Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Feasibility of dried blood spots for HIV viral load monitoring in decentralized area in North Vietnam in a test-and-treat era, the MOVIDA project

  • Tuan Anh Nguyen,

    Roles Funding acquisition, Project administration, Supervision, Writing – review & editing

    Affiliation National Reference Laboratory of HIV Molecular Biology, National Institute of Hygiene and Epidemiology, Hanoi, Vietnam

  • Tram Hong Tran,

    Roles Investigation, Supervision, Writing – review & editing

    Affiliation National Reference Laboratory of HIV Molecular Biology, National Institute of Hygiene and Epidemiology, Hanoi, Vietnam

  • Binh Thanh Nguyen,

    Roles Investigation, Writing – review & editing

    Affiliation National Reference Laboratory of HIV Molecular Biology, National Institute of Hygiene and Epidemiology, Hanoi, Vietnam

  • Tram Thi Phuong Pham,

    Roles Investigation, Project administration, Writing – review & editing

    Affiliation National Reference Laboratory of HIV Molecular Biology, National Institute of Hygiene and Epidemiology, Hanoi, Vietnam

  • Nhung Thi Hong Le,

    Roles Investigation

    Affiliation National Reference Laboratory of HIV Molecular Biology, National Institute of Hygiene and Epidemiology, Hanoi, Vietnam

  • Dung Viet Ta,

    Roles Data curation, Investigation

    Affiliation Hanoi University of Public Health, Hanoi, Vietnam

  • Huong Thi Thu Phan,

    Roles Funding acquisition

    Affiliation Vietnam Administration of HIV/AIDS Control, Ministry of health, Hanoi, Vietnam

  • Long Hoang Nguyen,

    Roles Funding acquisition

    Affiliation Vietnam Administration of HIV/AIDS Control, Ministry of health, Hanoi, Vietnam

  • Mohand Ait-Ahmed,

    Roles Project administration

    Affiliation Center for Translational Research, Institut Pasteur, Paris, France

  • Hien Thi Ho,

    Roles Data curation, Investigation, Writing – review & editing

    Affiliation Hanoi University of Public Health, Hanoi, Vietnam

  • Fabien Taieb ,

    Contributed equally to this work with: Fabien Taieb, Yoann Madec

    Roles Conceptualization, Data curation, Funding acquisition, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing

    Affiliations Center for Translational Research, Institut Pasteur, Paris, France, Emerging Diseases Epidemiology Unit, Institut Pasteur, Paris, France

  • Yoann Madec ,

    Contributed equally to this work with: Fabien Taieb, Yoann Madec

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Project administration, Supervision, Visualization, Writing – original draft, Writing – review & editing

    yoann.madec@pasteur.fr

    Affiliation Emerging Diseases Epidemiology Unit, Institut Pasteur, Paris, France

  • for the MOVIDA 2 study group

    The complete membership of the MOVIDA 2 can be found in the Acknowledgments.

Abstract

Background

Access to HIV viral load is crucial to efficiently monitor patients on antiretroviral treatment (ART) and prevent HIV drug resistance acquisition. However, in some remote settings, access to viral load monitoring is still complex due to logistical and financial constraints. Use of dried blood spots (DBS) for blood collection could overcome these difficulties. This study aims to describe feasibility and operability of DBS use for routine viral load monitoring.

Methods

From June 2017 to April 2018, HIV-infected adults who initiated ART were enrolled in a prospective cohort in 43 clinical sites across 6 provinces in North Vietnam. Following national guidelines, the first viral load monitoring was planned 6 months after ART initiation. DBS were collected at the clinical site and sent by post to a central laboratory in Hanoi for viral load measurement.

Results

Of the 578 patients enrolled, 537 were still followed 6 months after ART initiation, of which DBS was collected for 397 (73.9%). The median (inter quartile range) delay between DBS collection at site level and reception at the central laboratory was 8 (6–19) days and for 70.0% viral load was measured ≤30 days after blood collection. The proportion of patients with viral load ≥1000 copies/mL at the 6 month evaluation was 15.9% (n = 59). Of these, a DBS was collected again to confirm virological failure in 15 (24.4%) of which virological failure was confirmed in 11 (73.3%).

Conclusion

Delay of DBS transfer to the central laboratory was acceptable and most viral loads were measured in ≤30 days, in-line with routine follow-up. However, the level of DBS coverage and the proportion of patients in failure for whom a confirmatory viral load was available were suboptimal, indicating that integration of viral load monitoring in the field requires, among other things, careful training and strong involvement of the local teams. The proportion of patients experiencing virological failure was in line with other reports; interestingly those who reported being non-adherent and those with a low BMI were more at risk of failure.

Introduction

In Vietnam, according to the latest UNAIDS report, HIV prevalence has been estimated at 0.3% in the adult population [1]. The epidemic, however, is essentially concentrated in people who inject drug (PWID), men who have sex with men (MSM) and female sex-workers (FSW).

Access to antiretroviral therapy (ART) began in the mid-1990s, and the number of HIV-infected patients on ART rapidly increased in the 2000s. Between 2017 and 2018, it was estimated that the proportion of people infected with HIV who were on ART increased from 50% to 65% [1, 2]. National guidelines, in accordance with the World Health Organization (WHO) guidelines, recommend initiating an ART combination based on two nucleotide reverse transcriptase inhibitors associated with one non-nucleotide reverse transcriptase inhibitor [3].

