Chapter 3 Antiretroviral Therapy and Comprehensive HIV Care in Resource-Limited Settings

Moses R. Kamya, MD, MPH, Jonathan Mermin, MD, MPH, Jonathan E. Kaplan, MD

INTRODUCTION

In 2006, 39.5 million people worldwide were living with HIV, and over 4 million people were newly infected with the virus in 2006.¹ Combating the AIDS pandemic requires a multi-sectoral, multi-faceted approach with a goal to ensure that countries everywhere come as close as possible to achieving universal access to HIV prevention, antiretroviral therapy (ART), and basic care. In resource-poor settings in Africa and Asia, where 90% of people with HIV/AIDS live, access to ART is limited. Over the past four years, the increasing availability of cheaper generic drug formulations and the launch of initiatives by international agencies, including the Global Fund to Fight AIDS, Tuberculosis, and Malaria and the United States President’s Emergency Plan for AIDS Relief (PEPFAR), have permitted expanded HIV-treatment programs to take root in many countries. As of December 2005, an estimated 1.3 million people were receiving ART in low- and middle-income countries. This represented only 20% of the estimated 6.5 million people in urgent need of antiretroviral therapy in those regions.²

There is compelling evidence that ART can be feasibly administered in resource-limited settings (RLSs) in both adults and children.³^–⁵ Virologic and immunologic effectiveness, reduction in morbidity, and increased survival among patients in RLS are similar to those seen in Western countries.⁶^–¹¹ A review of multiple studies reported a substantial increase in CD4+ T-lymphocyte cell counts, and 73% of patients with an undetectable viral load by a median follow-up of 6 months).³ Many studies report high levels of treatment adherence that are comparable to or better than those of industrialized countries.^12,¹³

Several factors could limit the effectiveness of ART in RLS, including: (1) patients often starting ART with advanced HIV infection, e.g., CD4+ T-lymphocyte cell counts <100 cells/μL)³; (2) a high prevalence of co-infections, including tuberculosis (TB) and bacterial diseases¹⁴^–¹⁶; (3) Interruptions in treatment due to the high cost of antiretroviral (ARV) drugs or transportation to clinics^17,¹⁸ (e.g., the percentage lost to follow-up in RLS ranges from 3.7% to 44%¹⁹); (4) maintaining an uninterrupted supply of drugs^20,²¹; (5) monitoring the efficacy and safety of ARV treatment in countries with limited resources (which is problematic because of the lack of laboratory facilities that can conduct tests indicated for HIV care, such as CD4+ T-lymphocyte cell counts and viral loads, and even where such tests can be performed, most patients with HIV infection do not have access to or cannot afford them²²); and (6) other challenges of ART, including organizational constraints, inadequate human resources, and limited healthcare infrastructure.

Comprehensive care (ART and non-ART) is important in preserving the health of HIV-infected persons and improving quality of life. Ideally, programs should implement care that includes prevention and treatment of opportunistic and other common illnesses, psychosocial support, and prevention of transmission of HIV. To optimize care and treatment for persons with HIV in Africa a two-pronged approach of ARV therapy and prophylaxis/treatment of opportunistic illness is important. Whereas many of the strategies regarding ART are similar regardless of place, the opportunistic infections (OIs) affecting adults and children with HIV vary depending on climate, environment, local infections, and economic conditions. For example, certain illnesses, such as malaria, are common in sub-Saharan Africa; at the same time, some of the infections discussed in current US-based guidelines,²³ such as Mycobacterium avium complex, may be less common in some areas of Africa. Some interventions, such as provision of clean drinking water, are regularly provided by governmental services in the industrialized world, but are less available elsewhere. Because of these differences, it is useful to develop a locally applicable package of standard interventions that can be provided to patients with HIV. Cotrimoxazole (trimethoprim-sulfamethoxazole), which is inexpensive, widely available, and effective in reducing morbidity and mortality, should be the cornerstone of such a preventive care package.²⁴

Taking daily prophylaxis and using other preventive interventions has the additional benefit of providing a foundation for ARV implementation, both on an individual basis with regard to adherence and HIV disclosure to family members, and on a health system level in relation to establishing reliable drug and commodity distribution mechanisms.

