CSF Lactate Interpreter
CSF Lactate Interpreter
Interpret a CSF lactate against the 3.5 mmol/L (31.5 mg/dL) threshold for bacterial meningitis, in either unit — and with the two clinical situations that move the threshold rather than remove it.
CSF Lactate Interpreter
Value + unit → mmol/LCSF lactate reported as 6.4 mmol/L, community-acquired presentation, no antibiotics given before the lumbar puncture
Threshold and conversion
mg/dL ÷ 9.008 = mmol/L · mmol/L × 9.008 = mg/dL
Usual reference interval roughly 1.1–2.4 mmol/L (10–21.6 mg/dL)
- no paired serum needed
- the CSF:serum glucose ratio requires a blood glucose drawn at the same time, which is frequently forgotten, drawn late, or drawn after a glucose infusion has started. CSF lactate is a single measurement on a single tube, which is much of the reason it performs better in practice than it does on paper
- 9.008
- the molar mass of lactate in g/mol. It is the only arithmetic on this page, and getting it wrong by a factor of ten is the classic way a threshold is misread across the Atlantic
- antibiotics first
- the single biggest threat to this number. Prior antibiotic exposure lowers CSF lactate as it sterilises the culture, so the two tests that would have disagreed are silenced together
- post-neurosurgical
- the threshold rises rather than disappearing. Published postoperative cut-offs range from about 3.45 to 5.4 mmol/L, with a pooled sensitivity of 0.92 and specificity of 0.88 across that range
Worked example
CSF lactate reported as 6.4 mmol/L, community-acquired presentation, no antibiotics given before the lumbar puncture
The result is already in mmol/L, so no conversion is applied: 6.40 mmol/L
In mg/dL the same result would be 6.4 × 9.008 = 57.7 mg/dL, and the threshold of 3.5 mmol/L would appear as 31.5 mg/dL
6.40 is well above 3.5, which in an untreated community-acquired presentation is the strongest single biochemical pointer to a bacterial cause on the whole CSF profile
Note what did not have to be done: no paired serum glucose was needed, so this answer survives a night when the blood tube was never sent
It remains a pointer and not a diagnosis. Empirical antibiotics go in now, Gram stain, culture and PCR go off now, and the number is read next to the cell count, differential, protein and glucose ratio rather than instead of them
CSF lactate against the rest of the profile
| Parameter | Typical in bacterial meningitis | Typical in viral meningitis | Needs a paired blood sample? |
|---|---|---|---|
| Lactate | Median around 13 mmol/L | Median around 2.3 mmol/L | No |
| CSF:blood glucose ratio | Median around 0.1 | Median around 0.52 | Yes |
| Total white cell count | Median around 1,300/mm³ | Median around 143/mm³ | No |
| Protein | Median around 4,100 mg/L | Median around 830 mg/L | No |
| Gram stain | Positive in roughly 60–80% if untreated | Negative | No |
Where the 3.5 mmol/L threshold does not simply apply
| Situation | What happens to the lactate | How to read it |
|---|---|---|
| Antibiotics before the lumbar puncture | Falls, sometimes into the reference interval, within hours | The most dangerous case. A low value is weak reassurance; lean on the cell count, differential and PCR |
| Post-neurosurgical, craniotomy, external ventricular drain | Baseline rises from blood, bone dust and the surgical inflammatory response | The threshold moves up — published postoperative cut-offs run from about 3.45 to 5.4 mmol/L. Use your unit’s own, and follow the trend |
| Seizure, cerebral ischaemia or infarction | Rises modestly through anaerobic metabolism in brain tissue | A raised value has an alternative explanation; it does not exclude infection |
| Systemic hyperlactataemia — shock, sepsis, metformin | Contributes only modestly; CSF lactate is largely independent of blood lactate | Less confounding than expected, but check the arterial lactate before attributing everything to the CSF |
| Tuberculous or cryptococcal meningitis | Often raised, sometimes markedly | A high lactate does not mean a pyogenic organism. The differential and the specific tests decide |
| Traumatic tap | Little affected — the test tolerates blood contamination better than the cell count does | One of its quiet advantages over the white cell count, which needs correcting |
The meningitis marker that does not need a blood tube
CSF lactate distinguishes bacterial from viral meningitis better than any other single number on the cerebrospinal fluid profile, including the CSF:serum glucose ratio that most of us were taught to reach for first. In Giulieri and colleagues’ prospective cohort of community-acquired meningitis the median lactate was 13 mmol/L in bacterial meningitis and 2.3 mmol/L in viral meningitis, and a threshold of 3.5 mmol/L separated the two groups completely, while the glucose ratio, the protein and the cell count all overlapped. The conventional cut-off is therefore 3.5 mmol/L, which is 31.5 mg/dL in the unit much of North America reports. It is not the only figure in circulation: Sakushima’s meta-analysis of 33 studies proposed 35 mg/dL as the optimal cut-off, which is 3.9 mmol/L (35 ÷ 9.008 = 3.885), not 3.5. The two numbers come from different kinds of evidence — one cohort in which the groups separated completely, and a pooled analysis across studies using a range of cut-offs — and a result between 3.5 and 3.9 mmol/L is above one and below the other.
