The within-site patchy distribution of organisms represents a natural source of variability that can bias the precision of biotic indices. This study aims to evaluate the spatial distribution of benthic macroinvertebrates and its effects on the biotic index used in Italy for assessing the ecological status of rivers. Two main issues were considered: the sampling design and the minimum number of replicates which are necessary to achieve a given precision level. To this purpose, 30 surber replicates were taken, instead of the ten required for the application of the index, in a homogeneous reach of a mid-sized perennial river (Trebbia River, Northern Italy). The percentile confidence interval (p.c.i.) of the index was calculated from 1000 possible communities with the resampling method and the bootstrap procedure. The 95% p.c.i. varied from 0.195 to 0.227, which is great considering that the index ranges from 0 to 1. The metrics based on abundance data were especially affected by the heterogeneous distribution of organisms. The resulting precision of the index increased only partially when doubling the sampling effort.