Show Summary Details
Page of

Imaging the respiratory system in the critically ill 

Imaging the respiratory system in the critically ill
Imaging the respiratory system in the critically ill

Lawrence R. Goodman

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2022. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 18 May 2022

Key points

  • Routine radiographs are not cost effective in the ICU setting.

  • Most published guidelines agree that radiographs are worthwhile after insertion of tubes or catheters, and in patients receiving mechanical ventilation. Otherwise, they are required only for change in the patient’s clinical status.

  • Picture archiving and communication systems utilize digital imaging technology. They provide superior quality images, rapid image availability at multiple sites, and fewer repeat examinations, reducing both cost and patient radiation.

  • Disadvantages of picture archiving and communication systems include expensive equipment and personnel required to keep them functioning.

  • The majority of chest X-ray abnormalities in the ICU are best understood by paying careful attention to the initial appearance of the X-ray in relation to the patient’s onset of symptoms and the progression of abnormalities over the next few days.


In the 1960s and 1970s when intensive care units (ICUs) were novel, both clinicians and radiologists relied on the routine daily portable radiograph for patient care. As both groups gained sophistication in dealing with these critically-ill patients, and their various tubes and monitoring apparatus, many began to question the need for daily radiographs. Yet daily radiographs persist in many hospitals today despite rising concerns about radiation exposure (especially in patients under 40) and the rising costs of health care. There have been numerous studies looking at differences in outcome between daily and on-demand radiographs. Despite the wide differences in study design, patient mixes, and countries of origin, the results all point in the same direction [1,2,3,4,5,6,7,8,9]. Changing to on-demand radiographs decreases utilization by approximately 25–35% without measurable difference in outcome parameters (ICU or hospital mortality, length of stay, duration of mechanical ventilation, delay in addressing major unsuspected problems, etc.). Two recent meta-analyses [10,11] have confirmed these findings.

Although routine radiographs may not be rewarding, there are many scenarios where daily images are beneficial. The American College of Radiology has recently revised their guidelines for portable radiographs. Many of the recommendations are based on the articles discussed [1,2,3,4,5,6,7,8,9,10,11] with the ACR citing additional references where appropriate. Specifically, the ACR guidelines discuss four broad categories of clinical scenarios and present justification for their recommendations. Numerical ratings were also applied (less than 3 usually not appropriate, 4–6 may be appropriate, greater than 7 usually appropriate). These recommendations are for portable radiographs only, which produce a relatively low level of radiation (less than 0.1 mSv) [12].

  • Admission or transfer to the ICU—recommendation: new patients should have an admission radiograph (rating = 7).

  • Stable patients with no change in their clinical status. There is little justification in routine daily radiographs. Recommendation: radiographs should be ordered for change in clinical status only (rating = 3).

  • Insertion or tube or catheter

    • Endotracheal tube—between 12–15% tubes are misplaced and malpositioning is seldom detected clinically. Recommendation: chest x-ray should be obtained immediately after intubation (rating = 9).

    • Following intravenous catheter insertion—the overall pneumothorax rate is approximately 10%. In addition, chest x-rays frequently show malpositioned catheters with subclavian vein catheters twice as likely to be malpositioned compared with jugular venous catheters. Recommendation: radiograph after central venous catheter insertion (rating = 9). Follow-up only for suspected complications only.

    • Swan-Ganz catheters—malpositioned catheters are frequent (greater than 20%) while pneumothorax is unusual (2%). Recommendation: radiograph after insertion. Once pneumothorax is excluded and proper position demonstrated, repeat for clinical indications only (rating = 9).

    • Nasogastric tube—significant malposition is uncommon (less than 1%). Recommendation: chest x-ray after initial insertion and before first feeding (rating = 9). Follow-up not required for stable position.

  • Chest tube

    • At the time of initial insertion approximately 10% of chest tubes are not in “ideal” position. Many of these “malpositions” are clinically unimportant. Recommendation: chest x-ray on insertion to evaluate position, success of treatment, and complications related to intubation (rating = 9). Routine follow-up not required.

