High bicarbonate levels in narcoleptic children.

The objective of this study was to evaluate the levels of plasma bicarbonate levels in narcoleptic children. Clinical, electrophysiological data and bicarbonate levels were evaluated retrospectively in children seen in our paediatric national reference centre for hypersomnia. The cohort included 23 control subjects (11.5 ± 4 years, 43% boys) and 51 patients presenting de-novo narcolepsy (N) (12.7 ± 3.7 years, 47% boys). In narcoleptic children, cataplexy was present in 78% and DQB1*0602 was positive in 96%. The control children were less obese (2 versus 47%, P = 0.001). Compared with control subjects, narcoleptic children had higher bicarbonate levels (P = 0.02) as well as higher PCO2 (P < 0.01) and lower venous pH gas (P < 0.01). Bicarbonate levels higher than 27 mmol L(-1) were found in 41.2% of the narcoleptic children and 4.2% of the controls (P = 0.001). Bicarbonate levels were correlated with the Adapted Epworth Sleepiness Scale (P = 0.01). Narcoleptic patients without obesity often had bicarbonate levels higher than 27 mmol L (-1) (55 versus 25%, P = 0.025). No differences were found between children with and without cataplexy. In conclusion, narcoleptic patients had higher bicarbonate plasma levels compared to control children. This result could be a marker of hypoventilation in this pathology, provoking an increase in PCO2 and therefore a respiratory acidosis, compensated by an increase in plasma bicarbonates. This simple screening tool could be useful for prioritizing children for sleep laboratory evaluation in practice.