Skip to main content
Patient Care

Sepsis Alert


      Before her tragic death, nothing about Mariana Bridi da Costa's remarkable life had been surprising.

   Her rise to fame in the world modeling and pageant communities was really no surprise--she was strikingly beautiful. In 2006, she had been voted one of the four most beautiful faces in the world. And nobody was surprised when she placed fourth at the 2008 Miss World pageant in Brazil.

   The surprise didn't begin until January 2009, when Mariana was admitted to the hospital for complications from a recent bladder infection. The surprises continued later that month when it was announced that doctors would remove her hands and feet in a desperate attempt to save her life. Surprise turned to shock on January 24, when it was announced that Mariana da Costa, the rising Brazilian star, was dead at the age of 20 from multiple organ failure.1 The underlying culprit was septic shock.2 Thousands attended the funeral, each asking some form of the same question: "How could such a thing happen?"


   In truth, Mariana's case is neither rare nor isolated. The Mayo Clinic reports treating about 100 severe sepsis cases each year in otherwise young, healthy individuals. On average, one in four will die.3 Those numbers increase rapidly when we start adding at-risk patient populations to the equation. The very young, very old, those with recent surgeries, invasive procedures or illnesses and those with depressed immune systems have far greater infection and mortality rates.

   Nationwide, American hospitals are expected to treat approximately one million cases of severe sepsis in 2010. Of those patients, 30% will die within one month of diagnosis. An additional 20% will succumb over the next five months.4–6 By the time you finish reading this article, five more patients will have lost their lives to sepsis.

   That mortality rate is comparable to those of lung and breast cancer combined. It's also on par with the 28-day mortality rates of patients suffering acute myocardial infarctions in 1960.7 While the world medical establishment has attacked the AMI patient population with research, education and technology, sepsis remains a silent killer.

   It's also getting worse. Sepsis rates are predicted to rise at a rate of 1.5% a year. That will mean an additional million cases in the U.S. alone by 2020.8 But the silent killer at the bedside also presents a unique and exciting challenge to caregivers.

   The story of sepsis isn't all bad news. It's also a story of doctors looking for unique assessment tools to identify septic patients early. It's a story of aggressive treatment regimens that are showing promising results. And now it's a story that has come to the prehospital arena.

   When emergency room physician Wayne Guerra, MD, was tasked with standardizing in-hospital treatment of sepsis patients across the Adventist Division of the Centura Health hospital group in Colorado, he teamed up with John Riccio, MD, the local EMS medical director, to create a unique addition to area prehospital protocols: a one-of-a-kind prehospital sepsis alert program.

   The goal of the program is simple enough. "First we need to decide if we can identify these patients in the field," Riccio explains. "If we can't identify them, it's going to be hard to say we're making a difference. Second, does [identifying the sepsis patient] make a difference in their length of stay, complications, mortality, etc.?" It's a compelling question. But before we can address it, we need to answer an even more fundamental inquiry.


   Part of the challenge of addressing sepsis research and education is the lack of a clear answer to the question "What is sepsis?" Thankfully, the medical establishment is getting closer to a consensus. This lack of clear identification and treatment guidelines is part of the ongoing fight to reduce sepsis mortality.

   Another challenge is that our understanding of the sepsis process is still developing. Traditionally, we've viewed sepsis as the body's response to infection. Therefore, patients who die from sepsis have seemingly lost the battle with the microorganisms invading their bodies. But the process is far more complex.

   The issue with which medical research wrestles is this: Often, septic patients die even after the underlying infection has been brought under control. The microorganisms that initiated the sepsis cascade become bystanders in the process. Could the patient's own immune response create a greater risk than the underlying infection?

   At its core, sepsis is a problem with circulation. Ironically, this circulatory dysfunction is caused by the body's own overactive immune system. When the immune system senses a threat--bacterial, fungal, viral or parasitic--it mounts a response. Most caregivers are familiar with the white blood cell production and localized inflammatory responses that are standard immune system responses.

   In severe sepsis, the body's balance between inflammatory and anti-inflammatory chemical responses gets out of whack. A massive release of pro-inflammatory mediators creates an uncontrolled inflammatory response. This systemic inflammatory response is known as SIRS (systemic inflammatory response syndrome).

