Reversal of Vasodilatory Shock: Current Perspectives on Conventional, Rescue, and Emerging Vasoactive Agents for the Treatment of Shock

Jonathan H. Chow, MD; Ezeldeen Abuelkasem, MBBCh, MSc; Susan Sankova, MD; Reney A. Henderson, MD; Michael A. Mazzeffi, MD, MPH; Kenichi A. Tanaka, MD, MSc


Anesth Analg. 2019;130(1):15-30. 

In This Article

Rescue Agents


Few classes of drugs have been as controversial and highly scrutinized as corticosteroids in septic shock. Recent studies have provided evidence for their use, and they are frequently utilized as first-line rescue agents when vasodilatory shock is refractory to fluid resuscitation and conventional vasopressors.[65,66]

Cortisol is the primary corticosteroid released by the hypothalamic–pituitary–adrenal (HPA) axis. Corticotropin-releasing hormone is released by the hypothalamus and stimulates the pituitary to release corticotropin.[67] This stimulates the adrenal cortex to release cortisol, which has critical roles in glucose metabolism and modulation of cytokines.[68] In vasodilatory shock, corticosteroids play an important role in the mediation of vasoconstriction through regulation of vascular smooth muscle sensitivity to Ang-2, NE, and Epi. Conversely, they also mediate vasodilation by decreasing production of NO, which is a potent endogenous vasodilator.[68]

Clinical Studies. In 2002, Annane et al[69] studied the effect of hydrocortisone and fludrocortisone in septic shock in a large RCT of patients with adrenal insufficiency. They found that patients receiving hydrocortisone 50 mg every 6 hours and fludrocortisone 50 μg every day for 7 days had a significant reduction in 28-day mortality (53% corticosteroid group versus 63% placebo; HR = 0.67; 95% CI, 0.47–0.95; P = .02) and a significant reduction in vasopressor use at day 28 (57% corticosteroid group versus 40% placebo; HR = 1.91; 95% CI, 1.29–2.84; P = .001).[69] The Corticosteroid Therapy of Septic Shock (CORTICUS) trial, which utilized hydrocortisone only, was not able to replicate this mortality benefit (28-day mortality 34.3% hydrocortisone versus 31.5% placebo, P = .51), although it found that hydrocortisone reversed shock more rapidly than placebo.[66] The follow-up Adjunctive Glucocorticoid Therapy in Patients with Septic Shock (ADRENAL) trial also only utilized hydrocortisone and found faster resolution of shock (median duration of hydrocortisone 3 days [interquartile range {IQR}, 2–5] versus placebo 4 days [IQR, 2–9]; HR, 1.32; 95% CI, 1.23–1.41; P < .001), albeit with no significant 28- or 90-day mortality benefit.[70]

The most recent RCT, the Hydrocortisone Plus Fludrocortisone for Adults with Septic Shock (APROCCHSS) trial, utilized both hydrocortisone and fludrocortisone for 7 days and found that this intervention improved the number of organ failure–free days (14 days intervention group versus 12 days placebo, P = .003) and vasopressor-free days at day 28 (17 days intervention group versus 15 days placebo, P < .001).[65] Furthermore, the relative risk of death was improved with this intervention (RR = 0.88; 95% CI, 0.78–0.99), and there was also significantly decreased mortality at ICU discharge, hospital discharge, and day 180, although this effect was not statistically significant at day 28 (33.7% vs 38.9%, P = .06).[65] Some key differences between APROCCHSS and ADRENAL which may explain the different reported mortality benefits include quicker randomization from the onset of shock (<24 hours APROCCHSS versus 20 ± 90 hours ADRENAL), faster administration of steroids (bolus dose versus continuous infusion without loading dose, respectively), and higher degree of shock at enrollment (mean, 0.7 μg·kg−1·min−1 NE versus 0.3 μg·kg−1·min−1 NE, respectively).[65,70]

Taken together, the data from APROCCHSS indicate that the use of steroids in vasodilatory shock is safe.[65] Although they do increase the rate of hyperglycemia, steroids do not result in an increased rate of serious adverse events, gastroduodenal bleeding, or superinfections.[65] Because of the relatively wide margin of safety, many centers routinely use 7 days of hydrocortisone and fludrocortisone as first-line rescue therapy for vasodilatory shock refractory to conventional vasopressors.