In 2014, UNAIDS launched the 90-90-90 goals to help end the AIDS epidemic [4]. To increase the number of patients on ART and reach the first two 90 goals, Vietnam initiated a test-and-treat strategy through the implementation of large-scale testing programs and access to ART to all HIV-infected patients, whatever their clinical condition or immuno-virological level.

To help reach the last 90, expanded access to viral load monitoring is crucial. In 2017, it was estimated that about one third of those on ART received viral load testing in Vietnam [5]. This is particularly important as the immunological and clinical criteria to diagnose therapeutic failure have been shown to perform poorly [68]. Laboratories currently able to perform viral load measurements are concentrated in the large urban centers of Hanoi and Ho Chi Minh City, while HIV patients are spread all over the country. The gold standard for viral load monitoring relies on plasma samples, but this requires maintaining a cold-chain from the blood sampling site to the laboratory where the viral load will be measured, which are distant in both space and time. This is costly and subject to technical and logistical difficulties. Prior to this study, routine viral load monitoring was not available in remote settings, neither by plasma nor by DBS. To overcome the challenges of plasma transfer, dried blood spots (DBS) which are easy to collect and easy to transfer, have also performed well at detecting virological failure when compared to plasma [912].

Hence, the MOVIDA 2 project (Monitoring Of Viral load In Decentralised Area, Clinical trials ID: NCT03249493) was implemented to evaluate, in real-life conditions, the feasibility of DBS use for HIV viral load monitoring in remote provinces in North Vietnam, where no routine HIV viral load monitoring was available. Prior to this project, two laboratory evaluations were conducted establishing that, at the central laboratory in charge of viral load measurement, viral load results on DBS and plasma compared well and fulfilled WHO requirements [13, 14].

In the present study we aimed to describe i) the cohort of patients who initiated ART at the time of test-and-treat in remote areas in Vietnam, ii) the process of DBS transfer from remote areas to the central laboratory and iii) virological outcomes at 6 months of ART.

Methods

The MOVIDA 2 project, a prospective observational cohort study implemented in 6 provinces in North Vietnam, enrolled HIV infected adults who initiated ART in 43 clinical sites between June 2017 and April 2018. The choice of the provinces and the clinical sites was at the discretion of the Vietnam Administration of HIV/AIDS control (VAAC) which is a structure within the Ministry of Health in Vietnam. Follow-up in the cohort started at the time of ART initiation.

Following national and WHO guidelines, virological monitoring for patients on ART was scheduled at 6 months, 12 months and every 12 months thereafter [3]. Virological failure was defined at the threshold of 1000 copies/mL. In the case of a viral load above that threshold, guidelines are to offer adherence support to the patient and re-control the viral load within 3 to 4 months.

For viral load monitoring, a venous blood puncture was performed at the clinical site; then, using a calibrated pipette, 70μL of whole blood was deposited on each of the 5 pre-perforated spots of two Munktell TFN (Ahlstrom-Munksjö, Bärenstein, Germany) paper cards. The DBS cards were then left to dry for a minimum of 3 hours at ambient temperature before being packed into individual ziplock bags with 3 desiccants. They were then kept at ambient temperature until transfer.

Once a week, all the individual DBS envelopes available at the laboratory were collected into a larger envelope, with the address of the central laboratory pre-printed on it, along with a transfer form on which the name of the clinical site, the date of transfer and the number of DBS sent were indicated. A member of laboratory staff then personally transported this envelope to the nearest post office for postage. DBS cards were thus transferred weekly by postal service to the National Reference laboratory of HIV molecular biology at the National Institute of Hygiene and Epidemiology (NIHE), Hanoi (Vietnam). However, some clinical sites initially transferred the DBS to an intermediate laboratory in the provincial capital which then transferred the DBS to the laboratory in Hanoi. This concerned a limited number of DBS, and this procedure soon stopped after informing the clinical sites to transfer directly to the laboratory in Hanoi Upon DBS reception at the laboratory in Hanoi, viral load was measured and then DBS cards were stored in a freezer at -20°C.

Prior to this study, in each of the 6 provinces, one training session was organized by the staff from the central laboratory. Two laboratory technicians per clinical site were invited to attend this training and were in charge of dissemination to their fellow laboratory co-workers at clinical site level. The training focused specifically on calibration of blood quantities to drop on each spot, maintenance before packing and DBS storage conditions and transfer. In some sites, technicians had prior experience with DBS for early infant diagnosis, however the rules for blood collection for HIV viral load monitoring are different and justified training them nevertheless. During on-site monitoring visits, laboratory staff were informally interviewed and reminded of good practice.

Viral load measurement on DBS

HIV viral load was measured at the National Reference laboratory of HIV molecular biology at NIHE, Hanoi (Vietnam), using Abbott HIV-1 Real Time assay. Following the manufacturer’s recommendations, one single DBS spot of 70μL of whole blood was placed into a tube with 1.3 mL of Abbott mDBS buffer preparation (Abbott molecular, Des Plaines, IL). The tube was mixed by swirling and incubated for 30 minutes in a heating block (Benchmark Scientific) at 55°C before being directly loaded on the m2000sp platform for sample preparation. Amplification and detection was performed on the m2000rt platform using the open mode 1.0 mL HIV-1 RNA DBS IUO TT version 11 protocol. The lower limit of detection of the assay indicated by the manufacturer was 839 copies/mL.

Statistical analysis

Baseline characteristics were summarized using frequencies and proportions for categorical variables, median and inter quartile range (IQR) for continuous variables; they were compared between PWID and other patients using chi-2 test for categorical variables and student t-test for continuous variables.

