Episode 66- Surviving Sepsis: Updated recommendations and clinical pearls

Hello and welcome to Episode 66 of ER-Rx. In this Fresh Fruit series, I’ll give you what I think are some of the most important updates, changes, and reaffirmed recommendations of the new sepsis guidelines that came out in October of 2021. For more detail, I encourage all of you to read the guidelines yourselves- all 59 glorious pages of it. Since this is a pharmacy podcast, I’ll stick mostly to recommendations discussing medications. With that, let’s jump right into it.

New from the 2016 version is the recommendation against using qSOFA compared with other tools such as SIRS as a single screening agent for sepsis or septic shock. This is because since 2016, we’ve learned that qSOFA is more specific, but less sensitive than having two of four SIRS criteria for early identification of sepsis. Remember that neither SIRS nor qSOFA are ideal screening tools for sepsis when used alone (strong recommendation, moderate-quality evidence).

Also new is the downgrade of the recommendation for giving at least 30 mL/kg of IV crystalloid fluid within the first three hours of sepsis identification (was strong, low quality evidence). They now only suggest those 30 mL/kg of fluids be given (is now weak, low quality of evidence). I know we’re splitting hairs at this point, but I think that this is a great small change in the right direction because many providers get really frustrated with being forced to give this much fluid to all of their patients. Overall it seems better to treat the individual patient, monitor their specific resuscitation parameters, and only give as much fluids as necessary—especially in the era of COVID.

They continue to strongly recommend antibiotics within one hour of septic shock recognition, and they also recommend antibiotics within one hour in patients with “confirmed” or “very likely” sepsis – but remember the recommendation and quality of evidence is highest in the setting of septic shock specifically (strong, low quality of evidence). In patients with “possible” sepsis without shock, the guidelines now give you three hours to start antibiotics to give you time to exclude all noninfectious causes of illness (weak, very low quality of evidence). They suggest against using procalcitonin to decide whether or not to start antibiotics (weak, very low quality of evidence). This echoes the recent community-acquired pneumonia guidelines, which recommended starting antibiotics regardless of procal level if clinical evaluation suggests pneumonia. This is because procal is best used as a surrogate of when to stop antibiotics, not when to start them—and we’ll have more on this later. Regarding which antibiotics to use, they recommend adding coverage against MRSA only in patients with a high risk of MRSA (Best practice statement). In terms of multidrug resistant organism (MDRO) coverage, they suggest double-covering only in patients with sepsis or septic shock at high risk for a MDRO, and suggest only using one gram-negative agent in low-risk patients. Once the organism is identified and sensitivities are back, they again suggest against using dual gram-negative coverage (weak, very low quality of evidence). So overall, nothing really earth-shattering here, but it does get complicated and this recommendation will be highly specific to your patient, institution, and local prevalence of MRSA or MDRO. At my site, we always add vancomycin because our MRSA rate is >10-20% and we almost never double-cover gram negatives unless we have confirmed cultures showing a MDR strain in the past, or a patient with severe septic shock who has numerous risk factors for resistant organisms.

They suggest using empiric antifungal therapy in patients with high risk of fungal infection (weak, low quality of evidence). As you know, most fungal infections develop in patients already in the ICU, and studies show us that there was no improvement in mortality rates in patients given antifungals up-front. Typically, antifungals should be started in all patients with febrile neutropenia who fail to break fevers after 4-7 days of broad-spectrum antibiotics. But again, this is very specific to each individual patient, and we take into consideration whether or not they’re neutropenic, on TPN, or have had a history of invasive fungal infections.

