Use of Artificial Intelligence in Radiology: Impact on Pediatric Patients, a White Paper From the ACR Pediatric AI Workgroup.

In this white paper, the ACR Pediatric AI Workgroup of the Commission on Informatics educates the radiology community about the health equity issue of the lack of pediatric artificial intelligence (AI), improves the understanding of relevant pediatric AI issues, and offers solutions to address the inadequacies in pediatric AI development. In short, the design, training, validation, and safe implementation of AI in children require careful and specific approaches that can be distinct from those used for adults. On the eve of widespread use of AI in imaging practice, the group invites the radiology community to align and join Image IntelliGently (www.imageintelligently.org) to ensure that the use of AI is safe, reliable, and effective for children.

Assessment of fluid removal using ultrasound, bioimpedance and anthropometry in pediatric dialysis: a pilot study.

BACKGROUND: Fluid overload is associated with morbidity and mortality in children receiving dialysis. Accurate clinical assessment is difficult, and using deuterium oxide (D2O) to measure total body water (TBW) is impractical. We investigated the use of ultrasound (US), bioimpedance spectroscopy (BIS), and anthropometry to assess fluid removal in children receiving maintenance hemodialysis (HD). METHODS: Participants completed US, BIS, and anthropometry immediately before and 1-2 h after HD for up to five sessions. US measured inferior vena cava (IVC) diameter, lung B-lines, muscle elastography, and dermal thickness. BIS measured the volume of extracellular (ECF) and intracellular (ICF) fluid. Anthropometry included mid-upper arm, calf and ankle circumferences, and triceps skinfold thickness. D2O was performed once pre-HD. We assessed the change in study measures pre- versus post-HD, and the correlation of change in study measures with percent change in body weight (%∆BW). We also assessed the agreement between TBW measured by BIS and D2O. RESULTS: Eight participants aged 3.4-18.5 years were enrolled. Comparison of pre- and post-HD measures showed significant decrease in IVC diameters, lung B-lines, dermal thickness, BIS %ECF, mid-upper arm circumference, ankle, and calf circumference. Repeated measures correlation showed significant relationships between %∆BW and changes in BIS ECF (rrm =0.51, 95% CI 0.04, 0.80) and calf circumference (rrm=0.80, 95% CI 0.51, 0.92). BIS TBW correlated with D2O TBW but overestimated TBW by 2.2 L (95% LOA, -4.75 to 0.42). CONCLUSION: BIS and calf circumference may be helpful to assess changes in fluid status in children receiving maintenance HD. IVC diameter, lung B-lines and dermal thickness are potential candidates for future studies.

Longitudinal assessment of vascular calcification in generalized arterial calcification of infancy.

BACKGROUND: Generalized arterial calcification of infancy (GACI), also known as idiopathic infantile arterial calcification, is a very uncommon genetic disorder characterized by calcifications and stenoses of large- and medium-size arteries that can lead to end-organ damage. OBJECTIVE: To describe changes in imaging findings in 10 children with GACI at a single institution from 2010 to 2021. MATERIALS AND METHODS: In this retrospective study we reviewed initial and follow-up body imaging in children with genetic confirmation of GACI at our hospital. All initial images were analyzed for the presence and distribution of arterial calcifications, stenoses and wall thickening/irregularity within the chest, abdomen and pelvis. We compared available follow-up studies to the initial imaging findings. We extracted clinical information including prenatal and postnatal treatment from the children's medical records. RESULTS: We evaluated 10 children (five boys) with a diagnosis of GACI. Median age at first body imaging was 8 days (range: 1 day to 5 years). Six children were identified prenatally and four postnatally. Postnatal presentation included cardiac failure, seizures and hypertension. Images in newborns (n = 8) most commonly showed diffuse arterial calcifications (6/8; 75%), while stenoses were less common (2/8; 25%) during this period. Two children were diagnosed after the neonatal period - one in infancy and one during childhood. In total, half the children (5/10; 50%) had arterial stenoses - three cases visualized at first imaging and two identified on follow-up images during infancy. Stenoses had completely resolved in one child (1/5; 20%) at last follow-up. Eight children received prenatal or postnatal treatment or both. All children who received both prenatal and postnatal treatment (n = 4) had completely resolved calcifications at last follow-up. CONCLUSION: Children with GACI might have characteristic vascular calcifications at birth that raise the suspicion of this disease. Arterial calcifications decrease or disappear spontaneously or after treatment, but arterial stenoses usually persist. Calcifications and arterial stenoses can be easily identified and followed with non-contrast CT and CT angiography.

