Current and emerging therapeutic options for refractory septic shock: A systematic review

Table 1 Risk of bias assessment

Ref.	Randomization process	Deviations from intended interventions	Missing outcome data	Measurement of outcomes	Selection of the reported results	Overall risk of bias
Lyu et al[15], 2022	Low risk	Low risk	Low risk	Low risk	Low risk	Low risk
Ibarra-Estrada et al[16], 2023	Low risk	Low risk	Low risk	Low risk	Low risk	Low risk
Hajjar et al[17], 2019	Low risk	Low risk	Low risk	Low risk	Low risk	Low risk
Moskowitz et al[18], 2020	Low risk	Some concerns	Low risk	Low risk	Some concerns	Some concerns
Fujii et al[19], 2020	Low risk	Low risk	Low risk	Some concerns	Some concerns	Some concerns
Sevransky et al[20], 2021	Low risk	Low risk	Low risk	Low risk	Low risk	Low risk
Wang et al[21], 2022	Some concerns	Some concerns	Some concerns	Some concerns	Some concerns	Some concerns
Cherukuri et al[22], 2019	Low risk	Low risk	Low risk	Low risk	Low risk	Low risk
Douglas et al[23], 2020	Low risk	Low risk	Low risk	Low risk	Low risk	Low risk
Morelli et al[24], 2008	Some concerns	Some concerns	Some concerns	Some concerns	Some concerns	Some concerns
Albanèse et al[25], 2005	Low risk	High Risk	Some concerns	High Risk	High Risk	High Risk
Hyvernat et al[26], 2016	Low risk	Low risk	Low risk	Low risk	Low risk	Low risk
Annane et al[27], 2002	Low risk	Low risk	Low risk	Low risk	Low risk	Low risk
Liu et al[28], 2018	Some concerns	Low risk	Some concerns	Some concerns	Some concerns	Some concerns
Xiao et al[29], 2016	Some concerns	Some concerns	Some concerns	Some concerns	Some concerns	Some concerns

