Diagnostic Accuracy Studies in Radiology: How to Recognize and Address Potential Sources of Bias

Use of radiologic imaging to differentiate lipoma from atypical lipomatous tumor/well-differentiated liposarcoma: Systematic review

Background

Lipomas and atypical lipomatous tumors or well-differentiated liposarcomas (ALTs/WDLs), pose a diagnostic challenge due to their overlapping clinical and imaging features. Accurate differentiation is crucial as treatment strategies differ significantly between benign lipomas and malignant ALTs/WDLs. In recent years, medical imaging techniques have shown promise in distinguishing lipomas from ALTs/WDLs by providing enhanced visualization and assessment of various imaging parameters.

Objective

This systematic review aimed to investigate the use of magnetic resonance (MR) imaging and computed tomography (CT) scan to differentiate lipomas from ALTs/WDLs.

Methods

A systematic review was conducted by using MEDLINE, PubMed, PubMed Central, Cochrane Library, Google Scholar, and clinical trail.gov to identify imaging studies published between 2001 and 2022. Two independent reviewers reviewed 221 record to scrutinize the studies. The methodological quality of each included studies was assessed the using Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool.

Results

Thirteen retrospective cohort studies included 1,390 of total patients. Among them, 11 studies used MR imaging, 2 studies used CT scan and MR imaging both to differentiate lipoma from ALTs/WDLs. The significant diagnostic variables identified in the included studies were age, size, texture, mean intensity, contrast enhancement, location, septation, and nodularity. The overall, sensitivity, specificity, and accuracy of the included studies for diagnosis of lesions range from 66% to 100%, 37% to 100%, and 76% to 95%, respectively. The positive and negative predictive values range from 46.9% to 90% and 86% to 100%, respectively.

Conclusion

The most frequent diagnostic features of ALTs/ WDLs include tumors ⩾110 mm in size, often in patients over 60, predominantly in the lower extremities, with an irregular shape, incomplete fat suppression, contrast enhancement, nodularity, septation >2 mm, and predictive markers such as lactate dehydrogenase >220 and a short tau inversion recovery-signal intensity ratio >1.18.

Performance of an ultra-fast deep-learning accelerated MRI screening protocol for prostate cancer compared to a standard multiparametric protocol

OBJECTIVES

To establish and evaluate an ultra-fast MRI screening protocol for prostate cancer (PCa) in comparison to the standard multiparametric (mp) protocol, reducing scan time and maintaining adequate diagnostic performance.

MATERIALS AND METHODS

This prospective single-center study included consecutive biopsy-naïve patients with suspected PCa between December 2022 and March 2023. A PI-RADSv2.1 conform mpMRI protocol was acquired in a 3 T scanner (scan time: 25 min 45 sec). In addition, two deep-learning (DL) accelerated sequences (T2- and diffusion-weighted) were acquired, serving as a screening protocol (scan time: 3 min 28 sec). Two readers evaluated image quality and the probability of PCa regarding PI-RADSv2.1 scores in two sessions. The diagnostic performance of the screening protocol with mpMRI serving as the reference standard was derived. Inter- and intra-reader agreements were evaluated using weighted kappa statistics.

RESULTS

We included 77 patients with 97 lesions (mean age: 66 years; SD: 7.7). Diagnostic performance of the screening protocol was excellent with a sensitivity and specificity of 100%/100% and 89%/98% (cut-off ≥ PI-RADS 4) for reader 1 (R1) and reader 2 (R2), respectively. Mean image quality was 3.96 (R1) and 4.35 (R2) for the standard protocol vs. 4.74 and 4.57 for the screening protocol (p < 0.05). Inter-reader agreement was moderate (κ: 0.55) for the screening protocol and substantial (κ: 0.61) for the multiparametric protocol.

CONCLUSION

The ultra-fast screening protocol showed similar diagnostic performance and better imaging quality compared to the mpMRI in under 15% of scan time, improving efficacy and enabling the implementation of screening protocols in clinical routine.

CLINICAL RELEVANCE STATEMENT

The ultra-fast protocol enables examinations without contrast administration, drastically reducing scan time to 3.5 min with similar diagnostic performance and better imaging quality. This facilitates patient-friendly, efficient examinations and addresses the conflict of increasing demand for examinations at currently exhausted capacities.

KEY POINTS

Time-consuming MRI protocols are in conflict with an expected increase in examinations required for prostate cancer screening. An ultra-fast MRI protocol shows similar performance and better image quality compared to the standard protocol. Deep-learning acceleration facilitates efficient and patient-friendly examinations, thus improving prostate cancer screening capacity.

