Pediatric Radiology

https://doi.org/10.1007/s00247-022-05539-9

ESPR

Chest magnetic resonance imaging in cystic fibrosis: technique

and clinical benefits
Daniel Gräfe1   · Freerk Prenzel2   · Franz Wolfgang Hirsch1 

Received: 27 April 2022 / Revised: 31 May 2022 / Accepted: 14 October 2022

© The Author(s) 2022

Abstract
Cystic fibrosis (CF) is one of the most common inherited and life-shortening pulmonary diseases in the Caucasian popu-
lation. With the widespread introduction of newborn screening and the development of modulator therapy, tremendous
advances have been made in recent years both in diagnosis and therapy. Since paediatric CF patients tend to be younger and
have lower morbidity, the type of imaging modality that should be used to monitor the disease is often debated. Computed
tomography (CT) is sensitive to many pulmonary pathologies, but radiation exposure limits its use, especially in children
and adolescents. Conventional pulmonary magnetic resonance imaging (MRI) is a valid alternative to CT and, in most cases,
provides sufficient information to guide treatment. Given the expected widespread availability of sequences with ultra-short
echo times, there will be even fewer reasons to perform CT for follow-up of patients with CF. This review aims to provide
an overview of the process and results of monitoring CF with MRI, particularly for centres not specialising in the disease.

Keywords  Bronchiectasis · Chest · Children · Computed tomography · Cystic fibrosis · Eichinger score · Fourier
decomposition · Magnetic resonance imaging · Pulmonary · Ultra-short echo times

Introduction Why image the lungs?

Cystic fibrosis (CF) is one of the most common, inherited and
life-limiting diseases in the Caucasian population. The mean
prevalence in European Union (EU) countries is 0.74/10,000,
with an incidence of 1/3,000 to 1/6,000 live births [1, 2]. A gene
defect encoding a chloride channel causes this multisystemic
disease whereby progressive lung disease is the leading cause of
death [3]. While in the past CF often resulted in demise in child-
hood and adolescence, improvements in therapy and diagnostics
have increased the median life expectancy into mid-adulthood
[4]. Two factors have significantly altered the management of CF
and, thus, survival in recent years: the widespread introduction
of newborn screening, resulting in presymptomatic therapy [5]
and the development of modulator therapies, allowing treatment
of the defective chloride channel [6].
Physical examination, microbiological and pulmonary func-
tion tests and laboratory parameters are used to monitor the
progression of the disease. Structural lung changes occur in
infancy and during the preschool period and may be missed
in asymptomatic children [7, 8]. However, high-resolution
cross-sectional imaging can detect abnormalities in the lungs
more sensitively than lung function diagnostics and, thus,
direct therapy at an early stage [8, 9]. It has been shown that
in infants and young children, magnetic resonance imaging
(MRI) correlates well with the (sensitive) lung clearance
index [10]. In addition, imaging can sometimes identify
the cause of pulmonary deterioration or even distinguish
between fungal and other microbial causes of infection, ena-
bling targeted therapy [11].

* Daniel Gräfe
daniel.graefe@medizin.uni-leipzig.de
What do clinicians need to know?
1
Department of Pediatric Radiology,
Leipzig University Hospital, Ideally, the treating pulmonologist needs precise and com-
Liebigstraße 20a, 04103 Leipzig, Germany prehensible parameters by which to assess the progression
2
Department of Pediatrics, of the disease and from which they can immediately derive
Leipzig University Hospital, clues for patient management. The imaging modality is not
Liebigstraße 20a, 04103 Leipzig, Germany

