The Changing Nature of the Prevention and Management of Opportunistic Infections

Perspective - Opportunistic Infections Volume 10 Issue 5 November/December 2002

Perspective The Changing Nature of the Prevention and Management of Opportunistic Infections

As related by Henry Masur, MD, at the A second explanation for the oping PCP within 6 months of a medical International AIDS Society–USA CME absence of continued decline in oppor- visit, one quarter had a CD4+ cell count course in Washington, DC, in May 2002, tunistic infection rates is the continued greater than 200/µL. Thus clinicians opportunistic infection rates have spread and progression of HIV disease should keep in mind that the “thresh- declined since the advent of potent in patient groups that are not receiving olds” published for opportunistic infec- antiretroviral therapy. However, there adequate medical care. In some sites in tions are general guidelines, not abso- is evidence to indicate that these Washington, DC, for example, many lute boundaries of susceptibility. Both declines have stalled and perhaps patients are now presenting with oppor- prior to the era of potent antiretroviral reversed in some populations in associ- tunistic infections as the initial manifes- therapy and currently, opportunistic ation with long-term antiretroviral fail- tation of HIV disease. Adult/Adolescent infections will occasionally occur at ure and inadequate access to care. Spectrum of HIV Disease Project data on unusually high CD4+ cell counts. 2365 patients developing Pneumocystis There are data to indicate that the The use of chemoprophylaxis for oppor- carinii pneumonia (PCP) from 1996 CD4+ cell count nadir reached prior to tunistic infections and the advent and through 1999 indicate that 45% were not restoration of CD4+ counts to greater widespread use of potent antiretroviral in care at the time they developed PCP, than 200 cells/µL is not predictive of therapy were associated with a marked compared with 41% of 3863 patients increased risk for opportunistic infec- decline in rates of most opportunistic who developed PCP from 1993 through tions. As shown in a study by Miller and infections in HIV-infected individuals 1996. The proportion of patients devel- from the early 1990s through 1997. oping PCP who were in care and met the However, a recent report from the criteria for PCP prophylaxis but received Adult/Adolescent Spectrum of Disease no prophylaxis was 6% in the period Many patients are now Project indicates that additional reduc- from 1993 to 1996, and increased to 14% tions in rates for many opportunistic in 1996 to 1999. A large proportion of presenting with infections were not observed between patients developing PCP were in care 1997 and 1999 (McNaghten et al, 39th and met criteria for and received PCP opportunistic infections IDSA, 2001). One explanation for these prophylaxis, but developed PCP because observations is that more patients are of nonadherence or resistance to pro- as the initial experiencing immunologic decline in phylactic regimens. A smaller propor- association with long-term virologic fail- tion of patients continue to develop PCP manifestation of ure of antiretroviral therapy. An optimal at CD4+ cell counts higher than the level virologic response (ie, viral suppression established as the criterion for institu- HIV disease to below limits of assay detection) is not tion of primary PCP prophylaxis, which achieved in many patients with initial or is 200/µL or less. subsequent treatment. Continued virologic failure places such patients at Opportunistic Infection Risk colleagues (Ann Intern Med, 1999), oppor- increased risk of disease progression. and Prophylaxis tunistic infection incidence rates and For example, a 2000 review of 20 potent 95% confidence intervals in patients therapy arms in antiretroviral treatment CD4+ cell count still appears to be a reli- with CD4+ count nadirs of less than 200, studies (Bartlett et al, AIDS, 2001) indi- able predictor of risk for opportunistic 150, 100, or 50 cells/µL did not differ cated that, on intent-to-treat analysis in infections in the potent antiretroviral substantially (incidence rates of 3.7-8.1 the majority of the study arms, 50% or therapy era. In terms of heightening per 100 patient-years) when cell counts less of patients achieved plasma HIV-1 diagnostic vigilance for opportunistic increased to greater than 200/µL on RNA levels less than 50 copies/mL at 48 infections, however, it has long been potent therapy. However, the incidence weeks. An increased incidence of oppor- recognized that risk extends to patients rate was dramatically elevated to 72.9 tunistic infections during virologic fail- with CD4+ cell counts above the levels per 100 patient-years in patients with ure is not currently supported in the established as threshold levels for initi- persistently low CD4+ cell counts published literature, however. ation of prophylactic therapy. For exam- (Figure 1). ple, data from the Multicenter AIDS These and other data indicating that Dr Masur is Chief of the Critical Care Cohort study (Phair et al, N Engl J Med, CD4+ cell count increases on potent Medicine Department of the National 1990) prior to the advent of potent ther- antiretroviral therapy are associated Institutes of Health in Bethesda, Md. apy showed that among patients devel- with reduced risk for opportunistic infec-

