Anda di halaman 1dari 5

The American Journal of Surgery xxx (2016) 1e5

Contents lists available at ScienceDirect

The American Journal of Surgery


journal homepage: www.ajconline.org

Clinical relevance of the routine daily chest X-Ray in the surgical


intensive care unit
Shelby Resnick, Kenji Inaba*, Efstathios Karamanos, Dimitra Skiada, James A. Dollahite,
Obi Okoye, Peep Talving, Demetrios Demetriades
University of Southern California, Keck School of Medicine, Department of Surgery, Division of Acute Care Surgery, Los Angeles County General Hospital
(LACþUSC), 2051 Marengo St, IPT, C5L100, Los Angeles, CA 90033, USA

a r t i c l e i n f o a b s t r a c t

Article history: Introduction: A daily Chest X-ray (CXR) is obtained in many surgical intensive care units (SICU). This
Received 16 June 2016 study implemented a selective CXR protocol in a high volume, academic SICU and evaluated its impact on
Received in revised form clinical outcomes.
23 September 2016
Methods: All SICU patients admitted in 2/2010 were compared with patients admitted in 2/2012. Be-
Accepted 29 September 2016
tween the time periods, a protocol eliminating the routine daily CXRs was instituted.
Results: In 02/2010 and 02/2012, 107 and 90 patients were admitted to the SICU, respectively, for a total
Keywords:
of 1384 patient days. CXRs decreased from 365 (57.1% of patient-days) in 2010 to 299 (40.9% of patient
CXR
ICU
days; p < 0.001) in 2012. A greater proportion of Physician Directed CXRs (PDCXRs) had new findings
Resource utilization (80.8%) compared to Automatic Daily CXRs (ADCXRs) (23.5%, p < 0.001). There was no difference in
overall or SICU length of stay, ventilator-free days, morbidity or mortality.
Conclusion: Eliminating ADCXRs decreased the number of CXRs performed, without affecting LOS, me-
chanical ventilation, morbidity or mortality. Physician-directed ordering of CXRs increased the diagnostic
value of the CXR and decreased the number of clinically irrelevant CXRs performed.
© 2016 Elsevier Inc. All rights reserved.

1. Background ICUs are still performing ADCXRs.7,25e27 A 2012 survey of 104 ICUs
found that significant variations exist amongst ICUs regarding CXR
In many intensive care units (ICU) traditional practice mandates usage. Almost 1/3 of the ICUs in the study were still obtaining daily
a daily Chest X-Ray (CXR) for all ICU patients, despite a growing CXRs.26 A more recent study by Tolsma et al., in 2014 found that
body of literature demonstrating its limited efficacy.1e17 Pro- while only 7% were obtaining mandatory daily CXRs on all patients,
ponents of the automatic daily CXR (ADCXR) advocate for a routine 39% of ICUs had some form of routine CXR protocol in place.25
CXR citing the earlier diagnosis of pneumonias and pneumo- At our own institution the practice of obtaining routine CXRs
thoraces, the ability to detect the movement of lines or tubes, daily was the standard protocol in place for all newly admitted patients
fluid status evaluations and easier scheduling for radiology per- until the performance of this study. The purpose of this study was
sonnel.18e24 However, the ADCXR is associated with an increased to implement a selective CXR protocol in our surgical intensive care
radiation burden to the patient, higher hospital resource utilization, unit and to examine the impact of this change not only on patient
and the need for patient repositioning which may result in the outcomes but also the impact on practitioner ordering behaviors.
displacement of lines and tubes. Despite this controversy, many We hypothesized that institution of the protocol would lead to
changes in practitioner ordering behaviors, eliminating the daily
CXR from standard practice and decreasing resource utilization and
* Corresponding author. University of Southern California, Keck School of Medi-
radiation burden without negatively impacting patient outcomes.
cine, Department of Surgery, Division of Acute Care Surgery, Los Angeles County
General Hospital (LACþUSC), 2051 Marengo St, IPT, C5L100, Los Angeles, CA 90033,
USA. 2. Methods
E-mail addresses: shelby.resnick@uphs.upenn.edu (S. Resnick), kenji.inaba@
med.usc.edu (K. Inaba), ef.karamanos@gmail.com (E. Karamanos), dimitra.skiada@
gmail.com (D. Skiada), dollahit@med.usc.edu (J.A. Dollahite), obi.okoye@hotmail.
2.1. Study design
com (O. Okoye), peep.talving@usc.edu (P. Talving), demetria@usc.edu
(D. Demetriades). Approval for this study was obtained from the USC Institutional

http://dx.doi.org/10.1016/j.amjsurg.2016.09.059
0002-9610/© 2016 Elsevier Inc. All rights reserved.