13
Vol.:(0123456789)

the primary objective for the clinician. The radiologist, on
the other hand, must be aware of the perspective and requests
of the referring physician, even if they have not been explic-
itly phrased. Before any examination, consider:
a) Are there changes over time that have prognostic or
therapeutic relevance for the patient? Multiple scores for
each imaging modality are useful for such quantification of
longitudinal trends [12–15].
b) Is there an acute event that requires immediate ther-
apy? Examples include acute respiratory tract infective
exacerbations or pulmonary haemorrhage. The radiologic
assessment should, irrespective of the score, explicitly evalu-
ate and report this aspect.
The basis for each score is the morphological, patho-
logical alteration of the lung. These appear with different
emphasis depending on the imaging modality: bronchial wall
thickening, bronchiectasis, consolidations, bullae, mucus
plugs and hyperinflation (Figs. 1, 2, 3, 4, 5 and 6) [16, 17].
In addition, functional assessment of regional changes in
ventilation and perfusion is also desirable as complementary
information.
Any change in appearance of morphological lesions
forms the basis for prophylactic or therapeutic medical
treatment and physiotherapy. In infective exacerbations, the
imaging modality should ideally provide evidence of bacte-
rial or mycotic aetiology. In the rare case of haemoptysis,
angiographic imaging of the bronchial arteries is necessary
Fig. 1  Magnetic resonance
images in a 16-year-old
girl with bronchiectasis in the
upper lobes (arrows). a Coronal
respiratory-triggered T2 fast
spin echo and self-gated ultra-
short echo time sequences with
(b) coronal, (c) axial and (d)
sagittal multiplanar reconstruc-
tions. Please note that air trap-
ping distal to the dilated bronchi
is only depicted in the latter
13
Pediatric Radiology
since the haemorrhage may be treated by interventional
occlusion [18].
Why MRI?
Traditionally, imaging of the lungs has been based on con-
ventional projection radiography. Over the past decades,
multiple chest radiograph-based scores have been developed
to objectify the findings in CF [19, 20]. However, a consider-
able amount of subjectivity persists in reporting, especially
in the evaluation of conventional radiographic images. Many
lesions are only visualised as nonspecific surrogate findings
(e.g., ring shadows, mottled shadows and soft shadows)
[21]. Artificial intelligence-based software has potential to
enhance the consistency of radiographic assessment, but still
requires further development [22].
Given the limitations of conventional radiography, com-
puted tomography (CT) as a fast, high-resolution, cross-
sectional diagnostic technique is, therefore commonly used
by many centres for follow-up in CF [23]. In addition, unlike
MRI, CT achieves sufficient quality without sedation in chil-
dren younger than 5 years of age [24].
The benefits of CT notwithstanding, the risk–benefit
trade-off of routine CT follow-up needs to be reevaluated
[25], even when effective dose is very low [26]. As a result
of expanding treatment, the number of relatively healthy
Pediatric Radiology
Fig. 2  Coronal (a) and axial (b) T2 fast spin  echo images  in a
16-year-old girl with severe cystic fibrosis and infection with Staphy-
lococcus aureus and Pseudomonas aeruginosa show signs of bron-
chiectasis in the right upper lobe (arrows) and small mucus plugs in
the left upper lobe (arrowheads)
children who are regularly followed up has increased. Nowa-
days, the transition of paediatric patients to adult pulmonol-
ogy is a regular (if not routine) event. Thus, in the context
of the increasing life expectancy of patients with CF, an
increase in tumour disease might potentially occur among
patients as a late consequence of ionising radiation [27],
which should therefore be minimised.
Pulmonary MRI, the radiation-free alternative to radio-
graphs and CT, has the disadvantage of inadequate image
quality in about 10% of examinations [28]. However, with
simplified respiratory gating and new sequence patterns, as
described below, lung MRI examinations are increasingly
simpler to perform. This is expected to drastically reduce the
number of inadequate MRI lung studies. Today, lung MRI
is a viable alternative to chest CT [29, 30], as reflected by
the growing body of comprehensive reviews on lung MRI
in patients with CF [31, 32].
Fig. 3  A coronal ultra-short echo time sequence in a 10-year-old
boy with cystic fibrosis and good pulmonary  function. Self-gating
(a) with an acquisition time of  6:15 min. Single breath-hold tech-
nique (b) with an acquisition time of 17 s. Peripheral mucus is found
in both upper lobes (arrows)
The basic protocol
Fundamental problems associated with lung imaging using
MRI include the low density of protons, rapid signal loss
due to T2* decay and, finally, the persistent and ubiquitous
periodic motion of the diaphragm, thoracic wall, heart and
great vessels. Lung MRI is feasible at both 1.5 T and 3 T
[33]. Lower static magnetic field strength is beneficial for
lung imaging due to its slower T2* decay [34, 35].
The basis of any pulmonary MRI is a thin-slice
T2-weighted sequence [28] (Tables 1 and 2). Typically, this
allows hyperintense visualisation of most pathologies: intra-
bronchial and alveolar mucus plugs, bronchial wall thick-
ening, consolidations and abscesses. Given their excellent
13

