A white blood cell count in sepsis can swing in either direction, and both extremes carry clinical meaning. A count above roughly 12,000 cells per microliter (leukocytosis) signals that the immune system is mobilizing aggressively, while a count below about 4,000 (leukopenia) suggests the body’s defenses are being overwhelmed or consumed faster than they can be replaced. Perhaps most counterintuitively, about half of patients who show up at the hospital with bacteria in their blood have a completely normal white blood cell count, which can lull clinicians into a false sense of security.1PubMed Central. The complete blood count to diagnose septic shock Understanding what the WBC count does and does not reveal in sepsis matters for anyone trying to make sense of lab results during a frightening infection.
Why the WBC Can Go Either Way
When your body detects an infection, it ramps up production of white blood cells in the bone marrow and releases stored reserves into the bloodstream. In many cases, this produces leukocytosis, with counts climbing well above the normal range. But the picture is rarely that simple. In severe or fast-moving infections, white cells rush out of the bloodstream and into infected tissues so quickly that the circulating count actually drops. Neutrophils, the most abundant type of white blood cell, are consumed rapidly during this process as they attempt to trap and destroy bacteria.2PubMed Central. A Case of Severe Sepsis Presenting Marked Decrease of Neutrophils and Interesting Findings on Dynamic CT The result is that a single WBC reading, taken in isolation, tells you surprisingly little. A high count says the immune system is responding. A low count says it may be losing the fight. And a normal count may just mean it’s early.
The Original SIRS Criteria and Why They Changed
For years, clinicians used white blood cell counts as one of four criteria to identify sepsis. The older definition, established in 1991, required at least two of four signs to flag a patient as having systemic inflammatory response syndrome, or SIRS. Those four signs were a heart rate above 90, a respiratory rate above 20, a temperature above 38°C or below 36°C, and a WBC above 12,000, below 4,000, or with more than 10 percent immature cells (bands).3PubMed Central. SIRS, qSOFA and new sepsis definition If you had a suspected infection plus two of those criteria, that was sepsis under the old rules.
The updated definition, known as Sepsis-3, moved away from SIRS entirely. Clinicians recognized that SIRS criteria were too sensitive and not specific enough. Someone who ran a flight of stairs or had a minor cold could technically meet SIRS criteria without having sepsis. The newer framework focuses on organ dysfunction rather than inflammatory markers, using a scoring system that measures how well organs like the kidneys, liver, lungs, and brain are functioning.3PubMed Central. SIRS, qSOFA and new sepsis definition This does not mean the WBC count became irrelevant. It still appears on every standard blood panel in the emergency department and ICU. It just stopped being part of the formal definition of sepsis, because it is too unreliable as a standalone diagnostic tool.
What Band Cells Reveal That the Total WBC Misses
Beyond the headline WBC number, clinicians pay close attention to what types of white blood cells are circulating. One of the most informative clues is a “left shift,” which means the blood contains an unusually high proportion of immature neutrophils called band cells. When the bone marrow is under pressure to push out white cells faster than usual, it releases these not-fully-mature forms. A bandemia of more than 10 percent has a reasonably high specificity for infection, roughly 85 percent, meaning that when you see it, infection is the most likely explanation.4PubMed Central. The complete blood count to diagnose septic shock – Section: Left shift Other causes include recent surgery, major bleeding, and certain blood disorders, but in the right clinical context, a significant left shift is treated as evidence of sepsis until proven otherwise.
What makes bandemia particularly useful is that it can flag infection even when the total WBC looks reassuringly normal. A study of emergency department patients found that bandemia without leukocytosis was still associated with worse infectious outcomes, particularly sepsis.5PubMed. Outcomes of Bandemia without leukocytosis in the emergency department: A retrospective analysis In more concrete terms, another study found that patients with moderate or high band counts but normal overall WBC had roughly double to triple the odds of a positive blood culture, and their odds of dying in the hospital were three to nearly five times higher than patients without bandemia.6PubMed. Bandemia with normal white blood cell counts associated with infection In other words, the composition of the white cells can matter more than the total number.
