Diagnostic Value of Chest Ultrasound After Cardiac Surgery: A Comparison With Chest X-ray and Auscultation Antonella Vezzani, MD,* Tullio Manca, MD,* Claudia Brusasco, MD,† Gregorio Santori, MD, PhD,† Massimo Valentino, MD,‡ Francesco Nicolini, MD,* Alberto Molardi, MD,* Tiziano Gherli, MD,* and Francesco Corradi, MD, PhD*§ Objective: Chest auscultation and chest x-ray commonly are used to detect postoperative abnormalities and complications in patients admitted to intensive care after cardiac surgery. The aim of the study was to evaluate whether chest ultrasound represents an effective alternative to bedside chest x-ray to identify early postoperative abnormalities. Design: Diagnostic accuracy of chest auscultation and chest ultrasound were compared in identifying individual abnormalities detected by chest x-ray, considered the reference method. Setting: Cardiac surgery intensive care unit. Participants: One hundred fifty-one consecutive adult patients undergoing cardiac surgery. Interventions: All patients included were studied by chest auscultation, ultrasound, and x-ray upon admission to intensive care after cardiac surgery.
Measurements and Main Results: Six lung pathologic changes and endotracheal tube malposition were found. There was a highly significant correlation between abnormalities detected by chest ultrasound and x-ray (k ¼ 0.90), but a poor correlation between chest auscultation and x-ray abnormalities (k ¼ 0.15). Conclusions: Chest auscultation may help identify endotracheal tube misplacement and tension pneumothorax but it may miss most major abnormalities. Chest ultrasound represents a valid alternative to chest x-ray to detect most postoperative abnormalities and misplacements. & 2014 Elsevier Inc. All rights reserved.
P
and (6) pericardial effusion with or without cardiac tamponade. Correct positioning of the endotracheal tube also was checked. Anteroposterior bedside chest radiographs were obtained using portable x-ray equipment. The evaluation of the chest x-ray was performed by the radiologist on duty, unaware of chest ultrasound and auscultation findings. For data analysis, each hemithorax was divided into 6 regions.11 Pathologic entities were recognized according to the diagnostic criteria for bedside chest x-ray2 and using the terminology recommended by the Nomenclature Committee of the Fleischner Society.12 Consolidations were identified as an essentially homogenous opacity with loss of normal lucency. Alveolar-interstitial syndrome was identified as a pattern of opacity often symmetric and perihilar (Fig 1A). PLAPS was identified by the presence of an increased homogenous density in basal lung fields and/or pleural effusion producing obliteration of the costo-phrenic space (Fig 2A). Pleural effusion in the supine position was considered as an increased homogenous density superimposed over lung fields.13 Pneumothorax was identified by an increased radiolucency without lung markings in the costo-phrenic angles. Pericardial effusion was suspected when chest x-ray showed an enlarged cardiac silhouette with or without an epicardial fat-pad sign and with lungs typically clear.14 An endotracheal tube placed above the carina was considered as the correct position.15 Auscultation was performed by a physician immediately before chest ultrasound and bedside chest x-ray (CXR). Physical examination included auscultation of lungs and heart. Bronchial breath sounds were suggestive of alveolar consolidation. Fine crackles during the inspiratory phase, in anterior and lateral regions on either hemithorax, were suggestive of alveolar-interstitial syndrome. Reduction of breath sounds
OSTOPERATIVE CARDIAC SURGERY CARE in the intensive care unit (ICU) includes bedside chest x-ray and auscultation to detect postoperative abnormalities and complications. Despite the poor quality of images and radiation exposure, chest x-ray remains the daily reference for assessing lung status in critically ill patients. The American College of Radiology recommends a first chest x-ray to be obtained at ICU admission and subsequently only when required for specific clinical indications.1 Despite these guidelines, chest x-ray generally is used as a complement to physical examination even in the absence of specific indications because of a reported high prevalence of pathologic findings on bedside chest x-ray in critically ill patients.2 However, the usefulness of daily bedside chest x-ray has been questioned both in the general ICU3–5 and postoperative cardiac surgery;6,7 although, in a minority of patients, clinical evaluation alone may not be sufficient to detect early postoperative abnormalities and complications.8 Presently, bedside lung ultrasound commonly is applied to assess the respiratory condition of ventilated patients and may provide accurate information of diagnostic and therapeutic relevance.9,10 The aim of the present study was to evaluate whether chest ultrasound could be able to identify early abnormalities after cardiac surgery in comparison with chest auscultation and chest x-ray.
KEY WORDS: lung ultrasonography, postoperative complications, critical care, daily on-demand chest radiography
METHODS One hundred fifty-one consecutive adult patients undergoing cardiac surgery over a 4-month period at the Cardiac Surgery Intensive Care Unit of Azienda Ospedaliero-Universitaria of Parma were included (Table 1). The study was approved by the local ethics committee, and informed consent was obtained from each patient before surgery. All patients were admitted to the ICU directly after surgery and had chest auscultation, ultrasound, and x-ray upon arrival. Chest x-ray was considered as the reference method. Six pathologic entities were explored by each method: (1) consolidation, (2) alveolar-interstitial syndrome, (3) posterolateral alveolar and/ or pleural syndrome (PLAPS), (4) pleural effusion, (5) pneumothorax,
From the *Department of Surgery, Cardiac Surgery Intensive Care Unit, University Hospital of Parma, Parma, †University of Genova, Genova; and ‡S.O.C. Radiology Azienda Servizi Sanitari 3 - Friuli Venezia Giulia; and §Department of Critical Care, Intensive Care Unit, E.O. Ospedali Galliera, Genova, Italy. Address reprint requests to Antonella Vezzani, MD, Department of Surgery, Cardiac Surgery Intensive Care, University Hospital of Parma, Via Gramsci 14, 43100 Parma, Italy. E-mail: [email protected] © 2014 Elsevier Inc. All rights reserved. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2014.04.012
Journal of Cardiothoracic and Vascular Anesthesia, Vol 28, No 6 (December), 2014: pp 1527–1532
1527
1528
VEZZANI ET AL
Table 1. Baseline Characteristics of Patients Total number Age, mean ⫾ SD Gender, male, n (%) Type of surgery, n (% of total) Coronary artery bypass Valve surgery Combination of coronary artery bypass and valve surgery Other cardiac surgery Urgent surgery, n (%)
151 70 ⫾ 10 103 (68%) 73 (48%) 7 (5%) 64 (42%) 7 (5%) 23 (15%)
Abbreviations: n, number; SD, standard deviation.
