The present study compared four different sites and conditions for the measurement of maximal inspiratory pressure (MIP) and maximal expiratory pressure (MEP) in 38 spontaneous breathing tracheotomised patients. Of the patients, 28 had chronic obstructive pulmonary disease (COPD). The four different conditions were: 1) through a cuff inflated cannula (condition A); 2) through the mouth with a deflated cannula (condition B); 3) through the mouth with a phonetic uncuffed cannula (condition C); and 4) through the mouth after stoma closure (condition D). Five trials in each condition were performed using a standardised method. The measurement of both MIP and MEP differed significantly depending on the condition of measurement. MIP taken in condition A was significantly higher when compared with conditions B, C and D. MEP in condition A was significantly higher when compared with condition B and D. In condition A the highest frequency of the best measurement of MIP and MEP was observed at the fourth and fifth effort, respectively. The same results were obtained after the selection of only COPD patients. In conclusion, respiratory muscle assessment differs significantly depending on measurement condition. Measurement through inflated cannula tracheotomy yields higher values of both maximal inspiratory and maximal expiratory pressure.