Please cite this article in press as: Resnick S, et al., Clinical relevance of the routine daily chest X-Ray in the surgical intensive care unit, The
American Journal of Surgery (2016), http://dx.doi.org/10.1016/j.amjsurg.2016.09.059
2 S. Resnick et al. / The American Journal of Surgery xxx (2016) 1e5

Review Board. This was a prospective, observational study of pa- their ICU admission, February of 2010 or 2012. To account for dis-
tients admitted to the Surgical Intensive Care Unit (SICU) at The Los crepancies in the number of patients admitted to the ICU and their
Angeles County þ University of Southern California (LAC þ USC) ICU LOS, during the given month, CXR frequencies were calculated
Medical Center. The LAC þ USC Medical Center is an American Col- using patient days as the denominator and number of CXRs per-
lege of Surgeons (ACS) verified, Level 1 trauma center, admitting an formed as the numerator.
average of 5000 trauma patients, annually. The 30-bed SICU admits Continuous variables were compared using Student's t-test,
all critically ill trauma and surgery patients, excluding burn, cardiac while dichotomous variables were compared using [Chi2] or
surgery, and pediatric patients who are housed in separate ICUs and Fisher's exact test, as appropriate. Continuous variables were
cared for by a separate team of surgeons or pediatric intensivists. dichotomized using clinically relevant cut-points to include age
Annually, our SICU admits over 1200 critically ill patients, the ma- >55 years, APACHE II  20 and BMI <25, 25e29.9, and >30.
jority of whom require mechanical ventilation. Patients admitted to Values are reported as mean ± standard deviation (SD) for
the SICU are cared for by a dedicated team consisting of critical care continuous variables and as percentages for categorical variables. All
fellows, residents and nurse practitioners. The team is supervised by statistical analysis was performed using Statistical Package for Social
a critical care board certified attending surgeon or physician. Sciences (SPSS Windows©), version 17.0 (SPSS Inc., Chicago, IL.)
This prospective study was conducted in 2 distinct phases over 2
years. The first phase of data collection occurred in February 2010, 3. Results
during which time it was the standard practice for all patients
newly admitted to the SICU to receive a daily CXR between the In February 2010 and February 2012, 107 and 90 patients were
hours of midnight and 5am. This could continue for the entire SICU admitted to the ICU, respectively. The average age was 31 ± 12 years
stay or could be stopped by the critical care team at any point (range 18e71 years), 68.2% were male. The two populations were
during the stay. During this time period, the new CXR protocol was equivalent with respect to demographics, comorbidities and ICU
being drafted, however, all patients were subjected to the old interventions. The severity of illness was not significantly different
protocol. Beginning July 2010 a protocol was implemented elimi- between the two cohorts, 29% of patients in 2010 and 23% of pa-
nating the automatic ordering of the daily CXR. The ability to order tients in 2012 had APACHE II scores >20 (p ¼ 0.42). Of the 197 total
an automatic daily CXR on the ICU admission orders was removed ICU patients, the majority were admitted for traumatic injuries
and the critical care team had to order CXRs on a selective basis, (118, 60%), followed by oncologic diseases (29, 14.4%) and after
only when deemed necessary for patient care. In February 2012, general surgical procedures (29, 14.4%). All patient characteristics
once the physicians, nurses and ancillary staff became accustomed are depicted in Table 1.
to the new protocol, the second phase of data collection began. All patients underwent an initial CXR on admission. There was
