With some of our septic patients experiencing acute renal failure needing CVVHDF, I have posed the question, that as yet I can find no answer to.
With 'normal' patients in ARF on CVVHDF, they would become hypothermic to a temperature of 35 Degrees or below. Hence needing active warming through a Bear Hugger and filter heater aids.
With septic patients on CVVHDF, I have seen temperatures of 36-37 Degrees with no heater elements to aid this.
If this is the case, should we not then be complementing them with heater elements to achieve an active temperature of 38-38.5 Degrees? As you may see in the normal immune response to increase an individuals temperature.
Patients are not actively given antipyretics due to the researched based evidence of increased mortality.
If we are placing septic patients on CVVHDF, then we are actively cooling them!
As such, are we not increasing the patients chances of mortality and morbidity?
septic patients have by no means a normal immune system. it is severely compromised. fever increases the rate at which reactions are occuring in our organism - and it has been shown to be deleterious in a variety of conditions (e.g. after CPR, TBI, MCI, stroke). In contrast also hypothermia might be deleterious in these conditions (see these recent publications: Mourvillier B, JAMA. 2013;310(20):2174-2183. or Nielsen N, N Engl J Med. 2013 Dec 5). The current evidence supports the concept of normothemia. And although studies are missing in my opinion this would be the most intuitive treatment goal for a doctor - keep your patient warm, and fight fever.
Agreed with Peter. Although I'm no expert on CVVHDF, iatrogenic fever cannot be recommended to upgrade natural immune responses. Fever appears as a puzzling surrogate marker of natural immunology-physiological responses. I wouldn't think that iatrogenic warmth would even provide the defensive mechanisms one may expect from natural fever response.
Two studies on systemic warming during sepsis seem to be relevant (Anesthesiology. 2006 Jun;104(6):1216-22 and Surg Infect (Larchmt). 2007 Jun;8(3):387-95). The former is an experimental study demonstrating an increased survival in warmed sheep, the latter is just a pilot clinical study revealing some improvement in haemodynamic parameters. The main concern is that you propose 38.5, while both these studies set the warming at 40°C.
I agree that there pros and cons in warming, but as a thumb rule, I would avoid warming in sepsis when (there is the suspect that) energy supply from blood does not match the metabolic requirements from warmed cells, which could be the case in some cases of severe sepsis.
The question is not surprising, but the answer is difficult. This is because there are very few studies in this area. We prefer in the absence of other evidence look for normothermia. At this time In our service, nor hypothermia or hyperhermia will be welcome.
Very important question but the answer is unclear. Until new studies will published I think no hypothermia or hyperthermia We should keep our patients normothermic
Yes, we should definitely maintain normothermia till we get some evidence based support in favor of hypo or hyperthermia. We need more observational studies in our intensive care set-ups to come up with some deviating conclusion. Moreover, apart from biochemical milieu, the normal temperature allows various enzymatic reactions to proceed at optimal rate which are essential for maintaining homeostasis and survival in patients with sepsis.
It seems an interesting perspective in terms of physiology and inmune response. However, from the clinical point of view i dont know if a septic shock patient with high vasopressor requirements would benefit from "heating" him up tp 38 or 38.5ºC. However, i must confess that it could somehow seem reasonable to avoid blocking nature compensation mechanisms such as hyperthermia or even restore them. My only fear would be hemodynamics.
RRT heat removal raises vascular resistance and venous tone, as well as arterial blood pressure so for a long time a moderate cooling has been used to improve tolerance but nowadays this potentially beneficial effect has been challenged for the acutely ill patient. Differences on vascular reactivity in response to cooling are not completely elucidated and the effect of CRRT-induced cooling on regional perfusion and energy metabolism in critically ill patients has not been well defined. Recently, Rokita et al (Nephrol Dial Transplant, 2004; 19:623-30) reported that when arterial pressure raised after cooling, a significant decreases in cardiac output and systemic oxygen delivery developed. So, the final effect of temperature on hemodynamics and oxygen delivery is still controversial with some reports pointing to a positive impact by moderate cooling that may even hypothetically facilitate resuscitation and others suggesting that lowering fever may indeed be harmful by worsening tissue perfusion.