Pilot study for comparative assessment of dual-energy computed tomography and single-photon emission computed tomography V/Q scanning for lung perfusion evaluation in infants.

OBJECTIVE: To evaluate clinical applications of dual-energy computed tomography (DECT) in pediatric-specific lung diseases and compare ventilation and perfusion findings with those from single-photon emission computed tomography (SPECT-CT) V/Q. METHODS: All patients at our institution who underwent exams using both techniques within a 3-month period were included in this study. Two readers independently described findings for DECT, and two other readers independently analyzed the SPECT-CT V/Q scan data. All findings were compared between readers and disagreements were reassessed and resolved by consensus. Inter-modality agreements are described throughout this study. RESULTS: Eight patients were included for evaluation. The median age for DECT scanning was 3.5 months (IQR = 2). Five of these patients were scanned for both DECT and SPECT-CT V/Q studies the same day, and three had a time gap of 7, 65, and 94 days between studies. The most common indications were chronic lung disease (5/8; 63%) and pulmonary hypertension (6/8; 75%). DECT and SPECT-CT V/Q identified perfusion abnormalities in concordant lobes in most patients (7/8; 88%). In one case, atelectasis limited DECT perfusion assessment. Three patients ultimately underwent lobectomy with corresponding perfusion abnormalities identified by all reviewers on both DECT and SPECT-CT V/Q in all resected lobes. CONCLUSION: DECT is a feasible technique that could be considered as an alternative for SPECT-CT V/Q for lung perfusion evaluation in infants.

Validation of computed tomography angiography as a complementary test in the assessment of renal artery stenosis: a comparison with digital subtraction angiography.

BACKGROUND: Renal artery stenosis is an important cause of hypertension in children, accounting for 5-10% of cases. When suspected, noninvasive imaging options include ultrasound (US), computed tomography (CT) angiography and magnetic resonance (MR) angiography. However, digital subtraction angiography (DSA) remains the gold standard. OBJECTIVE: To investigate the accuracy and inter-reader reliability of CT angiography in children with suspected renal artery stenosis. MATERIALS AND METHODS: This is a retrospective study of patients suspected of having renal artery stenosis evaluated by both CT angiography and DSA between 2008 and 2019 at a tertiary pediatric hospital. Only children who underwent CT angiography within 6 months before DSA were included. CT angiography studies were individually reviewed by two pediatric radiologists, blinded to clinical data, other studies and each other's evaluation, to determine the presence of stenosis at the main renal artery and 2nd- and 3rd-order branches. The sensitivity, specificity and accuracy were calculated using DSA as the reference. The effective radiation dose for CT angiography and DSA was also calculated. Kappa statistics were used to assess inter-reader agreement. RESULTS: Seventy-four renal units were evaluated (18 girls, 19 boys). The patients' median age was 8 years (range: 1-21 years). Overall, CT angiography was effective in detecting renal artery stenosis with a sensitivity of 85.7%, specificity of 91.5% and accuracy of 88.9%. There was moderate inter-reader agreement at the main renal artery level (k=0.73) and almost perfect inter-reader agreement at the 2nd/3rd order (k=0.98). However, the sensitivity at the 2nd- and 3rd-order level was lower (14.3%). CT angiography provided excellent negative predictive value for evaluating renal artery stenosis at the main renal artery level (90.1%) and at the 2nd- or 3rd-order branches (82.7%). The median effective dose of CT angiography studies was 2.2 mSv (range: 0.6-6.3) while the effective dose of DSA was 13.7 mSv. CONCLUSION: CT angiography has high sensitivity and specificity at the main renal artery level with a lower radiation dose than previously assumed. Therefore, it can be used as a diagnostic tool in patients with low to medium risk of renal artery stenosis, and as a screening and treatment planning tool in patients at high risk.

