Best practices for MRI systematic reviews and meta‐analyses

Best Practice for MRI Diagnostic Accuracy Research With Lessons and Examples from the LI-RADS Individual Participant Data Group

Medical imaging diagnostic test accuracy research is strengthened by adhering to best practices for study design, data collection, data documentation, and study reporting. In this review, key elements of such research are discussed, and specific recommendations provided for optimizing diagnostic accuracy study execution to improve uniformity, minimize common sources of bias and avoid potential pitfalls. Examples are provided regarding study methodology and data collection practices based on insights gained by the liver imaging reporting and data system (LI-RADS) individual participant data group, who have evaluated raw data from numerous MRI diagnostic accuracy studies for risk of bias and data integrity. The goal of this review is to outline strategies for investigators to improve research practices, and to help reviewers and readers better contextualize a study's findings while understanding its limitations. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY: Stage 3.

Diagnostic accuracy of contrast-enhanced CT versus PET/CT for advanced ovarian cancer staging: a comparative systematic review and meta-analysis

PURPOSE

Accurate staging of ovarian cancer is critical to guide optimal management pathways. North American guidelines recommend contrast-enhanced CT as the primary work-up for staging ovarian cancer. This meta-analysis aims to compare the diagnostic accuracy of contrast-enhanced CT alone to PET/CT for detecting abdominal metastases in patients with a new or suspected diagnosis of ovarian cancer.

MATERIALS AND METHODS

A systematic review of MEDLINE, EMBASE, Scopus, the Cochrane Library, and the gray literature from inception to October 2022 was performed. Studies with a minimum of 5 patients evaluating the diagnostic accuracy of contrast-enhanced CT and/or PET/CT for detecting stage 3 ovarian cancer as defined by a surgical/histopathological reference standard ± clinical follow-up were included. Study, clinical, imaging, and accuracy data for eligible studies were independently acquired by two reviewers. Primary meta-analysis was performed in studies reporting accuracy on a per-patient basis using a bivariate mixed-effects regression model. Risk of bias was evaluated using QUADAS-2.

RESULTS

From 3701 citations, 15 studies (918 patients with mean age ranging from 51 to 65 years) were included in the systematic review. Twelve studies evaluated contrast-enhanced CT (6 using a per-patient assessment and 6 using a per-region assessment) and 11 studies evaluated PET/CT (7 using a per-patient assessment and 4 using a per-region assessment). All but one reporting study used consensus reading. Respective sensitivity and specificity values on a per-patient basis were 82% (67-91%, 95% CI) and 72% (59-82%) for contrast-enhanced CT and 87% (75-94%) and 90% (82-95%) for PET/CT. There was no significant difference in sensitivities between modalities (p = 0.29), but PET/CT was significantly more specific than CT (p < 0.01). Presumed variability could not be assessed in any single category due to limited studies using per-patient assessment. Studies were almost entirely low risk for bias and applicability concerns using QUADAS-2.

CONCLUSION

Contrast-enhanced CT demonstrates non-inferior sensitivity compared to PET/CT, although PET/CT may still serve as an alternative and/or supplement to CT alone prior to and/or in lieu of diagnostic laparoscopy in patients with ovarian cancer. Future revisions to existing guidelines should consider these results to further refine the individualized pretherapeutic diagnostic pathway.

Impact of Size Thresholds on the Diagnosis of Incidental Adrenal Lesions: A Systematic Review and Meta-Analysis

BACKGROUND

Preferred size-threshold recommendations for management of incidental adrenal lesions remain controversial.

PURPOSE

This meta-analysis aimed to compare the diagnostic accuracy of different size thresholds for detecting malignancy in patients with incidental adrenal lesions on imaging.

MATERIALS AND METHODS

A systematic review of MEDLINE, Embase, Scopus, the Cochrane Library, and the gray literature, covering the period from inception to September 2021, was performed. Studies with >10 patients evaluating the diagnostic accuracy of imaging size thresholds for detecting malignancy in patients with incidental adrenal lesions and no prior history of cancer were included. Study, clinical, imaging, and accuracy data for eligible studies were independently acquired by two reviewers. Primary meta-analysis was performed using a bivariate mixed-effects regression model. Risk of bias was evaluated using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool.

RESULTS

From 2,690 citations, 40 studies (9,794 patients with mean age ranging from 41 to 66 years) were included. Most (36 of 40) were retrospective single-center studies. CT with or without MRI served as the index test(s). Sensitivity and specificity values, respectively, by size threshold used in the included studies were as follows: 85% (95% confidence interval [CI] 74%-91%) and 39% (95% CI 23%-57%) for 3-cm thresholds; 85% (95% CI 78%-90%) and 75% (95% CI 62%-85%) for 4-cm thresholds; 70% (95% CI 56%-81%) and 74% (95% CI 59%-85%) for 5-cm thresholds; and 75% (95% CI 67%-82%) and 77% (95% CI 62%-87%) for 6-cm thresholds. No cause for variability in sensitivity or specificity was identified on subgroup analysis of the 4-cm threshold. Nearly half of the studies (19 of 40) had at least one QUADAS-2 domain with a high risk of bias.

CONCLUSIONS

A 4-cm size threshold demonstrates the highest combined sensitivity and specificity, with a preserved specificity compared with higher size thresholds, but with a trend toward improved sensitivity. Future research reevaluating 4-5 cm size thresholds while excluding characteristically benign lesions by imaging may help redefine a size threshold that has improved specificity but preserved sensitivity, compared with the existing 4-cm threshold.

