A standard protocol for blood pressure measurement in the newborn

Improvements in neonatal care have resulted in increasing survival of extremely premature infants whose hospital course often runs into weeks or months. Some interventions during the acute care of these neonates, such as umbilical catheterization and use of steroids, not infrequently result in elevation of blood pressure (BP). It is, therefore, essential that these infants be monitored accurately for possible hypertension during their convalescence. Unfortunately, normative data on BP in this population are scant and comparison of data from various studies is hampered by methodologic differences in design. Studies in adults address the necessity for a restful state, adopting a comfortable position, and attempts to reduce the startle response to initial cuff inflation. Studies in the newborn using the oscillometric technique have not addressed these concerns. A standard BP measurement protocol was studied to determine the effect of ensuring a restful state, startle response to cuff inflation, and infant position on BP in clinically stable low birth weight infants after the first week of life.

The Dianamap oscillometer was used to measure BP in infants with a birth weight <2500 g between 7 and 42 days postnatal age. Each infant was studied only once when they were clinically stable. BP was measured in two positions, prone and supine, in random order. Infants were studied at least 11/2 hours after their last feeding or medical intervention. An appropriate sized cuff was applied to the right upper arm and the infant was positioned according to randomization. The infant was then left undisturbed for at least 15 minutes or until the infant was sleeping or in a quiet awake state. Three successive BP recordings were taken at 2-minute intervals. The infant's position was then reversed and another 15 minutes of quiet time was allowed. Thereafter, a second set of three successive BP recordings were obtained. The most recent routine nursing BP measurement was also recorded. Data were analyzed using analysis of variance and are presented as means and standard errors of the mean.

Sixty-four infants were studied. Birth weights ranged from 901 to 2423 g and gestational ages from 26 to 37 weeks. Overall, mean BP was significantly lower in the prone than supine positions (45.7 +/- 0.7 vs 47.8 +/- 0.8 mm Hg, P < .002). In either position, the first measurement was significantly higher than the third (average difference was 3 mm Hg, P < .003). In general, the relationships among position and order of measurement were similar for systolic and diastolic BP. Mean BPs obtained by routine nurse measurements were significantly higher than those in either position using our standard protocol (54.4 vs 47.0 or 49.1 mm Hg, P < .003). Moreover, the routine nurse measurements varied more widely than did those obtained using the standard protocol. The standard deviation for the routine mean BP measurements by nurses was 11.4 compared with 6.8 and 8.2 for the first measurements in the prone and supine positions, respectively, with the standard protocol. The mean BP measurements made in the supine position (the highest measurements obtained) using the standard protocol were also significantly lower than published values: 57 of 64 measurements were less than the average mean BP for age described by Tan (J Pediatr. 1988; 112:266-270).

The statistically significant difference between the prone and supine position and among successive measurements in each position are not clinically relevant. The clinically significant differences between measurements obtained with this standard protocol and routine nursing measurements or published data are the result of ensuring a restful state after cuff application. We believe that measurements thus obtained are more representative of true resting BPs in these infants. (ABSTRACT TRUNCATED)