Diagnostic Radiology

Australasian Radiology (2001) 45, 444447

Efficacy of daily bedside chest radiography as visualized by digital luminescence radiography

Johannes Kirchner, Christoph A Stueckle, Esther M Schilling and Jutta Peters
Department of Diagnostic and Interventional Radiology, Klinikum Niederberg Velbert, University Essen, Germany

SUMMARY
To determine the diagnostic impact of daily bedside chest radiography in comparison with digital luminescence technique (DLR; storage phosphor radiography) and conventional film screen radiography, a prospective randomized study was completed in 210 mechanically ventilated patients with a total of 420 analysed radiographs. The patients were allocated to two groups: 150 patients underwent DLR, and 60 patients underwent conventional film screen radiography. Radiological analysis was performed consensually and therapeutic efficacy was assessed by the clinicians. There was no statistical significant difference between the frequency of abnormal findings seen on DLR and conventional film screen radiography. In total, 448 abnormal findings were present in 249 of 300 DLR and 97 of 120 conventional film screen radiographs. The most common findings were signs of overhydration (41%), pleural effusion (31%), partial collapse of the lung (11%) and pneumothorax (2%). One hundred and twenty-three of 448 (27%) of these abnormal findings were thought to have a considerable impact on patient management. The high rate of abnormal findings with significant impact on patient management suggests that the use of daily bedside chest radiography may be reasonable. Key words: bedside chest radiography; computed radiography, efficacy study; intensive care unit.

INTRODUCTION
The significance of bedside chest radiography in the intensive care unit (ICU) has been discussed and is controversial. Numerous investigators have considered daily bedside radiography to be highly effective,
110

METHODS
In a prospective randomized study, we compared DLR and conventional film screen radiography of the chest of surgical patients with mechanical ventilation for a period of at least 24 h. The patients in group A (n = 150) underwent the first and second bedside chest radiograph by digital luminescence technique after admission to the ICU. Group B included 60 patients who had their first and second bedside chest radiographs by conventional film screen technique. The study design was designed to be in accordance with the Helsinki Declaration. Digital luminescence radiography was performed by means of the DLR system PCR 9000 (Phillips Corporation, Hamburg, Germany) and the luminescence screen ST 5A (Fuji Corporation, Tokyo, Japan). The exposure parameters were identical for all patients (film focus distance 80120 cm, no grid, tube voltage 125 kV, 2 mm nominal focus size). All images were processed by use of two algorithms. The first algorithm

whereas others have found it to

be less effective.1116 At this time, there seems to be a trend in the radiological literature to question the value of the method in the daily monitoring of severely ill patients.11 Recently, some investigators described the incidence of findings requiring immediate intervention to be so low (< 1%) that routine daily radiographs could not be justified;14 whereas others suggested routine daily radiographs were only useful for patients with SwanGanz catheters.
11

The aim of the present study was to re-evaluate the diagnostic and therapeutic efficacy of daily bedside chest radiography in comparison with digital luminescence radiography (DLR), which is also called storage phosphor radiography, and conventional film screen radiography.

J Kirchner MD; CA Stueckle MD, EM Schilling MD; J Peters MD.

Correspondence: Johannes Kirchner, Medical Superintendent, Department of Diagnostic and Interventional Radiology, Robert Koch Strasse 2, 42549 Velbert, Germany. Email: biozol-welp@wtal.de
Submitted 27 November 2000; resubmitted 2 April 2001; accepted 5 April 2001.

EFFICACY OF DAILY BEDSIDE CHEST RADIOGRAPHY

produced an unenhanced image, whereas the second algorithm produced an edge-enhanced image by using an unsharp mask filter, an enhancement factor of 5 on a scale of 16 and a linear gradation curve. The conventional film screen radiographs (intensifying screen Cronex Quanta fast Detail; Dupont Corp, Bad Homburg, Germany) were performed without a grid (film focus distance 80120 cm, tube voltage 125 kV, 2 mm nominal focus size). All radiographs were evaluated consensually by two experienced radiologists blinded to clinical information before offering their interpretation of the findings. Standardized radiological analysis included quality assessment (exposure, centring), cardiac size, depth of inspiration, interstitial pattern, position of tubes and catheters, and presence of pleural effusion, pneumothorax or parenchymal opacification as a sign of atelectasis or pulmonary infiltration. The indications for performance of chest radiography were classified by the clinicians according to the following three categories: (i) assessment of catheters or tubes; (ii) specific indication (suspected infiltration, overhydration, atelectasis, pneumothorax); and (iii) follow-up examination. The radiological diagnoses were discussed during a daily conference by the clinical ICU team. Radiological findings were classified by the clinicans as follows: (i) no new information; (ii) unexpected new information without impact on patient management; and (iii) unexpected new information with impact on patient management. Following the suggestions of the American College of Radiology Efficacy Studies Committee the abnormal findings were classified as having either diagnostic efficacy only (E-1), which means unexpected new information without impact on therapy, or therapeutic efficacy (E-2) causing an immediate impact on patient management. Outcome efficacy (E-3) was not evaluated in the present study. Students t-test (in the case of Gaussian distribution) and Wilcoxons test were used for statistical analysis of quantitative values, and 2 test for qualitative findings. Probability values less than 0.05 were considered to be significant.