   Once SIRS is underway, impaired fibrinolysis interrupts the body's normal clotting cascade, and the blood begins to clot abnormally. Critical microvasculature, essential to supplying oxygen to the body's organs, begins to clog and fail. Systemic inflammation, vasodilatation and capillary leakage contribute to hypotension and the rapid slide into end-organ hypoxia and failure.

   When we understand the sepsis spiral, we see we are fighting against the body itself. The damage is systemic and remote from the initial source of infection.

   When, exactly, a patient crosses the line from uncomplicated sepsis to severe sepsis and then to septic shock is a subtle and somewhat subjective matter. The exact boundary is far from clear. The typical disease progresses something like this:


   Experienced by millions of people each year, uncomplicated sepsis is caused by a wide variety of viral and bacterial infections, such as the common flu, gastroenteritis, bacteremia, pneumonia, dental abscesses and, as in the case of Mariana, bladder infections. Some victims will manage themselves with over-the-counter anti-inflammatory medications for comfort. Many will receive oral antibiotics from their physicians. After trips to pharmacies, they'll go home to rest, hydrate and take a few days off. Most will recover without further intervention.


   A small percentage of patients experiencing uncomplicated sepsis will progress to severe sepsis. Severe sepsis is identified when one or more end organs begin to show signs of failure.

   Due to inhibited circulatory function, organs like the heart, kidneys, lungs and liver become oxygen-deprived and begin to fail. Core body temperature may rise, then fall. The heart rate will increase, and the blood pressure will fail. Respirations increase due to systemic hypoxia and inhibited lung function. This is the magic moment when recognition of the patient's condition can make a huge difference in their outcome. All of these patients will require hospitalization to stop the deadly inflammatory cascade within. Using typical treatment guidelines, 30%-35% will eventually die.9 These patients are the target of the program devised by Drs. Guerra and Riccio.

   Riccio is emphatic about the critical difference caregivers can make if we can recognize the sepsis progression at this moment and intervene. "If we can keep the patient from becoming hypotensive," he says, "we can save a lot of lives. Any episode of low blood pressure significantly increases your morbidity and mortality. If [prehospital caregivers] can recognize sepsis earlier and prevent it by resuscitating these patients early on, I think we're going to have a dramatic effect."


   If fluid resuscitation comes too late, even aggressive rehydration can fail. Sustained hypotension after fluid resuscitation is the hallmark of septic shock.10 These are patients who require more than fluids and traditional antibiotic schedules to overcome the systemic damage to their vasculature and organs.

   These patients will need advanced pharmacological intervention that may include vasopressors, corticosteroids, anticoagulants and immune stimulants to maintain organ perfusion, reverse the inflammation cycle and avoid multiple organ failure. Even with rapid admission and treatment in the ICU, more than half these patients will continue into multiple organ failure and die.9

   Sepsis can evolve rapidly. The first key to reducing mortality is identifying severe sepsis candidates early in the disease progression and initiating early goal-directed therapy. That's where prehospital caregivers come into play.


   When Christiana Care Health Services of Wilmington, DE, initiated a systemwide sepsis alert program focused on educating all caregivers in early recognition and rapid treatment of sepsis patients, leaders learned two important lessons. The first was that while medical providers throughout the facility received sepsis-recognition training, 85% of new sepsis patients were identified in the emergency room. The bulk of early sepsis recognition rested squarely on the shoulders of the facilities' emergency services.13

   The second was that early goal-directed therapy could be remarkably effective. Since the program's implementation, Christiana has cut its severe sepsis mortality rate in half--from 61.7% to 30.2%. The program also achieved a 34% decrease in average length of stay and a 188% increase in the number of patients discharged directly home. Those results won the Joint Commission's prestigious Ernest Amory Codman award.13

   Guerra was inspired by those results to try to take the concept even further. "We asked ourselves," he says, "if giving fluids early is so important in the treatment process, is there an opportunity to push that out even further into the field? Using the early goal-directed therapy concept used by [Dr. Emanuel] Rivers,14 could we use those criteria to identify patients with sepsis and institute the fluid bolus earlier? Would that make any change?"

