Case Report
Tuberculous meningitis is a difficult diagnosis to make. The presenting symptoms, signs, and laboratory values are often nonspecific and can be attributed to other underlying conditions. Imaging and cerebrospinal fluid analysis can be more helpful. Extra pulmonary tuberculosis is not uncommon, and tuberculous meningitis represents 5% of extrapulmonary cases in the United States with the risk for tuberculosis increased six times for foreign-born persons. We report the case of a foreign-born cirrhotic patient with tuberculous meningitis whose diagnosis was likely delayed because of signs and symptoms of her underlying hepatic disease.
Introduction
Mycobacterium tuberculosis meningitis (TBM) is a particularly devastating disease with a mortality that may exceed 50%.1 Of those who recover, up to 40% of patients will sustain persistent neurological deficits.2 Recovery is related to prompt recognition and initiation of appropriate therapy. In patients with comorbid conditions that may alter the level of consciousness, the diagnosis of TBM can be extremely challenging. We report the case of a patient with known cirrhosis that developed TBM that ultimately proved fatal, The patient's mental status changes were initially attributed to hepatic encephalopathy, and this may have contributed to a delay in diagnosis.
Case Report
A 59-year-old Asian woman with a past medical history significant for autoimmune hepatitis with cirrhosis presented with dyspnea. Ascites and a large right pleural effusion were discovered. There was concern that this finding represented decompensation of her underlying liver disease, and she was admitted for symptom management in conjunction with a liver transplant evaluation. The patient underwent repeated thoracentesis over a period of several months and ultimately required placement of a chest tube for recurrent pleural effusion. The fluid was transudative and yielded no growth on routine bacterial cultures. Three months after her initial presentation, she developed persistent fevers to 101 deg F, and routine cultures of blood, urine, and pleural fluid yielded no growth. One month later, a repeat urine culture yielded Escherichia coli and ampicillin-sulbactam therapy was instituted. However, her fever persisted despite sterilization of her urine. During this time, culture of chest tube drainage yielded coagulase-negative Staphylococcus. Ceftriaxone and vancomycin therapy was initiated, but no specific source for her continued fevers was identified. Four months after her initial presentation, the patient became socially withdrawn, and she talked extensively about death and suicide. Over time, there was excessive somnolence that was attributed initially to clinical depression and later to hepatic encephalopathy and metabolic anomalies. However, her fevers persisted to 101.8F despite continued antibiotic coverage, and this ultimately prompted an Infectious Diseases consultation.
The patient was originally from Thailand, and she immigrated to Pennsylvania in 1976. She lived in Japan from 1989 to 1992. She denied transfusions of blood products, risk factors for human immunodeficiency virus infection, alcohol, or tobacco use. She had a negative tuberculin tine test in 1992.
The patient's temperature was 101.1 deg F and she had a pulse of 75 beats per minute. Her blood pressure was 102/50 mm Hg, and her respiratory rate was 22 breaths per minute. She was frail, emaciated, and somnolent. She had surgical pupillary changes, no papilledema, and no scleral icterus. There was neck pain with passive flexion and extension. The cardiac examination was normal. Her breath sounds were decreased with dullness to percussion on the right, and a chest tube was in place. Her abdominal examination was notable for a nodular liver, and her spleen was palpable with minimal ascites. She had tracepitting edema bilaterally. She was unresponsive to verbal stimuli and had diminished corneal reflexes, but she responded to deep pain with grimace. She had markedly diminished deep tendon reflexes with an equivocal Babinski sign bilaterally.
The patient's laboratory studies were notable for hyponatremia and a serum ammonia level of 76 mmol/L.
An evaluation of her mental status was initiated in light of her deteriorating condition. Computed tomography of the head demonstrated a slight increase in the size of the third and lateral ventricles. Lumbar puncture was ultimately performed. This study revealed a cerebrospinal fluid (CSF) pleocytosis of 180/ mm' (5% segmented neutrophils, 3% monocytes, 92% lymphocytes), and red blood cell count of 295/mm3. The CSF protein was 3,018 mg/dL. The CSF glucose was 30 mg/dL with serum glucose of 157 mg/dL. There were no organisms identified on Gram stain or stains for fungi or acid-fast bacilli. Computed tomography of the chest (Fig. 1) demonstrated three pulmonary nodules within the right lung that had not been apparent on previous chest X-ray films. Based on these findings, an empiric four-drug regimen of isoniazid, rifampin, ethambutol, and pyrazinamide was instituted for presumed tuberculous meningitis. Tuberculin skin testing was planned. However, the patient continued to deteriorate and required transfer to the intensive care unit where endotracheal intubation and the initiation of pressors were required. In the acuity of the subsequent resuscitative efforts, the tuberculin skin test was never performed. The patient died shortly thereafter, approximately 4 months after her original presentation.