27 International AIDS Society–USA Topics in HIV Medicine

tions have led to recommendations for discontinuation of prophylaxis based on CD4+ cell count and adequate control of plasma HIV-1 RNA level. Currently, there are specific CD4+ cell count-based Incidence Rate/ 200 Nadir CD4+ of 100 Patient-Years guidelines for discontinuation of prima- L) Responders (95% Cl) µ ry prophylaxis for PCP, toxoplasmosis, µ 150 <200/ L 3.7 (3.3–4.1) and Mycobacterium avium complex (MAC) <150/µL 6.0 (3.4–10.0) infection and secondary prophylaxis for <100/µL 8.1 (4.5–13.4) these infections and for cytomegalovirus 100 <50/µL 5.9 (3.0–10.2)

(CMV) disease and cryptococcosis Nonresponders 72.9 (69.0–76.8) (USPHS/IDSA, 2001). Data are currently with Persistent

CD4+ Count (cells/ 50 insufficient to recommend discontinua- Low CD4+ tion of secondary prophylaxis for histoplasmosis or coccidioidomycosis based 0 on CD4+ cell count. However, many Time practitioners with considerable experience with these latter types of infections are of the opinion that discontinuation Figure 1. Effect of pretreatment nadir CD4+ cell count on incidence of opportunistic infec- of prophylaxis is safe with sufficient tions in patients whose CD4+ cell count increased to greater than 200/µL on potent CD4+ cell count increase on potent antiretroviral therapy (responders) and incidence in patients with persistently low cell count antiretroviral therapy. (nonresponders). Adapted with permission from Miller et al, Ann Intern Med, 1999.

HIV-Associated Pulmonary Disease same consult service from 1997 to 2000 lavage and immunofluorescence stain- indicate that manifestations were due to ing. In practiced hands, diagnosis can be Pulmonary disease remains one of the PCP in 9 (18%), other bacterial infection made in a high percentage of cases using most common infectious complications in 24 (47%), CMV in 2 (4%), and lym- induced sputum and immunofluores- encountered in patients with HIV dis- phoma in 8 (16%; Wolff and O’Donnell, cence. There is considerable interest in ease. PCP, tuberculosis, and pneumo- Chest, 2001). However, the types of polymerase chain reaction (PCR)-based coccal and Haemophilus infections remain opportunistic infections observed may diagnostic techniques for PCP. One common among pulmonary infections, depend on the patient populations recent study showed that PCR analysis of as does infection due to atypical and studied. As noted, experience at some simple oral washes was associated with viral pathogens. Less common causes of centers indicates that many HIV-infected 91% sensitivity, 94% specificity, 76% pos- disease include histoplasmosis and coc- individuals without adequate connec- itive predictive value, and 98% negative cidioidomycosis, toxoplasmosis, and tions to health care are now presenting predictive value (Fischer et al, J Infect Dis, Kaposi's sarcoma. Among the relatively with opportunistic infections as initial 2001). Quantitative PCR techniques for less common pulmonary complications, manifestations of HIV disease, and the PCP diagnosis currently are being devel- cases of disease caused by Aspergillus, frequency with which PCP is encoun- oped, and may be commercially avail- Staphylococcus, and lymphoma are in- tered is greater among such individuals. able in the relatively near future. creasing in frequency. Pathogens that Trimethoprim/sulfamethoxazole may cause pulmonary disease in other Current Status of PCP Diagnosis and (TMP-SMX) remains the drug therapy of immunosuppressed populations and Management choice for PCP, with adjunctive cortico- that cause other types of disease in HIV- steroids being used in patients with infected patients, but that remain rare PCP should be considered in any HIV- severe disease. Pentamidine is also an causes of pulmonary disease in the lat- infected individual with pulmonary dis- effective therapy, but is associated with ter, include CMV, MAC, herpes simplex ease. Although suspicion should be adverse effects. Other treatment options, virus, and Rhodococcus. heightened in patients with lower CD4+ including atovaquone and clindamycin- Changes in the spectrum of HIV- cell counts, it should be remembered primaquine, have not provided sufficient related pulmonary disease have been that PCP occurs at counts greater than efficacy to be considered as first-line observed during the potent antiretrovi- 200/µL in a sizable proportion of treatment. There is some evidence that P ral therapy era. Data from 204 patients patients. PCP has a variety of radiologic carinii can become resistant to TMP-SMX, seen at a Georgetown consult service presentations. Although symmetrical which may explain some proportion of from 1993 to 1995 indicate that pul- diffuse interstitial infiltrates are most the disease incidence in patients receiv- monary manifestations were due to PCP common, the presentation can include ing TMP-SMX prophylaxis. Since at- in 74 patients (36%), other bacterial asymmetrical infiltrates, nodules, lobar tempts to culture P carinii have met with infection in 59 (29%), CMV in 14 (7%), disease, and cavitation. Specific diagno- failure, conventional resistance testing and lymphoma in 7 (3%). By compari- sis can be made in a very high percent- of the organism has not been performed. son, data from 51 patients seen in the age of cases using bronchoalveolar However, within the past several years,