445
digital luminescence technique or conventional film screen examination. There was no significant difference between the two groups (Tables 1 and 2). The quality of 10/300 DLR and 12/120 conventional radiographs were classified by the observers as poor (10/10 DLR because of positioning, 8/12 conventional radiographs because of positioning, 7/12 conventional radiographs because of insufficient exposure). In total, 448 abnormal findings (Table 3) were seen on 349/420 radiographs (249/300 DLR and 97/120 conventional film screen radiographs). There was no significant difference between findings in DLR and conventional film screen radiographs at levels of P < 0.005. Pleural effusions were seen significantly more often in conventional film screen radiography at levels of P < 0.01. Pulmonary overhydration was diagnosed most often (173/420) followed by pleural effusion (149/420) and atelectasis (49/420). While pneumothorax was seen on 8/420 radiographs, only 6/420 radiographs showed signs of pulmonary infiltration.

Efficacy
Of 448 abnormal findings, 335 were classified to have no immediate impact on patient management and, therefore, showed only diagnostic efficacy (E-1). Of 448 abnormal findings, 123 also showed therapeutic efficacy (E-2). Here, incorrect positioning of tubes and catheters showed the highest impact on patient management followed by the detection of pneumothorax and unsuspected hyperhydration (Table 4).
Table 2. Indications for all 420 chest radiographs in group A (digital luminescence radiography) and group B (conventional film screen radiography) Cause of investigation Control Specific Follow up Total Group A (%) 105 57 138 30 (35) (19) (46) (100) Group B (%) 44 20 56 120 (36) (17) (47) (100)

RESULTS
Comparison of the demographic data as well as the underlying operations and indications confirms random selection to either

Control, postoperative control or control after intubation, catheter replacement or puncture; Specific, investigation because of suspected overhydration, fever or decline of oxygenation, etc; Follow up, investigation without specific indication. Percentages indicate percentages of all examinations.

Table 1. Demographic data and distribution of surgical procedures and trauma in group A (digital luminescence radiography) and group B (conventional film screen radiography) Group A No. patients (n) Age (years) Sex (male : female) Cardiac surgery Trauma Abdominal surgery Others 150 Median 62 (range 1592) 103:47 97 (65%) 26 (17%) 15 (10%) 12 (8%) Group B 60 Median 61 (range 1684) 39:21 40 (67%) 12 (20%) 5 (8%) 3 (5%)

446

J KIRCHNER ET AL.

Table 4 shows the findings on radiographs that were classified as follow-up studies and their impact on patient management. In 159/194 radiographs classified as follow-up studies, abnormal findings did not affect patient management. In 35/194 follow-up examinations, abnormal findings had an impact on patient management. Again, displacement of catheters and tubes had the highest impact on patient management.

technique has been demonstrated to be inferior to conventional film screen biplane chest radiography because of its lower spatial resolution.19,25 The benefits of DLR in chest radiography to the intensive care patient have been postulated repeatedly but, as yet, neither the diagnostic nor therapeutic efficacy of the method have been definitely determined.26 The American College of Radiology Efficacy Studies Committee defined three categories of efficacy: diagnostic efficacy, therapeutic efficacy and outcome efficacy (E-1, E-2, E-3).27 Most studies have been limited to diagnostic efficacy. In the present study, efficacy was defined as therapeutic efficacy (E-2), following the definition proposed by Gartenschlger et al.3 Beyond this, a comparison of the mentioned studies is difficult. There is no clear definition as to what a daily bedside radiograph means. In the present study, radiographs that were not specifically indicated by changes in clinical signs or symptoms or for documentation of catheter or tube positioning were defined as follow-up or routine examinations. In the present study, 349/420 radiographs demonstrated abnormal findings. This seems to be high, but is in accordance with the literature. Significant abnormal findings were found in up to 65% of daily radiographs in other efficacy studies.68 The relatively high share of significant abnormalities is possibly a result of the study design; to achieve comparable timing of examinations only the first and the second radiographs after admission to the ICU were evaluated. More abnormalities are expected during this stage. Conversely, the present design resulted in a low rate of pulmonary infiltrates. The radiographic features of overhydration and pleural effusion are the major abnormal findings in the present study, caused presumably by the high prevalence of patients undergoing cardiac surgery. Both findings showed an obviously low impact on patient management. Surprisingly, the difference between the impact of the abnormal findings in all radiographs and those that were classified as follow-up studies was not significant, with the exception of hyperhydration (Table 4).