We must always count on temperature losses that will induce a raise in blood pressure and vascular resistance but also a decrement in oxygen delivery and when this cooling triggers overt hypothermia, theoretically, a state of shock could be precipitated, As for a target of higher temperature, there are not to my knowledge any studies that address this important issue, so, under my point of view and until new data arrive, the safest approach would be always to aim for isothermal treatments
thermal regulation of the patients depends on the central nervous system. in severe sepsis, the functioning of sytem is defective. Increase temperature from 38 ° to 38.5 ° can not improve the immune response
Whereas spontaneous hypothermia in severe SIRS and septic shock increases hospital mortality rate ( BD < 35.6: 81.8% versus 55.3 for BD > 38.4 , OR 3.63 , F Brivet et al, CCM 1994), and if we take into consideration that “cooling” to obtain a normal temperature ( 36.8) permits an initial vasopressors dose decrease and reduces D14 mortality rate ( F Schortgen ‘s multicenter trial Am J Resp Crit Care Med 2012), we should prefer a normal temperature and avoid warmed CVVHD ( 38.5) for an hypothetical better immune response.
Whereas spontaneous hypothermia in severe SIRS and septic shock increases hospital mortality rate ( BD < 35.6: 81.8% versus 55.3 for BD > 38.4 , OR 3.63 , F Brivet et al, CCM 1994), and if we take into consideration that “cooling” to obtain a normal temperature ( 36.8) permits an initial vasopressors dose decrease and reduces D14 mortality rate ( F Schortgen ‘s multicenter trial Am J Resp Crit Care Med 2012), we should prefer a normal temperature and avoid warmed CVVHD ( 38.5) for an hypothetical better immune response.
We have more than 300 patients registered whom needed CRRT. Hypothermia during CRRT defined as
Great question!
I believe that there is confusion in the literature between patients who have a primarily neurological disease and those that have a primarily a systemic infection.
The "benefit" of normothermia is based on animal models (of TBI and stroke) and lots of observational clinical studies (again stroke and TBI) … so, manitaning normothermia may be reasonable for critically ill patients with TBI and Stroke, but the evidence is of a low grade as there are no RCTs (1,2).
For sepsis, we published a large observational study (3) that suggests for patients with sepsis, that elevated temperature may be beneficial. Others have found a similar relationship and also that antipretic use may be harmful (4). Additionally a small RCT suggest harm with fever supression (5). This conflicts with Frederique Schortgen's (excellent) RCT (6) … but vasopressor dose and 14-day mortality are not the preferred patient-centred outcomes for critical care trials (90-day is better).
There is unfortunately not enough high quality evidence to guide us at present … my preference is to keep the patient in teh hyperthermic range (between 38 and 39C) whilst on CRRT providing I am happy that the circulation is "managing".
It is interesting to me that my answer differs from most of the other answers. My bias is that we should not interfere too much with what teh body is trying to do … the response to stress or infection has evolved over millions of years and hopefully has been selected because it aids survival.
Our group is doing a bit of work in this field … trying to understand whether fever is beneficial for critically ill patients with sepsis and as a separate question (a much more controversial one) … does normothermia reduce disability and death for patients with stroke and TBI.
I hope this helps … references below.
REFERENCES
1.Saxena M, Andrews PJD, Cheng A. Modest cooling therapies (35 degrees C to 37.5 degrees C) for traumatic brain injury. Cochrane Database of Systematic Reviews. (3):CD006811. PubMed PMID: 18646169.
2. Den Hertog HM, van der Worp HB, Tseng M-C, Dippel DW. Cooling therapy for acute stroke. Cochrane Database of Systematic Reviews. 2009(1):CD001247. PubMed PMID: 19160194.
3. Young P, Saxena MK, Beasley CRW, Bellomo R, Bailey M, Pilcher D, et al. Early peak temperature and mortalityin critically ill patients with or without infection. Intensive Care Med. 2011;38(3):437-44.
4. Lee BH, Inui D, Suh GY, Kim JY, Kwon JY, Park J, et al. Association of body temperature and antipyretic treatments with mortality of critically ill patients with and without sepsis: multi-centered prospective observational study. Critical Care. 2012;16(1):R33.
5. Schulman C, Namias N, Doherty J, Manning R, Li P, Alhaddad A, et al. The Effect of Antipyretic Therapy upon Outcomes in Critically Ill Patients: A Randomized, Prospective Study. Surgical Infections. 2005;6(4):369-75.
6. Schortgen F, Clabault K, Katsahian S, Devaquet J, Mercat A, Deye N, et al. Fever control using external cooling in septic shock: a randomized controlled trial. American journal of respiratory and critical care medicine. 2012;185(10):1088-95.