This chapter considers some issues related to ART and non-ART care in RLS: when ART should begin, which regimens should be used, how to promote adherence to therapy and prevention of development of drug resistance, how treatment should be monitored, with specific reference to ART toxicity and efficacy, and when to change ART. We also address issues of organizing an ART program and ensuring that patients receive a comprehensive care package including a basic preventive care package and treatment of OIs. We do not address other aspects of HIV care, such as psychosocial and spiritual support and end-of-life care.

WHEN AND HOW TO INITIATE ART

When to Start ARV Treatment–WHO Eligibility Criteria

In April 2002, WHO published the first edition of the WHO ARV guidelines, a Public Health Approach, with simplified standardized tools for the delivery of ART. Updates of these guidelines have been published frequently, considering new scientific data and the experiences of increasing ART scale up in many countries.²⁵ Modified versions of the WHO guidelines are generally implemented in RLS.³ The 2006 criteria for initiating ART are shown in Table 3-1.

BOX 3-1 Revised WHO Clinical Staging of HIV/AIDS for Adults and Adolescents

Primary HIV Infection

• Asymptomatic

• Acute retroviral syndrome

Clinical Stage I

• Asymptomatic

• Persistent generalized lymphadenopathy (PGL)

Clinical Stage IV

Conditions where a presumptive diagnosis can be made on the basis of clinical signs or simple investigations:

• HIV wasting syndrome

• Pneumocystis pneumonia

• Recurrent bacterial pneumonia

• Chronic herpes simplex infection (orolabial genital or anorectal of more than 1 month’s duration)

• Oesophageal candidiasis

• Extrapulmonary TB

• Kaposi’s sarcoma

• Central nervous system (CNS) toxoplasmosis

• HIV encephalopathy

Conditions where confirmatory diagnostic testing is necessary:

• Extrapulmonary cryptococcosis including meningitis

• Disseminated nontuberculous mycobacteria infection

• Progressive multifocal leukoencephalopathy (PML)

• Candida of trachea, bronchi or lungs

• Chronic cryptosporidiosis

• Chronic isosporiasis

• Visceral herpes simplex infection

• Cytomegalovirus (CMV) infection (retinitis or of an organ other than liver, spleen or lymph nodes)

• Any disseminated mycosis (e.g., histoplasmosis, coccidiomycosis, penicilliosis)

• Recurrent nontyphoidal salmonella septicemia

• Lymphoma (cerebral or B cell non-Hodgkin)

• Invasive cervical carcinoma

• Atypical disseminated visceral leishmaniasis

• Symptomatic HIV-associated nephropathy or HIV-associated cardiomyopathy

Table 3-1 WHO Clinical and Immunologic Criteria for Initiating ART

Clinical Staging for Initiation of ART
Clinical Stage (see Revised WHO Clinical Staging, Box 3-1)	CD4+ T-lymphocyte Available	CD4+ T-lymphocyte Not Available
I	CD4+ T-lymphocyte guided	Do not treat
II	CD4+ T-lymphocyte guided	TLC guided
III	Consider CD4+ T-lymphocyte	Treat
IV	Treat	Treat

CD4+ T-lymphocyte Count Criteria for Initiation of ART
CD4+ T-Lymphocyte Count (cells/μL)	Actions
<200	Treat irrespective of clinical stage
200–350	Consider treatment, initiate before count drops below 200 cells/μL
>350	Do not initiate treatment

TLC, total lymphocyte count.

