World Journal of AIDS, 2011, 1, 149-154
doi:10.4236/wja.2011.14022 Published Online December 2011 (http://www.SciRP.org/journal/wja)
Copyright © 2011 SciRes. WJA
149
Role of HIV-1 Viral Load in Initiating
Antiretroviral Therapy
Nayana A. Ingole, Seema M. Kukreja, Preeti R. Mehta
Microbiology Department, Seth Gordhandas Sunderdas Medical College & King Edward Memorial Hospital, Mumbai, India.
E-mail: drkseema@gmail.com
Received August 3rd, 2011; revised September 10th, 2011; accepted September 27th, 2011.
ABSTRACT
Purpose: For commencement of Antiretroviral Therapy (ART), CD4 count and/or WHO clinical staging is used as the
guide in India. In western countries along with clinical and immunological criteria, HIV-1 viral load is also used to
start the patient on treatment. Th e present study was conducted to determ ine the role of vira l load in takin g decision on
ART commencement in HIV-1 infected treatment naïve individuals. Method: A cross sectional study was carried out at
the Integrated Counseling and Testing Centre (ICTC) in the Department of Microbiology at a Tertiary care teaching
hospital after Institutional Ethics Committee approval. After obtaining written informed consent, HIV-1 infected pa-
tients who were clinically a symptomatic, ART naïve, having CD4 count <250 cells/mm3 and age more th an or equal to
15 yrs were enrolled in this study. Blood sample was collected and viral load was estimated by COBAS® TaqMan®
HIV-1 Test. Result: During the study period of one year, 8966 HIV-1 infected patients were referred for CD4 count
estimation. Of these 1624 patients had CD4 count <250 cells/mm3 and 405 patients were treatment naïve. Of these 96
(23.70%) patients were clinica lly asymptomatic and were enrolled. Of those enro lled, ten (10.41%) had viral load less
than 5000 copies/ml. Co nclusion: Decision to start patient on ART can be made judiciously when viral load is used
along with CD4 count estimation.
Keywords: Viral Load, CD4 Counts, HIV, Monitoring, ART, Clinically Asymptomatic
1. Introduction
In developed countries viral load monitoring along with
CD4 count estimation is used to make decision regarding
commencement of Antiretroviral Therapy (ART) in an
HIV infected individual.
Viral load testing is important as a guide for clinical
decision making as regards initiation of treatment, swit-
ching to second-line treatment and optimizing the dura-
tion of the first-line treatment regimen. In addition, the
viral load is a particularly useful tool for monitoring ad-
herence to treatment, performing sentinel surveillance for
new infections/seroconversion, and diagnosing HIV in-
fection in children aged <18 months [1].
But for developing countries, World Health Organiza-
tion (WHO) recommends starting ART when CD4 count
falls below 350 cells/mm3 or when patient is in WHO
clinical stage III or IV if CD4 testing is not available [2].
National AIDS Control Organization (NACO) recom-
mends initiating therapy when patient is in WHO clinical
stage III and IV or when CD4 count < 250 cells/mm3 [3].
Although it is a standard practice in high-income
countries, determination of the HIV viral load is not
recommended in developing countries because of the
costs and technical constraints. But as more and more
countries establish capacity to provide second-line ther-
apy, and as costs and technological constraints associated
with viral load testing decrease, determination of the vi-
ral load becomes necessary [1].
Hence, this study was conducted to determine whether
performing HIV-1 viral load can lead to better decisions
in initiating therapy.
2. Material and Methods
A cross sectional study was carried out at the Integrated
Counseling and Testing Centre (ICTC) in the department
of Microbiology at a Tertiary care teaching hospital. In-
stitutional Ethics Committee approval was obtained prior
to commencement of the study. HIV-1 infected patients
who were clinically asymptomatic, ART naïve, having
CD4 count < 250 cells/mm3 and age more than or equal
to 15 yrs were enrolled in this study after obtaining writ-
ten informed consent. HIV-1 viral load was estimated by
Role of HIV-1 Viral Load in Initiating Antiretroviral Therapy
150
COBAS® TaqMan® HIV-1 test.
