To assess antenatal care in health care units, so as to obtain a baseline for future evaluation studies.A self-applied inquiry was conducted among municipal health managers within a probabilistic stratified random sample of 627 municipalities which, through expansion technique, extended the analysis to 5,507 municipalities. Data was collected from October 2003 to April 2004. The survey appraised information about the priority granted by the managers to each modality of care, as well as data concerning characteristics of the assistance provided and the declared estimate of the demand being covered. The Chi-square test and Student's t-test were performed in order to verify independence among the qualitative variables and mean differences, respectively.Almost half (43.8%; n=2,317) of the municipalities did not attend gestational risk; 81% (n=4,277) and 30.1% (n=1,592) reported that they attend over 75% of the demand for low and high risk antenatal respectively; 30.1% (n=1,592) attend over 75% of the demand for high risk care. Care for low risk (chi2=282,080; P<0.001 n=4,277) and for high risk pregnancies (chi2=267.924; P<0.001 n=5,280) were associated to geographic region, municipality's size and management modality within the Unified Health System. The guarantee of vacancy for labour and birth was also associated to management modality.There were gaps related to the provision and the quality of antenatal care within the Unified Health System. Municipal based health care extends the provision of antenatal care, but there are inequalities among regions and among municipalities according to demographic size.