Development and national scale implementation of an open-source electronic laboratory information system (OpenELIS) in Côte d'Ivoire: Sustainability lessons from the first 13 years

PURPOSE

Côte d'Ivoire has a tiered public health laboratory system of 9 reference laboratories, 77 laboratories at regional and general hospitals, and 100 laboratories among 1,486 district health centers. Prior to 2009, nearly all of these laboratories used paper registers and reports to collect and report laboratory data to clinicians and national disease monitoring programs.

PROJECT

Since 2009 the Ministry of Health (MOH) in Côte d'Ivoire has sought to implement a comprehensive set of activities aimed at strengthening the laboratory system. One of these activities is the sustainable development, expansion, and technical support of an open-source electronic laboratory information system (OpenELIS), with the long-term goal of Ivorian technical support and managerial sustainment of the system. This project has addressed the need for a comprehensive, customizable, low- to no-cost, open-source LIS to serve the public health systems with initial attention to HIV clients and later expansion to cover the general population. This descriptive case study presents the first published summary of original work which has been ongoing since 2009 in Côte d'Ivoire to transform the laboratory information management systems and processes nationally.

IMPACT

OpenELIS is now in use at 106 laboratories across Côte d'Ivoire. This article describes the iterative planning, design, and implementation process of OpenELIS in Côte d'Ivoire, and the evolving leadership, ownership, and capacity of the Ivorian MOH in sustaining the system. This original work synthesizes lessons learned from this 13-year experience towards strengthening laboratory information systems in other low resource settings.

Long-term immunological responses to treatment among HIV-2 patients in Côte d'Ivoire

BACKGROUND

Studies indicate that responses to HIV-2 treatment regimens are worse than responses to HIV-1 regimens during the first 12 months of treatment, but longer-term treatment responses are poorly described. We utilized data from Côte d'Ivoire's RETRO-CI laboratory to examine long-term responses to HIV-2 treatment.

METHODS

Adult (≥15 years) patients with baseline CD4 counts < 500 cells/μl that initiated treatment at one of two HIV treatment centers in Abidjan, Côte d'Ivoire between 1998 and 2004 were included in this retrospective cohort study. Patients were stratified by baseline CD4 counts and survival analyses were employed to examine the relationship between HIV type and time to achieving CD4 ≥ 500 cells/μl during follow up.

RESULTS

Among 3487 patients, median follow-up time was 4 years and 57% had documented ART regimens for > 75% of their recorded visits. Kaplan-Meier estimates for achievement of CD4 ≥ 500 cells/μl after 6 years of follow-up for patients in the lower CD4 strata (< 200 cells/μl) were 40% (HIV-1), 31% (HIV-dual), and 17% (HIV-2) (log-rank p < 0.001). Cox Regression indicated that HIV-1 was significantly associated with achievement of CD4 ≥ 500 cells/μl during follow-up, compared to HIV-2.

CONCLUSIONS

Sub-optimal responses to long-term HIV-2 treatment underscore the need for more research into improved and/or new treatment options for patients with HIV-2. In many West African countries, effective treatment of both HIV-1 and HIV-2 will be essential in the effort to reach epidemic control.

Use of pre-ART laboratory screening to identify renal, hepatic and haematological abnormalities in Côte d'Ivoire

BACKGROUND

High demand for HIV-services and extensive clinical guidelines force health systems in low-resource settings to dedicate resources to service delivery at the expense of other priorities. Simplifying services may reduce the burden on health systems and pre-antiretroviral therapy (ART) laboratory screening is among the services under consideration for simplification.

METHODS

We assessed the frequencies of conditions linked to ART toxicities among 34,994 adult, ART-naïve patients with specimens referred to the RETRO-CI laboratory in Abidjan, Côte d'Ivoire between 1998 and 2017. Screening included tests for serum creatinine, alanine aminotransferase (ALT) and haemoglobin (Hb) to identify renal dysfunction (estimated glomerular filtration rate < 50 mL/min), hepatic abnormalities (ALT > 5× upper limit of normal) and severe anaemia (Hb < 6.5 g/dL), respectively. We considered screening results across four eras and identified factors associated with the conditions in question.

