Neurosurgical Management of HIV-Related Focal Brain Lesions

Published on 13/03/2015 by admin

Last modified 13/03/2015

Print this page

This article have been viewed 2356 times

Chapter 146 Neurosurgical Management of HIV-Related Focal Brain Lesions

Human immunodeficiency virus (HIV) infection causes a disease of the immune system known as acquired immunodeficiency syndrome (AIDS).^1,² Although the condition was initially described in 1981 by the manifestations of opportunistic infections in homosexual men, the term AIDS was not coined by the Centers for Disease Control and Prevention (CDC) until September 1982.^3,⁴

HIV infection progressively reduces the effectiveness of the immune system and leaves patients susceptible to opportunistic infections and tumors. HIV is transmitted through direct contact of a mucous membrane or the bloodstream with a bodily fluid containing HIV, such as blood, semen, vaginal fluid, or breast milk. This transmission can involve anal, vaginal, or oral sex; blood transfusion; contaminated hypodermic needles; exchange between mother and baby during pregnancy, childbirth, or breastfeeding; or other exposure to one of the above bodily fluids. AIDS is now a pandemic with more than 33 million people infected worldwide.^5,⁶

Epidemiology

Central nervous system (CNS) symptoms occur in 30% to 50% of AIDS patients, and as many as 90% have evidence of CNS disease at autopsy.⁷^–¹⁰ Approximately 10% of AIDS patients have focal CNS lesions that might serve as a target for a stereotactic biopsy.^8,¹¹ One third of these patients have not had prior manifestations of AIDS when they are referred for biopsy.

Intracranial disease is second only to pulmonary disease as a cause of morbidity and mortality in patients with AIDS.^12,¹³ Of the approximately 10% of patients with focal lesions in the brain, 30% to 70% have Toxoplasma gondii abscesses, and another 15% to 30% have primary or metastatic CNS lymphoma.^8,^11,^12,¹⁴^–¹⁶ Although the incidence of toxoplasma abscesses has decreased in some studies with effective antiretroviral and antitoxoplasma therapy, the incidence of CNS lymphoma appears unchanged.¹⁷ The remaining 20% include a myriad of bacterial (e.g., Listeria, Escherichia coli), fungal (e.g., Cryptococcus, Aspergillus), and viral (e.g., HIV, herpes, cytomegalovirus [CMV]) processes that can manifest as focal lesions on computed tomography (CT) or magnetic resonance imaging (MRI) of the brain.^8,^11,¹⁸ Even the often diffuse process of progressive multifocal leukoencephalopathy (PML) commonly presents as a focal lesion.^15,¹⁸ Moreover, HIV-positive patients are also apparently susceptible to the same diseases as the HIV-negative population and, therefore, can have the usual primary and secondary brain tumors,^8,^10,¹⁵ as well as the vascular (hemorrhage and infarction) lesions that occur in the brains of AIDS patients and the 4% to 20% of biopsies from which no definite diagnosis can be made. A comprehensive list of causes of focal brain lesions in HIV-positive patients is provided in Table 146-1. Furthermore, a few patients with AIDS have several CNS lesions of more than one etiology.¹¹

TABLE 146-1 Etiologies of Focal Cerebral Lesions in HIV-Positive Patients

Infectious

Parasites

Toxoplasma gondii

Cysticercosis

Fungi

Cryptococcus

Candida

Histoplasma

Coccidiodies

Aspergillus

Mucor

Escherichia coli

Bacteria

Mycobacteria

Listeria

Nocardia

Salmonella

Spirochetes

Viruses

Cytomegalovirus

Herpesviruses

Varicella zoster

Papovavirus

Noninfectious

Neoplastic

Lymphoma (primary and secondary)

Kaposi’s sarcoma

Glioma

Metastatic carcinoma

Other^∗

Vascular

Infarcts

Hemorrhage

Nondiagnostic

∗ There is no reason to believe that HIV-positive patients are not susceptible to the same neoplastic brain diseases as the rest of the population.

Modified from Levy RM, Bresdesen DE, Rosenblum ML: Neurological manifestations of the acquired immunodeficiency syndrome (AIDS): Experience at UCSF and review of the literature. J Neurosurg 62:475-95, 1985.