The majority of patients in the developing world who present for ART have CD4+ T-lymphocyte cell counts far below 200 cells/μL. Most of these patients receive a diagnosis of HIV infection only when they present to the hospital with a life-threatening OI.²⁶ At the Infectious Diseases Institute clinic at Makerere University and Mulago Hospital in Kampala, Uganda, the median CD4+ T-lymphocyte cell count at the start of ART was 97.5 cells/μL for 918 patients who started therapy. There is urgent need to establish and scale up routine HIV testing in clinical and other settings to identify HIV-infected patients earlier in the course of HIV infection and to link them to HIV care programs.²⁷

Evaluating the level of immune suppression is important for making decisions regarding initiation of ART and is a challenge in RLS. Flow cytometry serves as the gold standard technique to quantify the CD4+ T-lymphocyte cell population; however, other methods are also available. Table 3-2 shows the different methods of CD4+ T-lymphocyte testing (standard flow cytometry vs alternatives), and some of the advantages and disadvantages of each method. FACSCalibur equipment (Becton-Dickinson (BD) Biosciences, San Jose, CA, USA and EPICS XL, Beckman Coulter, Miami, FL, USA) is expensive but has the highest throughput and less cost per test, making it appropriate for central laboratories. The FACSCount System (BD) for measuring CD4+ T-lymphocyte cell counts is simpler to perform but requires more expensive reagents and therefore costs $5–10 US/test. One of the most crucial needs in the developing world is universal access to affordable and locally usable CD4+ T-lymphocyte testing technology. As access to ART increases worldwide, alternative laboratory monitoring methods are being considered for use in RLS. For example, the Guava Easy CD4+ T-lymphocyte assay (Guava Technologies, Hayward, CA, USA) utilizes microcapillary cytometry for CD4+ T-lymphocyte cell enumeration. Due to small reagent amounts and no requirement for sheath fluid, costs can be minimized–about $1.25 per test. However, the relatively low throughput of the Guava means that this technology has to be applied in smaller health centers with lower demand for CD4+ T-lymphocyte testing. Josefowicz et al conducted a five-site evaluation in the US and Canada and demonstrated that the Easy CD4+ T-lymphocyte assay is precise and accurate when compared to flow cytometry.²⁸ Our recent study showed that the CD4+ T-lymphocyte cell count determined by the GuavaTech Easy CD4+ T-lymphocyte was comparable to that determined by flow cytometry, with a sensitivity and specificity for CD4+ T-lymphocyte below 200 cells/μL of 90% and 98%, respectively.²⁹ Low-cost manual methods for CD4+ T-lymphocyte cell enumeration such as the Dynabeads technique³⁰^–³² now exist. Studies evaluating the performance of the Dynabeads methods for the determination of CD4+ T-lymphocyte cells show high overall correlation coefficients between Dynabeads and FACSCount.^31,³³ Since the Dynabeads method is simpler and cheaper than flow cytometry, it provides an alternative option for the enumeration of CD4+ T-lymphocyte cells in laboratories with limited facilities. However, the manual methods (Dynabeads, Cytospheres) are much more labor intensive, and their application in ART scale-up in RLS is less clear. Considerations in making decisions regarding choice of CD4+ T-lymphocyte testing method besides instrument and reagent costs include training requirements, throughput, equipment maintenance, supply chain, compatibility with proficiency testing materials, and technician time.

Table 3-2 Laboratory Methods to Monitor CD4+ T-lymphocyte Counts in Resource-limited Settings

Studies from Africa show simple markers such as HIV-related symptoms (e.g., WHO clinical stage 4); anemia (e.g., hemoglobin less than 8 g/dL) and total lymphocyte count (TLC) (e.g., less than 1200 cells/μL) are relatively specific for a CD4+ T-lymphocyte cell count <200 cells/μL and can be used to determine initiation of ART; however, they are not sensitive predictors of CD4+ T-lymphocyte cell counts below 200 cells/μL.^34,³⁵ The TLC cutoff point has major impact on the sensitivity and specificity of this marker in predicting CD4+ T-lymphocyte cell counts.³⁶ A higher cutoff increases sensitivity, reducing the likelihood of failing to identify those patients who might benefit from ART. With a lower cutoff (e.g., 1000–1200 cells/μL), sensitivity is lost, but specificity increases.^35,³⁶ In RLS, a lower TLC cutoff point with maximal specificity may be a more practical and cost-effective public health approach for determining eligibility for initiation of ART in asymptomatic or mildly symptomatic HIV-infected patients where CD4+ T-lymphocyte testing is unavailable.³⁶ In RLS (as well as in industrialized countries), the clinical advantages of starting treatment at CD4+ T-lymphocytecell counts above 200 cells/μL are unclear.³⁷

Viral load testing is very costly and is not of much benefit for initiating ART, even in industrialized countries. It should not be part of screening algorithms for initiating ART in RLS.³⁸

How to Start ART

Confirming HIV Infection

Confirming HIV infection is critical before ART is initiated. Patients may present for ART with false-positive HIV serologic results or with only presumed HIV infection. In case of any doubt, clinicians should perform a repeat HIV test before initiating ART.

Baseline Clinical Assessment

Before any patient is started on ART, he/she should undergo baseline clinical assessment including a medical history, physical examination, laboratory investigations, staging of disease (WHO clinical criteria) and counseling and assessment of patients’ readiness to start therapy. The treatment of coexisting infection takes priority over starting ART. ART should not be started in patients with serious co-morbidities or those who are terminally ill. However, in most HIV treatment centers in RLS, the facilities to diagnose OIs are limited. Particularly, in countries with a high burden of TB (which includes most middle- and low-income countries), ART programs are challenged with diagnosing TB. In our experience, ART initiation commonly causes the unmasking of occult TB.³⁹ The rolling out of ART therefore requires that TB diagnostic capacity be strengthened. TB screening protocols should be part of any ART program in endemic areas.

The baseline medical history should include essential demographic characteristics; HIV risk behaviors; the past medical history including major illnesses (particularly TB) and sexually transmitted infections (STIs), hospitalizations and surgeries; the length of time since the diagnosis of HIV infection; symptoms and current medications (including anti-TB drugs, traditional therapies, etc) that could interact with ART or potentially cause additive toxic effect if co-administered with ARV drugs. For women, current or planned pregnancy and the access to contraceptive services should be reviewed.

The physical examination should include vital signs, weight, and detailing of any abnormalities of the skin, eyes (including fundi if possible), oral-pharynx (oropharyngeal thrush generally indicates severe immune suppression), lymph nodes, lungs, heart, abdomen, extremities, nervous system, and genital tract and evidence for symptomatic liver (e.g., jaundice, ascites) or kidney disease (e.g., body swelling).

Laboratory testing involves tests that are helpful before starting ART because they may discourage use of a particular regimen or may be helpful in diagnosing an underlying OI. Absence of laboratory testing should not be a barrier to initiating ART. In almost all cases, if laboratory testing is not available, it is still preferable to begin a clinically eligible patient on ART than to not provide her or him with therapy. If laboratory capacity is available, in addition to assessing the level of immune suppression by CD4+ T-lymphocyte cell testing, a full blood count (or at least a hemoglobin) for patients starting on a zidovudine (ZDV)-containing regimen and a pregnancy test for women of child-bearing potential starting on an efavirenz (EFV)-containing regimen, which is potentially teratogenic, are beneficial before initiating ART. Other baseline tests are important to assess the patient’s health and diagnose any preexisting HIV complications. These tests include:

• Chest radiograph for patients with cough.

• A sputum smear for patients who have coughed for more than 2–3 weeks.

• Syphilis screening (15% of an ART cohort in Botswana were RPR positive ²⁶).

• Symptom-directed laboratory tests needed to diagnose preexisting illnesses.

The cost-effectiveness of performing baseline serum chemistry tests (liver and renal function tests) at baseline in RLSs is not clear, but studies to answer this question are ongoing.⁴⁰ In Uganda, 11% of patients beginning ART had a positive serum cryptococcal antigen (CRAG) test at baseline.⁴¹ About two-thirds of patients with a positive CRAG test who did not have a prior history of cryptococcal infection developed cryptococcal meningitis while on ART. However, there are no specific data on the impact of an intervention of routine screening and treatment of CRAG-positive asymptomatic persons on morbidity or mortality. Table 3-3 summarizes a baseline clinical evaluation checklist for patients starting ART.