Low Counts Tend to Be More Dangerous Than High Ones
This is one of the findings that surprises people most. In sepsis, leukopenia generally carries a grimmer prognosis than leukocytosis. A large study of sepsis patients found that mortality was about 27 percent in those with low WBC counts, compared with roughly 20 percent in those with high counts and 15 percent in those with normal counts. After adjusting for age, sex, race, and other health conditions, the patients with leukopenia still had about 60 percent higher odds of dying compared to those with leukocytosis.7PubMed Central. Evaluation of leukopenia during sepsis as a marker of sepsis-defining organ dysfunction That association weakened once overall organ dysfunction was accounted for, suggesting that leukopenia is, at least in part, a marker of how sick the patient already is rather than a direct cause of death.
A separate large study of over 270,000 patients with infection in intensive care found a similar pattern: lower white cell counts, even those still within the normal range, were associated with higher mortality.8PubMed Central. How infection affects the relationship between leukocyte count and mortality risk in intensive care Notably, that study found that extremely high counts (around 40,000) did not substantially raise the risk of death in infected patients compared to moderately elevated counts. The danger was concentrated at the low end of the spectrum. The takeaway for patients and families is that a high WBC in sepsis, while alarming on paper, is the body doing what it should be doing. A low count means the body’s first-line defense is depleted, which is a more precarious situation.
The Trend Matters More Than Any Single Reading
A single WBC snapshot tells you where the patient is at that moment, but the trajectory of the count over hours and days is far more informative. Research tracking WBC patterns in septic shock identified several distinct trajectories, and the one most clearly linked to death was a steadily rising WBC count. Patients whose white cells kept climbing had roughly 3.4 times the mortality risk at 30 days compared to patients whose counts stayed stable and normal.9PubMed Central. White blood cell count trajectory and mortality in septic shock: a historical cohort study
A larger study examining over 15,000 sepsis patients identified four trajectory groups based on average WBC levels over time. Patients in the highest trajectory group (averaging around 24,000) had a 28-day mortality rate of about 31 percent, while patients in the lowest trajectory group (averaging around 6,600) had a mortality rate near 13 percent. The high-trajectory group also had significantly higher rates of secondary kidney injury.10PubMed Central. The Impact of White Blood Cell Count Trajectories on Prognosis and Secondary Acute Kidney Injury in Sepsis Patients This helps reconcile the earlier point about leukopenia being worse in single readings: a patient who starts low may be in acute danger, but a patient whose count keeps rising despite treatment may be failing to control the infection, which is its own kind of danger. Serial lab draws are the norm in ICU settings precisely for this reason.
The Neutrophil-to-Lymphocyte Ratio
Rather than looking at white cells as a single number, clinicians increasingly calculate the ratio between neutrophils (the infection-fighting cells) and lymphocytes (the cells involved in longer-term immune responses). In sepsis, neutrophil counts surge while lymphocyte counts drop, so the ratio climbs. A meta-analysis of nine studies covering more than 10,000 patients found that a higher neutrophil-to-lymphocyte ratio was associated with worse prognosis, with each increment corresponding to about 75 percent higher hazard of death.11PubMed. Prognostic value of neutrophil-to-lymphocyte ratio in sepsis: A meta-analysis
In ICU patients, the ratio averaged around 9.5 overall but was higher (around 10.3) in patients with septic shock compared to those with sepsis alone. A cut-off value above roughly 10 showed good specificity, about 78 percent, for identifying patients with more severe disease, though its sensitivity was limited.12PubMed Central. Neutrophil to Lymphocyte Ratio (NLR)—A Useful Tool for the Prognosis of Sepsis in the ICU The appeal of this ratio is that it comes free with any standard blood panel; no extra test is needed. It is not definitive on its own, but it adds a layer of information beyond the total WBC.