DISCUSSION
Complex medical problems are common in the ICU. Complications from ventilatory assistance, parenteral nutrition and diminished immunological defence occur frequently and demand immediate diagnosis and treatment. Furthermore, chest radiography commonly supports the daily physical and apparative investigation.
17,10,1518

Increasingly, DLR has been

used to improve the efficacy of radiological examinations.19 The most important advantage of DLR is its high dynamic range, eliminating over and under exposures, especially in bedside radiography.
18

The positions of tubes, catheters and drains,

the diaphragm and retrocardiac areas of the left lung can be assessed more confidently by DLR.1924 Digital luminescence
Table 3. Distribution of radiological findings (n = 448) in all digital luminescence radiography (n = 300) and conventional film screen radiographs (n = 120) Findings Group A Group B 23 17 56 50 16 4 2 2
2

No abnormal finding 51 Catheter/tubes 39 (13) Pleural effusion 93 (31) Hyperhydration 123 (41) Atelectasis 33 (11) Infiltrate 2 (0.7) Pneumothorax 6 (2) Others 5 (1.6)

(14) (45) (42) (13) (3) (2) (2)

NS *P < 0.005; P < 0.01 NS NS NS NS NS NS

*Not significant at P < 0.005. Significant at P < 0.01. Catheter/tubes indicates displacement or incorrect positioning of catheters, drains and tubes; NS, not significant at P < 0.05. Percentages are percentages of all findings.

Table 4. Classification of the efficacy of the 448 abnormal findings in all 346 radiographs and in the 194 radiographs, which were classified as follow-up studies Finding All examinations Without impact (%) With impact (%) 6 137 126 42 6 3 5 (11) (92) (73) (86) (100) (37) (71) 50 12 47 7 0 5 2 (89) (8) (27) (14) (0) (63) (29) Follow-up examinations Without impact (%) With impact (%) 1 106 84 29 6 3 5 (5) (96) (92) (88) (100) (60) (83) 21 4 7 4 0 2 1 (95) (4) (8) (12) (0) (40) (1) 2

Catheter/tubes Pleural effusion Hyperhydration Atelectasis Infiltrate Pneumothorax Others

NS NS P < 0.005 NS NS NS

Catheter/tubes indicates displacement or incorrect positioning of catheters, drains and tubes; NS, not significant at P < 0.05. Percentages are percentages of the specific findings.

EFFICACY OF DAILY BEDSIDE CHEST RADIOGRAPHY

However, 16% of all routine radiographs of the present study that showed unexpected findings had an immediate impact on clinical management, which is in accordance with Strain et al.15 The fact that no gold standard for assessment of diagnostic accuracy exists is a major limitation of the studys results. Therefore, the modalities can only be assessed in relation to each other. Finally, the correct assessment of the impact of negative findings on the further therapeutic management seems to be difficult. Negative findings may prevent unnecessary diagnostic procedure or therapy.10
17. 16. 15. 14. 13.