Table 3-3 Baseline Clinical Evaluation Checklist for Patients Starting ART

Assessment
1 History Level of understanding of HIV/AIDS; the length of time since the diagnosis of HIV infection Demographics and lifestyle: whether employed and nature of work History of previous ART, prior use of nevirapine during pregnancy Pregnancy risks: contraception options and choices, current or planned pregnancy, access to contraceptive services Sexual risks and disclosure: willingness to practice safer sex, disclosure of HIV serostatus, use of condoms, HIV counseling and testing of sex partners and children Symptoms of chronic pain and depression History of opportunistic infections and other significant illnesses, e.g., TB and STIs, hospitalizations, and surgeries Current medications (including anti-TB drugs, traditional therapies, etc)
2 Physical Exam Weight Nutritional status Functional capacity and level of disability Examination of vital signs, skin, eyes, oropharynx (presence of thrush), lymph nodes, lungs, heart, abdomen, genital tract (STIs), extremities, nervous system
3 Baseline Laboratory Testing to Assess Immunosuppression and Decide on Regimen Confirming HIV serostatus CD4+ T-lymphocyte testing A full blood count for patients starting on a ZDV- containing regimen A pregnancy test for women of child-bearing potential starting on EFV-containing regimen
4 Baseline Laboratory Testing to Assess General Health and Diagnose any Preexisting HIV Complications A sputum smear for AFB for patients who have coughed for more than 2–3 weeks and a chest radiograph for patients who have unproductive cough or whose AFB smears are negative Syphilis screening Serum chemistries (liver and renal function tests), if available Symptom directed laboratory tests to diagnose preexisting illnesses
5 Staging of Disease Using WHO Clinical Criteria
6 Counseling and Assessment of Patients’ Readiness to Start Therapy Including Assessment for Specific Education/Information/Counseling Support Needs

Treatment Education and Assessment of Readiness to Start

Initiation of ART requires a team effort, involving counselors, nurses, caretakers, doctors, and the patient. The preparation of the patient for ART should start with baseline counseling. Time should be spent with the patient to assess his/her ‘readiness’ to start ART. The issues discussed should include:

• Basic knowledge about HIV and available treatment options.

• Interest and motivation in taking therapy and understanding of the life-long commitment to ART.

• A review of the expected benefits and potential side effects of the regimen chosen and the need to report any perceived side effects of the medications. In some cases, the latter may be life saving, for example, the recognition of symptoms related to abacavir (ABC) hypersensitivity or nevirapine (NVP)-associated hepatotoxicity.

• A review of possible drug interactions (such as with oral contraceptives) and food restrictions.

• The critical need to maintain safe sex practices to prevent HIV transmission and re-infection.

• Reproductive health and how to access contraceptive services.

• The importance of drug adherence to a successful outcome. The patient must be willing to complete a personal adherence plan with a counselor, including a discussion on fitting drug dosing into the patient’s routine. If resources are limited, priority for ART slots should be given to patients most likely to be adherent, such as those who have been faithful to their appointments and have adhered to prophylactic medication.

• The importance of disclosure of HIV status to spouse and other family members to allow for adherence support and family HIV testing. Disclosure may be an important indicator that the patient will be adherent.

• The potential benefits of having a treatment supporter to observe treatment on a daily basis.

• Avoidance of drug sharing and drug shopping.

• Assessment of psychosocial and financial barriers that may interfere with clinic visits.

It is common for one or more orientation meetings to occur between the patient and his/her treatment supporter before ART is begun. Adherence should be emphasized at all visits. Motivation sessions that include presentations by persons living with HIV/AIDS are particularly useful. Patients should be informed about the symptoms of ARV drug toxicities and should be aware of the need to seek care and/or to stop therapy in the interim if the need so arises.

CHOICE OF ART REGIMENS IN ADULTS AND ADOLESCENTS

Table 3-4 shows the first- and second-line ARV regimens in adults and adolescents recommended by WHO in 2006.⁴² Most countries have their own guidelines that are generally derived from the WHO guidelines.