Patient Considerations

Despite incubation periods of up to, and possibly greater than, 10 years, as well as recent advances in treatment, AIDS remains an imminently fatal disease.¹⁹^–²² The CNS has been identified as one of the reservoirs for HIV that makes its eradication from the body elusive, despite highly effective therapies.^23,²⁴ Nonetheless, potentially effective therapies exist for approximately 90% of the identifiable causes of CNS disease in patients with AIDS.^8,¹³ This fact suggests the usefulness of brain biopsy in the diagnosis and treatment of AIDS patients with focal brain disease. Because stereotactic biopsy is diagnostic in up to 96% of AIDS cases,^15,¹⁶ and because almost all treatable AIDS-related focal brain lesions do not require resection, there is virtually no need for an open procedure, except in rare cases of impending herniation due to mass effect. Furthermore, stereotactic biopsy is a low-risk procedure,^15,^16,¹⁸ even in severely medically compromised patients (e.g., many of those with AIDS), because it can usually be performed with the patient under local anesthesia.¹ A need for definitive diagnosis and low risk to the patient are persuasive in the decision to undertake an invasive diagnostic procedure. On the other hand, the communicable and fatal nature of HIV greatly compounds this issue and raises the questions of risk to those performing the procedure and whether or not accurate diagnosis and treatment have an effect on outcome. Questions such as these have inherent ethical overtones, but some, and possibly tenuous, scientific data on these issues are available to help in the decision-making process. Fortunately, the reduced incidence of opportunistic infections as a result of modern antiviral strategies has limited the surgeon’s exposure to such complex considerations.^7,²⁵

Risk to Surgeons Operating on HIV-Positive Patients

Solid data regarding the risk to surgeons who operate on patients who are HIV positive is scant and difficult to obtain. First, there is no way to be absolutely certain what other risk factors a given surgeon who contracts the virus might have and no way to be certain what operating mishaps, if any, can lead to transmission of the virus. In addition, more controversial issues, such as privacy, have an effect on the reporting of such data. Notwithstanding these issues, a few relatively scientific attempts at quantifying the risk of seroconversion (becoming HIV positive) in surgeons have been reported.^26,²⁷ Schiff, for example, formulated a mathematical model for estimating a given surgeon’s cumulative lifetime risk.²⁶ The model accounts for seroprevalence in the patient population; the inoculation rate, roughly based on surgical subspecialty; and the estimated probability of infection for a given procedural mishap. One survey estimated a greater than 6% 30-year risk in the 10% of surgeons considered at highest risk.²⁷ Even if estimates of a surgeon’s risk of contracting HIV are of questionable accuracy, existing evidence raises the possibility of some risk, and any risk of contracting a fatal disease must be considered.

Preoperative Evaluation

Radiographic Studies

Since the introduction of highly active antiretroviral therapy (HAART) in the mid 1990s, the morbidity and mortality associated with HIV-1 infection has significantly decreased and AIDS has become a chronic disorder.²⁸ Accurate diagnosis and treatment of brain lesions in HIV-infected patients affects outcome, and given even the slightest risk of disease transmission during an invasive procedure, the question arises whether the diagnosis can be made noninvasively. Although radiologists have reported an ability to make some diagnoses confidently on routine neuroimaging studies (CT and MRI), the consensus has been that the definitive diagnosis must be made histologically.⁷ Dina, for example, found that if a mass in the brain of an HIV-positive patient enhanced, was unifocal, exhibited subependymal spread on CT or MRI, and was hyperdense on CT without contrast, it was never toxoplasmosis in his series of patients.²⁹ However, he still recommended that these lesions undergo biopsy.²⁹

Since then, other investigators have delineated a wider spectrum of CT and MRI findings for lymphoma, cautioning against confident radiographic diagnoses.³⁰ Some authors have suggested that metabolic imaging (positron-emission tomography [PET] and single photon emission computed tomography [SPECT]) make it possible to distinguish between malignant and nonmalignant lesions.^31,³² In fact, thallium-201 SPECT is more than 90% accurate in distinguishing between tumor (especially lymphomas) and infection,³² although its value in the presence of HAART appears reduced.³³ Despite eloquent attempts to characterize the various brain lesions seen on neuroimaging studies in HIV-positive patients, no constellation of findings is 100% diagnostic of any of the numerous disease entities.^34,³⁵ Moreover, the possibility of numerous etiologies for several brain lesions cannot be ignored in these patients (Fig. 146-1).

Figure 146-1 Contrast-enhanced computed tomographic scan of a 24-year-old man with AIDS. A, These lesions (arrows) improved and then stabilized after 1 year of anti-Toxoplasma therapy. B, After his symptoms recurred 1 year later, the patient had this scan, which revealed a new enhancing mass on the left. Biopsies of these lesions determined that those on the right side were due to toxoplasmosis, as expected, and the lesions on the left side were determined to be lymphoma.

Empirical Therapy

At best, neuroimaging is suggestive in the diagnosis of AIDS-related brain lesions. This does not mean, however, that one should proceed directly to brain biopsy. As noted earlier, most focal brain lesions in HIV-positive patients are T. gondii abscesses, which generally resolve when the patient is treated with adequate doses of sulfadiazine (or clindamycin) and pyrimethamine (anti-Toxoplasma therapy).^36,³⁷ Rosenblum and associates thus proposed that stable HIV-positive patients with several lesions on MRI receive 2 to 3 weeks of empirical anti-Toxoplasma therapy.³⁸ If follow-up images and physical examination fail to show improvement, these authors then recommend biopsy. This regimen appears prudent, and although up to 40% of patients cannot tolerate the therapy, leading to inadequate treatment and often to biopsy, the incidence of toxoplasmosis in biopsy series is decreasing.^15,^16,¹⁸ The algorithm should probably be extended to include any HIV-positive patient with one or more contrast-enhancing lesions on CT, because cerebral toxoplasmosis is not always multifocal.¹⁵ In addition, deterioration in a patient who is on anti-Toxoplasma medications might warrant early biopsy.

