JOURNAL ARTICLE EDITOR'S CHOICE
Absence of Cerebrospinal Fluid Pleocytosis in Encephalitis
Ralph Habis, Anna Kolchinski, Ashley N Heck, et al
Clinical Infectious Diseases, Volume 81, Issue 1, 15 July 2025, Pages 179–187, https://doi.org/10.1093/cid/ciae391
Abstract
Background
Early diagnosis of encephalitis involves identifying signs of neuroinflammation, including cerebrospinal fluid (CSF) pleocytosis. However, an absence of CSF pleocytosis in encephalitis has been described, most notably in autoimmune encephalitis. We examined clinical characteristics and outcomes associated with the absence or presence of CSF white blood cell pleocytosis (≥5 cells/µL), to inform timely diagnosis and management of encephalitis.
Methods
This retrospective study compares initial CSF profiles in 597 adult patients with all-cause encephalitis.
Results
Of the 597 patients, 446 (74.7%) had CSF pleocytosis while 151 (25.3%) did not. CSF pleocytosis occurred more commonly in infectious cases (200/446, 44.8%), along with 59 (13.2%) autoimmune cases, comprised chiefly of anti–N-methyl-D-aspartate receptor (anti-NMDAR) encephalitis (37/59, 62.7%). Notably, the group without pleocytosis comprised similar proportions of infectious (47/151, 31.1%) and autoimmune (38/151, 25.92%; P > .05) encephalitis. Among those with infectious encephalitis, 47/247 (19%) had an absence of pleocytosis, including 18/76 (23.7%) with HSV-1 encephalitis. The absence of pleocytosis was associated with a decreased rate of acyclovir administration (47.7% in patients without pleocytosis vs 71.1% in patients with pleocytosis; P < .001). Despite pleocytosis being associated with some measures of clinical severity at admission such as a Full Outline of UnResponsiveness (FOUR) score ≤14, it was not associated with mortality or prolonged hospitalization.
Table 1. Comparison of Patients With All-Cause Encephalitis Without and With Cerebrospinal Fluid Pleocytosis
| All Patients (N = 597) | Absence of CSF Pleocytosis (n = 151/597; 25.3%) | CSF Pleocytosis (n = 446/597; 74.7%) | P a | |
|---|---|---|---|---|
| Gender | .52 | |||
| Male | 302/597 (50.6) | 73/151 (48.3) | 229/446 (51.3) | |
| Female | 295/597 (49.4) | 78/151 (51.7) | 217/446 (48.7) | |
| Mean (95% CI) age,b y | 49.25 (47.8–50.7) | 51.4 (48.5–54.4) | 48.5 (46.9–50.1) | .08 |
| Immunocompromisedc | 137/597 (22.9) | 29/151 (19.2) | 108/446 (24.2) | .21 |
| CCI ≥2 | 275/597 (46.2) | 82/150 (55.3) | 192/445 (43.1) | .01* |
| Fever | 315/597 (52.8) | 54/151 (35.8) | 261/446 (58.5) | <.001* |
| Focal neurological signsd | 228/597 (38.2) | 61/151 (40.4) | 167/446 (37.4) | .52 |
| New-onset seizure | 198/596 (33.2) | 60/151 (39.7) | 138/445 (31) | .05* |
| Headache | 275/597 (46.1) | 45/151 (29.8) | 230/446 (51.6) | <.001* |
| Psychiatric symptoms | 348/597 (58.3) | 99/151 (65.6) | 249/446 (55.8) | .04* |
| Memory deficits | 170/597 (30) | 69/151 (45.7) | 110/446 (24.7) | <.001* |
| CSF glucose <45 mg/dL | 120/597 (20.1) | 20/151 (13.2) | 100/446 (22.4) | .015* |
| CSF protein >50 mg/dL | 417/597 (69.8) | 78/151 (51.7) | 339/446 (76) | <.001* |
| CSF WBC, median (IQR), cells/mm3 | 29 (IQR: 5–165) | 2 (IQR: 1–3) | 62 (IQR: 20–255)e | <.001* |