Table 2 Study description

Ref.	Type of study	Type of patients	Primary outcomes	Number of patients	Mean age	Mortality
Lyu et al[15], 2022	RCT, single-center, double-blind	Adult patients with septic shock	Mortality at 90 days	Intervention Group: 213 patients received a combination of hydrocortisone, vitamin C, and thiamine. Control Group: 213 patients received 0.9% saline	64	90-day Mortality: 40.4% in the intervention group vs 39.0% in the placebo group. 28-day Mortality: 37.1% in the intervention group vs 36.2% in the placebo group
Ibarra-Estrada et al[16], 2023	RCT, double-blind	Adults Diagnosed with septic shock due to suspected or confirmed infection, requiring NE to maintain a MAP ≥ 65 mmHg and serum lactate > 2 mmol/L after adequate fluid resuscitation	28-day mortality	Methylene blue group: 45 patients received methylene blue as an adjunctive therapy. Placebo group: 46	46-47	28-mortality: MB 33% and P 46%
Hajjar et al[17], 2019	RCT, double-blind	Adults (≥ 18 years) with cancer admitted to the ICU, showing documented or strong clinical suspicion of infection and meeting at least two criteria of the systemic inflammatory response syndrome	28-day all-cause mortality	Vasopressin group: 125 patients NE group: 125 patients	NR	28-day mortality: Vasopressin group: 56.8%. NE group: 52.8%. 90-day mortality: Vasopressin group: 72.0%. NE group: 75.2%
Moskowitz et al[18], 2020	RCT, multicenter, double-blind	Adult patients with septic shock with infusion vasopressors	Change in SOFA score	Intervention: 100 patients received a combination of vitamin C, hydrocortisone, and thiamine. Control: 100 patients received a placebo	68	30-day mortality: Intervention group: 34.7%, placebo group: 29.3%
Fujii et al[19], 2020	RCT, multicenter, open label	Adults with documented infection, increase of at least 2 points in the SOFA score. Lactate levels greater than 2 mmol/L. Vasopressor dependency for at least 2 hours prior to enrollment	Time alive and free from vasopressors	Intervention Group: 109 patients received intravenous vitamin C, hydrocortisone, and thiamine. Control group: 107 patients received intravenous hydrocortisone alone	61.7	Intervention group: 28.6% mortality. Control group: 24.5% mortality
Sevransky et al[20], 2021	RCT, double-blind, adaptive	Adults with sepsis requiring mechanical ventilation or high-flow oxygen support due to respiratory failure. Continuous vasopressor support to maintain adequate blood pressure	days free of vasopressors and mechanical ventilation in the first 30 days	Treatment group: 1000 patients received combination therapy with vitamin C, thiamine, and corticosteroids. Control group: 1000 patients received matching placebos for each active agent	65	Interrupted before ending
Wang et al[21], 2022	Pilot RCT	Adults (≥ 18 years) with septic shock. NE dose ≥ 15 µg/minute	Renal perfusion: Assessed through changes in urine output and renal function parameters	22 patients: TP group: 10 patients. Usual care group: 12 patients	62.5	27.3% for the TP group and 41.7% for the usual care group
Cherukuri et al[22], 2019	RCT, double-blind	Adult patients with sepsis or septic shock	Length of ICU stay: Days on ventilator: Days on intravenous blood pressure support: 28-day mortality rate	Intervention: 32 patients received 100000 IU of vitamin A intramuscularly daily for 7 days. Control: 32 patients received a blinded placebo	51	28-Day Mortality Rates. Vitamin A group: 34%. Placebo group: 28%
