High levels of fasting plasma insulin, reflecting insulin resistance, are an early event in the progression to type 2 diabetes. Experimental studies implicate ceramides in insulin resistance; however, whether circulating ceramides and related sphingolipids are associated with plasma insulin in humans is uncertain. We measured 15 ceramide and sphingomyelin species with different saturated fatty acids in baseline fasting plasma samples from the Strong Heart Family Study, a prospective cohort of American Indians. We examined associations of sphingolipids with plasma insulin, HOMA-IR and HOMA-B at baseline and follow-up (median 5 years later) using linear mixed models adjusted for family structure. Among 2086 participants without diabetes, higher levels of plasma ceramides carrying the fatty acids 16:0, 18:0, 20:0 or 22:0 were associated with higher plasma insulin and higher HOMA-IR at baseline and follow-up (Table). BMI modified associations of sphingomyelin species carrying 18:0, 20:0, 22:0 or 24:0 with insulin (p<0.003); higher sphingomyelin levels were associated with lower fasting insulin and HOMA indices only among those with normal BMI. Our study suggests lowering circulating ceramides might be a target in prediabetes, and targeting sphingomyelins should take into account BMI.

Table. Plasma Ceramides, Insulin and HOMAIR.

Species Fold change in geometric mean with 2 fold higher ceramide (95% CI), p-value 
 Baseline insulin Follow-up insulin Baseline HOMAIR Follow-up HOMAIR 
Cer-16:0 1.13 (1.07-1.20) 1.9x10-5 1.10 (1.01-1.21) 0.03 1.15 (1.08-1.22) 4.7x10-6 1.11 (1.01-1.22) 0.03 
Cer-18:0 1.11 (1.06-1.15) 1.0x10-6 1.12 (1.06-1.19) 1.0x10-5 1.12 (1.08-1.17) 6.3x10-8 1.13 (1.06-1.20) 2.0x10-4 
Cer-20:0 1.12 (1.07-1.17) 1.5x10-7 1.10 (1.03-1.18) 0.003 1.13 (1.08-1.18) 1.3x10-7 1.11 (1.04-1.19) 0.002 
Cer-22:0 1.13 (1.08-1.19) 3.9x10-7 1.12 (1.04-1.20) 0.004 1.15 (1.09-1.21) 1.0x10-7 1.13 (1.04-1.22) 0.004 
Cer-24:0 1.(0.99-1.11) 0.08 1.09 (1.01-1.19) 0.04 1.(1.00-1.13) 0.04 1.11 (1.01-1.21) 0.03 
Species Fold change in geometric mean with 2 fold higher ceramide (95% CI), p-value 
 Baseline insulin Follow-up insulin Baseline HOMAIR Follow-up HOMAIR 
Cer-16:0 1.13 (1.07-1.20) 1.9x10-5 1.10 (1.01-1.21) 0.03 1.15 (1.08-1.22) 4.7x10-6 1.11 (1.01-1.22) 0.03 
Cer-18:0 1.11 (1.06-1.15) 1.0x10-6 1.12 (1.06-1.19) 1.0x10-5 1.12 (1.08-1.17) 6.3x10-8 1.13 (1.06-1.20) 2.0x10-4 
Cer-20:0 1.12 (1.07-1.17) 1.5x10-7 1.10 (1.03-1.18) 0.003 1.13 (1.08-1.18) 1.3x10-7 1.11 (1.04-1.19) 0.002 
Cer-22:0 1.13 (1.08-1.19) 3.9x10-7 1.12 (1.04-1.20) 0.004 1.15 (1.09-1.21) 1.0x10-7 1.13 (1.04-1.22) 0.004 
Cer-24:0 1.(0.99-1.11) 0.08 1.09 (1.01-1.19) 0.04 1.(1.00-1.13) 0.04 1.11 (1.01-1.21) 0.03 

Covariates: age, sex, site, education, smoking, log(BMI), waist, physical activity.

Disclosure

R.N. Lemaitre: None. C. Yu: None. A. Hoofnagle: Research Support; Self; Waters Corporation. H. Nair: None. P.N. Jensen: None. A.M. Fretts: None. J.G. Umans: None. C. Sitlani: None. D. Siscovick: None. I. King: None. N. Sotoodehnia: None. B. McKnight: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.