During each phase, data was prospectively collected for all patients no difference in admission CXR findings between the two years.
admitted to the SICU during the first 28 days of February; the year The most common pathology found on the initial CXR was lung
2012 was a leap year, therefore, data from February 29th was not consolidation (43, 21.8%) and 14.7%29 were read as normal.
included in the analysis.
The total number of CXRs obtained during each time period was 3.1. Frequency of CXRs
then captured. Each CXR was reviewed for indication and compared
to the prior CXR for the presence of new findings, which included The 197 ICU patients accumulated a total of 1384 patient days in
atelectasis, pulmonary edema, and changes in lung aeration in the ICU. Six-hundred and sixty-four CXR were performed during
addition to new disease processes. The final attending radiologist the two time periods, 48% of patient-days. More CXRs were per-
read for the CXR was utilized for analysis. All CXRs were obtained as formed in 2010, 365 CXRs (57.1% of patient-days) compared to 299
portable, anterior-posterior images performed at bedside in the CXRs (40.9% of patient days) in 2012 (p < 0.01). The type of CXR also
SICU. Any read regarding adequacy of insertion, removal or new changed. Overall, 121 PDCXRs and 341 ADCXRs were performed.
positioning of any central venous catheters, swan-ganz catheters, The number of ADCXRs performed in 2010 (217, 34.0% of patient
endotracheal tubes, decompression or feeding tubes, and epidural days) was significantly higher than the number performed in 2012
catheters was considered a new finding. These findings were com- (129, 17.6% of patient days, p < 0.01). The number of PDCXRs
bined for analysis into a single group, tube and line management. increased from 6.4% of patient days in 2010 to 10.9% of patient days
All patients underwent an admission CXR. All subsequent CXRs in 2012. These changes in trends between years are depicted in
performed during the patient ICU LOS were categorized as those Fig. 1.
with and without a specific indication. All CXRs performed in 2010
as part of the ADCXR protocol were analyzed as an ADCXR. All CXRs 3.2. Clinical value
ordered in 2012 without a specific indication were also included in
this group. All CXRs performed, in both time frames, with a specific The clinical and diagnostic value of the ADCXRs and PDCXRs was
indication were categorized as PDCXRs. Provided a specific indi- assessed by reviewing the findings on all CXRs. In comparison to
cation for obtaining the CXR was documented, CXRs could still be the CXR prior, a greater percent of PDCXRs had new findings (97,
ordered for the following day, these CXRs were considered PDCXRs. 80.8%) than ADCXRs (80, 23.5%; p < 0.001). Atelectasis, pneumo-
In addition to information regarding the CXR, patient de- thoraces (PTXs) and improved aeration were found more often on
mographics, comorbidities, ICU interventions, and admitting sur- ADCXRs (p < 0.01). None of these findings were clinically signifi-
gical service were documented. Outcome data collected included cant, and would not have changed management. All PTXs that were
mortality, overall length of stay (LOS) and ICU LOS, mechanical identified were small apical PTXs that required no further treat-
ventilator free days, and complications, including pneumonia ment. In contrast, the PDCXRs identified more new pneumonias (0,
(PNA), acute respiratory distress syndrome (ARDS), and pulmonary 0.0% vs 9, 9.3%; p < 0.01) and were utilized more often to manage
embolism (PE). and confirm placement of tubes and lines (0, 0.0% vs 78, 80.4%;
p < 0.01). Pulmonary edema was also diagnosed more frequently in
2.2. Statistics the PDCXR group than the ADCXR group (0, 0.0% vs 5, 5.2%) but did
not reach statistical significance (p ¼ 0.065). Comparison of all CXR
The study population was compared based on the time period of findings are described in Table 2.