RESULTS

The prevalence of renal dysfunction, hepatic abnormalities and severe anaemia were largely unchanged over time and just 8.4% of patients had any of the three conditions. Key factors associated with renal dysfunction and severe anaemia were age > 50 years (adjusted odds ratio (aOR): 2.53; 95% confidence interval (CI): 2.19-2.92; P < 0.001) and CD4 < 100 cells/µl (aOR: 2.57; 95% CI: 2.30-2.88; P < 0.001).

CONCLUSION

The relative infrequency of conditions linked to toxicity in Côte d'Ivoire supports the notion that simplification of pre-ART laboratory screening may be undertaken with limited negative impact on identification of adverse events. Targeted screening may be a feasible strategy to balance detection of conditions associated with ART toxicities with simplification of services.

HIV type 1 drug resistance in adults receiving highly active antiretroviral therapy in Abidjan, Côte d'Ivoire

As antiretroviral therapy continues to scale-up in developing countries, there is concern that high levels of HIV drug resistance to antiretroviral drugs will occur. Here we describe rates of emergence of HIV-1 drug resistance and factors associated with their occurrence among adults who received antiretroviral therapy (ART) for >1 year through the Côte d'Ivoire national drug access program from 1998 to 2003. To detect genotypic drug resistance, we sequenced all 1- and 2-year specimens with detectable HIV RNA viral load. To assess factors associated with emerging drug resistance, we used log normal regression with interval censoring, including covariates in the model for self-reported drug adherence, CD4 cell count, and HIV viral load at therapy initiation, and observed changes in these measures, type of prescribed ART drugs, diagnoses of opportunistic illness, and demographic characteristics. An estimated 14.2% [95% confidence limits (CL) 11.7, 16.9] and 26.6% (95% CL 22.7, 30.8) of patients developed primary drug-resistant mutations within 1 year and 2 years after initiation of therapy, respectively. Factors associated with drug resistance included drug nonadherence, partial or lack of viral suppression, higher viral load or lower CD4 at initiation of therapy, and initiation of ART with what is now considered substandard dual combination therapy. Our results demonstrate the need to strengthen adherence and continuity in treatment programs in order to avoid interruption of ART drugs. Treatment programs should pay attention to indicators of emerging drug resistance: incomplete or lesser decreases in viral load or increases in CD4 cell counts following initiation of therapy, and the occurrence of AIDS opportunistic illnesses.

Virologic and immunologic response to antiretroviral therapy and predictors of HIV type 1 drug resistance in children receiving treatment in Abidjan, Côte d'Ivoire

We describe changes in HIV-1 viral load, CD4+ T cell percentage, and incidence of drug resistance and factors associated with drug resistance for 134 children receiving antiretroviral therapy (ART) for approximately 1 year in Abidjan. Between August 1998 and September 2003, ART was initiated for 395 HIV-infected children ages 0-15 years in the Côte d'Ivoire national drug access initiative. All 1-year samples with detectable HIV RNA >1000 copies/ml were tested for HIV-1 drug resistance and changes in viral load and CD4+ T cell counts were also determined. At treatment initiation, 80% of children had CD4+ T cell percentages <15% and a median viral RNA load of 5.6 log copies/ml. The median age at treatment initiation was 7 years with only 25% of patients less than 4 years of age. Of the 134 children receiving therapy, 72 (54%) had undetectable viral load. The estimated 1-year viral load decline was 1.9 log10 copies/ml and the CD4+ T cell percentage increase was 10.9%. The estimated 1-year cumulative probability for developing any class of drug resistance was 0.44 (95% CI, 0.35, 0.53). In a multivariate analysis, the magnitude of virologic response to therapy was inversely associated with development of drug resistance. Children with less CD4+ T cell rise from baseline values and the use of dual therapy were also associated with the development of drug resistance. Guidelines are needed for the treatment of pediatric HIV infection in Africa in order to minimize the occurrence of drug resistance and enhance better virologic, immunologic, and clinical outcomes.

