This report compares six Doppler ultrasound measures of blood flow: Pourcelot's pulsatility index, Gosling's pulsatility index, area under the curve, systolic amplitude, diastolic amplitude, and mean amplitude. Recordings from an in vitro arterial model and from the anterior cerebral arteries in 33 newborn infants with asphyxia, intraventricular hemorrhage, or normal state were analyzed. In the model, all measures had excellent correlation with flow. Neonates were correctly classified according to diagnosis by the Doppler measures as follows: Pourcelot's pulsatility index (100%), Gosling's pulsatility index (97%), diastolic amplitude (94%), mean amplitude (76%), area under the curve (70%), and systolic amplitude (58%). The best accuracy was obtained with the pulsatility indices.

Doppler ultrasound has been applied in the evaluation of blood flow in the anterior cerebral artery in the neonate with cerebro-vasculature disease. There is significant controversy as to which Doppler measure best reflects the status of the cerebrovasculature. We compared five commonly used Doppler indices of blood flow: Pourcelot's pulsatility index (PPI), Gosling's pulsatility index (GPI), diastolic amplitude (D), area under the curve (AUTC), and Systolic amplitude (S). The indices were derived from recordings obtained from an in-vitro model at different flow rates and from 33 newborn infants with either asphyxia (A), intraventricular hemorrhage (IVH), or “normal” state. In the in-vitro model, all had correlation coefficients > 0.93 when compared with an electromagnetic flowmeter, indicating that all indices correlate well with flow rate. The ability of the indices to classify the patients according to clinical diagnosis was evaluated using discriminant analysis. All indices appear to correlate well with flow in an in-vitro model with a fixed probe angle. In addition, PPI, GPI, and D were of equal value in detecting IVH and A. The wide range of diastolic measure appear to allow a clear separation between the groups. We suspect this because the other two angle sensitive measures, S and AUTC had smaller ranges and were less accurate.

We previously reported a correlation between brachial artery pulsatility index (continuous wave velocitometry) and left atrial enlargement (echocardiographic LA:Ao ratio) in neonates with PDA. To further define the limitations of neonatal Doppler ultrasound for PDA evaluation, we compared the pulsatility index (PI) obtained by continuous wave (CW) velocitometry with the index obtained from spectral analysis (SA) of the audio Doppler signal in ten patients with suspected PDA. Twenty-one brachial examinations were obtained. Linear regression of the PI values obtained by CW and SA revealed a highly linear correlation (r=0.89, p<0.001) but the regression line was not the line of identity. Additionally, the standard deviation of the residuals is greater than that expected for identical measurements. PI values by both methods resolved into two distinct groups, representing normal arterial flow and reverse flow secondary to left-to-right shunt. Although CW measurements were lower than SA measurements, there was no difference in sensitivity for identifying infants with significant PDA. We conclude that CW pulsatility index is a satisfactory predictor of PI obtained by the more precise SA technique. Lower CW values may be related to inherent limitations of the zero crossing detector used to produce the CW signal.

Cardiac output can be estimated with Doppler ultrasonography by measuring aortic root diameter and aortic blood velocity: Q (ml/min) = V × A × 60 (sec/min). Q=aortic blood flow, V=mean aortic velocity, and A=aortic cross sectional area. We evaluated the ability of a Diasonics duplex ultrasound scanner to estimate cardiac output in newborn piglets. Aortic diameter and mean peak aortic blood velocity were measured while simultaneously measuring cardiac output with the microsphere technique in six newborn piglets. Comparison of the 16 cardiac output measurements obtained with each technique demonstrated a correlation coefficient of .68. Duplex ultrasonography reliably estimated cardiac output in newborn piglets by measuring aortic diameter and mean peak blood flow velocity (p<0.05). However, in our hands, the correlation between these two techniques showed considerable scatter. Further studies to investigate the etiology of these deviations are necessary before utilizing Doppler derived cardiac outputs in the research or clinical setting.

A semi-automated system for evaluation of Doppler cerebral blood flow studies obtained from newborn infants is described. A low cost digitizer is used to convert the graphic data from the flow tracing to digital data. A small business computer is used to analyze the data and produce a chartable report. The reliability of the digitizer is also evaluated.

We examined the clinical significance of noninvasive intracranial pressure measurements and pulsatility indices in 74 infants with confirmed IC-IVh. The intracranial pressure measurements were obtained using the applanation principle, and the pulsatility indices were calculated from the Doppler flow velocity tracings of the anterior cerebral artery. Fifty-three infants (71.6%) who died had a significantly lower birth weight and gestational age than those who survived. Survival rate decreased significantly with increased intracranial pressure (P less than 0.0002) and increased pulsatility indices (P less than 0.0001). We found no significant relationship between outcome and the size of IC-IVH demonstrated by CT scan. Birth weight, intracranial pressure measurements, and cerebral arterial pulsatile flow changes appear to be major prognostic indicators in neonatal IC-IVH.

A low-cost microcomputer system is used to calculate cardiac parameters from echocardiographic (Echo) measurements and produce a chartable report. The system is simple, uses Basic language, and provides for data storage.