Drew E. Carlson

The absence of circadian cues during recovery from sepsis modifies pituitary-adrenocortical function and impairs survival.

Lighting and other environmental cues in the intensive care unit rarely adhere to a consistent daily pattern. To determine the influence of the daily light/dark (LD) cycle on recovery from sepsis, male Sprague Dawley rats were acclimated to lights-on condition at 6 AM and lights-off condition at 6 PM for 6 to 14 days. Catheter placement and cecal ligation and puncture (CLP) were performed under ketamine and xylazine. Rats were returned to the established LD cycle, to constant light (LL), or to constant dark (DD) at 6 PM. One-week survival was 83.33% during LD (n = 24), 62.5% during LL (n = 16), and 31.25% during DD (n = 16; P < 0.01 for difference from the LD group). Both plasma adrenocorticotropin (ACTH) and corticosterone levels in the morning of the first day after CLP were greater during DD than during LD (P < 0.05 in each case). The early elevation in ACTH was independent of survival. However, the greater frequency of nonsurviving DD rats accounted for the elevation of corticosterone in the DD group. Overall, most nonsurviors had a unique response pattern composed of an early elevation of corticosterone in relation to plasma ACTH that then declined to a value above the normal circadian peak despite a late increase in endogenous ACTH when death was imminent. We conclude that the circadian cues provided by the LD cycle improve survival after CLP. Removal of these cues by DD increases the early appearance and incidence of a hormonal response pattern that is associated with a lethal outcome.

Detrimental effects of rapid fluid resuscitation on hepatocellular function and survival after hemorrhagic shock

Because end-organ injury can occur with reperfusion following hemorrhage or ischemia, we hypothesized that aggressive intravenous fluid resuscitation would aggravate tissue injury in a fixed-volume model of hemorrhagic shock. Unanesthetized chronically prepared male rats were hemorrhaged 33–36 mL/kg for 2.5 h. Then Lactated Ringers Solution (3x hemorrhage volume) was infused over 5 min (FAST), 20 min (MEDIUM), 180 min (SLOW), or not at all (NO RESUS). Plasma ornithine carbamoyltransferase (OCT), lactate, and creatinine were measured as indices of hepatocellular injury, anaerobic metabolism, and renal function, respectively. At 1 h post-resuscitation (PR), MAP was greater after SLOW and MEDIUM treatment (tx) than after other txs (P < 0.05). OCT increased earliest after FAST tx to values greater than those after other txs from 30 min to 24 h PR (P < 0.01). Plasma lactate was elevated immediately before resuscitation in all groups (P < 0.01) and returned to baseline at 3 h PR after SLOW tx compared to 5 h PR after FAST tx (P < 0.05). Creatinine at 5 h PR was less in the groups treated with intravenous fluid compared to the NO RESUS group, P < 0.05. Survival at 72 h was reduced in the FAST (57%) and NO RESUS (58%) groups compared to the SLOW (87%) and MEDIUM (85%) groups (P < 0.05). Thus, overly aggressive fluid tx accelerates hepatocellular injury, is no better than lesser rates of resuscitation at correcting plasma lactate and preserving renal function, and provides no overall survival benefit.

Saline resuscitation after fixed-volume hemorrhage : role of resuscitation volume and rate of infusion

The authors have reported previously that small-volume resuscitation (1.8 χ bled volume) with 0.9% NaCl restores blood volume and attenuates hormonal responses after large hemorrhage without correction of arterial hypotension. The authors studied the role of rate of infusion in this observation in chronically prepared dogs (aortic flow probe, right atrial pressure and volume, and arterial catheters) after 30% hemorrhage (24.1 ± 0.4 mL/kg). After 30 minutes, subjects were observed either without treatment (no resuscitation) or with infusion of 43 mL/kg 0.9% NaCl over 3 hours by one of three protocols: (1) impulse infusion over 10 minutes, (2) variable rate infusion, bolus with tapering infusion, or (3) constant rate infusion. Significant improvement in cardiac output and in blood volume and significant decreases of vasopressin and arterial catecholamines were observed in all fluid-treated groups. This benefit was relatively independent of rate of infusion, although impulse infusion produced greater early improvement, which dissipated with time, and constant rate infusion produced better late results. In none of the fluid-treated groups were these improvements reflected in improved mean arterial pressure compared with the no resuscitation group. The authors conclude that small-volume, slow-rate saline infusion produces physiologic benefits that cannot be assessed by easily measured clinical parameters. Thus, early resuscitation after trauma could aid patients even if arterial pressure is unchanged. This benefit might be even greater in patients with uncontrolled bleeding because arterial pressure, and hence bleeding, may not be increased by resuscitation of this type. A reassessment of the value of prehospital fluid resuscitation in the injured patient is warranted.

Central neural distribution of immunoreactive Fos and CRH in relation to plasma ACTH and corticosterone during sepsis in the rat.

