OBJECTIVE--DDD pacing is better than VVI pacing in complete heart block and sick sinus syndrome but is more expensive and demanding. In addition, some patients have to be programmed out of DDD mode and this may have an important impact on the cost-effectiveness of DDD pacing. The purpose of this study was to determine how many patients remain in DDD mode over the long term (up to 10 years). DESIGN--A retrospective analysis of the outcome over 10 years of consecutive patients who had their pacemakers programmed initially in DDD mode. SETTING--A district general hospital. PATIENTS--249 patients with DDD pacemakers. Sixty two patients (24.9%) had predominantly sick sinus syndrome and 180 (72.3%) had predominantly atrioventricular conduction disease. Mean (range) complete follow up for this group of patients was 32 months (1-10 years). RESULTS--Cumulative survival of DDD mode was 83.5% at 60 months. Atrial fibrillation was the commonest reason for abandonment of DDD pacing. Atrial fibrillation developed in 30 patients (12%), with atrial flutter in three (1.2%). Loss of atrial sensing or pacing, pacemaker mediated tachycardia, and various other reasons accounted for reprogramming out of DDD mode in eight patients (3.2%). Overall, an atrial pacing mode was maintained in 91% and VVI pacing was needed in only 9%. CONCLUSIONS--With careful use of programming facilities and appropriate secondary intervention, most patients with dual chamber pacemakers can be maintained successfully in DDD or an alternative atrial pacing mode until elective replacement, although atrial arrhythmia remains a significant problem. There are no good reasons, other than cost, for not using dual chamber pacing routinely as suggested by recent guidelines and this policy can be achieved successfully in a district general hospital pacing centre.