Medical Chronicle November/December 2013 | Page 44

HIV/TB FORUM
PART 1

DR DAVID SPENCER Head, Infectious Diseases, Helen Joseph Hospital, Johannesburg

HIV Meds and autopsy reviews in SA
Recent data from South African researchers indicate that HIV-infected people are surviving longer and that
the spread of the virus among those ?24 years may be decreasing. However, there are still areas of concern.
Provided baseline CD4 counts at
the time of ART initiation are ?200c/
µl, computer estimates place life
expectancy at 70%-86% of that of ageand gender-matched HIV-negative

South African adults. Community viral
loads, i.e. the ‘average’ viral load of
all tested in a specified community
have fallen in both Johannesburg and
Cape Town since 2004-2011. But, is

44 MEDICAL CHRONICLE NOVEMBER/DECEMBER 2013

this a reliable marker of reduced viral
transmission in these cities? Fewer
newborns are being vertically infected.
Transmission rates have fallen to
?5.6-3.5%. The country’s PMTCT

programme appears to be working.
Nonetheless, areas of concern
remain. Beds in public hospitals are
filled with the HIV-infected. Not all go
home. TB, pneumonia, overwhelming
sepsis, end-organ failure, drug toxicity
and drug-interactions now define
public medicine in SA. Two local HIV
autopsy reviews speak to the causes
of death in those who are HIV positive.
Not surprisingly, TB is the primary
cause of death in both reports.
47 consecutive autopsies were
assessed in a 2007 study from two
Johannesburg hospitals, the Charlotte
Maxeke and Chris Hani Baragwanath.
All were HIV-positive and all had a
clinical diagnosis of TB before death.
None had received antiretrovirals
(ARVs). TB was confirmed in 37
(79%). It was disseminated in 60%.
Additional non-TB disease was often
present in the lungs, e.g. bacterial
pneumonia (26%), cytomegalovirus
(CMV) DNA (60%), Pneumocystis
jiroveci infection (11%). Disease was
also found outside the lungs, i.e. in the
adrenal glands, liver, heart, brain and
lymph nodes. Much of this was not
TB: Non-typhoidal salmonellae and
Mycobacterium avium complex were
grown from the spleens of 11 (23%)
and two (4%) respectively. Renal
disease was frequent: Interstitial
nephritis, pyelonephritis, acute
tubular necrosis, HIV nephropathy
and fungal (Cryptococcus neoformans)
infection of the kidneys. Although
death was caused by TB in the
majority, other conditions contributed
to the final outcome.
The second report (2012) in
part updates the earlier. It is from
the Charlotte Maxeke hospital. 39
consecutive autopsies were performed
on HIV-positive adults during 2009.
The majority had been on ART before
death. Although a prior diagnosis of
TB was not a protocol requirement,
TB was again the most frequent (67%)
cause of death. Does the timing of the
start of antiretroviral therapy (ART)
influence mortality? Yes. Death from
TB was more likely after starting ART.
TB was found in 57% prior to the start
of ART but increased to 87% of those
dying within 16-50 days of the start
of ART. Once ART had been taken for
many months, the risk of dying with
active TB decreased to 60%. The
increased likelihood of death from TB
in the ‘early’ ART group suggests that
immune reconstitution played a part in
the outcome.