Ongoing cytomegalovirus (CMV) replication
is associated with an increased mortality risk for people who start
antiretroviral therapy with a low CD4 cell count, Thai investigators report in
the online edition of Clinical Infectious
Over a quarter of patients had CMV viremia when HIV treatment was
initiated and 5% still had detectable virus six months later. A high CMV viral
load at baseline was associated with a more than sevenfold increase in
mortality risk during the first two years of antiretroviral therapy.
“CMV viremia was associated with mortality
irrespective of CD4 counts,” write the authors.
Very little is known about the prevalence
of CMV viremia and its association with HIV disease progression and death among people starting antiretroviral treatment in poorer countries.
An international team of researchers
therefore retrospectively analysed stored blood samples obtained from 293 people in Thailand who started HIV therapy when their CD4 cell count was
below 200 cells/mm3. The researchers examined the prevalence of CMV
viremia at baseline and after six months of treatment and its association with
mortality and progression to new AIDS-defining events. They also explored the
factors associated with the persistence of CMV during HIV treatment.
Most of the patients were men (57%) and
their median age was 33 years. The patients had severe immune suppression at
baseline, median CD4 cell count being just 82 cells/mm3 and median
viral load was approximately 80,000 copies/ml.
CMV was detected in 26% of patients on
entry to the study. The prevalence of CMV viremia was related to immune
suppression and was highest among patients with the lowest CD4 cell counts (below
50 cells/mm3 = 46% vs 101-200
cells/mm3 = 10%, p < 0.001).
Median baseline CMV viral load was 548
Overall, the patients had a good response
to HIV therapy. After six months, 80% had a HIV viral load below 200 copies/ml
and median CD4 cell count had increased to 198 cells/mm3, with 48%
of patients having a CD4 cell count above 200 cells/mm3.
Repeat CMV testing was performed for 199
patients at the six-month timepoint. Viremia was detected in nine people
(5%) and their median CMV viral load was 114 copies/ml.
A total of eight deaths and 28 new
AIDS-defining events were observed over two years. The overall mortality rate
was 1.42 per 100 person-years, and the median time to death was 7.5 months
after the initiation of treatment. The rate of new AIDS-defining illnesses was
5.39 per 100 person-years, with the median time to the diagnosis of an event
being 2.1 months after starting antiretrovirals. No deaths were directly
attributed to CMV and only two AIDS-defining illnesses corresponded to CMV
However, six of the eight deaths occurred
in patients with a baseline CMV viral load above 500 copies/ml.
Indeed, the investigators’ first analysis
showed that a high CMV viral load at baseline was the most important risk
factor for death (HR = 21.09; 95% CI< 4.26-104.53, p < 0.001). Also
significant were low haemoglobin (HR = 6.87; 95% CI, 1.72-27.49, p = 0.011) and
CD4 cell count, with each 50 cell/mm3 increase reducing the risk of
death by approximately 20% (p = 0.022).
However, after controlling for potential
confounders, a baseline CMV load above 500 copies/ml was the only factor
associated with an increased mortality risk (HR = 7.28; 95% CI, 1.32-40.29, p
The investigators’ initial analysis showed
that both immune suppression and CMV viremia were associated with the
development of new AIDS-defining events. However, in their multivariate analysis
only a low CD4 cell count retained a significant association with their outcome
(p = 0.048).
Baseline factors associated with ongoing
CMV replication after six months of HIV therapy were baseline haemoglobin below
10 g/dl and baseline CMV viral load above 500 copies/ml (HR = 6.63; 95% CI,
1.41-31.08, p = 0.016).
prevalence of CMV viremia in our study resembles that seen in several studies
in severely immune-suppressed HIV-infected patients in high-income countries,”
write the authors. “The association between CMV viremia and increased mortality
in HIV patients, even in the context of ART, has also been long reported.”
They believe their findings have
significant implications for the care of patients in resource-limited settings.
Many people in poorer countries initiate HIV treatment with severe immune
suppression and mortality rates during the early months of treatment are high.
The investigators recommended “a
large-scale prospective study be conducted to better document the extent of the
problem of CMV viremia in severely immune-suppressed HIV patients in
resource-limited settings…we believe that a well-powered trial is required to
convincingly answer if pre-emptive therapy could reduce early mortality in CMV
viremic HIV patients starting ART.”