   Today, under the guidance of Riccio and Guerra, prehospital services like Colorado's South Metro Fire Rescue are implementing a field sepsis alert program that puts EMTs and paramedics at the front lines of early sepsis recognition and care.

   The sepsis alert field criteria are similar to those used inside the receiving facilities. When patients meet the alert criteria, field providers activate the sepsis team prior to arrival, just like a stroke or cardiac alert program. But the reception at the hospital is markedly different.

   "The patient goes to a room large enough to put in a central line," says Guerra. "They get a monitored room. Ultrasound will be there, with a physician, a nurse and a lab technician to draw cultures. A respiratory therapist will be there to help with the airway or run the stat venous lactate. And a nursing supervisor is there to help get us an ICU bed."

   This team is ready to continue the early goal-directed therapy initiated by the field providers, who've already begun the most critical early steps, identification and aggressive fluid resuscitation. With this design, field providers are able to mobilize significant resources to benefit their patient. But none of it means anything if we can't identify the severe sepsis patient early in their presentation.


   Sepsis is not a simple process to identify. Early in the cycle, patients can look much like your run-of-the-mill sick person. Core temperatures can be high or low. The patient may be aware they're fighting an infection, but not always.

   Whether or not your EMS system currently has a sepsis alert program, recognizing patients transitioning into severe sepsis is an important clinical skill. Here are some guidelines.

  • Recognize high-risk patients: Sepsis is more likely to occur in several high-risk populations. Have a higher index of suspicion when evaluating the elderly or the very young, patients who are bed-confined or immobile, and patients who have had recent surgeries or invasive medical procedures.9 Be highly suspicious of patients receiving immunosuppressive treatments like chemotherapy or post-organ transplant medications. Recognize that some disease processes leave the patient naturally immunocompromised. This is the case with diabetes, liver cirrhosis, autoimmune disease and HIV/AIDS populations.9
  • Look for a source of infection: In many cases the source of infection is identifiable. Ask about recent illnesses, surgeries, invasive procedures or trauma. Has the patient had a respiratory infection or been feeling ill? Ask about symptoms of gastrointestinal or bladder infections, abdominal discomfort and unusual body or joint pain. Also ask about current or past prescriptions for antibiotics, steroids or immunosuppressants.
  • Pay attention to the patient's body temperature: We traditionally consider fevers when we envision the body's response to infection, but septic patients may also be mildly hypothermic. Ask the patient about recent fever or chills, and if you have a means to take an accurate temperature, take one. Be suspicious of core temperatures above 38°C (100.4°F) or below 36°C (96°F).10
  • Look for changes in vital signs: When an infection is confirmed or highly suspected, changes in vital signs become significant. Look for a pulse greater than 90 and a respiratory rate above 20 breaths per minute in combination with a blood pressure below 90 systolic or a mean arterial pressure below 65. Consider these markers tipping points for severe sepsis.10
  • Assess other subtle physical signs: Patients transitioning into severe sepsis will rapidly become fluid-depleted. Look for signs of dehydration like poor skin turgor, dry mucosa and decreased urine output.
  • Check lactate levels when possible: EMS crews participating in Denver's sepsis alert program are using a new assessment device, the lactate monitor. This tool was originally developed for athletes, but it's proving a useful patient assessment tool (see Why Monitor Lactate?, page 32).

   The monitors look and operate much like standard blood glucose monitors and display the patient's current blood lactate levels. Lactate is a sign of metabolic distress and may be an early indicator of severe sepsis.11 For crews with this capability, lactate levels above 4 mmol/L are an additional indicator of hypoperfusion.


   You don't need a sepsis alert protocol to begin treating sepsis patients in your system. Most of the assessment and treatment guidelines are already available. When you think you've identified a severe sepsis patient, keep these key points in mind.

  • Support the airway: End-organ hypoxia is the enemy of the sepsis patient, so we need to ensure we're oxygenating appropriately. Place patients with adequate respiratory drive on high-flow oxygen. For mild respiratory distress, consider CPAP at lower pressures and, as always, move on to advanced airways and intubation in cases of altered mentation and failure to breathe. Monitor the patient's SpO2 and EtCO2 where available. Keep in mind that sepsis patients are likely candidates for acute lung injury and acute respiratory distress syndrome. When mechanical ventilation is necessary, take care not to overventilate.10

  • Aggressive fluid resuscitation and hemodynamic stabilization: Septic patients are profoundly dehydrated. Beyond sepsis recognition, aggressive fluid resuscitation may be the single most important intervention we can begin in the prehospital environment. Establish bilateral IV lines. While local protocols will dictate the extent of fluid resuscitation permissible, a bolus of 500–1,000 ccs in the first 30 minutes of care is a good starting point.10 Call your base physician and consult on appropriate endpoints when your protocols do not specifically address sepsis. Patient age, history, weight, vital signs, medical history and individual physician preferences will all play roles in establishing an initial fluid resuscitation plan. Even patients who are traditionally considered fluid-restricted, such as those with congestive heart failure and renal failure, may still be indicated for well-monitored fluid challenges. Pay close attention to lung sounds and blood pressure during aggressive fluid resuscitation. Prehospital use of vasopressors for sepsis is rare and should be considered only after fluid resuscitation has proven inadequate. Some 40%-60% of severe sepsis patients will eventually receive vasopressors as part of their in-hospital courses, usually dopamine or dobutamine.12 Seek consultation with your base physician if you're considering dopamine after unsuccessful fluid administration.

  • Prevent hypothermia: Patients transitioning through severe sepsis become highly susceptible to hypothermia. This is especially true in the elderly and young.17,18 Abnormally low body temperatures increase mortality in these patients. While we don't want to bundle febrile patients, resist actively cooling septic patients and protect them from excessive heat loss from the environment and administration of cold fluids.
  • Trend the vital signs: Trend heart rate, blood pressure and respiratory rate in septic patients. Calculate the mean arterial pressure if you are trained to do so. Not only are these values excellent warning indicators of early end-organ failure, they are necessary signposts on the journey through sickness and into health. Baseline and ongoing lactate levels are proving helpful markers.11

   Don't be surprised if the attending physician requests a baseline blood sugar even if the patient has no diabetic history. If the initial serum blood sugar is high, the receiving facility may consider controlling it with insulin. Septic patients demonstrate increased survivability when their serum blood sugar is maintained below 150 mg/dL.10,15,18


   Death from infection is a puzzle that predates medicine. It has shaped the path of human history unlike any other medical condition and continues to do so today. While modern medicine may never completely eliminate tragedies like the death of Mariana da Costa, it is clear we can do better, and prehospital providers can play an important role.

   Riccio knows from experience the skepticism new alert programs can create. "It's not a gimmick," he says. "If we can get the engine revved up based on your call, it makes a tremendous difference in time. The difference can be multiple hours." And he's clear that those hours mean a lot in the world of sepsis treatment: "We've proven that early treatment can dramatically change the outcome. If there's a major killer out there and we can have a major positive effect on its outcome, that's something I get excited about. I'd like to see the same excitement for sepsis, because there are a lot of lives that can be saved."

   Sepsis mortality remains unacceptably high, but awareness projects, aggressive treatment guidelines and innovative systems like Christiana and Centura's Adventist Group are proving we can dramatically reduce the tragic stories and send more patients home well. The path toward reducing sepsis mortality begins with early recognition. That task will fall firmly on the shoulders of emergency medicine. This will, no doubt, be a journey of small steps. It begins with the simple recognition that we can do better.


The protocol being used for sepsis recognition in the Denver area is a combination of the original systemic inflammatory response syndrome (SIRS) criteria established in 1992 and some additional criteria used by Dr. Emanuel Rivers in his 2001 study of early goal-directed therapy.14 For a sepsis alert to be activated, patients need to meet three general criteria:
1) Physical signs of SIRS;
2) History consistent with infection;
3) Signs of hypoperfusion or hypotension.