Douglas et al[23], 2020	RCT, multicenter	Adults who presented with signs of sepsis and hypotension (MAP ≤ 65 mmHg) after receiving between 1 L and 3 L of fluids	Positive fluid balance at 72 hours or at ICU discharge, whichever occurred first	83 patients in the intervention group and 41 patients in the usual care group	62	Intervention group: 20.5%. Control group: 24.4%
Morelli et al[24], 2008	RCT	Adult patients with septic shock	The mortality rate is 28 days after treatment initiation. The outcome aimed to assess the effectiveness of combined dobutamine and TP treatment in improving survival rates in patients with septic shock	60	65	TG (dobutamine and TP): What is the primary and secondary outcome. 40% mortality. CG (standard treatment): 60% mortality
Albanèse et al[25], 2005	RCT, open-label study	Diagnosed with hyperdynamic septic shock after fluid resuscitation-Exhibited hemodynamic instability (MAP ≤ 60 mmHg). Had two or more organ dysfunctions	MAP achieved in patients with hyperdynamic septic shock after treatment with either NE or TP	20 patients, NE group: 10 patients, TP group: 10 patients	65	30% among the patients with hyperdynamic septic shock
Hyvernat et al[26], 2016	RCT, double-blind study	Persistent hypoperfusion despite adequate fluid resuscitation and NE administration. Defined as severe sepsis with arterial hypotension (SBP < 90 mmHg or MAP < 70 mmHg) despite adequate fluid resuscitation	Mortality in 28-day	59 in the 200 mg group and 63 in the 300 mg group	64.8	200 mg group: 52.5%. 300 mg group: 44.4%
Annane et al[27], 2002	RCT double-blind, parallel group	Adults (> 18 years). Required NE to maintain MAP. Urinary output less than 0.5 mL/kg/hour for at least 1 hour. Arterial lactate levels higher than 2 mmol/L	28-day survival	Corticosteroid group: Number of Patients: 151. Treatment: Received hydrocortisone (50 mg IV every 6 hours) and fludrocortisone (50 µg orally once daily) for 7 days. Placebo group: Number of patients: 149. Treatment: Received matching placebos for 7 days	63	Corticosteroid group: 43.0% (65 out of 151 patients). Placebo group: 49.7% (74 out of 149 patients)
Morelli et al[35], 2009	Pilot RCT	Patients diagnosed with septic shock. MAP below 65 mmHg despite adequate volume resuscitation	The primary outcome was the total NE dose required to maintain a MAP of 65-75 mmHg over the 48-hour treatment period	TP group: 15 patients. Vasopressin group: 15 patients. NE gsroup: 15 patients	65	Zero during 48 hours
Lv et al[36], 2017	RCT, double-blind	Adults (≥ 18 years). Onset of septic shock within 6 hours	28-day all-cause mortality	Hydrocortisone group: 58 patients. Placebo group: 60 patients	Hydrocortisone group: 65.4 years. Placebo group: 66.1 years	28-day mortality: Hydrocortisone group: 23.1%. Placebo group: 32.2%. In-Hospital mortality: Hydrocortisone group: 31.0%. Placebo group: 40.0%
Russell et al[32], 2008	RCT, multicenter, double-blind	Adults diagnosed with septic shock requiring vasopressor support, NE	28-day all-cause mortality	Vasopressin group: 396 patients. NE group: 382 patients	64	28-day mortality: Vasopressin group: 35.4%. NE group: 39.3%. 90-day mortality: Vasopressin group: 43.9%. NE group: 49.6%
