The late-presenting HIV-infected patient 30 years after the introduction of HIV testing: spectrum of opportunistic diseases and missed opportunities for early diagnosis

Abstract

Objectives: The aim of the study was to describe the characteristics of HIV-infected late presenters, opportunistic diseases at diagnosis and missed opportunities to diagnose HIV infection earlier.

Methods: In a retrospective cohort study, we reviewed the medical records of all adults with newly diagnosed HIV infection admitted to the Department of Infectious Diseases of the Vivantes Auguste-Viktoria Hospital, Berlin, Germany.

Results: In the 5-year period from 2009 to 2013, 270 late presenters were identified. The most common AIDS-defining conditions were oesophageal candidiasis (n = 136; 51%), wasting syndrome (n = 106; 40%) and pneumocystis pneumonia (n = 91; 34%). Fifty-five patients (21%) had presented with at least one HIV indicator condition on prior contact with health care services without being offered testing for HIV. Female patients and heterosexual men [not men who have sex with men ('non-MSM')] had a significantly higher chance of being among patients previously presenting with indicator conditions and not being tested [odds ratio (OR) 4.7; 95% confidence interval (CI) 2.2-10.0; P < 0.001; and OR 2.4; 95% CI 1.2-5.1; P < 0.01, respectively]. The most commonly missed indicator conditions were leucocytopenia (n = 13; 24%), thrombocytopenia (n = 12; 22%), oral candidiasis (n = 9; 16%), unexplained weight loss (n = 7; 13%), herpes zoster (n = 5; 9%) and cervical dysplasia/cancer (n = 4; 20% of women). The median time between presentation with an indicator condition and the diagnosis of HIV infection was 158.5 days [interquartile range (IQR) 40-572 days]. Patients with oral candidiasis and unexplained weight loss had the shortest time between the "missed opportunity" and the diagnosis of HIV infection. Fifty-five hospital admissions with a total cost of over EUR 500 000 and - most importantly - six in-hospital deaths might have been prevented if HIV testing had been performed in patients with documented indicator conditions.

Conclusions: Indicator conditions are still missed by clinicians. Women and 'non-MSM' are at highest risk of presenting with an indicator condition but not being tested for HIV infection.

Keywords: AIDS; HIV; HIV testing; female gender; indicator condition; late presentation.