Please cite this article in press as: Resnick S, et al., Clinical relevance of the routine daily chest X-Ray in the surgical intensive care unit, The
American Journal of Surgery (2016), http://dx.doi.org/10.1016/j.amjsurg.2016.09.059
S. Resnick et al. / The American Journal of Surgery xxx (2016) 1e5 3

Table 1
Patient characteristics.

Overall Feb 2010 Feb 2012 p value

(n ¼ 197) (n ¼ 107) (n ¼ 90)

Demographics
Age  55 82 (41.6%) 47 (43.9%) 35 (38.9%) 0.56
Gender (Male) 133 (67.5%) 70 (65.4%) 63 (70.0%) 0.54
Overweight (BMI ¼ 25e29.9) 63 (32.0%) 35 (32.7%) 28 (31.1%) 0.88
Obese (BMI  30) 77 (39.1%) 42 (39.3%) 35 (38.9%) 1.00
Comorbidities
DM 19 (9.6%) 10 (9.3%) 9 (10.0%) 1.00
Cardiovascular Disease 26 (13.2%) 11 (10.3%) 15 (16.7%) 0.21
Alcohol abuse 4 (2.0%) 2 (1.9%) 2 (2.2%) 1.00
Cirrhosis 5 (2.5%) 3 (2.8%) 2 (2.2%) 1.00
HIV infection 1 (0.5%) 0 (0.0%) 1 (1.1%) 0.47
Severity of disease
Apache II > 20 52 (26.4%) 31 (29.0%) 21 (23.3%) 0.42
Interventions
Chest Tube 40 (20.3%) 24 (22.4%) 16 (17.8%) 0.48
Mechanical Ventilation 93 (47.2%) 50 (46.7%) 43 (47.8%) 0.89
Days of Mechanical Ventilation 2.7 ± 1.6 2.5 ± 1.3 3.0 ± 2.4 0.64
Central Venous Catheter 61 (31.0%) 30 (28.0%) 31 (34.4%) 0.36
Surgical service
Trauma 118 (59.9%) 62 (57.9%) 56 (62.2%) 0.56
Oncology 29 (14.7%) 17 (15.9%) 12 (12.8%) 0.53
General 29 (14.4%) 16 (15.0%) 13 (13.8%) 0.70
Vascular 9 (4.6%) 4 (3.7%) 5 (5.6%) 0.54
Neurosurgery 7 (3.6%) 4 (3.7%) 3 (3.3%) 1.00
Urology 4 (2.0%) 3 (2.8%) 1 (1.1%) 0.40

Abbreviations: BMI: Body Mass Index; DM: Diabetes Mellitus; HIV: Human Immunodeficiency Virus.

Table 2
Findings on chest X-Ray (CXR).

Automatic Daily (n ¼ 341) Physician Directed (n ¼ 120) p value

New Findingsa 23.5% (80) 80.8% (97) <0.01


Atelectasis 66.2% (53) 1.0%1 <0.01
Pneumothorax 16.2% (13) 2.1% (2) <0.01
Improved Aeration 18.8% (15) 0% (0) <0.01
Pneumonia 0% (0) 9.3% (9) 0.04
Pulmonary Edema 0% (0) 5.2% (5) 0.65
Management of Tubes/Lines 0% (0) 80.4% (78) <0.01

Percentages calculated from total CXRs with new findings, not overall.
a
Compared to previous CXR.

3.3. Outcomes

The overall mortality was 5% (n ¼ 10) and did not differ between
the 2 years (p ¼ 0.52). Likewise, there was no difference between
CXR Frequencies
the two years in terms of overall LOS, ICU LOS, or ventilator free
2010 2012
60.0% 57.1% days. Outcome comparisons are demonstrated in Table 3. Addi-
tionally, there was no difference in complications, with equivalent
50.0%
numbers of PNA, ARDs, PE and DVT in each year.

40.9%
40.0% 4. Discussion
34.0%

30.0% The purpose of this study was to evaluate CXR usage in our SICU
before and after eliminating a protocol of routine daily CXRs and
the implementation of a physician directed, selective CXR ordering
20.0% 17.6%
protocol. Additionally, this study was an attempt to take the “next
10.9% step” in evidence based medicine, that is, to study not only patient
10.0%
6.4% outcomes but the behavioral changes of practitioners in an aca-
demic environment by observing practitioner ordering behaviors
0.0% after the implementation of our evidence-based CXR protocol. Also,
Overall ADCXR PDCXR we set out to evaluate the clinical relevance of the ADCXR in
Fig. 1. Chest X-Ray (CXR) Frequencies. All p values < 0.01. Values reported as percent of
comparison to the PDCXR. The study compared equivalent time
CXRs performed per number of patient days. ADCXR: Automatic Daily Chest X-Ray; periods before and after the institution of the selective ordering
PDCXR Physician Directed Chest X-Ray. protocol. Our analysis demonstrated a clinically important overall

Please cite this article in press as: Resnick S, et al., Clinical relevance of the routine daily chest X-Ray in the surgical intensive care unit, The
American Journal of Surgery (2016), http://dx.doi.org/10.1016/j.amjsurg.2016.09.059
4 S. Resnick et al. / The American Journal of Surgery xxx (2016) 1e5

