Sir—I read with interest the recent article by Dr. Davis and colleagues [1] regarding decreased CSF glucose levels in patients with herpes simplex virus (HSV) encephalitis. The authors' patients, who had HSV-1 encephalitis, also had decreased CSF glucose levels. In my experience, decreased CSF glucose levels in HSV-1 encephalitis are neither uncommon nor universal. The authors state that low CSF glucose levels are predictive of CNS bacterial infections because CNS viral infections are associated with normal CSF glucose levels [1]. Of course, other nonbacterial CNS infections are associated with low CSF glucose levels: for example, fungal infections, tuberculosis, amebic infections, brain abscess with ventricular leak, and parameningeal infections. The report by Davis et al. [1] should prompt clinicians to review the differential diagnostic possibilities of decreased CSF glucose levels in CNS viral infections. Nonviral infections that may present with mental confusion—for example, mycoplasma meningoencephalitis, neuroborreliosis, legionnaires' disease, rickettsial infection, and encephalopathy due to noninfectious disorders, such as hepatic encephalopathy and drug-induced meningitis/encephalitis, among others—are not associated with decreased CSF glucose levels [2–4]. Among the viral infections that present as aseptic meningitis, meningoencephalitis, or encephalitis, a decreased CSF glucose level may occur in CNS infections due to mumps virus, enteroviruses, and lymphocytic choriomeningitis virus, in addition to HSV-1 and viral infections presenting as aseptic meningitis, meningoencephalitis, or encephalitis. A decreased CSF glucose level is not a feature of arboviral encephalitis, including West Nile encephalitis [3]. CSF lactic acid levels are a useful way to determine the etiology of decreased CSF glucose levels by differentiating bacterial versus viral CNS infections [4–8]. In patients with meningitis/encephalitis, CSF lactic acid levels are an underused modality with diagnostic significance. CSF lactic acid concentrations can be determined within minutes, whereas results of PCR for HSV-1 usually take several days to be reported. Because early empiric treatment for bacterial infections and HSV-1 encephalitis is different, an early presumptive diagnosis is crucial, pending definitive HSV-1 PCR test results. The main diagnostic use of CSF lactic acid levels is rapid differentiation of viral meningitis/encephalitis (characterized by levels of ⩽3 mmol/L) and partially treated viral meningitis (with levels of 3–6 mmol/L) from bacterial CNS infections (with levels of ⩾6 mmol/L). For this purpose, determination of CSF lactic acid levels is without equal for rapid and inexpensive differentiation between these clinical entities [2, 3]. If CSF glucose levels are decreased, and fungal and tubercular meningitis are excluded as possibilities, the differential diagnosis is often between HSV-1 encephalitis and bacterial CNS infection. RBCs in the CSF not due to trauma and/or hemorrhage should suggest the possibility of HSV-1 encephalitis. RBCs in the CSF are not part of the clinical picture of bacterial CNS infection. Obtaining an electroencephalogram (EEG) is useful early in the diagnosis of HSV-1 encephalitis. In HSV-1 encephalitis, the EEG characteristically shows a localized temporoparietal focus, in contrast to other causes of viral encephalitis that demonstrate bilateral diffuse hemispheric EEG abnormalities. Levels of CSF lactic acid may be used to differentiate CNS infections with decreased CSF glucose levels from HSV-1 encephalitis. CSF lactic acid levels are highly elevated in bacterial meningitis ( >6 mmol/L) but are not elevated in HSV-1 encephalitis (<3 mmol/L) or other CNS viral infections. The number of RBCs in the CSF in HSV-1 encephalitis is directly proportional to the degree of parenchymal damage/hemorrhage. Metabolism of RBCs in the CSF produces lactic acid, which may increase lactic acid levels slightly. However, CSF lactic acid levels in bacterial CNS infections are highly elevated (6–20 mmol/L), even in the presence of RBCs in the CSF, and should not be confused with the minimal elevations seen in HSV-1 encephalitis with RBCs in the CSF (∼2–4 mmol/L) [2, 3, 9]. Viral CNS infections that may present with low CSF glucose levels include mumps, enteroviral infections, lymphocytic choriomeningitis infection, and HSV infection [9, 10]. Low CSF glucose levels may be present with a variety of noninfectious disorders, including, for example, sarcoid meningitis, carcinomatous meningitis, and posterior fossa syndrome [3]. It has been said that CSF lactic acid levels do not offer a diagnostic advantage over CSF glucose levels for the differential diagnosis of CNS disorders [10]. In patients with CNS infection and decreased CSF glucose levels, if the differential diagnosis is between viral (e.g., HSV-1) and bacterial infection, CSF lactic acid levels will clearly differentiate between them, whereas CSF glucose levels will not [9]. Conflict of interest. B.A.C.: No conflict.
No takes yet. Share an insight, caveat, or question.
B. A. Cunha (2004) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: