RPS 1104 - New applications of advanced CT and MRI

Purpose:

Cystic fibrosis (CF) is characterised by chronic obstructive lung disease, driven by recurrent infections and inflammation. Our aim was to elucidate the value of micro-CT (µCT) for monitoring disease progression in a mouse model of CF-like lung disease and for quantifying effects of neutrophil elastase (NE) knock-out as a potential target for therapeutic intervention.

Methods and materials:

60 conserved lungs from mice with four different genotypes combining CF-like lung disease (Scnn1b-Tg) or wild-type (WT) with neutrophil elastase knock-out (Scnn1b-Tg/NE^-/-, WT/NE^-/-) were scanned by µCT (8.6µm resolution) three times (newborn, 5-6, and 14-16 days old). Alveolar size measured by mean linear intercepts (LM), lung volume, and air proportion were quantified on µCT.

Results:

Average LM was similar in newborn mice among all four groups. At 5-6 days and 14-16 days, the alveolar size was larger in Scnn1b-Tg compared to WT (p<0.01), indicating the development of emphysema. Lung volume was significantly increased at 14-16d in Scnn1b-Tg (p<0.05) and air proportion was increased from 5-6 days (p<0.01). Importantly, Scnn1b-Tg/NE^-/- showed a significantly reduced air proportion at both 5-6 days and 14-16 days compared to Scnn1b-Tg alone, though the alveolar size and lung volume only showed a trend towards reduction be NE knock-out, indicating that NE knock-out reduces emphysema in CF-like lung disease.

Conclusion:

µCT detects early development of emphysema in a mouse model of CF-like lung disease, which can potentially be mitigated by NE deficiency. Non-invasive imaging of µCT allows for longitudinal quantification of CF-like lung disease and it may be used as an endpoint for novel therapeutic trials in this mouse model.

Limitations:

Data from adult mice has not yet been analysed and more parameters are being collected.

Ethics committee approval

Approved by the federal animal protection authority.

Funding:

Supported by the German BMBF.

Purpose:

To compare capabilities of pulmonary MR imaging with ultra-short echo time (UTE-MRI) for lung nodule detection and lung-RADS classification with thin-section low- and standard-dose CTs.

Methods and materials:

110 consecutive patients were examined with chest standard- and low-dose CTs (270 mA [SDCT] and 60 mA [LDCT]) and UTE-MRI. In each patient, the probability of presence at each pulmonary nodule was assessed on all three methods by means of a 5-point visual scoring system. In addition, all nodules were classified based on lung-RADS on each method by the same radiologists. To compare nodule detection capability, Jackknife alternative free-response receiver operating characteristic analysis was performed among all methods. To evaluate lung-RADS classification capability, inter-method agreements were also assessed by kappa statistics with χ² test performed.

Results:

On comparison of nodule detection capability by consensus reading, figure-of-merits (FOMs) of all methods (UTE-MRI: FOM=0.89, LDCT: FOM=0.86, SDCT: FOM=0.89) had no significant difference (p>0.05). For lung-RADS classification, interobserver agreement of each method was determined as almost perfect (UTE-MRI: κ=0.92, p<0.0001; LDCT: κ=0.93, p<0.0001; SDCT: κ=0.96, p<0.0001). In addition, inter-method agreements were also assessed as almost perfect (UTE-MRI vs. LDCT: κ=0.86, p<0.0001; UTE-MRI vs. SDCT: κ=0.87, p<0.0001; LDCT vs. SDCT: κ=0.95, p<0.0001).

Conclusion:

Pulmonary MR imaging with UTE is considered at least as valuable as low- and standard-dose CTs for lung nodule detection and lung-RADS classification.

Limitations:

This study was performed as a comparison of nodule detection capability and lung-RADS classification among MRI and low- and standard-dose CTs. However, there was no direct comparison of screening capability among them.

Ethics committee approval

This prospective study was approved by the institutional review board of Kobe University Hospital and written informed consent was obtained from each subject.

Funding:

This prospective study was supported by Canon Medical Systems Corporation.

Purpose:

To investigate the potential of non-contrast enhanced functional MRI modalities, including diffusion-weighted imaging (DWI) and chemical shift imaging (CSI), to characterise anterior mediastinal lesions and differentiate benign from malignant spectrum, as well as different subtypes in the same pathological category, in an attempt to improve non-invasive approaches for the diagnosis and follow-up of patients with mediastinal lesions.