28 Perspective - Opportunistic Infections Volume 10 Issue 5 November/December 2002

the target enzyme of TMP-SMX, dihy- was due to provocation of the immune hoped that advances can be made in dropteroate synthase (DHPS), has been response to PCP despite relative identifying and providing quality treat- sequenced, and it has been found that absence of the organisms. ment to the large number of HIV-infect- mutations in the enzyme are associated Similar cases have been described in ed individuals who currently are at with sulfa drug resistance in pneumococ- both patients with prior opportunistic increased risk of disease progression ci and other organisms. A study per- infections and those in whom presence because of lack of medical care. formed by Danish investigators indicates of an opportunistic pathogen was previ- that mortality due to PCP is greater in ously unrecognized. For example, it Presented in May 2002. First draft prepared from tran- patients exhibiting the DHPS mutations appears that nearly every patient with a scripts by Matthew Stenger. Reviewed and updated by Dr Masur in July 2002. (Helweg-Larsen et al, Lancet, 1999). history of CMV retinitis in whom potent Mutations appear to be more common in antiretroviral therapy is initiated will Financial Disclosure: Dr Masur has no affiliations with patients with prior exposure to sulfa develop some degree of retinal inflam- commercial organizations that may have interests related drugs, and have been found in stored mation due to immune reactivation syn- to the content of this article. samples from patients treated in the drome, with deterioration of vision 1990s and 1980s. There is thus some occurring in some. Other cases have concern that P carinii resistance to TMP- described reactivation syndromes in- SMX, although not a problem currently, volving, for example, cryptococcal pul- Suggested Reading could become a problem in the future. monary infection and MAC infection of mediastinal lymph nodes in patients Bartlett J, DeMasi R, Quinn J, Moxham C, Immune Reconstitution with no history of acute opportunistic Rousseau F. Overview of the effectiveness of Syndromes infection. In some cases, resolution of triple combination therapy in antiretroviral- the syndrome occurs without specific naive HIV-1 infected adults. AIDS. 2001;15:1369- Immune reconstitution syndromes in treatment and with continuation of 1377. patients initiating potent antiretroviral potent antiretroviral therapy. DeSimone JA, Pomerantz RJ, Babinchak TJ. therapy have been described in the set- Inflammatory reactions in HIV-1-infected per- tings of PCP and a variety of other Conclusion sons after initiation of highly active antiretrovi- pathogens, including CMV, MAC, tuber- ral therapy. Ann Intern Med. 2000;133:447-454. culosis, herpes simplex virus, varicella- Opportunistic infections are still occur- Fischer S, Gill VJ, Kovacs J, et al. The use of oral zoster virus, hepatitis C virus, and oral ring in HIV-infected individuals. In some washes to diagnose Pneumocystis carinii pneu- human papilloma virus. There are not yet settings, many of the patients with monia: a blinded prospective study using a good guidelines for distinguishing opportunistic infections are patients polymerase chain reaction-based detection sys- between reactivation syndromes and who are not receiving adequate medical tem. J Infect Dis. 2001;184:1485-1488. acute infections. The potential for care for HIV disease, including an immune reconstitution syndromes in Helweg-Larsen J, Benfield TL, Eugen-Olsen J, increasing number who present with Lundgren JD, Lundgren B. Effects of mutations patients initiating potent antiretroviral opportunistic infections as the initial in Pneumocystis carinii dihydropteroate syn- therapy should be taken into account in manifestation of HIV disease. The recent thase gene on outcome of AIDS-associated P. those cases in which HIV-infected leveling off, and perhaps increase, of carinii pneumonia. Lancet. 