Developmental hip dysplasia and hip ultrasound frequency in a large American payer database.

Developmental dysplasia of the hip (DDH) is an important contributor to musculoskeletal morbidity, but effective strategies to screen for DDH remain controversial. The current utilization of hip ultrasound (US) screening for DDH in the United States is not defined. This study utilized Optum's de-identified Clinformatics® Data Mart, a large commercial and Medicare Advantage claims database. The frequency of DDH and hip US utilization was estimated using billing data on an average of 2.9 million relevant beneficiaries included annually from 2007 through 2017. A total of 6806 DDH cases were identified with an average annual prevalence of 1.7 per 1000 infants, which was stable during the study period. Girls were more likely to be screened and diagnosed with DDH, comprising 72% of DDH cases with an OR of 2.55 (95% CI 2.42-2.69), p < 0.001. Hip US screening was employed in 0.9% of the infant population on average but increased substantially from 2007 (0.4%) to 2017 (2.2%). Most common billing diagnoses included hip deformity (27.4%), breech delivery (20.4%), and physical exam abnormality (17.7%). The average imaging costs per patient for all screened children was $108.94. Insurance claims reflect the current American practice of selective hip US with relative adherence to American Academy of Pediatrics guidelines based on reported diagnoses. While hip US utilization increased during the study period, prevalence of DDH diagnoses did not increase. Our results suggest that expansion of hip US screening may not effectively increase DDH detection although further investigation is needed to ascertain optimal screening strategies to improve patient outcomes.

Image Quality of ECG-Triggered High-Pitch, Dual-Source Computed Tomography Angiography for Cardiovascular Assessment in Children.

PURPOSE: Evaluate the feasibility and determinants of image quality of ECG-triggered High-Pitch Dual-Source Computed Tomography Angiography (CTA) for cardiovascular assessment in Children. MATERIAL AND METHODS: All children that underwent ECG-triggered High-Pitch Dual-Source CTA between August 2014 and September 2017 were identified. Scanner parameters and patients' information were retrieved. Objective image quality was evaluated measuring the Hounsfield units (HU) and standard deviation of regions of interests in the left ventricle, ascending and descending aorta. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated. Qualitative image quality was recorded independently by two pediatric radiologist blinded using a three-point scale: 1 - good image quality, 2 - mild artifacts, 3 - poor image quality. Continuous variables were presented as mean ± standard deviation. The interobserver agreement and non-parametric test were used. RESULTS: 93 patients (mean age 5.6 ± 7.1 years) were selected. Average cardiovascular attenuation, SNR and CNR were 406.2 ± 146.3 HU, 24.2 ± 16.0 HU and 52.1 ± 38.6 HU, respectively. Average image quality was 1.51 ± 0.48 and the inter-observer agreement was excellent (k = 0.8). Worse subjective quality scores were associated with lower age, height, weight, BSA, lower contrast dose and slower injection rates (p < 0.05). Higher heart rate was associated with high attenuation (p < 0.05), however, SNR and CNR did not show an association with heart rate (p = 0.80). CONCLUSIONS: ECG-triggered High-Pitch Dual-Source cardiac CTA is feasible and provides good or excellent image quality for the evaluation of cardiovascular diseases in children.

Cost-Effectiveness Analysis in Pediatric Imaging: The Evidence (or Lack Thereof) Thus Far.

