CSF versus serum leptin in narcolepsy: is there an effect of hypocretin deficiency?

To determine if hypocretin deficiency is associated with abnormally low serum leptin levels, a putative cause of increased body mass index in narcoleptics.

Cross-sectional controlled study.

Three hundred seventy subjects, including 111 healthy controls, 93 narcoleptic subjects with hypocretin deficiency (cerebrospinal fluid [CSF] hypocretin-1 levels < 110 pg/mL), 72 narcoleptic subjects with normal hypocretin levels, and 89 subjects with other sleep disorders

After completing the Stanford Sleepiness Inventory, participants underwent spinal taps and blood sampling for measurement of CSF leptin and hypocretin-1 levels, HLA DQB1*0602 phenotyping, and serum leptin and C-reactive protein levels.

Serum leptin levels were similar in narcoleptic subjects, whether hypocretin-deficient (13.2 +/- 1.7 ng/mL, mean +/- SEM) or not (13.0 +/- 1.8 ng/mL), controls (10.1 +/- 1.1 ng/mL) and subjects with other sleep disorders (11.5 +/- 1.6 ng/mL). Similarly, the CSF leptin levels and the CSF: serum leptin ratios (an indicator of brain leptin uptake) were not different between groups. Serum and CSF leptin levels were higher in women and in subjects with higher body mass indexes. Leptin brain uptake decreased in women, in the aged, and in more-obese subjects. In contrast with a presumed inhibitory effect of leptin on hypocretin-containing cells, CSF leptin levels tended to correlate positively with CSF hypocretin-1 levels. C-reactive protein was higher (4.2 +/- 0.9 mg/L) in narcoleptic subjects with hypocretin deficiency than in controls (1.4 +/- 0.3 mg/L, p = .0055), a difference still significant after adjustment on confounding factors.

Our data do not support a role for leptin in mediating increased body mass index in narcolepsy. A moderate but selective increase in C-reactive protein in hypocretin-1 deficient subjects should prompt research on inflammation in narcolepsy.