Golden hours in sepsis...

 Another Sepsis Day/Week has come and gone. We organized meetings and symposiums to raise awareness and issued press releases. Now that things have calmed down, I’d like to summarize what I took away from this year’s training sessions and offer a few suggestions. Around six years ago, I wrote a piece outlining my approach to sepsis treatment. They say 50% of medicine changes every ten years, but in the past six years, there hasn’t been much change in sepsis treatment. Early and effective intervention during the “golden hours” (the first 3-6 hours) reduces mortality. I’m unsure how much public awareness helps, but ensuring that nurses and residents, who have the most contact with patients, are well-informed will undoubtedly reduce mortality.

Here’s my first suggestion. In sepsis meetings, let’s keep discussing cytokines and mediators among ourselves, but for our target audience, let’s summarize the key actions in bullet points, create a poster, and hang it in all clinics.

The question I’m most frequently asked at meetings:

"Professor, we’re supposed to be monitoring lactate, but you recommend Ringer's lactate. Won’t that be confusing?"

Think of it this way: if there are no contraindications, you’ve administered the 30 ml/kg of fluids that you need to give over 24 hours within 2 hours. You've started stress doses, increased to max doses, but it's not enough, so you add adrenaline. Perfusion deficits begin in the knees, so you add dobutamine, and you might even infuse hydrocortisone. Antimicrobial therapy has already been started empirically. Sounds familiar, right? By this point, you're likely starting CVVHDF support as well.

Here’s what happens next: if you’re able to carry out all the steps mentioned before multiple organ dysfunction syndrome (MODS) develops, the patient’s mean arterial pressure (MAP) starts to improve, the need for adrenaline decreases, and naturally, lactate levels start to drop. But the goal isn’t to lower lactate; it’s to improve organ perfusion. If the patient’s MAP improves, they start producing urine, and their consciousness returns, do we still chase lactate levels? What if the patient’s condition improves while lactate rises? Should we panic? Is there anything to worry about? Is it wise to abruptly stop adrenaline because of its effect on lactate levels? Absolutely not. Please remember, lactate is not the cause of the patient's deterioration; it is the result. Lactate isn’t a toxic substance; it’s an energy source. I’ve treated many cases of septic shock with Ringer’s lactate without issues. However, I'm not a fanatic of RL.

In cases where liver disease or MODS due to septic shock is present, lactate won’t be metabolized in the liver, so we avoid using RL. Instead, we can administer Isolyte-S, which uses gluconate and acetate as buffers instead of lactate. I’m still against the use of saline. For patients with severe metabolic or lactic acidosis, pouring in saline will only result in hyperchloremic metabolic acidosis. I see this frequently in blood gases, and it doesn't seem like a smart approach. We've discussed those posters, but we sometimes miss the golden hours, and not all patients are this lucky, so lactate levels will continue to rise.

In my opinion, lactate monitoring doesn’t provide much benefit. Now, I want to highlight an even more critical point, probably the most important takeaway from today’s article. A study showed that, when sepsis is suspected, the infection consult is called, the resident arrives, thoroughly examines the patient, consults with their attending, and recommends the most appropriate antimicrobials. We place the orders, the medication is brought, the nurse administers the loading dose, and the time lost averages 220 minutes. The golden hours are gone!

There’s a great initiative from Hacettepe University’s Infectious Diseases Department, led by a distinguished professor, which they call a “Clinical Decision Support System.” If I’ve taken the name down incorrectly, I apologize. Here’s how it works: based on the most frequently isolated opportunistic agents and their antimicrobial resistance patterns in ICU units, a treatment algorithm is created. When a patient develops sepsis, the ICU resident follows this form and starts the recommended treatments, reducing the time to antimicrobial administration to 38 minutes. It’s a fantastic initiative and would make an excellent second poster to hang alongside the first.

In my practice, I do something similar: while administering first-line treatments, I administer a single dose of antibiotics for gram-positive, gram-negative, and, if risk factors are present, antifungal therapy from the available medications. The average time to start antimicrobial treatment is 8-10 minutes. Of course, if your hospital has the ability to perform rapid antibiogram testing, the success of empirical targeted therapy will be higher.

Another frequently asked question:

Which scoring systems should we use in diagnosing sepsis? To be honest, I don't believe any intensivist, unless conducting a study, uses scoring systems like SOFA, NEWS2, MEWS, MEDS, etc., to diagnose sepsis in the ICU. We have the patient's vitals, consciousness status, and urine output right in front of us at all times. These scores may hold some value when used in hospital wards, but in the ICU, they’re often unnecessary.

That being said, if you establish a sepsis team to monitor these scores, you can certainly detect many patients early. However, it’s crucial to remember that these early warning scores are not specific to sepsis. Some patients who you think have sepsis based on these scores and start empirical treatments may end up unnecessarily receiving broad-spectrum antibiotics.

Before closing the topic on antibiotics, I’d like to mention another critical aspect: the timing of stopping antibiotics. In the past, we used to administer colistin for 14-21, even up to 28 days. Now, treatment duration of 7-10 days, or even 5 days in some cases, is sufficient. De-escalation should also be carefully considered.

I’d also like to discuss vasopressin therapy, which has been recommended in sepsis bundles for a long time but has only recently become available in our country. When you've escalated norepinephrine to high doses and want to reduce the risk of complications like dysrhythmias, digital necrosis, and mesenteric ischemia, you can consider adding vasopressin either simultaneously or as a second-line agent. If this combination works, and the patient begins to improve, it’s suggested that you taper and discontinue one of the agents. Vasopressin is noted to be more effective if administered before the patient reaches the septic shock phase (i.e., before lactate levels exceed 1.4 mmol/L). It's been reported that administering it within the first 3 hours of sepsis can reduce the need for renal replacement therapy.

There’s also the thought that if a patient with sepsis has a high risk of arrhythmias with norepinephrine, which may destabilize their hemodynamics, vasopressin could be considered as a first-line agent. I don’t yet have personal experience with this, but we will see.

As we know, deaths in sepsis are not caused by the infection itself but by the exaggerated response we mount against the infection.

When sepsis occurs, we know that both inflammatory and anti-inflammatory cytokines increase, and sometimes immune paralysis can develop. In some sepsis cases, the presence of both has been termed Persistent Inflammation, Immunosuppression, and Catabolism Syndrome (PIICS). It is now increasingly believed that knowing which pathological metabolism is predominant—i.e., identifying sepsis subgroups—can improve treatment success. For example, in COVID-19 sepsis, where cytokine storms are prominent, administering Anakinra may be more beneficial. Additionally, if adrenomedullin levels exceed 70 picograms, mortality risk increases, and humanized monoclonal adrenomedullin antibody (Adrecizumab) has been reported to help reduce endothelial dysfunction and permeability.

If immune paralysis is suspected, immune-modulating treatment with Nivolumab may benefit bacterial sepsis. In cases where endotoxins or exotoxins are predominant, blood purification methods may be more effective. Toll-like receptors (TLRs), located on cell membranes, are responsible for recognizing pathogens. Agents that inhibit TLR signaling could be crucial in preventing cytokine storms in PIICS. Personalized treatments, much like those in cancer therapies, seem to be the future in sepsis care.

Lastly, it’s worth mentioning that non-adrenergic vasoactive agents like Angiotensin II (Angi-II) have been reported to increase vascular tone, improve mean arterial pressure, and reduce vasopressor requirements in septic patients, much like vasopressin.

These are the key points I took away from this year’s sepsis-related meetings.