Risk of Bias in Liver Imaging Reporting and Data System Studies Using QUADAS-2

Purpose: Use a tailored version of the Quality Assessment of Diagnostic Accuracy Studies tool to evaluate risk of bias and applicability across LIRADS related publications. Method: A tailored QUADAS-2 tool was created through consensus approach to assess risk of bias and applicability across 37 LI-RADS related publications. Studies were selected from 2017 to 2022 using the assistance of experienced hospital librarians to search for studies evaluating the diagnostic accuracy of CT, MRI, or contrast-enhanced ultrasound for HCC using LI-RADS through multiple different databases. QUADAS-2 assessments were performed in duplicate and independently by 2 authors with experience using the QUADAS-2 tool. Disagreements were resolved with a third expert reviewer. Consensus QUADAS-2 assessments were tabulated for each domain. Results: Using the tailored QUADAS-2 tool, 31 of the 37 included LI-RADS studies were assessed as high risk of bias, and 9 out of 37 studies demonstrated concerns for applicability. Patient selection (21 out of 37 studies) and flow/timing (24 out of 37 studies) domains demonstrated the highest risk of bias. 6 out of 37 studies in the index domain demonstrated high risk of bias. 2 out of 37 studies showed high risk of bias in the reference standard domain. Conclusion: A significant proportion of LI-RADS research is at risk of bias with concerns for applicability. Identifying risk of bias in such research is essential to recognize limitations of a study that may affect the validity of the results. Areas for improvement in LI-RADS research include reducing selection bias, avoiding inappropriate exclusions, and decreasing verification bias.

Validation of the Appendicitis Inflammatory Response (AIR) score during pregnancy: A nested case-control study

OBJECTIVES

Appendicitis poses diagnostic challenges. A correct diagnosis is important during pregnancy to avoid unnecessary surgery on the one hand and delayed surgery on the other hand, as both may negatively affect pregnancy outcomes. Clinical scores for risk-stratified management of suspected appendicitis are well established in adults but have not been validated during pregnancy. This nested case-control study evaluated the diagnostic accuracy of the Appendicitis Inflammatory Response (AIR) score and imaging during pregnancy.

METHODS

By cross-linking national Swedish health registries from a defined geographical area, we identified a cohort of 154 women who underwent appendectomy for suspected appendicitis during pregnancy and a matched cohort of 232 pregnant women admitted for acute abdominal pain and suspected appendicitis but with a discharge diagnosis of nonspecific abdominal pain (NSAP). All variables were extracted from medical records. The diagnostic value of AIR score and imaging was estimated for patients with a final diagnosis of appendicitis compared with patients with negative appendectomy and NSAP patients.

RESULTS

The final diagnoses for the operated patients were uncomplicated and complicated appendicitis in 49.4% and 26.6%, respectively, and negative appendectomy in 24.0%. Nearly half of all the patients underwent diagnostic imaging (41%), mainly by ultrasonography. The sensitivity and specificity of diagnostic imaging were 44.9% (95% CI 32.9%-57.4%) and 42.2% (95% CI 31.9%-53.1%), respectively. The area under the receiver operating characteristic curve of AIR score was 0.88 (95% CI 0.84-0.92) for all appendicitis and 0.90 (95% CI 0.84-0.95) for complicated appendicitis. The sensitivity for complicated appendicitis was 100% at a score of ≥4. The specificity for all appendicitis was 97% at a score of ≥9.

CONCLUSIONS

The results of this study suggest that the AIR score may be a suitable diagnostic tool for risk stratification of pregnant women with abdominal pain and suspected appendicitis but further validation among pregnant women is needed.

Diagnosis and staging of small intestinal neuroendocrine tumors with CT enterography and PET with Gallium-68: preoperative risk stratification protocol

PURPOSE

Pre-operative diagnosis and staging of small intestine neuroendocrine tumors (SI-NETs) remain sub-optimal, with open palpation during surgery still considered the gold standard. This limits a standardized implementation of minimally invasive surgery (MIS). The aim of this single-center retrospective study was to assess a tailored diagnostic work-up to identify candidates at low risk of undetected disease who may benefit from MIS.

METHODS

Patients diagnosed with SI-NETs between 2013 and 2022 who underwent contrast-enhanced computed tomography enterography (CTE) and Ga68-DOTATOC-positron emission tomography-CT (68 Ga DOTATATE PET/CT) preoperatively and subsequently underwent open surgical resection were included. Imaging studies were reassessed by two radiologists. Combined use of CTE and 68 Ga DOTATATE PET/CT in determining primary lesion disease burden (number of lesions) and LN disease stage (distal and proximal relative to superior mesenteric vessels) was assessed, using surgical reports and pathology as gold standard.

RESULTS

Overall, 56 patients were included. Sensitivity of CTE and 68 Ga DOTATATE PET/CT for at least one primary SI-NET was 100% and 94%, respectively. In the presence of concordance between studies, combined use of CTE and 68 Ga DOTATATE PET/CT for detection of single primary tumors improved specificity to 89% (n = 25/28) with a positive predictive value of 87.5% (n = 21/24). Distal LN disease was identified in 89.2% of cases (n = 33/37). The association of single lesion and distal LN disease was found pre-operatively in 32% of patients (n = 18).

CONCLUSION

Combined use of CTE and 68 Ga DOTATATE PET/CT enables identifying low-risk surgical candidates (single SI-NET lesions with distal LN disease).