Diagnostic accuracy of CT and MR features for detecting atypical lipomatous tumors and malignant liposarcomas: a systematic review and meta-analysis

OBJECTIVES

This systematic review and meta-analysis evaluated the diagnostic accuracy of CT and MRI for differentiating atypical lipomatous tumors and malignant liposarcomas from benign lipomatous lesions.

METHODS

MEDLINE, EMBASE, Scopus, the Cochrane Library, and the gray literature from inception to January 2022 were systematically evaluated. Original studies with > 5 patients evaluating the accuracy of CT and/or MRI for detecting liposarcomas with a histopathological reference standard were included. Meta-analysis was performed using a bivariate mixed-effects regression model. Risk of bias was evaluated using Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). This study is registered on PROSPERO, number CRD42022306479.

RESULTS

Twenty-six studies with a total of 2613 patients were included. Mean/median reported patient ages ranged between 50 and 63 years. The summary sensitivity and specificity of radiologist gestalt for detecting liposarcomas was 85% (79-90% 95% CI) and 63% (52-72%), respectively. Deep depth to fascia, thickened septations, enhancing components, and lesion size (≥ 10 cm) all demonstrated sensitivities ≥ 85%. Other imaging characteristics including heterogenous/amorphous signal intensity, irregular tumor margin, and nodules present demonstrated lower sensitivities ranging from 43 to 65%. Inter-reader reliability for radiologist gestalt within studies ranged from fair to substantial (k = 0.23-0.7). Risk of bias was predominantly mixed for patient selection, low for index test and reference standard, and unclear for flow and timing.

CONCLUSION

Higher sensitivities for detecting liposarcomas were achieved with radiologist gestalt, deep depth to fascia, thickened septations, enhancing components, and large size. Combined clinical and imaging scoring and/or radiomics both show promise for optimal performance, though require further analysis with prospective study designs.

CLINICAL RELEVANCE

This pooled analysis evaluates the accuracy of CT and MRI for detecting atypical lipomatous tumors and malignant liposarcomas. Radiologist gestalt, deep depth to fascia, thickened septations, enhancing components, and large size demonstrate the highest overall sensitivities.

KEY POINTS

• The summary sensitivity and specificity of radiologist gestalt for detecting liposarcomas was 85% (79-90% 95% CI) and 63% (52-72%), respectively. • Radiologist gestalt, deep depth to fascia, thickened septations, enhancing components, and large tumor size (≥ 10 cm) showed the highest sensitivities for detecting atypical lipomatous tumors/well-differentiated liposarcomas and malignant liposarcomas. • A combined clinical and imaging scoring system and/or radiomics is likely to provide the best overall diagnostic accuracy, although currently proposed scoring systems and radiomic feature analysis require further study with prospective study designs.

Comparative Performance of 2018 LI-RADS versus Modified LIRADS (mLI-RADS): An Individual Participant Data Meta-Analysis

BACKGROUND

LI-RADS version 2018 (v2018) is used for non-invasive diagnosis of hepatocellular carcinoma (HCC). A recently proposed modification (known as mLI-RADS) demonstrated improved sensitivity while maintaining specificity and positive predictive value (PPV) of LI-RADS category 5 (definite HCC) for HCC. However, mLI-RADS requires multicenter validation.

PURPOSE

To evaluate the performance of v2018 and mLI-RADS for liver lesions in a large, heterogeneous, multi-national cohort of patients at risk for HCC.

STUDY TYPE

Systematic review and meta-analysis using individual participant data (IPD) [Study Protocol: https://osf.io/duys4].

POPULATION

2223 observations from 1817 patients (includes all LI-RADS categories; females = 448, males = 1361, not reported = 8) at elevated risk for developing HCC (based on LI-RADS population criteria) from 12 retrospective studies.

FIELD STRENGTH/SEQUENCE

1.5T and 3T; complete liver MRI with gadoxetate disodium, including axial T2w images and dynamic axial fat-suppressed T1w images precontrast and in the arterial, portal venous, transitional, and hepatobiliary phases. Diffusion-weighted imaging was used when available.

ASSESSMENT

Liver observations were categorized using v2018 and mLI-RADS. The diagnostic performance of each system's category 5 (LR-5 and mLR-5) for HCC were compared.

STATISTICAL TESTS

The Quality Assessment of Diagnostic Accuracy Studies version 2 (QUADAS-2 was applied to determine risk of bias and applicability. Diagnostic performances were assessed using the likelihood ratio test for sensitivity and specificity and the Wald test for PPV. The significance level was P < 0.05.

RESULTS

17% (2/12) of the studies were considered low risk of bias (244 liver observations; 164 patients). When compared to v2018, mLR-5 demonstrated higher sensitivity (61.3% vs. 46.5%, P < 0.001), similar PPV (85.3% vs. 86.3%, P = 0.89), and similar specificity (85.8% vs. 90.8%, P = 0.16) for HCC.

DATA CONCLUSION

This study confirms mLR-5 has higher sensitivity than LR-5 for HCC identification, while maintaining similar PPV and specificity, validating the mLI-RADS proposal in a heterogeneous, international cohort.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 2.