Russell et al[31], 2009	Post hoc, multicenter, blinded RCT substudy	(≥ 16 years). Diagnosed with septic shock, ongoing hypotension requiring at least 5 µg/minute of NE infusion for a minimum of 6 hours	28-day mortality	Corticosteroid treatment group: Vasopressin + corticosteroids: 296 patients. NE + corticosteroids: 293 patients. Corticosteroid treatment group: Vasopressin: 93 patients. NE: 97 patients	60	28-day mortality overall: Corticosteroids + vasopressin: 35.9% corticosteroids + NE: 44.7%. No corticosteroids + vasopressin: 66.7%. No corticosteroids + NE: 62.9%
Yildiz et al[37], 2011	RCT, double-blind	Adults (≥ 18 years). Confirmed sepsis, characterized by infection and systemic inflammatory response syndrome. Patients were often classified based on severity scores, such as APACHE II or SOFA, indicating varying degrees of organ dysfunction	28-day mortality	Steroid group: 100 patients. Placebo group: 100 patients	60	Steroid group: 16 (59.3) and placebo group: 15 (53.6)
Keh et al[30], 2003	RCT, double-blind crossover	Proven or strongly suspected infection. Presence of three or more of the following conditions: Mechanical ventilation. Heart rate > 90 beats per minute. Temperature > 38 °C or < 36 °C. White blood cell count > 12000 cells/µL or < 4000 cells/µL, or > 10% immature cells. Sepsis-induced hypotension: SBP < 90 mmHg or a reduction of > 40 mmHg from baseline	Change in MAP and systemic vascular resistance after treatment with low-dose hydrocortisone compared to placebo	Hydrocortisone group: 20. Placebo group: 20	Hydrocortisone group: 54. Placebo group: 50	Hydrocortisone group: 30%. Placebo group: 30%
Keh et al[38], 2016	RCT, double-blind	Adults (≥ 18 years). Evidence of infection Systemic response to infection (at least 2 systemic inflammatory response syndrome criteria). Organ dysfunction present for no longer than 48 hours. Not in septic shock at the time of randomization	Development of septic Shock: Within 14 days of treatment	Hydrocortisone group: 190. Placebo group: 190	65	28-day mortality: Hydrocortisone group: 21.1%. Placebo group: 23.7%. 90-day mortality: Hydrocortisone group: 28.4%. Placebo group: 30.5%. 180-day mortality: Hydrocortisone group: 34.2%. Placebo group: 36.8%
Annane et al[34], 2018	RCT, multicenter, double-blind	Adults (≥ 18 years). Diagnosed with septic shock requiring vasopressor therapy	90-day all-cause mortality	Hydrocortisone plus fludrocortisone: 614. Placebo: 626	63	28-day mortality: Hydrocortisone plus Fludrocortisone: 20%. Placebo: 25%. 90-day all-cause mortality: Hydrocortisone plus fludrocortisone group: 30%. Placebo group: 36%
Venkatesh et al[33], 2019	RCT, multicenter, double-blind	Adults undergoing mechanical ventilation. Documented or strongly suspected infection. Treated with vasopressors or inotropic agents at least 4 hours prior to randomization	90-day all-cause mortality	Hydrocortisone: 1988. Placebo: 1902	63	28-day mortality: Hydrocortisone group: 197. Placebo group: 199. 90-day all-cause mortality: Hydrocortisone group: 511; placebo group: 526
Liu et al[28], 2018	RCT, multicenter, double-blind	Adults with hypotension despite adequate fluid resuscitation, and at least two diagnostic criteria for systemic inflammatory response syndrome	28-day mortality	TP group: 312 patients. NE group: 305 patients	61	28-day mortality: TP group: 40%; NE group: 38%
Xiao et al[29], 2016	RCT	Adults with SBP < 90 mmHg and MAP < 65 mmHg are required to use vasoactive agents	7-day mortality	NE group: 17, NE + TP group: 15	63.2	7-day mortality: NE group 76.5% and NE + TP 33.3%

