Association of maternal hematocrit with premature labor.

An examination of the association of maternal hematocrit with the occurrence of premature birth revealed a continuous relationship between these two factors with an increasing rate of prematurity occurring with decreasing maternal hematocrit. The lowest risk of premature birth was found when maternal hematocrit was between 41% and 44%. Statistically increases in prematurity were noted at all hematocrit levels of less than or equal to 38%. A woman with a hematocrit of 37% had twice the risk of having a premature birth as a woman whose hematocrit was between 41% and 44% (p less than 0.01). In a logistic regression analysis with premature birth as the dependent variable and hematocrit as the only predictor variable, it was noted that decreasing the hematocrit by a single point was associated with a 24% increase in the risk of prematurity (odds ratio = 1.24) while a five-point decrease in hematocrit was associated with a tripling in the risk of prematurity (odds ratio = 2.98). In a multivariable logistic regression analysis hematocrit was noted to explain more of the variation in prematurity than any of the other risk factors included in the model. An investigation of this association in specific groups at high risk of having premature births revealed that 46% of the increased risk of prematurity in black women and 49% of the increased risk of prematurity in women receiving welfare was related to hematocrit or some factor associated with hematocrit.