Maximal inspiratory and expiratory pressure measurement in tracheotomised patients

Measurement of maximal inspiratory pressure (MIP) and maximal expiratory pressure (MEP) is the most widely used test to assess muscle pressures, as it has no adverse effects, is noninvasive and relatively easy to perform 1. Although this test has possible limitations mainly due to no cooperation or noncompliance, several studies have demonstrated that MIP and MEP may be very useful in the diagnosis and follow-up of pulmonary and cardiac disease 2, 3. A reduction in MIP has been shown to be associated with a progressive clinical worsening in patients affected with myasthenia gravis 4 or Guillain–Barré syndrome 5. In addition, it has been suggested that an imbalance between the pressure generated by the inspiratory muscles and the MIP may lead to the development of hypercapnia in patients with chronic obstructive pulmonary diseases (COPD) 6. Indeed, both MIP and MEP have been shown to be useful in detecting the presence of iatrogenic myopathy, such as steroid-induced myopathy 7 or in predicting the post-operative pulmonary complications following coronary artery by-pass graft surgery 8. All of the aforementioned studies have been conducted in patients breathing through their natural airways. An imbalance between respiratory mechanical load and respiratory muscle capability is also a frequent cause of difficulty when weaning the patients from mechanical ventilation. Therefore, a tracheotomy is usually performed 14–21 days after intubation 9. In these patients, the MEP has been shown to provide a reliable estimate of cough efficacy and efficient airways clearance 10. Likewise, MIP is a very sensitive predictive index of successful weaning even though its specificity is rather low 11.

As far as the current authors are aware, no studies have verified the hypothesis that different sites of pressure recordings (i.e. at the mouth or directly at the stoma), and breathing conditions (i.e. inflated or deflated cannula), may affect measurements of both MIP and MEP in tracheotomised patients.

Therefore, the primary aim of the present study was to find differences, if any, in the measurement of respiratory pressures in tracheotomised patients who are difficult to wean, depending on the site and breathing conditions of measurement. The secondary aim of the study was to assess whether an underlying diagnosis would affect the results of the measurements.

METHODS

The present study was conducted in accordance with the Declaration of Helsinki, and all hospital Ethics Committees approved the protocol. All patients gave informed consent after specific information on tests, its clinical utility and on modality to perform the measurements.

Protocol

The study was conducted on two different days. The four conditions under study shown in figure 1⇓ are as follows. 1) Through a cuff inflated cannula (condition A), MIP and MEP were measured directly through the stoma (fig. 1a⇓) with the inner balloon cuffed and without any inner tube. 2) Through the mouth with the same cannula as in condition A with the inner balloon deflated (condition B) and without any inner tube. MIP and MEP were measured through the mouth (fig. 1b⇓) simultaneously occluding the external hole of the cannula by a therapist's finger. The respiratory therapist was confident that the seal was complete if patients were able to speak. Particular attention was also paid to avoid any possible leak from the cannula or around the cannula itself. 3) Through the mouth with a phonetic uncuffed cannula (Biesalsky® Rüsch, Kemen, Germany; condition C). MIP and MEP were measured directly through the mouth (fig. 1c⇓) with a phonetic uncuffed cannula and occluding the external hole of the cannula with a specific phonetic cap. This cannula presented an inner tube inside. 4) Through the mouth after closure of the stoma (condition D). MIP and MEP were measured directly through the mouth (fig. 1d⇓) 1 h from decannulation. The tracheotomy wound was covered with an adhesive support. A soft manual pressure was maintained by another operator throughout the measurement, to avoid any possible leak. A respiratory therapist was confident that the seal was complete if patients were able to speak. Only after this test were the measurements performed.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1—

A representative patient under the four studied conditions. a) Through a cuff inflated cannula (condition A). b) Through the mouth with the same cannula (as in the first condition) with the inner balloon deflated (condition B). c) Through the mouth with a phonetic uncuffed cannula (Biesalky® Rüsch; condition C). d) Through the mouth after closure of the stoma (condition D).

