Should monitoring come before treatment?

There are two main reasons why monitoring should be considered as soon as anyone is diagnosed with HIV, ideally before any treatment begins.

The first reason is to get information relevant to the need for treatment, and which will later show whether a person is responding to that treatment. This includes a clinical examination and possibly medical tests to ensure that any other treatable diseases have not been overlooked.

The second reason is to find out whether risk-factors for unwanted side effects of treatment are present. This may guide the choice of treatment and/or enable those side effects to be detected quickly and managed effectively if or when they occur.

This section introduces the WHO recommendations on monitoring which are set out in WHO guidelines on the use of antiretrovirals in resource-limited settings. Later sections go into more detail and describe some of the directions in which research and development efforts in this area are going.

Judging the need for treatment

Where treatment is limited, it may be considered entirely right to give priority to treating people who are ill and at the highest risk of becoming seriously ill and dying.

Some of the most valuable information can be gained without any laboratory tests, for example, by taking a clinical history to find out if a person has been unwell or has lost weight (or, in the case of a child, has stopped growing). Recording a person's weight when they are first seen, and at intervals after that, is a simple example of monitoring which can identify serious illness or record progress on treatmnt.

Some people with HIV can be diagnosed with AIDS and should be offered HAART, regardless of laboratory test results, along with treatment for any diagnosed condition they may have. It is not so easy to advise people with HIV who are symptomatic but do not have AIDS.

A study in Rakai District, Uganda, compared patterns of reported symptoms with CD4 counts in HIV positive pregnant women. Women were asked to report symptoms, which were counted as relevant if included among WHO criteria for major and minor symptoms linked to HIV disease. Of the women who reported no relevant symptoms at all, all had CD4 counts above 350. However, if one or more reported symptoms were used as a criterion for starting treatment, there would be considerable over-treatment. Setting the cut-off as 3 or more symptoms would pick up only 25% of those who should be offered treatment, based on CD4 counts (Nguyen).

Current best practice when someone is diagnosed with HIV includes a CD4 count (or in children, a CD4 and CD8 count) and a viral load test.

The first shows the extent to which the immune system has already been damaged by HIV and the second shows how active the virus is in the person's body.

Unfortunately, current CD4, CD8 and viral load tests are unavailable for most people with HIV. The options are either to find ways to make the tests cheaper and more widely available, or to find alternative tests that can be used in their place. Both are being explored and will be described.

It is equally important to make sure that there is no undetected serious illness, in addition to HIV, which could be treated or which may influence the choice of treatments that should be offered later. Tuberculosis screening is vital to identify people who can benefit from curative or preventive treatment for TB.

Anaemia (low haemoglobin count) has been identified in a number of populations as a risk factor for mortality in HIV disease. It may point to nutritional deficiencies or to treatable parasitic diseases. It is also a risk factor for side effects of ARVs, especially AZT, which is most likely to cause anaemia when taken by people with advanced HIV disease.

Undiagnosed and untreated sexually transmitted infections may place others at increased risk of HIV, if the person is sexually active, and should therefore be considered, investigated and treated as necessary.

Tests for pregnancy are needed, for women to be given options for their own treatment and also to prevent transmission of HIV to a child. If a woman is pregnant, nevirapine or other drugs may be preferred to efavirenz, since efavirenz has been found in animal studies to cause birth defects.

Preventing and managing side effects

Anaemia has already been mentioned as a risk factor for drug side effects, most relevant if AZT is to be considered for an initial regimen. Full blood counts for red cells and white cells can also detect general bone-marrow suppression.

Liver function tests are important, since a number of ARVs can cause liver damage. Hepatitis, if present, also increases the risk of side effects - especially from protease inhibitors, which are eliminated from the body through the liver.

Similarly, amylase tests can point to existing pancreatitis. Some ARVs - in particular ddI - should not be taken by people with pancreatitis.

Peripheral neuropathy is a risk associated with treatment using isoniazid (preventive treatment for tuberculosis) and with the antiretrovirals d4T (stavudine), ddI (didanosine) and ddC (zalcitabine). Peripheral neuropathy should be assessed before a person starts treatment and at regular intervals while a person is on treatment with drugs linked to it.

A later section Monitoring drug side effects discusses these and other side effects in more detail.

