Exfiltration storm sewer systems (ES) are a low impact development (LID) stormwater control designed to reduce stormwater runoff volumes by promoting infiltration. These systems are integrated into conventional storm conveyance systems and consist of a perforated pipe embedded in a gravel trench beneath the main storm sewer. While effective for runoff attenuation, ES may increase the risk of groundwater contamination, including from de-icing salts. This study evaluates the runoff reduction performance of an ES installed in low-permeability soils, its effect on groundwater levels and quality, and the influence of urban karst preferential pathways associated with subsurface infrastructure. A 75 m section of an ES installed in low permeability soils beneath a road in London, Ontario, was monitored over 13 months, including observations within the sewer system and surrounding groundwater. Despite low-permeability soils and a shallow (potentially perched) groundwater table, the ES achieved high runoff volume reductions (>83%). Storm-sewer chloride loads were reduced by ~99.8% via infiltration-driven volume losses, transferring chloride to the subsurface. Limited groundwater level and quality responses observed in piezometers installed in native soils, contrasted with significant groundwater level and EC fluctuations in a downgradient piezometer installed in high permeability backfill material (i.e. urban karst zone). The results indicate that the high hydrologic performance of the ES may be due to urban karst features which may also control the fate of infiltrated contaminants in the subsurface. Study findings provide insights into the effectiveness of ES for stormwater management and interactions between infiltration-based LIDs and anthropogenic urban karst features.