Cerebrospinal fluid to serum glucose ratio in non-hypoglycorrhachic neurological conditions.

OBJECTIVE To explore the relevance of cerebrospinal fluid to serum glucose ratio in non-hypoglycorrhachic conditions. DESIGN Retrospective observational study. SETTING Neurology ward, university teaching hospital, Hong Kong. PATIENTS Adult patients with conditions unrelated to hypoglycorrhachia who underwent lumbar puncture. MAIN OUTCOME MEASURES Cerebrospinal fluid and simultaneous serum glucose concentrations, and their ratio to each other. RESULTS Between September 1998 and August 2003, 170 cerebrospinal fluid and serum glucose samples were collected from 138 patients. Mean cerebrospinal fluid to serum glucose ratio was 0.61 (standard deviation, 0.142; range, 0.21-1.00). With the exception of cerebrospinal fluid protein level, laboratory parameters were similar among different diseases. The glucose ratio was lower than 0.6 in 43% and lower than 0.5 in 19% of samples. Cases with a low glucose ratio appeared to have higher serum glucose concentrations (significant among groups with different glucose ratios, P<0.001). The mean glucose ratio (0.65) was also significantly higher in patients with serum glucose concentration of lower than 7.8 mmol/L compared with those with serum glucose concentration between 7.8 and 11.1 mmol/L (mean, 0.46), or higher than 11.1 mmol/L (mean, 0.46) [P<0.001]. There was a strong negative correlation between the glucose ratio and serum glucose concentration (r= -0.704, P<0.001). CONCLUSION A lowered cerebrospinal fluid to serum glucose ratio is often seen in the absence of an appropriate disorder, especially when simultaneous serum glucose concentration is elevated. This may be explained by the saturation kinetics of glucose transportation in hyperglycaemia, and the time lag for cerebrospinal fluid and glucose to equilibrate when the blood level fluctuates.