Methods and materials:

This study included 71 patients with anterior mediastinal lesions detected by multi-slice computed tomography (MSCT). All cases were evaluated by diffusion-weighted imaging, ADC mean and minimum values, assessed and chemical shift MRI imaging, chemical shift ratio (CSR), and signal intensity index (SII), and were calculated and compared.

Results:

In diffusion-weighted imaging benign and malignant lesions cut off value was for the ADC mean (1.267 x10-3 mm2/sec) and for ADC min (1.215 x10-3 mm2/sec). For differentiation between sarcoidosis and malignant lymphadenopathy, the cut-off value for ADC mean was 1.298 x10-3 mm2/sec and ADC min was 1.215 x10-3 mm2/sec. In chemical shift imaging quantitative assessment, thymic hyperplasia showed lower CSR and higher SII values than the thymoma and thymic carcinoma, however, germ cell tumour showed borderline CSR and high SII.

Conclusion:

Functional MRI findings, both qualitative and quantitative assessment using diffusion-weighted images and chemical shift MR-images of the anterior mediastinum, has a diagnostic potential to differentiate between benign and malignant tumours. Patient’s demographics, the mediastinal compartment in question, and conventional MR findings were all taken into consideration.

Limitations:

The limited number of cases in each group.

Ethics committee approval

This study was conducted after institutional and departmental ethical committee approval number I- 091016 and informed consent was obtained from patients or their authorised representatives.

Funding:

No funding was received for this work.

Purpose:

To compare the capability of therapeutic effect prediction for chemoradiotherapy between chemical exchange saturation transfer (CEST) imaging at 3.5 ppm and FDG-PET/CT in non-small cell lung cancer (NSCLC) patients.

Methods and materials:

32 stage III NSCLC patients underwent CEST imaging, FDG-PET/CT, and chemoradiotherapy and follow-up examinations. All patients were then divided into recurrence (n=7) and non-recurrence (n=25) groups. From CEST data in each patient, magnetisation transfer ratio asymmetry (MTR_asym) map was computationally generated. In each lesion, MTR_asym and maximum value of standard uptake value (SUV_max) were assessed by ROI measurements. To compare each index between two groups, Student’s t-test was performed. Then, multivariate logistic regression analyses were performed to investigate the discriminating factors of the two groups. Finally, disease-free between responders and non-responders assessed by each index were compared by the Kaplan-Meier method followed by a log-rank test.

Results:

MTR_asym and SUV_max had significant differences between the two groups (p<0.05). Multivariate regression analyses identified MTR_asym (Odds ratio [OR]: 0.23, p=0.04) and SUV_max (OR: 0.09, p=0.008) as significant differentiators. Both indexes had significant differences in disease-free survival between the two groups (MTR_asym: p=0.01, SUV_max: p=0.0006).

Conclusion:

CEST imaging has potential for predicting the therapeutic effect of chemoradiotherapy and is considered at least as valuable as FDG-PET/CT in NSCLC patients.

Limitations:

No comparison of survival assessed by each index was considered as one of the limitations in this study.

Ethics committee approval

This prospective study was approved by the institutional review board of Kobe University Hospital and written informed consent was obtained from all patients.

Funding:

It was financially and/or technically supported by Canon Medical Systems Corporation.

Purpose:

To evaluate the potential of FD-MRI, a validated technique for lung perfusion assessment, in differentiating CF patients with respiratory tract exacerbation and its correlation with clinical parameters.

Methods and materials:

We selected 32 consecutive CF patients at their visit, performing a clinical examination, spirometry, and pulmonary MRI. Patients with exacerbations, identified with a Pex score, underwent antibiotic treatment, while those who were stable were not treated. A second examination was performed after therapy in exacerbated and after 30 days in the stable patients. For FD-MRI, an untriggered two-dimensional SSFP coronal sequence was used. The data obtained was converted into perfusion maps by a specific FD software. Two readers independently analysed the maps for perfusion defects using a field-based scoring system (0-18). We assessed intra- and inter-rater agreement (Cohen's k), differentiation between stable and exacerbated (OR), and correlation with spirometry parameters (Kendall‘s Tau).

