Treatment of infantile Pompe disease with recombinant human acid α-glucosidase has shown substantial improvement in survival, and in cardiac, motor and respiratory functions. We analyzed the outcome of all patients with infantile Pompe disease treated in the United Kingdom since the availability of the enzyme, using a questionnaire-based survey circulated to all treating centres. A total of 20 infants were treated from 2000 to 2009. Median ages at diagnosis and treatment were 5.75 months (range 0.25-31 months) and 6.5 months (0.5-32 months), respectively. Median duration of treatment was 31 months (1-102 months). Overall ventilator-free survival was 35% (7/20 infants), while 35% (7/20) died at a median age of 10 months (5.75-15 months) and 30% (6/20) were alive but ventilator-dependent. Endotracheal intubation for acute deterioration carried a high risk of failure of extubation and progression to long-term ventilation (LTV), but elective general anaesthesia, in contrast, was well tolerated. Overall outcome was worse than in the pivotal clinical trials; possible causes include later diagnosis and treatment in our patients and a higher incidence of infants at the severe end of the clinical spectrum. Careful consideration must be given to all possible outcomes, including LTV, before commencing enzyme replacement therapy in newly diagnosed infants.