All viral load results, measured between 4 and 10 months after the date of ART initiation, were considered as 6 month evaluation. Virological failure was defined as having a viral load >1000 copies/mL at 6 months of ART. Baseline factors associated with virological failure were identified using logistic regression models. Factors associated with a p-value <0.20 in univariate analysis were considered in the multivariate model. Then factors independently associated with virological failure were identified through backward stepwise procedure.

All statistical analyses were performed using Stata 14 (Stata Corp., College Station, TX, USA). A p-value <0.05 was considered significant.

Ethical considerations

The study protocol was approved by the Institutional Review Board from the National Institute of Hygiene and Epidemiology (Hanoi, Vietnam) (decision number 02/QD-VSDTTU), and the Institutional Review Board from Institut Pasteur (Paris, France) (decision number 2016-10/IRB/1). Authorization for data processing has been obtained from French legal authority (Commission Nationale Informatique et Liberté, decision number DR-2017-046). Only participants who provided written informed consent were enrolled, and a random anonymous identification number was assigned to each patient to guarantee confidentiality.

Results

Description of the population who initiated ART

A total of 578 patients were enrolled in the cohort and followed prospectively from the time of ART initiation.

Overall, 261 (45.1%) current or former PWID were enrolled (Table 1), of whom 97 (32.1%) were enrolled in a methadone-substitution program. Some care sites were within drug treatment centres and cared almost exclusively for PWID.

thumbnail
Table 1. Baseline characteristics between people who inject drug (PWID) and the other patients.

https://doi.org/10.1371/journal.pone.0230968.t001

Most of the patients enrolled were male, the proportion of males was even higher in PWID. PWID were also significantly older at ART initiation, although only by a few years

After HIV diagnosis, the median delay to ART initiation was 0.3 months and was significantly longer in PWID (p = 0.034). The 1st line ART combined lamivudine, tenofovir and efavirenz (3TC-TDF-EFV) in all patients. CD4 count was measured prior to ART initiation in only 162 (28%) patients, with this proportion being slightly lower in PWID (p = 0.040). When CD4 count was available, it did not differ between PWID and the other patients (p = 0.63); median CD4 level was 287 cells/mm3.

Overall, hepatitis B and C screening at the time of ART initiation was done in 49.7% and 47.1% of the patients, respectively. These proportions did not differ between PWID and other patients (p = 0.15 and p = 0.25, respectively). In those tested, 31 (10.8%) patients were HBV-infected, no difference was observed between PWID and the other patients (p = 0.98). On the other hand, significantly more patients were HCV-infected in PWID than in the other patients (91 (78.4%) and 14 (9.0%), respectively; p<0.001).

Description of the system for virological monitoring using DBS

Of the 578 patients who initiated ART, 537 (92.9%) were still being followed 6 months after ART initiation. Of these patients, 397 (73.9%) had a DBS collected for the 6 month virological evaluation, 21 (3.9%) did not have a DBS collected because they did not attend a clinical visit in the 4–10 month time interval after the date of ART initiation, 115 (21.4%) did not have a DBS collected although they attended at least one clinical visit in the 4–10 month time interval (Table 2). Finally, in 4 patients the DBS was lost at some point thus virological evaluation could not be performed.

thumbnail
Table 2. Description of DBS network and virological evaluation at 6 month of ART.

https://doi.org/10.1371/journal.pone.0230968.t002

After DBS collection at the HIV care site, the median (IQR) delay until reception at the central laboratory in Hanoi was 8 (6–19) days (Table 2). The median (IQR) delay from reception at the central laboratory to viral load measurement was 8 (7–15) days. Hence, the median (IQR) delay between DBS collection and viral load measurement was 22 (14–35) days, and was ≤30 days in 278 (70.0%) patients.

The quality of the DBS cards was evaluated upon reception at the central laboratory. Of the 397 samples received for the 6 month virological evaluation, 33 (8.3%) had desiccants showing signs of humidity and were then replaced; in 3 samples the number of desiccants did not fulfil recommendations (Table 2).

To try to understand who the patients without DBS collected for the 6 month virological evaluation were, we compared patients seen at M6 with DBS to those seen at M6 without DBS (S2 Table). The proportion of patients without DBS was significantly higher in the Thanh Hoa province compared to all other provinces, particularly in some clinical sites. Patients without DBS had lower CD4 count level at ART initiation (p = 0.04) and were more often not tested for HCV (p = 0.002). On the other hand, availability of DBS was not associated with gender, age, drug use, ethnicity, BMI level at ART initiation, WHO stage at ART initiation, remoteness to the clinical site or HBV status.

Virological evaluation at 6 months of ART

Viral load was measured on all 397 DBS. However, out of the 33 DBS that showed signs of humidity at DBS reception at the central laboratory, 24 led to undetected viral load and were discarded as humidity could have impaired quantification, bringing into question the validity of these undetectable results. Hence, the virological results of 373 patients were considered. Of these, the viral load was ≥1000 copies/mL in 59 (15.8%) patients. Following national guidelines, these 59 patients should have had a second DBS collected to confirm virological failure, but this was done in only 15 (25.4%) of them, of whom 11 (73.3%) were confirmed in virological failure (Table 2).