OBJECTIVE: To systematically review all published cost-effectiveness analyses (CEAs) of imaging technologies in children. METHODS: We identified all CEAs involving fetal and pediatric imaging included in a publicly available repository of CEAs published since 1976. Information on publication characteristics, methods, costs, quality of life weights, and incremental cost-effectiveness ratios (ICERs) was recorded for each article. RESULTS: Of 480 diagnostic CEAs, only 10 focused on fetal and pediatric imaging. The 10 studies reported 43 quality of life weights based on previously published adult data (n = 20, 46.5%); pregnant women perspective (n = 11, 25.6%), and treating physician perspective (n = 12, 27.9%). None of the studies elicited quality of life weights from children nor took into consideration the postnatal impact of disease on family's quality of life. All studies used a health care payer perspective without incorporating patients' incurred cost such as loss wages or travel related cost. Of 37 ICERs, 7 (18.9%) were cost saving and 6 (16.2%) were more expensive and less effective. The remaining ICERs ranged from $1,400 per quality-adjusted life-year (MRI versus ultrasound in newborns with moderate risk of occult spinal dysraphism) to $10,000,000 per quality-adjusted life-year (CT versus no imaging in children at low risk for craniosynostosis). CONCLUSION: There is a striking paucity of cost-effectiveness studies evaluating imaging technologies in children. Existing studies do not incorporate the patient and family perspectives regarding utilities (eg, impact of child's disease on families' quality of life) or cost (eg, loss wages, travel, time off). Future studies should elicit and incorporate the impact of disease on families to better reflect real-world scenarios.

Imaging Intussusception in Children's Hospitals in the United States: Trends, Outcomes, and Costs.

OBJECTIVE: To describe imaging utilization, outcomes, and cost in the management of intussusception between 2010 and 2017 in pediatric hospitals in the United States. METHODS: All children (under 18 years of age) with a primary diagnosis of intussusception in a large administrative database were identified. Demographics, imaging, and costs were described. RESULTS: There were 17,032 children (63.3% boys, 36.7% girls, mean age: 3.2 years) that had 20,655 hospital encounters for intussusception, and 88.5% were <5 years of age. The average length of stay was 2.8 days (median: 1 day), with rates of intensive care unit admission, 3.7%; 90-day readmission, 10.5%; and mortality, 0.2%. The surgical rate was 19.6%, and 93.5% (n = 19,301) of patients underwent imaging: 87.2% (n = 16,822) received ultrasound, 69.1% (n = 13,329) had fluoroscopy, 59% (n = 11,380) had abdominal radiographs, and 8.8% (n = 1,696) had CT. The reduction success rate for fluoroscopy was 77.9%. Surgery was more common in rural patients (26.8% versus 18.7% in urban patients, P < .001). Median encounter costs were $2,675 (interquartile range: $1,637-$5,465). Imaging cost represented a quarter (median $680, interquartile range: $372-1,069) of all costs. Higher costs (median) were associated with longer length of stay (<3 days: $858 versus >3 days: $5,342; use of CT ($4,168 versus $943 in patients without a CT), and surgery ($4,434 versus $860 without surgery). CONCLUSION: The management of intussusception is mainly nonsurgical, most frequently involving imaging with ultrasound and fluoroscopy, and resulting in excellent outcomes in the great majority of the cases. Despite playing a central role for diagnosis and management, imaging only represents a fraction of total cost.

Intrahepatic dynamic contrast MR lymphangiography: initial experience with a new technique for the assessment of liver lymphatics.

OBJECTIVES: To describe the technique and report on our initial experience with the use of intrahepatic dynamic contrast magnetic resonance lymphangiography (IH-DCMRL) for evaluation of the lymphatics in patients with hepatic lymphatic flow disorders. METHODS: This is a retrospective review of the imaging and clinical findings in six consecutive patients undergoing IH-DCMRL. The technique involves injection of a gadolinium contrast agent into the intrahepatic lymphatic ducts followed by imaging of the abdomen and chest with both heavily T2-weighted imaging and dynamic time-resolved imaging. RESULTS: In six consecutive patients, IH-DCMRL was technically successful. There were four patients with protein-losing enteropathy (PLE) and two with ascites in this study. In the four patients with PLE, IH-DCMRL demonstrated hepatoduodenal connections with leak of contrast into the duodenal lumen not seen by conventional lymphangiography. In one patient with ascites, IH-DCMRL demonstrated lymphatic leakage into the peritoneal cavity not seen by intranodal lymphangiography. In the second patient with ascites, retrograde lymphatic perfusion of mesenteric lymphatic networks and nodes was seen. Venous contamination was seen in two patients. No biliary contamination was identified. There were no short-term complications. CONCLUSIONS: IH-DCMRL is a cross-sectional technique which successfully evaluated hepatic lymphatic flow disorders and warrants further investigation. KEY POINTS: • Intrahepatic dynamic contrast magnetic resonance lymphangiography (IH-DCMRL) is a new imaging technique to evaluate hepatic lymphatic flow disorders such as protein-losing enteropathy. • In comparison to conventional liver lymphangiography, IH-DCMRL offers a 3D imaging technique and better distal lymphatic contrast distribution and does not use ionizing radiation.