Results:

There were 14 patients with exacerbation and 17 stable patients. The score showed an inter- and intra-rater agreement k of 0.85 and 0.93 (p<0.05). Exacerbated patients had a significantly higher score than stable patients in the first examination (OR:4.0, IC95%:1.4-12.3), while after therapy there wasn’t any difference (OR:1.1, IC95%:0.4–2.5). The score showed an inverse correlation with FEV1, FEF, and FVC in exacerbated patients, both in the first (Tau: -0.59,-0.61,-0.66, p<0,05) and in second examinations (-0.41,-0.43,-0.58, p<0.05). No correlation was found for the stable patients.

Conclusion:

The score applied to FD-MRI perfusion maps showed good reproducibility. It appeared to be effective in differentiating CF patients with pulmonary exacerbation and could play a role in monitoring as it seems correlated with spirometry parameters.

Limitations:

A small patient population. There were very few studies about the clinical use of FD-MRI.

Ethics committee approval

Approved by local ethics committee. Informed consent obtained.

Funding:

No funding was received for this work.

Purpose:

To investigate whether CT scans in maximum inspiration or expiration correlate more adequately with clinical data.

Methods and materials:

172 patients with COPD GOLD stage III or IV underwent non-enhanced thin-section MSCT at maximum inspiration and expiration, pulmonary function testing (PFT), exercise testing, and quality of life assessments within a narrow time frame. Quantitative CT (QCT) analysis was performed to determine total lung volume (TLV), total emphysema score (TES), and the 15th percentile of lung attenuation (P15) at both inspiration (IN) and expiration (EX) using MeVisPULMO 3D software. Spearman correlation analysis was used to correlate QCT analysis with PFT parameters and clinical measurements.

Results:

QCT parameters showed a varying correlation with PFT, exercise testing, and quality of life assessments: e.g. TLVIN showed a very strong correlation with TLC (r=0.81, P<0.001), a moderate to strong correlation with RV, FVC, and FEV1/FVC (r=0.60, 0.56, -0.49, P<0.001). TLVEX showed a moderate to strong correlation with TLC, RV, and FEV1/FVC (r=0.75, 0.66, -0.43, P<0.001). P15Diff showed a moderate correlation with KCO% and TLCO% (r= 0.41, 0.40, P<0.001).

Conclusion:

This study showed QCT parameters and PFT are complementary tools to evaluate pulmonary emphysema in patients of GOLD stages III and IV. TES and P15 at expiration might usefully reflect pulmonary function. Furthermore, P15 can be used as an effective parameter for evaluating lung diffusion capacity.

Limitations:

As we investigated CT scans and clinical data of patients with pulmonary emphysema, our results might not be applicable to the full extent to different subgroups.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

Recent studies support radiation-free morpho-functional magnetic resonance imaging (MRI) as a potential endpoint in cystic fibrosis (CF) lung disease and chronic obstructive pulmonary disease (COPD). Our aim was to determine the mid-term reproducibility of morphological and functional MRI abnormalities in clinically stable CF and COPD patients with repeat MRI studies one month apart, and to correlate morphological and perfusion abnormalities with spirometry.

Methods and materials:

15 CF (29.3±9.3y, FEV1%=66.59±15.83%) and 20 COPD (66.5±8.9y, FEV1%=42.04±13.26%) patients prospectively underwent clinical work-up, spirometry, and chest MRI twice one month apart (MRI1, MRI2). MRI scans were assessed using a validated MRI score, blinded for clinical and demographic data.

Results:

The MRI morphology, perfusion, and global score did not differ between MRI1 versus MRI2 in patients with CF and COPD, with the exception of the MRI global score which increased slightly in COPD from 14.6±3.8 to 15.2±3.8 (p<0.05). Limits-of-agreement between MRI1 and MRI2 were approximately ±4 for the MRI global score. Importantly, FEV1% also did not change from MRI1 to MRI2. MRI scores showed a moderate inverse correlation with FEV1% in CF (r=-0.57 to -0.70, p≤0.001), but not in COPD, which was mostly due to a narrow range of disease severity.

Conclusion:

Chest MRI abnormalities were highly reproducible within one month in clinically stable CF and COPD patients. Correlation with spirometry in CF further supports MRI’s role as a sensitive and robust modality for the assessment of regional disease severity.