    • Chest tube removal—pneumothorax complicates a small percentage of extubations and they are usually clinically apparent. There is seldom need for reinsertion of the tube. Recommendation: repeat chest x-ray for clinical indications only (rating = 5).

Another approach can be found in a survey of 190 intensivists from 34 ICUs (using the Delphi method) on the value of routine radiographs in 29 ICU scenarios (10 new medical devices, 10 existing medical devices and nine clinical situations) [13,14]. The clinicians’ recommendations were similar to the radiologists’ recommendations. A routine radiograph was considered appropriate after any tube insertion, except a nasogastric (NG) tube for reasons other than nutrition. There was strong, but not universal support for daily X-rays for patients with ARDS or oedema receiving mechanical ventilation, chest tubes, and for patients receiving non-invasive ventilation with a PaO2/FiO2 ratio <200 mmHg (Table 78.1).

Table 78.1 Intensivist consensus on need for chest X-ray for support and monitoring catheters


CXR after



CXR after



Endotracheal intubation



Non-invasive ventilation with PaO2/FiO2 ratio of ≤200 mmHg




In-dwelling chest tube


Subclavian central venous catheter



Invasive mechanical ventilation for ARDS



Internal jugular central venous catheter



Invasive mechanical ventilation for haemodynamic pulmonary oedema


Pulmonary artery catheter



In-dwelling endotracheal tube



Transvenous pacing lead



Non-invasive ventilation with PaO2/FiO2 ratio of >200 mmHg



Ballasted NG tube



In-dwelling pulmonary artery catheter



Chest tube



In-dwelling temporary transvenous pacing lead



NG tube for enteral nutrition



NG tube for enteral feeding



Non-feeding NG tube



Superior vena cava system catheter


* Attitude widely accepted.


0, not indicated.

Data from Hejblum G, et al., ‘A web-based Delphi study on the indications of chest radiographs for patients in ICUs’, Chest, 2008, 133(5), pp. 1107–12.

In four clinical scenarios, there was support for daily radiographs in patients with haemodynamic instability and ventilated, invasive ventilation for status asthmaticus, mechanical ventilation for respiratory failure in the immunocompromised, and non-invasive ventilation for respiratory failure in the immunocompromised (Table 78.2).

Table 78.2 Intensivist consensus on need for daily chest X-rays for specific clinical scenarios


Patient with



Invasive mechanical ventilation for haemodynamic instability


Invasive mechanical ventilation for status asthmaticus



Immunocompromised: invasive mechanical ventilation for acute respiratory failure


Immunocompromised: non-invasive ventilation for acute respiratory failure


Non-invasive ventilation for haemodynamic pulmonary oedema



Invasive mechanical ventilation for acute-on-chronic respiratory failure



Non-invasive ventilation for acute-on-chronic respiratory failure



Invasive mechanical ventilation for neurological or toxic coma, nl. respiratory function



CXR before extubation


* Attitude widely accepted.


0, negative.

Data from Hejblum G, et al., ‘A web-based Delphi study on the indications of chest radiographs for patients in ICUs’, Chest, 2008, 133(5), pp. 1107–12.

In an accompanying editorial, Lessnau [15] correctly pointed out that within each group of catheter insertions and monitoring, there are subgroups where commonly requested radiographs may not be necessary (e.g. following uncomplicated bronchoscopically-guided percutaneous tracheostomy, etc.). He also pointed out that, in some countries such as the USA, there are conflicting financial incentives. If the hospital bills for the radiograph, demand-only X-rays will decrease revenues. If the radiograph is part of a global fee, then there is an incentive to decrease daily films. In 1998, Graham et al. [5]‌ estimated that radiology expenses were 3.5% of overall inpatient costs. At their 1000 bed hospital in 1 year, over 69,000 portable films were obtained at a mean charge of $114.00 per examination; the portable radiograph bill was $8,000,000. In 100 patients after thoracic surgery, they estimated that over 80% of radiographs could have been eliminated, saving $286,000 ($725/patient).

