Public Health Nutrition

Research Article

Serum carotenoids and radiographic knee osteoarthritis: the Johnston County Osteoarthritis Project

Anneclaire J De Roosa1, Lenore Araba2, Jordan B Rennera3, Neal Crafta4, Gheorghe Lutaa5, Charles G Helmicka6, Marc C Hochberga7 and Joanne M Jordana8 c1

a1 National Cancer Institute, Bethesda, MD, USA

a2 Thurston Arthritis Research Center, Department of Epidemiology and Nutrition, University of North Carolina School of Public Health, Chapel Hill, NC, USA

a3 Thurston Arthritis Research Center, Department of Radiology, University of North Carolina School of Medicine, Chapel Hill, NC, USA

a4 Craft Technologies, Inc., Wilson, NC, USA.

a5 Thurston Arthritis Research Center, Department of Biostatistics, University of North Carolina School of Public Health, Chapel Hill, NC, USA

a6 Centers for Disease Control and Prevention, Atlanta, GA, USA

a7 Departments of Medicine, Epidemiology, and Preventive Medicine, University of Maryland School of Medicine, Baltimore, MD, USA

a8 Thurston Arthritis Research Center, 3310 Doc J Thurston Jr Building, CB* 7330, University of North Carolina, Chapel Hill, NC 27599-7330, USA


Objective: Antioxidant intake has been associated with less progression of radiographic knee osteoarthritis (OA), but studies of carotenoid biomarkers and OA have not been done. We examined associations between serum concentrations of nine naturally occurring carotenoids and radiographic knee OA.

Design: The study design was matched case–control. Sera were analysed by high-performance liquid chromatography for nine carotenoids: lutein, zeaxanthin, α- and β-cryptoxanthin, trans- and cis-lycopene, α-carotene, and trans- and cis-β-carotene. Conditional logistic regression was used to estimate the association between tertiles of each carotenoid and radiographic knee OA, independent of body mass index, education, serum cholesterol, and the other carotenoids.

Setting: Johnston County, North Carolina, United States of America.

Subjects: Two-hundred cases with radiographic knee OA (Kellgren–Lawrence grades ≥2) and 200 controls (Kellgren–Lawrence grade = 0) were randomly selected from the Johnston County Osteoarthritis Project, and were matched on age, gender and race.

Results: Participants with serum levels of lutein or β-cryptoxanthin in the highest tertile were approximately 70% less likely to have knee OA than controls {odds ratio (OR) [95% confidence interval (CI)] = 0.28 [0.11, 0.73] and 0.36 [0.14, 0.95], respectively}. Those in the highest tertile of trans-β-carotene (OR = 6.40 [1.86, 22.1]) and zeaxanthin (OR = 3.06 [1.19, 7.85]) were more likely to have knee OA.

Conclusions: While certain carotenoids may protect against knee OA, others may increase the odds of knee OA. Further study of carotenoids and knee OA are warranted before clinical recommendations about these substances and knee OA can be made.

(Received September 08 2000)

(Accepted February 07 2001)


c1 *Corresponding author: Email