Protocol optimization for cardiac and liver iron content assessment using MRI: What sequence should I use?

OBJECTIVE: To determine the optimal MRI protocol and sequences for liver and cardiac iron estimation in children. METHODS: We evaluated patients ≤18 years with cardiac and liver MRIs for iron content estimation. Liver T2 was determined by a third-party company. Cardiac and Liver T2* values were measured by an observer. Liver T2* values were calculated using the available liver parenchyma in the cardiac MRI. Linear correlations and Bland-Altman plots were run between liver T2 and T2*, cardiac T2* values; and liver T2* on dedicated cardiac and liver MRIs. RESULTS: 139 patients were included. Mean liver T2 and T2* values were 8.6 ±â€¯5.4 ms and 4.5 ±â€¯4.1 ms, respectively. A strong correlation between liver T2 and T2* values was observed (r = 0.96, p < 0.001) with a bias (+4.1 ms). Mean cardiac bright- and dark-blood T2* values were 26.5 ±â€¯12.9 ms and 27.2 ±â€¯11.9 ms, respectively. Cardiac T2* values showed a strong correlation (r = 0.81, p < 0.001) with a low bias (-1.0 ms). The mean liver T2* on liver and cardiac MRIs were 4.9 ±â€¯4.7 ms and 4.6 ±â€¯3.9 ms, respectively. A strong correlation between T2* values was observed (r = 0.96, p < 0.001) with a small bias (-0.2 ms). CONCLUSION: MRI protocols for iron concentration in the liver and the heart can be simplified to avoid redundant information and reduce scan time. In most patients, a single breath-hold GRE sequence can be used to evaluate the iron concentration in both the liver and heart.

Cost-Effectiveness Analysis: An Overview of Key Concepts, Recommendations, Controversies, and Pitfalls.

The field of radiology has witnessed a burst of technological advances that improve diagnostic quality, reduce harm to patients, support clinical needs, and better serve larger more diverse patient populations. One of the critical challenges with these advances is proving that value outweighs the cost. The use of cutting-edge technology is often expensive, and the reality is that our society cannot afford all the screening and diagnostic tests that are being developed. At the societal level, we need tools to help us decide which health programs should be funded. Therefore, decision makers are increasingly looking toward scientific methods to compare health technologies in order to improve allocation of resources. One of such methods is cost-effectiveness analysis. In this article, we review key features of cost-effectiveness analysis and its specific issues as they relate to radiology.

Contrast reaction training in US radiology residencies: a COARDRI study.

OBJECTIVE: To perform a survey-based assessment of current contrast reaction training in US diagnostic radiology residency programs. METHODS: An electronic survey was distributed to radiology residency program directors from 9/2015-11/2015. RESULTS: 25.7% of programs responded. 95.7% of those who responded provide contrast reaction management training. 89.4% provide didactic lectures (occurring yearly in 71.4%). 37.8% provide hands-on simulation training (occurring yearly in 82.3%; attended by both faculty and trainees in 52.9%). CONCLUSION: Wide variability in contrast reaction education in US diagnostic radiology residency programs reveals an opportunity to develop and implement a national curriculum.

Imaging Utilization for the Diagnosis of Appendicitis in Stand-Alone Children's Hospitals in the United States: Trends and Costs.

