The boxes represent the mid 50% of the data, the line within box represents the group mean value adjusted for age and body mass index. The high and low whiskers represent the 95th and 5th percentiles, respectively. The filled black circles represent outliers. Using the t test, the comparison between donors without diabetes and those positive for a single autoantibody only yielded a P value of .02; and for the comparison between donors without diabetes and those with type 1 diabetes yielded a P value of less than .001. A comparison between the donors positive for a single autoantibody only and those with type 1 diabetes was not performed. Statistical significance was indicated at a Bonferroni-corrected nominal α level of .025. Of note, although age and body mass index were poorly correlated with pancreas weight and failed to meet one of the assumptions of analysis of covariance, both were included into the final model because linear models that included age or body mass index by disease status group interaction terms showed that the interaction was not statistically significant.
Campbell-Thompson M, Wasserfall C, Montgomery EL, Atkinson MA, Kaddis JS. Pancreas Organ Weight in Individuals With Disease-Associated Autoantibodies at Risk for Type 1 Diabetes. JAMA. 2012;308(22):2337-2339. doi:10.1001/jama.2012.15008
Author Affiliations: Department of Pathology, University of Florida, Gainesville (Drs Campbell-Thompson and Atkinson, Mr Wasserfall, and Ms Montgomery); and Department of Information Sciences, City of Hope, Duarte, California (Dr Kaddis; email@example.com).
To the Editor: Autopsy and imaging studies suggest that human pancreata from adults with type 1 diabetes (T1D) are smaller and weigh less than those without the disease.1,2 However, it is unknown when pancreatic atrophy begins in T1D.
Therefore, we examined pancreas weight early in the natural history of T1D from at-risk individuals without diabetes but with disease-associated autoantibodies, obtained through the Network for Pancreatic Organ Donors with Diabetes program.3 All donors were identified by organ procurement organizations that coordinate organ and tissue donations for clinical transplantation or research.
Between October 16, 2006, and April 27, 2012, 193 pancreata were recovered from cadaveric multiorgan donors in the United States and considered for the study following written informed research consent from the next of kin. Study procedures were in accordance with the Declaration of Helsinki and the University of Florida institutional review board, and were detailed previously.3
Information on demographics and medical history were obtained from chart review or through a data-sharing agreement with the United Network for Organ Sharing. Organ recovery, processing, autoantibody testing methods, histopathological review, and disease classification procedures have been described.3 Prior to organ recovery, initial screening for autoantibodies to glutamic acid decarboxylase 65 and insulinoma-associated 2ic was performed. Following receipt of the pancreas, confirmatory testing was completed and also included autoantibodies to insulin and zinc transporter 8 antigens. T1D was confirmed by laboratory markers and medical chart review.
We included 23 controls without T1D, 8 T1D-free single islet autoantibody-positive donors, and 20 T1D cases. Cases were excluded with damaged pancreata (n = 17); missing organ weight (n = 26) or confirmatory autoantibody testing results (n = 10); other diseases or unresolved medical history (n = 34); unconfirmed (n = 1) or with multiple (n = 3) autoantibodies; pancreatitis (n = 20) or pancreatic endocrine tumor (n = 1); and donors aged 18 years or younger (n = 30).
The primary outcome variable was pancreas weight of donors in the 3 groups, examined using a general linear model to account for the influence of covariates on the end point (Table). Only age and body mass index (BMI) were retained in the final model. Other variables were not included due to multicollinearity, nonsignificance, or lack of previous evidence for an association.
All P values are unadjusted, 2-tailed, and statistically significant if less than .05, unless otherwise noted. Statistical analysis was performed using SAS software version 9.2 (SAS Institute Inc).
Of the 51 donors, 46 were classified as brain dead and 5 as nonheart beating. Baseline characteristics were similar across groups, with the exception of all diabetes-related variables, reflecting criteria used to categorize the donors (Table). Most donors received cardiovascular drugs prior to death, but there was variation between the 3 disease status groups (Table).
Differences existed in the mean pancreas weight of individuals stratified by disease status and adjusted for age and BMI (Figure). The mean weight of pancreata from those without diabetes (controls) was 81.4 g (95% CI, 73.0-89.8 g) compared with 61.3 g (95% CI, 46.8 g-75.8 g; P = .02) from the group positive for a single autoantibody only and 44.9 g (95% CI, 36.0 g-53.9 g; P < .001) from the T1D group.
In this study, the weight of pancreata in individuals without T1D, but with serum markers that potentially precede the clinical manifestations, as well as in individuals with T1D, was less than in controls. This suggests that early atrophy of the organ may be an important subclinical feature of T1D pathogenesis.
Limitations exist due to the small sample size and expected low incidence of subsequent T1D in single autoantibody-positive individuals.6 Additionally, functional assessment of pancreata was not possible, precluding evaluation of both the exocrine and endocrine compartments of the gland. In this data set, there is no evidence that cause of death or organ retrieval procedures affected organ weight.
Future studies should include validation and analysis of the potential mechanisms underlying this observation to understand whether early pancreatic atrophy contributes to the pathogenesis of T1D.
Author Contributions: Dr Kaddis had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Campbell-Thompson, Atkinson, Kaddis.
Acquisition of data: Campbell-Thompson, Wasserfall, Montgomery, Kaddis.
Analysis and interpretation of data: Campbell-Thompson, Atkinson, Kaddis.
Drafting of the manuscript: Atkinson, Kaddis.
Critical revision of the manuscript for important intellectual content: Campbell-Thompson, Wasserfall, Montgomery, Atkinson, Kaddis.
Statistical analysis: Kaddis.
Obtained funding: Campbell-Thompson, Atkinson.
Administrative, technical, or material support: Campbell-Thompson, Wasserfall, Montgomery, Atkinson.
Study supervision: Campbell-Thompson, Wasserfall, Atkinson.
Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: This research was performed with the support of the Network For Pancreatic Organ Donors with Diabetes program, a collaborative type 1 diabetes research project sponsored by the Juvenile Diabetes Research Foundation International, that provides tissues to approved investigators for the purposes of advancing the understanding of mechanisms that lead to type 1 diabetes (grant 6-2006-1140 to Dr Campbell-Thompson; and grant 25-2007-874 to Dr Atkinson, with a subcontract to Dr Kaddis). Organ procurement organizations partnering with the program to provide donor organ and tissue samples are listed at http://www.jdrfnpod.org/our-partners.php.
Role of the Sponsor: The funding organizations had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.
Disclaimer: The interpretation and reporting of organ donor data are the responsibilities of the authors and in no way should be seen as an official policy of or interpretation by the Organ Procurements and Transplantation Network or the US government.
Additional Contributions: Organ donor information was supplied by the United Network of Organ Sharing as the contractor for the Organ Procurements and Transplantation Network, or obtained from medical charts. Desmond Schatz, MD (University of Florida), and Alberto Pugliese, MD (University of Miami), critically reviewed medical charts, the manuscript, and participated in discussions about data analysis. Patrick A. Rowe, PhD (University of Florida), provided interpretation of the data. David Smith, PhD (City of Hope), provided assistance with constructing general linear models. No individuals received compensation for their contributions to this work.