was suggestive of pleural effusion. Reduction of vesicular breath sounds and/or bronchial breath sounds limited to lung bases was suggestive of PLAPS. Abolition of breath sounds on the anterior and lateral regions was suggestive of pneumothorax. Muffled heart sounds and tachycardia associated with hypotension, jugular venous distention, and pulsus paradoxus were taken as signs of cardiac tamponade. Endobronchial tube malposition was suspected when abolition or reduction of vesicular breath sounds involved the left hemithorax associated with other clinical signs, such as an oral endotracheal tube secured over the 21-cm mark in women and 23-cm mark in men.16 Chest ultrasound was performed immediately before chest radiography by a single intensivist (AV) skilled in echocardiographic examinations, using an HD 11 EX Philips ultrasound device with 3.5-Mh and 10-MHz transducers. The operator was unaware of clinical and chest x-ray findings. Lung ultrasonography examination was performed on the same regions evaluated by auscultation and chest x-ray. Normal lung pattern was identified by the presence of pleural sliding and A-lines.17 Alveolar consolidation was defined as a tissuelike pattern or coalescent B-lines in isolated region(s). Presence of a B-line in anterior and lateral regions was suggestive for alveolarinterstitial syndrome9,18 (Fig 1B).9,19 Lower lobe alveolar consolidation or minimal pleural effusion within the costo-phrenic angle was
Fig 1.
suggestive for PLAPS (Fig 2B).17 Pleural effusion was defined as an anechoic or hypoechoic pattern separating the virtual space between visceral and parietal pleura with changes during breathing.19,20 The absence of lung sliding associated with lung point sign was suggestive of pneumothorax.21 Cardiac tamponade was diagnosed on the basis of echocardiographic evidence.14 Endobronchial tube malposition was considered when lung sliding was absent on the left hemithorax with no paradoxical motion of the left diaphragm and visible lung pulse.22,23 Interobserver variability of chest ultrasonography was assessed in the first 40 consecutive patients enrolled in the present study using 2 chest ultrasound examinations obtained separately by 2 intensivists skilled in ultrasonography without knowledge of clinical and radiologic findings. The results are expressed as mean ⫾ standard deviation, counts, or percentages. Sample size was calculated prior to patient recruitment by taking into account an expected sensitivity of 0.95 and considering that the lower 95% confidence limit should not fall below 0.80 with 0.95 probability as previously described.24 A 35% expected prevalence of overall chest abnormalities on the basis of data from the literature was considered for sample size calculation.8 By using the approach previously reported for sample size estimation,24 at least 93 patients with chest abnormalities were required. The accuracy of each method was expressed as sensitivity, specificity, positive and negative predictive values, and diagnostic accuracy. To assess the agreements of chest ultrasound and chest auscultation with chest x-ray and interobservers,25 k statistics were used, with k values o0.40 indicating low agreement, 0.40 to 0.75 as low-to-good agreement, and 40.75 high agreement. Statistical analysis was performed using SPSS (IBM SPSS, Version 20.0. Armonk, NY: IBM Corp). RESULTS
Ninety-four of the 151 patients (62%) showed abnormalities on chest x-ray. Chest ultrasound correctly classified 144 patients and chest auscultation, 76. Chest ultrasound revealed abnormal
Chest x-ray (A) and ultrasound (B) image of alveolar-interstitial syndrome.
1529
CHEST ULTRASOUND V CHEST X-RAY AND AUSCULTATION
Fig 2.
Ultrasound (A) and chest x-ray (B) images of PLAPS on the left lung base.
findings not confirmed by chest x-ray in 3 patients, but failed to detect abnormalities revealed by chest x-ray in 4. Chest auscultation revealed abnormal findings not detected by chest x-ray in 2 patients, but failed to discover abnormalities shown by chest x-ray in 73. Abnormalities detected by chest x-ray were correlated significantly with those detected by chest ultrasound (k ¼ 0.90) but not chest auscultation (k ¼ 0.15). Sensitivity, specificity, PPV, NPV, and diagnostic accuracy of chest ultrasound and chest auscultation for each abnormalities are shown in Table 2. Consolidations were found in 14 patients (9%); 12 were minor linear atelectasis, 1 was a segmental opacity, and 1 a parahilar consolidation. Ultrasound failed to detect the parahilar consolidation and linear atelectasis in 1 case but showed 2 falsely positive minor consolidations. Auscultation gave 3 false positive and 11 false negative results for consolidation. Compared with chest x-ray in identifying consolidations, chest ultrasound had a sensitivity of 86% and a specificity of 99% while chest auscultation had a sensitivity of 21% and a specificity of 98%. Alveolar-interstitial syndrome was identified in 38 patients (25%). Chest ultrasound failed to detect only 2 alveolar-interstitial syndromes. Chest auscultation detected 6 alveolar-interstitial syndromes and failed to detect 32 alveolar-interstitial syndromes. When compared with chest x-ray in identifying alveolar-interstitial syndrome, chest ultrasound had a sensitivity of 95% and a specificity of 100% while chest auscultation had a sensitivity of 16% and a specificity of 100%. PLAPS was identified in 64 patients (42%). Chest ultrasound had 2 false positive and 2 false negative results. Chest auscultation had 60 false negative and 3 false positive results. When compared with chest x-ray in identifying PLAPS, chest ultrasound had a sensitivity of 97% and a specificity of 98% while chest auscultation had a sensitivity of 6% and a specificity of 97%. Pleural effusion was identified in 5 patients (3,3%) at chest
x-ray. Chest ultrasound identified all cases of pleural effusion while chest auscultation only 1; chest ultrasound had 1 false positive and chest auscultation 2 false positive results. Pneumothorax was identified in 3 patients; 1 was occult and 2 were tension pneumothoraces. Chest ultrasound detected all of them while chest auscultation failed to identify the occult one. Pericardial effusion was identified in 3 cases, 2 of them needing emergency drainage. Chest ultrasound detected all pericardial effusions while neither chest x-ray nor chest auscultation was able to identify them. Endotracheal tube malposition occurred in 2 patients identified by both chest ultrasound and chest auscultation. A 93% interobserver agreement (k ¼ 0.85; SE: 0.083; 95% CI: 0.69-1.01) on chest ultrasound findings between the 2 observers was found. DISCUSSION
The main findings of this study were that (1) chest ultrasound identified most of the pathologic abnormalities occurring early after cardiac surgery with a diagnostic accuracy that compared to chest x-ray, and (2) chest auscultation failed to identify major postoperative complications and showed a much poorer diagnostic accuracy than chest x-ray. Many studies have investigated the usefulness of chest ultrasound in emergency and critical care settings10,26 but not in a cardiothoracic surgery setting. To date, abandoning daily routine CXRs in cardiothoracic patients still is considered hazardous27 due to the low sensitivity of clinical assessment in predicting clinically important abnormalities detectable by CXRs.8 To the best of the authors’ knowledge this was the first study validating the usefulness of chest ultrasonography in cardiothoracic surgery patients in comparisons with CXR and clinical assessment.