OBJECTIVE: To describe the imaging utilization patterns for the diagnosis of appendicitis among children's hospitals in the United States over the last 10 years (2005-2014). METHODS: All patients with a primary discharge diagnosis of appendicitis included in a large administrative database of 45 pediatric institutions in the United States between 2005 and 2014 were selected. Demographics, imaging utilization, and costs were described. RESULTS: In all, 96,786 children with appendicitis (59% boys, 41% girls; mean age: 9.9 years) were studied. The average length of stay decreased from 5.0 days in 2005 to 3.4 days in 2014 (P < .01). The percentage of patients undergoing CT increased between 2005 and 2007 from 59.1% to 62.6%, respectively, followed by a decrease from 62.6% to 32.7% in 2014 (r2 = 0.93). Radiograph utilization decreased from 14.2% in 2005 to 3.6% in 2014 (r2 = 0.93), and ultrasound and MRI increased from 25% and 0.03% in 2005 to 61% and 1.0% in 2014 (r2 = 0.97 and 0.64), respectively. The mean total hospital costs increased from $11,700 in 2005 to $16,500 in 2014; imaging costs increased only slightly from $3,205 to $3,259. The imaging fraction of hospital costs decreased from 27.5% to 19.8%. CONCLUSION: There has been a significant decrease in utilization of CT and radiographs for the management of appendicitis in children, and ultrasound has continued to increase. Imaging costs have remained stable in comparison to rising hospital costs, generating a drop in the fraction of costs related to imaging.

Research Challenges and Opportunities for Clinically Oriented Academic Radiology Departments.

Between 2004 and 2012, US funding for the biomedical sciences decreased to historic lows. Health-related research was crippled by receiving only 1/20th of overall federal scientific funding. Despite the current funding climate, there is increased pressure on academic radiology programs to establish productive research programs. Whereas larger programs have resources that can be utilized at their institutions, small to medium-sized programs often struggle with lack of infrastructure and support. To address these concerns, the Association of University Radiologists' Radiology Research Alliance developed a task force to explore any untapped research productivity potential in these smaller radiology departments. We conducted an online survey of faculty at smaller clinically funded programs and found that while they were interested in doing research and felt it was important to the success of the field, barriers such as lack of resources and time were proving difficult to overcome. One potential solution proposed by this task force is a collaborative structured research model in which multiple participants from multiple institutions come together in well-defined roles that allow for an equitable distribution of research tasks and pooling of resources and expertise. Under this model, smaller programs will have an opportunity to share their unique perspective on how to address research topics and make a measureable impact on the field of radiology as a whole. Through a health services focus, projects are more likely to succeed in the context of limited funding and infrastructure while simultaneously providing value to the field.

Accuracy, risk and the intrinsic value of diagnostic imaging: a review of the cost-utility literature.

RATIONALE AND OBJECTIVES: The aim of this study was to systematically review the reporting of the value of imaging unrelated to treatment consequences and test characteristics in all imaging-related published cost-utility analyses (CUAs) in the medical literature. MATERIALS AND METHODS: All CUAs published between 1976 and 2008 evaluating diagnostic imaging technologies contained in the CEA Registry, a publicly available comprehensive database of health related CUAs, were screened. Publication characteristics, imaging modality, and the inclusion of test characteristics including accuracy, costs, risks, and the potential value unrelated to treatment consequences (eg, reassurance or anxiety) were assessed. RESULTS: Ninety-six published CUAs evaluating 155 different imaging technologies were included in the final sample; 27 studies were published in imaging-specialized journals. Fifty-two studies (54%) evaluated the performance of a single imaging modality, while 44 studies (46%) compared two or more different imaging modalities. The most common areas of interest were cardiovascular (45%) and neuroradiology (17%). Forty-two technologies (27%) concerned ultrasound, while 34 (22%) concerned magnetic resonance. Seventy-nine (51%) technologies used ionizing radiation. Test accuracy was reported or calculated for 90% (n = 133 and n = 5, respectively) and assumed perfect (reference test or gold-standard test without alternative testing strategy to capture false-negatives and false-positives) for 8% (n = 12) of technologies. Only 22 studies (23%) assessing 40 imaging technologies (26%) considered inconclusive or indeterminate results. The risk of testing was reported for 32 imaging technologies (21%). Fifteen studies (16%) considered the value of diagnostic imaging unrelated to treatment. Four studies incorporated it as quality-of-life adjustments, while 10 studies mentioned it only in their discussions or as a limitation. CONCLUSIONS: The intrinsic value of imaging (the value of imaging unrelated to treatment) has not been appropriately defined or incorporated in the existing cost-utility literature, which could be due to a lack of evidence on the issue. Thus, more research is needed on metrics for a more comprehensive evaluation of diagnostic imaging. Similarly, the incorporation of variations in imaging tests accuracy, inconclusive results and associated risks has lacked uniformity in the cost-utility literature. Acknowledgment of these characteristics in future cost-utility publications will enhance their value and provide results that more closely resemble routine clinical practice.

