Page 303 - SAHCS HIVMed Journal Vol 20 No 1 2019

Page 4 of 5 Original Research

HIV-positive and HIV-negative groups. There was a tendency patients confirmed to have measles on serological testing,
for complications to be more common in the HIV-positive thus excluding 12 cases with clinical features of measles that
group; however, the only significant difference was a longer tested negative for measles by serology and which may have
length of hospital stay. represented false-negative cases because of HIV co-infection
with deficient antibody synthesis. Fourthly, only patients
Over the 6-month period, 51 adult patients were admitted with more severe disease were included as they were
with suspected MV infection. Despite having clinical features hospitalised patients. Lastly, the cases were all from a single
suggestive of MV infection, 12 (23.5%) patients had negative centre, and therefore the results may not be generalisable.
MV serology. Negative serological testing may be attributed However, our study provides data on the largest series of
to possible laboratory error, to patients not having mounted hospitalised adults infected with HIV and co-infected with
an adequate immune response because of underlying measles. Unlike other published literature, we were also able
immunocompromise or to undetectable antibody levels to provide a comparison of adult patients infected and
within the first 72 h of the exanthem appearing. 30 uninfected with HIV, within the same cohort.

Of the 13 females who consented to HIV serological testing, Conclusion
12 tested HIV-positive. There were twice as many females as
males in the subgroup infected with HIV. This may reflect the Our findings confirm that MV is still an important cause of
burden of HIV infection among women in the South African morbidity and mortality, even among adult patients. Co-
population. 31 infection with HIV may be associated with worse outcomes.
Future studies with larger patient numbers may substantiate
In 2012, the estimated adult (15–49 years) prevalence rate of this conclusion. HIV testing should be carried out in all
HIV and/or AIDS in South Africa was 18.8%. However, in adults with suspected MV infection. ‘Mop-up’ vaccination
31
our study the prevalence of HIV-positivity was 18 of 24 campaigns should perhaps also target adults infected with
patients (75%). This higher rate of HIV infection among our HIV with the aim of attaining protective antibody levels and
measles cases may be the result of HIV-infected patients reducing the risk of developing disease.
being at increased risk of acquiring measles and requiring
hospitalisation ; firstly because HIV-induced immune Acknowledgements
26
deficiencies are compounded with the immune-suppressive
effect of the MV and secondly because of an inferior response Competing interests
to measles vaccination. 10,11,32,33,34,35 The authors have no conflict of interests.

Presenting symptoms, findings on clinical examination and
laboratory results revealed no significant differences between Authors’ contributions
the HIV-infected and HIV-uninfected subgroups. All of the N.E.D. conducted the study and wrote the manuscript. C.F.
patients infected with HIV presented with features typical supervised the study and reviewed the manuscript. All
of MV infection, including the occurrence of a morbilliform authors have read and approved the final manuscript.
rash. This contrasts with published data documenting
atypical findings in HIV-infected patients. 15,16 Furthermore, Funding Information
this is also despite the median CD4 cell count of 109 cells/
mm³ in the HIV-infected subgroup, suggesting advanced This research received no specific grant from any funding
retroviral disease and immunosuppression. agency in the public, commercial or not-for-profit sectors.

Measles is typically a self-limiting illness, but individuals who Data availability statement
are immunocompromised are at increased risk of severe
disease. This was mirrored in our study as half of the HIV- Data sharing is not applicable to this article as no new data
24
infected adults in our cohort developed complications related to were created or analysed in this study.
MV infection, as compared to only one patient in the HIV-
uninfected subgroup (OR = 5, 95% CI 0.48–51.8, p = 0.34). The Disclaimer
length of hospital stay was significantly higher in the HIV- The views expressed in the article are those of the authors
infected subgroup (p = 0.03). All three deaths recorded in our and not an official position of the institution or funder.
cohort occurred in the HIV-infected subgroup (OR = 2.9, 95% CI
0.13–65.3, p = 0.55), resulting in a case fatality rate of 16.7% in References
this group.
1. Babbott Jr FL, Gordon JE. Modern measles. Am J Med Sci. 1954;228(3):334–361.
Possible limitations of this study include the following. https://doi.org/10.1097/00000441-195409000-00013
Firstly, there were small patient numbers and this may have 2. WHO/UNICEF. WHO/UNICEF Joint Annual Measles Report, WHO Factsheet
[homepage on the Internet]. 2017 [cited April 2017]. Available from: www.who.
limited our ability to show statistical significance in some of int/mediacentre/factsheet/fs286/en/
the endpoints. Secondly, it was a retrospective study and so 3. Takla A, Wichmann O, Rieck T, Matysiak-Klose D. Measles incidence and reporting
trends in Germany, 2007–2011. Bull World Health Org. 2014;92:742–749. https://
the datasets were not complete. Thirdly, we included only doi.org/10.2471/BLT.13.135145

http://www.sajhivmed.org.za 296 Open Access