Autopsy findings included a fibrinous exudate at the base of the brain extending into both Sylvian fissures (Fig. 2). The lateral and third ventricles were slightly dilated. Histological sections of the brain revealed a meningoencephalitis with obliterative endarteritis and associated ischemic changes. The inflammatory infiltrate was composed of polymorphnuclear cells with variable numbers of lymphocytes, plasma cells, and histiocytes. No discrete granulomata were seen. Special stains demonstrated a few acid-fast bacilli in areas involved with the inflammatory infiltrate (Fig. 3). Examination of the lungs revealed a 1-cm nodule in the right middle lobe, which on microscopic sections showed necrotizing granulomata. Acid-fast stains of these sections were negative, but cultures of the lung nodule taken at autopsy were positive for M. tuberculosis. Other significant findings at autopsy included macronodular cirrhosis with changes of portal hypertension to include ascites, splenic congestion, and esophageal varices. Of note, approximately 3 weeks after the patient's death, the CSF cultures from the lumbar puncture yielded acid-fast organisms that were identified as M. tuberculosis.
Discussion
Approximately 4,000 cases of extrapulmonary tuberculosis occur each year in the United States with TBM accounting for approximately 5% of those cases.3 The prognosis of TBM is directly related to age, duration of symptoms, and the presence of neurological deficits.4 Mortality is particularly high for those over 50 years of age or those in whom illness has been present for over 2 months.1 A favorable prognosis requires an extremely high index of suspicion and the institution of appropriate therapy even before confirmatory diagnostic evidence is available.
Unfortunately, the clinical findings of TBM are relatively nonspecific and may easily be attributed to other conditions. Lowgrade fever, neurological deficits related to cranial nerve involvement, headache, malaise, decreased level of consciousness, and stiff neck are all commonly reported findings in TBM,5 but in patients with multiple comorbid conditions, any or all of these findings may have alternative explanations. Laboratory values may similarly be of little help. Patients may exhibit a mild anemia, leukocytosis, and decreased sodium as a result of inappropriate secretion of antidiuretic hormone,4 but these abnormalities are commonly seen in a variety of other clinical conditions, particularly in hospitalized patients, and by themselves they do not point to a specific diagnosis. The tuberculin skin test is positive in 80% of patients with TBM, however, a negative test occurs with sufficient frequency in all forms of active tuberculosis that it is of no help in excluding the diagnosis.4 Fifty percent of the patients with TBM have abnormalities on chest film,5,6 but in many instances there is no clinical or historical clue to suggest pulmonary tuberculosis.
The patient in this report had mental status changes that may have represented an early manifestation of TBM. However, because of the gradual onset and features consistent with clinical depression, these symptoms were not immediately evaluated further. Later, as she experienced increasing somnolence in the context of an elevated serum ammonia level, her symptoms were attributed to hepatic encephalopathy. In retrospect there were many features of this patient's presentation that placed her at increased risk for tuberculosis infection. The patient was originally from Thailand, and foreign-born individuals have been shown to have up to a sixfold increased rate of tuberculosis infection than U.S.-born persons.7 Immunosuppressed patients, particularly those with defects in cell-mediated immunity have been shown to be at particular risk for disseminated tuberculosis and therefore to meningeal disease.2 The immunosuppression associated with cirrhosis has not been well characterized, but there is evidence that patients with this condition may also be at increased risk for tuberculosis infection and that in such patients the prognosis may be particularly poor.8
The treatment of TBM is complicated by the variable penetration of antituberculous medications into the CSF. Depending on the drug susceptibility pattern of the isolate, finding an effective regimen that will penetrate the CSF can be quite challenging. Isoniazid achieves complete penetration of the CSF and its clinical efficacy in TBM is well established.9 Pyrazinamide also achieves complete CSF penetration.10 However, levels of rifampin in the CSF are generally only approximately 25% of serum levels, and some authors have recommended an increase in the dosing range of rifampin for patients with TBM. 10 Ethambutol is felt to achieve only very low levels in the CSF.11 Treatment of TBM should generally begin with a four-drug regimen that includes isoniazid, rifampin, and pyrazinamide unless the drug susceptibility of the isolate is already known. Once the susceptibility pattern of the isolate is confirmed, the regimen can be simplified to include only isoniazid and rifampin. If there is evidence of drug resistance, some authors recommend continuation of pyrazinamide in addition to two other agents for the duration of therapy because of excellent CSF penetration of this agent.10 A four-drug regimen to include ethambutol and pyrazinamide was initiated in this case to cover potentially resistant organisms in our patient. The duration of therapy for TBM will depend on the clinical course, but it is generally at least 12 months.2,10 More prolonged therapy may be required in the presence of drug resistance.
