Yes, sepsis can occur without fever. In fact, a substantial minority of septic patients present with normal temperature or hypothermia, and hypothermic sepsis carries a higher mortality rate than febrile sepsis. Temperature is a single data point, not a diagnostic gatekeeper.
Sepsis is a dysregulated host response to infection that causes life-threatening organ dysfunction. The body's thermoregulatory set point can fail or reset unpredictably during this cascade, especially in older adults, immunocompromised patients, and those with severe disease. Relying on fever as a screening trigger misses these presentations, sometimes with fatal consequences.
The thermoregulatory mechanism during sepsis
Body temperature is governed by the hypothalamus, which acts as a thermostat comparing core temperature to a set point. Infection normally triggers pyrogens — cytokines like interleukin-1, interleukin-6, and tumor necrosis factor — that raise the set point. The body then shaves, chills, and vasoconstricts to generate heat and reach that new target. This is why fever is common in early infection.
During sepsis, this system can break in several directions. In hypothermia, the hypothalamus may fail to mount or sustain a pyrogenic response, so the set point does not rise. In afebrile sepsis, the set point may rise only marginally or the body may be unable to generate enough heat to reach it. Elderly patients often have blunted cytokine responses and impaired thermogenesis. The result is a septic patient whose temperature reads normal or low while organ dysfunction progresses.
A plain way to hold it: fever is the body's furnace turning up, and sepsis without fever means either the thermostat stopped responding or the furnace cannot keep up.
How common is afebrile or hypothermic sepsis?
Fever is absent in a meaningful share of septic patients. In large cohort studies, roughly 10 to 30 percent of patients with sepsis present without fever, and hypothermia (core temperature typically below 36.0°C or 96.8°F) appears in a smaller subset. The proportions vary by setting and population. ICU cohorts tend to show more hypothermia than emergency department cohorts, reflecting more severe disease.
Hypothermic sepsis is consistently associated with worse outcomes. Mortality in hypothermic septic patients is often reported at roughly double that of febrile patients in observational studies, though estimates vary. Absence of fever is not a sign of mild disease. It can be a sign that the host response has failed.
Temperature distributions in sepsis also shift with age. Older adults mount fever less reliably, and the threshold for what counts as fever may be lower in that population. A temperature of 37.8°C (100.0°F) in a frail 85-year-old may be more clinically meaningful than 38.3°C (101.0°F) in a healthy 30-year-old.
Who is most likely to present without fever
Certain groups are far more likely to have sepsis without fever. Recognizing these populations matters more than any single thermometer reading.
Adults 65 and older, who often have blunted febrile responses. Patients on immunosuppressants, including corticosteroids and transplant medications. People with advanced cancer or receiving chemotherapy. Patients with cirrhosis, chronic kidney disease, or other organ failure. Newborns and young infants, whose temperature regulation is immature. Patients with indwelling devices, recent surgery, or prolonged ICU stays.
In these groups, clinicians should lower the threshold for suspecting sepsis and weight other signs — confusion, hypotension, tachypnea, and rising lactate — more heavily than temperature. A normal temperature never rules out sepsis.
The counterintuitive detail worth holding onto: severe sepsis can present as hypothermia rather than fever, and that presentation is the more dangerous one. The sicker patient may be the cooler one.
Fever thresholds, screening tools, and how they reconcile
Several tools use temperature as one criterion among many. The thresholds differ, and reconciling them prevents confusion. The following table compares commonly used screening frameworks.
| Tool or criterion | Temperature element | Notes |
|---|---|---|
| SIRS criteria | Temp above 38.0°C (100.4°F) or below 36.0°C (96.8°F) | Requires 2 of 4 criteria; temperature alone is neither necessary nor sufficient |
| qSOFA | No temperature criterion | Uses respiratory rate, systolic BP, mental status |
| NEWS2 | Temperature scored on a graded scale | Low and high temperatures both add points |
| Sepsis-3 definition | No temperature criterion | Organ dysfunction defined by SOFA score change |
| CDC adult sepsis guidance | Temperature among screening signs | Emphasizes clinical judgment |
The key reconciliation: SIRS uses a low-temperature threshold of 36.0°C (96.8°F) as abnormal, while many clinical definitions of hypothermia use the same 36.0°C cutoff. But fever thresholds vary — 38.0°C (100.4°F) in SIRS versus 38.3°C (101.0°F) in some older definitions. The practical takeaway is that no single number defines sepsis. Temperature is a screening prompt, not a diagnosis.