Do on-demand radiographs impact ordering of higher level imaging, such as computed tomography (CT) and ultrasound (US)? Kröner et al. [16] demonstrated in two consecutive five-month periods that the number of X-rays per patient per day dropped from 1.1 to 0.6 with the elimination of routine radiographs. Length of stay and mortality rates were unchanged. During the on-demand period, the number of CTs ordered did not increase, although there was a slight increase in the number of USs. In both periods, 38% of CTs or USs resulted in a change of therapy.

Digital imaging, picture archiving, and communication systems

At major hospitals centres, digital receptors of various kinds have replaced the traditional film screen cassette, and picture archiving, and communication systems (PACS) have replaced the traditional view box. Although the initial investment is steep (but decreasing) the advantages appear overwhelming [17,18].

  • Digital images are available within minutes for evaluation in radiology and the ICU. Monitors on newer portable equipment provide almost instantaneous bedside viewing for image quality and for STAT diagnosis.

  • Manipulation of the digital image contrast, magnification, and varied processing algorithms enhance diagnosis (e.g. pneumothorax, catheter placement, retro cardiac disease, etc.) and diminish the number of repeats, saving money and patient radiation.

  • Serial digital portable radiographs and other studies (CT, US) are easier to compare on monitors. This is critical in understanding short- and long-term trends of disease. Images can be viewed in multiple locations simultaneously.

  • Digital storage eliminates lost films, and images are easily transportable (disc or electronically) to other facilities.

There are, of course, disadvantages:

  • The initial cost is expensive, sophisticated IT support is required, and robust back-up systems are required.

  • If not properly monitored, patient radiation can increase using digital receptors.

  • Communication between radiology and the ICU often decreases.

Hains et al. [18] reviewed the 1990s literature and argued that digital imaging was a qualified success. Over the last two decades, digital imaging plates and monitors have improved, and PACS functionality, reliability, and storage capacity have improved dramatically as prices drop [17,18].

Typical ICU imaging cases

A complete review of diagnostic imaging in the ICU is beyond the scope of this chapter. Eight imaging problems frequently seen in the ICU follow. Many believe ‘all portables look alike’. Although this is true to some extent, there are many clues to arriving at a more specific diagnosis or a limited list of possibilities. Perhaps the most helpful approach is by considering the time of onset of the imaging findings relative to the patient’s disease or status, and the speed of progression. For example, lung contusion is usually worse within the first day, but fades within a day or two. Aspiration of acid contents is usually visible within a few hours, gets worse for a day or two, and then subsides. Heart failure can appear rapidly and disappear just as rapidly with treatment, while ARDS usually appears 1–3 days after a major insult, worsens over a few days, then plateaus, often for prolonged periods of time. Atelectasis, which is the most frequent chest X-ray abnormality in the ICU, is almost universal after surgery, whereas similar appearing consolidation from hospital acquired pneumonia usually do not appear until many days after admission [3,19].

Case 1

A patient presents with shortness of breath and dry cough over a few hours. The focal density in the right costophrenic angle (Fig. 78.1a) is non-specific, but does suggest a pulmonary infarction (Hampton’s hump). CT verifies infarction due to acute pulmonary embolism (PE) in the anterior basal segment (Fig. 78.1b—arrow). Although a Hampton’s Hump or wedge-shaped pleural densities are highly suggestive of a PE, they are only seen in a minority of PE patients.

Fig. 78.1 Case 1. (a) X-ray; (b) CT scan.

Fig. 78.1 Case 1. (a) X-ray; (b) CT scan.

Case 2

A patient presents with dyspnoea after cervical spine fracture repair. A chest X-ray (Fig. 78.2a) shows patchy extensive bibasilar infiltrates, cardiomegaly, normal vessels and endotracheal tube within 2 cm of the carina. CT shows patchy infiltrates confined to the lower lobes, worse medially than laterally (Fig. 78.2b). This is the typical X-ray appearance of aspiration pneumonitis in a post-operative patient. The posterior basal segments of the lower lobe are most frequently involved. The only likely differential diagnosis would be extensive post-operative atelectasis.

Fig. 78.2 Case 2. (a) X-ray; (b) CT scan.

Fig. 78.2 Case 2. (a) X-ray; (b) CT scan.