Although brain pathways activated by sepsis may respond acutely to endotoxin administration, the long-term central response to sepsis is not known. We prepared male rats for hormonal sampling at the circadian nadir (AM) and peak (PM) after cecal ligation and puncture (CLP) or sham surgery. Diurnal variation of corticosterone was present on postoperative day (D) 3 and D4 after sham surgery but not after CLP. CLP increased Fos immunostaining in the nucleus of tractus solitarius (NTS), ventrolateral medulla, medullary raphe, parabrachial nucleus, hypothalamus, amygdala, bed nucleus of stria terminalis, and preoptic region. Fos responses were generally greatest on D1 but persisted to the AM of D4. The number of Fos-positive cell nuclei in the NTS on D3 and D4 did not differ but had greater variance on D3 than on D4 (P<0.01) with a divergent response in the PM of D3 that was correlated with plasma ACTH (r=0.927, P<0.01) but not with corticosterone. CLP increased CRH-staining intensity in the hypothalamic paraventricular neurons uniformly from D1 through D4 (P<0.01). Similar to Fos in NTS, this response was correlated with plasma ACTH (r=0.738, P<0.05) and adrenal size (r=0.730, P<0.05) in the PM of D3. Neuronal CRH became detectable after CLP in specific medullary areas on D1 and in the preoptic region on D3 and D4. Thus, the suppression of circadian variation by CLP was associated with central neural responses that increased in relation to plasma ACTH without apparent influence on the release of corticosterone.

Cecal ligation and puncture in rats interrupts the circadian rhythms of corticosterone and adrenocortical responsiveness to adrenocorticotrophic hormone.

Objective:To determine the altered patterns of pituitary-adrenal activity and impaired adrenocortical sensitivity to adrenocorticotrophic hormone (ACTH) in the cecal ligation and puncture (CLP) model of sepsis. Design:Prospective, controlled experiment. Setting:Basic science laboratory. Subjects:Sprague-Dawley male rats 300–450 g. Interventions:Indwelling arterial catheters and CLP with either an 18-(CLP18) or a 21-gauge needle or sham surgery. Measurements and Main Results:Plasma ACTH and corticosterone recovered most rapidly after sham surgery and least rapidly after CLP18. From postoperative day 4 am through day 7, a robust diurnal rhythm of corticosterone (p < .001) with a modest rhythm of ACTH (p < .01) occurred only in sham rats, and the slope of the regression between plasma corticosterone and ACTH increased from am to pm after sham surgery (p < .05) but not after CLP. Corticosterone in response to intravascular ACTH (3, 10, and 250 ng/kg) 2 hrs after dexamethasone (0.25 mg/kg) only showed an am to pm difference after sham surgery. The pm sham responses to all doses of ACTH were greater (p < .01) than the respective am sham responses that were not different from the respective am or pm responses after CLP. Corticosterone after 10 ng/kg ACTH in the pm decreased as plasma macrophage migration inhibitory factor and IL-6 increased after CLP (r = −.691 and r = −.813, respectively; p < .02 in each case). Conclusions:The adrenocortical sensitivity to ACTH in the pm after CLP is suppressed progressively with the intensity of inflammation. This suppression appears to be a major factor in the interruption of circadian patterns of hormonal secretion in sepsis.

Potentiated Response of Corticotropin (ACTH) to Repeated Moderate Hemorrhage Requires Amygdalar Neuronal Processing1

We examined the role of the amygdala in the potentiation of the corticotropin (ACTH) response to a 10 mg/kg hemorrhage by a 1-hour episode of equivalent hypovolemia done 24 h earlier. Unanesthetized rats were studied on the fourth (D1) and fifth (D2) day after chronic implantation of arterial and venous catheters. Immunocytochemistry for Fos protein indicated that neurons in the central and medial nuclei of the caudal amygdala were activated by hemorrhage. We then tested the effect of excitotoxic destruction of the neurons in these areas by bilateral injections of ibotenic acid 10 days prior to catheter placement. In rats that were hemorrhaged on both D1 and D2, the responses of ACTH and corticosterone increased significantly from the first (H1) to the second hemorrhage (H2) in a control group injected with saline (p < 0.05) and in lesioned groups without bilateral damage of the Fos-responsive areas (p < 0.01). In the group with bilateral damage to these sites, the responses to H1 and H2 did not differ. Additional rats had H1 on D2 to control for the long-term effects of the chronic cannulation. The responses of ACTH to H1 on either D1 or D2 did not differ between the saline-injected controls and any of the lesioned groups. In contrast, the response of ACTH to H2 on D2 in rats with bilateral damage of the caudal amygdala was not significant and was less than the response of ACTH to H2 in both rats with unilateral damage of this area (p < 0.05) and those injected with saline (p < 0.05). We conclude that bilateral neuronal processing within the caudal amygdala is required for the potentiation of the response of ACTH to H2 by H1.