447
radiography in comparison with clinical and microbiological monitoring. Akt Radiol 1998; 8: 2530. Lotano R, Gerber D, Aseron C, Santarelli R, Pratter M. Utility of postintubation chest radiographs in the intensive care unit. Crit Care (Lond) 2000; 4: 503. Silverstein DS, Livingston DH, Elcavage J. The utility of routine daily chest radiography in the surgical intensive care unit. J Trauma 1993; 35: 6436. Strain DS, Kinasewitz GT, Vereen LE, George AB. Value of routine daily chest X-rays in the medical intensive care unit. Crit Care Med 1985; 13: 5346. Tape TG, Mushlin AI. The utility of routine chest radiography. Ann Int Med 1986; 104: 66370. Gray P, Sullivan G, Ostryzniuk P. Value of postprocedural chest radiographs in the adult intensive care unit. Crit Care Med 1992; 20: 151318. Witte G, Bause HW, Pothmann W, Schwermer B, Schofer M, Bcheler E. Die Thoraxaufnahme auf der Intensivstation als digitale Lumineszenzradiographie (DLR). RFO Fortschr Rntgenstr 1988; 149: 4537. Neufang KF, Krug B, Lorenz R, Steinbrich W. Kann die digitale Luminescenzradiographie (DLR) in der Thoraxdiagnostik die konventionelle Rntgenaufnahmetechnik ersetzen? RFO Fortschr Rntgenstr 1990; 152: 5019. Frank MS, Jost RG, Molina PL et al. High-resolution computer display of portable, digital, chest radiographs of adults: Suitability for primary interpretation. Am J Roentgenol 1993; 160: 4737. Fuhrman CR, Gur D, Good B, Rockette H, Cooperstein LA, Feist JH. Storage phosphor radiographs vs conventional films: Interpreters perceptions of diagnostic quality. Am J Roentgenol 1988; 150: 101114. Jennings P, Pradley SP, Hansell DM. Portable chest radiography in intensive care: A comparison of computed and conventional radiography. Br J Radiol 1992; 65: 8526. Sonoda M, Takano M, Miyahara J, Kato H. Computed radiography utilizing scanning laser stimulated luminescence. Radiology 1983; 148: 8338. Lyttkens K, Anderson B, Kehler M, Lindberg L. Bedside chest radiography using digital luminescence: A comparison between digital radiographs reviewed on a personal computer and as hardcopies. Acta Radiol 1992; 33: 42730. Hbener KH. Scanned projection radiography of the chest versus standard film radiography: A comparison of 250 cases. Radiology 1983; 148: 3637. Paulin S. Comment to: Wandtke. Bedside chest radiography. Radiology 1994; 192: 2824. Loop JW, Lusted LB. American College of Radiology diagnostic efficacy studies. Am J Roentgenol 1978; 131: 1738.

REFERENCES
1. Bekemeyer WB, Crapo RO, Calhoon S, Cannon CY, Clayton PD. Efficacy of chest radiography in a respiratory care unit: A prospective study. Chest 1985; 88: 6916. Brunel W, Coleman D, Schwartz D. Assessment of routine chest roentgenograms and the physical examination to confirm endotracheal tube position. Chest 1989; 96: 10435. Gartenschlger M, Busch H, Kussmann J, Nafe B, Beyermann K, Klose KJ. Radiologisches Thoraxmonitoring bei beatmeten Intensivpatienten. RFO Fortschr Rntgenstrahlen 1996; 164: 95101. Greenbaum DM, Marschall KE. The value of routine daily chest X-rays in intubated patients in the medical intensive care unit. Crit Care Med 1982; 10: 2930. Hall JB, White SR, Karrison T. Efficiacy of daily routine chest radiographs in intubated, mechanically ventilated patients. Crit Care Med 1991; 19: 68993. Henschke CI, Pasternack GS, Schroeder S, Hart KK, Herman PG. Bedside chest radiography: Diagnostic efficacy. Radiology 1983; 149: 236. Henschke CI, Yankelevitz DF, Wand A, Davis SD, Shiau M. Accuracy and efficacy of chest radiography in the intensive care unit. Radiol Clin North Am 1996; 34: 2131. Janower ML, Jennas-Nocera Z, Mukai J. Utility and efficacy of portable chest radiographs. Am J Roentgenol 1984; 142: 2657. Sivit CJ, Taylor GA, Hauser GJ, Pollack MM, Bulas DI, Guion CJ, Fearon T. Efficacy of chest radiography in pediatric intensive care. Am J Roentgenol 1989; 152: 5757. Wandtke JC. Bedside chest radiography. Radiology 1994; 190: 110. Fong Y, Whalen G, Hariri RJ. Utility of routine chest radiographs in the surgical intensive care unit. Arch Surg 1995; 130: 7648. Kirchner J, Albert S, Thomas H, Behne M, Jacobi V. Diagnosis of pneumonia in long-term ventilated patients: Digital luminescence 18.

2.

19.

3.

20.

4.

21.

5.

6.

22.

7.

23.

8. 9.

24.

25.

10. 11. 12.

26. 27.

Copyright of Australasian Radiology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.