| Acyclovirf | 389/597 (65.2) | 72/151 (47.7) | 317/446 (71.1) | <.001* |
| Etiology | ||||
| Infectious | 247/597 (41.4) | 47/151 (31.1) | 200/446 (44.8) | .003 |
| Autoimmune | 97/597 (16.2) | 38/151 (25.2) | 59/446 (13.2) | <.001* |
| Unknown | 253/597 (42.4) | 66/151 (43.7) | 187/446 (41.9) | .7 |
| Abnormal MRI | 289/487 (59.3) | 77/135 (57) | 212/352 (60.2) | .52 |
| Epileptiform activity on EEG | 84/368 (22.8) | 29/104 (27.9) | 55/264 (20.8) | .15 |
| Hyponatremia (<135 mEq/L) | 105/435 (24.1) | 27/115 (23.5) | 78/320 (24.4) | .85 |
| Thrombocytopenia (<150 × 103/μL) | 95/577 (16.5) | 18/140 (12.9) | 77/437 (17.6) | .19 |
| Leukocytosis (serum WBC >11 × 103/μL) | 187/577 (35.4) | 32/139 (23) | 155/438 (35.4) | .01* |
| SOFA >3 | 201/506 (39.7) | 38/121 (31.4) | 163/385 (42.3) | .03* |
| FOUR score ≤14 | 134/551 (24.3) | 20/132 (15.2) | 114/419 (27.2) | .005* |
| GCS ≤ 8 | 68/550 (12.4) | 12/129 (9.3) | 56/421 (13.3) | .23 |
| ICU admission | 239/586 (40.8) | 50/146 (34.2) | 189/440 (43) | .06 |
| Status epilepticus | 53/597 (8.9) | 16/151 (10.6) | 37/446 (8.3) | .39 |
| Mortality | 56/594 (9.4) | 16/151 (10.6) | 40/443 (9.0) | .36g |
| LOS,h median (IQR), d | 11 (IQR: 6–22) | 11 (IQR: 6–22) | 11 (IQR: 6–22) | .45i |
Abbreviations: CCI, Charlson comorbidity index; CI, confidence interval; CSF, cerebrospinal fluid; EEG, electroencephalography; FOUR, Full Outline of UnResponsiveness; GCS, Glasgow Coma Scale; HIV, human immunodeficiency virus; ICU, intensive care unit; IQR, interquartile range; LOS, length of stay; MRI, magnetic resonance imaging; SOFA, Sequential Organ Failure Assessment; WBC, white blood cell count.
aP value comparing the lack of CSF pleocytosis group with CSF pleocytosis group.
bAge range, 18–92 y.
cImmunocompromised is defined as HIV, recent chemotherapy (<1 m), solid-organ or bone marrow transplantation, receiving ≥20 mg of prednisone or equivalent for >1 m, or congenital immunodeficiency.
dSelf-reported symptoms suggestive of focal neurological manifestations such as acute-onset cranial nerve abnormalities or acute defects in sensorimotor abilities, including aphasia.
eMedian CSF WBC differentials in patients with pleocytosis. Neutrophils: 4% (IQR: 0–46%); lymphocytes: 79.5% (IQR: 32–92%); monocytes: 7% (IQR: 4–14%).
fProportion of patients who received empiric acyclovir.
gThe adjusted odds ratio for mortality was 1.35 (95% CI: .72–2.53); P = .36 after correcting for age, immunocompromised status, and type of encephalitis.
hLength of stay at the hospital.
iThe adjusted standardized β for LOS was −.32 (95% CI: −6.2 to 2.8); P = .45 after correcting for age, immunocompromised status, and type of encephalitis.
Table 2. Comparison of Patients With Infectious Encephalitis Without and With Cerebrospinal Fluid Pleocytosis
| All Patients (N = 247) | Absence of CSF Pleocytosis (n = 47/247; 19%) | CSF Pleocytosis (n = 200/247; 81%) | P a | |
|---|---|---|---|---|
| Gender | .77 | |||
| Male | 132/247 (53.4) | 26/47 (55.3) | 106/200 (53) | |
| Female | 115/247 (46.6) | 21/47 (44.7) | 94/200 (47) | |
| Mean (95% CI) age, y | 52.3 (50.3–54.3) | 50.7 (45.3–55.7) | 52.7 (50.5–54.9) | .46 |
| Immunocompromisedb | 80/247 (32.4) | 20/47 (42.6) | 60/200 (30) | .1 |