Individual Participant Data Meta-Analysis of LR-5 in LI-RADS Version 2018 versus Revised LI-RADS for Hepatocellular Carcinoma Diagnosis

Background A simplification of the Liver Imaging Reporting and Data System (LI-RADS) version 2018 (v2018), revised LI-RADS (rLI-RADS), has been proposed for imaging-based diagnosis of hepatocellular carcinoma (HCC). Single-site data suggest that rLI-RADS category 5 (rLR-5) improves sensitivity while maintaining positive predictive value (PPV) of the LI-RADS v2018 category 5 (LR-5), which indicates definite HCC. Purpose To compare the diagnostic performance of LI-RADS v2018 and rLI-RADS in a multicenter data set of patients at risk for HCC by performing an individual patient data meta-analysis. Materials and Methods Multiple databases were searched for studies published from January 2014 to January 2022 that evaluated the diagnostic performance of any version of LI-RADS at CT or MRI for diagnosing HCC. An individual patient data meta-analysis method was applied to observations from the identified studies. Quality Assessment of Diagnostic Accuracy Studies version 2 was applied to determine study risk of bias. Observations were categorized according to major features and either LI-RADS v2018 or rLI-RADS assignments. Diagnostic accuracies of category 5 for each system were calculated using generalized linear mixed models and compared using the likelihood ratio test for sensitivity and the Wald test for PPV. Results Twenty-four studies, including 3840 patients and 4727 observations, were analyzed. The median observation size was 19 mm (IQR, 11-30 mm). rLR-5 showed higher sensitivity compared with LR-5 (70.6% [95% CI: 60.7, 78.9] vs 61.3% [95% CI: 45.9, 74.7]; P < .001), with similar PPV (90.7% vs 92.3%; P = .55). In studies with low risk of bias (n = 4; 1031 observations), rLR-5 also achieved a higher sensitivity than LR-5 (72.3% [95% CI: 63.9, 80.1] vs 66.9% [95% CI: 58.2, 74.5]; P = .02), with similar PPV (83.1% vs 88.7%; P = .47). Conclusion rLR-5 achieved a higher sensitivity for identifying HCC than LR-5 while maintaining a comparable PPV at 90% or more, matching the results presented in the original rLI-RADS study. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Sirlin and Chernyak in this issue.

Patients With "Gray Zone" PSA Levels: Application of Prostate MRI and MRS in the Diagnosis of Prostate Cancer

Improving the detection rates of prostate cancer (PCa) and avoiding unnecessary prostate biopsies in men with prostate-specific antigen (PSA) levels within the gray zone require urgent attention. In this context, rapid advances in MR technology in recent years may offer a promising possibility. A systematic review to evaluate the applications of magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) in detecting PCa and clinically significant PCa (csPCa) in men with PSA levels within the gray zone. The study type is defined as systematic review. In July 2022, out of 229 studies identified by the database search and from other sources, 23 articles related to the selected topic of interest were included in this review. No field strength or sequence restrictions. The data including the study population, study characteristics, as well as basic MRI characteristics, from the final studies included in this review, were extracted independently by two reviewers. The major results of the original study were summarized and no additional statistical analysis was performed. Among the 23 studies included in this review, 17 focused on the applications of MRS and MRI for the prebiopsy diagnosis of PCa. Nine of these 17 articles used Prostate Imaging Reporting and Data System (PI-RADS) score to interpret MRI results, thereby confirming the practicality of the PI-RADS score in predicting PCa and csPCa. The remaining six articles evaluated the applications of MRI and MRS in guiding prostate biopsy. Although there was a variation in the biopsy modalities used in these studies, both MRI- and MRS-guided prostate biopsies were observed to improve the detection rates of PCa and csPCa in patients with PSA levels within the gray zone. MRS and MRI showed good performance in the detection of PCa and csPCa before biopsy. In addition, MRS- or MRI-guided prostate-targeted biopsies were able to improve the detection rates of PCa and csPCa. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Proportion of malignancy in Bosniak classification of cystic renal masses version 2019 (v2019) classes: systematic review and meta-analysis

OBJECTIVES

Determine the proportion of malignancy within Bosniak v2019 classes.

METHODS

MEDLINE and EMBASE were searched. Eligible studies contained patients with cystic renal masses undergoing CT or MRI renal protocol examinations with pathology confirmation, applying Bosniak v2019. Proportion of malignancy was estimated within Bosniak v2019 class. Risk of bias was assessed using QUADAS-2.

RESULTS

We included 471 patients with 480 cystic renal masses. No class I malignant masses were observed. Pooled proportion of malignancy were class II, 12% (6/51, 95% CI 5-24%); class IIF, 46% (37/85, 95% CI 28-66%); class III, 79% (138/173, 95% CI 68-88%); and class IV, 84% (114/135, 95% CI 77-90%). Proportion of malignancy differed between Bosniak v2019 II-IV classes (p = 0.004). Four studies reported the proportion of malignancy by wall/septa feature. The pooled proportion of malignancy with 95% CI were class III thick smooth wall/septa, 77% (41/56, 95% CI 53-91%); class III obtuse protrusion ≤ 3 mm (irregularity), 83% (97/117, 95% CI 75-89%); and class IV nodule with acute angulation, 86% (50/58, 95% CI 75-93%) or obtuse angulation ≥ 4 mm, 83%, (64/77, 95% CI 73-90%). Subgroup analysis by wall/septa feature was limited by sample size; however, no differences were found comparing class III masses with irregularity to class IV masses (p = 0.74) or between class IV masses by acute versus obtuse angles (p = 0.62).

CONCLUSION

Preliminary data suggest Bosniak v2019 class IIF masses have higher proportion of malignancy compared to the original classification, controlling for pathologic reference standard. There are no differences in proportion of malignancy comparing class III masses with irregularities to class IV masses with acute or obtuse nodules.