Fig. 4  An axial T2 fast spin echo image with fat saturation in a
12-year-old girl with severe pulmonary involvement in cystic fibro-
sis. Marked bronchiectasis and bronchial wall thickening affect both
upper lobes (arrows). In addition, minor consolidations are seen in
the right upper lobe (arrowhead)
visualisation on conventional MRI sequences, these find-
ings are also referred to as MR-plus pathologies. In con-
trast, smaller bullae without mass effect on surrounding
lung structure and interstitial lung lesions such as fibrosis
or bronchiectasis cannot reliably be detected with conven-
tional lung sequences unless they are thick-walled or filled
with mucus; thus, they are often referred to as MR-minus
pathologies [36, 37]. However, this term is losing its rele-
vance due to the more recent developments explained below
[38]. T2-weighted fast spin echo sequences (FSE) with
respiratory triggering are recommended for optimal qual-
ity, with an acquisition time of approximately 3 to 5 min,
depending on the patient’s breathing rate. Thin slices of
3–4 mm are recommended to minimize voxel averaging that
might prevent visualisation of micronodules or intralobu-
lar opacities. Often, substantially faster single-shot T2 spin
echo sequences are within a few breath-hold manoeuvres;
Fig. 5  A 16-year-old boy with cystic fibrosis. A  T2 fast spin echo
magnetic resonance (MR) image (a). The mucus in the bronchiectasis
in the basal right upper lobe is hypointense on T2 (arrow). An axial
unenhanced computed tomography (CT) image  (b). The mucus in
13
Pediatric Radiology
however, the clarity and hence the diagnostic confidence of
these fast techniques for smaller lesions is inferior to respir-
atory-triggered FSE sequences [39, 40]. Therefore, even in
older children, respiratory triggering for T2 imaging might
be preferred. The different options for respiratory trigger-
ing (respiratory belt, diaphragm scout, phase scout in the
liver or trigger systems integrated into the MR table) work
comparably well and differ predominantly regarding their
convenience of application [41, 42].
While T1 weighting is not mandatory for pulmonary MRI
in most other conditions, it should be performed routinely
in CF. The reason for this is the increased T1 signal in aller-
gic bronchopulmonary aspergillosis, which along with the
low T2 signal is considered characteristic of this disease
(Fig. 5) [11]. Usually, T1 weighting is achieved as a 3-D
gradient-echo acquisition in free breathing or during a single
breath-hold command. Diffusion weighting may emerge as
another parameter of disease severity in the future [43], but
in routine diagnostics it is reserved for those situations when
an abscess is suspected. Tables 1 and 2 provide exemplary
protocols for different ages.
For the ambitious
A major advance in lung MRI in recent years has been the
development of ultra-short echo time (UTE) sequences (or
zero-TE sequences) (Figs. 1 and 3). Extremely short echo
times of less than 0.05 ms precede the rapid T2* decay in the
lungs, resulting in augmented signal even from healthy lung
tissue [38]. At low flip angles, a proton-density weighted
impression is obtained. MR-minus pathologies particularly
can be delineated more adequately with UTE sequences than
with conventional MRI techniques [44]. Self-gated UTE
sequences are available in free breathing where even dif-
ferent breathing phases can be reconstructed. In addition,
highly optimised UTE sequences that allow imaging of the
the bronchiectasis in the basal right upper lobe is hyperdense on CT
(arrow). This combination of MR and CT findings is characteristic of
allergic bronchopulmonary aspergillosis and aspergilloma
Pediatric Radiology

Fig. 6  An example of an
unenhanced functional proton
magnetic resonance image
(MRI) (Fourier decomposition)
for the estimation of ventilation
by means of phase-resolved
functional lung MRI (PREFUL)
[16] in a free-breathing 6-year-
old boy after severe pneumonia
(acquisition time of 60 s). Bul-
lae can be observed in the left
lower lobe (basal) and the right
upper lobe, which ventilate with
delayed filling and emptying
of air (ventilation circle with
arrows around the image). The
automated computation map
(within the ventilation circle)
reveals healthy lung (green) as
well as extensive damage to the
lungs with perfusion deficits
(red) and ventilation-perfusion
mismatch (purple)

Table 1  Protocol for a T2 turbo spin echo PD-UTE T1 3-D Stack

15-min, non-contrast routine of Stars GRE
morphological follow-up (StarVIBE)
examination for cystic fibrosis
in children younger than 6 years Acquisition mode 2-D 3-D 3-D
of age (exemplary for Siemens
Orientation axial/coronal coronal axial
3 T)
Field of view (mm) 300 × 216 260 260
Matrix size (mm) 384 × 193 256 × 256 224
Slice thickness (mm) 3 3 (interpolated) 2.5 (interpolated)
Repetition time/echo time(ms) 1,000/50 3.7/0.05 3.7/2.46
Slice gap (%) 5 0 0
Flip angle (°) 120 5 9
Parallel imaging GRAPPA 2 n/a n/a
Respiratory compensation triggered self-gated free breathing
Duration 3–5 min each 1:40 min 1:14 min

