Exp Clin Endocrinol Diabetes 2007; 115(4): 240-243
DOI: 10.1055/s-2007-970577

© J. A. Barth Verlag in Georg Thieme Verlag KG · Stuttgart · New York

Early Retinopathy Progression in Four Randomized Trials Comparing Insulin Glargine and Nph Insulin

M. D. Davis 1 , R. W. Beck 2 , P. D. Home 3 , J. Sandow 4 , F. L. Ferris 5
  • 1Department of Ophthalmology and Visual Sciences, University of Wisconsin, Madison, WI, USA
  • 2Jaeb Center for Health Research, Tampa, FL, USA
  • 3School of Medical Sciences - Diabetes, University of Newcastle upon Tyne, Newcastle upon Tyne, UK
  • 4sanofi-aventis, Frankfurt am Main, Germany
  • 5Division of Biometry and Epidemiology, National Eye Institute, Bethesda, MD, USA
Further Information

Publication History

received 11. 9. 2006 first decision 11. 9. 2006

accepted 6. 11. 2006

Publication Date:
03 May 2007 (online)


Early worsening of diabetic retinopathy, characterized by cotton wool spots, intraretinal microvascular abnormalities and/or macular edema, can occur following improvement of glycemic control. In four randomized 28- to 52-week clinical trials comparing insulin glargine and NPH insulin in regard to glycemic control and frequency of hypoglycemia, ophthalmologic examinations and fundus photographs were included to assess frequency of early worsening of retinopathy or other early adverse ocular effects. Retinopathy progression rates at 28 weeks were 7-12% by clinical examination and 3-8% by photographic grading; corresponding rates of clinically significant macular edema (CSME) were 1-8% and 1-4%, respectively. Optic disc swelling was not observed clinically or in photographs. Two of the 24 possible comparisons (four trials, three outcomes, two assessment methods), both of which were photographic assessments in type 2 diabetes, were in/near the nominally significant range and favored NPH insulin: 28-week rates of ≥3-step retinopathy progression (insulin glargine: 16/213, 7.5%; NPH insulin: 6/220, 2.7%; p=0.028) and 52-week CSME rates (26/233, 11.2% and 14/214, 6.5%, respectively; p=0.098). Because the between-treatment differences were small and inconsistent across trials and assessment methods, and because overall rates were consistent with the natural course of diabetic retinopathy, we conclude that it is unlikely that insulin glargine carries a higher risk of early worsening or other early adverse effect than NPH insulin. These trials tended to exclude a large early adverse effect, such as optic disc swelling, but cannot assess longer-term effects; a 5-year randomized trial of insulin glargine versus NPH insulin has been initiated.Data from this manuscript have been presented as posters and published in abstract form at the European Association for the Study of Diabetes 2001 (Diabetologia 44(Suppl 1):I-IV(A287), 2001) and the Latin American Diabetes Association 2001 (11-15 November 2001, Punta del Este, Uruguay; Poster 180) congresses.



M. D. Davis

Fundus Photograph Reading Center

Department of Ophthalmology and Visual Sciences

University of Wisconsin-Madison

406 Science Drive

Suite 400


WI 53711-1068


Phone: +1/608/263/60 71

Fax: +1/608/262/18 99

Email: [email protected]