To the Editor:
The article by Landry et al entitled “Vasopressin Deficiency Contributes to the Vasodilation of Septic Shock” (Circulation. 1997;95:1122–1125) is original and provocative in its findings of apparently low plasma vasopressin concentration in septic shock, coupled with the observation that infusion of vasopressin rapidly restored arterial pressure. Two points can be made regarding the conclusions and subsequent speculation about the reason for low vasopressin levels in this setting. First, the data obviously do not establish an actual contributing role for diminished arginine vasopressin levels in the hypotension of the syndrome, only that levels are low and infusion of vasopressin restores blood pressure. This observation may have therapeutic, not mechanistic, implications because, as noted by Dr Reid in the accompanying editorial, infusion of other peptides might have had similar effects. Second, although autonomic or baroreceptor dysfunction is postulated to account for the low arginine vasopressin levels, the explanation may in fact be due to the hemodynamics of septic shock coupled with normal baroreflex function. The sinoaortic baroreceptor is the dominant component of the afferent limb for nonosmotic arginine vasopressin secretion in humans.1 These receptors respond to stretch, and pulsatile load may be a factor in governing their discharge frequency,2 although mean arterial pressure is frequently the only variable cited. There is evidence that pulsatile load, especially at lower pressures, is associated with greater inhibitory effects on sympathetic activity than comparable levels of pressure with a static load3 ; the same physiology may well apply for vasopressin secretion, because the pathways are closely related. In this regard, it is worth noting that although the arterial pressure was lower in the septic shock patients than in the subjects with cardiogenic shock (and high arginine vasopressin levels), the cardiac output was much higher in the septic shock patients. The sinoaortic baroreceptor therefore is presented with a very different load in the two syndromes. There is probably a threshold pressure below which arginine vasopressin secretion will be stimulated regardless of the mechanism of the hypotension, but even in normal humans, a moderate depression in blood pressure (if caused by vasodilation) does not stimulate arginine vasopressin.4 Thus, the response of arginine vasopressin in septic shock may be entirely predictable on the basis of the physiology of the reflex involved and may be teleologically understood as a protective mechanism to avoid the possible adverse vasoconstrictive effects of arginine vasopressin in critical organ beds such as the coronary circulation.
- Copyright © 1998 by American Heart Association
Goldsmith SR. Baroreflex control of vasopressin secretion in normal humans. In: Cowley AW, Liard J-F, Ausiello DA, eds. Vasopressin: Cellular and Integrative Functions. New York, NY: Raven Press; 1988:389–397.
Chapleau MW, Abboud FM. Contrasting effects of static and pulsatile pressure on carotid baroreceptor activity in dogs. Circ Res. 1987;61:648–658.
Chapleau MW, Hajduczok G, Abboud FM. Pulsatile activation of baroreceptors causes central facilitation of baroreflex. Am J Physiol. 1989;256(part 2):H1735–H1741.
We thank Dr Goldsmith for his interesting comments. Concerning his first point, that our results do not show that vasopressin deficiency is a contributor to the vasodilation of septic shock, it is worth remembering that vascular smooth muscle tone has no fixed set point and that during profound hypotension, vasopressin is normally released and constricts the vasculature. In contrast, our patients had inappropriately low levels of plasma vasopressin despite hypotension, and correction of this deficiency with exogenous vasopressin increased vascular tone. This suggests that had endogenous vasopressin been appropriately elevated, the vasodilation in these patients would have been less pronounced.
His second point, that the high cardiac output of patients in septic shock may be responsible for the low vasopressin in plasma, is difficult to answer because solid data elucidating the respective roles of the different components of the cardiovascular system on the baroreflex control of vasopressin secretion are sorely lacking. However, we note that patients with cirrhosis provide a well-known clinical example of increased baroreflex-mediated vasopressin secretion in the presence of a high cardiac output.