2.1. Estimation of HIV Viral Load
Principles of the Procedure
The COBAS® TaqMan® HIV-1 Test is based on three
major processes: 1) manual specimen preparation to ob-
tain HIV-1 RNA; 2) automated reverse transcription of
the target RNA to generate complementary DNA (cDNA)
3) simultaneous PCR amplification of target cDNA using
HIV-1 specific complementary primers, and detection of
cleaved dual fluorescent dye-labeled oligonucleotide
detection probes that permit quantitation of HIV-1 target
amplified product (amplicon) and HIV-1 Quantitation
Standard RNA, which is processed, amplified and de-
tected simultaneously with the specimen. The test can
quantitate HIV-1 RNA over the range of 47 - 10,000,000
copies/mL.
2.2. Specimen Collection and Storage
Three ml of venous blood sample was collected in EDTA
vacutainer from each subject. Within six hours of collec-
tion, the sample was centrifuged at 800 - 1600 × g for 20
minutes at room temperature to separate the plasma.
Separated plasma was transferred into sterile polypro-
pylene screw cap tube and was stored in frozen form at
70˚C till further use.
2.3. Specimen, Control and Reagent Preparation
1) Frozen plasma specimens were kept at room tempera-
ture until completely thawed and were vortexed for 5 -
10 seconds before use.
2) The reagents were allowed to reach ambient tem-
perature before proceeding. Heating Block(s) were pre-
heated to a temperature of 70˚C (±2˚C) and water bath to
a temperature of 50˚C (±2˚C) before starting the purifi-
cation reactions.
3) Batches of 21 samples and 3 controls (High positive
control, low positive control and negative control) were
processed at a time.
2.3.1. Reagent Preparation (For 24 Tests)
Reagent preparation was done under a laminar hood
1) The Inhibitor Removal Buffer was prepared by pi-
petting 20 ml of 96% - 100% ethanol to Inhibitor Re-
moval Buffer (IRB). Mixing was done by inverting 5 - 10
times.
2) The Wash Buffer was prepared by pipetting 80 ml
of 96% - 100% ethanol to the Wash Buffer (WASH).
Mixing was done by inverting 5 - 10 times.
3) The Elution Buffer (ELB) was preheated at 70˚C
(±2˚C) in a 2.0 mL screw-cap microfuge tube. Volume
eluted was 4 ml.
4) 0.5 ml of the Elution Buffer (ELB) was pipetted
into the Carrier RNA (CAR). Mixing was done by in-
verting; then vortexing was done until all of the Carrier
RNA was dissolved.
5) 5.0 ml of the Elution Buffer (ELB) was pipeted into
the Proteinase K (PK). Mixing was done by inverting;
then vortexing was done until all of the Proteinase K was
dissolved.
6) The Lysis/Binding Working Solution was prepared
by pipetting the volumes listed in the Table 1.
The indicated volume of Lysis/Binding Buffer was
added to a clean sterile 50 mL tube.
The indicated volume of reconstituted Carrier RNA
was added to the tube containing the Lysis/Binding
Buffer.
The HIV-1 QS was vortexed for 3 - 5 seconds and
indicated volume of HIV-1 QS was added to the tube
containing the Lysis/Binding Buffer and the reconsti-
tuted Carrier RNA.
Mixing was done by inverting 10 - 15 times.
The indicated volume of reconstituted Proteinase K
was added to the tube containing the Lysis/Binding
Buffer.
Mixing was done by inverting 10 - 15 times.
2.3.2. Specimen and Control Preparation
Specimen and control preparation was also done under
laminar hood.
1) 625 μL of Lysis/Binding Working Solution was pip-
peted into each well of the Lysis Rack (I, transparent).
2) 500 μL of specimen or control was pippeted into the
appropriate well.
3) After all specimens and controls have been added,
mixing was done by vortexing for approximately 10
seconds.
4) The Lysis Rack was incubated in a preheated 50˚C
(±2˚C) water bath for 10 minutes. Lysis Rack was dried
after removing from water bath.
5) The Lysis Rack was centifuged for 10 - 20 seconds
at a setting of 4600 × g in the micro-titer plate centrifuge.
6) 350 μL of isopropanol was pippeted into each well.
7) Specimens were mixed by inverting the rack three
times, then vortexing the rack for approximately 10 sec-
onds.
8) The Lysis Rack was centrifuged for 10 - 20 seconds
Table 1. Preparation of lysis/binding working solution.
Reagents 24
Lysis/Binding Buffer (mL) 14.0
Carrier RNA (μL) 280
HIV-1 QS (μL) 168
Proteinase K (mL) 2.8
Copyright © 2011 SciRes. WJA
Role of HIV-1 Viral Load in Initiating Antiretroviral Therapy151
at a setting of 4600 × g in the micro-titer plate centrifuge.
9) Opening one well at a time, 750 μL of specimen or
control mixture was transferred to the corresponding
wells of the Filter Tube Rack (II, yellow) with affixed
Waste Rack (white).
10) After all specimens or controls have been added,
the Filter Tube Rack assembly was centrifuged for 2
minutes at 4600 × g in the micro-titer plate centrifuge.
11) Opening one well at a time, the remaining speci-
men or control mixture was transferred to the corre-
sponding wells of the Filter Tube Rack.
12) The Filter Tube Rack assembly was centrifuged
for 2 minutes at 4600 × g in the micro-titer plate centri-
fuge.
13) The Filter Tube Rack was replaced with a new
Waste Rack.
14) 400 μL of Inhibitor Removal Buffer (IRB) was
pippeted down the side of each well.
15) The Filter Tube Rack assembly was centrifuged
for 2 minutes at 4600 × g in the micro-titer plate centri-
fuge.
16) 700 μL of Wash Buffer (WASH) was pippeted
down the side of each well.
17) The Filter Tube Rack assembly was centrifuged
for 2 minutes at 4600 × g in the micro-titer plate centri-
fuge.
18) The Filter Tube Rack was replaced with a new
Waste Rack
19) 700 μL of Wash Buffer (WASH) was pippeted
down the side of each well.
20) The Filter Tube Rack assembly was centrifuged for
3 minutes at 4600 × g in the micro-titer plate centrifuge.
21) The Filter Tube Rack was placed onto the Elution
Rack (IIIA, blue).
22) 75 μL of the prewarmed Elution Buffer (ELB) was
pipette onto the center of each filter without touching the
filter. The Elution Rack was incubated at room tempera-
ture for a minimum of 3 minutes after adding Elution
Buffer to the last well.
23) The Filter Tube Rack assembly was centrifuged for
3 minutes at 4600 × g in the micro-titer plate centrifuge.
24) The Filter Tube Rack was removed from the Elu-
tion Rack.
25) The Cover Rack (IIIB, blue) was placed onto the
Elution Rack (IIIA, blue).
26) The processed specimens and controls are then
used directly for PCR.
2.3.3. Reagent Preparation
Reagent preparation was done inside a PCR work station.
1) One vial of HIV-1 MMX and one vial CTM Mn2+
was equilibrated at ambient temperature for 30 minutes.
2) K-carrier was placed in a K-carrier holder.
3) New K-tubes were placed in the K-carrier without
touching the sides of the K-tubes.
4) The K-tubes were uncapped using the K-tube Cap-
per.
5) The Working MMX was prepared as follows:
For 24 tests, 170 μL of CTM Mn2+ was added to one
vial of HIV-1 MMX. Mixing was done inverting 10 times.
6) 50 μL of Working MMX was pippeted into each
K-tube.
7) 50 μL of each processed specimen and control were
added to the appropriate K-tube containing Working
MMX using a micropipettor with an aerosol barrier or
positive displacement tip.
Extracted RNA Reverse transcription, amplification
and detection were automatically performed by the CO-
BAS® TaqMan® 48 Analyzer.
The COBAS® TaqMan® 48 Analyzer:
1) Determined the Cycle Threshold value (Ct) for the
HIV-1 RNA and the HIV-1 Quantitation Standard RNA.
2) Determined the HIV-1 RNA titer based upon the Ct
values for the HIV-1 RNA and HIV 1 Quantitation Stan-
dard RNA and the lot-specific calibration coefficients.
3) Determined that the calculated copies/mL titer for
HIV-1 L (+) C and HIV-1 H (+) C fall within the as-
signed ranges.
2.4. Interpretation of Results
Result was checked for flags and comments to ensure
that the run is valid. For a valid run, result for each indi-
vidual specimen was interpreted as mentioned in Table 2.
3. Results
During the study period of one year from April 2009 to
March 2010, 8966 HIV-1 infected patients were referred
for CD4 count estimation to our ICTC. Of these, 1624
patients had CD4 count <250 cells/mm3 and 405 patients
were treatment naïve. Of these 96 (23.70%) patients were
clinically asymptomatic and were enrolled (Figure 1). Of
the 96 patients enrolled, 58 were males and 38 were fe-
males. Ten patients had viral load <5000 copies/ml. Of
these, eight were females and two were males (Table 3).
4. Discussion
CD4 count has been reported to be one of the best surro-
gate markers for monitoring the progression of HIV in-
fection and low CD4 counts are associated with in-
creased risk of developing AIDS or death [4]. Contrary to
this, in our study 23.70% of patients having CD4 count <
250 cells/mm3 were asymptomatic. Similar findings have
been reported by R. Kannangia et al. (2008) where ten
(15.2 %) of the 66 patients in CDC category C with CD4
count <200 cells/mm3 were asymptomatic [5]. Thus
Copyright © 2011 SciRes. WJA
Role of HIV-1 Viral Load in Initiating Antiretroviral Therapy
Copyright © 2011 SciRes. WJA
152
Table 2. Interpretation of results.
Titer Result Interpretation
Target Not Detected Ct value for HIV-1 above the limit for the assay or no Ct value for HIV-1 obtained. Results reported
as “HIV-1 RNA not detected”.
<4.70E+01 C/mL Calculated copies/mL were less than the Limit of Quantitation of the assay. Results reported as
“HIV-1 RNA detected less than 47 HIV-1 RNA copies/mL”.
4.70E+01 C/mL and 1.00E+07 C/mL Calculated results greater than or equal to 47 copies/mL and less than or equal to 1.00E+07 copies/mL
are within the Linear Range of the assay.
>1.00E+07 C/mL Calculated copies/mL were above the range of the assay. Results reported as “greater than 1.00E+07
HIV-1 RNA copies/mL”.
7342
1219
309
96
405
CD4>250 CD4<250 on ART
CD4<250 Not on ARTSymptomatic
Asymptomatic
Figure 1. Profile of study population.
Table 3. Distribution of viral load.
Viral load (copies/ml) Male female No of patients
<50 1 2 3
50 - 500 1 2 3
500 - 5000 0 4 4
5000 - 50000 12 4 16
50000 - 500000 27 21 48
>500000 17 5 22
Total 58 38 96
discordance between clinical and immunological profile
of the patients was observed.
National AIDS Control Organization (NACO) recom-
mends initiating therapy when patient is in WHO clinical
stage III and IV or when CD4 count is below 250
cells/mm3 [3]. In the present study for starting patient on
ART, if only clinical criteria were used it could have
been misleading.
Starting ART based on immunological criteria has its
own limitations. CD4 count is influenced by factors such
as age, sex, physical and psychological stress, ethnicity,
infection, malnutrition, pregnancy, corticosteroid use,
normal diurnal variation, and social isolation [6]. Sec-
ondly the mean CD4 count is inherently low in Indian
population compared to western population. Also, cut off
values of CD4 counts for starting the patient on ART are
determined based on findings in western population
which may not be appropriate in our settings [7]. The new
categories of CD4 counts proposed are cell count >300,
81 - 300, and <80 cells/mm3 instead of <500, 201 - 500,
<200 cells/mm3 by R. Kannangia et al. (2008) [5]. Attili
VS et al. (2005) have proposed Categories of CD4 counts
as >280, 120 - 280 and < or =120 cells/mm3 [8]. Thus a
separate cut off for CD4 count is recommended for In-
dian population. Thirdly it has been reported that CD4
count is not a good indicator for predicting viral load [9,
10]. CD4 criteria had a sensitivity of 53.9%, specificity
of 76.1%, and the positive predictive value of 44.2% in
detecting virological failure [10].