KEY POINTS

• The proportion of malignancy in Bosniak v2019 class IIF cystic masses is 46% (37 malignant/85 total IIF masses, 95% confidence intervals (CI) 28-66%). • The proportion of malignancy in Bosniak v2019 class III cystic masses is 79% (138/173, 95% CI 68-88%) and in Bosniak v2019 class IV cystic masses is 84% (114/135, 95% CI 77-90%). • Class III cystic masses with irregularities had similar proportion of malignancy (83%, 97/117, 95% CI 75-89%) compared to Bosniak class IV masses (84%, 114/135, 95% CI 77-90%) overall (p = 0.74) with no difference within class IV masses by acute versus obtuse angulation (p = 0.62).

CT/MRI and CEUS LI-RADS Major Features Association with Hepatocellular Carcinoma: Individual Patient Data Meta-Analysis

Background The Liver Imaging Reporting and Data System (LI-RADS) assigns a risk category for hepatocellular carcinoma (HCC) to imaging observations. Establishing the contributions of major features can inform the diagnostic algorithm. Purpose To perform a systematic review and individual patient data meta-analysis to establish the probability of HCC for each LI-RADS major feature using CT/MRI and contrast-enhanced US (CEUS) LI-RADS in patients at high risk for HCC. Materials and Methods Multiple databases (MEDLINE, Embase, Cochrane Central Register of Controlled Trials, and Scopus) were searched for studies from January 2014 to September 2019 that evaluated the accuracy of CT, MRI, and CEUS for HCC detection using LI-RADS (CT/MRI LI-RADS, versions 2014, 2017, and 2018; CEUS LI-RADS, versions 2016 and 2017). Data were centralized. Clustering was addressed at the study and patient levels using mixed models. Adjusted odds ratios (ORs) with 95% CIs were determined for each major feature using multivariable stepwise logistic regression. Risk of bias was assessed using Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) (PROSPERO protocol: CRD42020164486). Results A total of 32 studies were included, with 1170 CT observations, 3341 MRI observations, and 853 CEUS observations. At multivariable analysis of CT/MRI LI-RADS, all major features were associated with HCC, except threshold growth (OR, 1.6; 95% CI: 0.7, 3.6; P = .07). Nonperipheral washout (OR, 13.2; 95% CI: 9.0, 19.2; P = .01) and nonrim arterial phase hyperenhancement (APHE) (OR, 10.3; 95% CI: 6.7, 15.6; P = .01) had stronger associations with HCC than enhancing capsule (OR, 2.4; 95% CI: 1.7, 3.5; P = .03). On CEUS images, APHE (OR, 7.3; 95% CI: 4.6, 11.5; P = .01), late and mild washout (OR, 4.1; 95% CI: 2.6, 6.6; P = .01), and size of at least 20 mm (OR, 1.6; 95% CI: 1.04, 2.5; P = .04) were associated with HCC. Twenty-five studies (78%) had high risk of bias due to reporting ambiguity or study design flaws. Conclusion Most Liver Imaging Reporting and Data System major features had different independent associations with hepatocellular carcinoma; for CT/MRI, arterial phase hyperenhancement and washout had the strongest associations, whereas threshold growth had no association. © RSNA, 2021 Online supplemental material is available for this article.

Evaluation of class II cystic renal masses proposed in Bosniak classification version 2019: a systematic review of supporting evidence

PURPOSE

The Bosniak classification of cystic renal masses version 2019 (v.2019) includes an expanded number of types of masses in class II; such masses are considered benign in clinical practice. Data supporting these additions have not been well-documented. We aim to determine the proportion of malignant or probably malignant renal masses among the types added to Bosniak v.2019 class II.

METHODS

Multiple databases were searched for studies evaluating the proportion of malignant or probably malignant renal masses among new Bosniak v.2019 class II types, four for CT and two for MRI. Risk of bias and applicability was assessed using the QUADAS-2 tool.

RESULTS

Ten studies (2068 renal masses) met inclusion criteria. Among the four added class II types at CT, the proportion of malignancy among (1) 'homogeneous hyperattenuating (≥ 70 HU) masses at unenhanced CT' was 0% (0/32) in three studies; (2) 'homogeneous masses - 9 to 20 HU at unenhanced CT' was 0% (0/1454) in two studies, and (3) 'homogeneous masses 21 to 30 HU at portal-venous phase CT' was 0% (0/454) in four studies. Masses that are homogeneous, have low attenuation, and are too small to characterize on CT had no supportive evidence. Among the two added class II types at MRI, the proportion of malignancy among (1) 'homogeneous masses markedly hyperintense at unenhanced T2-weighted MR imaging (similar to CSF) was 0% (0/72) in one study, and (2) 'homogeneous masses markedly hyperintense at T1-weighted MR imaging (~ 2.5 × renal parenchyma signal intensity)' was 0% (0/32) and 5% (2/37) in two studies. Nine studies were at risk of bias within at least one QUADAS-2 domain.

CONCLUSION

The addition of six types of cystic renal masses to Class II in the Bosniak v.2019 proposal may be justified but based on limited evidence, with no evidence for 'homogeneous low attenuation masses that are too small to characterize' on CT, and thus considering them benign is in part based on expert opinion. Protocol Registration: PROSERO CRD42020196408.

