News Release

AI tool may enable opportunistic assessment of body composition from routine imaging, identify patients at cardiovascular risk

Peer-Reviewed Publication

American College of Physicians

Embargoed for release until 5:00 p.m. ET on Monday 29 September 2025   

Follow @Annalsofim on X, Facebook, Instagram, Bluesky, and Linkedin             
Below please find summaries of new articles that will be published in the next issue of Annals of Internal Medicine. The summaries are not intended to substitute for the full articles as a source of information. This information is under strict embargo and by taking it into possession, media representatives are committing to the terms of the embargo not only on their own behalf, but also on behalf of the organization they represent.   
----------------------------    

1. AI tool may enable opportunistic assessment of body composition from routine imaging, identify patients at cardiovascular risk

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-24-01863

URL goes live when the embargo lifts             

A prospective cohort study found that an artificial intelligence (AI) tool is accurate and feasible for enabling opportunistic assessment of body composition from routine MRI, helping to identify patients at high cardiometabolic risk. The study found that AI-derived body composition proportions were strongly associated with cardiometabolic risk. However, after adjusting for body mass index (BMI) and waist circumference, only visceral adipose tissue proportion and skeletal muscle fat fraction were associated with incident diabetes and major adverse cardiovascular events (MACE). The results are published in Annals of Internal Medicine.

 

Researchers from the Department of Diagnostic and Interventional Radiology at University of Freiburg and the Cardiovascular Imaging Research Center at Massachusetts General Hospital and Harvard Medical School studied data from 33,539 participants in the UK Biobank who had no history of diabetes, myocardial infarction, or ischemic stroke and underwent a whole-body MRI. The researchers used an open-source AI model to estimate 3D body composition volumes, including subcutaneous adipose tissue (fat just beneath the skin), visceral adipose tissue (fat located around internal organs), skeletal muscle volume and skeletal muscle fat fraction percentage from the whole-body MRI. Primary outcomes were incident diabetes and MACE. The study found that over a median follow-up of 4.8 years, 1.2% of female and 2.5% of male participants were diagnosed with incident diabetes and 1.0% and 2.4% were diagnosed with MACE, respectively. When adjusting for age, smoking, and hypertension, greater adiposity measures and low skeletal muscle proportion were associated with higher incidence of diabetes and MACE in both sexes. Additional adjustments for BMI and waist circumference revealed that associations between subcutaneous adipose tissue proportion and diabetes and MACE were reduced in males, whereas there was a negative associated between relative subcutaneous adipose tissue and future diabetes risk in females. Overall, the AI tool accurately extracted 3D body composition volumes from whole-body MRIs in less than 3 minutes per scan and demonstrated that as both sexes age, adipose tissue compartments and skeletal muscle fat fraction increase while skeletal muscle decreases. These results corroborate evidence that visceral adipose tissue, but not subcutaneous adipose tissue, is a key driver of adiposity-related cardiometabolic risk.

 

Media contacts: For an embargoed PDF, please contact Gabby Macrina at gmacrina@acponline.org. To contact corresponding author Jakob Weiss, MD please email jakob.benedikt.weiss@uniklinik-freiburg.de. 

----------------------------   

2. ACP warns politicization of science is driving U.S. health care system toward disaster

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-25-03999

URL goes live when embargo lifts

Authors from the American College of Physicians (ACP) warn that disaster looms ahead for the U.S. health care system, likely precipitated by an epidemic of vaccine-preventable and attenuating diseases. They say the politicalization of science, which includes the evisceration of public health infrastructure and funding, diminished access to care, and the proliferation of vaccine mis- and disinformation, has created a storm that the U.S. health care system will be ill-prepared to weather. The commentary is published in Annals of Internal Medicine.

 

The authors cite several recent examples of changes to the U.S. health care system that promise negative consequences for public health, including withdrawal from the World Health Organization, erosion of the CDC’s data-reporting process, and elimination of nearly a quarter of the CDC’s workforce. But the authors suggest that one of the most startling changes has taken place among patients, whose vaccine confidence is now at an all-time low. Misinformation and mistrust are being shown to contribute to reduced vaccine uptake, particularly among certain populations. Their confidence is being undermined by conflicting messaging the public is receiving, variable state policies on vaccination mandates, and payment coverage challenges.