| CCI ≥2 | 136/247 (55.1) | 30/47 (63.8) | 106/200 (53) | .18 |
| Fever | 167/247 (67.6) | 27/47 (57.4) | 140/200 (70) | .1 |
| Focal neurological signsc | 91/247 (36.8) | 20/47 (42.6) | 71/200 (35.5) | .37 |
| New-onset seizure | 46/247 (18.6) | 15/47 (31.9) | 31/200 (15.5) | .01* |
| Headache | 138/247 (55.9) | 20/47 (42.6) | 118/200 (59) | .04* |
| Psychiatric symptoms | 115/247 (46.6) | 26/47 (55.3) | 89/200 (44.5) | .18 |
| Memory deficits | 38/247 (15.4) | 12/47 (25.5) | 26/200 (13) | .03* |
| CSF WBC, median (IQR), cells/mm3 | 96 (12–75) | 1.6 (1–3) | 172 (43–381)d | <.001* |
| CSF glucose <45 mg/dL | 66/247 (26.7) | 10/47 (21.3) | 56/200 (28) | .35 |
| CSF protein >50 mg/dL | 202/247 (81.8) | 27/47 (57.4) | 175/200 (87.5) | <.001* |
| Etiology | ||||
| Viral | 190/247 (76.9) | 41/47 (87.2) | 149/200 (74.5) | .06 |
| Bacterial | 38/247 (15.4) | 6/47 (12.7) | 32/200 (16) | .25 |
| Fungal | 16/247 (6.4) | 2/47 (4.3) | 14/200 (7) | .64 |
| Abnormal MRI | 125/177 (70.6) | 24/36 (66.7) | 101/141 (71.6) | .56 |
| Epileptiform activity on EEG | 25/118 (21.2) | 6/26 (23.1) | 19/92 (20.7) | .79 |
| Hyponatremia (<135 mEq/L) | 70/224 (31.3) | 14/43 (32.6) | 56/181 (30.9) | .84 |
| Thrombocytopenia (<150 × 103/μL) | 59/246 (24) | 13/47 (27.7) | 46/199 (23.1) | .51 |
| Leukocytosis (serum WBC >11 × 103/μL) | 75/246 (30.5) | 9/47 (19.1) | 66/199 (33.2) | .06 |
| SOFA >3 | 102/201 (50.7) | 20/36 (55.6) | 82/165 (49.7) | .52 |
| FOUR score ≤14 | 54/240 (22.5) | 8/44 (18.2) | 46/196 (23.5) | .45 |
| GCS ≤8 | 23/230 (9.6) | 7/44 (15.9) | 16/195 (8.2) | .12 |
| ICU admission | 89/246 (36.2) | 21/47 (44.7) | 68/199 (34.2) | .18 |
| Status epilepticus | 7/247 (2.8) | 3/47 (6.4) | 4/200 (2) | .1 |
| Mortality | 22/247 (8.9) | 6/47 (12.8) | 16/200 (8) | .39e |
| LOS,f median (IQR), d | 10 (5–20.5) | 12 (5–25) | 10 (5–19) | .053g |
Abbreviations: CCI, Charlson comorbidity index; CI, confidence interval; CSF, cerebrospinal fluid; EEG, electroencephalography; FOUR, Full Outline of UnResponsiveness; GCS, Glasgow Coma Scale; HIV, human immunodeficiency virus; ICU, intensive care unit; IQR, interquartile range; LOS, length of stay; MRI, magnetic resonance imaging; SOFA, Sequential Organ Failure Assessment; WBC, white blood cell count.
aP value comparing the lack of CSF pleocytosis group with the CSF pleocytosis group.
bImmunocompromised is defined as HIV, recent chemotherapy (<1 m), solid-organ or bone marrow transplantation, receiving ≥20 mg of prednisone or equivalent for >1 m, or congenital immunodeficiency.
cSelf-reported symptoms suggestive of focal neurological manifestations such as acute-onset cranial nerve abnormalities or acute defects in sensorimotor abilities, including aphasia.
dMedian CSF WBC differentials in patients with pleocytosis. Neutrophils: 7% (IQR: 1–54%); lymphocytes: 73.5% (IQR: 22.5–90%); monocytes: 9% (IQR: 4–16%).
eThe adjusted odds ratio for mortality was 1.57 (95% CI: .56–4.39); P = .39 after correcting for age, immunocompromised status, and type of encephalitis.
fLength of stay at the hospital.
gThe adjusted standardized β for LOS was −.12 (95% CI: −12.9 to .8); P = .053 after correcting for age, immunocompromised status, and type of encephalitis.