Methylene Blue

Methylene blue (MB) inhibits NO-mediated stimulation of cGMP by inhibiting NOS and guanylyl cyclase (Figure 3).[71,72] Instead of raising blood pressure by causing smooth muscle contraction, it raises blood pressure by preventing vasodilation. Because it utilizes a mechanism that is distinct from conventional vasopressors, it has generated interest as a rescue medication for refractory vasoplegia.[73]

Pharmacology. The first medical use of MB was for a histochemical stain, and it is also commonly used as an indicator dye, reducing agent for methemoglobinemia, and vasoactive agent.[71,72,74,75] The mechanism of action of MB is different from the GPCR effector vasoconstrictors. It inhibits NO-mediated stimulation of cGMP by inhibiting NOS and guanylyl cyclase (Figure 3).[71,72] Thus, MB mitigates vasodilation instead of directly stimulating smooth muscle contraction.[73]

MB is typically administered intravenously and given at a dose of 1–2 mg kg−1 over 15 minutes and may be followed by an infusion of 0.5 mg/kg over 6 hours.[72,76,77] Toxicity in humans can include hemolytic anemia and development of methemoglobinemia in patients with glucose-6-phosphate dehydrogenase deficiency.[78–80] Due to its ability to inhibit monoamine oxidase, MB carries the potential for serotonin syndrome (confusion, hyperthermia, hypertonia, clonus, etc) in patients taking serotonin reuptake inhibitors.[81–83]

Clinical Data. MB's unique mechanism of action and relative safety have likely contributed to its rapid adoption into clinical practice for the treatment of shock. In 1992, the use of MB to treat 2 patients with septic shock refractory to NE was reported.[84] MB increased MAP and SVR without affecting the HR and only slightly affected the cardiac index.[84] MB was also reported to have a positive effect on SVR for VS during liver transplantation, although follow-up studies in this population have not been able to replicate these results.[85,86]

While appropriate doubt exists about its use as a first-line vasopressor, there is moderate-quality evidence supporting the use of MB for shock refractory to high-dose vasopressors.[87–90] One RCT of 638 patients with VS following CPB found MB to significantly decrease mortality (10.7% MB versus 3.6% placebo, P = .02), although this mortality benefit has not been consistently replicated.[76,91] A single-center retrospective analysis found higher rates of morbidity, renal failure, and a trend toward increased in-hospital mortality, while another study found that those receiving MB had higher rates of postoperative complications (46.6% MB versus 14.8% non-MB, P < .0001).[91,92] However, the MB-treated group did have a higher rate of preoperative comorbidities, so at baseline, it was more likely to experience major complications, regardless of MB exposure. This same study found that when MB was administered early in the operating room, these patients did experience a significant morbidity (35.4% early versus 52.0% late, P = .062) and mortality benefit (10.4% early versus 28.6% late, P = .018), suggesting that timing and early intervention may play a critical role in ensuring maximal benefit of MB.[92]

MB appears to be a promising adjunct to noradrenergic vasopressors and vasopressin in profound vasodilatory states. While not completely without adverse effects, it does have a relatively large margin of safety in most patients making it a reasonable choice for rescue in vasodilatory shock.


Thiamine is another medication that can be used to rescue patients with vasodilatory shock refractory to conventional vasopressors. Also referred to as vitamin B1, it is a water-soluble nutrient that is an essential cofactor in the Krebs cycle.[93] Deficiency of this cofactor interrupts the oxidative energy pathway, leading to an increase in anaerobic metabolism and eventual increase in lactic acid production.[94] Due to the central role of thiamine in aerobic cellular respiration and the increasing recognition of mitochondrial dysfunction as a contributor to multiorgan dysfunction, there has been an increased interest in the use of thiamine for metabolic resuscitation and clearance of lactate.[95,96]

Clinical Data. In 1988, Cruickshank et al[97] published a retrospective analysis of 158 patients admitted to the ICU who required parenteral nutrition support. They found that surviving patients had significantly higher thiamine levels than nonsurvivors. Among patients with evidence of thiamine deficiency, mortality was significantly higher than those who were not thiamine deficient (72% mortality versus 50%, respectively, P < .05).[97] This was one of the first studies to demonstrate the role for thiamine supplementation in the treatment of critically ill patients.

A single-center retrospective study of patients in septic shock propensity matched 123 patients who received thiamine supplementation within 24 hours of admission and 246 patients who did not.[98] The most common dose was 500 mg of intravenous (IV) thiamine, and it was administered for a median of 3 days. Treatment with thiamine was associated with improved lactate clearance (HR = 1.3; 95% CI, 1.002–1.704) and improved 28-day mortality (HR = 0.666; 95% CI, 0.490–0.905).[98] An RCT examining thiamine supplementation in septic shock found that 200 mg of twice daily thiamine for 7 days resulted in significantly improved 24-hour lactate (median, 2.1 mmol/L thiamine group versus 3.1 placebo, P = .03) and 30-day mortality (13% thiamine group versus 46% placebo, P = .047) in the subgroup of patients with thiamine deficiency.[99] In the general population, however, there was no difference in the time to shock resolution, ICU length of stay (LOS), hospital LOS, or mortality rate.

Current findings suggest that the benefit of thiamine is unlikely to be universal. However, its wide safety margin, benefits in those with thiamine deficiency, and the high prevalence of this deficiency in septic shock make thiamine an attractive adjunct in the management of shock refractory to conventional vasopressors.[99]