1530
VEZZANI ET AL
Table 2. Chest Ultrasound and Chest Auscultation Compared to Chest X-ray for Each Abnormality Sp Findings
CU/CA
CXR-
CXRþ
Consolidation
CUCUþ CACAþ CUCUþ CACAþ CUCUþ CACAþ CU CU þ CA CA þ CU CU þ CA CA þ CU CU þ CA CA þ CU CU þ
135 2 134 3 113 0 113 0 85 2 84 3 145 1 144 2 148 0 148 0 149 0 149 0 80 0
2 12 11 3 2 36 32 6 2 62 60 4 0 5 4 1 0 3 1 2 0 2 0 2 0 2
Alveolarinterstitial syndrome PLAPS
Pleural effusion
Pneumothorax
Endotracheal tube misplacement Central venous catheter misplacement
S%
PPV NPV
%
%
%
86
99
86
99
97 0.84
21
98
50
92
91 0.26
95 100 100
98
99 0.96
16 100 100
78
79 0.58
97
98
97
98
97 0.95
6
97
57
58
58 0.03
100
99
83 100
99 0.91
20
99
33
96 0.23
97
DA
k
100 100 100 100 100 1 66 100 100
99
99 0.8
100 100 100 100 100 1 100 100 100 100 100 1 100 100 100 100 100 1
NOTE. Each patient was characterized as positive (þ) or negative (-) for a specific abnormality found in a single region. Abbreviations: CA, chest auscultation; CU, chest ultrasound; CXR, chest x-ray; DA, diagnostic accuracy; NPV, negative predictive value; PLAPS, posterolateral alveolar and/or pleural syndrome; PPV, positive predictive value; S, sensitivity; Sp, specificity.
The authors have investigated 4 common abnormalities, ie, consolidations, alveolar-interstitial syndrome, pleural effusion and pneumothorax, which may have important implications in decision-making. Furthermore, they have assessed the presence of PLAPS, which is very frequent in the cardiac postoperative period and may represent an early sign of evolving disease, such as consolidation or pleural effusion, even if it does not have a clear pathologic significance. In the authors’ study, the number of consolidations was 14 out of 151. Only a segmental opacity needed a bronchoscopy, whereas the other consolidations disappeared after 24 hours without treatment. Although the largest number of consolidations was detected by ultrasound, 1 parahilar consolidation was not identified. Chest ultrasound showed a sensitivity of 86% and a diagnostic accuracy of 97% in identifying consolidations, which were lower than previously reported.11,17 One reason for this could have been that chest ultrasound may not identify even relevant consolidations if they do not reach the pleura,9 whereas it can detect even very small consolidations near the pleura, which are not detected easily by chest x-ray. Chest auscultation had a much lower sensitivity (21%) and diagnostic accuracy (91%) for consolidation, which was in accordance with results from other studies of ventilated patients.11
The number of alveolar-interstitial syndromes was 38 out of 151, and this complication was the one frequently needing therapeutic interventions (7 out of 38). Chest ultrasound was much more sensitive and specific than chest auscultation in identifying alveolar-interstitial syndrome. Chest ultrasound showed a sensitivity of 95% and a diagnostic accuracy of 99%, which were comparable to those reported in other studies.28 However, alveolar-interstitial syndrome was not identified in 2 patients. This may have been due to high levels of PEEP employed in these patients required to improve oxygenation. PLAPS was identified in 64 patients. Although the recognition of this pathologic condition did not result in an immediate therapeutic intervention, it allowed early identification of those that evolved into a significant pleural effusion or consolidation. It has been reported that ultrasound has an elevated sensitivity and specificity for pleural effusion and, when compared with thoracic computed tomography, performs better than chest x-ray for this abnormality.11,18 In the authors’ study, chest ultrasound showed 100% sensitivity and 99% diagnostic accuracy. The false positive result was probably due to the presence of a small pleural effusion not identified by chest x-ray. Chest auscultation showed a poor diagnostic accuracy for this abnormality. Only 3 cases of pneumothorax were present in the authors’ study. All pneumothoraces were detected by chest ultrasound, while chest auscultation detected only 2 cases. However, the pneumothorax missed by chest auscultation was small and did not require drainage. Finally, misplacement of an endotracheal tube occurred in only 2 cases and was detected accurately by both chest auscultation and ultrasound. Several studies defined lung ultrasound as a quick learning technique by any critical care physician.9,29 Furthermore, in cardiac surgery ICUs, many physicians already are able to perform echocardiography at least at base level, and, therefore, lung ultrasonography learning time could be even lower. In the authors’ group, 8-hour lecturing and 50 mentored examinations were considered sufficient for the acquisition of sufficient skill in lung ultrasonography. In addition, lung ultrasonography requires no more than 5 minutes.30 Finally, combining lung and heart ultrasonography allows critical care physicians to obtain a comprehensive assessment of the patient integrating ultrasound information with clinical data, which could not be done by an ultrasound technician unaware of the patient’s clinical condition. In fact, an integrated ultrasound examination of lung and heart allows a global assessment of congestive heart failure and extravascular lung water, a differential diagnosis of hemodynamic instability, and a target during resuscitation.31 Moreover, CXR is no longer considered necessary even for checking the correct positioning of central venous catheters after the introduction of a new ultrasound method able to directly visualize catheter tip positioning in the superior vena cava-atrial junction.32 In the same study, cost analysis of CXR and ultrasonography also were compared, resulting in slightly lower costs for ultrasonography, but with a considerable advantage in terms of reduction of ionizing radiation exposure for both patients and staff.32
1531
CHEST ULTRASOUND V CHEST X-RAY AND AUSCULTATION
All ultrasound examinations were performed by the same physician to avoid interobserver variability and focus only on comparison among the 3 different diagnostic techniques. Nevertheless, interobserver reproducibility between 2 different operators was tested on a subset of the study population to assess clinical applicability of chest ultrasound, showing a high interobserver concordance (93%) consistent with previous findings.33 This study had limitations. First, bedside chest x-ray was used as the reference method for comparison with chest auscultation and ultrasound despite poor-quality images with low sensitivity. Thoracic CT would be the real gold standard for lung imaging, but it cannot be performed as a routine postoperative investigation. Nevertheless, chest x-ray routinely is performed in postoperative cardiac surgery patients and, according to the American College of Radiology’s Appropriateness Criteria,1 is considered the reference method, leaving thoracic CT for special difficult cases.
