Thoracic ultrasound to differentially diagnose the cause of an opaque hemithorax (whiteout) when patients are referred for respiratory physiotherapy: A service evaluation

Introduction

An opaque hemithorax, commonly termed a ‘whiteout’ on chest radiograph (CXR) often results in a referral for urgent respiratory physiotherapy due to patient respiratory com- promise. This referral assumes that sputum plugging of either main bronchus has resulted in a whole lung collapse. The subsequent physiotherapy treatment consists of sputum removal followed by lung re-expansion. There are, however, many alternative causes of an opaque hemithorax that would not respond to physiotherapy treatment. Making a differential diagnosis of an opaque hemithorax by CXR alone (Figure 1) proves difficult as it can be caused by pathologies of pleural, parenchymal, diaphragmatic and mediastinal origin (Table 1) (Wu et al. 1989; Yu et al. 1993; Hayward and Hayward 2019).

Referring medical professionals often use the position of the mediastinum, or more specifically the trachea on CXR to identify the cause of an opaque hemithorax (Murfitt 2002). A shift of the trachea towards the side of the opaque hemithorax (ipsilateral shift) is thought to indicate a main bronchus plug and lung collapse therefore justifying a referral to physiotherapy. A recent review highlighted that even when a tracheal shift is present it does not appear to be a reliable way to identify the underlying cause of an opaque hemithorax (Hayward and Hayward 2019). This presents a problem when, unbeknown to the physiotherapist, they may be referred a patient who has a clinical condition that will not respond to physiotherapy interventions.

Thoracic ultrasound (TUS) has the potential to more accurately differentiate between pulmonary pathologies and could be used to better differentiate between pathologies causing an opaque hemithorax prior to any physiotherapeutic interventions (Winkler et al. 2018) (Figure 2). A TUS scan performed by either the referring professional or the attending physiotherapist themselves would assist clinicians in identifying those patients with conditions amenable to physiotherapy interventions. The benefits of using a more accurate diagnostic approach would be two fold. Firstly, the patient would not experience any delay in receiving the appropriate treatment. Secondly, an inappropriately referred patient would not undergo any unnecessary, and potentially harmful, physiotherapy treatments.

The aim of our service evaluation was to establish if LUS assisted physiotherapists to identify the cause of an opaque hemithorax to a greater extent than CXR when patients were referred for respiratory physiotherapy.

Methods

This service evaluation had a prospective design including all patients referred within the six-month service evaluation period (1st February to 1st August 2018). This evaluation period was deemed a realistic target considering no additional funding or resources were received for this evaluation.

Data was collected from patients referred for in-patient respiratory physiotherapy across specialties at Blackpool Victoria Hospital presenting with an opaque hemithorax on CXR. Acute in-patient specialities included medicine, surgery, orthopaedics, neurology, paediatrics and cardiothoracics. Other patients within the hospital that presented with opaque hemithoracies but were not referred for respiratory physiotherapy were not evaluated. The physiotherapist receiving the medical team referral contacted a member of the project team to inform them of the referral. A member of the project team trained in thoracic ultrasound (SH, LH, NW or CT) then accompanied the physiotherapist to review the patient. Once consent was gained verbally, or a decision to treat the patient in their best interests was made, a TUS scan was completed to assist in identifying the cause of the opaque hemithorax. If indicated, a respiratory physiotherapy treatment was performed. If no physiotherapy treatment was indicated the findings from the new TUS were reported back to the referring clinician. It was planned that if no TUS trained clinicians were available to complete a TUS scan, then the attending physiotherapist treated as they assessed clinically appropriate so as not to delay treatment to the patient.

The physiotherapists performing the TUS scans have gained accreditation to perform TUS through the Intensive Care Society (United Kingdom) Focused Ultrasound in Intensive Care (FUSIC) programme and have two years’ experience of performing TUS.

Data collected included: the patient’s CXR showing the side of the opaque hemithorax and direction of any tracheal shift, documented reason for referral to physiotherapy, TUS scan findings, final medical team findings and the treatment or management strategy for the patient following all investigations for the opaque hemithorax. The cause for each of the opaque hemithorax was established retrospectively from the medical notes.

Results

A total of nine patients were included in this service evaluation. These referrals were received within the 6-month evaluation period (1st February to 1st August 2018). To the best of the authors’ knowledge these were the only patients in the hospital presenting with an opaque hemithorax on CXR and referred for respiratory physiotherapy during this time period (Table 2).

Five of the referrals (5/9) presented with ipsilateral shifts with tracheas deviated towards the side of the opaque hemithorax, which has historically indicated volume loss/lung col- lapse. The pathologies that resulted in these ipsilateral shifts were two patients with pleural effusions, one with atelectasis due to sputum plugging and two patients with pneumonic consolidation. The remaining CXRs show the tracheas to be in a central position (4/9), which has historically indicated a lack of lung volume change. The pathologies resulting in these central tracheal positions were an empyema, two pleural effusions and atelectasis due to sputum plugging. None of the patients referred had a CXR showing a contralateral shift where the trachea deviates away from the side of the opaque hemithorax (Figure 3). The main documented reasons for a referral for respiratory physiotherapy in these nine cases were sputum plugging, consolidation or lung collapse (Table 2).

All nine patients underwent a TUS scan within an hour of the referral for respiratory physio- therapy. In order of frequency, the primary findings on TUS were pleural effusion (4/9), atelectasis (2/9), consolidation (2/9) and empyema (1/9) (Table 3). The results of the TUS scans were reported to the medical teams who originally referred the individual patients. In four cases the TUS findings highlighted that physiotherapy treatments remained indicated and interventions were provided. In five cases the TUS scans highlighted findings that were not immediately amenable to chest physiotherapy such as pleural effusions and empyema. More advanced imaging investigations requested by the medical teams confirmed the TUS findings and highlighted additional underlying causes of the opaque hemithorax (Table 2).