 

The authors say it is imperative to address these issues before a health disaster strikes. ACP calls on physicians to work together with local, state and federal officials to ensure that every person receives care in a system with appropriate capacity and resources. The current dismantling of the U.S. public health system must be stopped, as it impairs the nation from responding to health threats in an organized and effective manner. Public health has many definitions, however it is most clearly understood as “what we do together as a society to ensure the conditions in which everyone can be healthy,” they write.

 

Media contacts: For an embargoed PDF, please contact Gabby Macrina at gmacrina@acponline.org. To speak with someone at ACP please contact Angela Collom at acollom@acponline.org.

----------------------------   

3. Patients with severe obesity face barriers and biases when accessing subspecialty care

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-25-01720

URL goes live when the embargo lifts             

A brief research report describes access to subspecialty care among patients weighing more than 450 pounds. The study found that these patients face substantial limitations to accessing subspecialty care, and information on accessibility was difficult to obtain from clinics. The findings highlight a clear need for subspecialty clinics to be familiar with accommodations needed to care for patients of higher weights. The study is published in Annals of Internal Medicine.

 

One in 270 Americans now have a BMI of 60 kg/m2 or greater, which was once thought to be exceptionally rare. Researchers from Northwestern University Feinberg School of Medicine and the University of Chicago Pritzker School of Medicine used a “secret shopper” method to attempt to schedule an appointment for a hypothetical patient weight 465 pounds at practices across each of 5 subspecialties (dermatology, endocrinology, obstetrics and gynecology, orthopedic surgery, and otolaryngology) located in 4 metropolitan areas (Boston, Massachusetts; Cleveland, Ohio; Houston, Texas; and Portland, Oregon). Of the 300 practices included in the study sample, over half lacked basic standards of care for the hypothetical patient, and 41% declined to schedule an appointment. Only 117 practices were fully accessible, with endocrinology practices being the most willing to schedule and most likely to meet basic standards of care. Additionally, many staff members made potentially insensitive weight-related comments, highlighting a clear need for both better accommodations for patients of higher weights and staff training on inclusive care.  

 

Media contacts: For an embargoed PDF, please contact Gabby Macrina at gmacrina@acponline.org. To contact corresponding author Molly Hales, MD, PhD please email Kristin Samuelson at ksamuelson@northwestern.edu. 

----------------------------    

4. Sharp increase in gabapentin dispensing highlights need for awareness of adverse effects

Anticonvulsant is currently the 5th most prescribed medication in the U.S.

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-25-01750

URL goes live when the embargo lifts             

A cross-sectional study examined dispensing of anticonvulsant gabapentin from retail pharmacies in the United States from 2010 to 2024. The findings revealed a slow but increasing rate of gabapentin prescribing since 2026, with gabapentin becoming the 5th most prescribed medication in 2024 compared with the 10th most prescribed in 2017. The findings are published in Annals of Internal Medicine

 

Researchers from the Centers for Disease Control and Prevention (CDC) analyzed 2010 to 2024 data from the IQVIA National Prescription Audit and Total Patient Tracker all-payer pharmaceutical database to examine trends in gabapentin prescriptions and number of unique persons dispensed gabapentin from 2010 to 2024. They found that gabapentin prescriptions increased from 79.5 per 1,000 persons in 2010 to 177.6 per 1,000 persons in 2024, and persons dispensed gabapentin increased from18.9 per 1,000 persons in 2010 to 46.9 per 1,000 persons. These rates doubled between 2010 to 2016 but increased at a slower rate from 2016 to 2024. Substantial increases in prescribing were observed among nurse practitioners and physician assistants, which could reflect expanded prescribing authorities in certain states, but warrants further study. The researchers note that prescribing physicians and advanced care practitioners should be alert to the potential adverse effects of gabapentin.

 

Media contacts: For an embargoed PDF, please contact Gabby Macrina at gmacrina@acponline.org. To contact corresponding author Gery P. Guy Jr., PhD please email media@cdc.gov.

----------------------------    

Also new this issue:

Payment for physician services in the U.S.: Has a reckoning begun?

John Goodson, MD

Ideas and Opinions

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-25-03525

 


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.