This prospective study was undertaken to compare the utility of anorectal manometry (ARM) with that of anal electromyography (EMG) and cinedefecography (CD) in the diagnosis of paradoxical puborectalis syndrome (PPS). One hundred sixteen consecutive patients with a history of chronic constipation were prospectively assessed. These 35 males and 81 females were of a mean age of 60 years, ranging from 18 to 84 years. The incidences of PPS were 63 percent for ARM, 38 percent for EMG, and 36 percent for CD. The correlations of PPS were suboptimal: ARM and EMG, 70 percent; and ARM and CD, 61 percent. A two-tiered system for the manometric classification of PPS was developed. First, the evacuation pressure curve pattern was classified as a normal relaxed downward (Type A; n = 43), a nonrelaxed flat or equivocal (Type B; n = 36), and a paradoxical upward (Type C; n = 37). PPS was noted with increasing incidence within curve types (21 percent in Type A, 64 percent in Type B, and 95 percent in Type C). Second, an evacuation index (EI = evacuation pressure/squeeze pressure) was defined: Group I (EI < 0; n = 43), Group II (0 < or = EI < 0.25; n = 24), Group III (0.25 < or = EI < 0.5; n = 27), and Group IV (EI > or = 0.5; n = 18). The finding of PPS also correlated with the EI group: 21 percent in Group I, 67 percent in Group II, 74 percent in Group III, and 100 percent in Group IV. This subdivision of curve types and EI groups may provide a role in the diagnosis of PPS.