The natural history of human immunodeficiency virus (HIV) infection and acquired immunodeficiency syndrome (AIDS) has changed dramatically since the onset of the epidemic in the 1980s. The landmark of this process was the introduction of highly active antiretroviral therapy (ART) in 1996. Despite the progress made in the treatment and control of HIV infection, HIV/AIDS persists as one of the main causes of death in the world, affecting individuals from both high-income and low-income settings.1 In addition, although an increase in non-AIDS associated morbidity and mortality has been observed, opportunistic infections remain a major cause of hospitalization and death in people living with HIV/AIDS in high and low-income settings.2–4

Currently, in the post-ART period, opportunistic illnesses are mainly related with late diagnosis and/or presentation to care, non-adherence to ART and HIV resistance to antiretroviral drugs.2,5 Late diagnosis and/or presentation to care is one of the most challenging aspects of the HIV epidemic. In Brazil, 34% of the patients still present with an opportunistic illness at the moment of ART initiation.6 Furthermore, non-adherence to ART results in virologic failure and disease progression. Factors associated with non-adherence, such as low educational level, young age, unemployment, alcoholism and use of illicit drugs represent an important socioeconomic problem, in particular for low/middle-income settings.7,8 Finally, multidrug resistance to antiretroviral drugs is a consequence of HIV exposure to ART, particularly in settings where non-adherence prevails.9

In this study, we review the trends in opportunistic illnesses incidence rates and compare the results observed in high-income settings (HIS) with that for low/middle-income settings (LMIS), with special attention given to studies from Brazil. We evaluate the impact ART has had in three specific opportunistic infections of particular importance to Brazil and contrast the patterns in the countries evaluated.

Publications related to AIDS-associated opportunistic illnesses incidence were identified by systematically searching in Pubmed, Web of Science, Lilacs and Google scholar. Publications were restricted to the following languages: English, Portuguese, and Spanish. The databases were searched for studies published until January 2013 using the following search terms and Boolean operators, for matches under any field: (incidence) AND (HIV OR human immunodeficiency virus) AND (AIDS-defining illness OR opportunistic infection OR opportunistic disease OR AIDS-related opportunistic infection OR AIDS-related opportunistic illness). For the Lilacs database, search terms were translated into Portuguese language and separate searches with each term were conducted. Titles and available abstracts were scanned for relevance identifying papers requiring further consideration. Bibliographies of relevant articles were also checked. Inclusion criteria consisted in (1) presence of a person-time denominator and (2) results for all opportunistic illnesses and/or the three opportunistic infections of interest, namely: Pneumocystis carinii pneumonia (PCP), cerebral toxoplasmosis (NTX) and Mycobacterium avium complex (MAC). Exclusion criteria included: (1) results given only for hospitalization and/or severe diseases, (2) results given relative terms only (that is, as incidence rate ratios, odds ratios or relative risks), (3) results given only for Immune Reconstitution Inflammatory Syndrome, and (4) results that aggregate death and opportunistic infections in one outcome. The results, inclusion and exclusion criteria are shown in Fig. 1.

Fig. 1.

Search strategy and papers selection flowchart.

(0.23MB).

Thirty seven publications met the study's eligibility criteria, 25 from HIS and 12 studies from LMIS (Fig. 1).2,10–45 Out of the 12 studies from LMIS, four were from Latin America, specifically from Brazil. Results from these studies are summarized in the next sections with incidence rates in 100 person-years (100 PY) format.

Through a systematic review of the literature, we have shown that the incidence of opportunistic illnesses decreased over the 30 years of the HIV epidemic, markedly after ART availability. The significant reduction in the incidence rates was demonstrated for opportunistic illnesses overall and also for specific opportunistic infections, namely, PCP, NXT and MAC. In addition, the decreasing trends were shown for both HIS and LMIS where ART was made available. This result is extremely positive as it shows that opportunistic illnesses can be controlled while also pointing to the persistent challenge of a timely diagnosis of HIV infection. Indeed, in order to control opportunistic illnesses HIV infection status must be identified and earlier linkage to care needs to be facilitated. That is, a higher uptake of HIV testing with direct linkage to care of those found to be HIV-infected is urgently needed.

We found that the magnitude of the incidence rates and of the reduction of these rates as a function of ART varied between studies. Indeed, it is well known that there are geographical differences in the incidence of opportunistic illnesses.26 Other reasons for the differences in the baseline rates might include different study populations, including different sociodemographic subgroups evaluated in a specific study, for example, the MACS cohort that focuses on men who have sex men,27 as well as different inclusion criteria. Some studies included all HIV-infected individuals while others restricted the study population to individuals with specific CD4+ cell counts, for example, including only those with CD4+ cell count less than 100 cells/mm3.15 Moreover, different study definitions with respect to the diseases chosen to be included in any given study might have further contributed to the disparate results.

Out of the 37 studies included in the present review, almost 70% were from HIS. Of the 12 studies from LMIS, four studies were from Brazil.13–15,25 These studies reported incidence rates for opportunistic illnesses for the entire study period included in the respective study and not annual rates that could allow us to evaluate the temporal trends in incidence. Also, only one study from Brazil reported separately on the incidence rate of MAC. However, this study reported no cases of the disease, a finding that could have been due to the small sample size and/or short follow-up. For other important diseases that define the AIDS epidemic, namely, PCP, and NXT, no studies from Brazil were found. Furthermore, all are single center cohort studies, two from São Paulo (Hospital das Clínicas, Faculdade de Medicina da Universidade de São Paulo) and two from Rio de Janeiro (Instituto de Pesquisa Clínica Evandro Chagas, Fundação Oswaldo Cruz). We believe that the description of the trends in incidence rates of opportunistic illnesses is of paramount value to health care providers to guide clinical decision-making and policy makers to define priorities for care and prevention of opportunistic infections.

Strengths and limitations to the present study are worth mentioning. Through a systematic review conducted in four databases we found the epidemiological studies that reported on the incidence rate of opportunistic illnesses. We restricted the review to those studies reporting on rates (and not overall numbers or frequencies) since this epidemiological parameter is adjusted for population size and time under risk thus allowing for comparisons between studies. Though not a limitation of our study design and approach, the scarcity of studies from LMIS implies that we cannot adequately describe the patterns of incidence in these settings. In addition, the few studies found should also not be understood as representative of entire countries as they report from one center only. Finally, the different study methodologies such as inclusion criteria and diseases included, for example, limited the comparisons.

In conclusion, the incidence rate of opportunistic illnesses has decreased over time mainly due to the availability of highly effective, safe and well tolerated ART. However, a public health challenge remains for future years. Public health policies focusing on earlier HIV diagnosis and linkage to care, adherence and retention programs, and surveillance of HIV multidrug resistance in populations should be developed and implemented with the goal of improving the quality of life and reducing morbidity and mortality among HIV-infected individuals. To better understand the nuances of the epidemiology of opportunistic illnesses in LMIS, multicenter cohort studies should be encouraged. Finally, it is clear that studies from Brazil are urgently needed to assess the current burden of opportunistic illnesses in order to support the planning of HIV/AIDS health care services organization.

BG acknowledges funding from the National Council of Technological and Scientific Development (CNPq) and the Research Funding Agency of the State of Rio de Janeiro (FAPERJ). PML acknowledges funding from the National Council of Technological and Scientific Development (CNPq).

The authors declare no conflicts of interest.