Second, the number of pneumothoraces and endotracheal tube malpositions was too small to draw any conclusion about diagnostic accuracy of the 3 techniques. For these abnormalities, studies with a much larger number of patients would be necessary because of their very low prevalence. CONCLUSIONS
Due to the high number of abnormalities found in early postoperative cardiac surgery patients and the low accuracy showed by chest auscultation, routine imaging study is mandatory. In this scenario, chest ultrasound seems a valid alternative to chest x-ray to detect most postoperative abnormalities and misplacements. Furthermore, chest ultrasound avoids patients’ and staff’s radiation exposure, does not need additional resources outside the ICU, and easily can be repeated when the patient’s clinical condition changes.
REFERENCES 1. Radiology ACo: ACR Appropriateness Criteria. Routine chest radiographs in ICU. Available at: http://www.acr.org/Quality-Safety/ Appropriateness-Criteria/Diagnostic/Thoracic-Imaging. Accessed in November 2013. 2. Rubinowitz AN, Siegel MD, Tocino I: Thoracic imaging in the ICU. Critical Care Clinics 23:539-573, 2007 3. Krivopal M, Shlobin OA, Schwartzstein RM: Utility of daily routine portable chest radiographs in mechanically ventilated patients in the medical ICU. Chest 123:1607-1614, 2003 4. Clec'h C, Simon P, Hamdi A, et al: Are daily routine chest radiographs useful in critically ill, mechanically ventilated patients? A randomized study. Intensive Care Med 34:264-270, 2008 5. Oba Y, Zaza T: Abandoning daily routine chest radiography in the intensive care unit: Meta-analysis. Radiology 255:386-395, 2010 6. Mets O, Spronk PE, Binnekade J, et al: Elimination of daily routine chest radiographs does not change on-demand radiography practice in post-cardiothoracic surgery patients. J Thorac Cardiovasc Surg 134:139-144, 2007 7. Leong CS, Cascade PN, Kazerooni EA, et al: Bedside chest radiography as part of a postcardiac surgery critical care pathway: A means of decreasing utilization without adverse clinical impact. Crit Care Med 28:383-388, 2000 8. Tolsma M, Kroner A, van den Hombergh CL, et al: The clinical value of routine chest radiographs in the first 24 hours after cardiac surgery. Anesth Analg 112:139-142, 2011 9. Volpicelli G, Elbarbary M, Blaivas M, et al: International evidence-based recommendations for point-of-care lung ultrasound. Intensive Care Med 38:577-591, 2012 10. Ashton-Cleary DT: Is thoracic ultrasound a viable alternative to conventional imaging in the critical care setting? Brit J Anaesth 111: 152-160, 2013 11. Lichtenstein D, Goldstein I, Mourgeon E, et al: Comparative diagnostic performances of auscultation, chest radiography, and lung ultrasonography in acute respiratory distress syndrome. Anesthesiology 100:9-15, 2004 12. Tuddenham WJ: Glossary of terms for thoracic radiology: Recommendations of the Nomenclature Committee of the Fleischner Society. AJR. Am J Roentgenol 143:509-517, 1984 13. Woodring JH: Recognition of pleural effusion on supine radiographs: How much fluid is required? Am J Roentgenol 142:59-64, 1984 14. Restrepo CS, Lemos DF, Lemos JA, et al: Imaging findings in cardiac tamponade with emphasis on CT. Radiographics 27: 1595-1610, 2007
15. Lotano R, Gerber D, Aseron C, et al: Utility of postintubation chest radiographs in the intensive care unit. Crit Care 4:50-53, 2000 16. Owen RL, Cheney FW: Endobronchial intubation: A preventable complication. Anesthesiol 67:255-257, 1987 17. Lichtenstein DA, Meziere GA: Relevance of lung ultrasound in the diagnosis of acute respiratory failure: the BLUE protocol. Chest 134:117-125, 2008 18. Gargani L, Frassi F, Soldati G, et al: Ultrasound lung comets for the differential diagnosis of acute cardiogenic dyspnoea: A comparison with natriuretic peptides. European journal of heart failure 10: 70-77, 2008 19. Remerand F, Dellamonica J, Mao Z, et al: Multiplane ultrasound approach to quantify pleural effusion at the bedside. Intensive Care Med 36:656-664, 2010 20. Lichtenstein D, Hulot JS, Rabiller A, et al: Feasibility and safety of ultrasound-aided thoracentesis in mechanically ventilated patients. Intensive Care Med 25:955-958, 1999 21. Soldati G, Testa A, Sher S, et al: Occult traumatic pneumothorax: Diagnostic accuracy of lung ultrasonography in the emergency department. Chest 133:204-211, 2008 22. Weaver B, Lyon M, Blaivas M: Confirmation of endotracheal tube placement after intubation using the ultrasound sliding lung sign. Acad Emerg Med 13:239-244, 2006 23. Kerrey BT, Geis GL, Quinn AM, et al: A prospective comparison of diaphragmatic ultrasound and chest radiography to determine endotracheal tube position in a pediatric emergency department. Pediatrics 123:e1039-e1044, 2009 24. Flahault A, Cadilhac M, Thomas G: Sample size calculation should be performed for design accuracy in diagnostic test studies. J Clin Epidemiol 58:859-862, 2005 25. Landis JR, Koch GG: The measurement of observer agreement for categorical data. Biometrics 33:159-174, 1977 26. Silva S, Biendel C, Ruiz J, et al: Usefulness of cardiothoracic chest ultrasound in the management of acute respiratory failure in critical care practice. Chest 144:859-865, 2013 27. Lakhal K, Serveaux-Delous M, Lefrant JY, et al: Chest radiographs in 104 French ICUs: Current prescription strategies and clinical value (the RadioDay study). Intensive Care Med 38:1787-1799, 2012 28. Agricola E, Bove T, Oppizzi M, et al: “Ultrasound comet-tail images”: A marker of pulmonary edema: A comparative study with wedge pressure and extravascular lung water. Chest 127:1690-1695, 2005 29. Lichtenstein DA: Lung ultrasound in the critically ill. Ann Intensive Care 4:1, 2014
1532
30. Parlamento S, Copetti R, Di Bartolomeo S: Evaluation of lung ultrasound for the diagnosis of pneumonia in the ED. Am J Emerg Med 27:379-384, 2009 31. Lichtenstein D: FALLS-protocol: Lung ultrasound in hemodynamic assessment of shock. Heart, lung and vessels 5:142-147, 2013
VEZZANI ET AL
32. Vezzani A, Brusasco C, Palermo S, et al: Ultrasound localization of central vein catheter and detection of postprocedural pneumothorax: An alternative to chest radiography. Crit Care Med 38:533-538, 2010 33. Lichtenstein DA, Lascols N, Meziere G, et al: Ultrasound diagnosis of alveolar consolidation in the critically ill. Intensive Care Med 30:276-281, 2004
Measurements and Main Results: Six lung pathologic changes and endotracheal tube malposition were found. There was a highly significant correlation between abnormalities detected by chest ultrasound and x-ray (k ¼ 0.90), but a poor correlation between chest auscultation and x-ray abnormalities (k ¼ 0.15). Conclusions: Chest auscultation may help identify endotracheal tube misplacement and tension pneumothorax but it may miss most major abnormalities. Chest ultrasound represents a valid alternative to chest x-ray to detect most postoperative abnormalities and misplacements. & 2014 Elsevier Inc. All rights reserved.