WHO on laboratory requirements for use of ARVs (1999)

In 1999, WHO informally published a series of Guidance Modules on Antiretroviral Treatments. Module 5 deals with Laboratory Requirements for the Safe and Effective Use of Antiretrovirals - available here as a Microsoft Word document.

This listed the following areas where diagnostics may be needed:

• Diagnosis of HIV infection through antibody tests or PCR tests for viral genome

• Monitoring adverse reactions to ARV drugs

• Diagnosis of HIV-related infections

• CD4+ T-cell determinations

• Viral load monitoring

• Monitoring of resistance of HIV to ARV drugs

An additional category to be considered includes alternative (candidate) monitoring tests, some of which are commercially available.

WHO on clinical and laboratory monitoring (2002)

In April 2002, the World Health Organisation (WHO) published guidelines for scaling up antiretroviral therapy in resource-limited settings. These guidelines cover critical issues such as when to start therapy and recommended first-line regimens, as well as practical issues, such as the clinical and laboratory monitoring of antiretroviral use.

WHO recommends that in resource-limited settings the basic clinical assessment prior to the initiation of ART includes: documentation of past medical history, identification of current and past HIV-related illnesses, identification of co-existing medical conditions that may influence choice of therapy (such as TB or pregnancy) as well as current symptoms and physical signs.

In order to facilitate the scale up of ARV use in resource-limited settings, WHO prioritised currently available testing into four categories:

• Absolute minimum tests.

• Basic tests.

• Desirable tests.

• Optional tests.

Absolute minimum tests are prerequisites for introduction of ARV therapy in a national programme.

Basic tests are commonly used in the clinical setting and are needed to provide effective monitoring of most ARV regimens. In light of the urgency to provide potentially life-prolonging care to millions of people, the WHO wants to minimise the impediments to care. As such, the basic laboratory tests were not considered to be absolutely essential for programme implementation, although they need to be made available where resources allow.

Desirable tests would make monitoring and evaluation of programme effectiveness much more effective, while optional tests can be used in resource-rich settings.

The absolute minimum laboratory tests to have before initiating antiretroviral therapy are an HIV antibody test and a haemoglobin or haematocrit level. The rationale is that proof of HIV infection is needed prior to starting antiretroviral therapy in the first instance, and screening for anaemia is essential prior to starting AZT (zidovudine) containing regimens.

Basic testing should include a white blood cell count and differential (to permit assessment of neutropenic side-effects and the total lymphocyte count), serum alanine or aspartate aminotransferase level (to assess the possibility of hepatitis co-infection and to monitor for hepatotoxicity), serum creatinine and/or blood urea nitrogen (to assess baseline renal function), a serum glucose, and pregnancy tests for women.

While these tests are not absolutely essential, they are highly recommended in order to be able to provide monitoring for safe use of these agents and inform decisions about switching between regimens.

Desirable supplemental tests include bilirubin, amylase and serum lipids and CD4 cell testing. These tests, while not absolutely essential, are felt to provide significant information that would be beneficial for deciding whether therapy is required and for monitoring the effects of antiretroviral therapy in resource-limited settings.

CD4 cell counts in particular need to be made more widely available in these settings, because they are the best indicator for when to start treatment and to assess immunologic response to treatment.

Viral load testing is currently considered optional because of resource constraints.

Clinical monitoring is essential for the provision of safe and effective antiretroviral therapy. Where laboratory monitoring is limited, close clinical monitoring becomes even more crucial.

References

Nguyen RHN et al. Sensitivity, specificity and predictive value of using symptoms to screen for eligibility of lifelong anti-retroviral therapy among pregnant women. XIV International AIDS Conference, Barcelona, abstract WePeF6704, 2002.

Dodiya VJ et al. Enhancing the role of VCT to VCCTT. XIV International AIDS Conference, Barcelona, abstract ThPeD7777, 2002.

WHO 1999. Guidance Modules on Antiretroviral Treatments. Geneva, WHO, 1999.

WHO 2002. Scaling Up Antiretroviral Therapy in Resources Limited Settings: Executive Summary. Geneva: WHO, April 2002 (http://www.who.int/HIV_AIDS/CARE/CareIndex.htm )