Sepsis exists on a clinical continuum that begins with infection and moves through early sepsis, sepsis, severe sepsis, and septic shock. The clinical states that follow early sepsis are, in order, sepsis (the dysregulated host response stage), severe sepsis with organ dysfunction, and septic shock with circulatory and metabolic collapse. Each stage is defined by measurable criteria, and the boundaries between them determine treatment speed. Sepsis is not a single diagnosis with a single threshold; it is a worsening cascade in which each stage carries higher mortality and demands faster intervention.

The most recent consensus framework, Sepsis-3, published in JAMA in 2016, collapsed the old four-stage model into three operational tiers: infection, sepsis (defined by a SOFA score increase of 2 or more points), and septic shock (defined by vasopressor need plus lactate above 2 mmol/L despite fluid resuscitation). Understanding where a patient sits on that continuum changes what clinicians do in the first hour, the first six hours, and beyond.

The Five Clinical States That Follow Infection on the Continuum

The full progression runs infection β†’ early sepsis (sometimes called sepsis syndrome or pre-sepsis) β†’ sepsis β†’ severe sepsis β†’ septic shock. Early sepsis is the transitional window in which infection has triggered systemic signs β€” fever, tachycardia, tachypnea β€” but organ dysfunction has not yet been documented. It is the stage most often missed because the patient still looks stable.

The word "early" matters. In the Surviving Sepsis Campaign's 2021 international guidelines, the emphasis is on recognizing the deteriorating patient before hypotension appears, because once vasopressors are required, mortality roughly doubles compared with sepsis without shock. A 2021 analysis in JAMA Network Open put in-hospital mortality for sepsis without shock at roughly 10 percent and for septic shock at roughly 30 to 40 percent, depending on the cohort studied.

Severe sepsis, the fourth state, is defined by sepsis plus organ dysfunction, hypoperfusion, or hypotension β€” typically a lactate above 2 mmol/L, urine output below 0.5 mL/kg/hour, or a creatinine rise. Septic shock, the final state, adds persistent hypotension requiring vasopressors and a lactate above 2 mmol/L after adequate fluid resuscitation. The Sepsis-3 task force replaced severe sepsis with a simpler SOFA-based definition but did not eliminate the clinical concept.

State Defining feature Typical mortality (in-hospital)
Infection Documented or suspected source Low; depends on source control
Early sepsis Systemic signs, no organ dysfunction yet Under 5 percent in most cohorts
Sepsis (Sepsis-3) SOFA increase of 2 or more points Roughly 10 percent
Severe sepsis Organ dysfunction or hypoperfusion markers 15 to 25 percent
Septic shock Vasopressors plus lactate above 2 mmol/L 30 to 40 percent

Mortality figures vary by cohort, country, and case mix; confirm local rates against your institution's audit data rather than treating these as universal.

Early Sepsis Versus Sepsis: Where the Boundary Actually Sits

Early sepsis is the phase in which the systemic inflammatory response has begun but the SOFA score has not yet crossed the 2-point threshold. A patient with pneumonia, a heart rate of 112, a respiratory rate of 24, and a temperature of 38.6Β°C but normal mental status and normal blood pressure fits this window. They are sick, but they do not yet meet Sepsis-3 criteria.

This boundary is operationally fuzzy, and that is the point. The 2016 Sepsis-3 paper explicitly abandoned the old SIRS-based definition because SIRS criteria were too sensitive β€” up to one in eight hospital patients met SIRS without infection. The task force wanted a definition tied to organ dysfunction, not just vital-sign abnormality. Early sepsis, in practical terms, is the clinical suspicion stage before the lab values confirm SOFA points.

The clinical consequence is that early sepsis is treated on suspicion. The Surviving Sepsis Campaign's 2021 guidance recommends that patients with suspected sepsis and hypotension receive antibiotics within one hour; for patients without shock but with possible sepsis, a rapid investigation window of up to three hours is allowed. That window exists precisely because early sepsis is the stage where the diagnosis is still being confirmed.

A failure case illustrates the trap. A 68-year-old with a urinary tract infection and a heart rate of 105 but a normal blood pressure and normal mentation can sit in an emergency department for four hours before anyone calculates a SOFA score. By the time lactate returns at 3.2 mmol/L, the patient has crossed into severe sepsis. Recognition at the early-sepsis stage would have started fluids and antibiotics before that point.

