Secondary Logo

Journal Logo

SPECIAL ARTICLES: Guidelines Section


Author Information
Infectious Diseases in Clinical Practice: March-April 2002 - Volume 11 - Issue 3 - p 154-157
  • Free


[WHO http://www. _2002]

WHO has released guidelines for antiretroviral therapy for resource limited areas from an International Writing Committee led by Dr. Scott Hammer from Columbia University in New York City. The release of this document was coordinated with the release of the UN Global Fund that is intended to provide some of the resources for implementing these strategies. The stated goal of the WHO document is treatment of three million persons by 2005.

Laboratory Testing Requirements

Requirements for laboratory testing must be realistic for the areas where these guidelines are to be implemented, which means that common methods of evaluation such as the CD4 cell count may not be available and viral load testing is very unlikely. Listed below are those tests designated as absolute minimum requirements, those considered basic testing for evaluation and toxicity monitoring, other tests that are desirable, and the optional viral load assay.

  • Absolute Minimum: HIV Serology, Hgb or BCT
  • Basic testing: WBC, ALT or AST, creatinine or BUN, glucose, pregnancy test
  • Desirable: Bilirubin, amylase, lipids, CD4 count
  • Optional: Viral load

Indications To Treat

Indications for initiating antiretroviral therapy are divided into two categories: one for when the CD4 cell count is available and the other for when the CD4 count is not available. In the latter category, there will often be reliance on the total lymphocyte count (TLC), which is considered “less useful in asymptomatic patients.” The threshold for significance of the CD4 cell count is 200 cells/mm3 and for the TLC it is 1000–1200 cells/mm3 combined with HIV-related symptoms. Specific criteria based on laboratory testing and the WHO classification of HIV in stages I–IV are summarized below:

CD4 count available

  • WHO stage IV* (AIDS-defining diagnosis)
  • WHO stage I–III* plus CD4 < 200 cells/mm3

CD4 count not available

  • WHO stage IV*
  • WHO stage II* or III* plus TLC
  • <1000–1200 cells/mm3 (includes thrush and recurrent invasive bacterial infections; does not include asymptomatic patient regardless of TLC)

Clinical stages

  • Clinical Stage I: Asymptomatic or PGL and/or normal activity
  • Clinical Stage II: weight loss <10%, minor mucocutaneous conditions, zoster <5 years, recurrent URIs and/or symptomatic plus normal activity
  • Clinical Stage III: Weight loss >10%, unexplained diarrhea >1 mo, unexplained fever >1 mo, thrush, oral hairy leukoplakia, pulmonary TB in past year, or severe bacterial infection and/or bed-ridden <50% of days in the past month
  • Clinical Stage IV: CDC-defined AIDS and/or bed-ridden >50% of days in the past month

Selection of Antiretroviral Regimen

Of the 16 FDA-approved antiretroviral agents, 13 were selected for inclusion; delavirdine, ddC and amprenavir were excluded. The preferred nucleosides are AZT/3TC, based on safety in pregnancy and the availability of a co-formulated form. The second dual nucleoside regimen advocated is d4T/3TC with “others” including d4T/ddI and AZT/ddI. For initial treatment, the recommendation is for two nucleosides (usually AZT/3TC) combined with one of the following: an NNRTI (efavirenz or nevirapine), abacavir, nelfinavir or a ritonavir-boosted PI including indinavir, saquinavir (hard gel or soft gel formulations) or lopinavir. The RTV-boosted regimens all use the low dose of ritonavir with 100 mg twice daily. Important considerations in the selection of these regimens are the simplicity of the treatment in terms of pill burden, frequency of administration, pregnancy or possibility of pregnancy, side effects, the need for co-administration of rifampin for tuberculosis and the need for refrigeration of ritonavir to maintain stability with storage over 30 days. Table 1 summarizes recommendations including the PI dosage regimens and Table 2 identifies advantages and disadvantages of the drug(s) added to the two nucleosides.

WHO Recommended Initial Therapy for Resource Poor Settings
Selecting the “Third Drug”

Antiretroviral Therapy with Pregnancy

Multiple considerations apply in treating pregnant women, but the “bottom line” is the preferred regimen of AZT + 3TC in combination with either nelfinavir or nevirapine. This reflects the availability of data to support efficacy and safety of these drugs in pregnancy. The main alternative is AZT + 3TC combined with saquinavir and ritonavir. The issues relevant to drug selection in pregnancy are outlined below, and are based on established merit in preventing perinatal transmission, drugs that have adequate pharmacokinetic studies to assure adequate levels in the pregnant state, drugs known to be toxic or theoretically toxic in pregnancy, and drugs which show trans-placental transmission. Recommendations follow.