Case 3

An elderly ICU patient with mild respiratory distress has a chest X-ray to rule out pulmonary embolism (Fig. 78.3a). The X-ray shows borderline enlarged heart, with large indistinct pulmonary arteries, mild peribronchial cuffing, and Kerley lines. CT shows Kerley lines laterally(arrows). The pulmonary artery is bigger than the adjacent bronchus and there is perivascular oedema (Fig. 78.3b circled). The radiographic findings of congestive heart failure (CHF) are well known. One would not order CT to diagnose CHF, but not infrequently, CT done for other reasons (e.g. PE or ARDS) reveals subtle CHF.

Fig. 78.3 Case 3. (a) X-ray; (b) CT scan.

Fig. 78.3 Case 3. (a) X-ray; (b) CT scan.

Case 4

A middle-aged male has a chest X-ray one day after cervical spine surgery for trauma (Fig. 78.4a). The trachea (ETT) and oesophagus (NG tube) are deviated to the left (arrow) of the spine rods. CT (Fig. 78.4b) shows seroma/haematoma deviating the trachea, the oesophagus, and the jugular vein to the left and the carotid artery to the right (surgically confirmed haematoma).

Fig. 78.4 Case 4. (a) X-ray; (b) CT scan.

Fig. 78.4 Case 4. (a) X-ray; (b) CT scan.

Case 5

A patient has respiratory distress one day after pancreatic surgery. A chest X-ray (Fig. 78.5a) shows total collapse of the right middle, right lower, and left lower lobes—the superior margin of the density on the right is the minor fissure. Both the right heart border and the right diaphragm are not visible (silhouette sign). Likewise, the left diaphragm is not visible. There are air bronchograms indicating patent airway (arrows). Absent air bronchograms indicate mucus plugging (Fig. 78.5b).

Fig. 78.5 Case 5. (a) X-ray; (b) X-ray (arrows show air bronchograms indicating patent airways).

Fig. 78.5 Case 5. (a) X-ray; (b) X-ray (arrows show air bronchograms indicating patent airways).

Case 6

A patient has fever of unknown origin several days after abdominal surgery. A chest X-ray shows vague focal densities throughout both lungs (Fig. 78.6a). CT Shows focal areas of consolidation varying from nodules to nodular cavities to focal parenchymal infiltrates (Fig. 78.6b). With septic emboli, the initial chest X-rays are often negative or vaguely positive. CT is virtually diagnostic early on with nodules, cavities and infiltrates in different stages of development.

Fig. 78.6 Case 6. (a) X-ray; (b) CT scan.

Fig. 78.6 Case 6. (a) X-ray; (b) CT scan.

Case 7

A patient has shortness of breath several days after caecal perforation. A chest X-ray shows vague interstitial thickening, right greater than left (Fig. 78.7a). The pulmonary vessels and heart are not enlarged. A CT, done for possible PE, shows bilateral asymmetrical ground glass opacification, not gravity dependent (Fig. 78.7b). The pulmonary vessels are not distended and there are no Kerley lines. In this case, the oedema became symmetrical within 24 hours. Otherwise, this is the typical X-ray and CT appearance of ARDS as opposed to CHF.

Fig. 78.7 Case 7. (a) X-ray; (b) CT scan.

Fig. 78.7 Case 7. (a) X-ray; (b) CT scan.

Case 8

An elderly man in the ICU for 2 weeks with respiratory distress after a perforated appendix has a chest X-ray (Fig. 78.8a). This is a case of established ARDS with diffuse symmetrical patchy infiltrates changing little from day-to-day. The heart and vessels are normal. There is a lucency in the right costophrenic angle (deep sulcus sign). This indicates air in the pleural space collecting anteriorly and laterally (arrow). The deep sulcus sign is not infallible and a confirmatory upright or decubitus film is required for confirmation. A small amount of air is also seen medially along the diaphragm (Fig. 78.8b circle).

Fig. 78.8 Case 8. (a) X-ray; (b) X-ray (circle shows small amount of air medially along the diaphragm).

Fig. 78.8 Case 8. (a) X-ray; (b) X-ray (circle shows small amount of air medially along the diaphragm).