The addition of corticosteroids to the TBM treatment regimen is a controversial issue, but it has been shown to decrease mortality and improve the neurological outcome in some studies.12 Patients in whom corticosteroid regimens seem beneficial are those with moderate severity of disease. Studies have indicated both short- and long-term benefits, including increases in survival with the addition of corticosteroids to antituberculous regimens. Patients with more severe manifestations of TBM consistently have poor outcomes despite corticosteroid use.12
Conclusion
More than 15% of the cases of TBM remain undiagnosed until the death of the patient.13 To reduce the morbidity and mortality associated with this treatable condition, it is necessary for clinicians to have an extremely high index of suspicion for TBM, particularly in those with comorbid conditions who may have other risk factors for this infection. Once suspected, rapid implementation of therapy is required, even before the diagnosis is confirmed. Our case demonstrates the difficulties of identifying TBM in a patient with nonspecific findings of fever, mental status changes, and hyponatremia. Earlier imaging studies and CSF evaluation may have provided the information necessary for earlier diagnosis and initiation of therapy.
References
1. Haas DW: Mycobacterium tuberculosis. Mandell, Douglas, and Bennett's Principles and Practice of Infectious Diseases, Ed 5, pp 2576-606. Edited by Mandell GL, Bennett JE, Dolin R. Philadephia, PA. Churchill Livingstone, 2000.
2. Pottage JC, Harris AA: Chronic meningitis. Infectious Diseases, Ed 2, pp 141524. Edited by Gorbach SL, Bartlett JG, Blacklow NR. Philadephia, PA WB Saunders Company, 1998.
3. Reported Tuberculosis in the United States, 1999. Atlanta, GA, Centers for Disease Control and Prevention. 2000.
4. Leonard JM, Des Prez RM: Tuberculous meningitis. Infect Dis Clin North Am 1990: 4: 769-87.
5. American Thoracic Society: Diagnostic standards and classification of tuberculosis in adults and children. Am J Respir Crit Care Med 2000; 16: 1376-95.
6. Clark WC, Metcalf JC, Muhlbauer MS, Dohan FC, Robertson JH: Mycobacterium tuberculosis meningitis: a report of 12 cases and a literature review. Neurosurgery 1986; 18: 604-10.
7. Reported Tuberculosis in the United States. 1999. Atlanta, GA, Centers for Disease Control and Prevention. 2000.
8. Thulstrop AM, Molle 1, Svendsen N, Sorensen HT: Incidence and prognosis of tuberculosis in patients with cirrhosis of the liver: a Danish Nationwide Population Study. Epidemiol Infect 2000; 124: 221-5.
9. Kucers A, Crowe SM, Grayson ML, et al: The Use of Antibiotics: A Clinical Review of Antibacterial. Antifungal and Antiviral Drugs, p 1187. Oxford, Butteroworth-- Heinemann, 1997.
10. Iseman MD: A Clinician's Guide to Tuberculosis, pp 173-81. Philadelphia, PA, Lippincott Williams and Wilkins, 2000.
11. Kasik JE: Central nervous system tuberculosis. In Tuberculosis, Ed 3. Edited by Schlossberg D. New York, Springer-Verlag, 1994.
12. Kucers A, Crowe SM, Grayson ML, et at: The use of antibiotics: a clinical review of antibacterial. In Antifungal and Antiviral Drugs, p 1214. Oxford, Butteroworth-- Heinemann, 1997.
13. Dooley DP, Carpenter JL, Rademacher S: Adjunctive corticosteroid therapy for tuberculosis: a critical reappraisal of the literature. Clin Infect Dis 1997; 25: 872-7.
14. Rieder H, Bloch A, Snider D: Tuberculosis diagnosed at death in the United States. Chest 1991; 100: 678-81.
Guarantor: CPT Mark G. Carmichael, MC USA
Contributors: CPT Mark G. Carmichael, MC USA*; MAJ Jennifer C. Thompson, MC USAtt; MAJ Thomas B. Buttolph, MC USA; MAJ Jeffrey A. Hooke, MC USA
*Department of Medicine, William Beaumont Army Medical Center, 5005 North Piedras Street, El Paso, TX 79920; e-mail: mark.carmichael@amedd.army.mil.
^Infectious Disease Service, Walter Reed Army Medical Center, Washington, DC 20307.
^^ Current address: Infectious Diseases Service, William Beaumont Army Medical Center, 5005 North Piedras Street, El Paso, TX 79920: e-mail: jennifer.thompson@ amedd.army.mil.
(sec)Department of Pathology, Walter Reed Army Medical Center, Washington, DC 20307.
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense.
This manuscript was received for review in December 2001. The revised manuscript was accepted for publication in June 2002.
Reprint & Copyright by Association of Military Surgeons of U.S., 2003.
Copyright Association of Military Surgeons of the United States Mar 2003
Provided by ProQuest Information and Learning Company. All rights Reserved