Article Text

Download PDFPDF

Indications for thoracic ultrasound in chest medicine: an observational study
  1. A R L Medford1,
  2. J J Entwisle2
  1. 1Department of Respiratory Medicine, Glenfield Hospital, University Hospitals of Leicester NHS Trust, Leicestershire, UK
  2. 2Department of Radiology, Glenfield Hospital, University Hospitals of Leicester NHS Trust, Leicestershire, UK
  1. Correspondence to Dr Andrew RL Medford, Department of Respiratory Medicine, Glenfield Hospital, University Hospitals of Leicester NHS Trust, Leicester LE3 9QP, Leicestershire, UK; andrewmedford{at}


Introduction Thoracic ultrasound (TUS) is increasingly used in chest medicine in secondary care. The indications for TUS are well known but less is known about their relative frequency. The purpose of this observational study was to describe the common indications for TUS and their relative frequency and the impact of TUS on management in a consecutive group of patients.

Methods 80 consecutive inpatients and outpatients referred for TUS by the same operator in a UK National Health Service teaching hospital were included. Demographic data, clinical indication, findings and effect of TUS on clinical management were noted.

Results The most common clinical indication was to assess a pleural effusion in 60/80 cases (75%), but other indications included assessment of diaphragmatic function, pleural thickening and chest wall masses. TUS significantly altered patient management in 52/80 cases (65%): it resolved equivocal chest radiograph (CXR) findings and excluded pathology in 20/80 cases (25%), detected effusions not visible on CXR in 14/80 cases (18%), localised a safe site for medical thoracoscopy in 11/80 cases (14%) when not clinically apparent, and detected unexpected septation in 7/80 cases (9%). TUS guided pleural cytology diagnosed pleural fluid metastases in 9/22 cases aspirated (41%).

Conclusion There are many clinical indications for TUS but the most common is pleural effusion assessment. TUS can diagnose inoperable pleural metastases, allow safe day case pleural intervention, exclude significant pleural pathology not visible on CXR, and triage further investigation.

  • thoracic ultrasound
  • pleural disease
  • intervention
  • lung cancer
  • chest imaging
  • risk management
  • thoracic medicine
  • respiratory tract tumours
  • ultrasound

Statistics from


Thoracic ultrasound (TUS) is increasingly used in chest medicine, especially for assessing pleural effusions.1 The British Thoracic Society guidelines highlight the role of TUS as a safe and accurate method of obtaining fluid for small or loculated effusions.2 TUS has become more commonly used because pleural procedures can result in complications, with 12 deaths and 15 serious complications noted in a 3 year period from 2005 to 2008.3 As a result, the National Patient Safety Agency has recommended that TUS is used for all pleural procedures except emergencies.

TUS helps pleural intervention. After a failed blind pleural tap (occurring in 20% of cases), TUS can help obtain fluid in 97–100% of cases in those with loculated or small effusions.4 5 TUS also improves the safety and accuracy of pleural puncture. Diacon et al6 noted 15% of proposed pleural puncture sites by clinical methods were inaccurate and that TUS prevented organ puncture in 10% of cases.6 TUS is also portable and can be performed at the bedside if necessary.

TUS can also provide further diagnostic information. Yang et al reported that septated (figure 1) and echogenic effusions (figure 2) are always exudates, whereas hypoechoic effusions can be transudates or exudates (figure 3).7 8 TUS can also reliably differentiate between fluid and solid pleural lesions (up to 93% in one study5) as well as guiding more invasive pleural procedures such as medical thoracoscopy.9 TUS can differentiate benign from malignant pleural disease, although this was conducted by two very experienced thoracic radiologists in a centre with high mesothelioma incidence and may not apply to every centre performing TUS.10

Figure 1

A right sided septated effusion in a 74-year-old man with suspected malignant mesothelioma, about to undergo medical thoracoscopy. Because of the septations (and evidence of adhesions at thoracoscopy), he did not receive a talc pleurodesis. A: Skin surface. B: Right pleural effusion. C: Septation. D: Collapsed lung. E: Cranial direction. F: Caudal direction.

Figure 2

A 62-year-old man with known non-small cell lung cancer (NSCLC) and a right upper lobe mass with a large right pleural effusion (13 cm depth). There is internal echogenicity in the posterior chest. A: Pleura. B: Right pleural effusion. C: Internal echogenicity within effusion. D: Right hemidiaphragm. E: Liver. F: Caudal direction. G: Cranial direction.