Impact of genotypic drug resistance mutations on clinical and immunological outcomes in HIV-infected adults on HAART in West Africa

OBJECTIVES

To analyse the association between the presence of resistance mutations and treatment outcomes. The impact of HIV-1 drug resistance mutations in African adults on HAART has so far never been reported.

METHODS

In 2004 in Abidjan, Côte d'Ivoire, 106 adults on HAART had plasma viral load measurements. Patients with detectable viral loads had resistance genotypic tests. Patients were followed until 2006. Main outcomes were serious morbidity and immunological failure (CD4 cell count < 200 cells/microl).

RESULTS

At study entry, the median previous time on HAART was 37 months and the median CD4 cell count was 266 cells/microl; 58% of patients had undetectable viral loads, 20% had detectable viral loads with no major resistance mutations, and 22% had detectable viral loads with one or more major mutations. The median change in CD4 cell count between study entry and study termination was +129 cells/microl in patients with undetectable viral loads, +51 cells/microl in those with detectable viral loads with no mutations and +3 cells/microl in those with detectable viral loads with resistance mutations. Compared with patients with undetectable viral loads, those with detectable viral loads with resistance mutations had adjusted hazard ratios of immunological failure of 4.32 (95%CI 1.38-13.57, P = 0.01). One patient died. The 18-month probability of remaining free of morbidity was 0.79 in patients with undetectable viral loads and 0.69 in those with resistance mutations (P = 0.19).

CONCLUSION

In this setting with restricted access to second-line HAART, patients with major resistance mutations had higher rates of immunological failure, but most maintained stable CD4 cell counts and stayed alive for at least 20 months.

Presence of CRF09_cpx and complex CRF02_AG/CRF09_cpx recombinant HIV type 1 strains in Côte d'Ivoire, West Africa

Based on partial env and pol (protease and RT) subtyping, we recently documented that the majority (>80%) of the HIV-1 strains that circulate in Côte d'Ivoire were CRF02_AG and about 11% were recombinants or could not be clearly assigned to a known subtype or CRF. In order to determine in more detail the precise structure of these viruses we sequenced the full-length genomes for six such strains. Bootscan and phylogenetic tree analysis showed that four strains were complex and unique CRF02_AG/CRF09_cpx recombinants, one was a CRF02_AG/CRF06_cpx recombinant, and one was a pure CRF09_cpx. Reanalysis of the remaining recombinants asserted the predominance of CRF09_cpx within intersubtype recombinants and circulation of CRF09_cpx in Côte d'Ivoire. More detailed analysis of the CRF09_cpx strains revealed also that part of the pol gene belonged to subtype K. This is the first time that such recombinants are described.

Performance of drug-resistance genotypic assays among HIV-1 infected patients with predominantly CRF02_AG strains of HIV-1 in Abidjan, Côte d’Ivoire

OBJECTIVE

We evaluated the performance of three genotypic assays to detect resistant mutations among HIV-1 infected patients with known antiretroviral drug resistance profile in Abidjan, Cote d'Ivoire, most of whom had the circulating recombinant form (CRF02_A/G) of HIV-1.

METHODS

The 56 patients analyzed in this study were enrolled in a pilot program to make available antiretroviral therapy (ART) to HIV-infected patients in Abidjan through the UNAIDS Drug Access Initiative (DAI). These patients had failed ART, as demonstrated by rebound in RNA viral load. Their plasma samples had been previously analyzed for ART genotypic drug-resistance by VircoGEN (Mechelen, Belgium) and were known to have primary and secondary resistance mutations, and also had phenotypic drug-resistance by a recombinant virus assay technology (Mechelen, Belgium). The two assays we evaluated were: VircoGEN, TruGene HIV-1, and ViroSEQ HIV-1 assays.