Diagnostic Accuracy of MRI for the Detection of Malignant Peripheral Nerve Sheath Tumors: A Systematic Review and Meta-Analysis

OBJECTIVE. This systematic review and meta-analysis evaluates the diagnostic accuracy of MRI for differentiating malignant (MPNSTs) from benign peripheral nerve sheath tumors (BPNSTs). MATERIALS AND METHODS. A systematic review of MEDLINE, Embase, Scopus, the Cochrane Library, and the gray literature from inception to December 2019 was performed. Original articles that involved at least 10 patients and that evaluated the accuracy of MRI for detecting MPNSTs were included. Two reviewers independently extracted clinical and radiologic data from included articles to calculate sensitivity, specificity, PPV, NPV, and accuracy. A meta-analysis was performed using a bivariate mixed-effects regression model. Risk of bias was evaluated using QUADAS-2. RESULTS. Fifteen studies involving 798 lesions (252 MPNSTs and 546 BPNSTs) were included in the analysis. Pooled and weighted sensitivity, specificity, and AUC values for MRI in detecting MPNSTs were 68% (95% CI, 52-80%), 93% (95% CI, 85-97%), and 0.89 (95% CI, 0.86-0.92) when using feature combination and 88% (95% CI, 74-95%), 94% (95% CI, 89-96%), and 0.97 (95% CI, 0.95-0.98) using diffusion restriction with or without feature combination. Subgroup analysis, such as patients with neurofibromatosis type 1 (NF1) versus those without NF1, could not be performed because of insufficient data. Risk of bias was predominantly high or unclear for patient selection, mixed for index test, low for reference standard, and unclear for flow and timing. CONCLUSION. Combining features such as diffusion restriction optimizes the diagnostic accuracy of MRI for detecting MPNSTs. However, limitations in the literature, including variability and risk of bias, necessitate additional methodologically rigorous studies to allow subgroup analysis and further evaluate the combination of clinical and MRI features for MPNST diagnosis.

Diagnostic Accuracy of MRI for Differentiation of Benign and Malignant Pancreatic Cystic Lesions Compared to CT and Endoscopic Ultrasound: Systematic Review and Meta-analysis

BACKGROUND

Differentiation of benign and malignant pancreatic cystic lesions on MRI, computed tomography (CT), and endoscopic ultrasound (EUS) is critical for determining management.

PURPOSE

To perform a systematic review evaluating the diagnostic accuracy of MRI for diagnosing malignant pancreatic cystic lesions, and to compare the accuracy of MRI to CT and EUS.

STUDY TYPE

Systematic review and meta-analysis.

DATA SOURCES

MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, Web of Science, and Scopus were searched until February 2020 for studies reporting MRI accuracy for assessing pancreatic cystic lesions.

FIELD STRENGTH

1.5T or 3.0T.

ASSESSMENT

Methodologic and outcome data were extracted by two reviewers (AU and MA, 2 years of experience each). All studies of pancreatic cystic lesions on MRI were identified. Studies with incomplete MRI technique were excluded. Risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS)-2 tool.

STATISTICAL TESTS

Sensitivity/specificity was pooled using bivariate random-effects meta-analysis with 95% confidence intervals (95%CI). Pairwise-comparison of MRI to CT and EUS was performed. The impact of gadolinium-based contrast agents, mucinous lesions, and risk of bias were explored using meta-regression.

RESULTS

MRI pooled sensitivity was 76% (95%CI 67% to 84%) and specificity was 80% (95%CI 74% to 85%) for distinguishing benign and malignant lesions. MRI and CT had similar sensitivity (P = 0.822) and specificity (P = 0.096), but MRI was more specific than EUS (80% vs. 75%, P < 0.05). Studies including only contrast-enhanced MRI were more sensitive than those including unenhanced exams (P < 0.05). MRI sensitivity and specificity did not differ for mucinous lesions (P = 0.537 and P = 0.384, respectively) or for studies at risk of bias (P = 0.789 and P = 0.791, respectively).

DATA CONCLUSION

MRI and CT demonstrate comparable accuracy for diagnosing malignant pancreatic cystic lesions. EUS is less specific than MRI, which suggests that, in some cases, management should be guided by MRI findings rather than EUS.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2.

Completeness of reporting for systematic reviews of point-of-care ultrasound: a meta-research study

OBJECTIVES

Systematic reviews are often considered among the highest quality of evidence. Completely reported systematic reviews, however, are required so readers can assess for generalisability of the research to practice and risk of bias. The objective of this study was to assess the completeness of reporting for systematic reviews assessing the diagnostic accuracy of point-of-care ultrasound (POCUS) using the Preferred Reporting Items for Systematic Reviews and Meta-analyses for Diagnostic Test Accuracy (PRISMA-DTA) checklist that was published in 2018.

DESIGN AND SETTING

In this meta-research study, MEDLINE, EMBASE and Cochrane Library databases were searched, with no date restriction, on March 1st, 2020 for systematic reviews assessing the diagnostic accuracy of POCUS. Adherence to PRISMA-DTA for the main text and abstract was scored independently and in duplicate using a modified checklist. Prespecified subgroup analyses were performed.

MAIN OUTCOME MEASURES

The primary outcome was the mean PRISMA-DTA checklist adherence for the full-text and abstract.

RESULTS

A total of 71 studies published from 2008 to 2020 met the inclusion criteria. The overall adherence for the full-text was moderate: 19.8 out of 26.0 items (76%) and for the abstract was 7.0 out of 11.0 items (64%). Although many items in the PRISMA-DTA checklist were frequently reported, several were r infrequently reported (<33% of studies), including item 5 (protocol registration), item D2 (minimally acceptable test accuracy) and item 14 (variability in target condition, index test and reference standards). Subgroup analyses showed a higher PRISMA-DTA mean adherence (SD) for high impact journals (20.9 (2.52) vs 18.9 (1.95); p<0.001), studies including supplemental materials (20.6 (2.48) vs 18.9 (2.28); p=0.004), studies citing adherence to PRISMA reporting guidelines (20.4 (1.95) vs 19.0 (3.00); p=0.038) and studies published in journals endorsing PRISMA guidelines (20.2 (2.47) vs 18.6 (2.37); p=0.025). There was variable adherence based on journal of publication (p=0.006), but not for study population (adult vs paediatric vs mixed) (p=0.62), year of publication (p=0.94), body region (p=0.78) or country (p=0.40). There was no variability in abstract adherence based on whether the abstract was structured with subheadings or not (p=0.25). A Spearman's correlation found moderate correlation between higher word counts and abstractand full-text adherence (R=0.45, p<0.001 and R=0.38, p=0.001), respectively.