P
and (6) pericardial effusion with or without cardiac tamponade. Correct positioning of the endotracheal tube also was checked. Anteroposterior bedside chest radiographs were obtained using portable x-ray equipment. The evaluation of the chest x-ray was performed by the radiologist on duty, unaware of chest ultrasound and auscultation findings. For data analysis, each hemithorax was divided into 6 regions.11 Pathologic entities were recognized according to the diagnostic criteria for bedside chest x-ray2 and using the terminology recommended by the Nomenclature Committee of the Fleischner Society.12 Consolidations were identified as an essentially homogenous opacity with loss of normal lucency. Alveolar-interstitial syndrome was identified as a pattern of opacity often symmetric and perihilar (Fig 1A). PLAPS was identified by the presence of an increased homogenous density in basal lung fields and/or pleural effusion producing obliteration of the costo-phrenic space (Fig 2A). Pleural effusion in the supine position was considered as an increased homogenous density superimposed over lung fields.13 Pneumothorax was identified by an increased radiolucency without lung markings in the costo-phrenic angles. Pericardial effusion was suspected when chest x-ray showed an enlarged cardiac silhouette with or without an epicardial fat-pad sign and with lungs typically clear.14 An endotracheal tube placed above the carina was considered as the correct position.15 Auscultation was performed by a physician immediately before chest ultrasound and bedside chest x-ray (CXR). Physical examination included auscultation of lungs and heart. Bronchial breath sounds were suggestive of alveolar consolidation. Fine crackles during the inspiratory phase, in anterior and lateral regions on either hemithorax, were suggestive of alveolar-interstitial syndrome. Reduction of breath sounds
OSTOPERATIVE CARDIAC SURGERY CARE in the intensive care unit (ICU) includes bedside chest x-ray and auscultation to detect postoperative abnormalities and complications. Despite the poor quality of images and radiation exposure, chest x-ray remains the daily reference for assessing lung status in critically ill patients. The American College of Radiology recommends a first chest x-ray to be obtained at ICU admission and subsequently only when required for specific clinical indications.1 Despite these guidelines, chest x-ray generally is used as a complement to physical examination even in the absence of specific indications because of a reported high prevalence of pathologic findings on bedside chest x-ray in critically ill patients.2 However, the usefulness of daily bedside chest x-ray has been questioned both in the general ICU3–5 and postoperative cardiac surgery;6,7 although, in a minority of patients, clinical evaluation alone may not be sufficient to detect early postoperative abnormalities and complications.8 Presently, bedside lung ultrasound commonly is applied to assess the respiratory condition of ventilated patients and may provide accurate information of diagnostic and therapeutic relevance.9,10 The aim of the present study was to evaluate whether chest ultrasound could be able to identify early abnormalities after cardiac surgery in comparison with chest auscultation and chest x-ray.
KEY WORDS: lung ultrasonography, postoperative complications, critical care, daily on-demand chest radiography
METHODS One hundred fifty-one consecutive adult patients undergoing cardiac surgery over a 4-month period at the Cardiac Surgery Intensive Care Unit of Azienda Ospedaliero-Universitaria of Parma were included (Table 1). The study was approved by the local ethics committee, and informed consent was obtained from each patient before surgery. All patients were admitted to the ICU directly after surgery and had chest auscultation, ultrasound, and x-ray upon arrival. Chest x-ray was considered as the reference method. Six pathologic entities were explored by each method: (1) consolidation, (2) alveolar-interstitial syndrome, (3) posterolateral alveolar and/ or pleural syndrome (PLAPS), (4) pleural effusion, (5) pneumothorax,
From the *Department of Surgery, Cardiac Surgery Intensive Care Unit, University Hospital of Parma, Parma, †University of Genova, Genova; and ‡S.O.C. Radiology Azienda Servizi Sanitari 3 - Friuli Venezia Giulia; and §Department of Critical Care, Intensive Care Unit, E.O. Ospedali Galliera, Genova, Italy. Address reprint requests to Antonella Vezzani, MD, Department of Surgery, Cardiac Surgery Intensive Care, University Hospital of Parma, Via Gramsci 14, 43100 Parma, Italy. E-mail: [email protected] © 2014 Elsevier Inc. All rights reserved. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2014.04.012
Journal of Cardiothoracic and Vascular Anesthesia, Vol 28, No 6 (December), 2014: pp 1527–1532
1527
1528
VEZZANI ET AL
Table 1. Baseline Characteristics of Patients Total number Age, mean ⫾ SD Gender, male, n (%) Type of surgery, n (% of total) Coronary artery bypass Valve surgery Combination of coronary artery bypass and valve surgery Other cardiac surgery Urgent surgery, n (%)
151 70 ⫾ 10 103 (68%) 73 (48%) 7 (5%) 64 (42%) 7 (5%) 23 (15%)
Abbreviations: n, number; SD, standard deviation.