GRAPPA generalized autocalibrating partially parallel acquisition, GRE gradient echo, n/a not applicable,
PD proton density, UTE ultra-short echo time, VIBE volume interpolated breath-hold examination

complete lung in a single breath-hold manoeuvre are pos-
sible. High-resolution UTE sequences are often acquired as
isovoxel 3-D sequences in free breathing and with a voxel
size of less than 1.5 mm, like CT, allow multiplanar recon-
struction. Once manufacturers adopt these UTE sequences
into their standard portfolio, they will become an integral
part of routine CF imaging.
Contrast-enhanced 3-D sequences with high temporal res-
olution rendered by keyhole imaging techniques are widely
available techniques for qualitative assessment of lung per-
fusion [45]. Through hypoxic pulmonary vasoconstriction,
perfusion also yields indirect information about ventilation.
However, in cases of haemoptysis, with bronchial artery
hypertrophy as a treatable cause of bleeding, CT angiogra-
phy remains the modality of choice [46]. Whether routine
use of contrast-enhanced functional assessments should be
recommended outside the research setting remains unclear.
Acknowledging that even macrocyclic contrast agents lead

13
 Pediatric Radiology

Table 2  Protocol for a T2 turbo spin echo PD-UTE T1 3-D Stack

14-min, non-contrast, routine of Stars GRE
morphological follow-up (StarVIBE)
examination for cystic fibrosis
in children older than 8 years Acquisition mode 2-D 3-D 3-D
of age (exemplary for Siemens
Orientation axial/coronal coronal axial
3 T)
Field of view (mm) 350 × 284 450 400 × 263
Matrix size (mm) 384 × 218 288 384 × 189
Slice thickness (mm) 3 2 (interpolated) 3 (interpolated)
repetition time/echo time (ms) 2,000/50 2.9/0.05 4/1.85
Slice gap (%) 5 n/a n/a
Flip angle (°) 120 5 5
Parallel imaging GRAPPA 2 SPIRiT 2 CAIPI 4
Resp. compensation triggered single breath-hold single breath-hold
Duration 3–5 min each 0:20 min 0:20 min

CAIPI controlled aliasing in parallel imaging, GRAPPA generalized autocalibrating partially parallel acqui-
sition, GRE gradient echo, n/a not applicable, PD proton density, SPIRiT iterative self-consistent parallel
imaging, UTE ultra-short echo time, VIBE volume interpolated breath-hold examination

to deposition in the paediatric brain [47] and considering
that progressively healthier and younger patients will be
given an early diagnosis and new therapies in the future, the
risk–benefit of routine contrast agent administration must be
questioned. Another promising new method is non-contrast
perfusion, such as arterial spin labelling [48] and variants of
Fourier decomposition [49]. Since the latter provides highly
encouraging results [50] and is not far from transitioning
into clinical practice, it will be discussed in the next section.
For specialists
The clearest and most elaborate visualisation of ventilation
is achieved with hyperpolarised gases such as 3helium and
129
xenon. Thereby, a 3-D lung volume is acquired during a
single breath-hold manoeuvre after the preceding inhalation
of roughly 1 litre of the polarised gas. Though these methods
represent ventilation disorders with high sensitivity [51, 52],
three drawbacks have hindered their introduction into the
clinical arena since their initial description 25 years ago:
First, the sourcing and subsequent on-site polarisation of the
gases is costly. Second, dedicated MR hardware enhance-
ments are required, since standard high-frequency generators
and receiver coils are designed for proton MRI [53]. Finally,
handling of gas inhalation followed by immediate scanning
requires extensive training. Therefore, this technique is per-
formed only in a few specialised institutions and is not suit-
able for routine clinical use even in the medium term.
Fourier decomposition for the assessment of perfusion
and ventilation is an important and elegant innovation in
functional lung MRI and an alternative to both contrast
administration and ventilation imaging with hyperpolarised
gases (Fig. 6) [54, 55]. This promising technique combines
perfusion and ventilation measurement and is based on
ordinary proton MRI as opposed to sophisticated methods
employing hyperpolarised gases. The requisite for this tech-
nique is a fast gradient echo sequence, which is available on
every scanner; hence, the need for expensive investment is
avoided. Thick coronal slices of about 10 mm are continu-
ously acquired with high temporal resolution at a single slice
position in free breathing. For each desired slice position,
the acquisition time required for the reconstruction of one
virtual ventilation and perfusion cycle is about 60 s. Subse-
quently, in postprocessing, ventilation and perfusion video
clips as well as a quantitative mismatch map are calculated
on any computer using dedicated software [56]. To date, this
software is not commercially available, which means that
the technology is currently only being used in a research
context; however, clinical rollout is expected soon.
T1 mapping of the lung as a quantitative parameter has
been proposed and explored for CF in the research context.
But so far, this method has not found its way into routine
clinical practice [57].
How to report?
As a first step, an assessment of whether the quality of the spe-
cific MRI examination is sufficient or whether some minor or
pronounced diagnostic limitations remain is recommended.
This assessment informs the pulmonologist whether the present
examination is of sufficient quality to answer all questions. In
case of pronounced diagnostic limitations, a short-term repeat
MRI examination or CT must be arranged, if of acute relevance.
After quality assessment, a free or structured report is prepared
regarding acute changes in comparison to the previous exami-
nations focusing on findings that require immediate treatment.

