Klaske van Sluis

Expiratory muscle strength training 123 at baseline [20]. Palmer et al. [24] reported a good MEP at baseline in their group after partial laryngectomy and reasoned their candidates must have good pulmonary support to tolerate some amount of aspiration during recovery. We observed no changes over time for PEF outcomes. Contrary to the MEP scores, baseline scores were lower than predicted for seven participants com- pared with reference values [37]. No normative values for PEF in TL patients are found. In the group of partial laryngectomy patients performing EMST a significant increase in peak cough flow (L/min) was found, from below normal prior to the intervention to within normal range after training.24 Peak expira- tory flow rate also increased significantly with training in a group of elderly.26 It is worth questioning why PEF values did not improve in the studied group of TL participants. A possible reason is impaired cough technique. As there is no glottic closure, the higher built up pressure does not lead to increased flow. Besides, no special attention was paid at the exhaling technique (i.e. generating the force primarily using the rectus abdominis muscle, and keeping the thorax maximally expanded during the first part of the forced expiration) during the training. The increase in MEP and vocal range in dB did not lead to an improvement in the clinical relevant outcome measures regarding pulmonary function, physi- cal exertion, fatigue, and vocal functioning. It was disappointing to see that no clinically relevant benefits were found in the self-reported outcome measures. This might be the result of a sample of relatively fit TL participants which showed high baseline values in MEP and who mentioned no specific pulmonary complaints. Despite an improvement in loudness, related to the dynamic range in dB after four weeks of training, no changes in self-reported vocal functioning (VHI-10) were seen. Although this pilot study offered useful insights, there are some limitations which should be mentioned related to the included group, use of EMST device and performance of training and measurements. Because of the small number of participants the results of this first EMST study in a TL group must be interpreted with caution. The outcomes cannot be generalized to the entire TL population. In particular, because the participants in our sample were all male, all fluent tracheoesophageal speakers, and relatively fit from the start, they may have progressed less and perceived less benefit, compared to what might be expected from TL patients who are less fit and report explicit coughing problems. Participants who achieved a MEP-score above 187.5 cmH 2 O contin- ued the training with the device set at the maximum work load of 150 cmH 2 O. It remains unclear if the increase in MEP would have been even larger in case of training with an EMST device with a wider range. To measure the effect of EMST with manometry, spirometry and CPET, adjustments were needed for use on the tracheostoma. Problems creating an airtight seal and plugging the voice prosthesis and the resulting air leakage could have influenced the results of the training and the measured outcomes negatively. VHI-10, SFQ, CCQ and Borg scales were best available questionnaires, although not specifically validated for TL population, and should therefore be interpreted with caution.