was suggestive of pleural effusion. Reduction of vesicular breath sounds and/or bronchial breath sounds limited to lung bases was suggestive of PLAPS. Abolition of breath sounds on the anterior and lateral regions was suggestive of pneumothorax. Muffled heart sounds and tachycardia associated with hypotension, jugular venous distention, and pulsus paradoxus were taken as signs of cardiac tamponade. Endobronchial tube malposition was suspected when abolition or reduction of vesicular breath sounds involved the left hemithorax associated with other clinical signs, such as an oral endotracheal tube secured over the 21-cm mark in women and 23-cm mark in men.16 Chest ultrasound was performed immediately before chest radiography by a single intensivist (AV) skilled in echocardiographic examinations, using an HD 11 EX Philips ultrasound device with 3.5-Mh and 10-MHz transducers. The operator was unaware of clinical and chest x-ray findings. Lung ultrasonography examination was performed on the same regions evaluated by auscultation and chest x-ray. Normal lung pattern was identified by the presence of pleural sliding and A-lines.17 Alveolar consolidation was defined as a tissuelike pattern or coalescent B-lines in isolated region(s). Presence of a B-line in anterior and lateral regions was suggestive for alveolarinterstitial syndrome9,18 (Fig 1B).9,19 Lower lobe alveolar consolidation or minimal pleural effusion within the costo-phrenic angle was
Fig 1.
suggestive for PLAPS (Fig 2B).17 Pleural effusion was defined as an anechoic or hypoechoic pattern separating the virtual space between visceral and parietal pleura with changes during breathing.19,20 The absence of lung sliding associated with lung point sign was suggestive of pneumothorax.21 Cardiac tamponade was diagnosed on the basis of echocardiographic evidence.14 Endobronchial tube malposition was considered when lung sliding was absent on the left hemithorax with no paradoxical motion of the left diaphragm and visible lung pulse.22,23 Interobserver variability of chest ultrasonography was assessed in the first 40 consecutive patients enrolled in the present study using 2 chest ultrasound examinations obtained separately by 2 intensivists skilled in ultrasonography without knowledge of clinical and radiologic findings. The results are expressed as mean ⫾ standard deviation, counts, or percentages. Sample size was calculated prior to patient recruitment by taking into account an expected sensitivity of 0.95 and considering that the lower 95% confidence limit should not fall below 0.80 with 0.95 probability as previously described.24 A 35% expected prevalence of overall chest abnormalities on the basis of data from the literature was considered for sample size calculation.8 By using the approach previously reported for sample size estimation,24 at least 93 patients with chest abnormalities were required. The accuracy of each method was expressed as sensitivity, specificity, positive and negative predictive values, and diagnostic accuracy. To assess the agreements of chest ultrasound and chest auscultation with chest x-ray and interobservers,25 k statistics were used, with k values o0.40 indicating low agreement, 0.40 to 0.75 as low-to-good agreement, and 40.75 high agreement. Statistical analysis was performed using SPSS (IBM SPSS, Version 20.0. Armonk, NY: IBM Corp). RESULTS
Ninety-four of the 151 patients (62%) showed abnormalities on chest x-ray. Chest ultrasound correctly classified 144 patients and chest auscultation, 76. Chest ultrasound revealed abnormal
Chest x-ray (A) and ultrasound (B) image of alveolar-interstitial syndrome.
1529
CHEST ULTRASOUND V CHEST X-RAY AND AUSCULTATION
Fig 2.
Ultrasound (A) and chest x-ray (B) images of PLAPS on the left lung base.
findings not confirmed by chest x-ray in 3 patients, but failed to detect abnormalities revealed by chest x-ray in 4. Chest auscultation revealed abnormal findings not detected by chest x-ray in 2 patients, but failed to discover abnormalities shown by chest x-ray in 73. Abnormalities detected by chest x-ray were correlated significantly with those detected by chest ultrasound (k ¼ 0.90) but not chest auscultation (k ¼ 0.15). Sensitivity, specificity, PPV, NPV, and diagnostic accuracy of chest ultrasound and chest auscultation for each abnormalities are shown in Table 2. Consolidations were found in 14 patients (9%); 12 were minor linear atelectasis, 1 was a segmental opacity, and 1 a parahilar consolidation. Ultrasound failed to detect the parahilar consolidation and linear atelectasis in 1 case but showed 2 falsely positive minor consolidations. Auscultation gave 3 false positive and 11 false negative results for consolidation. Compared with chest x-ray in identifying consolidations, chest ultrasound had a sensitivity of 86% and a specificity of 99% while chest auscultation had a sensitivity of 21% and a specificity of 98%. Alveolar-interstitial syndrome was identified in 38 patients (25%). Chest ultrasound failed to detect only 2 alveolar-interstitial syndromes. Chest auscultation detected 6 alveolar-interstitial syndromes and failed to detect 32 alveolar-interstitial syndromes. When compared with chest x-ray in identifying alveolar-interstitial syndrome, chest ultrasound had a sensitivity of 95% and a specificity of 100% while chest auscultation had a sensitivity of 16% and a specificity of 100%. PLAPS was identified in 64 patients (42%). Chest ultrasound had 2 false positive and 2 false negative results. Chest auscultation had 60 false negative and 3 false positive results. When compared with chest x-ray in identifying PLAPS, chest ultrasound had a sensitivity of 97% and a specificity of 98% while chest auscultation had a sensitivity of 6% and a specificity of 97%. Pleural effusion was identified in 5 patients (3,3%) at chest
x-ray. Chest ultrasound identified all cases of pleural effusion while chest auscultation only 1; chest ultrasound had 1 false positive and chest auscultation 2 false positive results. Pneumothorax was identified in 3 patients; 1 was occult and 2 were tension pneumothoraces. Chest ultrasound detected all of them while chest auscultation failed to identify the occult one. Pericardial effusion was identified in 3 cases, 2 of them needing emergency drainage. Chest ultrasound detected all pericardial effusions while neither chest x-ray nor chest auscultation was able to identify them. Endotracheal tube malposition occurred in 2 patients identified by both chest ultrasound and chest auscultation. A 93% interobserver agreement (k ¼ 0.85; SE: 0.083; 95% CI: 0.69-1.01) on chest ultrasound findings between the 2 observers was found. DISCUSSION
The main findings of this study were that (1) chest ultrasound identified most of the pathologic abnormalities occurring early after cardiac surgery with a diagnostic accuracy that compared to chest x-ray, and (2) chest auscultation failed to identify major postoperative complications and showed a much poorer diagnostic accuracy than chest x-ray. Many studies have investigated the usefulness of chest ultrasound in emergency and critical care settings10,26 but not in a cardiothoracic surgery setting. To date, abandoning daily routine CXRs in cardiothoracic patients still is considered hazardous27 due to the low sensitivity of clinical assessment in predicting clinically important abnormalities detectable by CXRs.8 To the best of the authors’ knowledge this was the first study validating the usefulness of chest ultrasonography in cardiothoracic surgery patients in comparisons with CXR and clinical assessment.
