Ventricular and lumbar cerebrospinal fluid analysis in 77 HIV-negative patients with Cryptococcal meningitis who received a ventriculoperitoneal shunt


A total of 100 patients with CM were admitted to the Department of Neurosurgery of Lingnan Hospital, Branch of the Third Affiliated Hospital of Sun Yat-sen University, from January 2016 to August 2020. Only 77 patients with CM were retrospectively analyzed. The inclusion criterion was the definite diagnosis of CM. Exclusion criteria included the following: (1) Patients without VPS placement; (2) Patients who had received a shunt prior to admission to our hospital; (3) Patients who had VPS placement by other teams of neurosurgeons and patients without lumbar CSF test within 48 h before VPS (Fig. 1).

Figure 1
figure 1

Research flow. A total of 100 patients with CM were admitted in neurosurgery department between January 2016 to August 2020. 23 patients were excluded. The number of patients who had CSF analysis at the indicated time was shown. The last lumbar CSF test before VPS was done within 48 h before surgery.

Diagnosis of CM

A definite diagnosis of CM required at least one of the following criteria17,24: (1) Positive India ink staining of CSF with centrifuged sediment for Cryptococcus; (2) Positive culture of Cryptococcus from CSF; (3) Compatible histopathology (5- to 10-μm encapsulated yeast observed in brain tissue); (4) Probable CM was considered in patients with clinical symptoms of meningitis and a positive test of Cryptococcal antigen, and/or positive CSF findings by metagenomic next-generation sequencing. All CM patients in this study had a definite diagnosis of CM. Tuberculosis was ruled out by T-SPOT, acid-fast bacilli (AFB) smear, and ELISA.


Antifungal treatment before VPS placement was different based on the attending physician’s preference and the patient’s tolerance of the drug. Therapeutic options are limited to amphotericin B, flucytosine, fluconazole, and voriconazole in our hospital. A relatively unified antifungal regimen of flucytosine plus fluconazole was used after VPS except for five patients (Table 1). Corticosteroids were administered to patients with immune reconstitution inflammatory syndrome. Patients with increased CSF pressure were treated with intravenous mannitol and underwent a repeated lumbar puncture before VPS placement. A VPS was considered when patients suffered from neurological deterioration caused by intractable elevated intracranial pressure and/or progressive ventricular enlargement. No patients in this study received EVD, as this option has not been usually considered at our department since 2016 due to the high frequency of secondary infection. Lumbar puncture was used to dynamically monitor CSF changes after VPS placement.

Table 1 Demographic and clinical features of patients.

VPS placement

The tip of the ventricular catheter was placed via a frontal approach into the foramen of Monro or the third ventricle by means of the anterior horn of the ipsilateral lateral ventricle, based on a prior definition of accurate ventricular catheter placement25. The non-dominant hemisphere side was the preferred side of placement. Accurate placement, which is very important to ensure all the drainage holes are in the ventricle, is essential for patients with CM without ventriculomegaly. Two types of adjustable shunt devices were used, the proGAV2.0 valve (Miethke) and the Strata II valve (Medtronic). Usually, the initial valve pressure was set to 200 mm H2O, or gear 2.5, respectively. The valve pressure was adjusted according to patient’s clinical response. The valve pressure would be adjusted down when a patient exhibited persistent neurological disorder, progressive ventricular enlargement, and a high opening pressure on lumbar puncture. The valve pressure would be adjusted up when a patient exhibits orthostatic headache and a low opening pressure on lumbar puncture. Prophylactic anti-epileptic drugs were usually administered peri-operatively. Shunt complications were recorded within six months after VPS. Three patients who had a shunt obstruction were successfully revised. Two patients were adjusted the depth of the tube insertion; one patient was revised by the replacement of a new shunt tube.

Acquisition and detection of CSF parameters

We recorded the CSF testing results, which met the following requirements: Lumbar CSF samples were obtained by lumbar puncture within 48 h before VPS placement, and 1–3 days (stage 1), 5–7 days (stage 2), 14 ± 2 days (stage 3), 30 ± 5 days (stage 4), 60 ± 7 days (stage 5), 90 ± 7 days (stage 6), and 180 ± 14 days (stage 7) after VPS placement. Ventricular CSF was obtained during shunt insertion directly from the lateral ventricle (Fig. 1). CSF analysis included white blood cell (WBC) count and red blood cells (RBCs) count, total protein concentration, CSF/blood glucose ratio, chloride ion level, and Cryptococcal count. When visible blood contamination occurred, one leukocyte was subtracted per 1000–1500 red blood cells to estimate the true CSF WBC count26. Cryptococcal count was determined by counting the number of Cryptococcus per milliliter of CSF via India ink staining.

Definition of ventriculomegaly

Ventriculomegaly was diagnosed on the basis of dilation of the temporal horn of the lateral ventricle and/or an Evans’ index of > 0.3, as determined by CT and/or MRI. Evans’ index refers to the ratio of the ventricular width of the bilateral frontal horn to the maximum biparietal diameter22.

Six-month neurological outcome

All patients in the study exhibited neurological symptoms before VPS placement, such as headache, visual changes, hearing impairments, disturbance of consciousness, muscle weakness, seizures, cognitive impairment and urinary incontinence. Neurological outcomes, which were evaluated at six months after VPS placement, were divided into four levels: cured, better, worse, and death.

Statistical analysis

All statistical analysis was performed using SPSS version 22 software (SPPS Inc., Chicago, IL, USA). All numerical variables of CSF parameters were presented as median and range or interquartile range as they were non-normally distributed data. Categorical variables were expressed as counts and percentages. A Wilcoxon signed-rank test was used to compare lumbar and ventricular CSF parameters and analyze lumbar CSF change after shunting surgery. The Mann–Whitney U test were used to analyze CSF parameters in patients with and without ventriculomegaly, and patients with and without surgical complications. Chi-squared test or Fisher’s exact test was used for comparisons of binary categorical variables, as appropriate. Statistical significance was set at a value of p < 0.05.

Ethics approval and consent to participate

The Clinical Research Ethical Committee of the Third Affiliated Hospital of Sun Yat-sen University approved the study (02-167-01) and waived the requirement of written informed consent. No experiments were conducted with human participants (or their tissue). All information was kept anonymous. The study complies with the Declaration of Helsinki.

Source link

Back to top button