Considering all these factors viral load has important
implications in taking decision on ART commencement.
To start patient on ART based on viral load, there are
no recommendations from WHO/British HIV Associa-
tion (BHIVA).
United States Department of Health and Human Treat-
ment Services/Henry J. Kaiser Foundation recommend
starting patient on ART when viral load is >10,000 to
20,000 copies per ml or when CD4 cell count is less than
500 cells per mm3 [11]. International AIDS Society-USA
Panel recommends starting patient on ART when viral
load is more than 30,000 copies per ml or when CD4+
cell count is less than 350 cells per mm3 and ART is de-
ferred if viral load is less than 5000 copies per ml [11].
Of the 96 patients studied, ten (10.41%) patients had
viral load less than 5000 copies/ml.
In resource limited settings like ours where CD4 is
used as a guide to start patient on ART, these ten
(10.41%) patients would have been eligible for ART but
if viral load is considered then ART can be deferred in
these ten patients.
Starting patient on ART is easy but as ART is a life-
long treatment, adherence to therapy is a problem. Ad-
herence rate reported in different studies ranges from
60.4 % to 74.3% [12,13]. ART can also lead to toxicities.
Fatu Forna et al. (2005) and Reto Nuesch et al. (2004)
have reported toxicities in 39.0% and 24.2% of the pa-
tients respectively [14,15]. Increasing resistance to ART
is another issue. In a study by Adge C et al. (2005) 57%
Role of HIV-1 Viral Load in Initiating Antiretroviral Therapy153
of the patient had genotypic resistance to at least one
reverse transcriptase inhibitor (RTI) or protease inhibitor
(PI) [16]. In Surveillance on HIV antiretroviral drug re-
sistance in treated individuals in England HIV drug re-
sistance reported was 55% [17].
Hence, for rationale use of ART, viral load can be a
better guide compared to CD4 alone. When viral load
and CD4 count both are used for predicting progression
to AIDS better estimate of risk of progression is provided
compared to when either marker is used alone [18]. Also
if baseline viral load is available one can monitor the
response to treatment by documenting the fall or rise in
viral load [11].
Viral load monitoring is significantly superior to CD4
monitoring in assessing viral suppression and treatment
failure in patients on ART [9,10] and as clinical failure is
an even later development, defining treatment failure on
clinical grounds alone is equally suboptimal [1].
Hence, under the National AIDS Control Programme
(NACP), the State Reference Laboratories (SRL) can be
given the responsibility of performing baseline viral load
estimation and monitoring response to therapy especially
in case of suspected drug resistance.
In the present study, eight out of 38 females and two
out of 58 males had viral load <5000 copies/ml.
In a meta-analysis for gender difference in viral load,
Sonia Napravnik et al. (2010) observed that a given
population of women have, on average, lower plasma
HIV RNA levels compared with a population of men
with similar CD4 lymphocyte counts and stage of HIV
disease [19]. Shade SB et al. (2000) also observed similar
gender differences in viral load. Women with advanced
HIV disease have slightly lower levels of viral load than
men at corresponding levels of CD4+ cell counts [20].
These gender differences in viral load could necessi-
tate gender specific recommendations for initiation of
ART. Hence when HIV-1 RNA thresholds are used to
form treatment recommendations for initiating ART, a
lower threshold is likely indicated for women compared
with men.
To conclude, decision to start patient on ART can be
made judiciously when viral load is used along with CD4
count estimation. Also, if baseline viral load value is
available it will help in assessing subsequent response to
therapy once initiated. Gender differences in viral load
necessitate gender specific recommendations for initia-
tion of ART. Virological monitoring should be combined
with clinical and immunological monitoring for better
patient management.