First, when we debate about septic patients rewarming 38-38.5 during CVVHDF process, I do not believe that we made confusion with stroke and and TBI.
Second we discuss about sepsis/infection and not all febrile ICU patients.
Third, Bellomo’s group main topic is AKI and RRT, but its observational tri-center study included less than 10 percent of the febrile patients, and 20% of them were febrile at day 3, moreover 50 to 70 % received paracetamol. Conclusion may have a lot of biases.
Schulman et al demonstrated that in their Trauma Intensive care unit, acetaminophen and cooling worsened the outcome – study stopped after the first interim analysis;
Lee BH et al demonstrated that NSAIDs or acetaminophen should be avoided in case of febrile sepsis, mortality rate OR > 2. The recent review published by Niven et al is inconclusive , since the studies were underpowered.
Thus, “should septic patients be rewarmed during RRT” remains a matter of debate, Xosé Perez may answer the question with its 300 pts treated with CRRT, in multivariate model hypothermia was not a predictive factor; these results should be published, before a prospective multicenter trial. Unfortunately the temperature was not recorded in our database of RRT…
The question generates several subquestions :
1) Shortgen (AJRCCM 2012, 185, 1088-95) and Brivet (see above) have a point : normothermia (36.5-37°C) reduces the vasopressor requirements and improves 14d survival. Beyond that, we need a larger trial.
2) Our cardiologist insists that we should rewarm peripherally the patient (Bair Hugger) to vasodilate the microcirculation (lowered afterload), when the patient is on extra renal replacement therapy (ERRT) with a core temperature circa 35-37°C (minimal or no active rewarming on the ERRT machine).
3) Another issue is the pharmacological modification ("modulation") of the inflammation.
a) Sympatholytics improve survival in experimental models (Smith IM, Kennedy LR, RegnÇ-Karlsson MH, Johnson VL, Burmeister LF. Adrenergic mechanisms in infection. III : and beta-receptor blocking agents in treatment. Am J Clin Nutr 1977;30:1285-8.;Seeley EJ, Barry SS, Narala S, Matthay MA, Wolters PJ. Noradrenergic neurons regulate monocyte trafficking and mortality during gram-negative peritonitis in mice. J Immunol 2013;190:4717-24. ).
b) Alpha-2 agonists lower inflammation (full references in Pichot C. Dexmedetomidine and clonidine : from second- to first-line sedative agents in the critical care setting ? J Intensive Care Med 2012;27:219-37.). They also lower temperature (see Belleville ref in same Pichot 2012 ref).
c) "surprisingly", alpha-2 agonists increase the pressure response to noradrenaline in patients presenting with septic shock (case report : 1. Pichot . Increased pressor response to noradrenaline during septic shock following clonidine? Anaesth Intensive Care 2010;38:784-5.) and in animals submitted to experimental sepsis ( Geloen . Clonidine and dexmedetomidine increase the pressor response to norepinephrine in experimental sepsis: a pilot study. Crit Care Med 2013;41:e431-8)
Taken together, it is tempting (but UNPROVEN) to lower temperature to normothermia (see Shortgen) in septic/septic shock patients by using no active rewarming on ERRT and using sedation with alpha-2 agonists (dexmedetomidine, clonidine).
Yes, but be careful with experimental series which enroll 7 rats in each group and use mean and SD to analyze SBP. Furthermore, baseline SBP was normal, near 150 mmHg, even in LPS group. Being completely naïve in experimental series, I suppose that increasing bolus of NE (base or bitartrate?) from 0.05 to 20 microg/kg, is the gold standard, but the differences seem to be significant with NE dose greater than 4-6 microg/kg. As highlighted by the authors the next step should be hypotensive sepsis model to approximate human septic shock.
I would like to thank all those throughout the world whom have given their kind feedback on this subject, both through this site and via emails.
As yet we do warm our patients to a normothermic temperature, if they at all need warming due to infection. Some of our patients remaining normothermic despite having no heater elements on to keep them warm.
Your comments have thrown up some interesting points for me to consider. One such point of question is 'who decided what temperature is actually normothermic'?
Normothermia is the range of temperatures where body feels comfortable and therefore does not feel the need to respond with a physiological reflex such as vasoconstriction or vasodilatation. In daily practice, most people prefers to use a single two-digit number (e.g. 37 degree C) instead of the range of normothermia.