Messenger RNA for neuropeptide Y in the arcuate nucleus increases in parallel with plasma adrenocorticotropin during sepsis in the rat

Loss of appetite occurs in the cecal ligation and puncture (CLP) model of sepsis in conjunction with the activation of central neural stress pathways. Neuropeptide Y (NPY) in the arcuate nucleus of the hypothalamus is upregulated by several stressors and is stimulatory to feeding. To examine the response of NPY messenger RNA in the arcuate nucleus to sepsis, we used biotinylated RNA probes and a quantitative non-isotopic in situ hybridization approach in cryo-preserved sections from rats made septic by CLP. The mRNA in arcuate neurons was upregulated from the first day after CLP. By the afternoon of the third day through the morning of the fourth day, the average grey level of NPY mRNA clusters was 30% greater after CLP than after sham surgery (P<0.05), and the integrated optical density based on both the grey level and the amount of area with detectable mRNA was 60% greater after CLP than after sham surgery (P<0.03). Both the average grey level and area with detectable staining were positively correlated to plasma ACTH (r=0.953 and 0.917, respectively, n=10 and P<0.01 in each case). Thus sepsis increases the expression of the mRNA for NPY in the arcuate nucleus in proportion to the magnitude of the stress response. However, the suppression of feeding behavior in the CLP model suggests that sepsis activates additional mechanisms that negate the orexigenic contribution of the neuronal increase in NPY mRNA.

Acute Trauma Response

The acute trauma response consists of multiple biological components reacting to various signals from local wounded tissue and systemic alterations. In severe injury, exaggerated and unchecked release of mediators result in detrimental systemic effects and may lead to multiple organ failure. In addition to host endogenous modulation of these mediators, timely therapeutic interventions may modify local inflammation, neuroendocrine signals, and systemic alterations. Recovery from trauma ultimately depends upon the coordination of the neuroendocrine, immune, and metabolic systems toward restoring homeostasis and survival. Future research in this area may elucidate new therapeutic options for the critically injured trauma patient.

Hypotensive hemorrhage increases calcium uptake capacity and Bcl-XL content of liver mitochondria.

We tested the hypothesis that the response of mitochondrial uptake of calcium and content of Bcl proteins to reversible hemorrhagic shock increases the vulnerability for hepatocellular death. Pentobarbital-anesthetized rats were bled to a mean arterial pressure of 30 to 40 mmHg for 1 h. A subset was then resuscitated (isotonic sodium chloride solution, three times shed volume). Liver mitochondria were isolated at the end of hemorrhage and 1.5 h after the onset of resuscitation. Resuscitation accelerated mitochondrial respiration in the presence of adenosine diphosphate (state 3) above control (P < 0.01). The respiratory control ratio ([RCR] state 3/state 4) was calculated using the respiratory rate in the presence of carboxyatractyloside (state 4). The RCR was depressed at the end of hemorrhage and recovered completely in response to resuscitation (P < 0.05). The mitochondrial capacity for calcium uptake increased at the end of hemorrhage and remained greater than control (P < 0.01) after resuscitation when plasma ornithine carbamoyltransferase (an index of hepatocellular injury) was greater than control (P < 0.05). At this time, the capacity for calcium uptake was correlated with plasma ornithine carbamoyltransferase (r = 0.819, P < 0.01). Mitochondrial content of Bcl-xL, an antiapoptotic protein, was increased at the end of hemorrhage (P < 0.03) with no further change after resuscitation and no change in mitochondrial Bak, a proapoptotic protein. Thus, mitochondrial mechanisms are triggered early during reversible hypovolemia that may limit the intensity of intracellular calcium signaling and its potential to cause cellular injury and death.

A circulating shock protein that depolarizes cells in vitro depresses myocardial contractility and rate in isolated rat hearts.

Previously, we identified the presence of a circulating shock protein (CSP) in the plasma of hemorrhaged rats that depolarizes a variety of cells in vitro. In isolated perfused rat hearts, partially purified CSP produced dose-dependent decreases in contractility and heart rate associated with an increase in coronary perfusion pressure (CPP). Electrical pacing failed to prevent the negative inotropic effects. Preventing the coronary vasoconstriction with nitroglycerin or attenuating it with a cyclooxygenase inhibitor also failed to prevent the inotropic or chronotropic effects of CSP. Carbocyclic thromboxane A2 (50ng/min) caused a similar increase in CPP to CSP but had no effect on contractility or rate during the first minute of infusion. These data indicate that the protein that appears in rat plasma after hemorrhage produces negative inotropic and chronotropic effects on the isolated heart that are independent of changes in CPP. Vasoactive arachidonic acid metabolites elicited by CSP are partially responsible for the increase in coronary vascular resistance.