and qualifying congenital cardiovascular diseases were elaborated. The following "minor" criteria were taken for suspecting a CHD: 1) cardiorespiratory distress following birth, 2) sequentially repeated Apgar score below normal, 3) "pneumonia" symptoms with respiratory distress, dyspnoea and cyanosis, attacks of unconsciousness, 4) feeding difficulties, failure to thrive, inexplicable irritability, 5) presence of other congenital anomalies. The almost certain presence of serious heart disease should be recognized in children, showing the following "major" symptoms: 1) permanent cyanosis, pallor or greyish colour, 2) cardiorespiratory failure (resembling usually symptoms of pneumonia), 3) ECG patterns indicating ventricular hypertrophy signs, 4) other significantly abnormal ECG patterns (e.g. AV and intraventricular conduction disturbances), 5) cardiac enlargement and lung vascularity abnormalities in chest X-rays, 6) weak, or impalpable arterial, particularly femoral pulses, femoral arterial pressures significantly lower, than at upper extremities, bounding pulses and high-pressure amplitude in arms and legs, 7) abnormal heart sounds and pathologic heart and vascular murmurs. A diagnostic "key", based upon evaluation of the "major criteria" facilitates the diagnosis and differentiation of the most important CHD's at neonatal and infantile age. When using this "key" one should keep in mind the relative frequency of incidence of particular lesions. The initial diagnoses by the above "key" were verified in 354 patients by cardiovascular catherisation, angiocardiography, surgical exploration, and for by autopsy. The diagnoses were perfectly accurate in 83.6% cases, in further 11.3% cases being also accurate but were supplemented by some details, and had to be corrected in only 5.1% cases.