1999;354:1347-1351. patients not receiving antiretroviral opportunistic infection rates is also like- treatment present with an acute oppor- ly attributable in part to immunologic Hogg RS, Yip B, Chan KJ, et al. Rates of disease progression by baseline CD4 cell count and viral tunistic infection. Delay of antiretroviral decline in the large number of patients load after initiating triple-drug therapy. JAMA. treatment until the acute infection has maintained on virologically failing 2001;286: 2568-2577. completely resolved should be consid- antiretroviral regimens for prolonged ered in these cases. Randomized clinical periods. Renewed attention to guide- Kaplan JE, Masur H, Holmes KK, US Public trials of immediate versus deferred lines for prophylaxis and treatment for Health Service, Infectious Diseases Society of America. Guidelines for preventing opportunis- antiretroviral therapy in acutely ill opportunistic infections is warranted. tic infections among HIV-infected persons— patients are needed. Additional work is needed in providing 2002. Recommendations of the U.S. Public In an illustrative case (Wislez et al, information on distinguishing between Health Service and the Infectious Diseases Am J Respir Crit Care Med, 2001), a patient acute opportunistic infections and Society of America. MMWR Recomm Rep. 2002; 51(RR-8):1-52. presenting with PCP and a PO2 of 59 was immune reactivation syndromes and started on TMP-SMX and showed normal defining optimal management of the lat- Karavellas MP, Azen SP, Macdonald JC, et al. PO2 and considerably improved radio- ter. For the future, it is hoped that better Immune recovery vitritis and uveitis in AIDS: logic findings on day 15. After initiation antiretroviral regimens and the addition clinical predictors, sequelae, and treatment out- of potent antiretroviral therapy, the of immunotherapy to antiretroviral comes. Retina. 2001;21:1-9. patient returned on day 26 in respiratory treatment will permit more patients to failure; PO was 69 and an infiltrate was achieve viral control adequate to pre- Lederman MM, Valdez H. Immune restoration 2 with antiretroviral therapies: implications for observed on x-ray. Biopsy showed the vent immunologic deterioration. For clinical management. JAMA. 2000;284:223-228. presence of very few P carinii organisms example, novel approaches to raising and an intense inflammatory response, CD4+ cell counts such as interleukin-2 McNaghten AD, Hanson DL, Nakashima AK, indicating that the respiratory syndrome have considerable promise. It is also Swerdlow DL. Incidence of AIDS-defining oppor-

29 International AIDS Society–USA Topics in HIV Medicine

tunistic illnesses in the US may be leveling from results from the EuroSIDA study. Ann Intern Med. Wislez M, Bergot E, Antoine M, et al. Acute res- 1998 to 1999. [Abstract 751.] 39th Annual 1999;130:570-577. piratory failure following HAART introduction in Meeting of the Infectious Diseases Society of patients treated for Pneumocystis carinii pneu- America. October 25-28, 2001; San Francisco, Phair J, Munoz A, Detels R, Kaslow R, Rinaldo C, monia. Am J Respir Crit Care Med. 2001;164:847- Calif. Saah A. The risk of Pneumocystis carinii pneu- 851. monia among men infected with human Miller V, Mocroft A, Reiss P, et al. Relations immunodeficiency virus type 1. Multicenter Wolff AJ, O'Donnell AE. Pulmonary manifesta- among CD4 lymphocyte count nadir, antiretrovi- AIDS Cohort Study Group. N Engl J Med. tions of HIV infection in the era of highly active ral therapy, and HIV-1 disease progression: 1990;322:161-165. antiretroviral therapy. Chest. 2001;120:1888-1893.