Inflated, deflated and uncuffed cannulas were chosen so as to maintain the same inner diameter. No fenestrated cannulas were used during the study.

Condition D was not performed if MIP and MEP measured in other conditions were <20 and >30 cmH₂0, respectively. This was because patients were considered too weak to be safely decannulated. A 1-h pause was always allowed between the two conditions studied in each day. Patients showing clinical signs of respiratory failure relapse, desaturation <89% in spite of increasing oxygen supply (F_I,O2 >40%) or refusing to complete the study were considered drop-outs.

RESULTS

From the 50 enrolled patients, five subjects were excluded from the study due to lack of cooperation and seven due to the presence of anatomical tracheal deformities. The remaining 38 consecutive tracheostomised patients were then considered for analysis. Table 1⇓ shows the anthropometric data of the 38 consecutive tracheotomised patients. Of these, 23 were from the centre in Gussago, six from Padova, five from Gaiato and four from the Pavia centre.

Comparison of the four different sites and conditions of measurement

The results of the measurement of MIP and MEP, in the four conditions under study, are shown in figure 2⇓.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2—

Effect of different modality/site of measurement on a) maximal inspiratory pressure and b) maximal expiratory pressure. The best values were chosen among five repeated manoeuvres in each modality. A–D correspond to the conditions described in figure 1⇑. Boxes represent the inter-quartile range (25–75%) with the line representing the median. The bar lines represent minimum-maximum values. ***: p<0.001; ^#: p = 0.012, Bonferroni's adjustment for pairwise comparison; ^¶: p = 0.005; *: p = 0.05.

ANOVA showed a significant difference (p<0.001) of the measure of MIP depending on the condition of measurement. The measure of MIP taken at the cannula with the cuff inflated (condition A) was significantly higher when compared with the measurement at the mouth with the cuffed cannula and the inner balloon deflated (condition B: pairwise comparison with Bonferroni's adjustment versus A; p<0.001), to the measure at the mouth with a phonetic uncuffed cannula in site (condition C versus A; adjusted p<0.001) and to the measure taken at the mouth after closure of the stoma (condition D versus A; adjusted p = 0.012; fig. 2a⇑).

Similarly, ANOVA showed a significant difference (p<0.001) of the measure of MEP depending on the condition of measurement. MEP measured at the cannula with the cuff inflated (condition A) was significantly higher when compared with the measurement at the mouth with the cuffed cannula and the inner balloon deflated (condition B: pairwise comparison with Bonferroni's adjustment versus condition A; p = 0.005) and to the measurement taken at the mouth after closure of the stoma (condition D versus A; adjusted p = 0.049). No statistical difference was reached in the pairwise comparison between condition A and condition C (measurements taken at the mouth with a phonetic uncuffed cannula in site; adjusted p = 0.43; fig. 2b⇑).

The frequency of the best within subject's measurement of MIP and MEP among the five repeated efforts is shown in figure 3⇓. The highest frequency of the best measurement of MIP was observed in the fifth effort in all of the conditions, except in condition A (i.e. at the cannula with the cuff inflated) where the best value was more frequently observed at the fourth effort. The best measure has seldom been observed in the first two efforts, in all of the conditions.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3—

Frequency of the best within subject's measure of a) maximal inspiratory pressure and b) maximal expiratory pressure among the five repeated efforts in each modality/site of measurement. □: A; ░: B; ▒: C; └: D. A–D correspond to the conditions described in the Protocol section.

Likewise, during the MEP measurments, the highest value was more frequently observed in the fifth effort, in all of the conditions, compared particularly with the first and second effort.

