A majority of the previous investigations on the hydrodynamic sensitivity of mammalian cells have focused on lethal effects as determined by cell death or lysis. In this study, we investigated the effect of hydrodynamic stress on CHO cells in a fed-batch process using a previously reported system which subjects cells to repetitive, high levels of hydrodynamic stress, quantified by energy dissipation rate (EDR). The results indicated that cell growth and monoclonal antibody production of the test cells were very resistant to the hydrodynamic stress. Compared to the control, no significant variation was observed at the highest EDR tested, 6.4 × 106 W/m3. Most product quality attributes were not affected by intense hydrodynamic stress either. The only significant impact was on glycosylation. A shift of glycosylation pattern was observed at EDR levels at or higher than 6.0 × 104 W/m3, which is two orders of magnitude lower than the EDR where physical cell damage, as measured by lactate dehydrogenase release, was observed. While not as extensively investigated, a second monoclonal antibody produced in a different CHO clone exhibited the same glycosylation change at an intensive EDR, 2.9 × 105 W/m3. Conversely, a low EDR of 0.9 × 102 W/m3 had no effect on the glycosylation pattern. As 6.0 × 104 W/m3, the lowest EDR that triggers the glycosylation shift, is about one order of magnitude higher than the estimated, maximum EDR in typically operated, large-scale stirred tank bioreactors, further studies in a lower EDR range of 1 × 103–6.0 × 104 W/m3 are needed to assess the glycosylation shift effect under typical large-scale bioreactor operation conditions. Follow-up studies in stirred tanks are also needed to confirm the glycosylation shift effect and to validate the repetitive hydrodynamic stress model. Biotechnol. Bioeng. 2009;103: 1103–1117. © 2009 Wiley Periodicals, Inc.