Table 3. Comparison of Patients With HSV-1 Encephalitis Without and With Cerebrospinal Fluid Pleocytosis
| All Patients (N = 76) | Absence of CSF Pleocytosis (n = 18/76; 23.7%) | CSF Pleocytosis (n = 58/76; 76.3%) | P a | |
|---|---|---|---|---|
| Gender | .59 | |||
| Male | 38/76 (50) | 10/18 (55.6) | 28/58 (48.3) | |
| Female | 38/76 (50) | 8/18 (44.4) | 28/58 (51.7) | |
| Mean (95% CI) age, y | 56.1 (52.5–59.7) | 60.7 (52.9–68.6) | 54.7 (50.6–58.8) | .08 |
| Immunocompromisedb | 17/76 (22.4) | 5/18 (27.8) | 12/58 (20.7) | .53 |
| CCI ≥ 2 | 42/76 (55.3) | 12/18 (66.7) | 30/58 (51.7) | .27 |
| Fever | 54/76 (71.1) | 8/18 (44.4) | 46/58 (79.3) | .004* |
| Focal neurological signsc | 36/76 (40.4) | 8/18 (44.4) | 25/58 (43.1) | .92 |
| New-onset seizure | 28/76 (36.8) | 8/18 (44.4) | 20/58 (34.5) | .44 |
| Headache | 44/76 (57.9) | 6/18 (33.3) | 38/58 (65.5) | .02* |
| Psychiatric symptoms | 40/76 (52.6) | 10/18 (55.6) | 30/58 (51.7) | .78 |
| Memory deficits | 12/76 (15.8) | 5/18 (27.8) | 7/58 (12.1) | .11 |
| CSF WBC, median (IQR), cells/mm3 | 76.5 (8.25–256.5) | 1 (1–2.3) | 166 (49–300.8)d | <.001* |
| CSF glucose <45 mg/dL | 10/76 (13.2) | 2/18 (11.1) | 8/58 (13.8) | .77 |
| CSF protein >50 mg/dL | 59/76 (77.6) | 11/18 (61.1) | 48/58 (82.8) | .05 |
| Abnormal MRI | 49/61 (80.3) | 9/15 (60) | 40/46 (87) | .02* |
| Epileptiform activity on EEG | 21/54 (38.9) | 5/13 (38.5) | 16/41 (39) | .97 |
| Hyponatremia (<135 mEq/L) | 25/75 (33.3) | 5/18 (27.8) | 20/57 (35.1) | .57 |
| Thrombocytopenia (<150 × 103/μL) | 17/75 (22.7) | 5/18 (27.8) | 12/57 (21.1) | .55 |
| Leukocytosis (serum WBC >11 × 103/μL) | 15/75 (20) | 3/18 (16.7) | 12/57 (21.1) | .68 |
| SOFA >3 | 23/66 (34.3) | 6/14 (42.5) | 17/52 (32.7) | .48 |
| FOUR score ≤14 | 10/72 (13.9) | 4/16 (25) | 6/56 (10.7) | .14 |
| GCS ≤8 | 9/73 (12.3) | 5/17 (29.4) | 4/56 (7.1) | .01* |
| ICU | 28/75 (37.3) | 9/18 (50) | 19/57 (33.3) | .2 |
| Status epilepticus | 4/76 (5.3) | 1/18 (5.6) | 3/58 (5.2) | 1 |
| Mortality | 4/76 (5.3) | 1/18 (5.6) | 3/58 (5.2) | .91e |
| LOS,f median (IQR), d | 10 (5.3–21.8) | 12 (7–22.5) | 8 (5–21.5) | .96g |
Abbreviations: CCI, Charlson comorbidity index; CI, confidence interval; CSF, cerebrospinal fluid; EEG, electroencephalography; FOUR, Full Outline of UnResponsiveness; GCS, Glasgow Coma Scale; HIV, human immunodeficiency virus; ICU, intensive care unit; IQR, interquartile range; LOS, length of stay; MRI, magnetic resonance imaging; SOFA, Sequential Organ Failure Assessment; WBC, white blood cell count.
aP value comparing the lack of CSF pleocytosis group with the CSF pleocytosis group.
bImmunocompromised is defined as HIV, recent chemotherapy (<1 m), solid-organ or bone marrow transplantation, receiving ≥20 mg of prednisone or equivalent for >1 m, or congenital immunodeficiency.
cSelf-reported symptoms suggestive of focal neurological manifestations such as acute-onset cranial nerve abnormalities or acute defects in sensorimotor abilities, including aphasia.
dMedian CSF WBC differentials in patients with pleocytosis. Neutrophils: 3% (IQR: 0–10%); lymphocytes: 84% (IQR: 62–94%); monocytes: 9.5% (IQR: 6–16%).
eThe adjusted odds ratio for mortality was .86 (95% CI: .08–9.6); P = .91 after correcting for age, immunocompromised status, and type of encephalitis.
fLength of stay at the hospital.
gThe adjusted standardized β for LOS was −.01 (95% CI: −7.8 to 7.4); P = .96 after correcting for age, immunocompromised status, and type of encephalitis.
Conclusions
CSF pleocytosis is an important criterion for encephalitis diagnosis, but 25.3% of patients with all-cause encephalitis and 23.7% of those with HSV-1 encephalitis exhibit an absence of pleocytosis on initial LP. Acyclovir initiation should not be delayed in the absence of pleocytosis in patients with suspected encephalitis.
