Objective. To extend knowledge relevant for the individualized optimization of training in the context of rehabilitation and sports by presenting results on exercise and training in hand cycling, a promising mode of upper body exercise for wheelchair-dependant patients. Major content of paper. The transfer from leg to arm work for daily ambulation that accompanies a lower leg impairment stresses the importance of regular upper body exercise to maintain fit and healthy in daily life. A very suitable mode of exercise to train the upper body is hand cycling, but research is still scarce. Not much is known on training parameters of hand cycling such as peak oxygen consumption (VO2), peak power output (PO) and peak heart rate (HR), and no training guidelines exist specific for upper body exercise. More general guidelines of the ACSM for healthy physical activity prescribe a minimum intensity of 40%VO2max or 55%HRR, but are aimed at the larger muscle mass of the lower body in healthy individuals. No data on patients are included. A first step in developing individualized training guidelines in the context of rehabilitation and sports is to study training parameters in hand cycling in healthy individuals and in patients. 7 Healthy subjects conducted an incremental hand cycling test. Their exercise capacity was described in terms of peak PO, peak VO2 and peak HR. As a next step, training parameters of wheelchair dependant persons are included: How do peak VO2, peak PO and peak HR in a mixed group of persons with paraplegia (PP: Th6, th12, th12), amputees and multitrauma (n=6) and a group of persons with tetraplegia (TP; n=22), measured during an incremental test, relate to healthy subjects? Results. Peak values for VO2, PO and HR for healthy subjects were 2.56±0.32l/min, 143.0±18.0W. 169±12bpm and are much lower than in lower body exercise (maximal values for cycling in healthy young males are: 4.1±0.2l/min, 345.0±22.9 and 189±5bpm respectively, n=8). Values for the mixed group were 2.14±0.43l/min, 111.0±16W and 172±5bpm and for TP: 1.21±0.32l/min, 38.4±16.7W and 122±16bpm. Though values for paraplegics during hand cycling approached those of healthy subjects, peak values were clearly lower in TP. Conclusion. Since only a relatively small muscle mass can contribute to propulsion, peak VO2, PO and HR are clearly lower in hand cycling compared to lower body cyclic exercise. This might have consequences for optimal training and more research is necessary to be able to individualize training of the upper body. A second important aspect in individualizing training is that patients might have a lower physical capacity than healthy subjects. Based on the present data, PP (Th6 and lower) reach values comparable to healthy subjects, important for generalizing arm training of healthy subjects to patients. Peak values in TP are clearly lower and must receive special attention in training.