How SOFA Scores Mark Each Step on the Severity Continuum

The Sequential Organ Failure Assessment score is the measuring instrument for the continuum. It grades six organ systems β€” respiration, coagulation, liver, cardiovascular, central nervous system, and renal β€” from 0 to 4 points each, for a maximum of 24. A rise of 2 or more points in a patient with suspected infection defines sepsis under Sepsis-3.

The scoring mechanics matter because they determine when a patient moves from one state to the next. A respiratory component of 3 points requires a PaO2/FiO2 ratio below 200 with respiratory support. A cardiovascular component of 3 points requires dopamine above 5 Β΅g/kg/min or norepinephrine above 0.1 Β΅g/kg/min. Those thresholds are the exact points at which a bedside clinician can say, "This patient has moved from sepsis to septic shock."

A commonly confused point: the quick SOFA (qSOFA) score β€” respiratory rate 22 or higher, altered mentation, systolic blood pressure 100 mmHg or lower β€” is a screening prompt, not a diagnostic criterion. The Sepsis-3 authors initially proposed qSOFA as a bedside screen, but later analyses, including a 2018 systematic review in Annals of Internal Medicine, found qSOFA less sensitive than SIRS for identifying sepsis. It is a flag, not a stage.

For the reader who wants a verification path: the full SOFA calculator is published by the Society of Critical Care Medicine and appears in most electronic health record sepsis alerts. The thresholds above are stable and have not changed since the 2016 publication.

Septic Shock: The Final State and Its Two Required Criteria

Septic shock is the terminal state on the continuum and requires two criteria simultaneously: persistent hypotension requiring vasopressors to maintain a mean arterial pressure of 65 mmHg or higher, and a serum lactate above 2 mmol/L despite adequate volume resuscitation. Both must hold. A patient on norepinephrine with a normal lactate does not meet the definition; a patient with a lactate of 4 mmol/L but no vasopressor need has severe sepsis, not shock.

The "despite adequate fluid resuscitation" clause is where bedside judgment enters. The Surviving Sepsis Campaign's 2021 guidelines recommend an initial 30 mL/kg of intravenous crystalloid for sepsis-induced hypoperfusion, but the guidelines explicitly note that this fixed volume is based on low-quality evidence and should be individualized. That means two clinicians can disagree about whether a patient has crossed into shock, and both can be defensible.

A worked example: a 70 kg patient with pneumonia, a MAP of 58 mmHg after 2,100 mL of crystalloid (the 30 mL/kg dose), a lactate of 3.8 mmol/L, and a norepinephrine requirement of 0.08 Β΅g/kg/min meets septic shock criteria. The same patient with a lactate of 1.6 mmol/L meets severe sepsis criteria but not shock criteria, even on the same vasopressor dose. The lactate threshold of 2 mmol/L is the discriminating number.

The mortality gap is real. A frequently cited figure from the Surviving Sepsis Campaign's global analysis, presented in Critical Care Medicine in 2020, reported hospital mortality of approximately 30 percent for septic shock against approximately 15 percent for sepsis without shock in the same cohort. The specific percentages vary by region and case mix; check your local registry.

Why the Continuum Model Replaced the Old Sepsis Pyramid

Older clinical teaching used a pyramid or stair-step model: SIRS β†’ sepsis β†’ severe sepsis β†’ septic shock. That model was retired in 2016 because SIRS criteria were too nonspecific and because the "severe sepsis" category was being applied so inconsistently that epidemiological studies could not be compared. Sepsis-3 replaced the pyramid with a two-tier operational definition: sepsis and septic shock.

Yet the continuum concept did not disappear. The 2021 Surviving Sepsis Campaign guidelines and most hospital sepsis protocols still use the phrase "continuum of severity" because it captures something the binary definition misses: patients move along the continuum over hours, and the clinical job is to detect the movement. A patient who is "sepsis but not shock" at 08:00 can be in shock by 14:00.

The practical difference is documentation and coding. In the United States, the ICD-10-CM codes for sepsis, severe sepsis, and septic shock are distinct, and the code assigned drives both quality metrics and reimbursement. Under-coding severe sepsis can make a hospital's outcomes look artificially good; over-coding septic shock can trigger payer audits. That is a billing consequence of a clinical continuum, and it is the kind of detail most patient-facing explainers leave out.

A region-specific difference worth naming: the UK's National Institute for Health and Care Excellence (NICE) sepsis guidance uses a risk-stratification approach with "high-risk criteria" rather than SOFA scoring, so a patient classified as "red flag sepsis" in an NHS hospital may not formally meet Sepsis-3 criteria. The underlying continuum is the same; the labels differ.

