In discussing this issue, I always start from the premise that you should never do anything that you truly believe will cause actual harm to your patient, regardless of the CMS core measure. That's just common sense. I typically imagine this type of situation applying to a true reason not to resuscitate with 30 mL/kg - critical aortic stenosis in a 90 year old frail female with chronic renal failure and peripheral vascular disease. In that case, I'm willing to fail the core measure and not worry too much.
That said, I would ask you to consider that we have EXACTLY ZERO evidence that patients who receive the full 30ml/kg bolus suffer adverse consequences from any large trial in septic shock patients. Moreover, mechanical ventilation is not necessarily an adverse outcome. I have facetiously said before, "I'd rather be intubated than dead." Sometimes patients will need to be ventilated to restore adequate perfusion. If we do not intubate in those cases we risk further organ failure due to hypoperfusion. Fear of the ventilator is largely overblown, especially when VAP rates are uniformly low across the country. Typically if I gave someone too much fluid in the morning and they were intubated, I'd extubate in the afternoon with a little furosemide. Interesting data as far back as the Rivers trial suggest that patients who received more fluids in the EGDT arm of his trial sustained a LOWER rate of intubation than those who did not receive fluids EVEN WITH CONCOMMITANT CHF as a diagnosis. In Rivers, 70.6% of patients were intubated by 72 hours, whereas those who received EGDT and significantly more fluid 55.6% of patients were intubated. These data were statistically significant and it should be noted that 30.2% of controls had CHF and 36.7% in EGDT had CHF. So despite having more CHF patients in the intervention group and receiving more fluids fewer patients were intubated. Why? The reason is that these patients are not presenting to you with their co-morbidity of renal failure or CHF as their primary problem - they are presenting with shock. Shock needs resuscitation. Consider that the American College of Surgeons Trauma guidelines suggest that Class III trauma patients have lost 1500-2000 mL of blood. This represents a 30-40% blood volume loss and results in symptoms we see in established septic shock: pulse greater than 120, detectably low blood pressure and pulse pressure, tachypnea 30-40 per minute and diminished urine output 5-15 mL/hour. The fluid replacement rule in this circumstance in trauma is to administer both crystalloid and fluid in a 3:1 ratio compared to the loss. A Class II patient has lost 750-1500 mL or a 15-30% volume loss, blood pressure may remain normal, pulse pressure is decreased, respiratory rate 20-30 and the fluid replacement for those patients is crystalloid (no blood) in a 3:1 ratio. Clearly the 3:1 ratio in trauma exceeds needs in septic shock because ongoing losses are less rapid, however even with these degrees of quantifiable initial volume loss (up to 2 liters in Class III patients), a 30 mL/kg bolus is essentially just replacing the loss that produced the observable physiology in an 70 kg patient. Thus, in shock states, these rates of fluid resuscitation are not unusual and correspond to signs and symptoms we see clinically with our "attentive evaluation" of septic shock patients. Finally, to make the final point, in the three most recent trials, again patients with CHF and chronic renal failure were not excluded from enrollment in the trials. If you look at the volume of fluid these patients received it quickly reaches the 30mL/kg threshold and goes beyond. In ProCESS the EGDT group received 2805 +/- 1957 mL, the usual care group received 2279 +/- 1881 mL. In ARISE, EGDT received 1964 +/1 1415 mL, usual care 1713 +/1 1401 mL. In ProMISE EGDT received 2000 mL average, usual care 1784 mL average. In general, I usually present this information to others by saying, "first do not harm, but think about whether the harm you are not doing is really there." Second, decide if mechanical ventilation (if the evidence were wrong that rates are lower with fluid resuscitation) is really that harmful. Third, remember you are treating shock, comorbidities are secondary issues at best if physiological parameters are as abnormal as we see in Class II and Class III trauma patients. Lastly, in all the large trials patients tolerated these fluid boluses. From: Sepsisgroups [mailto:[email protected]] On Behalf Of Reynolds, Stuart Sent: Wednesday, September 16, 2015 4:20 PM To: [email protected] Subject: [Sepsis Groups] Sepsis Core Measure Follow Up I agree with Dr Allen's comments below. The mandated 30ml/kg IVF bolus may be harmful; if 20ml/kg or 10ml/kg is sufficient to improve perfusion; particularly in patients with acute lung injury, or those with underlying heart or kidney disease. This reflex dosing takes away from the "attentive evaluation". It is this "attentive evaluation" which requires incentive. Many will need to practice down to these bundles and put their patient at risk to meet the CMS Core requirements. 