Here are the objective findings and treatment sections of the Denver protocol:

Specific Objective Findings
1. Initiate sepsis alert for the following patients:
   a. 18 years old and NOT pregnant
   b. At least two of the SIRS criteria and:
      i. Temperature greater than 38ºC (100.4ºF) or lower than 36ºC (96ºF)
      ii. Pulse greater than 90
      iii. Respiratory rate greater than 20
   c. Suspected or documented infection
   d. Hypoperfusion as manifest by one of the following:
      i. Systolic BP less than 90
      ii. MAP less than 65
      iii. Lactate level greater than 4 mmol/L

1. Administer high-flow oxygen
   a. NRB masks are to be used exclusively
2. Establish IV access with two large-bore angiocaths and draw blood samples
    a. Baseline blood values will be important
    b. Administer IV fluid boluses (20 cc/kg), rapid infusion
       i. Reassess after infusing 500-ml increments:
          1. Blood pressure
          2. Breath sounds
    c. Contact base if BP remains less than 90 systolic
3. Reassess patient on a regular basis. Document appropriately the following:
   a. Vital signs—BP must be auscultated
    b. Breath sounds
    c. Capnography
    d. Pulse oximetry
    e. Blood sugar
    f. Monitor cardiac rhythm
4. Scene times should be less than 15 minutes, with emergent transport to definitive care. Consider emergent return with these patients.

Why Monitor Lactate?

   The single biggest challenges Guerra and Riccio faced when they designed their prehospital sepsis alert program was developing criteria that would be specific enough to identify most severe sepsis patients while minimizing the volume of false alerts. At first the task seemed impossible--until someone posed the idea of adding a lactate reading to the criteria.

   "We looked at the SIRS criteria, and it just looked like it would pick up a ton of people," says Riccio. "It was like, 'Jeez, that matches everyone who's sick.' And if you cry wolf, then the hospitals are going to say, 'You guys don't know what you're doing.' The hospital staff just gets tired of it."

   Faced with the possibility of daily false alerts, Centura physicians initially dismissed the sepsis alert idea. It returned several months later. "We found the lactate meter," says Riccio, "and we thought, Well, we can't do a white blood cell count, but this is something that we can measure that might make us more accurate." Only time and practice will tell for sure, but for now the agencies participating in the research have this new tool.

   Using portable lactate testing required a short leap of imagination. Lactate has been identified as a potential leading indicator for metabolic distress in shock and end-organ failure for some time. It was an inclusion criterion for Dr. Emanuel Rivers' landmark sepsis research in 2001,14 and it remains a hopeful candidate for an early, easy-to-assess physical sign.

   While new to the prehospital arena, portable lactate meters have been around athletic circles for years. Olympic training coaches were the first to popularize the idea of using lactate to determine true anaerobic thresholds in top athletes. The hand-held meters look and work much like the glucometer technology of a decade ago. After a quick morning calibration, they are ready to go and display accurate blood lactate in under a minute.

   While lactate will certainly be helpful in confirming shock in classic sepsis presentations, its real advantage might be in helping identify cryptic septic shock. "There are a percentage of severe sepsis patients who have normal mean arterial numbers," says Guerra, "and they won't be hypotensive yet, but their lactate will be high. That's called cryptic septic shock, and they may be a significant number of septic patients. If you're not checking lactate, you're going to miss them."

The Future of Sepsis Monitoring

   As research continues to demonstrate the dramatic changes in outcome that can be achieved with early recognition and rapid intervention, the pressure increases to accurately recognize severe sepsis and intervene. But the physical assessment markers to help us identify patients transitioning from severe sepsis to septic shock remain cryptic and subtle.

   There is a need for more advanced monitoring tools that can help us look more closely into the patient's true metabolic status and warn us when a systemic inflammatory response is underway. Serum lactate may be one such indicator, but the future holds even more possibilities. Two new and innovative monitoring devices loom on the horizon. Preliminary research has already demonstrated great promise. If their performance remains strong through clinical trials, the future of sepsis recognition and treatment may get a whole lot brighter.

   Total Hemoglobin Monitoring (SpHb)--Pulse oximetery, a staple of prehospital monitoring, tells providers the percentage of oxygen bound to each individual hemoglobin molecule, but nothing about how much hemoglobin exists in the blood. Until recently, the only way to accurately measure hematocrit (SpHb) was to draw the patient's blood, send it to a lab and wait for results.

