Assessment of expiratory flow limitation in chronic obstructive pulmonary disease: a new approach.

Patients with severe chronic obstructive pulmonary disease (COPD) usually experience expiratory flow limitation (EFL) during spontaneous breathing at rest [1]. As EFL reduces the effectiveness of expiration, it results in dynamic hyperinflation with consequent dyspnoea, which is one of the major complaints of patients with COPD. In these patients, the consequences of EFL are markedly increased during exercise [2]. Bearing in mind that EFL is a good predictor of dyspnoea in COPD patients [3, 4], simple methods for detecting EFL without perturbing normal breathing are of clinical interest. In 1995 KOULOURIS et al. [5] proposed a method to detect EFL based on the external application of a negative pressure at the mouth during tidal expiration (negative expiratory pressure (NEP)). EFL is detected by comparing the magnitude of expiratory flow before and after application of NEP: in case of EFL, the expiratory flow does not rise when increasing the driving pressure. More recently, NINANE et al. [6] proposed the estimation of EFL during spontaneous breathing with a similar approach. To increase the driving pressure during expiration, these authors applied a positive abdominal pressure by manual compression of the patient9s abdominal wall. When compared with the NEP technique, this approach has the advantage of instrumental simplicity, but suffers from the drawbacks of applying abdominal pressure of uncontrolled magnitude and lack of automatisation. Both procedures can be applied during breathing at rest as well as during exercise [7, 8]. In this issue of the European Respiratory Journal (ERJ), DELLACÀ et al. [9] present interesting novel data showing that the forced oscillation technique (FOT) can be useful for noninvasively detecting EFL during spontaneous breathing. The FOT, which was proposed in the 1950s [10], is based on applying a small-amplitude oscillation pressure at the mouth. Using the FOT the patient9s respiratory mechanics can be determined by simply recording the oscillatory pressure and flow signals at the mouth, provided that the oscillation frequency is much higher than the breathing rate. The relationship between the pressure and the flow oscillation is the impedance of the respiratory system, which has two components: respiratory resistance (Rrs) and reactance (Xrs). In a simple interpretation, Rrs is attributed to airways and tissue resistances, whereas Xrs is determined by the inertial and compliant properties of the respiratory system. The rationale of using FOT for assessing EFL is that under this flow regime the forced oscillation applied at the mouth cannot reach the alveoli given the fact that the choke point shifts upstream of the lung during EFL. Therefore, the effective impedance viewed from the FOT device corresponds to only a fraction of the respiratory system: from the mouth to the choke point. According to this hypothesis, the associated decrease in the effective respiratory compliance results in a reduction of the measured Xrs during expiration when compared with the non flow-limited inspiration. The notion that phasic changes in Xrs during the breathing cycle could be an index of EFL is not new. This was suggested in 1993 by PESLIN et al. [11] when studying mechanically ventilated COPD patients, and the hypothesis was more specifically substantiated in a controlled analog model [12] and animal studies [13]. However, the usefulness of the FOT for detecting EFL in spontaneously breathing COPD patients was not studied in detail before the work of DELLACÀ et al. [9] published in this issue of the ERJ. These authors analysed the sensitivity and specificity of FOT to detect EFL in each individual breathing cycle by computing the phasic changes in Xrs within the cycle. As a reference technique to assess whether EFL was present or not in the same breathing cycle, DELLACÀ et al. [9] used the MEAD and WHITTENBERGER [14] method. This is based on the graphic analysis of the loop determined by the flow signal and by the resistive component of the transpulmonary pressure signal, which is measured by means of an oesophageal balloon. The results reported by DELLACÀ et al. [9] strongly support their hypothesis that the FOT applied at a frequency of 5 Hz is a useful tool for detecting EFL in spontaneously breathing COPD patients. Specifically, the authors report that they were able to find thresholds for the indices representing the phasic changes in Xrs that resulted in a 100% sensitivity and specificity. Obviously, these excellent results, which were obtained from a research pilot study including a limited number of patients and controls, need to be further confirmed by more extensive studies. In this regard, a key issue is whether the thresholds found by the authors would keep their high sensitivity and specificity when applied to a wider sample of patients. However, according to the data and reasoning provided by the authors, it is plausible to expect further support for their conclusions when the technique is applied in future studies. Another major issue warranting future studies is to what extend the FOT allows determination of the part expiration under EFL. Moreover, the technique should be assessed during exercise, particularly to determine whether a FOT frequency of 5 Hz is sufficiently high to keep an acceptable signal-to-noise ratio given the expected increase in the patient9s breathing rate. The promising results reported by DELLACÀ et al. [9] as regards the application of the FOT to assess EFL also warrant further studies applying this technique and the NEP method in the same patients to compare the practical advantages and disadvantages of both approaches. Besides its specific interest as a technique to assess EFL, the work of DELLACÀ et al. [9] is also relevant to better clarify some issues on the interpretation of Rrs and Xrs values measured by the FOT in routine applications in patients who Unitat de Biofisica i Bioenginyeria, Facultat de Medicina, Universitat de Barcelona-IDIBAPS, Barcelona, Spain.