1530
VEZZANI ET AL
Table 2. Chest Ultrasound and Chest Auscultation Compared to Chest X-ray for Each Abnormality Sp Findings
CU/CA
CXR-
CXRþ
Consolidation
CUCUþ CACAþ CUCUþ CACAþ CUCUþ CACAþ CU CU þ CA CA þ CU CU þ CA CA þ CU CU þ CA CA þ CU CU þ
135 2 134 3 113 0 113 0 85 2 84 3 145 1 144 2 148 0 148 0 149 0 149 0 80 0
2 12 11 3 2 36 32 6 2 62 60 4 0 5 4 1 0 3 1 2 0 2 0 2 0 2
Alveolarinterstitial syndrome PLAPS
Pleural effusion
Pneumothorax
Endotracheal tube misplacement Central venous catheter misplacement
S%
PPV NPV
%
%
%
86
99
86
99
97 0.84
21
98
50
92
91 0.26
95 100 100
98
99 0.96
16 100 100
78
79 0.58
97
98
97
98
97 0.95
6
97
57
58
58 0.03
100
99
83 100
99 0.91
20
99
33
96 0.23
97
DA
k
100 100 100 100 100 1 66 100 100
99
99 0.8
100 100 100 100 100 1 100 100 100 100 100 1 100 100 100 100 100 1
NOTE. Each patient was characterized as positive (þ) or negative (-) for a specific abnormality found in a single region. Abbreviations: CA, chest auscultation; CU, chest ultrasound; CXR, chest x-ray; DA, diagnostic accuracy; NPV, negative predictive value; PLAPS, posterolateral alveolar and/or pleural syndrome; PPV, positive predictive value; S, sensitivity; Sp, specificity.
The authors have investigated 4 common abnormalities, ie, consolidations, alveolar-interstitial syndrome, pleural effusion and pneumothorax, which may have important implications in decision-making. Furthermore, they have assessed the presence of PLAPS, which is very frequent in the cardiac postoperative period and may represent an early sign of evolving disease, such as consolidation or pleural effusion, even if it does not have a clear pathologic significance. In the authors’ study, the number of consolidations was 14 out of 151. Only a segmental opacity needed a bronchoscopy, whereas the other consolidations disappeared after 24 hours without treatment. Although the largest number of consolidations was detected by ultrasound, 1 parahilar consolidation was not identified. Chest ultrasound showed a sensitivity of 86% and a diagnostic accuracy of 97% in identifying consolidations, which were lower than previously reported.11,17 One reason for this could have been that chest ultrasound may not identify even relevant consolidations if they do not reach the pleura,9 whereas it can detect even very small consolidations near the pleura, which are not detected easily by chest x-ray. Chest auscultation had a much lower sensitivity (21%) and diagnostic accuracy (91%) for consolidation, which was in accordance with results from other studies of ventilated patients.11
The number of alveolar-interstitial syndromes was 38 out of 151, and this complication was the one frequently needing therapeutic interventions (7 out of 38). Chest ultrasound was much more sensitive and specific than chest auscultation in identifying alveolar-interstitial syndrome. Chest ultrasound showed a sensitivity of 95% and a diagnostic accuracy of 99%, which were comparable to those reported in other studies.28 However, alveolar-interstitial syndrome was not identified in 2 patients. This may have been due to high levels of PEEP employed in these patients required to improve oxygenation. PLAPS was identified in 64 patients. Although the recognition of this pathologic condition did not result in an immediate therapeutic intervention, it allowed early identification of those that evolved into a significant pleural effusion or consolidation. It has been reported that ultrasound has an elevated sensitivity and specificity for pleural effusion and, when compared with thoracic computed tomography, performs better than chest x-ray for this abnormality.11,18 In the authors’ study, chest ultrasound showed 100% sensitivity and 99% diagnostic accuracy. The false positive result was probably due to the presence of a small pleural effusion not identified by chest x-ray. Chest auscultation showed a poor diagnostic accuracy for this abnormality. Only 3 cases of pneumothorax were present in the authors’ study. All pneumothoraces were detected by chest ultrasound, while chest auscultation detected only 2 cases. However, the pneumothorax missed by chest auscultation was small and did not require drainage. Finally, misplacement of an endotracheal tube occurred in only 2 cases and was detected accurately by both chest auscultation and ultrasound. Several studies defined lung ultrasound as a quick learning technique by any critical care physician.9,29 Furthermore, in cardiac surgery ICUs, many physicians already are able to perform echocardiography at least at base level, and, therefore, lung ultrasonography learning time could be even lower. In the authors’ group, 8-hour lecturing and 50 mentored examinations were considered sufficient for the acquisition of sufficient skill in lung ultrasonography. In addition, lung ultrasonography requires no more than 5 minutes.30 Finally, combining lung and heart ultrasonography allows critical care physicians to obtain a comprehensive assessment of the patient integrating ultrasound information with clinical data, which could not be done by an ultrasound technician unaware of the patient’s clinical condition. In fact, an integrated ultrasound examination of lung and heart allows a global assessment of congestive heart failure and extravascular lung water, a differential diagnosis of hemodynamic instability, and a target during resuscitation.31 Moreover, CXR is no longer considered necessary even for checking the correct positioning of central venous catheters after the introduction of a new ultrasound method able to directly visualize catheter tip positioning in the superior vena cava-atrial junction.32 In the same study, cost analysis of CXR and ultrasonography also were compared, resulting in slightly lower costs for ultrasonography, but with a considerable advantage in terms of reduction of ionizing radiation exposure for both patients and staff.32
1531
CHEST ULTRASOUND V CHEST X-RAY AND AUSCULTATION
All ultrasound examinations were performed by the same physician to avoid interobserver variability and focus only on comparison among the 3 different diagnostic techniques. Nevertheless, interobserver reproducibility between 2 different operators was tested on a subset of the study population to assess clinical applicability of chest ultrasound, showing a high interobserver concordance (93%) consistent with previous findings.33 This study had limitations. First, bedside chest x-ray was used as the reference method for comparison with chest auscultation and ultrasound despite poor-quality images with low sensitivity. Thoracic CT would be the real gold standard for lung imaging, but it cannot be performed as a routine postoperative investigation. Nevertheless, chest x-ray routinely is performed in postoperative cardiac surgery patients and, according to the American College of Radiology’s Appropriateness Criteria,1 is considered the reference method, leaving thoracic CT for special difficult cases.
