Endpoints of resuscitation

Todd Fraser on 05-04-2010

Intensive care practice is built around the prompt resuscitation of patients with critical illness. It has long been recongnised that clinical parameters are poor indicators of adequacy of resuscitation, and the search for laboratory or monitoring endpoints has raged. With the apparent demise of the pulmonary artery catheter, markers of the actual cardiac output, or its adequacy in maintaining perfusion of the tissues have become the focus of research. In many ways it is incongruous that such a pillar of our practice is based on such spartan evidence. The publication of Rivers' seminal paper on early goal directed therapy in septic shock renewed interest in Central Venous Oxygen Saturations as an endpoint of sepsis resuscitation. While criticised on many fronts the Rivers paper guides current practice guidelines worldwide. A recent publication in the Journal of the American Medical Association compared serum lactate with Central venous oxygen saturations in septic shock. 300 patients were enrolled in this unblinded study. Much like the Rivers study, patients received therapy largely based on a protocol in the first 6 hours of their hospital stay. A measure of adequacy of tissue oxygenation (ScvO2 or lactate) was used to determine if blood transfusion or inotropic agents were required. The paper has been criticised in that fewer than 10% of patients required an intervention based on this measure of tissue perfusion. On the other hand, more than 90% of patients appear to have the same outcome whether complicated monitoring is used or not. So where are we now with resuscitation endpoints?

6 Comments

Got something to say?

Log in to your account

Forgotten your password?

AMIT KANSAL wrote 04-23-2010 11:04:45 pm
I start with volume resuscitation (cautious fluid use in Known heart failure/ suspected heart failure from clinical history)
Fluid challenge (250 to 500 ml crystalloid) guided by clinical/ laboratory parameters (lactate)
END POINT - few litres/ if no immediate improvement in first 1-3 Litres - CVL & guide with CVP (Cautious interpretation with High PEEP/ known RHF/ suspected Right heart failure)
CVP - 10 to 12 mmHg (may require 4-6 litres fluid), improving lactate/ urine output
MAP - 60-70 mmHg (vasopressors as needed)
That's it! not many end-point numbers at this stage!

Assess cardiac output (clinically/ ScvO2/ Pco2 (CV-Arterial) difference/ - Consider dobutamine accordingly if volume status appears optimum
Like everyone else, I believe in EGDT; but the right approach is still controversial (I don't think there is enough evidence to push Hemoglobin to 10, Above 7 is ok)
End point - limited utility of Scvo2 > 70% (a significant number will not tolerate dobutamine, my experience!), I rely on CO measurement using Echo/ PICCO

If no clinical improvement (worsening lactic acidosis/ oliguria/ increasing vasopressor requirement) -
Echo (a very accurate snap-shot) vs PICCO - I believe PAC is dead in this setting -
Echo/ PICCO will help in volume status Ax as well - give more volume,
decide vasopressin vs dobutamine accordingly.

Summary for endpoints -
CVP 10-12 mmHG
MAP - 60-7-mmHG
ScvO2 > 70% (how to achieve that?)
Improving urine output/ lactate/ CO

Todd Fraser wrote 06-22-2010 09:05:58 pm
Interesting paper published online ahead of print in the American Journal of Respiratory and Critical Care Medicine in May. It used lactate as an end point of resuscitation, in a modern context (restrictive transfusion threshold, ScvO2 monitoring etc). It STILL showed a reduction in in-hospital mortality. Am. J. Respir. Crit. Care Med., May 2010; doi:10.1164/rccm.200912-1918OC

Todd Fraser wrote 05-18-2010 02:34:19 pm
I think your comments probably reflect what most of us are doing. There is so little evidence that chasing any one endpoint will improve outcomes that we all end up doing "what feels right".

Ultimately we need a reliable measure of the microcirculation, but to my knowledge there just isn't one available.

James Doyle wrote 07-09-2010 12:11:50 pm
Although i agree with the management detailed in the above comments in terms of fluid resuscitation and vasopressor indications based on clinical assessmnet, MAP and CVP. I would agree that complex monitoring is unlikley to improve outcome with initial resuscitation (my opninon) but as far as the PAC being dead in the assessment of cardiac output, i am yet to be convinced. Anaesth Intensive Care 2010; 38: 295-301, Details some of the evidence thus far for pulse contour analysis. Only 3 out of 16 (small) studies have shown clinical reliability with a correlation % error of less than 30% (the lowest still being a high 27%). I would therfore suggest that modalities such a PICCO / Vigileo are only useful in monitoring trend rather than absolutes. I am not suggesting that the PAC is a gold standard but it may have some shelf life yet.

Todd Fraser wrote 07-10-2010 05:14:18 pm
Thanks for your comments James. Its complex, isn't it.

We have very limited measures of adequacy of tissue perfusion. From my perspective, this is the end point I'd most like to know. We can only infer that a particular cardiac output is a surrogate for this, but the numbers we choose (either a cardiac output, as measured by PAC, ultrasound, thoracic impedance or any other measure, or by some surrogate such as CvO2 sats) are arbitrary. We then have to assume that artificially increasing cardiac output or organ perfusion pressure somehow alters outcomes (ie there is scant evidence for any inotrope or vasoconstrictor).

Even if we do accept all the above, and want to give fluid to increase perfusion pressure or cardiac output, we have fairly imprecise measures of probability of an increase (Stroke Volume Variation, Systolic Pressure Variation etc), much less evidence of outcome benefit.

Todd Fraser wrote 07-11-2010 09:47:45 pm
The perils of using surrogate endpoints is highlighted by a fascinating retrospective study in JAMA in June 2010. It demonstrated that patients resuscitated from cardiorespiratory arrest who were hyperoxygenated did WORSE than those who were hypoxic.

Be careful what you wish for....

Kilgannon H, Jones A, Shapiro N, et al. Association between arterial hyperoxia following resuscitation from cardiac arrest and in-hospital mortality. JAMA 2010; 303:2165-2171.