DISCUSSION

The present study demonstrates that the assessment of respiratory muscles' pressures in critically ill tracheotomised patients differs significantly, depending on the measurement condition. In particular, the measurement through tracheotomy, with inflated cuff, yields higher values of both MIP and MEP, when compared with the other conditions. The highest value of the five efforts in each condition was more frequently observed in the last attempts of measurement, thus, indirectly demonstrating a learning effect of the manoeuvre. These results, obtained in a miscellaneous sample of difficult-to-wean tracheotomised patients, were not significantly different from when only the patients with COPD diagnosis were selected.

Several studies have demonstrated that the measurement of respiratory muscle force by means of MIP and MEP may be very useful in the diagnosis and follow-up of pulmonary and cardiac diseases 2, 3, as well as in patients affected with myasthenia gravis 4 or Guillain–Barré syndrome 5.

Likewise, the finding of abnormally low maximal respiratory pressures is able to predict the onset of post-operative pulmonary complications in patients undergoing coronary artery by-pass graft surgery 8. However, in COPD and asthma patients who have undergone prolonged steroid treatment, it is highly suggestive that there is a presence of iatrogenic myopathy 7. The pressure generated by the respiratory muscles to MIP ratio (inspiratory pressure/maximal pressure) has been shown to be significantly impaired in patients with COPD 6. Both the reduced capacity of the respiratory muscle (i.e. force) and an increased respiratory load (i.e. elastic and resistive) variably contribute to the development of respiratory muscle fatigue and to the impossibility of successfully sustaining spontaneous breathing in the critically ill patient. In this particular setting, it is of utmost importance that a reliable assessment of generated respiratory pressures constitutes, as far as possible, a real estimate of respiratory muscle force. The reliability of these measurements, which is highly dependent upon volitional effort, may be significantly reduced, particularly where lack of collaboration and compliance exist. This may represent a frequent case in the ICU setting, where patients with altered consciousness are admitted and sedation is often used. This also happens in the respiratory intermediate ICU. In fact, in the present study a quarter of patients didn't fulfil the inclusion criteria for enrolment in the study.

Despite the aforementioned limitations, respiratory muscle force assessment by measuring maximal respiratory pressures have been proposed, among others, as useful indices for decision making for extubation 14, 15, discontinuation from mechanical ventilation or removal of tracheal cannula 16 and assessing cough efficiency 10, 17 to start a specific rehabilitative programme 18, 19 as well as to assess its effectiveness 18.

Although a standardised approach to measurement of respiratory muscles strength and endurance has been proposed by the ATS and European Respiratory Society (ERS) 1, demographic and anthropometrical factors may variably contribute to the range of the normal values in healthy adults 20–26. Methodological factors may account for measurement variability among different subjects with possible confusing factors which contribute to the difficulty of both measurement and interpretation of respiratory muscles function 1, 17, 18, 22, 26–29. Finally, and more importantly, no particular technical issues and detailed instructions have been recommended by the ATS and ERS document regarding when and where respiratory pressures should be measured in patients not breathing through their natural airways.

In a questionnaire survey conducted among respiratory therapists from nine hospitals, Soo Hoo and Park 15 showed that, despite its clinical usefulness in the weaning process, a wide range of variation resulted from measurement of MIP by respiratory therapists in an ICU setting. Major issues were concerned with methods of obtaining respiratory pressures, lack of occlusion of the airway for the recommended amount of time, reproducibility, interpretation of the maximal effort and test-to-test variation. In that survey, 87% of respiratory therapists used a manometer, 38% using the ventilator's software and 25% using both of them 15. Moreover, the duration of airway occlusion time ranged from 2–4 s in 56% of the cases. Finally, 68% of respiratory therapists performed at least three measurements, with 75% of them only recording the highest value 15. The ATS/ERS 1 document suggested MIP measurement in intubated patients with poor cooperation should be performed using the so-called Marini method 30, whose reproducibility has been subsequently shown to be poor. Marini et al. 30 used a unidirectional valve to allow the output, but not the input of air. They also measured MIP starting from residual volume (RV) and performed 15–20 consecutive tests after 15–20 s of airway occlusion. In patients under mechanical ventilation, Sprauge and Hopkins 31 measured MIP using five to 10 consecutive repetitions, starting from RV and recording the average value among three. Bruton 32 performed the measurement according to the operator's decision and then recorded only the highest value. Caruso et al. 33 did not describe the method employed, while El-Khatib et al. 34 performed three tests and recorded the average. No uniformity exists in the measurement of maximal respiratory pressures in the intensive care setting and in particular in difficult-to-wean and consequently tracheotomised patients. It is obvious that any parameter can only be useful if it accurately reflects the patient's condition. Therefore, a call for further standardisation of measurement of respiratory pressures in these patients is surely needed.