Second, the number of pneumothoraces and endotracheal tube malpositions was too small to draw any conclusion about diagnostic accuracy of the 3 techniques. For these abnormalities, studies with a much larger number of patients would be necessary because of their very low prevalence. CONCLUSIONS
Due to the high number of abnormalities found in early postoperative cardiac surgery patients and the low accuracy showed by chest auscultation, routine imaging study is mandatory. In this scenario, chest ultrasound seems a valid alternative to chest x-ray to detect most postoperative abnormalities and misplacements. Furthermore, chest ultrasound avoids patients’ and staff’s radiation exposure, does not need additional resources outside the ICU, and easily can be repeated when the patient’s clinical condition changes.
REFERENCES 1. Radiology ACo: ACR Appropriateness Criteria. Routine chest radiographs in ICU. Available at: http://www.acr.org/Quality-Safety/ Appropriateness-Criteria/Diagnostic/Thoracic-Imaging. Accessed in November 2013. 2. Rubinowitz AN, Siegel MD, Tocino I: Thoracic imaging in the ICU. Critical Care Clinics 23:539-573, 2007 3. Krivopal M, Shlobin OA, Schwartzstein RM: Utility of daily routine portable chest radiographs in mechanically ventilated patients in the medical ICU. Chest 123:1607-1614, 2003 4. Clec'h C, Simon P, Hamdi A, et al: Are daily routine chest radiographs useful in critically ill, mechanically ventilated patients? A randomized study. Intensive Care Med 34:264-270, 2008 5. Oba Y, Zaza T: Abandoning daily routine chest radiography in the intensive care unit: Meta-analysis. Radiology 255:386-395, 2010 6. Mets O, Spronk PE, Binnekade J, et al: Elimination of daily routine chest radiographs does not change on-demand radiography practice in post-cardiothoracic surgery patients. J Thorac Cardiovasc Surg 134:139-144, 2007 7. Leong CS, Cascade PN, Kazerooni EA, et al: Bedside chest radiography as part of a postcardiac surgery critical care pathway: A means of decreasing utilization without adverse clinical impact. Crit Care Med 28:383-388, 2000 8. Tolsma M, Kroner A, van den Hombergh CL, et al: The clinical value of routine chest radiographs in the first 24 hours after cardiac surgery. Anesth Analg 112:139-142, 2011 9. Volpicelli G, Elbarbary M, Blaivas M, et al: International evidence-based recommendations for point-of-care lung ultrasound. Intensive Care Med 38:577-591, 2012 10. Ashton-Cleary DT: Is thoracic ultrasound a viable alternative to conventional imaging in the critical care setting? Brit J Anaesth 111: 152-160, 2013 11. Lichtenstein D, Goldstein I, Mourgeon E, et al: Comparative diagnostic performances of auscultation, chest radiography, and lung ultrasonography in acute respiratory distress syndrome. Anesthesiology 100:9-15, 2004 12. Tuddenham WJ: Glossary of terms for thoracic radiology: Recommendations of the Nomenclature Committee of the Fleischner Society. AJR. Am J Roentgenol 143:509-517, 1984 13. Woodring JH: Recognition of pleural effusion on supine radiographs: How much fluid is required? Am J Roentgenol 142:59-64, 1984 14. Restrepo CS, Lemos DF, Lemos JA, et al: Imaging findings in cardiac tamponade with emphasis on CT. Radiographics 27: 1595-1610, 2007
15. Lotano R, Gerber D, Aseron C, et al: Utility of postintubation chest radiographs in the intensive care unit. Crit Care 4:50-53, 2000 16. Owen RL, Cheney FW: Endobronchial intubation: A preventable complication. Anesthesiol 67:255-257, 1987 17. Lichtenstein DA, Meziere GA: Relevance of lung ultrasound in the diagnosis of acute respiratory failure: the BLUE protocol. Chest 134:117-125, 2008 18. Gargani L, Frassi F, Soldati G, et al: Ultrasound lung comets for the differential diagnosis of acute cardiogenic dyspnoea: A comparison with natriuretic peptides. European journal of heart failure 10: 70-77, 2008 19. Remerand F, Dellamonica J, Mao Z, et al: Multiplane ultrasound approach to quantify pleural effusion at the bedside. Intensive Care Med 36:656-664, 2010 20. Lichtenstein D, Hulot JS, Rabiller A, et al: Feasibility and safety of ultrasound-aided thoracentesis in mechanically ventilated patients. Intensive Care Med 25:955-958, 1999 21. Soldati G, Testa A, Sher S, et al: Occult traumatic pneumothorax: Diagnostic accuracy of lung ultrasonography in the emergency department. Chest 133:204-211, 2008 22. Weaver B, Lyon M, Blaivas M: Confirmation of endotracheal tube placement after intubation using the ultrasound sliding lung sign. Acad Emerg Med 13:239-244, 2006 23. Kerrey BT, Geis GL, Quinn AM, et al: A prospective comparison of diaphragmatic ultrasound and chest radiography to determine endotracheal tube position in a pediatric emergency department. Pediatrics 123:e1039-e1044, 2009 24. Flahault A, Cadilhac M, Thomas G: Sample size calculation should be performed for design accuracy in diagnostic test studies. J Clin Epidemiol 58:859-862, 2005 25. Landis JR, Koch GG: The measurement of observer agreement for categorical data. Biometrics 33:159-174, 1977 26. Silva S, Biendel C, Ruiz J, et al: Usefulness of cardiothoracic chest ultrasound in the management of acute respiratory failure in critical care practice. Chest 144:859-865, 2013 27. Lakhal K, Serveaux-Delous M, Lefrant JY, et al: Chest radiographs in 104 French ICUs: Current prescription strategies and clinical value (the RadioDay study). Intensive Care Med 38:1787-1799, 2012 28. Agricola E, Bove T, Oppizzi M, et al: “Ultrasound comet-tail images”: A marker of pulmonary edema: A comparative study with wedge pressure and extravascular lung water. Chest 127:1690-1695, 2005 29. Lichtenstein DA: Lung ultrasound in the critically ill. Ann Intensive Care 4:1, 2014
1532
30. Parlamento S, Copetti R, Di Bartolomeo S: Evaluation of lung ultrasound for the diagnosis of pneumonia in the ED. Am J Emerg Med 27:379-384, 2009 31. Lichtenstein D: FALLS-protocol: Lung ultrasound in hemodynamic assessment of shock. Heart, lung and vessels 5:142-147, 2013
VEZZANI ET AL
32. Vezzani A, Brusasco C, Palermo S, et al: Ultrasound localization of central vein catheter and detection of postprocedural pneumothorax: An alternative to chest radiography. Crit Care Med 38:533-538, 2010 33. Lichtenstein DA, Lascols N, Meziere G, et al: Ultrasound diagnosis of alveolar consolidation in the critically ill. Intensive Care Med 30:276-281, 2004