13
Pediatric Radiology
Table 3  Eichinger score (modified after [13]) for morphological
imaging of the lung in cystic fibrosis. Individual findings are scored
according to whether pathology is absent (0), present in less than
R-UL R-ML
Bronchiectasis/wall thickening 2 2
Mucus plugging 2 2
Abscesses/sacculations 2 2
Consolidation 2 2
Pleural findings 2 2
Lobar score (total) 10 10
L left, LG lingula, LL lower lobe, ML middle lobe, R right, UL upper lobe
The challenge of MRI reporting arises from the fact that
the lesions are often multiple and very unevenly distrib-
uted in the lung and that the irregularly confined typical
CF lesions are more difficult to measure than, for example,
nodules. Further, the T2 signal intensity as a relative value of
pathology is not readily quantifiable, although there are early
encouraging results in this regard [13, 58, 59]. Attempts
have been made to address these challenges with several
semiquantitative CF scores for MRI [12–14]. To quantify
the extent of pulmonary pathology and to reflect disease
progression in a single parameter, a semiquantitative score
can now be applied. It must be emphasised that CF scores,
whether for CT or MRI, have a certain investigator depend-
ence and are not well validated in terms of their predic-
tive power for clinically relevant outcomes [60]. However,
there is sufficient evidence that they correlate adequately
with other surrogate parameters of disease severity, such as
forced expiratory volume in 1 s [12] or the lung clearance
index [10].
A simple and feasible score is the Eichinger score [13],
which consists of a morphological and an independent func-
tional (MR perfusion) component (Table 3). An extension
of this score for mosaic signal intensities has been recently
proposed [61]. To estimate this score, the different lobes of
the lung (upper lobe, lower lobe, middle lobe and lingula)
are considered separately. For each lobe, alterations of the
bronchial wall, mucus plugging, abscesses, consolidation
and pleural alterations are scored semiquantitatively on a
scale from zero to two, depending on whether the lesion is
absent in the corresponding lobe or whether it affects less
or more than half of the lobe. An improvement in the quan-
titative assessment of hyperinflation can be expected in the
future from the incorporation of ventilation inhomogenei-
tes as shown by contrast-enhanced or unenhanced perfu-
sion, ventilation or UTE imaging. A generic disadvantage
of scores is that it is often uncertain what weights should
be assigned to the individual components. This might be
revolutionized by deep-learning algorithms that can inde-
pendently determine prognostic criteria. First approaches
half (1) or present in more than half (2) of a lobe. If lung perfusion is
recorded, an additional line can be added, which would increase the
total score to 72
R-LL L-UL L-LG L-LL Pathology
score (total)
2 2 2 2 12
2 2 2 2 12
2 2 2 2 12
2 2 2 2 12
2 2 2 2 12
10 10 10 10 60
to the computer-aided diagnosis of CF have already been
reported for CT [62, 63]. Adapting these algorithms to MRI
is complicated by the more heterogeneous image quality of
lung MRI compared with chest CT, but automated quantifi-
cation seems possible [59, 64].
What to expect for the future?
Ultra-short echo time sequences will become standard
additions to conventional CF lung MRI as soon as these
sequences are made widely available by manufacturers. In
addition, with non-contrast perfusion and ventilation meas-
urements of the lung, these sequences will eliminate the
need for contrast administration, which is still standard in
some centres due to its sensitivity.
A few fundamental disadvantages of MRI compared
to CT will linger on: First, the lengthy acquisition time of
MRI sequences usually requires sedation in children up to
about 5 years of age. Second, there are no studies on the
cost effectiveness of MRI compared to CT. However, it is
likely that such a comparison will favor CT even beyond
the age of infants. Finally, standardisation of MRI protocols
between different manufacturers is much more challenging
than for CT. Standardisation of MRI protocols and report-
ing is strongly advocated for the future [65] and is being
advanced by the imaging group of the Clinical Trial Network
of the European Cystic Fibrosis Society.
Even today, lung MRI for pulmonary monitoring in CF is
performed as a preferred method in many specialised insti-
tutions [31, 32]. However, the preference for lung MRI in
CF in these institutions has so far been based on individual
experience. A robust assessment of advantages and disad-
vantages regarding sensitivity of MRI compared to CT can
be anticipated from a large ongoing multicentre study in
France (https://​clini​caltr​ials.​gov/​ct2/​show/​NCT03​357562).
As a full evaluation of the lung can be achieved in free
breathing in about 15 min, pulmonary MRI might soon
replace chest CT as a standard in CF imaging in children.
13