A cool new recommendation is the suggestion of using prolonged infusions of beta-lactam antibiotics after giving the initial bolus dose (weak, moderate-quality of evidence). This is to take advantage of their time-dependent action. There isn’t much great data for this, but this intervention is pretty simple and has a low likelihood of causing harm. We started this practice at my site with Zosyn, where we give the initial load of 4.5 grams over 30 minutes, and then give a maintenance dose of 3.375 g Q8H but over 4 hours, assuming the patient has good kidney function. This not only optimizes pharmacokinetic and pharmacodynamic parameters, it also leads to an overall reduced drug use and drug cost. They even specifically shout out the use of clinical pharmacists here to help with optimizing dosing strategies of all antibiotics based on their specific PK/PD parameters (best practice statement).

As the guidelines suggest, please remember to look at antibiotics daily and de-escalate as soon as possible to help improve outcomes, reduce length of stay, reduce the risk of antibiotic resistance, and save money. Also, stick to the shortest course possible for the given infection. If the optimal duration of therapy is not clear, they suggest using procal in addition to clinical evaluation to help determine when to discontinue antibiotics—assuming we have source control and the patient is improving (weak, very low quality of evidence). This is a very low-risk intervention and should be used if procal is available at your site- but remember that the algorithms using procal for de-escalation differ in the frequency of monitoring and thresholds of percentage change in the procal for antibiotic discontinuation.

They reaffirm the recommendation to use crystalloids as the first-line fluid (strong, moderate-quality evidence), with a new suggestion of using balanced fluids (like lactated ringers) over normal saline (NS) (weak, low quality of evidence). This is a change from 2016, which recommended either balanced fluids or NS. And this is because we have newer data that associates NS with hyperchloremic metabolic acidosis, renal vasoconstriction, and AKI. This has been fueled by studies like the SMART trial, which showed improved mortality in patients receiving balanced fluids over NS. We haven’t really implemented more lactated ringers use at my site, since most of the providers aren’t really convinced by the data, and it opens up a whole new issue of drug compatibility.

Norepinephrine continues to be the first-line agent for septic shock (strong, moderate quality evidence), with the suggestion of adding vasopressin as the second agent once Norepi is running at > 0.25 mcg/kg/min (weak, moderate quality of evidence). Epinephrine can be used as the third line agent (weak, low quality evidence). At my site, we carry Angiotensin II and we try to use it as a third- or fourth-line agent, which is discussed in these guidelines as well. As a new recommendation, they recommend giving vasopressors peripherally rather than delaying their start for a central line (weak, very low-quality evidence)- especially if the vasopressor is administered for a short period of time (< 6 hours) and is given in a peripheral line proximal to the antecubital fossa.

They suggest using IV corticosteroids at a dose of 200 mg/d (typically 50 mg Q6H) if patients are on norepi or epi at a dose > 0.25 mcg/kg/min for at least 4 hours (weak, moderate-quality evidence). Now, this is new from the 2016 version which did not support the use of steroids unless fluids and vasopressors failed to stabilize the patient. They don’t recommend specific durations, but most trials gave them for 5-7 days or until ICU discharge or death.

In the setting of ARDS, they suggest using intermittent boluses over continuous infusions of paralytics (weak, moderate quality evidence). This is a change from the 2016 guidelines which recommended a continuous infusion of paralytics for 48 hours. Since that time, the ROSE trial was published and it showed that continuous infusions didn’t improve mortality when compared to light sedation with as needed paralytic boluses. Remember to make sure your patients are at a RASS of -4 or -5 with continuous infusions or boluses of opioids and benzos prior to giving any paralytic. Because being awake and paralyzed would be terrifying.

And lastly, my favorite new recommendation, they do away with Vitamin C therapy—suggesting against its use (weak recommendation, low quality of evidence). This is because a few trials and meta-analyses found no difference in outcomes with using the infamous vitamin C cocktail. Check out Episode 44 for my take on the Vitamin C hype.

As always, thank you so much for your time. Don’t forget to click on the show-notes, where you can find links to where you can leave me a review, subscribe to the pod, check out my references, and even support the show on BuyMeACoffee.com. Also please read the disclaimer found in the show notes and on the website, and remember that everything you hear or read on the podcast is for educational purposes and should not be used as medical advice or without consulting your own references.