CONCLUSIONS

Overall, the reporting of POCUS diagnostic accuracy systematic reviews and meta-analyses was moderate. We identified deficits in several key areas including the preregistration of systematic reviews in an online repository, handling of multiple definitions of target conditions, index tests and reference standards and specifying minimally acceptable test accuracy. Prospective registration of reviews and detailed reporting as per PRISMA-DTA during the research process could improve reporting completeness. At an editorial level, word count and supplemental material limitations may impede reporting completeness, whereas endorsement of reporting guidelines on journal websites could improve reporting.

Accuracy of liver and spleen stiffness on magnetic resonance elastography for detecting portal hypertension: a systematic review and meta-analysis

INTRODUCTION

The purpose of this systematic review and meta-analysis was to evaluate the diagnostic accuracy of liver and spleen stiffness on magnetic resonance elastography (MRE) for detecting clinically significant portal hypertension.

METHODS

A systematic review of MEDLINE, EMBASE, Scopus, the Cochrane Library, and the Grey Literature through to 15 August 2019 was performed. Original articles with >10 patients evaluating liver and/or spleen stiffness on MRE using a reference standard of portal hypertension defined as intractable ascites, esophageal varices, encephalopathy and/or death were included in analysis. Patient, clinical, MRI, and diagnostic performance was independently acquired by two reviewers. Meta-analysis was performed using a bivariate mixed-effects regression model.

RESULTS

Fourteen studies were included with 12 studies evaluating liver stiffness and eight studies evaluating spleen stiffness. The pooled and weighted sensitivity, specificity, and area under the curve (AUC) values for liver stiffness on MRE were 83% [95% confidence interval (CI) 72-90%], 80% (95% CI 70-88%), and 88% (95% CI 85-91%), respectively. The pooled and weighted sensitivity, specificity, and AUC values for spleen stiffness on MRE were 79% (95% CI 61-90%), 90% (95% CI 80-95%), and 92% (95% CI 89-94%), respectively. The liver and spleen stiffness sensitivity and specificity values were comparable when evaluating for esophageal varices only at of 80% (95% CI 66-89%) and 76% (95% CI 62-86%) for liver stiffness, and 75% (95% CI 52-90%) and 89% (95% CI 70-96%) for spleen stiffness.

DISCUSSION

Liver and spleen stiffness on MRE can serve as a supplemental noninvasive assessment tools for detecting clinically significant portal hypertension. Spleen stiffness may be more specific and accurate than liver stiffness for detecting portal hypertension.

Myelin quantification with MRI: A systematic review of accuracy and reproducibility

OBJECTIVES

Currently, multiple sclerosis is treated with anti-inflammatory therapies, but these treatments lack efficacy in progressive disease. New treatment strategies aim to repair myelin damage and efficacy evaluation of such new therapies would benefit from validated myelin imaging techniques. Several MRI methods for quantification of myelin density are available now. This systematic review aims to analyse the performance of these MRI methods.

METHODS

Studies comparing myelin quantification by MRI with histology, the current gold standard, or assessing reproducibility were retrieved from PubMed/MEDLINE and Embase (until December 2019). Included studies assessed both myelin histology and MRI quantitatively. Correlation or variance measurements were extracted from the studies. Non-parametric tests were used to analyse differences in study methodologies.

RESULTS

The search yielded 1348 unique articles. Twenty-two animal studies and 13 human studies correlated myelin MRI with histology. Eighteen clinical studies analysed the reproducibility. Overall bias risk was low or unclear. All MRI methods performed comparably, with a mean correlation between MRI and histology of R²=0.54 (SD=0.30) for animal studies, and R²=0.54 (SD=0.18) for human studies. Reproducibility for the MRI methods was good (ICC=0.75-0.93, R²=0.90-0.98, COV=1.3-27%), except for MTR (ICC=0.05-0.51).

CONCLUSIONS

Overall, MRI-based myelin imaging methods show a fairly good correlation with histology and a good reproducibility. However, the amount of validation data is too limited and the variability in performance between studies is too large to select the optimal MRI method for myelin quantification yet.

Diagnostic accuracy of dual-energy CT for the detection of bone marrow edema in the appendicular skeleton: a systematic review and meta-analysis

OBJECTIVES

This meta-analysis evaluated the diagnostic accuracy of dual-energy CT (DECT) for detecting bone marrow edema (BME) in the appendicular skeleton.

METHODS

A systematic review of MEDLINE, EMBASE, Scopus, the Cochrane Library, and gray literature from inception through January 31, 2020, was performed. Original articles with > 10 patients evaluating the accuracy of DECT for detecting BME in the appendicular skeleton with a reference standard of MRI and/or clinical follow-up were included. Study details were independently extracted by two reviewers. Meta-analysis was performed using a bivariate random-effects model with multivariable meta-regression. Risk of bias (RoB) was evaluated with QUADAS-2.