Gram-Negative Versus Gram-Positive Infections
The type of bacteria causing sepsis can influence how the WBC count behaves. Research on ICU patients with bloodstream infections found that gram-negative bacteria (the kind that include E. coli, Klebsiella, and Pseudomonas) produced substantially higher WBC counts and sharper rises in the days around diagnosis compared to gram-positive bacteria (the kind that include Staphylococcus and Streptococcus).13PubMed Central. Dynamics of C-reactive protein and white blood cell count in critically ill patients with nosocomial Gram positive vs. Gram negative bacteremia: a historical cohort study Gram-positive infections tended to produce more modest changes in the WBC, sometimes barely moving the needle. This does not help clinicians decide which antibiotic to use on its own, but it adds context: a dramatic WBC spike in a critically ill patient may be a soft clue pointing toward a gram-negative source.
Immune Function Degrades Even When Numbers Recover
One of the harder concepts to grasp is that white blood cells can be present in normal or even high numbers yet function poorly. Several days after the initial septic event, patients enter an immunosuppressive phase. During this window, typically between days three and eight, neutrophils show markedly impaired ability to migrate toward infection sites, produce the reactive oxygen species needed to kill bacteria, and carry out their normal defensive functions. Studies have also found increased numbers of circulating immature granulocytes during this phase, and these functional deficits were associated with higher risk of death from septic shock.14Journal of Leukocyte Biology. Marked alterations of neutrophil functions during sepsis-induced immunosuppression This means a patient whose WBC count normalizes after the first few days is not necessarily out of the woods. The cells present may be functionally exhausted or immature replacements that have not yet developed full killing capacity.
Age Changes Everything About White Cell Interpretation
The WBC count does not mean the same thing at every age. In older adults, the immune system’s ability to mount a proper inflammatory response is blunted by age-related changes. The febrile response and the recruitment of white blood cells may both be diminished, meaning older adults with serious sepsis can present without the high temperatures or the leukocytosis that would raise red flags in a younger patient.15Critical Care Nursing Quarterly. Severe Sepsis in Older Adults Three of the four traditional hallmark responses to systemic infection (temperature, heart rate, and white cell count) can be muted in older people. This is one reason sepsis in elderly patients is so frequently diagnosed late.
At the other end of the age spectrum, newborns present their own interpretive challenges. In neonatal sepsis, a low total white cell count and a low absolute neutrophil count are more strongly associated with infection than a high count. High ratios of immature-to-total neutrophils also flag infection. But the sensitivity of all these measures is frustratingly low, sometimes catching fewer than half of infected newborns, even though specificity is high.16PubMed Central. Use of the Complete Blood Cell Count in Early-Onset Neonatal Sepsis A separate study confirmed that total white cell count and absolute neutrophil count were significantly lower in septic neonates compared to healthy controls, with the neutrophil count showing particularly good specificity, above 90 percent.17PubMed Central. Role of complete blood cell count parameters in the diagnosis of neonatal sepsis The practical implication for parents is that doctors may pursue a sepsis workup in a newborn even when the WBC count looks “normal,” because normal ranges are different at that age and a normal number does not rule out infection.
How WBC Compares to Other Sepsis Biomarkers
The WBC count is universally available, cheap, and fast, but it is far from the best biomarker for diagnosing or predicting sepsis. Several studies have compared it directly against other common blood tests. Procalcitonin, a protein that rises more specifically in response to bacterial infections, has consistently shown better diagnostic and prognostic power than WBC, either alone or when combined with WBC.18PubMed. Comparison between white blood cell count, procalcitonin and C reactive protein as diagnostic and prognostic biomarkers of infection or sepsis in patients presenting to emergency department Lactate, which measures how much tissue is being starved of oxygen, had the strongest independent association with sepsis in one emergency department study, while WBC and neutrophil counts, though statistically significant, all had sensitivity and specificity below 70 percent at their best cut-offs.19PubMed. Evaluation of lactate, white blood cell count, neutrophil count, procalcitonin and immature granulocyte count as biomarkers for sepsis in emergency department patients Research in pediatric respiratory infections similarly found that procalcitonin and C-reactive protein outperformed WBC for distinguishing bacterial infections, though the combination of all three was most useful in practice.20PubMed Central. Usefulness of procalcitonin (PCT), C-reactive protein (CRP), and white blood cell (WBC) levels in the differential diagnosis of acute bacterial, viral, and mycoplasmal respiratory tract infections in children
None of this means the WBC is useless. It arrives first, often within minutes of a blood draw, while procalcitonin results can take an hour or more depending on the facility. It also provides the differential count (the breakdown of neutrophils, lymphocytes, bands, and other cell types) that yields the ratio and left-shift information discussed earlier. Clinicians rarely make sepsis decisions based on any single number. They combine the WBC with lactate, procalcitonin, organ function scores, vital signs, and clinical appearance.