Factors associated with initial failure (i.e. not considering the confirmatory measurement) were investigated (Table 3). In multivariate analysis, patients who had a low BMI at ART initiation and who self-reported not being fully adherent were more likely to be in virological failure (p = 0.011 and p<0.001, respectively). In some provinces, the risk of failure was increased. After adjusting for these factors, the distance to the care site was no longer associated with virological failure. Former or current drug use was not significantly associated with virological failure. HCV status was not known in about 50% of the patients, so this factor was not considered in the multivariate analysis.

thumbnail
Table 3. Factors associated with virological failure (1st viral load ≥1000 copies/mL, N = 373).

https://doi.org/10.1371/journal.pone.0230968.t003

Discussion

Despite the introduction of DBS use into the WHO guidelines and repeated arguments advocating for its use to widen access to HIV viral load monitoring in remote settings [15, 16], HIV viral load monitoring on DBS remains essentially limited to research settings where DBS results are compared to plasma results. Prior to this study, DBS use was reported in Democratic Republic of Congo, where HIV prevalence was low and patients widespread across the country, and where placement of new laboratories to measure viral load was not considered sustainable [17]. The non-governmental organization Médecins Sans Frontières (MSF; doctors without borders) also implemented viral load monitoring using DBS in some of its programmes, with variable viral load coverage [18]. Another study in Malawi evaluated the feasibility of DBS use for HIV viral load monitoring [19]. Our study is, therefore, one of the very few to use DBS routinely and to evaluate practical aspects of DBS use for routine HIV viral load monitoring in the field.

By simplifying transfer of blood samples and reducing its cost, DBS use is a pragmatic and immediately available option to offer routine virological monitoring to patients on ART in remote areas. We documented the various delays of viral load monitoring to see how DBS fitted into routine care. In Vietnam, clinical visits are planned monthly, or bimonthly when the patient is in a steady health state regarding HIV infection. The median delay between DBS collection at the clinical site and reception in Hanoi at the central laboratory was 8 days, showing the strong commitment of clinical teams. The overall delay between DBS collection and viral load measurement at the central laboratory was 22 days, and was ≤30 days in 70.0% of the patients. In the vast majority of patients, this delay was in line with the time interval between two consecutive clinical visits, meaning that the viral load results could be given back to the patient at the following the visit after DBS was collected. In some patients however, the delay was such that the result would be provided to the patient at the second clinical visit following DBS collection. Longer delays occurred when the clinical site did not transfer the samples weekly, or when viral load measurement was not performed in a timely manner at the central laboratory level. Logistical constraints or lack of familiarity with virological monitoring and DBS collection could explain these delays; these aspects must be addressed to offer better HIV care.

Worryingly, 4 DBS were declared to be lost after collection, with no ability to identify when, where and how this occurred. This study was conducted in real-life conditions, and although not anticipated, this was fortunately marginal. However, the clinical team should have been aware of this loss due to the lack of viral load result, but this was not the case, and no new DBS was collected. This illustrates that tracking of prescribed biological exams does not seem to exist. This has a direct impact at the individual level as these patients were not virologicaly evaluated. However, because these patients expected a viral load result that never came, this could lead to a mistrust in the health system and negatively affect compliance to clinical care and adherence to ART.

At reception at the central laboratory, some DBS exhibited defects that could have impaired virological evaluation. In particular, in 8.3% of the samples, desiccants showed signs of humidity. Rarely (n = 3, 0.8%), the number of desiccants did not comply with recommendations. Training dedicated to laboratory staff took place prior to the implementation of this study, concerning collection of DBS with calibrated blood quantities with single-use cones to prevent contamination as well as procedures for drying, packing, storage and transfer. The deficiencies observed could be due to lack of attention or a changing of staff and should be addressed by regular on-site training and awareness campaigns. Monitoring these quality indicators was important to prevent returning results that may falsely indicate virological control.

Another important issue to address is the absence of DBS collected in a significant proportion of patients at 6 months of ART. This was especially true in the Thanh Hoa province, and particularly so amongst some clinical sites. This situation occurred despite the fact that, prior to the implementation of the study, training were organised dedicated to clinical teams to define when DBS must be collected to comply with the national algorithm. This illustrates that one relies on the involvement and the good will of the clinical teams to manage routine virological monitoring. The impact on workload, at both the clinical and laboratory level, of the introduction of virological monitoring, in sites that did not previously have access to it, should not be minimized and could explain the lack of DBS collection. A recent review emphasized the need and the importance to train clinicians on how to incorporate viral load monitoring in their practice [15]. MSF also insists on the importance of continuous training as to viral load monitoring [18]. A study in Malawi revealed that the introduction of routine viral load monitoring could be felt as a new burden in the case of a shortage of staff [20]. During the course of the study, staff departure or transfer occurred, which could have resulted in increased workload and the arrival of untrained staff which may have induced the flaws apparent in the routine virological monitoring. This also calls for regular campaigns and training to ingrain clinical staff as to the importance of routine viral load monitoring.

This DBS evaluation study was implemented shortly after Vietnam implemented a test-and-treat strategy in order to increase the number of patients on ART. In 2018, it was estimated that 65% of those who are HIV-infected are on ART [1]. In those newly initiating ART, nearly 50% were PWID, formerly or currently using drugs. This result follows the elevated HIV incidence in PWID in Vietnam [1] and is in line with previous findings [21, 22]. Patients newly initiating ART were essentially men, a finding also highlighted in other studies in Vietnam [6, 2326]. This is partly explained by the large representation of PWID in this study; nearly all PWID were male, as previously reported [27]. In Vietnam, the HIV epidemic is also concentrated in MSM and female sex worker (FSW). However, few men identified as MSM and only 3 patients identified as FSW in this study. The under-representation of women in HIV programmes in Vietnam may result from the HIV-related communication being essentially targeted towards key-populations; it is possible that the general population does not feel at risk and hence does not get tested, especially women [28].