Table 3-4 First- and Second-Line ARV Drug Regimens in Adults and Adolescents (2006) as Recommended by WHO ⁴²

	Second-Line Regimens
First-Line Regimens	RTI Component^a	PI Component^b
(ZDV or d4T) + (3TC or FTC) + (EFV or NVP)	ABC + ddI or	ATV/r or
	ABC + TDF or	LPV/r or
	TDF + 3TC ± ZDV	SQV/r
TDF + (3TC or FTC) + (EFV or NVP)	ABC + ddI or	ATV/r or
	ddI + 3TC ± ZDV	LPV/r or
		SQV/r
ABC +(3TC or FTC) + (EFV or NVP)	ddI + 3TC ± ZDV	ATV/r or
		LPV/r or
	or	SQV/r
	TDF + 3TC ± ZDV
(ZDV or d4T) + (3TC or FTC) + (ABC or TDF)	EFV or NVP ± ddI

ZDV, zidovudine; d4T, stavudine; 3TC, lamivudine; FTC, emtricitabine; EFV, efavirenz; NVP, nevirapine; ABC, abacavir; ddI, didanosine; TDF, tenofovir; ATV/r, ritonavir-boosted atazanavir; LPV/r, ritonavir-boosted lopinavir; SQV/r, ritonavir-boosted saquinavir.

a 3TC can be considered to be maintained in second-line regimens to reduce viral fitness.

b Nelfinavir (NFV) or ATV in places without cold chain. TDF cannot be used with unboosted ATV.

Choices for First Line

First-line regimens are generally recommended at the national level to cover the majority of patients; the choice of preferred regimens is largely driven by cost and availability.

Rationale for Choice of Initial ART Regimens

Currently the initial treatment regimens that are widely used in RLS are: non-nucleoside reverse transcriptase inhibitor (NNRTI): EFV or NVP plus a nucleoside reverse transcriptase inhibitor (NRTI) backbone: stavudine (d4T) plus lamivudine (3TC) or ZDV plus 3TC. These first-line regimens prolong life, have a low pill burden, and have the lowest cost at the present time. The current treatment regimens permit rapid scale-up. However, they are also associated with drug toxicities that may be irreversible or lethal.⁴³ Patients at highest risk of these toxicities are those with advanced disease who are often given the highest priority in ART programs in RLS. Currently the data on the incidence and severity of ARV drug toxicities in African patients are still limited.

Nucleoside Reverse Transcriptase Inhibitors

Concerns are increasingly being voiced about the widespread use of stavudine (d4T), given its propensity to cause a progressive disabling peripheral neuropathy especially among patients with advanced HIV and preexisting peripheral neuropathy ²⁶ and those taking concomitant anti-TB drugs.⁴⁴ An ongoing program in rural Uganda is following a cohort of 1029 adult patients on a stavudine-containing regimen. Thirty-six per cent of the patients developed neuropathy (9% severe grade) after 11,268 patient months of observation.⁴⁵ Stavudine is also associated with stigmatizing facial lipoatrophy,⁴⁶ limb fat loss, and lactic acidosis (rare but potentially fatal). Since stavudine is widely used in first-line regimens, patients on stavudine should be monitored closely for neuropathy and lipoatrophy and should be switched to ZDV in a timely fashion to minimize effects of stavudine cumulative toxicity.

Zidovudine also is not without problems; ZDV is associated with anemia, which may lead to blood transfusions and rarely to death. In the DART trial, 21% of patients on ZDV developed a new episode of anemia following initiation of therapy, and 6.6% developed new grade-4 anemia by week 48.⁴⁷

There are alternative drugs without toxicities of this magnitude, although they are all considerably more expensive. Tenofovir (TDF) is at least as potent as d4T and does not have the toxicity profile of d4T or ZDV. TDF-containing regimens may be considered for alternative first-line therapy for the following reasons:

• Relatively low toxicity profile.

• Can be used in pregnancy, and concurrently with TB medication.

• Gilead has recently reduced prices of TDF and emtricitabine (FTC) for the 97 developing countries served by the Gilead Access Program and the new agreement of Gilead with Aspin to produce a generic version may bring the cost down further.

• Once-daily administration: currently TDF is co-formulated with FTC. A single combination pill of EFV, TDF, and FTC has been developed. This combination allows for the possibility of a 1 pill per day regimen, with the obvious potential for improved adherence.

• Added advantage of TDF for treatment of patients co-infected with hepatitis B. Hepatitis co-infection (hepatitis B surface antigenemia) is present in ∼10% of HIV infected patients beginning ART in Botswana.²⁶ This high prevalence has implications for empirical treatment with drugs that have activity against both HIV and hepatitis B.

