Opportunistic Infections: Treatment

Case 4: Discussion - Disseminated MAC

The patient in this case had blood cultures that turned positive for Mycobacterium avium complex (MAC) 8 days after they were drawn, and the patient was started on clarithromycin (Biaxin) and ethambutol (Myambutol). The following discussion will focus on the clinical features, diagnosis, and treatment of disseminated MAC. The epidemiology, transmission, and primary prevention of MAC are discussed in Case 3 in the section Opportunistic Infections: Prophylaxis.

Clinical and Laboratory Features of Disseminated MAC

Mycobacterium avium complex organisms enter through the lung or gastrointestinal tract and can rapidly disseminate in persons with advanced immune suppression[1,2]. Patients at risk for MAC have a weak immune response to the organisms and experience minimal tissue destruction. Instead, MAC-related clinical manifestations result from the huge burden of organisms, which interfere with tissue function and alter cytokine production[1]. Patients with disseminated MAC typically have CD4 counts less than 50 cells/mm³ and present with non-specific clinical features, including fatigue, fever, fatigue, weight loss, diarrhea, and abdominal pain (Figure 1)[2]. Less frequently, patients develop diarrhea or symptoms of extrahepatic obstruction. Common abnormal laboratory studies include anemia, increased alkaline phosphatase (often with normal bilirubin, normal hepatic transaminases levels), and increased serum lactate dehydrogenase levels[1,2,3]. The anemia associated with disseminated MAC predominantly results from a failure in the maturation of red blood cell precursors, presumably mediated by a soluble factor in the serum that suppresses erythroid progenitor cells[4]. Other factors may play a role in the MAC-associated anemia. Abdominal CT scan abnormalities include multiple large retroperitoneal and mesenteric lymph nodes[5], hepatomegaly, splenomegaly, and thickened small bowel wall. Although disseminated MAC is rarely a direct cause of death, it is an independent predictor of increased mortality[3]. A normal abdominal CT scan does not rule out the diagnosis of disseminated MAC.

Diagnosis of Disseminated MAC

The diagnosis of disseminated MAC is usually made by isolating the organism from a normally sterile body site. Use of mycobacterial blood cultures (BACTEC(TM) or DuPont isolator system) to isolate MAC has become the preferred method of diagnosis, with most studies showing a greater than 90% yield with two blood cultures. With disseminated MAC, the laboratory will usually detect mycobacterial growth by day 14, at which point it can rapidly identify the species using DNA probes[6]. Although there are rare reports in which a MAC was diagnosed by culture of a bone marrow aspirate after blood culture failed to grow MAC, performing a bone marrow aspirate or biopsy is unnecessary in most cases. Stained bone marrow samples can rapidly reveal the presence of acid-fast bacilli, but this test is not specific for MAC. Although a positive culture from stool or respiratory system indicates an increased risk of developing disseminated infection[7], it is not diagnostic of disseminated infection.

Treatment for Disseminated MAC Infection

Most data on treatment of disseminated MAC has come from studies performed prior to the widespread use of HAART. The use of combinations that include either clarithromycin or azithromycin (Zithromax) has markedly improved response rates compared with therapies that did not include a macrolide. In an early study clarithromycin monotherapy sterilized blood in most patients, but relapse with clarithromycin-resistant MAC was common[8]. In a randomized trial that clearly established the importance of macrolides in the treatment of MAC, Canadian investigators treated 229 patients with disseminated MAC and found the 3-drug regimen of clarithromycin, ethambutol, and rifabutin (Mycobutin) led to more frequent and more rapid resolution of bacteremia than the 4-drug regimen of ethambutol, rifampin, clofazimine (Lamprene), and ciprofloxacin (Cipro)[9]. Several studies have shown good overall response rates with azithromycin-containing regimens[10,11]. In the best comparative trial of azithromycin and clarithromycin as part of combined therapy for MAC, there was a trend toward better response rates and fewer relapses with clarithromycin, but these differences were not statistically different (Figure 2). Most regimens for disseminated MAC have also included ethambutol after a clinical trial showed lower relapse relates with inclusion of ethambutol in a treatment regimen. In a trial of clarithromycin plus clofazimine, with or without ethambutol, patients who received ethambutol had much lower relapse rates (Figure 3)[12]. Whether rifabutin provides significant benefit remains unclear based on conflicting results from two trials[13,14]. In one trial, patients who received clarithromycin, ethambutol, and rifabutin had better survival than those who received clarithromycin and ethambutol. In the other trial, in which patients received clarithromycin and ethambutol with or without rifabutin, the use of rifabutin did not improve bacterial response or survival, but did decrease the risk of developing clarithromycin-resistant MAC[14]. Trials have established that clofazimine does not provide any benefit when added to a standard regimen of clarithromycin plus ethambutol, and may increase mortality[15,16]. Similarly, amikacin (Amikin) does not improve responses[17]. Taken together, these data support the use of clarithromycin plus ethambutol, with or without rifabutin (Figure 4). In patients intolerant to clarithromycin, azithromycin can be used in its place. Initial treatment regimens should not include either clofazimine or amikacin. There are no clear guidelines regarding the management of clarithromycin- or azithromycin-resistant MAC, and expert advice should be obtained in this situation.