As a probable consequence of the test-and-treat strategy, CD4 measurements were no longer measured systematically prior to ART initiation, indeed only 28% of the patients had a CD4 measurement. In this group, the median (IQR) CD4 level was 287 (83–449) cells/mm3, which is higher than in previous reports in Vietnam [24, 29] and in line with the global increase of CD4 level at ART initiation observed in resource-limited settings [30].

In terms of virological results, the proportion of patients with a viral load <1000 copies/mL at 6 months was 84.2%. This is lower than other evaluations which took place at later time points [23, 24, 31], but similar to the proportion observed in a test-and-treat evaluation study in PWID performed in Vietnam [29]. At later time points, the higher rate of virological success could be explained by a survival bias effect as those who were not fully adherent are no longer in active care.

A lower BMI at ART initiation was significantly associated with increased risk of failure. Lower BMI has been found to be associated with mortality [26, 32, 33]. Reasons behind low BMI should be addressed to improve BMI status and thus ART efficacy. Interestingly, those self-identifying as not fully adherent were more likely to be in virological failure, suggesting that targeting patients who need strengthened support for adherence could prove useful. Finally, the rate of virological success differed significantly according to the province. This reflects disparities between clinical sites that were not captured by the factors at our disposal.

Although the national guidelines are, in case of a viral load ≥1000 copies/mL, to provide adherence support and additional viral load testing within 3 to 4 months, this was not achieved in all patients. Only 14 (25.8%) out of 58 patients with primary failure had a confirmatory viral load. Other studies have reported low proportions of patients receiving confirmatory viral load on time, with even worse results [34, 35]. Prior to this study, care sites did not have access to routine viral load monitoring and training was organized to explain the algorithm. It is possible that health staffs working at OPC did not sufficiently understand the need and conditions for confirmatory viral load. It has also been reported that the attendance of all visits necessitated by viral load monitoring was challenging for patients [20]. All this could have contributed to the lack of confirmatory viral load. However, a better understanding of the reasons behind this lack of confirmatory viral load is crucial for the program to work efficiently and to be able to be addressed.

Of those with confirmatory viral load, the majority were confirmed as failures. This is a lot more than usually observed [36], but could be explained by a selection bias from the clinical staff.

Our study has some limitations, the main one being that it is a descriptive study in real-life conditions. Apart from training, no intervention was implemented at the clinician or laboratory levels, thus explaining the delays and DBS coverage. We reported a slightly lower rate of virological success than previous studies [23, 24, 31], and one could object that DBS use could explain this due to the proviral DNA amplification [37]. However, the technique used for viral load measurement on DBS, which is CE-marked, performed well in an evaluation in the same laboratory [14]. Of note, another evaluation in PWID in Vietnam reported a very similar virological failure rate [29].

This study was the first one to use DBS for routine HIV viral load monitoring in the country. The results and progresses from this study are regularly discussed with the Vietnam Administration of HIV/AIDS control (VAAC) that is part of the Ministry of Health. These results have informed the decision from the Ministry of Health to allow DBS use for HIV viral load monitoring in March 2019. Discussions are still on-going on the scaling-up of DBS use in remote settings. In a first stage, DBS use could be opened to all HIV patients followed in the clinical sites that participated in the study, and in a second stage progressively be opened to all the clinical sites caring for HIV patients in these provinces and eventually in other remote provinces where DBS could help implement routine viral load monitoring.

Conclusion

This study documents that DBS use for routine viral load monitoring is feasible in remote areas and that its use fits well within the framework of HIV care. Suboptimal coverage for viral load monitoring and for confirmation of virological failure illustrates the need of adapted training to ensure the strong involvement of clinical teams.

Supporting information

S1 Table. Baseline characteristics of the patients.

https://doi.org/10.1371/journal.pone.0230968.s001

(DOCX)

S2 Table. Comparison of patients with and without DBS at 6 months.

https://doi.org/10.1371/journal.pone.0230968.s002

(DOCX)

Acknowledgments

We are grateful to the following persons for their cooperation and support in the conduct of this study.