The present study is the first, to the authors' knowledge, that assesses maximal respiratory pressure measurement in different patient's conditions, at the same time evaluating the effect of different tracheotomy cannulas and breathing conditions when compared with after-stoma closure.

The highest values of both MIP and MEP were obtained with the direct measurement at the cannula with the cuff inflated (condition A). This evidence is not surprising and is possibly explained by the fact that direct measurement of pressures at the trachea most likely resembles maximal oesophageal 35, 36 or cough gastric pressures measured by means of oesophageal and gastric balloons. In particular, clinical situations in which standard measurement of respiratory pressures is difficult or scarcely reliable, such as in patients with facial muscle weakness or bulbar dysfunction 37, 38, maximal sniff manoeuvre and cough pressures, are considered accurate and reproducible estimates of respectively inspiratory and expiratory muscle strength. Moreover, in the measurement at the trachea with the cuff inflated, air leaks can be reasonably ruled out. It is no surprise, then, that no statistically significant differences were otherwise observed among the other conditions studied. Despite a rigorous standardised procedure in the measurement of respiratory pressures, air leaks around the cannula, insufficient sealing by the therapist's finger occluding the cannula (when measuring at the trachea) or collapse of the upper airways (when measuring at the mouth) cannot be theoretically excluded. However, it seems that air leaks would be a major concern mainly in the measurement of MIP rather than MEP. Multiple comparisons among different conditions demonstrate lower levels of significance when measuring expiratory pressures than when measuring inspiratory pressures. The lack of statistical significance between the measure at the cannula with the cuff inflated and the measure at the mouth with the phonetic cannula would further confirm this observation.

As previously outlined, the number of repetitions of the measurement is extremely variable in the literature, ranging from three to 20 in the different studies. In the present study, the highest value of the measurement in each condition, both of MIP and MEP, was rarely observed in the first efforts. In contrast, the best performance was most frequently observed in the last two efforts, thus, suggesting a learning effect of the manoeuvre. A further improvement of the performance with increasing the number of efforts cannot be excluded. In fact, a major limitation of the present study is that the decision was made to set the limit to five efforts in each condition of study for both MIP and MEP measurement. On one hand, the current authors decided more efforts than in usual practice were needed (when respiratory therapists usually consider the best of three consecutive efforts), but on the other hand did not want to risk producing fatigue. In fact, a previous pilot study conducted on a limited sample of patients, such as those enrolled for the present protocol, had shown a lack of cooperation with more than five efforts. Other limitations of the study were that the respiratory therapists were not blinded to the study and those measurements were not recorded on a computer or a paper tracing.

In conclusion, the present study shows that respiratory muscles assessment differs significantly only when measurement is performed through the tracheotomy with the cuffed cannula, which yields the highest values of both maximal inspiratory pressure and maximal expiratory pressure measurements. Values of both maximal inspiratory pressure and maximal expiratory pressure obtained in all the other studied conditions of measurement (i.e. through the mouth with the cuff deflated and the cannula occluded, through the mouth with a phonetic cannula and, conventionally, through the mouth after stoma closure) were not statistically different. Finally, in this particular setting of patients, at least five measurements could be taken as a reasonable compromise between the possibility to achieve a better performance due to a learning effect and the risk of onset fatigue and lack of cooperation.