REFERENCES
[1] A. Calmy, N. Ford, B. Hirschel, S. J. Reynolds, L. Lynen,
E. Goemaere, F. G. de la Vega, L. Perrin and W. Rodri-
guez, “HIV Viral Load Monitoring in Resource-Limited
Regions: Optional or Necessary?” Clinical Infectious Dis-
eases, Vol. 44, No. 1, 2007, pp. 128-134.
doi:10.1086/510073
[2] WHO, “Antiretroviral Therapy for HIV Infection in
Adults and Adolescents, Recommendations for a Public
Health Approach 2010 Revision,” 2010.
http://whqlibdoc.who.int/publications/2010/97892415997
64_eng.pdf
[3] NACO, “Antiretroviral Therapy Guidelines for HIV In-
fected Adults and Adolescents Including Post Exposure
Prophylaxis,” Ministry of Health and Family Welfare,
2007.
http://www.nacoonline.org/upload/Policies%20&%20Gu
idelnes/1.%20Antiretroviral%20Therapy%20Guidelines%
20for%20HIVInfected%20Adults%20and%20Adolescent
s%20Including%20Post-exposure.pdf
[4] Y. Pirzada, S. Khuder and H. Donabedian, “Predicting
AIDS-Related Events Using CD4 Percentage or CD4
Absolute Counts,” AIDS Research and Therapy, Vol. 3,
2006, p. 20. doi:10.1186/1742-6405-3-20
[5] R. Kannangai, A. J. Kandathil, D. L. Ebenezer, G. Nithy-
anandam, P. Samuel, O. C. Abraham, T. D. Sudarsanam,
S. A. Pulimood and G. Sridharan, “Evidence for Lower
CD4+ T Cell and Higher Viral Load in Asymptomatic
HIV-1 Infected Individuals of India: Implications for
Therapy Initiation,” Indian Journal of Medical Microbi-
ology, Vol. 26, No. 3, 2008, pp. 217-221.
doi:10.4103/0255-0857.42031
[6] M. Irwin, “Low CD4+ T Lymphocyte Counts: A Variety
of Causes and Their Implications to a Multifactorial
Model of AIDS,” Virusmyth Homepage, 2001.
“http://www.virusmyth.com/aids/hiv/milowcd4.htm
[7] K. Ray, S. M. Gupta, M. Bala, S. Muralidhar and J.
Kumar, “CD4/CD8 Lymphocyte Counts in Healthy, HIV-
Positive Individuals & AIDS Patients,” Indian Journal of
Medical Research, Vol. 124, 2006, pp. 319-330.
[8] V. S. Attili, S. Sundar, V. P. Singh and M. Rai, “Validity
of Existing CD4+ Classification in North Indians, in Pre-
dicting Immune Status,” Journal of Infection, Vol. 51, No.
1, 2005, pp. 41-46. doi:10.1016/j.jinf.2004.08.022
[9] A. D. Kimmel, M. C. Weinstein, X. Anglaret, S. J. Goldie,
E. Losina, Y. Yazdanpanah, E. Messou, K. L. Cotich, R.
P. Walensky and K. A. Freedberg, “Monitoring of Anti-
retroviral Therapy in Resource-limited Settings: Discus-
sion,” Journal of Acquired Immunodeficiency Syndrome,
Vol. 54, No. 3, 2010, pp. 258-268.
doi:10.1097/QAI.0b013e3181d0db97
[10] B. Boughton, “WHO-Mandated CD4 Monitoring for HIV
Shows Poor Sensitivity, Specificity,” 17th Conference on
Retroviruses and Opportunistic Infections (CROI), Ab-
stract 111, San Francisco, 18 February 2010.
http://www.medscape.com/viewarticle/717302
[11] E. Mylonakis, M. Paliou and J. D. Rich, “Plasma Viral
Load Testing in the Management of HIV Infection,”
American Family Physician, Vol. 63, No. 3, 2001, pp.
483-490.