• The current ‘cost’ of toxic first-line drugs that are cheaper to purchase than less toxic alternatives may be underestimated. The cost of the extra clinic visits, treatment switches, and laboratory monitoring of drug toxicity may be more than purchasing less toxic first-line agents.

• If TDF is used in the first-line regimen, this approach reserves the thymidine analogs (d4T and ZDV) and PIs for second-line therapy.

• Use of TDF, FTC, and EFV has superior treatment outcomes compared to ZDV/3TC plus EFV.⁴⁸

The use of ABC is currently limited by cost.

Non-Nucleoside Reverse Transcriptase Inhibitors

The use of an NNRTI as the third drug is preferable to use of a protease inhibitor (PI), since an NNRTI-containing first-line regimen:

• is less expensive;

• provides the option to use a PI at a later date;

• appears to be safe during pregnancy (NVP, not EFV); and

• allows treatment of TB co-infected patients who are on rifampicin (EFV, not NVP).

Using the fixed-dose combination (currently the least expensive regimen) of d4T, 3TC, and NVP poses a complication with regard to starting and stopping treatment. The recommended NVP dosing regimen starts with a lower lead-in dose of 200 mg q.d. for 2 weeks, followed by 200 mg b.i.d. thereafter. This is based on analyses suggesting that this regimen is less frequently associated with rash. Starting a fixed-dose regimen of combination NNRTI–NRTI treatment without the ‘lead in’ dose of NVP may be associated with increased toxicity. In addition, NVP has a very long pharmacokinetic tail after discontinuation, and a low genetic barrier to resistance. Therefore, stopping use of the fixed-dose combination without continuing the NRTIs for an additional 5 days may lead to NNRTI resistance.⁴⁹

Concern about the use of NNRTIs as first-line ARV treatment is growing because of the increased use of NVP monotherapy to prevent mother-to-child transmission of HIV infection.⁵⁰ One dose of NVP during delivery has been shown to cause significant resistance mutations.^51,⁵² Johnson et al⁵³ showed that resistance mutations emerge in at least 65% of the women after single dose NVP. In a study in Thailand, women who received intrapartum NVP were less likely to have virologic suppression after 6 months of postpartum treatment with an NVP-containing regimen than those who had not received NVP.⁵⁰ The HIV-1 RNA level was less than 50 copies/mL in 49% of the women who had received intrapartum NVP, compared with 68% of the women who had not received intrapartum NVP (P = 0.03). More recent data do not support an adverse impact of NVP monotherapy for PMTCT on subsequent response to NVP-based ART if ART is initiated 6 months or more after receipt of a single, peripartum dose of nevirapine.⁵⁴ Further research to determine clinical implications is needed.

Regimens that contain EFV should not be used by women at risk of pregnancy because of the teratogenic potential for the fetus.⁵⁵ The potential teratogenicity of EFV means that effective contraception is strongly recommended for women taking this drug. Despite the increasing attention to ART in RLS, support for reproductive choice in HIV-infected individuals has lagged behind.⁵⁶ More research is needed to understand reproductive decisions in HIV-infected women and the factors that influence them. Because of drug interactions with rifampin, the use of NVP is generally avoided in patients who require both TB and ART.⁵⁷

PI-Based Regimens

PI-based regimens are an accepted standard of care for initial regimens. However, their high cost relative to NNRTI-based regimens makes their use problematic in RLS seeking to achieve rapid scale-up of therapy. Disadvantages are higher pill counts, food and water requirements in some cases, significant interactions with other drugs that preclude or complicate their use with TB treatment regimens including rifampin, metabolic abnormalities, and the need for a functioning cold chain for ritonavir (RTV)-boosted regimens. In RLS, PIs are generally reserved for second-line therapy.²⁴

Triple NRTIs

The triple-nucleoside combination of ABC, ZDV, and 3TC was associated with an inferior virological outcome, compared with a regimen containing EFV and two or three nucleosides.⁵⁸ However, emerging data suggest that select triple NRTI regimens are simple to administer, have low pill burden, are associated with acceptable HIV RNA reductions and CD4+ T-lymphocyte cell increases, are relatively safe, have fewer drug–drug interactions and are less likely than NNRTI-containing regimens to cause ARV drug resistance when treatment is interrupted. These regimens are being studied⁵⁹ and may be good choices for patients with TB or in whom a preferred or alternative NNRTI- or PI-based regimen may be less desirable due to concerns about toxicity, drug interactions, or regimen complexity. Examples of triple NRTI regimens include: ZDV + 3TC + TDF; ZDV + 3TC + ABC.