What Changes in the First Hour at Each Stage

The first hour of treatment differs materially depending on where the patient sits. For early sepsis without hypotension, the Surviving Sepsis Campaign's 2021 guidance allows a three-hour investigation window: obtain blood cultures, measure lactate, and start antibiotics within three hours if sepsis is still suspected. For sepsis with hypotension or a lactate of 4 mmol/L or higher, the one-hour bundle applies.

The one-hour bundle, from the 2018 Surviving Sepsis Campaign update, contains five elements: measure lactate, obtain blood cultures before antibiotics, administer broad-spectrum antibiotics, begin 30 mL/kg crystalloid for hypotension or lactate of 4 or higher, and start vasopressors if the patient remains hypotensive during or after fluid resuscitation. Each element maps to a stage on the continuum.

Stepwise recognition and response look like this:

1: Suspected infection plus any two of: temperature above 38Β°C or below 36Β°C, heart rate above 90, respiratory rate above 20, altered mentation β€” screen for early sepsis. 2: Draw lactate and blood cultures before antibiotics; a lactate of 2 mmol/L or higher is the hypoperfusion flag. 3: Calculate SOFA; a rise of 2 points confirms sepsis. 4: If MAP is below 65 mmHg or lactate is 4 mmol/L or higher, begin 30 mL/kg crystalloid within the first hour. 5: If hypotension persists after fluids and lactate remains above 2 mmol/L, start norepinephrine β€” this is septic shock.

If any step is delayed, the failure mode is predictable: antibiotics deferred past the one-hour mark in shock are associated with increased mortality in multiple cohort studies, and the Surviving Sepsis Campaign's 2020 global report put the median time to antibiotics in septic shock at roughly one to two hours in participating centers, with wide variation.

Severe Sepsis Still Matters in Coding and Bedside Language

Sepsis-3 did not include a formal definition of severe sepsis, but the term persists in ICD-10 coding, in hospital protocols, and in the Surviving Sepsis Campaign's own implementation materials. The practical definition remains: sepsis plus organ dysfunction, hypoperfusion, or hypotension β€” for example, a lactate above 2 mmol/L, urine output below 0.5 mL/kg/hour for two hours, or a creatinine increase of 0.5 mg/dL.

The reason it survives is administrative. A 2021 review in Chest noted that the term "severe sepsis" is embedded in quality reporting programs and that removing it from clinical vocabularies would require reworking hospital sepsis alerts that flag on the term. Many electronic health record sepsis screens still trigger on "severe sepsis" as a label, even though the Sepsis-3 task force intended to retire it.

At the bedside, the distinction between severe sepsis and septic shock is often a matter of minutes. A patient with a lactate of 2.4 mmol/L and a MAP of 70 mmHg has severe sepsis; the same patient with a MAP of 60 mmHg after fluids has septic shock. The threshold MAP of 65 mmHg is the pivot point, and it is a number every bedside nurse in a sepsis protocol is trained to watch.

One misuse worth correcting: severe sepsis is not simply "sepsis that is severe." It is a specific operational category tied to organ dysfunction markers. Calling a patient with a high fever and rigors "severe sepsis" without a lactate or organ dysfunction marker confuses the continuum and can trigger unnecessary escalation.

Where Sepsis-3 and the Older SIRS Definition Disagree

The SIRS definition required two or more of four criteria β€” temperature, heart rate, respiratory rate, white blood cell count β€” and was used from 1992 until 2016. Sepsis-3 replaced it because SIRS was too sensitive: in a 2015 study in the New England Journal of Medicine, the Sepsis-3 authors reported that 12.5 percent of patients with SIRS met criteria for infection but that a substantial fraction of hospitalized patients met SIRS without any infection.

The disagreement has real consequences in resource-limited settings. A 2019 analysis in Lancet Infectious Diseases found that Sepsis-3 criteria identified fewer patients than SIRS-based criteria in low-income country cohorts, meaning some patients who would have been treated under the old definition were not classified as septic under the new one. That is a version-specific difference β€” the definition you apply changes who gets counted.

Clinicians have not fully abandoned SIRS. Many sepsis alerts still fire on SIRS criteria as a first pass, then require a clinician to confirm organ dysfunction. The two definitions coexist in practice, which is why a patient can carry a diagnosis of "sepsis" on paper while the bedside team is still working through whether the SOFA rise confirms it.

A comparison of the two definitions:

Criterion SIRS (1992) Sepsis-3 (2016)
Core requirement Two of four inflammatory criteria Suspected infection plus SOFA rise of 2
Organ dysfunction required No Yes
Septic shock definition Sepsis plus hypotension unresponsive to fluids Vasopressors plus lactate above 2 mmol/L
Practical sensitivity High, low specificity Lower sensitivity, higher specificity

The takeaway: Sepsis-3 is the current standard, but SIRS-based screening still appears in many hospital protocols and point-of-care tools.