RESULTS

Twenty studies evaluating 790 patients for BME in the appendicular skeleton were included in analysis. The summary sensitivity, specificity, and AUC values for BME in the appendicular skeleton were 86% (95% confidence interval [CI] 82-89%), 93% (95% CI 90-95%), and 0.95, respectively. Quantitative analysis had a higher sensitivity than qualitative analysis on meta-regression (p = 0.01), but no difference in specificity (p = 0.28). No other covariates contributed to variability in accuracy (etiology (trauma vs non-trauma); location (upper vs lower extremity); and RoB). Studies demonstrated generally low or unclear RoB and applicability. Eight studies included index tests with high RoB when quantitative assessments used a retrospective cut-off value.

CONCLUSIONS

DECT demonstrates a higher specificity than sensitivity and AUC > 0.9. In scenarios where MRI availability is limited or contraindicated, DECT could be an alternative to MRI for detecting BME in the appendicular skeleton. However, limitations in sources of variability and RoB warrant continued study.

KEY POINTS

• Twenty studies evaluating 790 patients for bone marrow edema in the appendicular skeleton with dual-energy CT were included in analysis. • The summary sensitivity, specificity, and AUC values for detecting bone marrow in the appendicular skeleton were 86% (95% confidence interval [CI] 82-89%), 93% (95% CI 90-95%), and 0.95, respectively. • In scenarios where MRI availability is limited or is contraindicated, DECT could be an alternative to MRI for detecting bone marrow edema in the appendicular skeleton.

Diagnostic Accuracy of Limited MRI Protocols for Detecting Radiographically Occult Hip Fractures: A Systematic Review and Meta-Analysis

OBJECTIVE. This systematic review and meta-analysis aimed to evaluate the diagnostic accuracy of limited MRI protocols for detecting radiographically occult proximal femoral fractures. MATERIALS AND METHODS. A systematic review of MEDLINE, Embase, Scopus, the Cochrane Library, and the gray literature through November 15, 2019, was performed. Original articles with 10 or more patients evaluating limited MRI protocols for the diagnosis of radiographically occult proximal femoral fractures compared with multiparametric MRI with or without clinical outcome as the reference standard were included in the analysis. Patient, clinical, MRI, and performance parameters were independently acquired by two reviewers. Meta-analysis was performed using a bivariate mixed-effects regression model. RESULTS. Eleven studies with 938 patients and 247 proximal femoral fractures met inclusion criteria, and five of these studies were included in the meta-analysis. The pooled and weighted summary sensitivity and specificity and the area under the summary ROC curve for limited MRI protocols in detecting radiographically occult hip fractures were 99% (95% CI, 91-100%), 99% (95% CI, 97-100%), and 1 (95% CI, 0.99-1), respectively. The aggregate sensitivity and specificity values for a single-plane T1-weighted sequence only, STIR sequence only, T1-weighted and STIR sequences, and T2-weighted sequence only were as follows: 97% (89/92) and 100% (76/76), 99% (126/127) and 99% (865/873), 100% (118/118) and 99% (867/874), and 86% (51/59) and 97% (137/141), respectively. Sensitivity was 100% (58/58) when images were acquired on 3-T scanners only and 99% (284/288) when interpreted only by certified radiologists. The mean scanning time for the limited MRI protocols was less than 5 minutes. CONCLUSION. Limited MRI protocols can be used as the standard of care in patients with a suspected but radiographically occult hip fracture. A protocol composed of coronal T1-weighted and STIR sequences is 100% sensitive.

Diagnostic accuracy of 99mTc-sestamibi SPECT/CT for detecting renal oncocytomas and other benign renal lesions: a systematic review and meta-analysis

PURPOSE

The primary objectives of this systematic review and meta-analysis were to evaluate the diagnostic accuracy of 99mTc-sestamibi SPECT/CT for detecting renal oncocytoma versus (1) all other renal lesions and (2) chromophobe renal cell carcinoma (ChrRCC) alone.

METHODS

A systematic review of MEDLINE, EMBASE, Scopus, the Cochrane Library, and the Gray Literature was performed. Original articles with > 5 patients evaluating oncocytomas versus other renal lesions with SPECT/CT using a pathological reference standard were included. Patient, clinical, imaging, and performance parameters were independently acquired by two reviewers. Meta-analysis was performed using a bivariate mixed-effects regression model.

RESULTS

Four articles with a total of 117 renal lesions were included in analysis. The pooled and weighted sensitivity and specificity values of 99mTc-sestamibi SPECT/CT for detecting (1) renal oncocytoma versus other renal lesions were 92% (95% CI 72-98%) and 88% (95% CI 79-94%), respectively, and (2) 89% and 67%, respectively, for renal oncocytoma versus ChrRCC. The specificity for the detecting the oncocytoma-ChrRCC spectrum was 96% (95% CI 84-99%). The sensitivity and specificity for detecting benign versus malignant renal lesions were 86% (95% CI 66-95%) and 90% (95% CI 80-95%), and 88% and 95% when HOCTs were characterized as benign. All reporting studies used a cut-off tumor-to-background renal parenchyma radiotracer uptake ratio of > 0.6 for positive studies.

CONCLUSION

99mTc-sestamibi SPECT/CT demonstrates a high sensitivity and specificity for characterizing benign and low-grade renal lesions. This test can help improve the diagnostic confidence for patients with indeterminate renal masses being considered for active surveillance.