Table 3 ventilation. These findings are in line with multiple other studies
Outcomes. including the largest meta-analysis performed on this top-
Overall Feb 2010 Feb 2012 p value ic.1e3,9,10,13e16,19 This meta-analysis analyzed 9 studies with a
(n ¼ 197) (n ¼ 107) (n ¼ 90)
combined total of 39,358 CXRs for 9611 patients and found no
difference in ICU or hospital mortality, ICU or overall LOS or dura-
Mortality 10 (5.1%) 4 (3.7%) 6 (6.7%) 0.52
tion of mechanical ventilation.10 Based on these findings and the
Complications
Pneumonia 24 (12.2%) 13 (12.1%) 11 (12.2%) 1.00 findings of our own study we conclude that ADCXRs can safely be
ARDS 4 (2.0%) 3 (2.8%) 1 (1.1%) 0.48 eliminated from the ICU.
Pulmonary Embolism 6 (3.0%) 3 (2.8%) 3 (3.3%) 1.00 The majority of studies advocating daily routine CXRs were
DVT 3 (1.5%) 2 (1.9%) 1 (1.1%) 1.00
performed over 15 years ago. These studies concluded that signif-
Bacteremia/Sepsis 14 (7.1%) 10 (9.3%) 4 (4.4%) 0.18
UTI 9 (4.6%) 4 (3.7%) 5 (5.6%) 0.73 icant injuries, including pneumothoraces in mechanically venti-
ARF 14 (7.1%) 6 (5.6%) 8 (8.9%) 0.37 lated patients, new pneumonias and malpositioned lines or tubes,
CVI 15 (7.6%) 7 (6.5%) 8 (8.9%) 0.54 could be missed without daily CXRs.18e24 However, more recent
Ileus 2 (1.0%) 0 (0.0%) 2 (2.2%) 0.22 studies, including ours have shown the rate of missed injuries to be
SSI 14 (7.1%) 6 (5.6%) 8 (8.9%) 0.37
quite low. Explanations for this shift in findings over time maybe
Length of Stay
Hospital LOS 4.54 ± 2.37 4.21 ± 2.10 4.35 ± 3.11 0.67 related to the multiple changes that have occurred in the care of the
ICU LOS 2.13 ± 1.65 2.80 ± 1.63 1.95 ± 2.55 0.35 critically ill patient over the last 2 decades. Improvements in critical
Ventilator Free Days 2.03 ± 2.00 1.98 ± 1.05 2.30 ± 1.87 0.46 care have led to decreased mechanical ventilation days and a
Abbreviations: ARDS: Acute Respiratory Distress Syndrome; DVT: Deep Venous reduction in the number of ventilator-associated pneumonias
Thrombosis; UTI: Urinary Tract Infection; ARF: Acute Renal Failure; CVI: Cardio- (VAP), invasive lines and tubes. There is now a higher awareness of
vascular Incidents; SSI: Surgical Site Infections; LOS: Length of Stay. the importance of early extubation, implementation of QI check-
lists, a better understanding of ARDS and VAP, use of decreased
sedation protocols, and a push toward less invasive mon-
decrease in the number of CXRs performed by 17%. This compares itoring.29e34 After a review of the literature in 2013, The American
favorably with the current literature ranging from 22% to College of Radiology changed their recommendations regarding
52%.2,8,9,13,28 CXRs in ICU.7,9,15,35 Routine CXRs are no longer recommended for
In regards to changes in practitioner ordering behaviors, it is the stable patient, regardless of mechanical ventilation status. They
interesting to note that, even in 2012, on 17.6% of patient days, a continue to recommend CXR after placement of endotracheal tube,
CXR was ordered and performed without a specific indication. A central venous lines, Swan-Ganz catheters, nasogastric tubes,
possible explanation for this, as was shown in the study by Rao feeding tubes and chest tubes.35
et al., is the learning curve that exists after implementation of a new As healthcare costs in our nation continue to rise, overutilization
protocol. Although the providers were instructed to not order daily has been identified as a major contributor to these inflated
CXRs many still did, perhaps out of habit or comfort level.14 Lakhal healthcare costs.36,37 Reducing patient length of stays, eliminating
also observed an element of non-adherence to the reported pro- excess radiographic studies and implementing more cost effective