Funding  Open Access funding enabled and organized by Projekt
DEAL.
Declarations  16. Glandorf J, Klimeš F, Voskrebenzev A et al (2020) Comparison of
Conflicts of interest None
Open Access  This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.
References
1. Scotet V, L’Hostis C, Férec C (2020) The changing epidemi-
ology of cystic fibrosis: Incidence, survival and impact of the
CFTRGene discovery. Genes (Basel) 11:589
2. Farrell PM (2008) The prevalence of cystic fibrosis in the Euro-
pean Union. J Cyst Fibros 7:450–453
3. Elborn JS (2016) Cystic fibrosis. Lancet 388:2519–2531
4. Balfour-Lynn IM, King JA (2020) CFTR modulator therapies –
Effect on life expectancy in people with cystic fibrosis. Paediatr
Respir Rev 42:3–8
5. Southern KW, Mérelle MM, Dankert-Roelse JE, Nagelkerke AD
(2009) Newborn screening for cystic fibrosis. Cochrane Database
Syst Rev 2009:CD001402
6. Ramsey BW, Davies J, McElvaney NG et al (2011) A CFTR
potentiator in patients with cystic fibrosis and the G551D muta-
tion. N Engl J Med 365:1663–1672
7. de Jong PA, Nakano Y, Lequin MH et al (2004) Progressive dam-
age on high resolution computed tomography despite stable lung
function in cystic fibrosis. Eur Respir J 23:93–97
8. Stahl M, Steinke E, Graeber SY et al (2021) Magnetic resonance
imaging detects progression of lung disease and impact of new-
born screening in preschool children with cystic fibrosis. Am J
Respir Crit Care Med 204:943–953
9. Tiddens HAWM (2006) Chest computed tomography scans should
be considered as a routine investigation in cystic fibrosis. Paediatr
Respir Rev 7:202–208
10. Stahl M, Wielpütz MO, Graeber SY et al (2016) Comparison of
lung clearance index and magnetic resonance imaging for assess-
ment of lung disease in children with cystic fibrosis. Am J Respir
Crit Care Med 195:349–359
11. Dournes G, Berger P, Refait J et al (2017) Allergic bronchopulmo-
nary aspergillosis in cystic fibrosis: MR imaging of airway mucus
contrasts as a tool for diagnosis. Radiology 285:261–269
12. Schaefer JF, Hector A, Schmidt K et al (2018) A semiquantitative
MRI-Score can predict loss of lung function in patients with cystic
fibrosis: Preliminary results. Eur Radiol 28:74–84
13. Eichinger M, Optazaite DE, Kopp-Schneider A et al (2012) Mor-
phologic and functional scoring of cystic fibrosis lung disease
using MRI. Eur J Radiol 81:1321–1329
14. Tepper LA, Ciet P, Caudri D et al (2016) Validating chest MRI
to detect and monitor cystic fibrosis lung disease in a pediatric
cohort. Pediatr Pulmonol 51:34–41
13
Pediatric Radiology
15. Helbich TH, Heinz-Peer G, Eichler I et al (1999) Cystic fibrosis:
CT assessment of lung involvement in children and adults. Radiol-
ogy 213:537–544
phase-resolved functional lung (PREFUL) MRI derived perfusion
and ventilation parameters at 1.5T and 3T in healthy volunteers.
PLoS One 15:e0244638
17. Hansell DM, Bankier AA, MacMahon H et al (2008) Fleisch-
ner Society: Glossary of terms for thoracic imaging. Radiology
246:697–722
18. Ittrich H, Bockhorn M, Klose H, Simon M (2017) The diagnosis
and treatment of hemoptysis. Dtsch Arztebl Int 114:371–381
19. Terheggen-Lagro S, Truijens N, Van Poppel N et al (2003) Cor-
relation of six different cystic fibrosis chest radiograph scoring
systems with clinical parameters. Pediatr Pulmonol 35:441–445
20. Chrispin AR, Norman AP (1974) The systematic evaluation of the
chest radiograph in cystic fibrosis. Pediatr Radiol 2:101–105
21. Davis SD, Fordham LA, Brody AS et al (2007) Computed tomog-
raphy reflects lower airway inflammation and tracks changes in
early cystic fibrosis. Am J Respir Crit Care Med 175:943–950
22. Zucker EJ, Barnes ZA, Lungren MP et al (2020) Deep learning to
automate Brasfield chest radiographic scoring for cystic fibrosis.
J Cyst Fibros 19:131–138
23. Smyth AR, Bell SC, Bojcin S et al (2014) European Cystic Fibro-
sis Society Standards of Care: Best Practice guidelines. J Cyst
Fibros 13(Suppl 1):S23–S42
24. Kino A, Zucker EJ, Honkanen A et al (2019) Ultrafast pediatric