APACHE II: Acute Physiology and Chronic Health Evaluation II; ICU: Intensive care unit; MAP: Mean arterial pressure; NE: Norepinephrine; RCT: Randomized clinical trial; SOFA: Sequential Organ Failure Assessment; SBP: Systolic blood pressure; TP: Terlipressin.

Table 3 Secondary endpoints of the studies

Ref.	Secondary outcomes	Adverse events
Lyu et al[15], 2022	28day mortality; ICU mortality; Hospital mortality; reversal of shock; time to shock reversal; 12 hours delta SOFA score; ICU free days; vasopressor free days; ventilator support free days; length of stay in ICU; length of stay in hospital	Hyperglycemia; hypernatremia; fluid overload
Ibarra-Estrada et al[16], 2023	Hemodynamic parameters; organ dysfunction; length of stay; adverse events	Hypotension; serotonin syndrome; there was a risk of serotonin syndrome, particularly in patients taking selective serotonin reuptake inhibitors. Allergic Reactions to methylene blue. Other minor events; nausea, vomiting, and local infusion site reactions
Hajjar et al[17], 2019	90 days all cause mortality; days free from advanced support; adverse effects	Arrhythmia
Moskowitz et al[18], 2020	Kidney failure; 30 day mortality; ventilator free days; shock free; ICU free days	Hyperglycemia; hypernatremia; new hospital acquired infections
Fujii et al[19], 2020	90 days mortality; duration of ICU stay; duration of hospital stay; The total length of time patients remained hospitalized. Organ dysfunction scores; adverse events	Intervention, 10 (10.2%) vs usual care, 7 (15.6%). Not related
Sevransky et al[20], 2021	Mortality at 30 days, ICU mortality, mortality at 180 days, length of ICU stay, length of hospital stay, and longterm emotional and cognitive outcomes at 180 days
Wang et al[21], 2022	Mortality rate at 28 days. Hemodynamic parameters; Including MAP and NE requirements. Adverse events related to TP use. Changes in serum creatinine levels and other renal function indicators	Cardiovascular events; increased heart rate or arrhythmias. Renal events; worsening renal function or acute kidney injury. Gastrointestinal events; ischemic colitis or other gastrointestinal complications. New infections or worsening of existing infections. Injection Site Reactions; Localized reactions at the site of TP administration
Cherukuri et al[22], 2019	Serum vitamin A levels; cortisol response; incidence of adverse effects	Nausea; injection site reactions
Douglas et al[23], 2020	RRT; mechanical ventilation; discharge rates; mortality rates	Arrhythmias; NE extravasation
Morelli et al[24], 2008	NE requirements; systemic and regional hemodynamic; organ function; SvO2	TP; cardiac ischemia; decreased cardiac output; gastrointestinal ischemia; skin reactions as pallor due to vasoconstriction. Dobutamine; tachyarrhythmias; increased myocardial oxygen demand; hypotension
Albanèse et al[25], 2005	Cardiac index; oxygen delivery index; oxygen consumption index; renal function; blood lactate levels	NE; tissue ischemia due to excessive vasoconstriction; arrhythmias or increased heart rate; increased oxygen demand of tissues; decreased mesenteric blood flow; TP; decreased cardiac index and potential for reduced oxygen delivery; bradycardia; fluid overload is used with caution in patients with renal impairment
Hyvernat et al[26], 2016	Vasopressor requirements, infectious and digestive complications, hemodynamic responses, days free from mechanical ventilation, vasopressors, and renal replacement therapy, adverse events	Haemorrhagic events; superinfections
Annane et al[27], 2002	Mortality rates; overall mortality rates at various time points (e.g., 7 days, 14 days). Time to vasopressor withdrawal; adverse events; clinical improvement	Infections; increased incidence of new infections in both groups. Gastrointestinal issues; stress ulcers or gastrointestinal bleeding. Hyperglycemia; observed in the corticosteroid group. Electrolyte Imbalances; Including hyponatremia and hyperkalemia. Neurological Events; Such as delirium or altered mental status
Morelli et al[35], 2009	Hemodynamic parameters; organ function; adverse events; mortality	Tachyarrhythmias; 1 in the vasopressin group. 4 in the NE group. Changes in bilirubin levels; higher total and direct bilirubin concentrations in the vasopressin and NE groups; platelet count; a decrease in platelet count was observed only in the TP group over time. Renal function; RRT was noted in the NE group
Lv et al[36], 2017	In hospital mortality; reversal of shock; duration of ICU stay; duration of hospital stay	Hyperglycemia; hydrocortisone group; 90.9%; placebo group; 81.5%
Russell et al[32], 2008	90 days mortality; organ dysfunction; adverse events; vasopressor requirements	Cardiac events (e.g., arrhythmias); Ischemic events (e.g., limb ischemia); other complications related to the use of vasopressors
Russell et al[31], 2009	Duration of vasopressor support; sock reversal; organ dysfunction; length of ICU stay; adverse events	Arrhythmias; limb ischemia; gastrointestinal ischemia
Venkatesh et al[33], 2018	28 days all cause mortality; Evaluated to assess shorter term effects. Duration of vasopressor use; measured to determine the impact on hemodynamic support. Duration of mechanical ventilation; assessed to evaluate respiratory support needs. Organ dysfunction; quality of life	Hyperglycemia; gastrointestinal bleeding; secondary infections
Yildiz et al[37], 2011	Adverse events; hormonal levels; incidence of adrenal insufficiency (AI) and Relative Adrenal Insufficiency (RAI); Assessment of adrenal function in patients; APACHE II and SOFA scores; clinical characteristics	No serious adverse events were reported in either the steroid or placebo groups
Keh et al[30], 2003	Inflammatory markers; NE requirements; immune function; mortality rates	Infections; increased incidence of secondary infections in both groups, but no significant difference between hydrocortisone and placebo. Gastrointestinal bleeding, though it was not significantly higher in the hydrocortisone group. Hyperglycemia; the hydrocortisone group required insulin management for some patients
Keh et al[38], 2016	Time until septic shock; mortality 28 days, 90 days, 180 days; secondary infections; weaning failure; muscle weakness; hyperglycemia	Hyperglycemia; secondary infections; muscle weakness; weaning failure
Annane et al[34], 2018	Mortality rates; days alive and free of vasopressors; organ failure; mechanical ventilation	Gastrointestinal bleeding; superinfection; hyperglycemia; neurologic sequelae
Liu et al[28], 2018	change in sofa score; days alive and free of vasopressors; incidence of serious adverse events	Digital ischemia; severe diarrhea; arrhythmias; intestinal ischemia
Xiao et al[29], 2016	hemodynamic parameters; complications; success rate for 6 hours resuscitation goals	Cardiac arrhythmias; skin necrosis at infusion sites; other organ dysfunctions

APACHE II: Acute Physiology and Chronic Health Evaluation II; ICU: Intensive care unit; NE: Norepinephrine; SOFA: Sequential Organ Failure Assessment; TP: Terlipressin; SvO₂: Oxygen venous saturation.

Table 4 Vasoactive drugs

Ref.	NE dosage	TP dosage	Vasopressin dosage	Dobutamine dosage
Lyu et al[15], 2022	Not specified
Ibarra-Estrada et al[16], 2023	Not specified		0.03 IU/minute if NE dose reached ≥ 0.25 μg/kg/minute
Hajjar et al[17], 2019	Infusion was started at 5 mL/hours and increased by 2.5 mL/hours every 10 minutes to reach a maximum target rate of 30 mL/hours		Ranged from 0.01 to 0.06, IU/minute
Moskowitz et al[18], 2020	Not specified
Fujii et al[19], 2020	Initial dose: 0.05 to 0.5 μg/kg/minute, Titration: Dose adjusted based on the patient's blood pressure response, aiming to maintain a MAP ≥ 65 mmHg
Sevransky et al[20], 2021	Initial dose: Start at 0.05 to 0.1 μg/kg/minute. Titration: Adjust as necessary to achieve a MAP ≥ 65 mmHg
Wang et al[21], 2022	NE greater than or equal to 15 μg/minute (Norepinephrine ≥ 15 μg/minute)
Cherukuri et al[22], 2019	Not specified
Douglas et al[23], 2020	Initial dose: 0.05 to 0.1 μg/kg/minute; Titration: Dose increased based on the patient's response, often up to 0.5 μg/kg/minute or higher if necessary to achieve and maintain a MAP of ≥ 65 mmHg
Morelli et al[24], 2008	Continuous infusion: 0.9 mg/kg/minute to maintain MAP (MAP) at 70 mmHg	Single dose: 1 mg administered as a bolus infusion		Initial dose: 3 mg/kg/minute. Titration: The dose was progressively increased in increments of 1 to 3 mg/kg/minute to reverse the anticipated decrease in mixed SvO2 caused by the infusion of TP
Albanèse et al[25], 2005	Initial Dose: 0.3 μg/kg/minute. Titration: Increased by increments of 0.3 μg/kg/minute at 4-minute intervals to achieve a target MAP of 65 to 75 mmHg	Initial bolus: 1 mg (equivalent to 0.03-0.04 UI/minute). Second bolus: An additional 1 mg was given if the MAP remained below 65 mmHg after 20 minutes
Hyvernat et al[26], 2016	Initial dosage: The starting dose typically ranged from 0.05 to 0.5 μg/kg/minute and was titrated based on the patient's response. Titration: Doses were adjusted as needed to achieve the target MAP
Annane et al[27], 2002	Systolic arterial pressure lower than 90 mmHg for at least 1 hours despite adequate fluid replacement and administration of more than 5 μg/kg of dopamine, or current treatment with epinephrine or NE
Morelli et al[35], 2009	NE 15 μg/minute	Terlipressin 1.3 μg/kg/hour	Vasopressina 0.03 U/minute
Lv et al[36], 2017	Not specified
Russell et al[32], 2008	5 μg of NE per minute
Russell et al[31], 2009	5 μg of NE per minute
Venkatesh et al[33], 2019	Not specified
Yildiz et al[37], 2011	Not specified
Keh et al[30], 2003	Not specified
Keh et al[38], 2016	Not specified
Annane et al[34], 2018	0.5 to 1.0 µg/kg/minute to maintain MAP
Liu et al[28], 2018	starting dose of 4 µg/min, which could be titrated up to a maximum of 30 µg/minute	20 µg/hour, with the option to titrate up to a maximum of 160 µg/hour
Xiao et al[29], 2016	From 0.5 to 2.22 µg/kg/minute to maintain MAP	1.3 mg/kg/hour

MAP: Mean arterial pressure; NE: Norepinephrine; SvO₂: Oxygen venous saturation; TP: Terlipressin.