A re-examination of the recent RCT on fluid restriction

Fluid administration in septic shock is one of the most important therapy adopted to improve haemodynamics when preload dependency is present. Indeed, recent guidelines[1] suggest to give up to 30 ml/kg of crystalloids to patients diagnosed with septic shock, in order to achieve adequate haemodynamic stabilisation in the first hours since shock recognition. Thereafter, careful evaluation of preload adequacy using common indexes of fluid responsiveness is suggested. However, a recent survey [2] has highlighted how fluid administration is more often guided by clinical parameters than rigorous evaluation of fluid responsiveness. Many papers [3-5] have been published in the last few years, investigating the effect of excessive fluid load with positive fluid balance, on survival. All the findings lead to the conclusion that patients with more positive fluid balance are more likely to die, also after correction for confounding variables.

Hjortrup et al. [6], recently published a pilot feasibility RCT to explore the effect of a protocol restricting resuscitation fluid after initial resuscitation in ICU patients with septic shock compared to standard therapy. The authors enrolled 151 adult patients with septic shock, who had fulfilled sepsis criteria within 24 hours and they tested a more restrictive protocol for fluid administration, which resulted in a significant reduction of resuscitation fluids given to the treatment group both at day five and during ICU stay (mean differences −1.2 L [95 % CI −2.0 to−0.4), p < 0.001 and −1.4 L [95% CI −2.4 to −0.4] respectively, p < 0.001). Though, total fluid balance at day five and during ICU stay was not significantly different between groups (−1148 mL (−2531 to 235), p=0.06 and −475mL (−2254 to 1304), p=0.60, respectively). No differences in mortality were found, but a significant reduction in worsening of acute kidney injury was seen in treatment group (27/73 vs. 39/72 (OR 0.46 [95%CI 0.23–0.92] p = 0.03), showing that a more restrictive protocol of fluid resuscitation after initial management is feasible and effective in reducing resuscitation fluids and may even reduce worsening of AKI in septic patients. However findings regarding fluid resuscitation and AKI are still controversial, because there is some evidence showing that, on one hand, under-resuscitation may increase the incidence of AKI [7], while on the other, too many fluids may be harmful for the kidney in the early phase of resuscitation [8-10].

"…a more restrictive protocol of fluid resuscitation after initial management is feasible and effective in reducing resuscitation fluids and may even reduce worsening of AKI in septic patients."

Also, cumulative fluid balance between groups is quite similar, meaning that patients in the restrictive group received more non-resuscitation fluids compared to the control group. If the impact of fluids given for haemodynamic stabilisation is compared to fluids given for other reasons, whether there is any real difference may be questionable, and there is not any robust scientific confirmation as of yet. However, balance at day five shows a clear trend towards lower values in restrictive group, despite that the study is underpowered to show significance in terms of fluid balance. This may imply that lowering cumulative fluid balance at day five might have a significant impact in terms of renal outcome, at least. Indeed, some evidence has shown that elevated fluid balance in the first hours of resuscitation are related to higher mortality [11]. 

Unfortunately, the authors did not test a protocol evaluating fluid responsiveness before fluid loading (i.e. passive leg-raising, end-expiratory occlusion, etc.) [12], so we do not know how many fluids have been prescribed to patients who actually did not need them. Indeed, some data found in the literature suggests that half of patients who present with clinical signs of tissue hypoperfusion are non-fluid responders [13]. Besides this, neither more specific haemodynamic data (i.e. cardiac output, filling pressure, end-diastolic volumes, etc.) or metabolic data are presented (i.e. veno-arterial CO2 difference, ScvO2, etc.), so we do not know the effect of fluids in terms of increasing tissue perfusion and oxygen utilisation. Yet, the absence of difference in clinical major secondary exploratory end-points suggests that even giving fluids with more restrictive criteria during the first days following shock recognition did not significantly impair tissue oxygenation and may reduce potentially harmful fluid administration.

Finally, the exploratory nature of the study hampers the ability to draw a definitive conclusion regarding the effectiveness of a more restrictive strategy of fluid management in septic patients in terms of survival. Larger RCTs with more clinical endpoints are indeed warranted, although, this trial does seem to show that more restrictive resuscitation fluid management in septic patients appears safe, without major complication.

