Research article

Covariates of corticotropin-releasing hormone (CRH) concentrations in cerebrospinal fluid (CSF) from healthy humans

James N Baraniuk¹ , Hilda Maibach¹ , Gail Whalen¹ and Daniel J Clauw²

¹Division of Rheumatology, Immunology and Allergy, Room GL-002, Lower Level Gorman Building, Georgetown University, 3800 Reservoir Road, NW, Washington, DC 20007-2197

²Center for the Advancement of Clinical Research, The University of Michigan, Ann Arbor, MI

author email corresponding author email

BMC Neuroscience 2004, 5:58doi:10.1186/1471-2202-5-58

Published:	17 December 2004

Abstract

Background

Define covariates of cerebrospinal corticotropin-releasing hormone (CRH) levels in normal humans. CRH_CSF was measured in 9 normal subjects as part of an intensive study of physiological responses stressors in chronic pain and fatigue states. CRH_CSF was first correlated with demographic, vital sign, HPA axis, validated questionnaire domains, baseline and maximal responses to pain, exercise and other stressors. Significant factors were used for linear regression modeling.

Results

Highly significant correlations were found despite the small number of subjects. Three models were defined: (a) CRH_CSF with blood glucose and sodium (explained variance = 0.979, adjusted R² = 0.958, p = 0.02 by 2-tailed testing); (b) CRH_CSF with resting respiratory and heart rates (R² = 0.963, adjusted R² = 0.939, p = 0.007); and (c) CRH_CSF with SF-36 Vitality and Multidimensional Fatigue Inventory Physical Fatigue domains (R² = 0.859, adjusted R² = 0.789, p = 0.02).

Conclusions

Low CRH_CSF was predicted by lower glucose, respiratory and heart rates, and higher sodium and psychometric constructs of well being. Responses at peak exercise and to other acute stressors were not correlated. CRH_CSF may have reflected an overall, or chronic, set-point for physiological responses, but did not predict the reserves available to respond to immediate stressors.