Letter to the editor: “Is hypotension a major risk factor for neurological morbidity at term age in very preterm infants”

January 12, 2004

Dr. David R. Harvey, Editor-in-Chief

Department of Paediatrics

Queen Charlotte's and Chelsea Hospital

Goldhawk Road

London W6 OXG

United Kingdom

Dear Dr. Harvey:

Martens et al. [1] present the title of their recent report in the form of a question: “Is hypotension a major risk factor for neurological morbidity at term age in very preterm infants?” Unfortunately, the design of the study did not allow the authors to answer this important question.

The major limitation is the use of a single mean arterial blood pressure value (<30 mm Hg) to define hypotension. In 2000, we reported blood pressure data in 116 neonates with birth weights ≤1500 g [2]. In these very-low-birth-weight neonates, 95% confidence intervals for mean blood pressure extended well below 30 mm Hg, and for those infants with birth weights ≤1000 g, they extended slightly below 20 mm Hg. When the neonates were separated into gestational age groups, a clear relationship between lower gestational age and lower mean blood pressure was observed during the first 72 h postnatal.

An additional concern regarding the current study [1] is the failure to account for the postnatal rise in mean blood pressure. Although not clearly specified, it appears that all blood pressure measurements obtained during the initial hospitalization were included in the analysis. We [2] and others [3] have shown that blood pressure increases significantly during the first week of life. Analyzing the blood pressure data in light of postnatal age [1] would have clarified further which neonates were truly hypotensive.

In Table 3, Martens et al. [1] report a highly significant (p<0.001) relationship between gestational age <27 weeks and hypotension, and it is noted that 22 of the 30 neonates in this gestational age category (73%) were hypotensive using the study criteria (<30 mm Hg). Although the authors attempted to control for confounding variables, the value for the adjusted lower confidence interval bound for hypotension was 1.02, i.e., just above 1.00 (Table 2). These results suggest that further evaluation of the data [1] using gestational and postnatal age-specific data for mean blood pressure would provide clarification of the relationship between hypotension and outcome in preterm neonates.

Sincerely,

William D. Engle, MD

Charles R. Rosenfeld, MD