Of the nine cases, four received respiratory physiotherapy treatment with two of these requiring escalation to a bronchoscopy to facilitate tenacious sputum removal; three cases had an inter-pleural drain inserted to manage a pleural effusion or an empyema and two of the patients were started on end-of-life care. At the time of writing eight of the patients had not survived to the end of the six-month evaluation period.

Discussion

The use of CXR to accurately differentiate between the causes of an opaque hemithorax has previously been questioned by Wu et al. (1989) and Yu et al. (1993). The use of the position of the trachea on CXR to accurately determine pathology and clinically justify the need for a respiratory physiotherapy referral also appears to be unreliable (Yu et al. 1993). In our evaluation all nine of the CXRs presented with an ipsilateral (5/9) or central (4/9) tracheal position. No referral was received by physiotherapy to review a patient with a contralateral shift. This suggests that the referring clinicians may be using the position of the trachea on CXR, along with the clinical picture, as one way to justify or exclude the need for a respiratory physiotherapy referral.

As can be seen from Table 2, our small patient group does not fit the historical pattern of tracheal position being associated with an underlying pathology causing the opaque hemithorax. In our evaluation sputum plugging and pleural effusions have both caused ipsilateral and central tracheal positions. Our patient group appears to substantiate the findings of Yu et al. (1993) and Wu et al. (1989) that using this method of differential diagnosis is unreliable. This may however just be a coincidence and more data from more opaque hemithoraces could be collected as part of a larger evaluation in the future.

It appears from our small patient group that only four of the nine patients had the potential to respond to physiotherapy interventions. The remaining five patients would have received unnecessary and ineffective physiotherapy treatments when the causes of the opaque hemithorax were due to pleural effusions or empyema. The use of physiotherapy- initiated TUS has allowed five patients to avoid receiving inappropriate treatments and in- stead facilitated well-timed clinically appropriate interventions. For the four patients with the potential to respond to physiotherapy, treatment was initiated in confidence knowing that the opaque hemithorax was most likely caused by whole lung atelectasis or consolidation, with other confounding pathologies having been ruled out.

Since the six-month period for this service evaluation there has been an informal change in how the acute in-patient physiotherapy team manage patients referred with an opaque hemithorax on CXR. If any acute in-patient area receives a referral for a ‘whiteout’ the receiving physiotherapist will contact one of the TUS accredited physiotherapists if there are any doubts about the potential cause of the patients ‘whiteout’. There are ongoing discussions regarding integrating TUS into the management of patients presenting with an opaque hemithorax, while balancing this with avoiding delays to potentially important time-sensitive physiotherapy interventions as part of a hospital trust quality improvement project.

The final diagnosis as to the most likely cause of the opaque hemithorax for each patient was taken from either the patient’s hospital discharge letter or the notification of death sent to the patient’s family doctor or general practitioner. Considering all of these patients had originally been referred for pathologies thought to be amenable to physiotherapy treatments, the medical findings show that only four of these were correct. The remaining five were beyond the scope of physiotherapy treatment with some needing further investigations for potentially serious pathology. Following the use of TUS by physiotherapy to differentiate between lung pathologies causing the opaque hemithorax, alternate medical techniques such as pleural drain insertion, advanced imaging and palliation were em- ployed to manage the patient’s clinical presentation. Without the use of TUS, it is possible some of these important decisions could have been delayed for many hours, if not days, resulting in potentially worse patient outcomes.

One aspect of this case series that was not predicted prior to its commencement was the mortality rate in these nine patients. It was understood prior to the initiation of data collection that an opaque hemithorax was a serious clinical finding on CXR, although it was not appreciated how potentially fatal this finding could be. Nine patients were referred for physiotherapy with an opaque hemithorax but eight of them did not survive to the end of the six-month data collection period. It would appear from our cohort that an opaque hemithorax on CXR is a clinical finding that represents the latter stages of some serious life limiting pathologies.

No funding was secured for this service evaluation, which has resulted in some limitations such as a small sample size of only nine patients. There is also potential inter-rater variability due to different physiotherapists performing the TUS scans.

Conclusion

It would appear that referring clinicians could be using the position of the trachea on CXR as an indication for a referral to physiotherapy. However, this method does not appear to accurately differentiate between underlying pathologies. As we predicted, physiotherapists can use TUS to more accurately identify the causes of an opaque hemithorax in order to confidently initiate physiotherapy interventions or limit delays in alternate treatment being provided when physiotherapy is not indicated.

Key points

• Caution should be used when using the position of the trachea to differentiate the underlying causes of an opaque hemithorax.
• Appropriately trained physiotherapists can use TUS to differentiate between the causes of an opaque hemithorax.
• Further work needs to be completed around how physiotherapy-initiated TUS will fit patient pathways when managing an opaque hemithorax.

Contributions

SH developed the service evaluation methodology. All authors were involved in performing patient scans. SH wrote the manuscript, LH, CT, DS and NW provided feedback on manuscript structure and content.

Acknowledgements

Thanks go to the physiotherapy department, critical care teams and the library services at Blackpool Teaching Hospitals NHS Foundation Trust.

Ethical Approval

Following application to Blackpool Teaching Hospitals NHS Foundation Trust R&D department ethical approval was not required for this service evaluation.

Funding

No source of funding was provided for this review.