Carlos A. Santacruz, Adriano J. Pereira et el published a systematic review in Critical Care Medicine about trials which showed reduced mortality.
All trials published since inception to April, 2019 were searched in MEDLINE and pubmed (Adult, multicenter randomized controlled). A total of 212 trials were included. 27 trials reported reduction in mortality. 16 trials reported increase in mortality.
Here is the summary of trials which showed improvement in mortality :-
| Author, Year | Study | Results | p value |
| Amendola 2018 | Goal directed therapy versus standard care in patient with acute kidney injury | 31% in-hospital mortality in goal-directed therapy versus 51% in usual care | 0.048 |
| Annane 2018 | Hydrocortisone and fludrocortisone versus placebo in sepsis | 43% versus 49% mortality in steroid group | 0.03 |
| De Jong 2016 | Procalcitonin guided antibiotic therapy in patient with sepsis | 20% -28 day mortality in procalcitonin group versus 25% in usual | 0.001 |
| Guerin 2013 | Prone versus supine position in ARDS | 16% mortality in prone position versus 33% in supine position | 0.001 |
| Guntupalli 2013 | Talactoferrin versus placebo in sepsis | 21% mortality in the treatment group at 6 months versus 35% | 0.03 |
| Nava 2011 | Noninvasive ventilation versus standard practice in acute hypercapnic respiratory failure | Six months mortality was 11.7% in the treatment group versus 50% in the standard practice | 0.014 |
| Papazian 2010 | Cisatracurium versus placebo in ARDS | 90 day hospital mortality was 31% in treatment group versus 44.6% in standard practice | 0.04 |
| Ferrer 2009 | Non invasive ventilation after extubation in hypercapnic patients versus conventional oxygen in chronic respiratory failure | 11% mortality in noninvasive ventilation group versus 31% | 0.02 |
| de Smet 2009 | Selective digestive tract decontamination (SDD) and selective oropharyngeal decontamination (SOD) versus usual care | Reduction of mortality by 3.5% in Selective oral decontamination group | 0.045 |
| Ferrer 2006 | Early noninvasive ventilation in ICU to prevent reintubation | Intensive care unit mortality was 3% in noninvasive ventilation group versus 14% | 0.015 |
| Villar 2006 | high PEEP, low tidal volume ventilation in ARDS | 32% mortality in treatment group versus 53% | 0.04 |
| Panacek 2004 | Afelimomab in patient with severe sepsis and elevated interleukin 6 levels | 43.6 % mortality in the treatment group versus 47.6% | 0.041 |
| Ferrer 2003 | Noninvasive ventilation during persistent weaning failure in ICU | 90% ICU survival in the treatment group versus 59 % | 0.04 |
| Ferrer 2003 | Noninvasive ventilation versus high concentration oxygen therapy in acute hypoxic respiratory failure | 18% ICU mortality in noninvasive group versus 39% | 0.028 |
| Annane 2002 | Hydrocortisone plus fludrocortisone versus placebo in sepsis | 53% , 28 day survival in treatment group versus 63% | 0.02 |
| Bernard 2001 | Drotrecogin alfa activated versus placebo in patients with severe sepsis | 24 .7% mortality in the treatment group versus 30.8% | 0.005 |
| Brower 2000 | 6 mL/kg of tidal volume versus 12mL per kg in ARDS | 31% mortality in lower tidal volume group versus 39.8% | 0.007 |
| Esteban 2000 | Pressure control ventilation versus volume control ventilation in ARDS | 51% Hospital mortality in the PCV group versus 78% in volume group | 0.02 |
| Fagon 2000 | Invasive management strategy (bronchoscopic protected specimen brush samples or bronchoalveolar lavage) for ventilator associated pneumonia | 16.2%, 14 day mortality in the invasive group versus 26% | 0.022 |
| Nava 1998 | Noninvasive mechanical ventilation in the weaning of patients with COPD who are on mechanical ventilation | 60 day survival was 92% in non invasive pressure support versus 72% in invasive group | 0.009 |
| Amato 1998 | Protective ventilation strategy in ARDS( 6 mL/kg of tidal volume, lower driving pressures with permissive hypercapnia) | 38% mortality in protective group versus 71% in conventional Group | 0.001 |
| Baudo 1998 | Anti thrombin III (in patients with low levels <70%) versus placebo in sepsis | replacement therapy reduces mortality in the subgroup of septic shock patients only | 0.03 |
| Brochard 1995 | Noninvasive ventilation in acute exacerbation of COPD | 9% in hospital mortality in noninvasive group versus 29% in standard | 0.02 |
| Fisher 1994 | Human recombinant interleukin-1 receptor antagonist in the treatment of sepsis syndrome |
human recombinant IL-1ra provided does related mortality benefit, 16% versus 44% in the highest dose | 0.015 |
| Gutierrez 1992 | Gastric intramucosal pH as a therapeutic index of tissue oxygenation in critically ill |
for those admitted with normal pHi, survival was significantly greater in the protocol than in the control group (58% vs 42%; p less than 0.01) | <0.01 |
| Dominioni 1991 | High-dose IgG versus placebo in sepsis | 38% ICU mortality in IgG group versus 67% in placebo | 0.05 |
| Ziegler 1991 | Gram-negative bacteremia and septic shock with HA-1A human monoclonal antibody against endotoxin |
30% mortality in treatment group for says 49% in placebo | 0.014 |
BOTTOMLINE
As you can see, there are whole bunch of trials, which showed improved mortality with various interventions but could not be reproduced in the future. Any new intervention, however promising , should be taken with a grain of salt. Trial of vitamin C in sepsis is eagerly awaited.