抄録
Two new indices for an objective sorting of discontinuous adventitious lung sounds (rales) were measured from time-expanded waveforms of the rales from twenty patients with fibrosing lung disease (FLD) and from eighteen patients with chronic airflow obstruction (CAO). One index is a period from a starting point of the waveform to its first peak (1/4 cycle duration: Tb) and the other is one to the third peak (9/4 cycle duration: Tf). Ten types of model rales from a lung-sound teaching tape by Murphy and a record by Ueda were also studied. To clarify pathophysiological meanings of these indices, body mass index (BMI) and some parameters for pulmonary function (VC, FEV1, DLCO, PaO2 and PaCO2) were measured simultaneously and a correlative study with Tb or Tf was performed.
1) These indices were easy to measure because the values were not affected by the trend of the baseline and showed good reproducibility.
2) On the Tb-Tf two dimensional plane, crackles from FLD patients (Fcr) and those from CAO patients (Ccr) were clearly divided to form an oval-shaped area individually separated by a vertical line Tb=0.5msec. “Crepitant rales” (Ueda) and “fine crackles” (Murphy) located in the lower portion of Fcr area, and the rest of the model crackles distributed over Ccr area widely.
3) Significant positive correlations were found between Tf and VC (p<0.05), and Tf and BMI (p<0.05) in idiopathic interstitial pneumonia, Tb and FEV1 (p<0.05), Tf and FEV1 (p<0.01), and Tb and BMI (p<0.01) in Ccr.
4) An experimental study proved that positive correlations between BMI and Tb or Tf are due to a filtering effect on higher frequency components proportional to the thickness of transmission media.
5) Correlations between spirogaphic data and these two indices were discussed in relation to the mechanisms of crackle generation and transmission.
6) Our indices seem to have better specificity for differentiating rales than the IDW method by Murphy. Because of simplicity in measuring and correlation with some functional parameters, these indices hold greater advantages for clinical applications.