Maintenance Therapy and Discontinuing Maintenance Therapy

Maintenance therapy to prevent recurrence of disseminated MAC consists of the same recommended regimen used to treat acute disseminated MAC, given at the same doses (Figure 4). Several studies have shown that patients on treatment for disseminated MAC who have a sustained response to HAART have a low risk of developing recurrent disseminated MAC if they discontinue maintenance therapy for MAC[18,19,20]. In a retrospective, multi-center Canadian study, 52 patients who had received effective therapy for disseminated MAC subsequently discontinued therapy following immunologic recovery with HAART[19]. At the time MAC therapy was stopped, most patients had a CD4 count greater than 100 cells/mm³ (median 230 cells/mm³), all had well-controlled HIV RNA on HAART, and all had received extensive therapy for MAC (median 32 months). At a median of 20 months after stopping therapy, only 1 of 52 patients had recurrence of MAC. In a prospective, non-randomized ACTG study, 48 patients with immunologic responses to HAART who had completed at least 12 months of macrolide-based therapy for disseminated MAC subsequently discontinued MAC therapy[20]. These patients were asymptomatic for MAC, had received at least 16 weeks of HAART, and had CD4 counts greater than 100 cells/mm³. At the time of discontinuation of MAC therapy, the median CD4 count was 240 cells/mm³. Overall, 47 of the 48 remained free of MAC, and 1 developed localized MAC osteomyelitis[20]. In a retrospective French study, 26 patients with a history of disseminated MAC discontinued maintenance therapy, and 4 of the 26 relapsed[21]. The relapse rate for MAC was higher in this study, but the median CD4 count at the time MAC therapy was lower (105 cells/mm³), and less than 50% of the patients had undetectable HIV RNA levels. These studies led to a revision in the USPHS/IDSA guidelines for preventing opportunistic infections, which now state that secondary prophylaxis to prevent recurrence of disseminated MAC may be discontinued in patients who have a sustained increase in CD4 count (greater than 6 months) to greater than 100 cells/mm³ in response to HAART if they have completed 12 months of MAC therapy and have no ongoing symptoms or signs attributable to MAC[22].

Management of Immune Reconstitution-Associated MAC

In recent years, several reports have described patients who developed localized MAC infection with negative blood cultures soon after starting HAART[23,24,25]. These patients previously had very low CD4 cell counts and these localized manifestations have included focal inflammatory lymphadenitis[23], osteomyelitis[24], pulmonary infection[25], and gastrointestinal disease. Biopsies typically demonstrate a marked localized inflammatory response with a paucity of organisms. Some of these cases likely represent an unmasking of a subclinical infection caused by an upregulation in immunity following antiretroviral therapy. In the setting of symptomatic localized disease, clinicians have used corticosteroids along with standard MAC therapy, but rapid tapering of the corticosteroids has proven difficult. The effectiveness and optimal dosing of corticosteroids in this setting has not clearly been established. In addition to using combination antimicrobial therapy for MAC, it would be reasonable to use prednisone at a dose of 40-60 mg per day, continue for a 4-6 week period, and gradually taper the dose as tolerated during the final 2 weeks. Recurrence of symptoms should prompt an increase in dose followed by a more gradual taper.