Medications That Distort the Count
Corticosteroids are one of the most common confounders when interpreting a WBC count in a sick patient. These drugs are frequently given in sepsis for shock that is not responding well to fluids and vasopressors, and they are also used for a wide range of other conditions. Steroids predictably raise the white blood cell count by mobilizing neutrophils from the bone marrow and vessel walls and reducing their migration into tissues. In patients with acute infections who were receiving steroid therapy, overall maximum WBC counts averaged around 15,000 compared to about 13,000 in patients not on steroids. In patients chronically treated with steroids, the average WBC bump attributable to the drug was roughly 5,000 cells per microliter.21PubMed. Estimations of a degree of steroid induced leukocytosis in patients with acute infections Clinicians managing sepsis in patients on corticosteroids must mentally subtract this steroid effect when evaluating trends; a “reassuringly stable” WBC in a steroid-treated patient may actually mask a declining count.
Chemotherapy is the other major medication class that skews interpretation, though in the opposite direction. Patients receiving cancer treatment often have profoundly low white cell counts (neutropenia) at baseline, making the WBC count nearly meaningless as a sepsis indicator. In these patients, any fever is treated as a potential emergency regardless of the WBC number.
Lab Artifacts and Point-of-Care Limitations
Not all WBC counts are created equal from a technical standpoint. Point-of-care devices, which are increasingly used in resource-limited settings and field hospitals, can produce misleading results under certain conditions. Evaluation of one widely used portable WBC analyzer found that specific interfering substances could distort measurements. Patients with sickle cell disease or thalassemia were especially prone to falsely high counts because the device could not distinguish nucleated red blood cells from white blood cells, and because abnormal red cells resisted the lysis step that normally clears them from the sample.22PubMed Central. Point-of-care method for total white cell count: an evaluation of the HemoCue WBC device The device also lacked the ability to flag results affected by elevated reticulocyte counts, only alerting users at extreme WBC values. In a sepsis workup, an artificially inflated or deflated WBC reading from a portable device could send clinicians down the wrong path, which is why lab-confirmed counts remain the standard in hospital settings.
G-CSF for Dangerously Low Counts
When sepsis drives the neutrophil count dangerously low, clinicians sometimes consider granulocyte colony-stimulating factor (G-CSF), a drug that stimulates the bone marrow to produce more white cells. The evidence here is mixed and somewhat sobering. One study of septic patients with low neutrophil counts found that those whose bone marrow responded well to G-CSF had a 90 to 100 percent survival rate, while patients whose marrow did not respond at all had a zero percent survival rate. The drug worked best in patients who had low levels of immature neutrophils and whose own G-CSF production was not already maxed out, while it had little effect in patients whose marrow was already severely depleted and pushing out immature cells as fast as it could.23PubMed. Difference in the responses after administration of granulocyte colony-stimulating factor in septic patients with relative neutropenia
Despite these intriguing findings, the broader evidence has not supported routine G-CSF use in sepsis. Reviews of large randomized trials have concluded that G-CSF as an add-on to standard sepsis care does not clearly improve survival or reduce complications in the general sepsis population.24American Journal of Health-System Pharmacy. Use of granulocyte colony-stimulating factor in patients with severe sepsis or septic shock The drug remains an option considered on a case-by-case basis, especially in patients with chemotherapy-induced neutropenia who develop sepsis, but it is not a standard part of the sepsis treatment bundle.