National Institute of Hygiene and Epidemiology (NIHE), Hanoi, Vietnam: Anh Duc DANG, Huyen Thi Thanh DAO, Nhung Thi Hong LE, Phuong Khanh NGO, Binh Thanh NGUYEN, Tuan Anh NGUYEN, Tram Thi Phuong PHAM, Ha Thi TONG, Tram Hong TRAN; Institut Pasteur, Paris, France: Mohand AIT-AHMED, Yoann MADEC, Fabien TAIEB; Vietnam Administration of HIV/AIDS Control (VAAC), Hanoi, Vietnam: Nhan Thi Do, An Linh NGUYEN, Huong Thi Thu PHAN, Long Hoang NGUYEN; Hanoi University of Public Health, Hanoi, Vietnam: Hien Thi HO, Dung Viet TA, Phuong Anh Thi NGUYEN; Center for disease control in Lai Chau, Lai Chau, Vietnam: Kien Vu TRAN, Hai Hong NGUYEN, Tinh Duc TRAN; Nam Nhun hospital, Lai Chau province, Vietnam: Thuy Van LO, Manh To PHAM, Diep Thi MAO; Outpatient clinic at the center for disease control in Lai Chau, Lai Chau province, Vietnam: Huyen Thi NGUYEN, Sy Tien HA; Phong Tho hospital, Lai Chau province, Vietnam: Long Thanh TRAN; Tam Duong hospital, Lai Chau province, Vietnam: Cuong Van PHAM, Duyen Thi Quynh TONG, Hanh Huu HA; Tan Uyen hospital, Lai Chau province, Vietnam: Ha Thi LO, Thin Cong TRAN, Hang Thi NGUYEN; Than Uyen hospital, Lai Chau province, Vietnam: Tan Viet LO, Lan Thi DO, Hoa Thi NGUYEN; Center for disease control in Lao Cai, Lao Cai, Vietnam: Suu Van NGUYEN, Hieu Minh TRAN, Quyen Thi Thanh DAO, Thang Xuan NGUYEN; Bao Thang hospital, Lao Cai province, Vietnam: Hien Ngoc VU, Hanh Thi NGUYEN; Bao Yen hospital, Lao Cai province, Vietnam: Thang Duc LE, Toan Khanh TRUONG, Dung Thi Kim LUONG; Bat Xat hospital, Lao Cai province, Vietnam: Tam Thi NGUYEN, Lien Thi Kim NGUYEN, Thang Quang TRAN; Provincal hospital, Lao Cai province, Vietnam: Ninh Thi Thuy LE, Duc Manh NGUYEN, Ly Thi HOANG; Van Ban hospital, Lao Cai province, Vietnam: Thom Thi TRAN, Xuan Thi Thanh TONG, Cuong Xuan NGUYEN; Center for disease control in Phu Tho, Phu Tho, Vietnam: Oanh Trong NGUYEN, Ngoc Xuan NGUYEN, Trung Quang HO; Center for disease control in Thai Nguyen, Thai Nguyen, Vietnam: Minh Binh TRUONG, Truong Van NGUYEN; Dong Hy Medical Center, Thai Nguyen province, Vietnam: Hoa Thi Ngoc NGO, Anh Hoang CAO; Hospital A, Thai Nguyen province, Vietnam: Van Thi HOANG, Manh The NGUYEN; Thai Nguyen City Medical Center, Thai Nguyen province, Vietnam: Van Thị HOANG, Giang Ngoc NGUYEN; Phu Luong Medical Center, Thai Nguyen province, Vietnam: Nhung Thi Kim LUU, Thuat Thien NGUYEN; Pho Yen Medical Center, Thai Nguyen province, Vietnam: Vien Thị LY, Hoanh The NGUYEN; Center for disease control in Thanh Hoa, Thanh Hoa, Vietnam: Truong Ngoc LUONG, Tue Tri LE, Son Truong LE, Nam Vu Dinh, Tam Thi NGUYEN, Ly Thi LE, Lan Thi Thanh LE; Cam Thuy hospital, Thanh Hoa province, Vietnam: Giang Thi LE, Cam Quang HA, Anh Khac CU; Muong Lat hospital, Thanh Hoa province, Vietnam: Tung Thanh TONG, Khang Thi Quach; Ngoc Lac hospital, Thanh Hoa province, Vietnam: Thanh Van NGUYEN, Hien Thi CAO, Hong Thi LE; Nhu Thanh hospital, Thanh Hoa province, Vietnam: Duong Van LE, Thuy Thi NGUYEN; Nhu Xuan hospital, Thanh Hoa province, Vietnam: Phuong Thi Hac, Giap Cong DOAN; Quan Hoa hospital, Thanh Hoa province, Vietnam: Hoanh Cong DINH, Thu Thi TRINH, Tu Thi VY; Quang Xuong hospital, Thanh Hoa province, Vietnam: Dung Phuong Thi TRINH, Hien Thi PHAN, Trang Thi NGUYEN; Sam Son hospital, Thanh Hoa province, Vietnam: Duong Van NGUYEN, Giang Thi LE, Toan Thi HOANG; Thach Thanh hospital, Thanh Hoa province, Vietnam: Toan Manh VO, Thanh Thi PHAM, Phu Dinh NGUYEN; Thanh Hoa City Hospital, Thanh Hoa province, Vietnam: Phong Anh LE, Hien Thi NGUYEN, Binh Xuan DAO; Thanh Lam hospital, Thanh Hoa province, Vietnam: Phuong Thi NGUYEN; Thanh Phong hospital, Thanh Hoa province, Vietnam: Van Thi LE; Tho Xuan hospital, Thanh Hoa province, Vietnam: Ha Thi NGO, Tam Thanh HOANG, Anh Thi TRINH; Thuong Xuan hospital, Thanh Hoa province, Vietnam: Hung Phong LE, Vinh Thi CAM, Thu Thanh Thi LE; Tinh Gia hospital, Thanh Hoa province, Vietnam: Han Quang PHAM, Phiet The LE, Dung Van LE; Center for disease control in Yen Bai, Yen Bai, Vietnam: Van Ha NGUYEN, Tuoi Thi HOANG, Tieu Van PHAN, Hue Thi DOAN, Trieu Thi PHAM; Hong Ca, Yen Bai province, Vietnam: Giang Truong NGUYEN, Thuan Cong HA; Nghia Lo Regional General Hospital, Yen Bai province, Vietnam: Hai Quang NGUYEN, Cuc Thi Thu NGUYEN; Van Chan Medical Center, Yen Bai province, Vietnam: Thanh Thi Phuong DO, Hau Thi NGUYEN, Y Thi HOANG; Van Yen Medical Center, Yen Bai province, Vietnam: Khanh Thi PHUONG, Le Thi TRAN; Yen Bai City Medical Center: Duong Viet VU, Cham Thi TRIEU; Yen Binh Medical Center, Yen Bai province, Vietnam: Ha Thu MAI, Chien Ngoc TRAN, Thao Thi Phuong TRUONG.