Figure 3

A 13 cm right sided hypoechoic effusion in a 62-year-old woman with an adnexal lesion. A: Skin surface. B: Right pleural effusion. C: Right hemidiaphragm. D: Liver. E: Caudal direction. F: Cranial direction.

Although the indications for TUS are relatively well known, there are no published studies to our knowledge that have specifically looked at the indications for TUS in secondary care. The objective of this observational study is to describe the common indications for TUS and their relative frequency in a consecutive group of patients, as well as demonstrating how TUS can alter patient management and promote patient safety. The study centre was a UK National Health Service teaching hospital with six consultant thoracic radiologists, 14 consultant respiratory physicians, 520 beds and 63 respiratory beds.


Study population

Eighty consecutive inpatients and outpatients referred to a thoracic radiology department ultrasound list underwent TUS in a 6 month period between October 2008 and March 2009 in a UK NHS teaching hospital. Patients were consecutively referred to a TUS list for a single operator in the radiology department at a single centre. Patients were identified by the request for ‘TUS’ on the request form. All TUS requests were included consecutively for inpatients and outpatients. None of the patients were known to have malignancy at the time of pleural intervention. All effusions had a prior chest radiograph (CXR).


TUS was performed by ARLM (a post-Certificate of Completion of Training (CCT) respiratory and general medicine physician having completed level 1 Royal College of Radiologists (RCR) training in TUS and completing further training in interventional pulmonology) and were verified at the time of scanning by JJE (a consultant thoracic radiologist and level 3 RCR TUS operator) or experienced sonographer. Level 1 RCR training requires 3 months training with a thoracic radiologist and scanning and interpreting a required number of common indications for TUS.11

Criteria for assessing effusion size by TUS

The size of effusions on TUS was defined according to the depth of measurement: ‘small’ (under 5 cm), ‘moderate’ (5–10 cm) and ‘large’ (over 10 cm). All ‘small’ effusions were not visible or equivocal on CXR.

Criteria for assessing clinical impact of TUS

The impact of TUS on clinical management was regarded as significant if one of the following criteria was met:

  1. Resolution of equivocal changes on CXR, (ie, exclusion of effusion, pleural thickening and diaphragmatic dysfunction)

  2. Detection of effusions not visible on CXR

  3. Localisation of safe/optimal sites for lateral decubitus thoracoscope insertion when not clinically apparent

  4. Detection of significant unexpected septated effusions prompting the need for further imaging or intervention and avoidance of pleurodesis

  5. Resolution of equivocal clinical examination findings (ie, chest wall masses).

Data collection

Data were prospectively collected including patient age, gender, the indications for TUS, the size of any effusion on TUS (see below) and whether TUS altered patient management (see below). The impact of TUS was assessed by review of the medical records, other radiological investigations and pleural cytology results. Mean values are reported with SD in parentheses.


The mean age of the 80 patient cohort was 65 years (SD 15.8) with a male preponderance (52/80, 65%). The majority of patients were outpatients (55/80, 69%).

The clinical indications for TUS and findings are shown in table 1. The most common indication for TUS was pleural effusion assessment in 60/80 cases (75%) followed by diaphragm assessment in 12/80 cases (15%). The mean size of scanned effusions was 1.75 cm (SD 0.35) (small, n=12), 7.6 cm (SD 0.42) (moderate, n=5), and 11.8 cm (SD 1.99) (large, n=12). All the small effusions were not visible or equivocal on CXR. Two other moderate effusions were detected when not expected on CXR.

Table 1

Breakdown of clinical indications for TUS and findings

Table 2 demonstrates a breakdown of 52/80 cases (65%) where TUS significantly altered patient management. Most commonly, TUS resolved equivocal CXR changes in 20/80 cases (25% overall) and excluded mainly pleural pathology (most commonly effusion) or diaphragm dysfunction. In 14/80 cases (18%), TUS detected an effusion not visible on the CXR; in seven cases this was too small to sample even under TUS, and the remainder were sampled safely under TUS. In 11/80 cases (14%), TUS located the best site for medical thoracoscopy when this was not clinically apparent, allowing safe and effective access. In most of the remaining 7/80 cases (9%), TUS detected unexpected septated effusions (figure 4) not seen on computed tomography (CT) (figure 5). This altered the need for drainage, decisions about pleurodesis, and referral for decortication (in 1 case).

