A deeper analysis in thyroid research: A meta-epidemiological study of the American Thyroid Association clinical guidelines.

BACKGROUND: The American Thyroid Association (ATA) uses the GRADE or the American College of Physicians (ACP) system to develop recommendations. Recommendations based on low-quality evidence should spur for the conduction of clinical studies, if feasible. The extent to which recommendations by the ATA based on low-quality of evidence are being actively researched remains unknown. METHODS: Clinical guidelines produced by the ATA using the GRADE or the ACP system to classify evidence were deemed eligible. Reviewers, in duplicate and independently, extracted therapeutic recommendations based on low-quality evidence, whereas recommendations with higher quality of evidence, aimed at diagnosis, or best practice statements were excluded. Eligible recommendations based on low-quality evidence were deconstructed to their components using the PICO format. We then searched on clinicaltrials.gov to identify ongoing research. Trials were deemed eligible if they addressed the PICO question with at least one of the intended outcomes. RESULTS: A total of 543 recommendations were retrieved, of which 305 (56%) were based on low-quality of evidence and only 90 were deemed eligible. Of these, we found that 33 (37%) recommendations were actively being researched in 53 clinical trials. Most of the trials were randomized and funded by non-profit organizations. Many clinical trials studied thyroid nodules and differentiated thyroid cancer (26/53; 49%), whereas few studied were aimed at anaplastic thyroid cancer (2/53; 4%). CONCLUSION: One out of three of gaps in evidence, identified as low quality during the development of ATA guidelines, are currently actively researched. This finding calls for the need to develop a better research infrastructure and funding to support thyroid research.

American thyroid association guidelines are inconsistent with Grading of Recommendations Assessment, Development, and Evaluations-A meta-epidemiologic study.

OBJECTIVES: The Grading of Recommendations Assessment, Development, and Evaluations (GRADE) working group has recognized some scenarios in which strong recommendations may be supported by low-quality evidence, the so-called paradigmatic situations. The adherence to these paradigmatic situations by the American Thyroid Association (ATA) guidelines, however, remains unknown. STUDY DESIGN AND SETTING: Clinical guidelines from the ATA were retrieved and deemed eligible if created using GRADE or the American College of Physicians (ACP) system. Reviewers, independently and in duplicate, assessed all strong recommendations based on low-quality evidence and judged their alignment with GRADE paradigmatic situations. The study was conducted at KER Unit Mexico. RESULTS: A total of five clinical guidelines, one using the GRADE and four using the ACP system, were eligible for analysis. We assessed a total of 518 recommendations, of which 355 (69%) were labeled as "strong" and 163 (31%) as "weak". A total of 151 strong recommendations were based on low-quality evidence, of which 36 (24%) were congruent with one of the five GRADE paradigmatic situations, whereas 115 (76%) were not consistent with any paradigmatic situations and should have been categorized as weak (23% [26/115]) or best-practice recommendations (77% [89/115]). CONCLUSION: ATA clinical guidelines are discordant with GRADE guidance. Future guidelines should carefully evaluate the quality of evidence and recognize its limitations when developing recommendations.

Acanthosis Nigricans in the Knuckles of Infants: A Novel Clinical Marker of High Metabolic Risk.

INTRODUCTION: Acanthosis nigricans (AN) is an early clinical sign of insulin resistance (IR) primarily in adults. The prevalence and association of AN and IR in infants, however, remains uncertain. We aimed to describe the prevalence of AN and its association with IR in a group of Latin-American infants. METHODS: We studied a random sample of 227 healthy infants between 9 and 24 months of age. After a complete clinical history was obtained and a physical examination was performed, fasting plasma glucose and serum insulin were measured. Three blinded evaluators assessed AN in each patient. Infants with AN were categorized as cases. The HOMA-IR index cutoffs of ≥ 90th and ≥ 95th percentiles were considered IR. RESULTS: There were 49 infants with AN (21.6%) (cases) and 178 without AN (78.4%) (controls). Cases had a significantly higher mean serum insulin, fasting plasma glucose, and HOMA-IR levels of 3.67 ± 2.56 µU/ml vs. 2.42 ± 1.45 µU/ml, P = 0.005; 84.2 ± 12.6 mg/dL vs. 77 ± SD 9.9 mg/dL, P ≤ 0.001; HOMA-IR 0.77 ± 0.54 vs. 0.46 ± 0.28, P ≤ 0.001, respectively. More cases than controls presented HOMA-IR levels ≥ 95th percentile (cases 18.4%; controls 0.5%, P ≤ 0.001) and ≥ 90th percentile (cases 32.7%; controls 1.6%, P ≤ 0.001). AN in the knuckles had a high sensitivity and a negative predictive value (NPV) for detecting patients with HOMA-IR levels above the 95th percentile (sensitivity 90%; NPV 99.4%) and above the 90th percentile (sensitivity 84.2%; NPV 98.3%). CONCLUSION: AN in the knuckles is a prevalent, non-invasive, costless, and reliable screening clinical tool that can be used for early detection of infants with IR and a high metabolic risk.