Cost-utility analyses of diagnostic laboratory tests: a systematic review.

OBJECTIVE: To review and evaluate the literature of cost-utility analyses (CUAs) regarding diagnostic laboratory testing. METHODS: We reviewed all articles related to diagnostic laboratory testing in the Tufts Medical Center Cost-Effectiveness Analysis Registry (www.cearegistry.org), which contains detailed information on over 2000 published CUAs through 2008. We analyzed the extent to which the studies adhered to recommended practices for conducting and reporting cost-effectiveness analyses. We also recorded whether the studies contained information on diagnostic test accuracy and costs, and whether any account was taken of potential benefits or harms of testing that are unrelated to subsequent treatment, such as the reassurance value of testing. RESULTS: We identified 141 published CUAs pertaining to diagnostic laboratory testing published through 2008 which contained 433 separate incremental cost-effectiveness ratios. Prior to 2000, there were only 20 CUAs published, but the number averaged 13.4 annually thereafter. Most studies focused on hematology/oncology (n = 42, 30%) and obstetrics/gynecology (n = 36, 26%) applications. Approximately 63% (n = 89) of studies clearly reported information about the accuracy of the test, but only 10% (n = 14) mentioned test safety or possible risks. A small number (n = 10, 7%) mentioned or considered the potential value or harm of testing unrelated to treatment consequences. Over 55% of the reported incremental cost-effectiveness ratios (ICERs) were either dominant (more quality-adjusted life years for less cost), or below $50,000 per quality-adjusted life years gained (in 2008 US dollars). CONCLUSION: The number of CUAs investigating laboratory diagnostic testing has increased substantially with applications to diverse clinical areas. The literature reveals many areas in which testing represents good value for money. The vast majority of studies have not considered preferences for test information unrelated to treatment consequences.

Cost-effective diagnostic cardiovascular imaging: when does it provide good value for the money?

To summarize the results of all original cost-utility analyses (CUAs) in diagnostic cardiovascular imaging (CVI) and characterize those technologies by estimates of their cost-effectiveness. We systematically searched the literature for original CVI CUAs published between 2000 and 2008. Studies were classified according to several variables including anatomy of interest (e.g. cerebrovascular, aorta, peripheral) and imaging modality under study (e.g. angiography, ultrasound). The results of each study, expressed as cost of the intervention to number of quality-adjusted life years saved ratio (cost/QALY) were additionally classified as favorable or not using $20,000, $50,000, and $100,000 per QALY thresholds. The distribution of results was assessed with Chi Square or Fisher exact test, as indicated. Sixty-nine percent of all cardiovascular imaging CUAs were published between 2000 and 2008. Thirty-two studies reporting 82 cost/QALY ratios were included in the final sample. The most common vascular areas studied were cerebrovascular (n = 9) and cardiac (n = 8). Sixty-six percent (21/32) of studies focused on sonography, followed by conventional angiography and CT (25%, n = 8, each). Twenty-nine (35.4%), 42 (51.2%), and 53 (64.6%) ratios were favorable at WTP $20,000/QALY, $50,000/QALY, and $100,000/QALY, respectively. Thirty (36.6%) ratios compared one imaging test versus medical or surgical interventions; 26 (31.7%) ratios compared imaging to a different imaging test and another 26 (31.7%) to no intervention. Imaging interventions were more likely (P < 0.01) to be favorable when compared to observation, medical treatment or non-intervention than when compared to a different imaging test at WTP $100,000/QALY. The diagnostic cardiovascular imaging literature has growth substantially. The studies available have, in general, favorable cost-effectiveness profiles with major determinants relating to being compared against observation, medical or no intervention instead of other imaging tests.