Safety of Intrathecal Administration of Gadolinium-based Contrast Agents: A Systematic Review and Meta-Analysis

Background The use of MR cisternography with intrathecal administration of gadolinium-based contrast agents (GBCAs) is limited by a lack of understanding of the relationship between intrathecal GBCA exposure and dose-related adverse events. Purpose To perform a systematic review to establish an understanding of the dose-response relationship of intrathecal GBCAs and to characterize related adverse events, particularly at higher doses. Materials and Methods Medline, Embase, CINAHL, and Central databases were searched for studies reporting intrathecal GBCA use. Data extraction included studies focused on rates and types of adverse events after intrathecal GBCA exposure. A two-tailed independent sample t test statistic was used to evaluate the relationship between GBCA dose and the presence of serious versus nonserious adverse events. Meta-analysis was used to determine the overall incidence of adverse events. Study quality and publication bias were assessed using the modified Newcastle-Ottawa scale and a funnel plot (effect size measured using Hedges' g followed by the Egger test), respectively. Results Fifty-three studies with a total of 1036 patients were included for analysis. The overall rate of adverse events after intrathecal administration of GBCA was 13% (95% confidence interval [CI]: 9.3%, 18%). Meta-analysis revealed moderate heterogeneity (I² = 62%). Serious adverse event rates could not be determined with meta-analysis. They were reported in 10 studies and were primarily neurologic in nature, with two cases of coma-one resulting in death. Serious adverse events were associated with significantly higher GBCA doses when compared with nonserious adverse events (mean difference, 4.5 mmol; 95% CI: 2.3 mmol, 6.6 mmol; P = .008). For serious adverse events, there was no clear dose-dependent increase in severity above 2.0 mmol. Conclusion Overall, intrathecal administration of GBCAs at doses greater than 1.0 mmol are associated with serious neurotoxic complications with relative clinical safety at lower doses. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Kanal in this issue.

Digital breast tomosynthesis for breast cancer detection: a diagnostic test accuracy systematic review and meta-analysis

OBJECTIVES

No consensus exists on digital breast tomosynthesis (DBT) utilization for breast cancer detection. We performed a diagnostic test accuracy systematic review and meta-analysis comparing DBT, combined DBT and digital mammography (DM), and DM alone for breast cancer detection in average-risk women.

METHODS

MEDLINE and EMBASE were searched until September 2018. Comparative design studies reporting on the diagnostic accuracy of DBT and/or DM for breast cancer detection were included. Demographic, methodologic, and diagnostic accuracy data were extracted. Risk of bias was assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS)-2 tool. Accuracy metrics were pooled using bivariate random-effects meta-analysis. The impact of multiple covariates was assessed using meta-regression.

PROSPERO ID

CRD 42018111287.

RESULTS

Thirty-eight studies reporting on 488,099 patients (13,923 with breast cancer) were included. Eleven studies were at low risk of bias. DBT alone, combined DBT and DM, and DM alone demonstrated sensitivities of 88% (95% confidence interval [CI] 83-92), 88% (CI 83-92), and 79% (CI 75-82), as well as specificities of 84% (CI 76-89), 81% (CI 73-88), and 79% (CI 71-85), respectively. The greater sensitivities of DBT alone and combined DBT and DM compared to DM alone were preserved in the combined meta-regression models accounting for other covariates (p = 0.003-0.006). No significant difference in diagnostic accuracy between DBT alone and combined DBT and DM was identified (p = 0.175-0.581).

CONCLUSIONS

DBT is more sensitive than DM, while the addition of DM to DBT provides no additional diagnostic benefit. Consideration of these findings in breast cancer imaging guidelines is recommended.

KEY POINTS

• Digital breast tomosynthesis with or without additional digital mammography is more sensitive in detecting breast cancer than digital mammography alone in women at average risk for breast cancer. • The addition of digital mammography to digital breast tomosynthesis provides no additional diagnostic benefit in detecting breast cancer compared to digital breast tomosynthesis alone. • The specificity of digital breast tomosynthesis with or without additional digital mammography is no different than digital mammography alone in the detection of breast cancer.

MRI-Derived Biomarkers Related to Sarcopenia: A Systematic Review

BACKGROUND

MRI allows quantitatively assessing muscle quantity and quality.

PURPOSE

To summarize the role of MRI as a noninvasive technique for the identification of in vivo surrogate biomarker of sarcopenia.

STUDY TYPE

Systematic review.

POPULATION

In April 2019, a systematic literature search (Medline/EMBASE) was performed to identify articles on the topic at issue.

FIELD STRENGTH/SEQUENCE

No field strength or sequence restrictions.

ASSESSMENT

After a literature search, study design, aim, sample size, demographics, magnetic field strength, imaged body region, MRI sequences, and imaging biomarker were extracted.

STATISTICAL TESTS

Data are presented as frequencies and percentages.

RESULTS

From 69 records identified through search query, 18 articles matched the inclusion criteria. All articles were published from 2012 and had a mainly prospective design (14/18, 78%). Sample size ranged from 9 to 284 subjects, for a total of 1706 enrolled subjects. Healthy subjects were enrolled or retrospectively selected in 8/18 (44%) articles, corresponding to 658 (39%) healthy subjects. Magnetic field strength was 1.5 or 3T in 14/18 (78%) studies. The most analyzed body regions were the thigh (7/18, 39%) and the trunk (6/18, 33%). Stratifying studies according to their aim, 13/18 (72%) studies focused on muscle quality and quantity, 3/18 (17%) studies on outcome prediction, and 2/18 articles (11%) addressed both aims. A wide set of MRI biomarkers have been proposed. Muscle cross-sectional area was the most used for muscle quantity estimation, while quantitative biomarkers of muscle fat content or fiber architecture were proposed to assess muscle quality.

DATA CONCLUSION

The proposed biomarkers were assessed using different MRI sequences for different body regions in different subjects/patient cohorts, pointing out a lack of standardization on this topic. Future studies should test and compare the performance of proposed MRI biomarkers for sarcopenia characterization and quantification using a standardized experimental setup.

LEVEL OF EVIDENCE

1 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2020;51:1117-1127.