Sympathetic nerve activity and heart rate variability during severe hemorrhagic shock in sheep.

INTRODUCTION: In this study we explored direct and indirect measures of autonomic nervous system function, as well as changes in cardiovascular complexity, during hemorrhagic shock (HS).

METHODS: HS was induced in anesthetized sheep (n=8) by removing 40 ml/kg of blood in four 10 ml/kg steps over 40 min. Resuscitation was performed with lactated Ringer's and re-infusion of shed blood. Renal sympathetic nerve activity (RSNA) was measured by microneurography. Spectral analysis of heart rate variability (HRV) employed fast-Fourier transformation of the R-to-R interval (RRI) of the EKG. This generated the normalized high-frequency (HFnu) and low-frequency (LFnu) powers of the RRI, and their ratio (LFnu/HFnu, a proposed index of sympatho-vagal balance). Additionally, non-linear methods were applied: RRI complexity was measured by approximate (ApEn) and sample (SampEn) entropy methods; RRI fractal dimension was measured by curve lengths (FDCL). Plasma catecholamines were determined by HPLC.

RESULTS: The model caused profound HS; 2/8 animals survived till the end of resuscitation. RSNA increased in 7/8 sheep and, as HS progressed, multiple burst complexes were identified followed by sympathetic withdrawal. Concomitant decreases in HFnu and increases in LFnu/HFnu occurred after 20 ml/kg blood loss. ApEn and FDCL decreased after withdrawal of 40 ml/kg of blood. Catecholamine concentrations increased throughout HS. LFnu/HFnu and RSNA were not linearly correlated.

CONCLUSIONS: HS led to an increase in RSNA with subsequent withdrawal. LFnu/HFnu increased during HS in association with vagal withdrawal and loss of RRI complexity. RRI complexity may in part reflect vagal modulation of the heart rate. Changes in directly measured tonic sympathetic traffic do not correlate with non-invasive measures of autonomic modulation of the heart.