tocols in their 2012 survey.26 We attempted to mitigate this effect protocols are critical for controlling expenditures.38 When factoring
by designing a break-in phase of two years and removing the ability in the additional costs of transport and radiology personnel the
to order automatic daily CXRs, however, as was seen here, a small daily CXR can increase hospital expenses considerably. Price et al.
number of providers continued to obtain CXRs without a clinical provided a conservative estimate of savings of $45,475.92 over a 2.5
indication. year period however other studies have estimated cost savings to
Perhaps more important than the 17% decrease in imaging is the be twice this and as high as $650,000 per year.2,8,12,17 While direct
clinical relevance of the imaging obtained. Compared to the ADCXR, costs cannot be compared given the difference in when and where
the PDCXR was of high diagnostic value. In 80% of the PDCXRs a these studies were performed the message is still clear, a reduction
change from the prior CXR was found, the majority of these findings in CXRs saves money.
pertinent to the overall treatment of the patient. This is in stark Future studies should not focus on the diagnostic value of the
contrast to only 23% of new findings discovered on the ADCXR, none PDCXR, as this has already been shown to be higher than ADCXR in
of which changed the management of the patient. In a study of 1354 multiple studies. Rather, the next step is to examine the clinical and
CXRs from 167 patients in a respiratory ICU, Bekenmeyer et al., physical exam findings and objective data which should trigger the
found that while the PDCXR had higher diagnostic value the re- ordering of a CXR, in order to optimize their diagnostic val-
ported rate of new findings in the ADCXR was 45% of which 40% ue.1e3,5,16,17,21 Our study is limited in that we did not evaluate for the
required intervention.21 They therefore concluded that routine CXR clinical indications that led physicians to order CXRs. By evaluating
in the ICU was useful and necessary. However, a larger investigation what prompts the ordering of diagnostic CXRs, i.e changes in
with a study design similar to ours in a mixed medical-surgical ICU, oxygenation or respiratory status, fever, increased white blood cell
by Graat et al., found that both the diagnostic and therapeutic value counts, we may be able to create and institute more accurate protocols.
of the on-demand CXR was increased with the elimination of The results of our study show that by instituting a selective CXR
ADCXRs.3 Another Netherlands based study also found that routine protocol we safely and successfully decreased the number of CXRs
CXRs had very low diagnostic and therapeutic yields, with only 4.4% performed and increased the diagnostic value of the CXR when
of the routine CXRs providing clinically relevant information and physician-directed ordering was applied. Our data adds to the
less than half of those resulting in a change in patient management.2 growing body of literature supporting the removal of daily routine
Clec'h et al., looked at a different patient population than our study, CXRS from ICUs. Additionally it shows that implementation of a
all mechanically ventilated patients in a mixed medical and surgical unit-wide protocol can be successfully instituted in a busy Level I
ICU, and again showed a low diagnostic value for the routine CXR. trauma center SICU where there is high turnover of housestaff.
Routine daily CXRs were therapeutic in only 5% whereas 56% of the
non-routine CXRs resulted in an intervention (p ¼ 0.01).16 5. Conclusion
Overall, we found no differences in outcomes, including mor-
tality, complications, LOS or requirement for mechanical The routine, automatically ordered daily CXR was found to have

Please cite this article in press as: Resnick S, et al., Clinical relevance of the routine daily chest X-Ray in the surgical intensive care unit, The
American Journal of Surgery (2016), http://dx.doi.org/10.1016/j.amjsurg.2016.09.059
S. Resnick et al. / The American Journal of Surgery xxx (2016) 1e5 5

very low clinical relevance. Institution of a SICU protocol elimi- 14. Rao PSR, Abid Q, Khan KJ, et al. Evaluation of routine postoperative chest X-rays
in the management of the cardiac surgical patient. Eur J Cardio Thoracic Surg.
nating the routine CXR from daily practice was successful in
1997 Dec 8;12:724e729.
decreasing the number of CXRs performed, increasing the diag- 15. Oba Y, Zaza T. Abandoning daily routine chest radiography in the intensive care
nostic yield and therapeutic value of the CXRs ordered. This did not unit: meta-analysis1. Radiology. 2010 May 13;255(2):386e395. Radiological
negatively affect ICU LOS, overall LOS, need for mechanical venti- Society of North America, Inc.
16. Clec’h C, Simon P, Hamdi A, et al. Are daily routine chest radiographs useful in
lation, morbidity or mortality. By eliminating the routine CXR, cost- critically ill, mechanically ventilated patients? A randomized study. Intensive
effective care is increased without detriment to the patient. Care Med. 2007 Nov 10;34(2):264e270.
17. Silverstein DS, Livingston DH, Elcavage J, Kovar L, Kelly KM. The utility of
routine daily chest radiography in the surgical intensive care unit. J Trauma Inj
Conflict of interest statement Infect Crit Care. 1993 Oct;35(4):643e646.
18. Horst HM, Fagan B, Beute GH. Chest radiographs in surgical intensive care
All authors, Shelby Resnick, Kenji Inaba, Efstathios Karamanos, patients: a valuable “routine”. Henry Ford Hosp Med J. 1986;34(2):84e86.
19. Brainsky A, Fletcher RH, Glick HA, Lanken PN, Williams SV, Kundel HL. Routine
Dimitra Skiada, James A. Dollahite, Obi Okoye, Peep Talving, and portable chest radiographs in the medical intensive care unit: effects and costs.
Demetrios Demetriades, have no conflicts of interest or other Crit Care Med. 1997 May;25(5):801e805.
disclosures. 20. Hall JB, White SR, Karrison T. Efficacy of daily routine chest radiographs in
intubated, mechanically ventilated patients. Crit Care Med. 1991 May;19(5):
689e693.
Prior manuscript presentations 21. Bekemeyer WB, Crapo RO, Calhoon S, Cannon CY, Clayton PD. Efficacy of chest
radiography in a respiratory intensive care unit. A prospective study. Chest J.
1985 Nov;88(5):691e696.
Annual Meeting of AAST and Clinical Congress of Acute Care
22. Hauser GJ, Pollack MM, Sivit CJ, Taylor GA, Bulas DI, Guion CJ. Routine chest
Surgery, poster presentation, San Francisco, CA, September 2013. radiographs in pediatric intensive care: a prospective study. Pediatrics. 1989
Apr;83(4):465e470.
Funding 23. Henschke CI, Pasternack GS, Schroeder S, Hart KK, Herman PG. Bedside chest
radiography: diagnostic efficacy. Radiology. 1983 Oct;149(1):23e26.
24. Greenbaum DM, Marschall KE. The value of routine daily chest x-rays in
This research did not receive any specific grant from funding intubated patients in the medical intensive care unit. Crit Care Med. 1982
agencies in the public, commercial, or not-for-profit sectors. Jan;10(1):29e30.
25. Tolsma M, Rijpstra TA, Schultz MJ, Mulder PG, van der Meer NJ. Significant
changes in the practice of chest radiography in Dutch intensive care units: a
References web-based survey. Ann Intensive Care. 2014;4(1):10.
26. Lakhal K, Serveaux-Delous M, Lefrant JY, Capdevila X, Jaber S. Chest radio-
1. Krivopal M, Shlobin OA, Schwartzstein RM. Utility of daily routine portable graphs in 104 French ICUs: current prescription strategies and clinical value
chest radiographs in mechanically ventilated patients in the medical ICU*. (the RadioDay study). Intensive Care Med [Internet]. 2012 Aug 2;38(11):
Chest J. 2003 May 24;123(5):1607e1614. 1787e1799. Available from: http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.
2. Hendrikse KA. Low value of routine chest radiographs in a mixed medical- fcgi?dbfrom¼pubmed&id¼23011527&retmode¼ref&cmd¼prlinks.
surgical ICU*. Chest J. 2007 Sep 1;132(3):823. 27. Hejblum G, Ioos V, Vibert J-F, et al. A web-based Delphi study on the in-
3. Graat ME, Kroner A, Spronk PE, et al. Elimination of daily routine chest ra- dications of chest radiographs for patients in ICUs. Chest J. 2008 May;133(5):