Drugs that have established efficacy in preventing perinatal transmission

  • Placenta crossing (which presumably protects infant, but reduction in viral load also protects infant): All NRTIs and NVP cross the placenta. PIs do not cross the placenta.
  • Breast milk passage: Documented with AZT, 3TC and NVP; suspected with all antiretrovirals.
  • AZT, 3TC, d4T, ddI, NVP and NFV have pharmacokinetic studies to justify dosage regimens similar to non-pregnant patients. ABC, EFV and LPV/RTV have not been studied
  • Larger doses may be required for IDV and SQV, but RTV boosting may provide adequate levels

Toxicities in pregnancy

  • EFV: Teratogenic, avoid in 1st trimester
  • d4T + ddI: Lactic acidosis (avoid)
  • PIs: Glucose intolerance
  • IDV: Theoretical risk of neonatal hyperbilirubinemia


  • AZT has the largest experience for safety and efficacy
  • NFV has the best studies and is the best tolerated “third drug”

Recommendations for ART in Pregnancy

Patient diagnosed with HIV during pregnancy

  • Consider delay ART until after first trimester (10 weeks)
  • Preferred Regimen: AZT + 3TC + NFV or NVP
  • Alternative regimen: AZT + 3TC + SQV/r
Patient becomes pregnant while taking ART
  • Suspend treatment for toxicity (especially hyperemesis) or until after the first trimester, then continue or change.
  • Suspension in the 1st trimester is often not feasible but if done, all ART drugs should be stopped
  • Change agents: EFV, d4T/ddI

Patient is taking ART and breast-feeding:

Continue ART

Patient takes short course ART to prevent perinatal transmission:

Prior exposure to NVP, AZT or AZT/3TC should not deter use of these drugs pending further study.

Antiretroviral Therapy With Tuberculosis Co-infection

This is an important issue for the developing world where rates of active tuberculosis co-infection are exceptionally high. Dual treatment is confounded by the extraordinary pill burden and drug toxicity. The highest priority is given to immediate treatment of tuberculosis with the timing of introduction of ART dependent on the stage of HIV infection. The selection of agents is based on drugs that do not pose problems for interaction with rifampin. The specific recommendations in terms of timing of ART and the regimen selected are summarized below.

Antiretroviral therapy in patients with TB co-infection

  • Regimens for ART: AZT + 3TC + either ABC, EFV, SQV/r or NVP

When to start antiretroviral therapy

  • CD4 <50 cells/mm3 or extrapulmonary TB: Start as soon as TB therapy tolerated
  • CD4 50–200 cells/mm3 or TLC <1000–1200: Start after TB therapy × 2 months
  • CD4 >200 cells/mm3 or TLC >1000–1200: Monitor CD4 counts if available

Treatment Failure

Three definitions of treatment failure are provided based on clinical observations (new HIV-related complication after sufficient time for immune reconstitution), immunologic failure (a CD4 cell count increase of at least 30%) or virologic failure (no consensus except that persistent detectable virus generally indicates failure). In resource-limited regions it is anticipated that viral load testing will generally not be available, requiring dependence on clinical observations or CD4 cell counts to define failures. These are relatively crude and late indicators, so a high frequency of resistance mutations is anticipated. Guidelines for “rescue regimens” based on the initial regimen are provided below; it is assumed that resistance testing will not be available for individual patients (but may be available in sentinel labs under supervision of the IAS).


  • Clinical Failure: Development of a new AIDS-defining opportunistic infection or tumor when HAART given for sufficient time for immune restoration (must be differentiated from “immune reconstitution syndrome”)
  • Immunologic Failure: CD4 count decreases >30% from peak or returns to or below pre-treatment baseline
  • Virologic Failure: No consensus definition and not easily applied to resource-limited settings; “continued detectable viremia indicates incomplete viral suppression”

Factors that contribute to HIV resistance

  • Lack of viral load monitoring to detect treatment failure
  • Intrinsic resistance of HIV-2 and Group 0 HIV-1 subtype to NNRTIs
  • Failure to follow safe sex practices and other harm-reduction interventions
  • Interruptions in drug supply (if one component of a regimen is not available, the entire regimen should be temporarily stopped)
  • Potential contribution of regimens of limited potency (NVP or AZT/3TC) to prevent perinatal transmission is unknown.

Prophylaxis for Opportunistic Infections

Although somewhat beyond the mission statement of antiretroviral therapy, these guidelines also provide a useful summary of WHO recommendations for OI prophylaxis and include the use of TMP-SMX in patients who satisfy the criteria for antiretroviral therapy as discussed above. There is also a recommendation for fluconazole prophylaxis to prevent cryptococcosis in areas where there is a high prevalence of this complication when combined with a CD4 cell count <100 cells/mm3. INH prophylaxis for tuberculosis is recommended in those with a positive PPD, but only when resources are available to exclude active tuberculosis and with the caveat that this may not be feasible in many areas. These recommendations are summarized in Table 4.


OI Prophylaxis
Rescue Therapy for Treatment Failure*
© 2002 Lippincott Williams & Wilkins, Inc.