1. Bekemeyer WB, Crapo RO, Calhoon S, Cannon CY, and Clayton PD. (1985). Efficacy of chest radiography in a respiratory intensive care unit. A prospective study. Chest, 88(5), 691–96.Find this resource:

2. Clec’h C, Simon P, Hamdi A, et al. (2008). Are daily routine chest radiographs useful in critically ill, mechanically ventilated patients? A randomized study. Intensive Care Medicine, 34(2), 264–70.Find this resource:

3. Goodman LG and Putman CE. (1992). Critical Care Imaging. Philadelphia, PA: W.B. Saunders Company.Find this resource:

4. Graat ME, Kröner A, Spronk PE, et al. (2007). Elimination of daily routine chest radiographs in a mixed medical-surgical intensive care unit. Intensive Care Medicine, 33(4), 639–44.Find this resource:

5. Graham RJ, Meziane MA, Rice TW, et al. (1998). Postoperative portable chest radiographs: Optimum use in thoracic surgery. Journal of Thoracic and Cardiovascular Surgery, 115(1), 45–52.Find this resource:

6. Hendrikse KA, Gratama JWC., ten Hove W, Rommes JH, Schultz MJ, and Spronk PE. (2007). Low value of routine chest radiographs in a mixed medical-surgical ICU. Chest, 132(3), 823–8.Find this resource:

7. Henschke G, Paternack GS, Schroeder S, Hart KK, and Herman PG. (1983). Bedside chest radiography: diagnostic efficacy. Radiology, 149, 23–6.Find this resource:

8. Janower M, Jennas-Nocera Z, and Mukai J. (1984). Utility and efficacy of portable chest radiographs. American Journal of Roentgenology, 142(2), 265–7.Find this resource:

9. Leong CS, Cascade PN, Kazerooni EA, Bolling SF, and Deeb GM. (2000). Bedside chest radiography as part of a postcardiac surgery critical care pathway: a means of decreasing utilization without adverse clinical impact. Critical Care Medicine, 28(2), 383–8.Find this resource:

10. Ganapathy A, Adhikari NK, Spiegelman J, and Scales DC. (2012). Routine chest X-rays in intensive care units: a systematic review and meta-analysis. Critical Care, 16(2), R68.Find this resource:

11. Oba Y and Zaza T. (2010). Abandoning daily routine chest radiography in the intensive care unit: meta-analysis1. Radiology, 255(2), 386–95.Find this resource:

12. American College of Radiology. (2011). The American College of Radiology Appropriateness Criteria for Routine Chest Radiographs in ICU patients. Available at:

13. Hejblum G, Chalumeau-Lemoine L, Ioos V, et al. (2009). Comparison of routine and on-demand prescription of chest radiographs in mechanically ventilated adults: a multicentre, cluster-randomised, two-period crossover study. Lancet, 374(9702), 1687–93.Find this resource:

14. Hejblum G, Ioos V, Vibert J-F, et al. (2008). A web-based Delphi study on the indications of chest radiographs for patients in ICUs. Chest, 133(5), 1107–12.Find this resource:

15. Lessnau K-D. (2008). From Delphi to knowledge and comfort. Chest, 133(5), 1060–2.Find this resource:

16. Kröner A, Binnekade JM, Graat ME, et al. (2008). On-demand rather than daily-routine chest radiography prescription may change neither the number nor the impact of chest computed tomography and ultrasound studies in a multidisciplinary intensive care unit. Anesthesiology, 108(1), 40–5.Find this resource:

17. Tan SL and Lewis RA. (2010). Picture archiving and communication systems: a multicentre survey of users experience and satisfaction. European Journal of Radiology, 75(3), 406–10.Find this resource:

18. Hains IM, Georgiou A, and Westbrook JI. (2012). The impact of PACS on clinician work practices in the intensive care unit: a systematic review of the literature. Journal of the American Medical Informatics Association, 19(4), 506–13.Find this resource:

19. Goodman LR and Putman C. (1992). Intensive Care Radiology: Imaging of the Critically Ill. Critical Care Imaging, 3rd edn. Philadelphia, PA: WB Saunders.Find this resource: