European Radiology (2020) 30:1839–1846

https://doi.org/10.1007/s00330-019-06551-8

COMPUTED TOMOGRAPHY

Patients undergoing recurrent CT exams: assessment of patients

with non-malignant diseases, reasons for imaging
and imaging appropriateness
Madan M. Rehani 1 & Emily R. Melick 1 & Raza M. Alvi 1 & Ruhani Doda Khera 1 & Salma Batool-Anwar 2 &
Tomas G. Neilan 1 & Michael Bettmann 3

Received: 30 July 2019 / Revised: 15 October 2019 / Accepted: 25 October 2019 / Published online: 2 December 2019
# European Society of Radiology 2019

Abstract
Objective To determine percent of patients without malignancy and ≤ 40 years of age with high cumulative radiation doses
through recurrent CT exams and assess imaging appropriateness.
Methods From the cohort of patients who received cumulative effective dose (CED) of ≥ 100 mSv over a 5-year period, a sub-set
was identified with non-malignant disease. The top 50 clinical indications leading to multiple CTs were determined. Clinical
decision support (CDS) system scores were analyzed using a widely adopted standard of 1–3 (red) as “not usually appropriate,”
4–6 (yellow) “may or may not be appropriate,” and 7–9 (green) “usually appropriate.” Clinicians reviewed patient records to
assess compliance with appropriate use criteria (AUC).
Results 9.6% of patients in our series were with non-malignant conditions and 1.4% with age ≤ 40 years. CDS scores
(rounded) were 2% red, 38% yellow, 27% green, and 33% unscored CTs. Clinical society guidelines for CT exams,
wherever available, were followed in 87.5 to 100% of cases. AUCs were not available for several clinical indications
as also referral guidelines for serial CT imaging. More than half of CT exams were unrelated to follow-up of a
primary chronic disease.
Conclusions We are faced with a situation wherein patients in age ≤ 40 years require or are thought to require many CT exams
over the course of a few years but the radiation risk creates concern. There is a fair number of conditions for which AUC are not
available. Suggested solutions include development of CT scanners with lesser radiation dose and further development of
appropriateness criteria.
Key Points
& We are faced with a situation wherein patients in age group 0–40 years and with non-malignant diagnosis require or are
thought to require many CT exams over the course of a few years.
& More than half of CT exams were unrelated to follow-up of a primary chronic disease.
& Imaging guidelines and appropriateness use criteria are not available for many conditions. Wherever available, they are for
initial work-up and diagnosis and there is a lack of guidance on serial CT imaging.

Keywords Radiology . Referral and consultation . Radiation protection . Patient safety . Radiologic technology

Abbreviations
ACC American College of Cardiology
* Madan M. Rehani ACCM American College of Critical Care Medicine
madan.rehani@gmail.com; mrehani@mgh.harvard.edu ACR American College of Radiology
ADIS Advanced diagnostic imaging services
1
Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, AUC Appropriateness use criteria
USA CDS Clinical decision support
2
Brigham Women’s Hospital, Boston, MA 02115, USA CED Cumulative effective dose
3 CMS Centers for Medicare & Medicaid Services
Wake Forest University School of Medicine,
Winston-Salem, NC 27101, USA COPD Chronic obstructive pulmonary disease
1840
CT Computed tomography
ED Effective dose
IDSA Infectious Disease Society of America
IRB Institutional Review Board
MRI Magnetic resonance imaging
NCCN National Comprehensive Cancer Network
PAMA Protecting Access to Medicare Act
PET Positron emission tomography
RPDR Research Patient Data Registry
SNMMI Society of Nuclear Medicine and
Molecular Imaging
Introduction
Radiologists and clinicians assess imaging appropriate-
ness when they make choices about diagnostic imaging
in routine clinical care [1]. The professional societies
have created appropriateness use criteria (AUC), defined
as clinical imaging guidelines using sound, reproducible,
widely accepted methodology, and high-quality evidence
to as great an extent as possible [2–4]. In the USA, the
AUCs are widely accepted and form the basis for
Centers for Medicare & Medicaid Services (CMS) di-
rectives regarding the Protecting Access to Medicare
Act (PAMA). This act requires referring providers to
consult AUC prior to ordering advanced diagnostic im-
aging services (ADIS) such as computed tomography
(CT), magnetic resonance imaging (MRI), and nuclear
medicine (including positron emission tomography
(PET)) for certain clinical conditions in Medicare pa-
tients [5], with educational and operation testing period
from January 2020.
In a recent study, using the data provided by the dose track-
ing platform Radimetrics (Ver 2.6 Bayer HealthCare), the au-
thors identified a group of 8952 patients who underwent a
median of 19, mean of 21, and a maximum of 109 CTs during
the 5-year period resulting in cumulative effective doses
(CED) ranging from 100 to 1185 mSv [6]. There is reasonable
degree of consensus among official international and national
organizations [7–10] on radiation risks at this level of radia-
tion dose.
This created an urgent need to study if there are sizeable
number of patients in relatively younger age group of ≤ 40
years with non-malignant conditions with potential for linger
life expectancy and study imaging appropriateness in this po-
tentially vulnerable group. This retrospective study was un-
dertaken at a single hospital to determine how often the CT
exams met accepted criteria for imaging. To the best of our
knowledge, neither identification of such at-risk group nor
focused assessment of imaging appropriateness in such a
group has been previously reported.
Eur Radiol (2020) 30:1839–1846
Methods
As this is a retrospective study using no intervention on the
patient cohort, the Institutional Review Board (IRB) waived
the requirement to obtain informed consent. As a matter of
nomenclature, a single visit for a CT study (one accession
number) was counted as single CT “exam” and thus if multi-
ple phases (CT scans) were performed during the study visit, it
was still counted as one CT exam.
Stratification of study group
The authors first obtained all diagnoses listed in the electronic
medical records (Epic) of the 8,952 patients with CED ≥
100 mSv with the help of the Research Patient Data Registry
(RPDR) through a query system. Using keyword filtering of
the diagnosis lists, patients were identified as having malig-
nant or non-malignant condition(s). Keywords from the diag-
nosis field used to define a patient as having malignant disease
include the following: “chemotherapy,” “antineoplastic,” “ra-
diotherapy,” “malignant,” “malignancy,” “malignancies,”
“carcinoma,” “leukemia,” “lymphoma,” “sarcoma,” “metasta-
sis,” “metastatic,” and “cancer.” If a patient’s diagnosis list did
not include any of these keywords, the patient was categorized
as having no known malignant condition(s). If a patient had a
diagnosis of “neoplasm” that included the term “benign,” “un-
certain,” or “unspecified” and no malignant term, this field
alone would not define the patient as having a malignant
condition.
The top 50 most frequently occurring conditions in the co-
hort of patients without malignancies and age ≤ 40 years were
identified for further analysis of imaging appropriateness.
Assessment of imaging appropriateness
The American College of Radiology (ACR) has created
CareSelect, a comprehensive, national standards-based clini-
cal decision support (CDS) database comprising over 3,000
clinical scenarios and 15,000 criteria [11]. Our hospital has
adopted the ACR system, which incorporates not only ACR
Appropriateness Criteria® [2] but also criteria from the
American College of Cardiology (ACC) [12], the National
Comprehensive Cancer Network (NCCN) [13], and the
Society of Nuclear Medicine and Molecular Imaging
(SNMMI) [14]. This study being focused on CT exams, the
ACR criteria® are of primary concern.
Imaging appropriateness was studied by the following
methods, collectively aimed at assessing both the availability
(or non-availability) and relevance of AUC.
(a) Identification of clinical conditions in patient cohort of
age ≤ 40 years with recurrent CTs for which AUC are not
available from ACR.
Eur Radiol (2020) 30:1839–1846 1841

(b) Analysis of appropriateness scores for CTs, as provided Table 2 Top 50 most frequently occurring clinical reasons for CT
imaging in patients with non-malignant diagnoses of age ≤ 40 years
by CDS. This includes all AUCs inbuilt into the CDS and
covers all CT exams the sub-set of patients underwent. Clinical condition Number of times
(c) Manual review of patient medical records by: this condition
occurred among
this group
(i) the cardiovascular physician who analyzed only
chest CTs and CTAs for appropriateness based on Other non-specific abnormal finding of lung field 353
guidelines from the AHA/ACC [15] and supplement- Fever, unspecified 247
Encounter for fitting and adjustment of vascular 203
ed by clinical judgement.
catheter
(ii) the pulmonologist who chose to analyze appropriate- Abdominal pain, unspecified site 147
ness of chest CTs in patients with the following Cough 147
known or suspected conditions: pneumonia, other in- Shortness of breath 141
Tachycardia, unspecified 135
filtrate, interstitial lung disease, pleural effusion, and Cardiac arrhythmia, unspecified 133
chronic obstructive pulmonary disease (COPD). The Other ascites 131
AUC, if available from corresponding clinical socie- Unspecified abdominal pain 131
Solitary pulmonary nodule 130
ty, are mentioned in the results and discussion part. Encounter for fitting and adjustment of non-vascular 129
catheter NEC
This provided assessment based on AUCs of clinical soci- Pain in limb 127
Headache 125
eties in addition to ACR as well as review by clinicians. Unspecified pleural effusion 123
Encounter for adjustment and management of vascular 119
access device
Other dyspnea and respiratory abnormality 119
Atelectasis 118
Results Pleural effusion, not elsewhere classified 116
Anemia, unspecified 115
Table 1 presents age distribution for the 8,952 patients, broken Enlargement of lymph nodes 115
down into malignant and non-malignant diagnosis sub-groups. Pulmonary collapse 115
Observation for suspected cardiovascular disease 114
Nearly 90% of patients (n = 8,091; 90.4%) had malignant di- Other specified soft tissue disorders 108
agnoses and only 10% (n = 861; 9.6%) had non-malignant Chest pain, unspecified 106
diagnoses. The current study focused on relatively young pa- Nausea with vomiting 98
Pneumonia, organism unspecified 97
tients (≤ 40 years) with non-malignant diagnoses (123 patients, Other specified preoperative examination 94
14.3% of all patients without malignancy (861) and 1.4% of all Routine general medical examination at a health care 94
the total of 8952 patients). These patients had mean age of 31.9 facility
Swelling of limb 94
years and median of 32.7 years with range of 18–40 years. Observation for other specified suspected conditions 89
Table 2 provides the top 50 most frequently occurring clin- Dizziness and giddiness 81
ical reasons that led to multiple CTs in 123 patients of age ≤ 40 Cervicalgia 79
Observation for unspecified suspected mental 79
condition
Table 1 Age distribution for patients with malignant and non- Other diseases of lung, not elsewhere classified 77
malignant conditions Backache, unspecified 76
Abdominal pain, other specified site; multiple sites 75
Age Malignant % of Non-malignant % of Encounter for other preprocedural examination 73
range count total count total Encounter for general adult medical examination 72
without abnormal findings
≤ 20 3 0.0 4 0.0 Other malaise and fatigue 72
21–30 119 1.3 50 0.6 Preoperative cardiovascular examination 72
Pulmonary congestion and hypostasis 72
31–40 298 3.3 69 0.8 Unspecified disorder of immune mechanism 72
41–50 630 7.0 102 1.1 Hypotension, unspecified 71
51–60 1542 17.2 155 1.7 Other long term (current) drug therapy 70
Respiratory failure 70
61–70 2412 26.9 194 2.2 Encounter for other specified aftercare 69
71–80 2134 23.8 166 1.9 Urinary tract infection, site not specified 69
81–90 836 9.3 98 1.1 Essential (primary) hypertension 68
Hypoxemia 68
≥ 91 117 1.3 23 0.3
Total 8091 90.4 861 9.6
Overall 8952
total years with only non-malignant diagnoses. It is clear that a
wide variety of clinical reasons led to multiple CTs in these
1842
patients. Most patients had multiple reasons for imaging, as their
clinical conditions changed over time and the change does not
allow a denominator for estimating percent value of frequency.
Imaging appropriateness
Table 3 provides a list of 14 conditions not specifically cov-
ered by current ACR Appropriateness Criteria at the time of
writing this manuscript.
Imaging appropriateness scores
Our hospital had in-house CDS system starting 2004, but cur-
rently, imaging appropriateness scores of the CDS system are
built into Epic at our hospital [16]. Epic was launched at our
hospital on April 2, 2016 and CDS data in current system is
not retrievable prior to April 2016. Therefore, CDS analysis in
this study is based on information available from April 2016
onwards, providing imaging appropriateness scores for 203
out of 305 CTs in 44 patients as available in Epic.
The CDS system provides both a quantitative and color
classification of appropriateness:
& Score 1–3 = Red (low utility, “Usually not appropriate”)
& Score 4–6 = Yellow (marginal, “May or may not be
appropriate”)
& Score 7–9 = Green (indicated, “Usually appropriate”)
The system also has an “Acceptable” category that pertains
to unscored indications for which criteria are not available.
Research-specific scans go unscored and pass CDS. To avoid
confusion, we have labeled them in Table 4 as unscored.
Table 4 shows that for all CT exams combined, 2% were red
Table 3 Clinical
conditions for which Follow-up for pharyngocutaneous fistula
appropriateness criteria Encounter for adjustment and
is not available from the management of vascular access device
ACR
Atelectasis
Enlargement of lymph nodes
Anemia, unspecified
Other long term (current) drug therapy
Encounter for general adult medical
examination without abnormal findings
Encounter for immunization
Hypoxemia
Other specified counseling
Observation for unspecified suspected
mental condition
Other malaise and fatigue
Pulmonary congestion and hypostasis
Unspecified disorder of immune
mechanism
Eur Radiol (2020) 30:1839–1846
Table 4 Analysis of imaging appropriateness scores for CT exams of 44
patients for which information was available from the CDS system
Number of Red Yellow Green Unscored
CTs (%) (1–3) (%) (4–6) (%) (7–9) (%) (%)
Chest 81 (26.6) 1 (1.2) 45 (55.6) 13 (16.0) 22 (27.1)
Others 224 (73.4) 5 (2.2) 70 (31.3) 69 (30.8) 80 (35.7)
Total 305 (100) 6 (2.0) 115 (37.7) 82 (26.9) 102 (33.4)
(low utility), 38% were yellow (marginal), 27% were green
(indicated), and 33% were unscored.
Analysis of imaging appropriateness by clinical
specialists
The cardiovascular physician randomly selected 30 patients of
≤ 40 years of age. These patients had a total of 535 CTs of
which 112 (21%) were chest CTs, including 59 (53%) CT
angiograms and CT pulmonary angiograms. Twenty-three
chest CTs were performed outside the study institution’s
healthcare system and were not included in the clinical anal-
ysis. In the 89 CT exams analyzed, 26% were performed to
rule out pulmonary embolism. Other indications included
chest trauma to rule out vascular damage (12%), suspicion
for aortic dissection, and follow-up after aortic aneurysm re-
pair (20%). The remaining chest CT indications included pul-
monary empyema (n = 5), pharyngocutaneous fistula follow-
up (n = 10), Takayasu’s arteritis (n = 2), necrotizing pancrea-
titis (n = 2), follow-up for mycotic aneurysm (n = 2),
suspected LVAD thrombosis (n = 2), and to rule out CAD in
symptomatic young patients (n = 2). Of the 89 CTs, 62 (69%)
had guidelines available from clinical societies like ACC [15].
These guidelines were followed in 100% of the cases. For
example, to evaluate the appropriateness of the CTs performed
to rule out pulmonary embolism, we used the Well’s criteria.
All of the CTs were done on patients with Well’s score ≥ 3
with at least moderate risk of pulmonary embolism. All pa-
tients who had CTs done for follow-up for or suspected aortic
disease such as aortic dissection, aortic aneurysm, mediastinal
hematoma, or post-operative surveillance were appropriate as
per ESC and ACCF/AHA guidelines [15 17]. All patients who
had CT done to rule out dissection presented with high-risk
chest pain features (sudden onset of sharp chest pain radiating
to the back), and neither ECG, CXR, nor history pointed to-
wards alternative diagnoses. Among patients post-TEVAR,
CT is recommended as the first-choice imaging for follow-
up after 1 month, 6 months, 12 months, and then annually
[17]. These recommendations were followed in this cohort.
Definite guidelines for the remaining 31% of chest CTs are
lacking. These indications were varied and included follow-up
s e r i a l C Ts f o r e m p y e m a ( n = 5 ) , f o l l o w - u p f o r
pharyngocutaneous fistula (n = 10), evaluation for necrotizing
Eur Radiol (2020) 30:1839–1846
pancreatitis (n = 2), evaluation for hematemesis (n = 1), follow-
up for mycotic aneurysm (n = 2), evaluation for septic pulmo-
nary emboli (n = 5), rule-out chest source of fever (n = 1), and
follow-up after ventriculo-thoracic shunt placement (n = 1).
An assessment was also made on the number of CTs per-
formed for follow-up on a primary condition. It is of interest to
note that out of the 112 CT chest exams, 48 CTs (nearly 43%)
were follow-up CTs for primary conditions such as (1) empy-
ema, (2) pharyngocutaneous fistula, (3) right subclavian artery
dissection, (4) pseudo aneurysm after TEVAR, (5) aortic dis-
section, (6) mycotic aortic aneurysm after repair, and (7) septic
pulmonary embolism. The remaining 64 CTs (57%) evaluated
conditions subsequent to the primary clinical diagnosis: acute
pulmonary embolism, acute aortic dissection, coronary artery
disease, vascular damage after trauma, and rule-out LVAD
thrombosis.
The analysis by the pulmonologist used the inclusion
criteria mentioned above and further excluded cases of
pulmonary embolism, aortic dissection, and those who
underwent aortic dissection repair surgery and had
follow-up imaging. Similarly, trauma victims imaged to
assess the extent of injury (60 CTs) were excluded as
adherence to imaging guidelines may be subjective. This
resulted in 24 patients and 95 chest CTs. The distribu-
tion of CTs in different conditions was eleven patients
with history of intravenous drug use (IVDU) or another
immunocompromised condition accounted for 36 chest
CTs (an average of 3.3 scans per patient) and seven
patients with either sepsis or were critically ill
accounted for 28 chest CTs (an average of four scans
per patient). Three patients with empyema had a total of
15 chest CTs (an average of five exams per patient).
One patient had quadriplegia and had nine chest CTs
performed, one patient with HIV and cardiomyopathy
had two chest CT exams performed, and one patient
with encephalitis had five chest CTs performed.
Twenty-one of 24 patients in this cohort (87.5%) had an
appropriate CT performed for known or suspected pulmo-
nary infiltrates; all of these patients were septic or immune-
compromised. The remaining three patients (12.5%) had a
total of 15 CT exams for pleural effusion and/or after tube
drainage.
Discussion
There are a number of new findings in this study including
identification of how many patients with cumulative radiation
dose of ≥ 100 mSv have non-malignant conditions (around
10%) and are in a relatively younger age of ≤ 40 years (nearly
14% of these 10% or 1.4% of total patients with CED ≥ 100
mSv); determination of the top 50 most frequently occurring
clinical reasons that led to multiple CT exams of all regions of
1843
the body; identification of the clinical conditions from the top
50 for which AUC are not currently available from ACR;
assessment of imaging appropriateness using CDS scores
(only 2% red, but 38% yellow); assessment of imaging appro-
priateness for chest CTs using the guidelines currently avail-
able from clinical professional societies (showing 87.5 to
100% compliance); assessment of the fraction of CT exams
of chest (31%, that is, nearly one-third) for which guidelines
are not available from clinical professional societies; the sub-
jective assessment of imaging appropriateness by the clini-
cians that indicate almost all CT exams were appropriate;
and finally, realization that nearly half (precisely 43% in our
group) of the serial CT exams were for reasons other than the
follow-up of main disease condition.
This study is not aimed at assessing cancer risks from multi-
ple CT exams the patients undergo. We defined and identified
the at-risk population. Earlier studies have not identified the size
and profile of this population. The focus of past studies has been
on assessing practice in terms of overall magnitude of CT usage,
trends in usage, and assessment of radiation doses and risks
without analyzing the appropriateness of CT exams [18–23].
Indication-based appropriateness studies have been report-
ed in some specific clinical conditions such as CT
enterography [24], non-traumatic abdominal emergencies for
women [25], emergency department CT in renal colic patients
[26], and suspected pulmonary embolism [27]. These studies
have not assessed appropriateness of serial CT imaging in
individual patients with varying clinical conditions. Our re-
sults show that nearly half of CT exams were unrelated to
follow-up of a primary chronic disease.
CDS utilizing valid clinical imaging (and other) guidelines
has long been advocated as a means to encourage evidence-
based medical practice. The study institution has used CDS sys-
tem for over a decade with positive outcomes reported [16,
28–30].
It is well known that despite availability of integrated
CDS software, a number of advanced imaging requests
are for uncertain or inappropriate indications, and a sig-
nificant percentage may not be able to be scored by
CDS [31]. Due to limited guidelines and clinical varia-
tions, CDS cannot be expected to always provide an-
swers, and that is the case in this situation. A detailed
description of the CDS system is outside the scope of
this paper, but information is available in a number of
other publications [16, 28–33]. Manual review by the
clinical specialists is required and was done in our
study for a selected group of patients. As noted, AUC
are not able to cover all clinical situations. In this study,
14 (Table 3) out of 50 (Table 2) clinical reasons (i.e.,
28%) and 33.4% of the CTs for which CDS data was
available (unscored in Table 4) fell under this category
at the time of analysis of our data in later part of 2018.
AUC keep getting updated continuously and this study
1844
provides this information for developers of appropriate-
ness criteria.
There are comparable non-invasive modalities that one
might consider for confirming or ruling out PE in patients
who are at least at moderate risk for PE, particularly for those
who have had multiple CTs. For example, an IRM-EP study
demonstrated that MRI has a sensitivity and specificity for
diagnosing PE of 79–85% and 99–100%, respectively [34].
Review by a pulmonologist confirms that according to
American College of Critical care Medicine (ACCM) and
Infectious Disease Society of America (IDSA) guideline
(2008), chest x-ray or CT chest should be obtained when clin-
ically indicated, particularly among immunocompromised pa-
tients (level 1) [35]. Similarly, the American Association for
Thoracic surgery consensus guideline (2010) [36] recom-
mends imaging in all patients with signs and symptoms of
pneumonia or unexplained sepsis (class 1, LOEB) [37]. For
pleural effusion, plain chest radiographs that impart very low
radiation dose are usually the initial imaging modality. The
radiographic features of pleural effusion are characteristic
and may be evident with as little as 50 ml of pleural fluid.
However, in the intensive care setting, the radiographs are
usually done on patients in a supine position, which often
underestimates the volume of pleural fluid present. Pleural
ultrasound is more accurate than the plain chest radiograph
in these cases for estimating the volume of pleural fluid and
also assists in thoracentesis. The pleural ultrasound is there-
fore recommended in addition to conventional chest x-ray
(class 1, LOEB). Pleural ultrasound has been found to be very
effective in identifying small effusions, and bedside availabil-
ity makes it an ideal tool [38]. It is also suggested that ultra-
sound detects septations within the pleural fluid with greater
sensitivity than CT [39].
In cases of diagnostic difficulty, contrast-enhanced CT may
help. The British Thoracic Society guidelines (2010) recom-
mend that CTs should be performed in the investigation of all
undiagnosed exudative pleural effusions (level of evidence C)
and should be requested for complicated effusions when initial
drainage has been unsuccessful (level C) [40]. While these
guidelines are available for initial work-up and diagnosis, there
is a lack of guidance on serial imaging. CT is also particularly
helpful in distinguishing empyema from lung abscess [41, 42].
Limitations
There are some limitations of the current study. First,
the study institution moved from an in-house to a com-
mercial CDS in April 2016. Thus, the CDS scores of
patients with CTs prior to April 2016 were not avail-
able. Further, the CDS system currently at the study
institution is not the same as ACR CareSelect as there
are many unique features which occurred during transi-
tion from in-house to ACR-based system. These
Eur Radiol (2020) 30:1839–1846
occurred as two in-house CDS systems; one of the
study institution and another of its sister hospitals were
combined. When both were merged, there was a need to
merge a large number of clinical indications that were
used in one but not in the other. Thus, the results of
this CDS system may differ from others where ACR
CareSelect is used. These limitations led to a small
sample size for assessment of CDS score and appropri-
ateness. Another limitation includes super specialization
at the study institution. Patients had a variety of clinical
conditions for which CTs performed require several
physicians to review these exams for research as in this
study but most physicians at the study institution have a
highly specialized area of expertise. It becomes difficult
to collect a group of several specialist clinicians to look
at CT scans of a single region of the body like chest
and determine appropriateness. The study needs to be
extended to a larger group to include also abdominal
imaging and corresponding experts. It must also be re-
membered that results of this study cannot be extrapo-
lated to other settings as most hospitals do not have
CDS system in place. Our hospital has almost 15 years
of experience with CDS and therefore a red rate of only
2% for all CT imaging in the study group and 1.2% for
chest CT can be understandable. Institutions without
CDS can be expected to have manifold higher inappro-
priate imaging.
Future studies
Our study creates compelling needs for research, develop-
ment, and actions by various stakeholders to make patients
safer. First, it should be possible to conduct needed CT exams
with radiation doses that do not lead patients into a zone of
known radiation risk. Imaging industry needs to meet this
challenge through research and development. Second, clinical
professionals need to develop and implement imaging appro-
priateness criteria; otherwise, a fairly large number of patients
fall into at-risk zone as per magnitude assessed in our earlier
papers [6, 43]. Third, current radiation risk management pol-
icies for patients need to be re-evaluated as they have not been
as effective as other areas like flying, travel by car, or even
occupational safety, especially as benefit-risk ratio is becom-
ing questionable. National and international organizations
need to realize that their policies are not able to stop this
situation from happening and need to review policies.
Fourth, radiation epidemiologists and radiation biologists
need to provide better evidence of radiation risks at CED ≥
100 mSv [44]. Finally, there is a need for assessment of im-
aging appropriateness for high cumulative radiation dose co-
hort at institutions without at CDS system in place that will
reflect the true situation of need for introducing actions for
improving appropriateness in imaging.
Eur Radiol (2020) 30:1839–1846
Conclusions
Our study shows that a sizeable fraction of patients who
receive high cumulative doses of 100 mSv+ are those
with non-malignant condition, and further, those who
fall in relatively younger age group of ≤ 40 years are
not an ignorable fraction. Non-availability of AUC for
many clinical conditions is of concern. Our results come
from an institution where per CT exam doses are much
lower than US national figures [6] and where CDS has
been in use for nearly 15 years. The institutions where
CDS is not in use and where optimization in CT dose is
not as good as at our institution are likely to have much
more magnitude of at-risk patients. In addition, there
were circumstances identified where MRI or ultrasound
could have been used instead of CT. It also appears that
while appropriateness criteria are available for initial
work-up and diagnosis, there is a lack of guidance on
serial imaging. There is also a strong need for industry
to develop CT equipment that delivers lesser radiation
doses (sub-mSv) for clinically indicated CT exams. We
should envisage a future where radiation risk should not
be an issue for patients who justifiably need recurrent
CT exams.
Acknowledgements The authors sincerely acknowledge the subject mat-
ter expertise provided by Prof. Fred Mettler, Emeritus Professor,
University of New Mexico and of Dr. Bob Liu, MGH for his help in
dealing with IRB.
Funding information The authors state that this work has not received
any funding.
Compliance with ethical standards
Guarantor The scientific guarantor of this publication is Madan Rehani.
Conflict of interest The authors of this manuscript declare no relation-
ships with any companies whose products or services may be related to
the subject matter of the article.
Statistics and biometry No statistical analysis expertise was availed
besides authors'.
Informed consent Written informed consent was waived by the
Institutional Review Board.
Ethical approval Institutional Review Board waived the requirement to
obtain informed consent
Study subjects or cohorts overlap None.
This paper is continuation of the study reported in paper Rehani et al
DOI: https://doi.org/10.1007/s00330-019-06523-y (reference 6). That
paper had only radiation dose information and no medical /clinical infor-
mation. The current paper analyses the clinical information/ diagnoses of
patients and assesses imaging appropriateness.
1845
Methodology
• Retrospective
• Observational
• Performed at one institution
References
1. Sistrom CL (2009) The appropriateness of imaging: a comprehen-
sive conceptual framework. Radiology 251(3):637–649
2. American College of Radiology (ACR) (2018) ACR
Appropriateness Criteria [Internet] [updated 2019; cited 2019
July 29]. https://www.acr.org/Clinical-Resources/ACR-
Appropriateness-Criteria. Accessed 7 Nov 2019
3. The Royal College of Radiologist. iRefer: radiological investigation
guidelines tool, RCR, London https://www.rcr.ac.uk/clinical-
radiology/being-consultant/rcr-referral-guidelines/about-irefer.
Accessed 7 Nov 2019
4. European Society of Radiology. Referral guidelines for imaging.
https://www.myesr.org/referral-guidelines-imaging. Accessed 7
Nov 2019
5. Centers for Medicare & Medicaid Services (CMS), Department of
Health and Human Services (HHS) (2016) Appropriate Use Criteria
Program. https://www.cms.gov/Medicare/Quality-Initiatives-
Patient-Assessment-Instruments/Appropriate-Use-Criteria-
Program/index.html. Accessed 7 Nov 2019
6. Rehani MM, Yang K, Melick ER, Heil J, Šalát D, Sensakovic WF, Liu
W (2019) Patients undergoing recurrent CT scans: assessing the mag-
nitude. Eur Radiol. https://doi.org/10.1007/s00330-019-06523-y
7. (2007) The 2007 recommendations of the International
Commission on Radiological Protection. ICRP Publication 103.
Ann ICRP 37(2-4):1–332
8. United Nations Scientific Committee on the Effects of Atomic
Radiation (2018) Sources, effects and risks of ionizing radiation.
UNSCEAR 2017 Report [Internet]. New York, NY: UNSCEAR].
Available from: http://www.unscear.org/docs/publications/2017/
UNSCEAR_2017_Annex-B.pdf. Accessed 7 Nov 2019
9. National Council of Radiation Protection and Measurements (2018)
Implications of recent epidemiologic studies for the linear-non
threshold model and radiation protection. NCRP Commentary
No. 27 [Internet]. Bethesda, Maryland: NCRP; https://ncrponline.
org/shop/commentaries/commentary-no-27-implications-of-recent-
epidemiologic-studies-for-the-linear-nonthreshold-model-and-
radiation-protection-2018/. Accessed 7 Nov 2019
10. National Research Council (2006) Health risks from exposure to
low levels of ionizing radiation: BEIR VII Phase 2 [Internet]. The
National Academies Press, Washington. https://doi.org/10.17226/
11340
11. ACR Select – National Decision Support Company. CareSelect
imaging [Internet]. [updated 2019; cited 2019 July 29]. Available
from: http://nationaldecisionsupport.com/acrselect/. Accessed 7
Nov 2019
12. Hendel RC, Lindsay BD, Allen JM et al (2018) ACC Appropriate
Use Criteria methodology: 2018 Update: A Report of the American
College of Cardiology Appropriate Use Criteria Task Force. J Am
Coll Cardiol 71(8):935–948. https://doi.org/10.1016/j.jacc.2018.
01.007
13. National Comprehensive Cancer Network (NCCN). NCCN
Imaging Appropriate Use Criteria (NCCN Imaging AUCTM)
https://www.nccn.org/professionals/imaging/default.aspx.
Accessed 7 Nov 2019
14. Society of Nuclear Medicine and Molecular Imaging (SNMMI)
(2013) SNMMI Appropriate Use Criteria (AUC) http://www.
snmmi.org/auc
1846
15. Hiratzka LF, Bakris GL, Beckman JA, et al (2010) 2010
ACCF/AHA/AATS/ACR/ASA/SCA/SCAI/SIR/STS/SVM guide-
lines for the diagnosis and management of patients with thoracic
aortic disease: executive summary. A report of the American
College of Cardiology Foundation/American Heart Association
Task Force on Practice Guidelines, American Association for
Thoracic Surgery, American College of Radiology, American
Stroke Association, Society of Cardiovascular Anesthesiologists,
Society for Cardiovascular Angiography and Interventions,
Society of Interventional Radiology, Society of Thoracic
Surgeons, and Society for Vascular Medicine. Catheter Cardiovas
Interv 76:E43-86.
16. Rosenthal DI, Weilburg JB, Schultz T et al (2006) Radiology order
entry with decision support: Initial clinical experience. J Am Coll
Radiol 3(10):799–806
17. Erbel R, Aboyans V, Boileau C et al (2014) 2014 ESC Guidelines on
the diagnosis and treatment of aortic diseases: document covering
acute and chronic aortic diseases of the thoracic and abdominal aorta
of the adult. The Task Force for the Diagnosis and Treatment of
Aortic Diseases of the European Society of Cardiology (ESC). Eur
Heart J 35(41):2873–2926. https://doi.org/10.1093/eurheartj/ehu281
18. Sodickson A, Baeyens PF, Andriole KP et al (2009) Recurrent CT,
cumulative radiation exposure, and associated radiation-induced
cancer risks from CT of adults. Radiology 251(1):175–184.
https://doi.org/10.1148/radiol.2511081296
19. Griffey RT, Sodickson A (2009) Cumulative radiation exposure and
cancer risk estimates in emergency department patients undergoing
repeat or multiple CT. AJR Am J Roentgenol 192(4):887–892.
https://doi.org/10.2214/AJR.08.1351
20. Desmond AN, O'Regan K, Curran C et al (2008) Crohns disease:
factors associated with exposure to high levels of diagnostic radia-
tion. Gut 57(11):1524–1529. https://doi.org/10.1136/gut.2008.
151415
21. Smith-Bindman R, Wang Y, Yellen-Nelson TR et al (2017)
Predictors of CT radiation dose and their effect on patient care: a
comprehensive analysis using automated data. Radiology 282(1):
182–193. https://doi.org/10.1148/radiol.2016151391
22. Smith-Bindman R, Moghadassi M, Wilson N et al (2015) Radiation
doses in consecutive CT examinations from five university of
California medical centers. Radiology. 277(1):134–141. https://
doi.org/10.1148/radiol.2015142728
23. Brenner DJ, Elliston CD, Hall EJ, Berdon WE (2001) Estimates of
the cancer risks from pediatric CT radiation are not merely theoret-
ical: comment on “Point/Counterpoint: In x-ray computed tomog-
raphy, technique factors should be selected appropriate to patient
size. Against the Proposition”. Med Phys 28(11):2387–2388
24. Guimarães LS, Fidler JL, Fletcher JG et al (2010) Assessment of ap-
propriateness of indications for CT enterography in younger patients.
Inflamm Bowel Dis 16(2):226–232. https://doi.org/10.1002/ibd.21025
25. Giannitto C, Campoleoni M, Maccagnoni S et al (2017)
Unindicated multiphase CT scans in non-traumatic abdominal
emergencies for women of reproductive age: a significant source
of unnecessary exposure. Radiol Med 123(3):185–190. https://doi.
org/10.1007/s11547-017-0819-6
26. Raja AS, Pourjabbar S, Ip IK et al (2018) Impact of a health infor-
mation technology–enabled appropriate use criterion on utilization
of emergency department CT for renal colic. AJR Am J Roentgenol
212(1):142–145. https://doi.org/10.2214/AJR.18.19966
27. Mills AM, Ip IK, Langlotz CP, Raja AS, Zafar HM, Khorasani R
(2018) Clinical decision support increases diagnostic yield of computed
tomography for suspected pulmonary embolism. Am J Emerg Med
36(4):540–544. https://doi.org/10.1016/j.ajem.2017.09.004
28. Weilburg JB, Sistrom CL, Rosenthal DI et al (2017) Utilization
management of high-cost imaging in an outpatient setting in a large
Eur Radiol (2020) 30:1839–1846
stable patient and provider cohort over 7 years. Radiology 284(3):
766–776. https://doi.org/10.1148/radiol.2017160968
29. Sistrom CL, Dreyer KJ, Dang PP et al (2009) Recommendations for
additional imaging in radiology reports: multifactorial analysis of
5.9 million examinations. Radiology. 253(2):453–461. https://doi.
org/10.1148/radiol.2532090200
30. Vartanians VM, Sistrom CL, Weilburg JB, Rosenthal DI, Thrall JH
(2010) Increasing the appropriateness of outpatient imaging: effects
of a barrier to ordering low-yield examinations. Radiology 255(3):
842–849. https://doi.org/10.1148/radiol.10091228
31. Moriarity AK, Klochko C, Obrien M, Halabi S (2015) The effect of
clinical decision support for advanced inpatient imaging. J Am Coll
Radiol 12(4):358–363. https://doi.org/10.1016/j.jacr.2014.11.013
32. Palen TE, Sharpe RE Jr, Shetterly SM, Steiner JF (2019)
Randomized clinical trial of a clinical decision support tool for
improving the appropriateness scores for ordering imaging studies
in primary and specialty care ambulatory clinics. AJR Am J
Roentgenol 16:1–6. https://doi.org/10.2214/AJR.19.21511
33. Schneider E, Zelenka S, Grooff P, Alexa D, Bullen J, Obuchowski
NA (2015) Radiology order decision support: examination-
indication appropriateness assessed using 2 electronic systems. J
Am Coll Radiol 12(4):349–357. https://doi.org/10.1016/j.jacr.
2014.12.005
34. Revel MP, Sanchez O, Couchon S et al (2012) Diagnostic accuracy
of magnetic resonance imaging for an acute pulmonary embolism:
results of the ‘IRM-EP’ study. J Thromb Haemost 10(5):743–750.
https://doi.org/10.1111/j.1538-7836.2012.04652.x
35. O’Grady NP, Barie PS, Bartlett JG et al (2008) Guidelines for eval-
uation of new fever in critically ill adult patients: 2008 update from
the American College of Critical Care Medicine and the Infectious
Diseases Society of America. Crit Care Med 36(4):1330–1349.
https://doi.org/10.1097/CCM.0b013e318169eda9
36. Shen KR, Bribriesco A, Crabtree T et al (2017) The American
Association for Thoracic Surgery consensus guidelines for the man-
agement of empyema. J Thorac Cardiovasc Surg 153(6):e129–
e146. https://doi.org/10.1016/j.jtcvs.2017.01.030
37. Ebell MH, Siwek J, Weiss BD et al (2004) Strength of recommenda-
tion taxonomy (SORT): a patient-centered approach to grading evi-
dence in the medical literature. Am Fam Physician 69(3):548–556
38. Kocijancic I, Vidmar K, Ivanovi-Herceg Z (2003) Chest sonogra-
phy versus lateral decubitus radiography in the diagnosis of small
pleural effusions. J Clin Ultrasound 31(2):69–74
39. Kearney S, Davies C, Davies R, Gleeson F (2000) Computed to-
mography and ultrasound in parapneumonic effusions and empy-
ema. Clin Radiol 55(7):542–547
40. Maskell NA, Butland RJ (2003) BTS guidelines for the investiga-
tion of a unilateral pleural effusion in adults. Thorax 58(90002):8ii-
17.
41. Mcloud TC, Flower CD (1991) Imaging the pleura: sonography,
CT, and MR imaging. AJR Am J Roentgenol 156(6):1145–1153
42. Waite RJ, Carbonneau RJ, Balikian JP, Umali CB, Pezzella AT,
Nash G (1990) Parietal pleural changes in empyema: appearances
at CT. Radiology 175(1):145–150
43. Brambilla M, Vassileva J, Kuchcinskac A, Rehani MM (2019)
Multinational data on cumulative radiation exposure of patients
from recurrent radiological procedures: call for action. Eur Radiol
https://doi.org/10.1007/s00330-019-06528-7
44. Rehani MM (2019) Looking for solutions: vision and a call-for-
attention for research scientists. Int J Radiat Biol 17:1–4. https://
doi.org/10.1080/09553002.2019.1569775
Publisher’s note Springer Nature remains neutral with regard to jurisdic-
tional claims in published maps and institutional affiliations.