Sepsis-3, the consensus definition, removed temperature from the diagnostic criteria entirely. It defines sepsis as life-threatening organ dysfunction caused by a dysregulated host response to infection, operationalized as an increase in the SOFA score of 2 or more points. Temperature is nowhere in that equation. This is the clearest statement that fever is neither required nor sufficient.
The time course of afebrile sepsis
Sepsis evolves over hours to days, and temperature can shift across that arc. A patient may be febrile early and become hypothermic as shock develops. Tracking the timeline helps clinicians interpret a single reading.
| Phase | Typical timeframe | Temperature pattern |
|---|---|---|
| Infection onset | Hours to days | Fever common but not universal |
| Early sepsis | 0–6 hours | Fever, or normal temp in at-risk groups |
| Progression to severe sepsis | 6–24 hours | Temp may normalize or fall |
| Septic shock | Hours after onset | Hypothermia increasingly common |
| Resuscitation | 24–72 hours | Temp fluctuates; may remain low |
| Recovery | Days to weeks | Fever may recur with secondary infection |
The clinical lesson: a falling temperature in a deteriorating patient is a red flag, not reassurance. Serial measurements and the trajectory matter more than any single value.
Practical actions and red flags
For clinicians and caregivers, the practical approach is to treat temperature as one input among several. The following steps are reasonable and concrete.
Assess for organ dysfunction regardless of temperature: check mental status, respiratory rate, blood pressure, urine output, and skin perfusion. Use a validated screening tool (qSOFA, NEWS2, or SIRS) as a prompt, not a verdict. Obtain lactate and blood cultures when sepsis is suspected, even if the patient is afebrile. Lower the fever threshold in older or immunocompromised patients — consider 37.8°C (100.0°F) or even lower as concerning. Recheck temperature serially; a falling temperature in a worsening patient warrants escalation.
Red-flag warnings that should prompt urgent evaluation:
New confusion or altered mental status in an older adult. Systolic blood pressure below 90 mmHg or a drop of more than 40 mmHg from baseline. Respiratory rate of 22 breaths per minute or higher. Lactate of 2 mmol/L or higher, or rising on repeat measurement. Core temperature below 36.0°C (96.8°F) in a patient with suspected infection. Mottled skin, prolonged capillary refill, or decreased urine output.
These signs indicate possible organ dysfunction and should trigger immediate clinical assessment. They are not diagnostic on their own, but they justify escalation.
Limitations and what temperature cannot tell you
Temperature is a crude signal. It varies by measurement site — rectal and core readings run higher than oral, and oral higher than axillary — so a "normal" reading depends partly on how it was taken. A temperature of 37.5°C (99.5°F) measured orally may correspond to a core temperature near 38.0°C (100.4°F). Site and technique matter.
Thermoregulation also varies with time of day, medications, and underlying disease. Antipyretics can mask fever. Beta-blockers and other drugs can blunt the response. Chronic conditions alter baseline temperature. No single value reliably separates septic from non-septic patients.
The most important limitation is that sepsis is a syndrome defined by organ dysfunction, not by temperature. A patient can meet Sepsis-3 criteria with a completely normal temperature. Conversely, fever is common in many non-septic conditions. Neither the presence nor absence of fever should override clinical judgment.
A final caveat: this article describes general mechanisms and does not constitute medical advice. Anyone with suspected sepsis needs prompt evaluation by a clinician, regardless of what the thermometer reads.