chest computed tomography: comparison of free-breathing vs.
breath-hold imaging with and without anesthesia in young chil-
dren. Pediatr Radiol 49:301–307
25. Joyce S, Carey BW, Moore N et al (2021) Computed tomography
in cystic fibrosis lung disease: a focus on radiation exposure. Pedi-
atr Radiol 51:544–553
26. Moloney F, Kavanagh RG, Ronan NJ et al (2021) Ultra-low-dose
thoracic CT with model-based iterative reconstruction (MBIR)
in cystic fibrosis patients undergoing treatment with cystic fibro-
sis transmembrane conductance regulators (CFTR). Clin Radiol
76:393.e9-393.e17
27. Miglioretti DL, Johnson E, Williams A et al (2013) The use of
computed tomography in pediatrics and the associated radiation
exposure and estimated cancer risk. JAMA Pediatr 167:700–707
28. Hirsch FW, Sorge I, Vogel-Claussen J et al (2020) The current
status and further prospects for lung magnetic resonance imaging
in pediatric radiology. Pediatr Radiol 50:734–749
29. Renz DM, Scholz O, Böttcher J et al (2015) Comparison between
magnetic resonance imaging and computed tomography of the
lung in patients with cystic fibrosis with regard to clinical, labo-
ratory, and pulmonary functional parameters. Investig Radiol
50:733–742
30. Hatabu H, Ohno Y, Gefter WB et al (2020) Expanding applica-
tions of pulmonary MRI in the clinical evaluation of lung disor-
ders: Fleischner Society Position Paper. Radiology 297:286–301
31. Dournes G, Walkup LL, Benlala I et al (2021) The clinical use of
lung MRI in cystic fibrosis. Chest 159:2205–2217
32. Woods JC, Wild JM, Wielpütz MO et al (2020) Current state of
the art MRI for the longitudinal assessment of cystic fibrosis. J
Magn Reson Imaging 52:1306–1320
33. Biederer J, Beer M, Hirsch W et al (2012) MRI of the lung (2/3).
Why... when … how? Insights Imaging 3:355–371
34. Campbell-Washburn AE, Malayeri AA, Jones EC et al (2021)
T2-weighted lung imaging using a 0.55-T MRI system. Radiol
Cardiothorac Imaging 3:e200611
35. Chassagnon G, Martin C, Ben Hassen W et al (2019) High-reso-
lution lung MRI with ultrashort-TE: 1.5 or 3 Tesla? Magn Reson
Imaging 61:97–103
Pediatric Radiology
36. Dournes G, Menut F, Macey J et al (2016) Lung morphology
assessment of cystic fibrosis using MRI with ultra-short echo time
at submillimeter spatial resolution. Eur Radiol 26:3811–3820
37. Wielpütz M, Kauczor H-U (2012) MRI of the lung: state of the
art. Diagn Interv Radiol 18:344–353
38. Wielpütz MO, Triphan SMF, Ohno Y et al (2019) Outracing
lung signal decay – potential of ultrashort echo time MRI. RöFo
191:415–423
39. Cieszanowski A, Lisowska A, Dabrowska M et al (2016) MR
imaging of pulmonary nodules: detection rate and accuracy of size
estimation in comparison to computed tomography. PLoS ONE
11:e0156272
40. Hirsch W, Sorge I, Krohmer S et al (2008) MRI of the lungs in
children. Eur J Radiol 68:278–288
41. Morita S, Ueno E, Suzuki K, Machida H, Fujimura M, Kojima
S, Hirata M, Ohnishi T, Imura C (2008) Navigator-triggered pro-
spective acquisition correction (PACE) technique vs. conventional
respiratory-triggered technique for free-breathing 3D MRCP: an
initial prospective comparative study using healthy volunteers. J
Magn Reson Imaging 28:673–677
42. Santelli C, Nezafat R, Goddu B, Manning WJ, Smink J, Kozerke
S, Peters DC (2011) Respiratory bellows revisited for motion com-
pensation: preliminary experience for cardiovascular MR. Magn
Reson Med 65:1097–1102
43. Ciet P, Serra G, Andrinopoulou ER et  al (2016) Diffusion
weighted imaging in cystic fibrosis disease: beyond morphologi-
cal imaging. Eur Radiol 26:3830–3839
44. Gräfe D, Anders R, Prenzel F et al (2021) Pediatric MR lung
imaging with 3D ultrashort-TE in free breathing: are we past the
conventional T2 sequence? Pediatr Pulmonol 56:3899–3907
45. Wielpütz MO, Puderbach M, Kopp-Schneider A et al (2014) Mag-
netic resonance imaging detects changes in structure and perfu-
sion, and response to therapy in early cystic fibrosis lung disease.
Am J Respir Crit Care Med 189:956–965
46. Monroe EJ, Pierce DB, Ingraham CR et al (2018) An interven-
