View Record

TitleSurvival of South-African HIV infected patients
AuthorPost, Frank A
SubjectHIV epidemiology
SubjectSurvival analysis
AbstractIn sub-Saharan Africa, resource-limitation results in scarce availability of HIV prognostic tools such as CD4+ T-Lymphocyte (CD4) count and HIV viral load. To facilitate counselling and clinical decisions in this setting, widely available and inexpensive markers of prognosis are required. Chapter one gives an overview of the epidemiology and pathophysiology of HIV infection (with particular reference to sub-Saharan Africa), and its clinical manifestations. Staging systems for HIV infection and aspects of management in resource-poor environments are briefly discussed. Chapter two describes the epidemiological, pathophysiological and clinical aspects of tuberculosis (TB) in HIV infected patients, the commonest opportunistic infection in sub-Saharan Africa. It further provides HIV and TB prevalence data from the Western Cape, South Africa. In chapter three a study is presented demonstrating the usefulness of the total lymphocyte count (TLC) in combination with the World Health Organisation (WHO) clinical staging system to predict outcome in 831 HIV positive patients. A TLC of 1250/μL was found to be the equivalent of a CD4 count of 200/μL. Patients with early HIV disease (WHO stage 1&2) had low annual rates of progression to AIDS : 3-4% if the TLC was above 1250/μL, 12-14% if the TLC was below 1250/μL. Annual progression to AIDS increased to 25% and 46% in patients with clinical stage 3 and a TLC above or below 1250/μL respectively. Patients with AIDS had 30-55% one-year mortality rates depending on the TLC. Chapter four illustrates that pulmonary tuberculosis (PTB) in HIV infected patients presents with a radiographic spectrum reflecting the degree of HIV induced immune suppression. Chest radiographs and pre-treatment total lymphocyte counts provide prognostic information. Upper zone cavitatory infiltrates typical of reactivation PTB were associated with a preserved CD4 count (mean 389/μL) and predicted a 100% two-year survival. Pleural effusions were associated with a mean CD4 count of 184/μL and predicted 65% two-year survival. Patients with atypical radiographic presentation, including lower and mid-zone infiltrates, hilar and mediastinal adenopathy or interstitial patterns, had low CD4 counts (mean 105/μL) and a 36% survival at two years. Rather than classifying every patient with pleura-pulmonary tuberculosis as WHO stage 3, incorporation of the prognostic value of the chest radiograph into the clinical staging system, such that typical reactivation PTB becomes stage 2, tuberculous pleural effusion stage 3 and atypical PTB stage 4, would enhance the prognostic accuracy of HIV related tuberculosis. Chapter five demonstrates that patients with AIDS could be categorized accord ing to one of three survival patterns, relating to the type of opportunistic illness. One-year survival rates were highest for extra-pulmonary tuberculosis and herpes simplex virus infection (70% ); intermediate for oesophageal candidiasis, cryptococcal meningitis, kaposi sarcoma and pneumocystis carinii pneumonia (45%) ; and poorest for the HIV wasting syndrome, AIDS-dementia complex and performance status 4 (20%). Despite the overall poor prognosis associated with the acquired immunodeficiency syndrome, a substantial proportion of patients survive, even in the absence of anti-retroviral therapy, for a number of years. Chapter six concludes by proposing how the data presented in this thesis could be used in the clinical management of patients with HIV infection in a resource limited environment.
PublisherUniversity of Cape Town
PublisherFaculty of Health Sciences
PublisherDesmond Tutu HIV Centre