Its second advantage is procedural rather than statistical, and it is the one that decides how often the test is actually usable. The glucose ratio needs a blood glucose drawn at the same time as the lumbar puncture. That sample is regularly forgotten in a busy department, drawn hours later, or drawn after dextrose has been started, and each of those quietly invalidates the ratio without anyone noticing. CSF lactate is one measurement on one tube. It also tolerates a traumatic tap far better than the white cell count does, so it survives the two commonest ways a CSF sample is spoiled.
Two situations change what a result means, and they run in opposite directions. Antibiotics given before the tap lower CSF lactate towards the reference interval within hours, at the same time as they sterilise the culture — so the two tests that might have disagreed are silenced together, and a reassuring lactate in a pre-treated patient is much weaker evidence than the same number in an untreated one. That is the failure mode to fear, because it is the one that argues for stopping treatment.
The other is post-neurosurgical meningitis, and here the usual warning is put too strongly. Lactate does not stop working after a craniotomy: in meta-analysis of postoperative neurosurgical patients it had a pooled sensitivity of 0.92 and a specificity of 0.88. What changes is the threshold. Blood, bone dust, devitalised tissue and the surgical inflammatory response raise the baseline, and the cut-offs used in those studies ranged from about 3.45 to 5.4 mmol/L, averaging 4.41. Applying 3.5 to a patient three days after a posterior fossa operation will label a great many uninfected people. Use the local postoperative threshold, prefer the trend across serial samples to any single value, and remember throughout that this number supports a clinician’s judgement about whether to treat — it does not replace it, and bacterial meningitis is treated on suspicion long before any result returns.
Frequently asked questions
What CSF lactate level indicates bacterial meningitis?
The threshold this page uses is 3.5 mmol/L, which is 31.5 mg/dL, from Giulieri and colleagues’ 2015 prospective cohort, in which it separated bacterial from viral meningitis completely. Sakushima’s 2011 meta-analysis proposed 35 mg/dL instead, which is 3.9 mmol/L, so a result between 3.5 and 3.9 mmol/L is positive on one source and not the other. In an untreated community-acquired presentation, a value at or above it supports a bacterial cause more strongly than any other single CSF parameter, and a value below about 2.4 mmol/L argues against one. Neither result is a diagnosis on its own.
Is CSF lactate better than the CSF to serum glucose ratio?
On the published comparisons, yes — it discriminates better, and it does not need a blood glucose drawn at the same time as the lumbar puncture. That paired sample is frequently missing, drawn late, or drawn after dextrose has been started, any of which invalidates the ratio silently. The lactate is one measurement on one tube.
Do antibiotics affect CSF lactate?
Yes, and this is the most important caveat. A dose given before the tap can bring the lactate down towards the reference interval within hours, while also sterilising the culture. A normal lactate in a patient who has already had antibiotics is therefore much weaker reassurance than the same value in an untreated patient.
Can CSF lactate be used after neurosurgery?
Yes, but not against the 3.5 mmol/L threshold. Blood, bone dust and the surgical inflammatory response raise the baseline, so published postoperative cut-offs run from about 3.45 to 5.4 mmol/L. Across that range, pooled sensitivity was 0.92 and specificity 0.88. Use your unit’s own postoperative threshold and follow serial values.
Does a traumatic tap invalidate the CSF lactate?
Much less than it invalidates the cell count. Blood contamination has only a modest effect on CSF lactate, which is one of the reasons it is useful when the tap has been difficult — the white cell count in the same sample may need correcting before it can be read at all.
Related calculators
References
- Giulieri SG, Chapuis-Taillard C, Jaton K, et al. CSF lactate for accurate diagnosis of community-acquired bacterial meningitis. Eur J Clin Microbiol Infect Dis. 2015;34(10):2049–2055.
- Sanaei Dashti A, Alizadeh S, Karimi A, Khalifeh M, Shoja SA. Diagnostic value of lactate, procalcitonin, ferritin, serum-C-reactive protein, and other biomarkers in bacterial and viral meningitis: a cross-sectional study. Medicine (Baltimore). 2017;96(35):e7637.
- Sakushima K, Hayashino Y, Kawaguchi T, Jackson JL, Fukuhara S. Diagnostic accuracy of cerebrospinal fluid lactate for differentiating bacterial meningitis from aseptic meningitis: a meta-analysis. J Infect. 2011;62(4):255–262.
- Xiao X, Zhang Y, Zhang L, Kang P, Ji N. The diagnostic value of cerebrospinal fluid lactate for post-neurosurgical bacterial meningitis: a meta-analysis. BMC Infect Dis. 2016;16:483.
- van de Beek D, Cabellos C, Dzupova O, et al. ESCMID guideline: diagnosis and treatment of acute bacterial meningitis. Clin Microbiol Infect. 2016;22(Suppl 3):S37–S62.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.