tionalist’s guide to hemoptysis in cystic fibrosis. Radiographics
38:624–641
47. Blumfield E, Swenson DW, Iyer RS, Stanescu AL (2019) Gad-
olinium-based contrast agents — review of recent literature on
magnetic resonance imaging signal intensity changes and tissue
deposits, with emphasis on pediatric patients. Pediatr Radiol
49:448–457
48. Donnola SB, Dasenbrook EC, Weaver D et al (2017) Preliminary
comparison of normalized T1 and non-contrast perfusion MRI
assessments of regional lung disease in cystic fibrosis patients. J
Cyst Fibros 16:283–290
49. Fischer A, Weick S, Ritter CO et al (2014) SElf-gated Non-Con-
trast-Enhanced FUnctional Lung imaging (SENCEFUL) using a
quasi-random fast low-angle shot (FLASH) sequence and proton
MRI. NMR Biomed 27:907–917
50. Behrendt L, Smith LJ, Voskrebenzev A et al (2022) A dual center
and dual vendor comparison study of automated perfusion-
weighted phase-resolved functional lung magnetic resonance
imaging with dynamic contrast-enhanced magnetic resonance
imaging in patients with cystic fibrosis. Pulm Circ 12:e12054
51. Thomen RP, Walkup LL, Roach DJ et al (2017) Hyperpolarized
129Xe for investigation of mild cystic fibrosis lung disease in
pediatric patients. J Cyst Fibros 16:275–282
52. Bannier E, Cieslar K, Mosbah K et al (2010) Hyperpolarized 3 He
MR for sensitive imaging of ventilation function and treatment
efficiency in young cystic fibrosis patients with normal lung func-
tion. Radiology 255:225–232
53. Roos JE, McAdams HP, Kaushik SS, Driehuys B (2015) Hyperpo-
larized gas MR imaging: technique and applications. Magn Reson
Imaging Clin N Am 23:217–229
54. Couch MJ, Munidasa S, Rayment JH et al (2021) Comparison
of functional free-breathing pulmonary 1H and hyperpolarized
129Xe magnetic resonance imaging in pediatric cystic fibrosis.
Acad Radiol 28:e209–e218
55. Bauman G, Puderbach M, Heimann T et al (2013) Validation of
Fourier decomposition MRI with dynamic contrast-enhanced MRI
using visual and automated scoring of pulmonary perfusion in
young cystic fibrosis patients. Eur J Radiol 82:2371–2377
56. Behrendt L, Voskrebenzev A, Klimeš F et al (2020) Validation
of automated perfusion-weighted phase-resolved functional lung
(PREFUL)-MRI in patients with pulmonary diseases. J Magn
Reson Imaging 52:103–114
57. Neemuchwala F, Ghadimi Mahani M, Pang Y et al (2020) Lung
T1 mapping magnetic resonance imaging in the assessment of
pulmonary disease in children with cystic fibrosis: a pilot study.
Pediatr Radiol 50:923–934
58. Puderbach M, Eichinger M, Haeselbarth J et al (2007) Assessment
of morphological MRI for pulmonary changes in cystic fibrosis
(CF) patients. Investig Radiol 42:715–724
59. Benlala I, Hocke F, Macey J et al (2020) Quantification of MRI
T2-weighted high signal volume in cystic fibrosis: a pilot study.
Radiology 294:186–196
60. Calder AD, Bush A, Brody AS, Owens CM (2014) Scoring of
chest CT in children with cystic fibrosis: state of the art. Pediatr
Radiol 44:1496–1506
61. Leutz-Schmidt P, Stahl M, Sommerburg O et al (2018) Non-
contrast enhanced magnetic resonance imaging detects mosaic
signal intensity in early cystic fibrosis lung disease. Eur J Radiol
101:178–183
62. Konietzke P, Weinheimer O, Wielpütz MO et al (2018) Validation
of automated lobe segmentation on paired inspiratory-expiratory
chest CT in 8–14 year-old children with cystic fibrosis. PLoS ONE
13:e0194557
63. Chassagnon G, Martin C, Burgel PR et al (2018) An automated
computed tomography score for the cystic fibrosis lung. Eur
Radiol 28:5111–5120
64. Benlala I, Point S, Leung C et al (2020) Volumetric quantification
of lung MR signal intensities using ultrashort TE as an automated
score in cystic fibrosis. Eur Radiol 30:5479–5488
65. Ciet P, Bertolo S, Ros M et al (2022) State-of-the-art review of
lung imaging in cystic fibrosis with recommendations for pul-
monologists and radiologists from the “iMAging managEment
of cySTic fibROsis” (MAESTRO) consortium. Eur Respir Rev
31:210173
Publisher’s note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
13