Table 2

Breakdown of cases where TUS finding changed patient management significantly

Figure 4

A 74-year-old man with a septated right sided pleural effusion with possible malignant features on CT awaiting medical thoracoscopy. The septation was not noted on CT (see figure 5). As a result, talc pleurodesis was not performed, and the patient underwent early decortication. A: Skin surface. B: Right pleural effusion. C: Septation. D: Right hemidiaphragm. E: Liver. F: Caudal direction. G: Cranial direction.

Figure 5

Corresponding CT for figure 4. Minimal septation noted on CT. Nodular pleural enhancement of visceral and parietal pleura noted. A: Right pleural effusion with minimal septation. B: Collapsed right lung. C: Parietal pleural thickening. D: Visceral pleural thickening.

The results of TUS guided diagnostic pleural tap, performed in 22 of the 29 effusions, are shown in table 3. TUS pleural cytology diagnosed metastatic malignancy in 9/22 cases (41%) of effusions aspirated in patients without a diagnosis of malignancy.

Table 3

Breakdown of diagnostic tap results (no evidence of known malignancy for all cases)


This UK urban secondary care study has demonstrated that pleural effusion assessment is the most common indication for TUS (in 60/80 cases, 75%). Other less common indications are assessment of diaphragm function, pleural thickening and chest wall masses. We have shown that clinical management was affected by TUS in 52/80 cases (65%), most commonly resolving equivocal CXR findings, allowing safe sampling of effusions not visible on CXR, allowing safe access for medical thoracoscopy, and detecting unexpected septation. Finally, TUS pleural cytology diagnosed pleural metastatic disease in 9/22 cases sampled (41%).

The clinical implications of this study are that providing a TUS service in secondary care is worthwhile on diagnostic, therapeutic and safety grounds. In particular, the major benefit of TUS is in clarifying and assessing the presence of pleural effusion when this has not been possible on a CXR. Pleural effusions (or CXR appearances mimicking pleural effusion) can occur in medical and surgical patients for a myriad of reasons. With an ageing population with increased comorbidity but prolonged survival, there is likely to be an ever increasing demand for TUS to assess pleural effusions, allow safe sampling, and diagnose or direct further investigation.

The indications for TUS are well known11 but there are fewer data on their relative frequency. We have demonstrated the superiority of TUS to CXR for assessing pleural effusion and diaphragm function. TUS should be seen as complementing CXR. TUS also allows safe pleural intervention such as aspirating small effusions. Detection of septated effusions alters further treatment including pleurodesis and the need for further procedures.

The diagnostic yield of TUS pleural fluid cytology in this study is lower than that in other series2 because of the variable pre-test probability and an unselected population in this study. However, TUS diagnosed malignancy unsuitable for radical treatment in 9/22 cases (41%). The high proportion of cases (52/80, 65%) in which TUS altered management is consistent with other studies for example Yuan et al12 who demonstrated TUS altered clinical management in 45% and contributed to management of 82% of patients, albeit in an emergency department setting with acute chest complaints and CXR opacities.

We acknowledge the limitations of our study. This was a single centre observational study. Our study size was therefore small and we cannot exclude selection bias. We did not assess service demand or perform a cost analysis, but TUS may be cost saving by allowing day case investigation and by avoiding other diagnostic tests.

In summary, our study demonstrates the indications for TUS and their relative frequency in a UK NHS teaching hospital. TUS significantly impacts on patient management to improve diagnosis, direct further investigation or treatment, and improve safety of pleural procedures.

Key learning points

  • The main indication for TUS in respiratory radiology departments is for assessment of pleural effusions.

  • Other less common indications include assessment of pleural thickening, diaphragm function and chest wall masses.

  • TUS can significantly alter patient management by resolving equivocal CXR changes, detecting effusions not visible on CXR, locating a safe site for pleural intervention or detecting unexpected septation.

  • TUS guided pleural cytology can diagnose pleural metastasis in a significant proportion of patients as the first diagnosis.


We acknowledge Jody Welton, Alison Rathod (sonographers) and other staff at the Department of Radiology, Glenfield Hospital.



  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

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.