1. 3 hours to volume resuscitate 2. 6 hours to assess volume status (which is only required if on vasopressors) 3. 6 hours to add vasopressors in the setting of fluid unresponsiveness 4. Normal Saline (pH 5.5, Cl 154) versus balanced salt solutions like plasmalyte (pH 7.4, Cl 98) Here's the irony: * If one is responsive to the patient's needs and prescribes 20 ml/kg in the setting of hypotension, because reassessment reveals improved perfusion and reversal of hypotension, this attentiveness results in penalty. If one however in unobservant, automatic and simply prescribes 30ml/kg IVF over 3 hours, starts norepinephrine within 6 hours (patient may have required at hour 1 or 2), and evaluates the patient within 6 hours - they've checked all the boxes and have met criteria with respect to volume, vasopressor initiation and clinical assessment. * Don't forget a patient with SBP <90 mmHg whose response to a fluid bolus of 20-30ml/kg over 30min resulted in SBP >90 would not have met inclusion into the EGDT trial. Arise and Process also required fluid unresponsiveness before being enrolled. * It's the unresponsiveness to fluids that determined enrollment; as determined by an "Attentive Evaluation" after a fluid bolus. Sincerely, Stuart F Reynolds, MD FRCPE FCCP Director Critical Care Services Clinical Professor Critical Care Medicine MUSC AHEC [cid:[email protected]] <http://www.spartanburgregional.com/> 101 East Wood Street | Spartanburg, SC 29303 o: 864-560-6531 e: [email protected]<mailto:[email protected]> | w: SpartanburgRegional.com<http://www.spartanburgregional.com/> Sean, I attended your webinar on Sepsis Core measures last week and was left with a number of concerning questions: 1. If we are using the logic that "there is no evidence to show it doesn't hurt" to justify follow-up physical exam measures for evaluation of response to resuscitation, then why does the same logic not apply to the use of Normsol and other chloride-balanced crystalloids? I would argue that there is a growing body of evidence that normal saline may indeed "hurt" (JAMA. 2012;308(15):1566-1572.; Br J Surg 102 (1):24-36. Crit Care Med 2014; 42:1585-1591. 2. In defending the use of many of these unproven metrics of volume responsiveness and distal perfusion, you described many of these measures as a "proxy" measure of "attentive evaluation" and intensive care. I full agree, and practice this way. I believe these measures help represent a collective epi-phenomenon of intensive and regimented care. Using the same reasoning, why then is there no provision in any of this for providers to document their own rationale for diverging from some of these restrictive mandates when judged to be clinically justified. Is this not also a worthy "proxy" of intensive and attentive care? 3. When does the clock really start ticking? Our hospitals still don't have a solid and reliable answer to this question. Is it when the physician documents their suspicion of sepsis, 3 hours after a fever and hypotension? When blood cultures are first ordered one hour after the fever? Or when an MD orders Tylenol, a CBC, lactate, and blood cultures on someone he/she suspects may be either bleeding, in pain, or possibly infected post-Op? When do these types of patients really "declare" themselves septic. The efforts to try to "capture" every element of Goal-directed care in an "all-or-none" pass/fail algorithm dooms itself from the beginning. Why didn't CMS just start off with the 3 hour bundle, monitor how others do with the 6 hour bundle, and try to figure out where (and why) their algorithm is succeeding, or failing, to capture (and enforce) best practice? I've augured to my group that there is absolutely no excuse for not getting blood cultures, a lactate, and fluids on board within one hour of a high suspicion of sepsis. This is a low bar we should all be meeting, but probably aren't. Why not simply start there, and work our way forward? Gilman B. Allen, MD Associate Professor Department of Medicine Director of Adult Critical Care Services University of Vermont / Fletcher Allen Healthcare HSRF 220, 149 Beaumont Ave Burlington, VT 05405-0075 (802)656-9004 Fax: (802) 656-8926 [email protected]<mailto:[email protected]<mailto:[email protected]%3cmailto:[email protected]>
_______________________________________________ Sepsisgroups mailing list [email protected] http://lists.sepsisgroups.org/listinfo.cgi/sepsisgroups-sepsisgroups.org