References

  1. 1. UNAIDS. http://www.unaids.org/en/regionscountries/countries/vietnam last accessed: October 16, 2019. 2017.
  2. 2. Son VH, Abdul-Quader A, Suthar AB. Using strategic information for action: lessons from the HIV/AIDS response in Vietnam. BMJ Glob Health. 2018;3(5):e000793. pmid:30294455
  3. 3. WHO. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection, second edition -http://www.who.int/hiv/pub/arv/arv-2016/en/ (last accessed: October 17, 2019). 2016.
  4. 4. UNAIDS. 90-90-90: An ambitious treatment target to help end the AIDS epidemic (available at: http://www.unaids.org/sites/default/files/media_asset/90-90-90_en.pdf; last accessed: October 17, 2019). 2014.
  5. 5. HIV and AIDS data hub for Asia Pacific (available at: https://www.aidsdatahub.org/Country-Profiles/Viet-Nam; last accessed: April 11, 2019). 2017.
  6. 6. Le NK, Riggi E, Marrone G, Vu TV, Izurieta RO, Nguyen CKT, et al. Assessment of WHO criteria for identifying ART treatment failure in Vietnam from 2007 to 2011. PloS one. 2017;12(9):e0182688. pmid:28877173
  7. 7. Rawizza HE, Chaplin B, Meloni ST, Eisen G, Rao T, Sankale JL, et al. Immunologic criteria are poor predictors of virologic outcome: implications for HIV treatment monitoring in resource-limited settings. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2011;53(12):1283–90.
  8. 8. Mee P, Fielding KL, Charalambous S, Churchyard GJ, Grant AD. Evaluation of the WHO criteria for antiretroviral treatment failure among adults in South Africa. Aids. 2008;22(15):1971–7. pmid:18784460
  9. 9. Monleau M, Aghokeng AF, Eymard-Duvernay S, Dagnra A, Kania D, Ngo-Giang-Huong N, et al. Field evaluation of dried blood spots for routine HIV-1 viral load and drug resistance monitoring in patients receiving antiretroviral therapy in Africa and Asia. J Clin Microbiol. 2014;52(2):578–86. pmid:24478491
  10. 10. Rutstein SE, Kamwendo D, Lugali L, Thengolose I, Tegha G, Fiscus SA, et al. Measures of viral load using Abbott RealTime HIV-1 Assay on venous and fingerstick dried blood spots from provider-collected specimens in Malawian District Hospitals. J Clin Virol. 2014;60(4):392–8. pmid:24906641
  11. 11. Schmitz ME, Agolory S, Junghae M, Broyles LN, Kimeu M, Ombayo J, et al. Field Evaluation of Dried Blood Spots for HIV-1 Viral Load Monitoring in Adults and Children Receiving Antiretroviral Treatment in Kenya: Implications for Scale-up in Resource-Limited Settings. Journal of acquired immune deficiency syndromes. 2017;74(4):399–406. pmid:28002185
  12. 12. Zeh C, Ndiege K, Inzaule S, Achieng R, Williamson J, Chih-Wei Chang J, et al. Evaluation of the performance of Abbott m2000 and Roche COBAS Ampliprep/COBAS Taqman assays for HIV-1 viral load determination using dried blood spots and dried plasma spots in Kenya. PloS one. 2017;12(6):e0179316. pmid:28622370
  13. 13. Taieb F TH, Ho TH, et al. Evaluation of Two Techniques for Viral Load Monitoring Using Dried Blood Spot in Routine Practice in Vietnam (French National Agency for AIDS and Hepatitis Research 12338). Open Forum Infectious Diseases. 2016.
  14. 14. Taieb F, Tran Hong T, Ho HT, Nguyen Thanh B, Pham Phuong T, Viet Ta D, et al. First field evaluation of the optimized CE marked Abbott protocol for HIV RNA testing on dried blood spot in a routine clinical setting in Vietnam. PloS one. 2018;13(2):e0191920. pmid:29425216
  15. 15. Peter T, Ellenberger D, Kim AA, Boeras D, Messele T, Roberts T, et al. Early antiretroviral therapy initiation: access and equity of viral load testing for HIV treatment monitoring. Lancet Infect Dis. 2017;17(1):e26–e9. pmid:27773596
  16. 16. Roberts T, Cohn J, Bonner K, Hargreaves S. Scale-up of Routine Viral Load Testing in Resource-Poor Settings: Current and Future Implementation Challenges. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2016;62(8):1043–8.
  17. 17. Boillot F, Serrano L, Muwonga J, Kabuayi JP, Kambale A, Mutaka F, et al. Implementation and Operational Research: Programmatic Feasibility of Dried Blood Spots for the Virological Follow-up of Patients on Antiretroviral Treatment in Nord Kivu, Democratic Republic of the Congo. Journal of acquired immune deficiency syndromes. 2016;71(1):e9–15. pmid:26413848
  18. 18. Médecins Sans Frontières. Making viral load routine—Successes and challenges in the implementation of routine HIV viral load monitoring (available at: https://www.msf.org/sites/msf.org/files/making_viral_load_routine_part_1_programmatic_strategies.pdf; last accessed: April 5, 2019). 2016.