In short, a presumptive diagnosis of cerebral toxoplasmosis can be given confidently to an HIV-positive patient whose brain lesions resolve after administration of anti-Toxoplasma drugs. This regimen eliminates the need for biopsy in a population of patients who can tolerate the therapy and have no focal cerebral lesions of another etiology. However, radiographic evidence of improvement can be complicated by the concomitant administration of steroids, because some lesions improve with the administration of dexamethasone or prednisone—lymphoma, in particular.

Laboratory Evaluation

Rosenblum and colleagues ³⁸ also suggested that in the absence of mass effect, cerebrospinal fluid (CSF) should be evaluated in HIV-positive patients with focal brain lesions. However, sensitivity and specificity for most causes of HIV-related focal lesions is less than optimal.^37,³⁹^–⁴² Only a few patients with lymphoma have abnormal CSF cytology, for example.^39,⁴¹ However, Epstein-Barr virus (EBV) may be a marker in the CSF for primary CNS lymphoma (PCNSL). The assay for EBV in the CSF requires gene amplification (polymerase chain reaction [PCR]).^43,⁴⁴ Furthermore, although the CSF is abnormal in toxoplasmosis patients, the findings are nonspecific.³⁷ In serologic tests (whether on the blood or the CSF), as expected, antibody titers are unreliable in immunocompromised patients.

Polymerase Chain Reaction

In recent years, the analysis of CSF in patients with HIV disease has been revolutionized by the advent of the PCR technique, reducing the need for brain biopsy.^25,⁴⁵ Even minute amounts of DNA from various viral and bacterial particles can be detected in the CSF using PCR. JC virus (JCV, associated with PML), herpes simplex virus (HSV), varicella zoster virus (VZV), CMV, EBV (associated with PCNSL), T. gondii, and Mycobacterium tuberculosis have been detected in the CSF of HIV-positive patients. Threshold levels for detection of many of these antigens have now been defined.⁴⁶ Furthermore, the technique may be used in certain cases to monitor the efficacy of treatment or the progression of disease.^47,⁴⁸

Antinori and associates analyzed the CSF for EBV, JCV, and T. gondii by PCR in 66 of 136 HIV patients with CNS lesions. Patients in whom anti-Toxoplasma therapy had failed also underwent stereotactic biopsy. These investigators found that if EBV DNA or T. gondii DNA tests were positive, the probability of PCNSL or toxoplasmic encephalitis, respectively, exceeded 0.96.⁴⁹ On the other hand, the absence of T. gondii DNA positivity in the CNS does not rule out the diagnosis of toxoplasmic encephalitis.^49,⁵⁰ The same is true of bacteriologic examination of the CSF for tuberculosis.⁵¹ Others report less sensitivity.⁵² More recently, Portolani and colleagues reported data that cast doubt on the role of EBV in HIV-related CNS disease.⁵³ Nonetheless, lumbar puncture can provide enough diagnostic certainty to treat a patient without a histologic evaluation of tissue obtained by biopsy.

Surgery

If unable to make a confident diagnosis by noninvasive or minimally invasive means for an HIV-positive patient who has a focal cerebral lesion that has not responded to anti-Toxoplasma therapy, a biopsy must be considered. Such a measure is taken if the clinician and the patient and ultimately the surgeon agree that the risk of achieving a definitive diagnosis is outweighed by the potential for improving outcome. Chappell and colleagues reported their experience in the management of HIV-related focal brain lesions in an attempt to define the role of brain biopsy ¹⁵ and developed an algorithm that minimizes unnecessary risk to patients and surgeons. Decision making and the application of such an algorithm can be further refined by the use of PCR analysis of the CSF.

Surgical Technique

CT-guided stereotaxis was used for all patients in the series. Single- or double-dose administration of intravenous contrast material with and without delayed image acquisition was used. De La Paz and Enzmann suggested that a double-dose delayed-contrast CT provides the best results,⁵⁴ but MRI is still more sensitive and possibly more specific in the differential diagnosis of HIV-related brain lesions.^33,^35,^54,⁵⁵ If a compatible stereotactic system is available, MRI guidance is a suitable alternative to CT. The efficacy of an ultrasound-guided technique merits its inclusion as a suitable alternative as well.⁴² However, this procedure involves a more-involved surgical opening and thus possibly an increased risk to the surgeon and the patient. The availability of frameless stereotactic systems allows the minimal invasion of a twist-drill or bur hole procedure without the discomfort of a headframe.

In most cases, intravenous sedation with monitored anesthetic care provides adequate anesthesia. General anesthesia is used at the patient’s request, if it can be tolerated. The incidence of operative complications is low, but it may be prudent to monitor patients overnight in an intensive care setting.

Buy Membership for Neurosurgery Category to continue reading. Learn more here