Article review submitted by EJRC member Antonio Dell’Anna.

References

1.    Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, Sevransky JE, Sprung CL, Douglas IS, Jaeschke R, Osborn TM, Nunnally ME, Townsend SR, Reinhart K, Kleinpell RM, Angus DC, Deutschman CS, Machado FR, Rubenfeld GD, Webb S, Beale RJ, Vincent JL, Moreno R, Surviving Sepsis Campaign Guidelines Committee including The Paediatric S, (2013) Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive care medicine 39: 165-228
2.    Cecconi M, Hofer C, Teboul JL, Pettila V, Wilkman E, Molnar Z, Della Rocca G, Aldecoa C, Artigas A, Jog S, Sander M, Spies C, Lefrant JY, De Backer D, (2015) Fluid challenges in intensive care: the FENICE study : A global inception cohort study. Intensive care medicine 
3.    Acheampong A, Vincent JL, (2015) A positive fluid balance is an independent prognostic factor in patients with sepsis. Critical care 19: 251
4.    Neyra JA, Li X, Canepa-Escaro F, Adams-Huet B, Toto RD, Yee J, Hedayati SS, Acute Kidney Injury in Critical Illness Study G, (2016) Cumulative Fluid Balance and Mortality in Septic Patients With or Without Acute Kidney Injury and Chronic Kidney Disease. Critical care medicine 44: 1891-1900
5.    Myburgh JA, Mythen MG, (2013) Resuscitation fluids. The New England journal of medicine 369: 1243-1251
6.    Hjortrup PB, Haase N, Bundgaard H, Thomsen SL, Winding R, Pettila V, Aaen A, Lodahl D, Berthelsen RE, Christensen H, Madsen MB, Winkel P, Wetterslev J, Perner A, Group CT, Scandinavian Critical Care Trials G, (2016) Restricting volumes of resuscitation fluid in adults with septic shock after initial management: the CLASSIC randomised, parallel-group, multicentre feasibility trial. Intensive care medicine 42: 1695-1705
7.    Luo JC, Qiu XH, Pan C, Xie JF, Yu T, Liu L, Yang Y, Qiu HB, (2015) Increased cardiac index attenuates septic acute kidney injury: a prospective observational study. BMC anesthesiology 15: 22
8.    Wang N, Jiang L, Zhu B, Wen Y, Xi XM, Beijing Acute Kidney Injury Trial W, (2015) Fluid balance and mortality in critically ill patients with acute kidney injury: a multicenter prospective epidemiological study. Critical care 19: 371
9.    Teixeira C, Garzotto F, Piccinni P, Brienza N, Iannuzzi M, Gramaticopolo S, Forfori F, Pelaia P, Rocco M, Ronco C, Anello CB, Bove T, Carlini M, Michetti V, Cruz DN, investigators NEeCI, (2013) Fluid balance and urine volume are independent predictors of mortality in acute kidney injury. Critical care 17: R14
10.    Garzotto F, Ostermann M, Martin-Langerwerf D, Sanchez-Sanchez M, Teng J, Robert R, Marinho A, Herrera-Gutierrez ME, Mao HJ, Benavente D, Kipnis E, Lorenzin A, Marcelli D, Tetta C, Ronco C, DoRe Msg, (2016) The Dose Response Multicentre Investigation on Fluid Assessment (DoReMIFA) in critically ill patients. Critical care 20: 196
11.    Boyd JH, Forbes J, Nakada TA, Walley KR, Russell JA, (2011) Fluid resuscitation in septic shock: a positive fluid balance and elevated central venous pressure are associated with increased mortality. Critical care medicine 39: 259-265
12.    Guerin L, Monnet X, Teboul JL, (2013) Monitoring volume and fluid responsiveness: from static to dynamic indicators. Best Pract Res Clin Anaesthesiol 27: 177-185
13.    Monnet X, Julien F, Ait-Hamou N, Lequoy M, Gosset C, Jozwiak M, Persichini R, Anguel N, Richard C, Teboul JL, (2013) Lactate and venoarterial carbon dioxide difference/arterial-venous oxygen difference ratio, but not central venous oxygen saturation, predict increase in oxygen consumption in fluid responders. Critical care medicine 41: 1412-1420