  19. 19. Rutstein SE, Hosseinipour MC, Kamwendo D, Soko A, Mkandawire M, Biddle AK, et al. Dried blood spots for viral load monitoring in Malawi: feasible and effective. PloS one. 2015;10(4):e0124748. pmid:25898365
  20. 20. Rutstein SE, Golin CE, Wheeler SB, Kamwendo D, Hosseinipour MC, Weinberger M, et al. On the front line of HIV virological monitoring: barriers and facilitators from a provider perspective in resource-limited settings. AIDS Care. 2016;28(1):1–10. pmid:26278724
  21. 21. Tran BX, Nguyen NP. Patient satisfaction with HIV/AIDS care and treatment in the decentralization of services delivery in Vietnam. PloS one. 2012;7(10):e46680. pmid:23071611
  22. 22. Tran BX, Dang AK, Truong NT, Ha GH, Nguyen HLT, Do HN, et al. Depression and Quality of Life among Patients Living with HIV/AIDS in the Era of Universal Treatment Access in Vietnam. Int J Environ Res Public Health. 2018;15(12).
  23. 23. Dat VQ, Duong BD, Nhan DT, Hai NH, Anh NTL, Thu HHK, et al. Viral load suppression and acquired HIV drug resistance in adults receiving antiretroviral therapy in Viet Nam: results from a nationally representative survey. Western Pac Surveill Response J. 2018;9(3):16–24. pmid:30377546
  24. 24. Tanuma J, Matsumoto S, Haneuse S, Cuong DD, Vu TV, Thuy PTT, et al. Long-term viral suppression and immune recovery during first-line antiretroviral therapy: a study of an HIV-infected adult cohort in Hanoi, Vietnam. Journal of the International AIDS Society. 2017;20(4).
  25. 25. Auld AF, Shiraishi RW, Mbofana F, Couto A, Fetogang EB, El-Halabi S, et al. Lower Levels of Antiretroviral Therapy Enrollment Among Men with HIV Compared with Women—12 Countries, 2002–2013. MMWR Morb Mortal Wkly Rep. 2015;64(46):1281–6. pmid:26605861
  26. 26. Tran DA, Shakeshaft A, Ngo AD, Mallitt KA, Wilson D, Doran C, et al. Determinants of antiretroviral therapy initiation and treatment outcomes for people living with HIV in Vietnam. HIV Clin Trials. 2013;14(1):21–33. pmid:23372112
  27. 27. Lan CW, Lin C, Thanh DC, Li L. Drug-related stigma and access to care among people who inject drugs in Vietnam. Drug Alcohol Rev. 2018;37(3):333–9. pmid:28762584
  28. 28. Nguyen TA, Oosterhoff P, Hardon A, Tran HN, Coutinho RA, Wright P. A hidden HIV epidemic among women in Vietnam. BMC Public Health. 2008;8:37. pmid:18221565
  29. 29. Nguyen HH, Bui DD, Dinh TT, Pham LQ, Nguyen VT, Tran TH, et al. A prospective "test-and-treat" demonstration project among people who inject drugs in Vietnam. Journal of the International AIDS Society. 2018;21(7):e25151. pmid:29984473
  30. 30. IeDea Collaborations CC. Global Trends in CD4 Cell Count at the Start of Antiretroviral Therapy: Collaborative Study of Treatment Programs. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2018;66(6):893–903.
  31. 31. Rangarajan S, Donn JC, Giang le T, Bui DD, Hung Nguyen H, Tou PB, et al. Factors associated with HIV viral load suppression on antiretroviral therapy in Vietnam. J Virus Erad. 2016;2(2):94–101. pmid:27482442
  32. 32. Madec Y, Laureillard D, Pinoges L, Fernandez M, Prak N, Ngeth C, et al. Response to highly active antiretroviral therapy among severely immuno-compromised HIV-infected patients in Cambodia. Aids. 2007;21(3):351–9. pmid:17255742
  33. 33. Naidoo K, Yende-Zuma N, Augustine S. A retrospective cohort study of body mass index and survival in HIV infected patients with and without TB co-infection. Infect Dis Poverty. 2018;7(1):35. pmid:29690932
  34. 34. Mwau M, Syeunda CA, Adhiambo M, Bwana P, Kithinji L, Mwende J, et al. Scale-up of Kenya's national HIV viral load program: Findings and lessons learned. PloS one. 2018;13(1):e0190659. pmid:29324811
  35. 35. Free Virus Elution Protocol for HIV Viral Load Testing from DBS on Roche cobas® AmpliPrep/cobas® TaqMan HIV-1 Test v2.0. Available at http://www.idc-dx.org/resources/free-virus-elution-protocol-for-hiv-viral-load-testing-from-dbs-on-roche-cobas. Accessed on january 2016
  36. 36. Etoori D, Ciglenecki I, Ndlangamandla M, Edwards CG, Jobanputra K, Pasipamire M, et al. Successes and challenges in optimizing the viral load cascade to improve antiretroviral therapy adherence and rationalize second-line switches in Swaziland. Journal of the International AIDS Society. 2018;21(10):e25194. pmid:30350392
  37. 37. Monleau M, Montavon C, Laurent C, Segondy M, Montes B, Delaporte E, et al. Evaluation of different RNA extraction methods and storage conditions of dried plasma or blood spots for human immunodeficiency virus type 1 RNA quantification and PCR amplification for drug resistance testing. J Clin Microbiol. 2009;47(4):1107–18. pmid:19193835