   Now, Masimo's Rainbow SET CO-oximetry unit is demonstrating comparable accuracy to lab testing with a noninvasive, instantaneous and continuous monitor. In combination with pulse oximetry, SpHb readings can give providers a glimpse of something much more useful than SpO2: total oxygen content (SpOC).

   Hematocrit monitoring could have numerous prehospital applications, but the boon for sepsis care might be its use as an early indicator of shock and its ability to identify accurate endpoints for fluid resuscitation guidelines. Sepsis patients require massive fluid replacement, but dropping the patient's hematocrit could negatively affect total oxygen content. SpHb monitoring would help optimize that process.

   Tissue Oxygenation Monitoring (StO2)--Classic oxygen saturation (SpO2) monitors tell caregivers how much oxygen is being carried around in the blood, but nothing about how well that oxygen is being utilized by the tissues. The InSpectra StO2 system is a noninvasive monitor that measures hemoglobin saturation in the patient's microcirculation and gives a close approximation of how much oxygen is actually being used by the tissues.

   Inadequate tissue perfusion is the hallmark definition of shock. InSpectra aims to allow caregivers a front-row view of cellular oxygen usage. With StO2 monitoring, providers can look into the microvasculature and see shock at its inception. Having shown potential in several clinical trials, InSpectra's monitor could become the leading indicator researchers have been looking for, showing changes at the microvascular level before total oxygen content or serum lactate begin to measurably deviate.

   Both of these monitors have great potential for use in shock identification and treatment, but they are only now entering the hospital environment. It could be years before the first prehospital providers begin adding them to their patient assessment ensemble. For now we'll need to rely on accurate history taking, thorough physical examinations and sound clinical judgment. Those tools will always be the foundation of prehospital medicine, but advanced monitoring devices like these could provide valuable assistance.


1. Associated Press. Mariana Bridi, Brazil Model Who Lost Hands, Feet, Dies. Boston Herald, Jan. 24, 2009.

2. Hitti M. Sepsis Threatens Brazilian Model's Life. WebMD,

3. Cohen E. Sepsis can strike, kill shockingly fast.,

4. Bone RC, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 101(6): 1,644–55, Jun 1992.

5. Organisation for Economic Co-operation and Development. OECD Health Report 2000.

6. Kanji S, Devlin JW, Piekos K, Racine E. Recombinant human activated protein C (drotrecogin alfa activated): A novel therapy for severe sepsis. Pharmacotherapy 21: 1,389–1,402, 2001.

7. The Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology. Acute myocardial infarction: Pre-hospital and in-hospital management. Eur Heart J 17(1): 43–63, Jan 1996.

8. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: Analysis of incidence, outcome, and associated costs of care. Crit Care Med 29(7): 1,303–10, Jul 2001.

9. Surviving Sepsis Campaign. Sepsis: What You Should Know,

10. Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med 36(1): 296–327, Jan 2008.

11. Vincent J, Carlet J, Opal S, eds. The Sepsis Text, pp. 323–38. Norwell, MA: Kluwer Academic Publishers, 2002.

12. Yu DT, Black E, Sands KE, et al. Severe sepsis: Variation in resource and therapeutic modality use among academic centers. Crit Care 7(3): R24–34, Jun 2003.

13. Christiana slashes sepsis mortality rate. Hosp Peer Rev 33(6): 86–8, Jun 2008.

14. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 345(19): 1,368–77, Nov 8, 2001.

15. Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med 348(2): 138–50, Jan 9, 2003.

16. Kramer MR, Vandijk J, Rosin AJ. Mortality in elderly patients with thermoregulatory failure. Arch Intern Med 149(7): 1,521–3, Jul 1989.

17. Oda K, Matsuo Y, Nagai K, Tsumura N, Sakata Y, Kato H. Sepsis in children. Pediatr Int 42(5): 528–33, Oct 2000.

18. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med 345(19): 1,359–67, Nov 8, 2001.

   Steve Whitehead, NREMT-P, is a firefighter/paramedic with the South Metro Fire Rescue Authority in Colorado. He is a primary instructor for South Metro's EMT program and a lifelong student of emergency medicine. Reach him through his blog,

Back to Top