RESULTS

For the reverse transcriptase gene, all 27 samples that had the T215Y/F mutation were detected by VircoGEN , ViroSEQ, and TrueGene. All 19 (100%) samples that had the K70R/E mutation detected by VircoGEN were detected by ViroSEQ, and 18 (94.7%) by TrueGene. All ten samples with the M184V mutation, three with the K65R, two with the G190A mutation, one with the K103N mutation, and one with the V75T mutation were detected similarly by all three assays. For the protease gene, all three assays detected the I84V (n = 1), M46I (n = 1), and L90M (n = 1) mutations.

CONCLUSION

These results suggest that any of these assays should be considered for monitoring the occurrence of drug resistance among HIV-infected patients receiving antiretroviral therapy in West Africa.

Field evaluation of the gag-based heteroduplex mobility assay for genetic subtyping of circulating recombinant forms of human immunodeficiency virus type 1 in Abidjan, Côte d'Ivoire

The gag-based heteroduplex mobility assay (gag-HMA) was evaluated for its ease and reliability in subtyping circulating recombinant forms (CRFs) of human immunodeficiency virus type 1 (HIV-1) in Côte d'Ivoire. One hundred thirty-two plasma samples were analyzed blindly for HIV-1 subtypes by sequencing the pol gene and by gag-HMA. DNA sequencing was used as the "gold standard." Of the 132 samples sequenced, 108 (82%) were CRF02_AG, 14 (11%) were pure subtype A, 5 (4%) were subtype G, 3 (2%) were subtype D, 1 was CRF01_AE, and 1 was subtype H. The gag-HMA correctly classified 126 (95.5%) of the samples. Of the 108 samples that were classified as CRF02_AG by DNA sequencing, 107 (99%) were correctly identified by gag-HMA, resulting in a positive predictive value of 96.4%. The gag-HMA seems to be a valuable tool for understanding the molecular epidemiology of HIV-1 CRF02_AG in Côte d'Ivoire and West Africa, which could be important for developing and evaluating AIDS vaccines, although DNA sequencing remains necessary for accurate molecular epidemiology.

Prevalence of genotypic and phenotypic HIV-1 drug-resistant strains among patients who have rebound in viral load while receiving antiretroviral therapy in the UNAIDS-Drug Access Initiative in Abidjan, Côte d'Ivoire

OBJECTIVE

To determine the prevalence of genotypic and phenotypic antiretroviral (ARV) drug-resistant HIV-1 strains among patients with viral load rebound while receiving ARV therapy in Abidjan, Côte d'Ivoire.

METHODS

Between August 1998 and April 2000, we selected all patients (n = 241) who had received ARV drug therapy for at least 6 months in the UNAIDS-Drug Access Initiative (DAI), in Abidjan. We analyzed for genotypic and phenotypic drug resistance among 97 (40%) of the 241 patients who had a rebound in plasma viral load, defined as an initial decrease of > 0.5 log10 copies/ml followed by a subsequent increase of > 0.25 log10 copies/ml.

RESULTS

Of the viruses isolated from the 97 patients, 86 (88.7%) had usable sequences and 68 (79%) of the 86 patients had genotypic resistance to at least one reverse transcriptase inhibitor (RTI) or protease inhibitor (PI). Resistant mutations were found for zidovudine in 50 (78%) of 64 patients who had received the drug, 11 (68.7%) of 16 patients on lamivudine, for nevirapine in two (2%), for indinavir in one (1%), and for ritonavir in one (1%). Phenotypic resistance to at least one nucleoside RTI was seen in 45 (56%) of the 80 patients tested, to non-nucleoside RTIs in eight (10%), and to PIs in one (1.3%). Multivariate regression analysis showed factors associated with resistance to be initial treatment with dual therapy (P = 0.04) compared with highly active antiretroviral therapy, and maximal initial viral load response (P = 0.006).

CONCLUSION

Our results demonstrate a high prevalence of ARV drug resistance associated with dual ARV therapy. These results indicate the limited role for dual ARV therapy.