diographs in a mixed medical-surgical intensive care unit. Intensive Care Med. 1107e1112.
2007 Apr;33(4):639e644. 28. Bhagwanjee S, Muckart DJ. Routine daily chest radiography is not indicated for
4. Leong C, Cascade PN, Kazerooni EA, Bolling SF, Deeb GM. Bedside chest radi- ventilated patients in a surgical ICU. Intensive Care Med. 1996 Dec;22(12):
ography as part of a postcardiac surgery critical care pathway: a means of 1335e1338.
decreasing utilization without adverse clinical impact. Crit Care Med. 2000 Feb 29. Checkley W, Brower R, Korpak A, Thompson BT. Effects of a clinical trial on
1;28(2):383e388. mechanical ventilation practices in patients with acute lung injury. Am J Respir
5. Chahine-Malus N, Stewart T, Lapinsky SE, et al. Utility of routine chest radio- Crit Care Med. 2008 Jun;177(11):1215e1222.
graphs in a medical-surgical intensive care unit: a quality assurance survey. Crit 30. Shoemaker WC, Belzberg H, Wo CCJ, et al. Multicenter study of noninvasive
Care. 2001 Oct;5(5):271e275. monitoring systems as alternatives to invasive monitoring of acutely III
6. Strain DS, Kinasewitz GT, Vereen LE, George RB. Value of routine daily chest x- emergency patients. Chest J Am Coll Chest Physicians. 1998 Dec 1;114(6):
rays in the medical intensive care unit. Crit Care Med. 1985 Jul;13(7):534e536. 1643e1652.
7. Graat ME, Choi G, Wolthuis EK, et al. The clinical value of daily routine chest 31. DuBose JJ, Inaba K, Shiflett A, et al. Measurable outcomes of quality improve-
radiographs in a mixed medical-surgical intensive care unit is low. Crit Care. ment in the trauma intensive care unit: the impact of a daily quality rounding
2006 Feb;10(1):R11. checklist. J Trauma Inj Infect Crit Care. 2008 Jan;64(1), 22e27e discussion 27e9.
8. Price MB, Grant MJ, Welkie K. Financial impact of elimination of routine chest 32. Shoemaker WC, Thangathurai D, Wo CC, et al. Intraoperative evaluation of
radiographs in a pediatric intensive care unit. Crit Care Med. 1999 Aug;27(8): tissue perfusion in high-risk patients by invasive and noninvasive hemody-
1588e1593. namic monitoring. Crit Care Med. 1999 Oct;27(10):2147e2152.
9. Hejblum G, Chalumeau-Lemoine L, Ioos V, et al. Comparison of routine and on- 33. Cocanour CS, Peninger M, Domonoske BD, et al. Decreasing ventilator-
demand prescription of chest radiographs in mechanically ventilated adults: a associated pneumonia in a trauma ICU. J Trauma Inj Infect Crit Care. 2006
multicentre, cluster-randomised, two-period crossover study. Lancet. 2009 Jul;61(1), 122e129e discussion 129e30.
Nov;374(9702):1687e1693. 34. Vogt KN, Frankel H. Maintaining comfort, cognitive function, and mobility in
10. Ganapathy A, Adhikari NKJ, Spiegelman J, Scales DC. Routine chest x-rays in surgical intensive care unit patients. J Trauma Acute Care Surg. 2014 Aug;77(2):
intensive care units: a systematic review and meta-analysis. Crit Care. 364e375.
2012;16(2):R68. 35. Amorosa JK, Bramwit MP, Mohammed T-LH, et al. ACR appropriateness criteria
11. Cruz J, Ferra M, Kasarabada A, Gasperino J, Zigmund B. Evaluation of the clinical routine chest radiographs in intensive care unit patients. J Am Coll Radiol. 2013
utility of routine daily chest radiography in intensive care unit patients with Mar;10(3):170e174.
tracheostomy tubes: a retrospective review. J Intensive Care Med. 2016 36. Emanuel E, Tanden N, Altman S, et al. A systemic approach to containing health
Jun;31(5):333e337. care spending. N. Engl J Med. 2012 Sep;367(10):949e954.
12. Fong Y, Whalen GF, Hariri RJ, Barie PS. Utility of routine chest radiographs in 37. Emanuel EJ, Fuchs VR. The perfect storm of overutilization. JAMA. 2008
the surgical intensive care unit. A prospective study. Arch. Surg. 1995 Jun;299(23):2789e2791.
Jul;130(7):764e768 (Chicago, Ill: 1960). 38. Mathias JS, Feinglass J, Baker DW. Variations in US hospital performance on
13. Krinsley JS. Test-ordering strategy in the intensive care unit. J Intensive Care imaging-use measures. Med Care. 2012 Sep;50(9):808e814.
Med. 2003 Nov 1;18(6):330e339.

Please cite this article in press as: Resnick S, et al., Clinical relevance of the routine daily chest X-Ray in the surgical intensive care unit, The
American Journal of Surgery (2016), http://dx.doi.org/10.1016/j.amjsurg.2016.09.059