MONITORING AND SWITCHING ART

Once patients have been started on ART, they must be monitored for efficacy and toxicity. In developed countries, the efficacy of an ARV treatment regimen is monitored by viral load testing, and decisions to switch therapy are based on increasing viral load, decreasing CD4+ T-lymphocyte cell counts and resistance testing.⁶⁰ Western style monitoring is not affordable and is severely limited by laboratory infrastructure in RLS. Unlike the prices for ARV drugs, the prices for laboratory reagents have not decreased significantly in recent years, although there are efforts to find new and less expensive ways to monitor efficacy of treatment.^61,⁶² The lack of laboratory monitoring should not prevent initiating or continuing ART. In fact, the minimum laboratory testing (in addition to clinical monitoring) that is adequate for monitoring for drug toxicity in RLS is not known. Several studies looking at the effect of different levels of monitoring on patient health outcomes are underway.^40,⁵⁹ In the meantime, WHO recommends clinical monitoring, with symptom- and ART drug regimen-directed laboratory assessment as needed for monitoring ART.

Clinical Monitoring

Without laboratory support, especially in rural settings, clinicians rely more on clinical monitoring only, using symptoms and signs to indicate treatment failure or drug toxicity. Once therapy has begun, clinical assessment should cover: the patient’s perception of how he/she is doing on treatment; changes in body weight over the course of therapy (assuming a reliable food supply); intercurrent symptoms and signs (including changes in the frequency and/or severity of HIV-associated symptoms); signs like fever, diarrhea, thrush; signs of immune reconstitution syndromes; and symptoms and signs of drug toxicities.

Patients must be evaluated for the range of potential adverse events caused by their medications. Patients may complain of minor difficulties in tolerating ARVs, which are often transient and do not necessitate stopping the drug. In contrast, drug toxicities may be severe enough to necessitate change of therapy or temporary discontinuation to assess whether the adverse effect is due to the suspected drug. Major toxicities of commonly used ARV drugs include ZDV-related anemia, EFV-associated CNS symptoms, NVP-associated hepatotoxicity and rash and d4T-related neuropathy, lipoatrophy, and lactic acidosis.

Using clinical symptoms to assess the possibility of treatment failure is a challenge and may be inaccurate, especially during the first 3–6 months of ART. Patients may develop symptoms that are not caused by treatment failure but instead represent the side effects of ARV therapy, an immune reactivation inflammatory syndrome,⁶³ OIs that continue to appear because the patient is still immunocompromised, or an infection/reinfection by a common endemic pathogen such as TB or malaria. If a patient has been adherent with a good ART regimen and has not had previous ARV exposure, very few treatment failures will occur during the first 6 months.⁶⁴ After 6 months to 1 year of ART, clinical manifestations will be more useful for predicting treatment success/failure. This will be particularly so in patients who were symptomatic (WHO clinical stage III and IV) at the start of ART, a group that currently represents the majority of those who start ART in countries with limited resources. Development of a new OI or loss of weight may signal treatment failure. Following the effect of ART on relatively mild HIV-related symptoms and signs such as prurigo may be useful markers of treatment response but remain unproven. Prurigo occurs in at least 10% of African patients with advanced HIV disease.⁶⁵ Once prurigo appears, in the absence of ARV treatment the itching and the papular eruption generally persist and symptomatic treatment is ineffective. In a Ugandan study, pruritic papular eruption significantly improved and in some cases disappeared among patients with advanced HIV infection started on ART (Colebunders Robert, Byakwaga Helen, personal observation).