Article Text

Download PDFPDF

Squawks in pneumonia
  1. R Paciej,
  2. A Vyshedskiy,
  3. D Bana,
  4. R Murphy
  1. Faulkner/Brigham and Women Hospitals, Boston, MA, USA
  1. Correspondence to:
    Dr R Murphy
    1153 Centre St, Suite 4990, Boston, MA 02130, USA;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Squawks are short inspiratory wheezes that have been described in hypersensitivity pneumonitis and other fibrotic disorders. Little attention has been paid to the fact that they also occur in patients with pneumonia. In the course of studying the correlation of automated lung sound analysis with disease states in patients at a community teaching hospital, we noticed that squawks appeared to be more common in patients with pneumonia than we expected. We therefore examined the occurrence of squawks more systematically in 500 subjects who had been examined with a multichannel lung sound analyser (Stethographics Model STG-1602), as previously described.1

Seventy eight of the subjects in this population had a clinical diagnosis of pneumonia. All participants had been asked to breathe more deeply than normal with their mouths open. Two 20 second samples were taken. The Institutional Review Board of the Faulkner Hospital approved the study. Two experienced observers, blinded to the clinical diagnosis, used playback and waveform displays to identify squawks. They were defined according to the criteria initially described by Earis et al and adopted by CORSA.2,3 All channels from each subject were replayed and the waveforms of the data in the time domain were simultaneously examined. Only those sounds which fit both the auditory and waveform characteristics were considered to be squawks for the purposes of this study.

Squawks were present in 12 of the 78 patients with pneumonia and in none of 224 patients considered to have no significant lung disease. They were found in four of 18 patients with interstitial pulmonary fibrosis, two of 41 patients with bronchial asthma, one of 79 patients with COPD, and in none of 56 patients with congestive heart failure. We also noted squawks in a patient with radiation pneumonitis and in one of the two patients in our database with hypersensitivity pneumonitis.

In nine of the 12 patients with pneumonia the squawks were in the same location as the radiological opacifications. In one patient the squawk was in a different location and in another the chest radiograph did not show evidence consistent with pneumonia until 4 days after the squawk was detected. In the remaining patient a squawk was heard over the left posterior mid chest. The portable chest radiograph was interpreted as technically suboptimal due to the patient’s body habitus. The clinicians caring for this patient made a diagnosis of pneumonia and treated him accordingly.

Interestingly, one patient who clearly had congestive heart failure on a number of occasions had a squawk when we examined him. On re-examining his record he also had a diagnosis of pneumonia on that day. Similarly, there were two patients with squawks (one with COPD and one with asthma) who, on the day that the squawks were noted, had acute febrile illnesses consistent with pneumonia. In one patient the presence of a squawk led to reinterpretation of the chest radiograph as showing an area of opacification consistent with pneumonia.

The squawks in this study all had a distinctive sound that is readily distinguished from crackles, rhonchi, rubs, and most wheezing noises. Occasionally wheezes can be short and have a similar sound, but this occurs rarely as an isolated finding in inspiration. The squawks in this study all had sinusoidal waveforms as illustrated in the time-amplitude plot shown in fig 1. The mean (SD) duration of these sounds was 64 (49) ms (range 16–228) and the mean (SD) frequency was 425 (110) Hz (range 200–667). These findings are similar to those of Earis et al.3

When a squawk is accurately identified, the question arises—what does it mean? In a patient who is not acutely ill, investigations to rule out hypersensitivity pneumonitis and the other fibrotic conditions mentioned above should be considered. In a patient with a clinical picture consistent with pneumonia, the presence of a squawk offers some objective evidence to support the diagnosis. It would seem reasonable to suggest that the patient should be followed up to be sure that the squawk disappears when the acute illness resolves to exclude the possibility that the acute illness was the mode of presentation of a chronic pulmonary disorder. Squawks can be helpful in providing evidence for pneumonia in lung areas where radiological visualisation may be suboptimal, such as below the dome of the diaphragm or in the retrocardiac region.

In summary, short inspiratory wheeze-like sounds are found in pneumonia. Other conditions that can cause them are chronic restrictive disorders but these are relatively uncommon compared with pneumonia. When there is no evidence of these restrictive disorders and an acute syndrome consistent with respiratory infection is present, squawks can provide relatively specific—although not very sensitive—evidence of pneumonia.

Figure 1

Time-amplitude plot of a sound recorded at the lung bases posteriorly (channel 14) in a patient with pneumonia. Waveforms are presented in both the unexpanded (A) and expanded (B) modes. (A) The unexpanded waveform shows one full breath. The solid bars above the unexpanded wave mark the respiratory cycle (the light bar indicates inspiration and the dark bar indicates expiration). The normal inspiratory sound can be seen as having almost random waveform fluctuations. Fine crackles (c) look like spikes on an unexpanded waveform. A squawk is present at the end of inspiration. One fine crackle and a wheeze are present during expiration. (B) The expanded squawk waveform exhibits monophonic sinusoidal fluctuations lasting approximately 60 ms. (C) The squawk sound in the frequency domain shows a single peak at 600 Hz. In this patient similar squawks appeared in three consecutive breaths during 20 s of deeper than normal breathing. They occurred approximately in the same location on the chest and at the same time in the respiratory cycle. These observations were typical of the squawks detected in our study. FFT = fast fourier transform.


View Abstract