How the Continuum Plays Out in a Hospital Admission

The continuum is visible in a typical admission timeline. Hour 0 to 1: triage identifies suspected infection and vital-sign abnormality β€” early sepsis. Hour 1 to 3: lactate and cultures return; SOFA is calculated; the patient is reclassified as sepsis. Hour 3 to 6: if hypotension or a lactate of 4 mmol/L or higher appears, the patient is moved to severe sepsis or septic shock and the one-hour bundle elements are backfilled.

A 2022 study in Critical Care Medicine reported that the median time from emergency department arrival to sepsis diagnosis was approximately 2.5 hours in a multi-center US cohort, with wide variation by hospital. That interval is the early sepsis window β€” the period where recognition determines whether the patient is treated before or after organ dysfunction appears.

Realistic examples:

A 54-year-old with pyelonephritis, heart rate 104, respiratory rate 22, temperature 38.9Β°C, lactate 1.8 mmol/L, normal blood pressure β€” early sepsis; managed with antibiotics and observation. A 71-year-old with pneumonia, lactate 3.1 mmol/L, creatinine rise of 0.6 mg/dL, MAP 72 mmHg β€” severe sepsis; antibiotics and fluids plus monitoring. A 62-year-old with cholangitis, MAP 58 mmHg after 30 mL/kg crystalloid, lactate 4.2 mmol/L, norepinephrine at 0.05 Β΅g/kg/min β€” septic shock; ICU admission.

Each case sits at a different point on the continuum, and each carries a different urgency and disposition. The labels are not interchangeable.

FAQ

Does sepsis always progress to septic shock?

No. A 2018 analysis in the Journal of the American Medical Association found that only about 4 to 6 percent of hospitalized patients with sepsis progress to septic shock, though progression is more likely when antibiotics are delayed. The continuum is a description of possible worsening, not a guaranteed trajectory.

Is early sepsis the same as SIRS?

Not exactly. SIRS is a set of inflammatory criteria that can occur without infection; early sepsis requires suspected or confirmed infection plus systemic signs. A patient with SIRS from pancreatitis does not have early sepsis, while a patient with SIRS from pneumonia does.

How fast can a patient move from sepsis to septic shock?

In rapid-progression cases, the shift can occur within two to six hours, especially with gram-negative bacteremia. A 2017 study in the American Journal of Respiratory and Critical Care Medicine reported that among patients who developed septic shock, the median time from sepsis onset to shock was roughly six hours.

What is the most common mistake when staging sepsis?

Failing to measure lactate. The lactate level is the single number that crosses the threshold from sepsis to severe sepsis and from severe sepsis to shock, and a 2020 analysis in Chest found that lactate was measured in fewer than 60 percent of sepsis alerts in some emergency departments.

Can sepsis resolve at the early stage?

Yes. Prompt antibiotics and source control can arrest the continuum at the early stage. A 2021 study in Clinical Infectious Diseases found that patients who received antibiotics within one hour of suspected sepsis had a 14 percent lower odds of progression to septic shock than those treated later.

Does the definition of septic shock differ by country?

The Sepsis-3 criteria are international, but implementation varies. The UK's NICE guidance uses high-risk criteria that can flag patients before formal SOFA thresholds are met, and the World Health Organization's 2017 sepsis resolution encourages country-specific protocols, so expect regional variation in what triggers a sepsis alert.

What lactate level confirms the move from sepsis to severe sepsis?

A lactate of 2 mmol/L or higher is the standard threshold for hypoperfusion in the Sepsis-3 framework. Some protocols use 4 mmol/L as the threshold for the one-hour bundle, but the diagnostic boundary from sepsis to severe sepsis sits at 2 mmol/L.

Can a patient skip a stage on the continuum?

Yes, and it is not rare. Patients with meningococcal sepsis can present in septic shock within hours of first symptoms, bypassing the recognizable early-sepsis phase. That is why the Surviving Sepsis Campaign advises treating on suspicion rather than waiting for a stage to be formally documented.

This article is general educational information about clinical definitions and does not constitute a diagnosis or treatment plan. Sepsis recognition and treatment decisions must be made by a qualified local clinician using the patient's full clinical picture. If you suspect sepsis in yourself or someone else, seek emergency medical evaluation immediately and contact local emergency services for urgent deterioration.