Limitations:

Our study was performed in adult CF patients and the results may not be readily transferable to paediatric CF patients.

Ethics committee approval

This study was approved by the institutional ethics committee and written informed consent was obtained from all participants.

Funding:

This study was funded by Boehringer Ingelheim Pharma GmbH & Co. KG, Germany.

Purpose:

To establish a link between total coronary calcification (CAC), detected on routine chest CT exams following a visual ordinal scoring system (OSS), and the possibility of developing a cardiovascular event (CE) over time.

Methods and materials:

A retrospective case-control study was conducted selecting patients referred to our hospital for CE during 2015-16 who had a previous chest CT done, then two same-gender control patients with chest CT performed the same day and in previous dates were added. A total of 84 patients were included. An OSS was used to evaluate length coronary calcification and correlated with the presence or absence of CE.

Probabilities and curves of disease-free survival were calculated with the Kaplan-Meier method. The differences between curves were evaluated with the log-rank test.

SPSS 20.0 statistical analysis software was used, including Pearson χ2 test, t-Student, or Mann-Whitney tests and ANOVA or Kruskal-Wallis for group comparison.

Statistical significance was considered at a 5% probability level (p<0.05).

Results:

CAC was observed in 92.85% of cases and 64.81% of the controls, with the descent coronary artery being the most commonly calcified.

A significative relationship between the number of calcified arteries and the total score obtained in the visual scoring system with having suffered a CE (Pearson correlation 17 and 21 with p< 0.001, respectively) was found using the Chi-square test (p< 0.001) and survival Kaplan-Meier curve.

A score of >4.9 determined a strong posibility of CE in a 36 month follow-up period.

Conclusion:

Performing a visual score of CAC in routine chest CT can provide prognostic information concerning cardiovascular risk and should be incorporated in all radiological reports.

Limitations:

A retrospective study and small sample.

Ethics committee approval

n/a

Funding:

No funding received for this work.

Purpose:

Various quantitative evaluation techniques of computed tomographic (CT) images have been proven to deliver useful and objective biomarkers describing lung parenchyma and airways. Because low dose (LD) CT is becoming the standard in lung CT imaging, the question arises how an ultra-low dose (ULD) CT at dose level like that in chest x-ray influences quantitative CT (QCT).

Methods and materials:

In the present study, 198 individuals (age: 54.4±16.4yrs) with a clinical indication for low-dose CT, such as malignancy screening or follow-up of a known lung nodule, were scanned twice, once with LD (mean effective dose: ~2.18mSv) and once with ULD (~0.18mSv). Advanced modelled iterative reconstruction (ADMIRE) was used. LDCT was reconstructed with strength level 2 (LDCT₂), ULDCT with strength level 2, and 4/5 (ULDCT₂, ULDCT_4/5). Numerous fully automated QCT parameters were generated on global and lobar generation-based levels. LDCT served as ground truth.

Results:

The mean lung volume (LV) for ULDCT₂ was 0.58% smaller compared to LDCT₂ (ULDCT_4/5: 0.60%) and the differences were not significant. The mean lung density (MLD) was significantly increased by 6.56HU (ULDCT_4/5: 3.73HU), linear regression parameters for ULDCT₂~LDCT₂: r²=0.97, slope=0.99, intercept=0.52HU (ULDCT_4/5~LDCT₂: r²=0.97, slope=1.01, intercept=13.6HU), largest deviations for the lingula with 6.85HU (2.15HU), r²=0.94, slope=0.98, and intercept=-13.0HU (r²=0.9, slope=0.86, intercept=-100.0HU). Airway wall percentages (WP) were significantly increased by 2.59% (1.04%), r²=0.92, slope=0.89, and intercept=7.98% (r²=0.93, slope=0.89, intercept=7.16%).

Conclusion:

There are significant differences between QCT parameters in LDCT and ULDCT, however, they are strongly correlated. The accuracy of QCT parameters decreases for the lobes and for higher airway generations. Regarding these findings, ULDCT may be an alternative to LDCT. ADMIRE strength 4/5 generally delivered preferable results.

Limitations:

Results are influenced by inspiration level.

Ethics committee approval

Study under Mayo Clinic IRB approval: #14-002156.

Funding:

Dr. Fletcher received a research grant from Siemens Healthcare.