Copyright © 2011 SciRes. WJA
Role of HIV-1 Viral Load in Initiating Antiretroviral Therapy
Copyright © 2011 SciRes. WJA
154
[12] M. B. Cauldbeck, C. O’Connor, M. B O’Connor, J. A.
Saunders, B. Rao, V. G. Mallesh, N. Kotehalappa, P.
Kumar, G. Mamtha, C. McGoldrick, R. B. S. Laing and K.
S. Satish, “Adherence to Anti-Retroviral Therapy among
HIV Patients in Bangalore, India,” AIDS Research and
Therapy, Vol. 6, 2009, p. 7.
doi:10.1186/1742-6405-6-7
[13] R. Murri, A. Ammassari, A. De Luca, A. Cingolani, P.
Marconi, A. W. Wu and A. Antinori, “Self-Reported Non-
adherence with Antiretroviral Drugs Predicts Persistent
Condition,” HIV Clinical Trials, Vol. 2, No. 4, 2001, pp.
323-329. doi:10.1310/KDM0-RU5W-NVTW-N9MC
[14] F. Forna, C. A. Liechty, P. Solberg, F. Asiimwe, W. Were,
J. Mermin, P. Behumbiize, T. Tong, J. T. Brooks and P. J.
Weidle, “Clinical Toxicity of Highly Active Antiretrovi-
ral Therapy in a Home-Based AIDS Care Program in Ru-
ral Uganda,” Journal of Acquired Immunodeficiency Syn-
drome, Vol. 44, No. 4, 2007, pp. 456-462.
doi:10.1097/QAI.0b013e318033ffa1
[15] R. Nuesch, P. Srasuebkul, J. Ananworanich, K. Ruxrung-
tham, P. Phanuphak and C. Duncombe, “Monitoring the
Toxicity of Antiretroviral Therapy in Resource Limited
Settings: A Prospective Clinical Trial Cohort in Thailand,”
Journal of Antimicrobial Chemotherapy, Vol. 58, No. 3,
2006, pp. 637-644. doi:10.1093/jac/dkl313
[16] C. Adje, R. Cheingsong-Popov, T. Roels, G. Djomand, et
al., “High Prevalence of Genotypic Antiretroviral (ART)
Drug-Resistant HIV-1 Strains among HIV-1-Infected Pa-
tients Receiving ART in Abidjan, Ivory Coast,” Confer-
ence on Retroviruses and Opportunistic Infections, San
Francisco, 30 January - 2 February 2000.
http://gateway.nlm.nih.gov/MeetingAbstracts/ma?f=1022
43620.html
[17] P. Scott, E. Arnold, B. Evans, A. Pozniak, G. Moyle, M.
Shahmenesh, D. White, J. Shirley, P. Cane and D. Pillay,
“Surveillance of HIV Antiretroviral Drug Resistance in
Treated Individuals in England:1998-2000,” Journal of
Antimicrobial Chemotherapy, Vol. 54, No. 3, 2003, pp.
469-473.
[18] J. W. Mellors, A. Munoz, J. V. Giorgi, J. B. Margolick, C.
J. Tassoni, P. Gupta, L. A. Kingsley, J. A. Todd, A. J. S.
R. Detels, J. P. Phair and C. R. Rinaldo Jr., “Plasma Viral
Load and CD4+ Lymphocytes as Prognostic Markers of
HIV-1 Infection,” Annals of Internal Medicine, Vol. 126,
1997, pp. 946-954.
[19] S. Napravnik, C. Poole, J. C. Thomas and J. J. Eron Jr.,
“Gender Difference in HIV RNA Levels: A Meta-Analy-
sis of Published Studies,” Journal of Acquired Immuno-
deficiency Syndrome, Vol. 31, No. 1, 2002, pp. 11-19.
doi:10.1097/01.QAI.0000026906.47530.AB
[20] S. B. Shade, C. Rappoport, D. LeSueur and M. C. Eberly,
“Gender Differences in Viral Load: Results from Six
Clinical Trials,” International Conference on AIDS, Dur-
ban, 9-14 July 2000.
http://gateway.nlm.nih.gov/MeetingAbstracts/ma?f=1022
38735.html